report 2016 - economic and financial developments · 2017-03-17 · by 25 basis points in december...

Report 2016

Economic and financial developments

Globaleconomyandeuroarea

43Economicandfinancialdevelopments ❙ GloBAl EcoNoMy AND EuRo AREA ❙

1. Globaleconomyandeuroarea

1.1 Modestbutstableglobalgrowth

In 2016, theglobaleconomymaintainedmodestgrowthdespiteahighlyuncertainenvironment.Followingahesi‑tantstart,thereweresignsofgrowingmomentuminthesecond half of the year. Supported by monetary policyaccommodation and the continuing low energy prices,theadvancedeconomiesdisplayedsomeresiliencedespite

the fears aroused by Brexit in particular. Therewas thusonly a slight dip in growth in the euro area, the UnitedKingdom and Japan. Conversely, there was a markedslowdownintheUnitedStates.Intheemergingeconomiesand commodity-exporting countries, growth was stable,butverywidevariationspersistedbetweencountries.Pastreductionsincommoditypricesandgenerallymodestwageincreasescontinuedtorestraininflationworldwide.

Table 1 GDP OF THE MAIN ECONOMIES

(percentage changes in volume compared to the previous year, unless otherwise stated)

p.m. Contribution to

global GDP growth

p.m. Share of

global GDP (1)

2014

2015

2016

2016

2010

2016

Advanced countries . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 2.1 1.6 0.7 46.3 41.9

of which :

United States . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 2.6 1.6 0.3 16.8 15.6

Euro area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 2.0 1.7 0.2 13.7 11.9

Japan (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 1.2 0.9 0.0 4.9 4.1

United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . 3.1 2.2 2.0 0.0 2.5 2.3

Emerging countries . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 4.1 4.1 2.2 53.7 58.1

of which :

China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 6.9 6.7 1.1 13.9 17.9

India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 7.6 6.6 0.5 5.9 7.3

Russia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.7 −3.7 −0.6 0.0 3.6 3.1

Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.1 −3.8 −3.5 −0.1 3.2 2.6

World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 3.2 3.1 3.1 100.0 100.0

p.m. World trade (3) . . . . . . . . . . . . . . . . . . . . . . . . . 3.8 2.7 1.9

Source : IMF.(1) According to the IMF definitions and calculated on the basis of purchasing power parities.(2) The data for 2015 and 2016 take account of the switch from the System of National Accounts 1993 to System of National Accounts 2008 and the updating of the reference year

from 2005 to 2011. The Japanese authorities made these changes in December 2016.(3) Average of exports and imports of goods and services.

44 ❙ ECONOMICANDFINANCIALDEVELOPMENTS ❙ NBB Report 2016

Stabilisation in the emerging economies

China encountered no major problems in continuingtorebalance itseconomyinfavourofdomesticgrowthbased on the strengthening of consumption and a gradual shift in activity from industry towards services.GDP growth dipped again slightly, to an average of6.7% in the first three quarters of 2016, down from6.9%in 2015.Butitremainedrobustandstayedwithinthe official target range of between 6.5% and 7%.The strong credit expansion invigorated the propertymarket, while the additional fiscal stimulus measuresbolstered consumption andgave a significant boost topublic investment. In contrast, the growth of privateinvestmentweakened considerably, despite a revival inresidentialinvestment.Thatslowdown,reflectedinafallintotal investment, ispartlyduetotheincreasedshareof services in theChineseeconomy–asmany servicesarenotopentoprivatecapital–andtoexcesscapacityinsomeindustrialsectors.Netexportscontractedsharplyowingtoanaemicglobaldemandandthecountry’slossofcompetitivenessinlowvalueaddedsectors.

Monetary policy, which has been gradually easedsince 2014,remainedexpansionaryin 2016.ThePeople’sBank of china still faces the challenge of supporting the economy while safeguarding financial stability. Againstthebackdropofstrongcreditexpansionwhichhasfuelledthe rapid rise in corporate debt and driven up propertypricessignificantlyinsomeregions,itchosetokeepitskeyinterestratesforloansanddepositsunchangedat4.35%and1.5%respectively.Inaddition,itmadeonlyonead‑justment to the reserve requirement coefficient, cuttingitfrom17.5%to17%inMarch.Inordertocombatthecountry’s debt problems, the central government intro‑duced policies aimed at supporting the debt restructuring of local authorities and public enterprises.

Since the US Federal Reserve was expected to raise itskeyinterestrates,thecapitaloutflowfromChinaalreadyapparent in 2015continued,varying in intensityaccord‑ingtochangesinthefinancialmarkets’perceptionoftheeconomicsituationinChina. Inthesecircumstances,theexchangerateremainedunderpressure.Thus,attheendof 2016,therenminbislumpedtoitslowestlevelagainsttheUSdollarinmorethaneightyears.ThePeople’sBankofChinadippedheavilyintoitsexchangereservestosup‑portthedomesticcurrency.Butitstillhadtoresolvetoac‑companyitsdepreciationbygraduallyreducingitscentralrate against the us dollar.

In the commodity-exporting countries, economic activ‑ity in general continued to suffer from the weaknessof demand and hence the low level of prices for those

products, due in particular to the ongoing transition oftheChineseeconomy.Brazil languishedinadeepreces‑sionwhichhadbegunin 2015,exacerbatedbyapoliticalcrisis that seriously dented confidence in the country.However,theslowdowninactivityseemedtoeaseduringtheyear,inthewakeofarecentriseincommoditypricesandaslightexpansionofexports,resultingfromthepastdepreciationof the real.TheRussianeconomy,draggeddownbythefall inoilpricesandtheinternationalsanc‑tionsimposedsincemid-2014inresponsetotheconflictinUkraine,showedsignsofstabilisationin 2016follow‑ing the rise in crude oil prices. Finally, Indiamaintainedsound economic growth supported by an improvementinthetermsoftrade,afallininflation,andanumberofreformsfavourabletothebusinessclimate.Nevertheless,thegovernment’sdecisionon8 Novembertodemonetisethe500 and1000 rupeedenominationstoputastoptotheinformaleconomymayhavecurbedeconomicactivityfromtheendoftheyear.

The advanced economies remained resilient

Whilethedollar’sappreciationandthefallinoilpriceshadputatemporarybrakeonexportsandinvestmentattheendof 2015and inearly 2016, theAmericaneconomygainedmomentum from the thirdquarter. The strongerGDPgrowthwasbasedvery largelyonamarkedexportrecoveryand,toalesserextent,fasterstock-buildingandarevivalinfederalgovernmentspending.Incontrast,pri‑vatebusinessinvestmentremainedweakwhileconsumerspending,which is still themainengineofactivity,con‑tinuedtorisesteadily.Itwassupportedinparticularbyastrengtheninglabourmarketandfurtherwageincreases.Theunemploymentratedroppedbelow5%,whilehourlywages increased towards the end of the year at theirstrongest rate since 2009. Labour market participationstabilised, thoughat just under63%at the endof theyear,itremainedbelowitspre-crisislevelofaround66%.

Despite signs of gradual improvement on the jobsmar‑ket, the Federal Reserve adopted a prudent approach.In the face of uncertainty over the real dynamism ofthe economy, and as inflation still remained below itstargetof2%,themembersoftheFederalOpenMarketCommittee(FOMC)leftthekeyinterestratesunchangedforalmosttheentireyear.Thestatusquowhichhaslongprevailedcontrastswiththeexpectationsconcerningthefederal funds rates published in December 2015,whenthe Federal Reserve had raised its key interest rates forthe first time since 2008, from a range of 0.00-0.25%to0.25-0.50%.Atthattime,allmembersoftheFOMCwereexpectingat least two rate increasesduring 2016.Nonetheless, takingaccountof theactualandexpecteddevelopmentsonthelabourmarketandontheinflation


front, theFedeventually increased its interest ratebandby 25 basis points in December 2016. It also foreseesfurtherrisesin 2017.

Thefiscalpolicystancewaseasedslightlyduringthefiscalyear 2016.Thepublicdeficitgrewfrom4.4%to5%ofGDPbetweenOctober 2015andSeptember 2016,pro‑pellingthepublicdebtto115.6%ofGDP.

InJapan,theeconomymaintainedmodestgrowthover‑all, underpinnedbybuoyant consumptionas a resultofthegradualimprovementinemploymentandhouseholdincomes. InNovember, theunemployment ratestoodat3.1%,afterhavingdroppedbackto3%inJuly,itslowestlevelsinceJanuary1995.However,thisfavourablepictureneedstobequalifiedslightlyasitstemsfrombothariseindemand for labour and a contraction of the labour force. Japanese economic growth benefited from extremelyfavourablefinancial conditions andgovernment supportmeasures. Conversely, it was restrained by weaker de‑mandfromtheemergingeconomies,particularlyinAsia,andtheappreciationoftheyen,twofactorsthatconsid‑erablydepressedJapan’sexports.

DespitetherelativelyhealthystateoftheJapaneseecon‑omy, inflation still returned tonegativefigures in 2016.The Bank of Japan therefore maintained its asset purchase programme aimed at expanding the monetary base bysome80000 billionyen(oraround€615 billion)ayear.It also took various steps to ease its monetary policyfurther.Inviewoftheworldwidefinancialtensionatthebeginning of the year and the uncertainty surroundingthe economic outlook in the emerging countries, theinterestrateondepositswasthustakenintonegativeter‑ritoryat–0.1%,attheendofJanuary.Inaddition,anewmonetarypolicyframeworkwasintroducedattheendofSeptember,aimedinparticularatraisinginflationexpecta‑tionstowardsthetargetof2%,butalsointendedtopre‑ventanydecline inthefinancialsector’sprofitability inalowinterestrateenvironment.Thatframeworkcomprisestwomainelements.Thefirstaimsto influencetheyieldcurvebytightercontroloverlong-andshort-terminterestrates.Thesecondisaformalcommitmenttoexpandthemonetarybaseuntilinflationremainsabovethe2%tar‑get“inastablemanner”.TheBankofJapanalsostatedthat it did not rule out the option of cutting its main policyratefurther,andifnecessaryadjustingitsassetpur‑chases.Turningtothebudget,theconsolidationprocesswhichbeganin2013 wasinterruptedinfiscalyear 2016.Threefiscalstimulusplanswereproduced,whileafurtherincrease inVAT,originallyscheduledforApril 2017,waspostponedtoOctober 2019.Inthatcontext,theprimarydeficithoveredaround5%ofGDPandthegrosspublicdebtexceeded230%ofGDP.

In theUnitedKingdom, the referendumon23 JuneanditsoutcomeinfavourofBrexithadnosignificantadverseeffectoneconomicactivity.QuarterlyGDPgrowth,whichcameto0.6%inthethirdquarter,washardlyany lowerthan in thesecondquarter,andconsiderablyhigher thanexpected. Thus, supported by the vigour of the servicessectorwhichrepresentsalmost80%oftheeconomy,the2%activitygrowthover 2016asawholewasonlyslightlybelow the 2015 figure. The unemployment rate, whichstood at 4.8% inNovember, dropped to its lowest levelfortenyears,thusstimulatingprivateconsumption.Thesefavourabledevelopmentsareprobablydue inpart to theBankofEngland’snewmonetarypolicymeasures.Atthebeginning of August, it cut its benchmark Bank Rate by25 basispointsto0.25%.Italsodecidedtostepupitstotalpurchasesofgovernmentbondsfrom£375 billion(around€509 billion)to£435 billion(around€590 billion),topur‑chase£10 billionworthofbondsissuedbynon-financialcorporations, and to introduce long-term refinancingoperations in order to ensure the effective transmission of itsmonetaryeasing tohouseholdsandbusinesses.Whilethe slump in sterling following the referendumdoes notseem to have produced benefits as yet for sectorsmoregearedtowardsexports,suchas industryandagriculture,it has brought higher inflation. Nonetheless, at 1.6% inDecember, inflation is stillwell below the target of 2%.AlthoughBrexit’smacroeconomic impacthasbeenminorsofar,itislikelytobepainfulinthelongterm.ThequestionmarkovertheformoffuturerelationsbetweentheEUandtheUnitedKingdomisexpectedtosubduenewinvestmentandjobcreation,andsomesignsofthatwerealreadyap‑parentin 2016.With thehigherbarrierstotrade,financial

CHART 1 LOWINFLATIONINTHEADVANCEDECONOMIES

(consumerpriceindex,changecomparedtothecorrespondingperiodofthepreviousyear)

2011 2012 2013 2014 2015 2016–2

–1

0

1

2

3

4

5

6

United Kingdom

Japan

United States

Euro area

Sources:EC,Japan’sMinistryofInternalAffairsandCommunications,U.S.Bureauof labor.


transactions and the movement of people that could result from Brexit, that uncertainty could thus ultimately curbeconomicgrowthinboththeUnitedKingdomanditstrad‑ing partners.

Growth of world trade slows again

Thegrowthofworld tradeslackenedpaceagain in 2016.Thatwasduemainlytoaslowdownoftradeintheemerg‑ingcountriesandinNorthAmerica.Theworldwidesluggish‑nessofeconomicactivity,andmorespecifically investment–thedemandcomponentwiththelargestimportcontent –,isconsideredtobeamajorreasonforthisslowergrowth.China’s transition to an economy geared to consumptionand services – and therefore less dependent on imports of rawmaterials andmachinery – and the decline in capitalspendingbycountriesexportingcommoditiesareprobablythe dominant factors here. Other reasons put forward toexplain theweaker expansion ofworld trade also includethe lack of progress in trade liberalisation – or an actual rise in protectionism as evidenced by the restrictions observedbytheWorldTradeOrganisation(WTO)–andthefactthatglobalvaluechainsarenolongerbeinglengthened,in con‑trasttotheirrapidextensionoverthepasttwodecades.

Rising commodity prices

While thepriceperbarrel ofBrent crudehaddroppedbelow $30 in mid-January, oil prices staged a strongrecovery throughout the first half of 2016, exceeding

the $50 mark during the summer. That rise was duelargelytoinvoluntaryproductionstoppagesthatreducedsupplies,thoughoversupplypersisted.Conflictsofvari‑ous kinds and difficult weather conditions interruptedproductioninanumberofmajoroil-producingcountriessuch as Libya, Nigeria, Venezuela and Canada. At thesametime,globaldemandprovedstrongerthanexpect‑ed,thoughwithoutregaining itsprevioushigh level. Inthesecondhalfoftheyear,oilpricesweremorevolatile.AttheendofNovember,theOPECcountriesagreedtocutproduction,andthatdroveoilpricesabove$50 perbarrel again.

Overall,pricesofcommoditiesexcludingenergyalsore‑cordedamodestriseinthefirsthalfoftheyear,beforestabilisingforatime.Thatincreaseisdirectlyrelatedtotherecoveryofenergyprices,asenergyisamajorinputboth for industrial raw materials and for agriculturalproduction.On topof that, foodpricesalso roseasaresult of disappointing harvests,while the increase inmetalprices likewise reflects a slight strengtheningofglobaldemand.Attheendoftheyear,thestatementsmade by the US President-elect concerning the intro‑duction of an infrastructure stimulus plan caused prices to surge in anticipation.

Highly favourable financial conditions

Supported by the accommodative monetary policy, fi‑nancialconditionsremainedveryfavourableoverall,even

CHART 2 FURTHERSLOWDOWNINWORLDTRADEAGAINSTTHEBACKDROPOFRISINGCOMMODITYPRICES

2011 2012 2013 2014 2015 2016–4

–2

0

2

4

6

8

10

12

14

2011 2012 2013 2015201425

45

65

85

105

125

145

165

185

2016

INTERNATIONAL TRADE(seasonally adjusted three-month moving averages of export and import volumes ; percentage changes compared to the previous year)

United States

Emerging countries

World Euro area Food commodities

Brent crude (per barrel)

Industrial commodities(index 2007 = 100)

COMMODITY PRICES(daily data, in $)

Sources:CPB,OECD,ThomsonReutersDatastream.


thoughthefinancialmarketsexperiencedseveralperiodsof turbulence.

A first bout occurred in late 2015 and early 2016,against the backdrop of new fears over the outlookforgrowthintheemergingcountries,especiallyChina.Acrosstheworld,Januarybroughtoneofthemostin‑tensesell-offsofriskyfinancialpositionssincethe2008financial crisis. The tension spread to the advancedeconomies and triggered a flight to quality. In thesecircumstances, the yield curve flattened while stockmarketsfalteredandriskpremiumswentup.Thesede‑velopmentswerefollowedbydisruptionthathadapar‑ticularimpactonthebankingsector,wheresharepricesfell sharply following publication of disappointing re‑sults. More generally, at the end of January, doubtsover the strengthof theAmerican recovery combinedwiththeBankofJapan’sdecisiontotakeitsovernightdepositrateintonegativeterritoryreinforcedthepros‑pect of a lengthy period of low interest rates, givingrisetofearsfortheprofitabilityofthefinancialsector.In parallel with brighter prospects for world growth,investors nevertheless soon regained their appetite for risk, which calmed volatility, supported asset pricesand lowered risk premiums. Sentiment regarding theemerging economies also improved as the advanced economies expected interest rates to remain low foraprolongedperiod, concernoverChina’sgrowthwasdiminishing, and commodity prices were picking up.The renewed flow of capital to those countries wasaccompanied by a strong stock market rally and theappreciationofthosecountries’currencies.

At theendof June, themarkets reactedstrongly to thevote in favour of the United Kingdom’s departure fromtheEU.Awaveofoptimismhaddrivenupassetpricesintheprecedingdays,buttheunexpectedreferendumresultcaused stock markets to tumble and depressed risk‑free yields.At the same time, sterling slumpedwhile theUSdollarandtheyenappreciated.Althoughtheinitialreac‑tion was strong, the turmoil subsided once the centralbanks stated that theywere ready to provide sufficientliquidity and, if need be, adopt new monetary easingmeasures such as those taken by the Bank of England.The tension then dissipated further against the back‑ground of a hesitant improvement in the outlook for the globaleconomy.Nonetheless,sterlingremainedlowand,following the reappraisal of the future monetary policystance,yieldsonsovereignbondsremainedclosetotheirhistoricalfloor.

The unexpected outcome of the American presidentialelectionon8 Novemberalso sent shockwaves throughboth thefinancialmarketsand thecommoditymarkets.The prospect of a massive infrastructure investment plan andtaxcuts,asadvocatedbythePresident-elect,hadabeneficialeffectonequitymarketsingeneral.Followingabriefpanicintheimmediateaftermathofthevote,stockmarkets rallied overall, especially in the United Stateswheretheindicesreachednewrecordlevels.Bycontrast,inviewoftheinflationriskofsuchafiscalstimulusplanfor the American economy in a virtually full employ‑ment situation, and therefore the likelihood that theFederalReservemight raise itskey interest rates sooner,investors turned away from the bond markets. Yields

CHART 3 HISTORICALLYFAVOURABLEFUNDINGCONDITIONS

–1

0

1

2

3

4

40

60

80

100

120

140

160

2011 2012 2013 20152014 2016 2011 2012 2013 20152014 2016

Germany US Japan

STOCK MARKET PRICES(daily data, indices 2007 = 100)

EURO STOXX (EA) S&P 500 (US)

Nikkei 225 (JP) Emerging countries

YIELD ON TEN-YEAR GOVERNMENT BONDS(in %)

source : Thomson Reuters Datastream.


onUS Treasuries climbed steeply during theweek aftertheelection,withtheten-yearyieldrisingbymorethan40 basispoints.TheriseinAmericaninterestratesspreadto other regions of the world, including the euro areaand the emerging economies.On the foreign exchangemarkets, the US dollar appreciated against most othercurrenciesandparticularly in relation to someemergingcurrenciessuchastheMexicanpesoandtheBrazilianreal.ThestatementsmadebythePresident-electsparkedfearsoverthefuturetradingrelationshipsoftheUnitedStates,and hence over the macroeconomic outlook for a number of countries. The changes at the top of the American ad‑ministrationrekindledtheuncertaintysurroundingfutureUS fiscalandtradepoliciesinparticular.

TheeuroremainedfairlystableagainsttheUSdollarformuchof 2016beforedepreciatingattheendoftheyear.OwingtothestatusquolongmaintainedbytheFederalReserve following the December 2015 rise in the keyinterestrates,thedollarwasnotexpectedtostrengthen.Conversely,thesinglecurrencyedgedupwardsineffec‑tive terms, but that rise masked divergences betweencurrencies.Itwasdueessentiallytoastrongappreciationagainstthepoundsterling,whichmorethanoffsetthedeclineagainsttheyenandemergingcurrenciessuchasthe Russian rouble and the Brazilian real. A significantfactor behind the strengthening of the yen until theautumnwas its status as a safe haven in an uncertainworld. The appreciation of the Russian rouble and the

Brazilian real reflectsboth the rise incommoditypricesand the more encouraging macroeconomic outlook for those countries.

1.2 Lowinflationinthecontextofamodestrecoveryintheeuroarea

Intheeuroarea,theslowbutsteadygrowthofeconomicactivity evident sincemid-2013 continued in 2016. ThevolumegrowthofGDPcameto1.7%asanannualaver‑age,downslightlyagainstthe 2015figureof2%.

Activitywassupportedbydomesticdemandonthebackof favourable funding conditions resulting in particular fromafurthereasingofEurosystemmonetarypolicyandaslightlyexpansionaryfiscalpolicy intheeuroareaasawhole. Employment was also dynamic. Conversely, theimpetusgeneratedbytheearlierdepreciationoftheeuroand the fall inoilprices in 2014and 2015fadedaway.External demand remained weak. There were also sev‑eral factors that continued to hamper the dynamismofthe economy. They included theongoingbalance sheetadjustment in a number of sectors and countries, theinsufficientimplementationofstructuralreforms–whichshouldboostpotentialgrowthinthelongtermandbol‑sterconfidenceanddemandintheshortterm–,andtheemergence of numerous uncertainties. Overall, produc‑tioncapacity remainedunderused in 2016, though toasteadilydiminishingdegree.Despitesubstantial jobcrea‑tion,unemploymentrateswerestillhigh,andtherecentinvestmentrevivalwasnotenoughtofillthegap.Thislastfactorisakeydeterminantoftheexceptionallyslowpaceofrecoveryfromthecrisis.

Among the sources of uncertainty affecting the globaleconomy, those relating to theUnitedKingdom’s futureexit fromtheEUwereofparticularconcern to theeuroarea.Sofar,however,theyhavehadlittleimpactoncon‑fidence.Despite the progress achieved in stabilising thefinancial system and strengthening the capitalisation ofEuropeanbanks,fearsaboutthebanks’profitabilityand,in some Member States, the high outstanding amountof non-performing loans on their balance sheet mayhave impeded their ability to fund business activity andinvestment.Finally,inaclimateofmistrustamongsomesectionsofthepopulation,asuccessionofscheduledelec‑tions and referendums made it harder for governments to conductpoliciesthathelptostrengthentheeconomyinthelongtermbutmayhavetransitionaleffectswhichareperceived as negative.

With growth still insufficient to put a rapid end to theunder-utilisationofproductioncapacity intermsofboth

CHART 4 EXCHANGERATESSUBJECTTOPOLITICALUNCERTAINTY

(exchangeratesagainsttheUSdollar,indices2011=100)

30

40

50

60

70

80

90

100

110

2011 2012 2013 2014 2015 2016

Euro area

Brazil

Russia

United Kingdom

China

Japan

source : Thomson Reuters Datastream.


labourandcapital,andanoutlookcloudedbynumerousuncertainties, inflation remained very low in the euroarea. Although the year-on-year change in the harmo‑nised index of consumer prices (HICP),which had beenzeroin 2015,didbecomeslightlypositiveagainin 2016atanannualaverageof0.2%,thatmeagrerisewasduemainlytothedisappearanceofthenegativebaseeffectsof theenergycomponent followingtherise inoilpriceson the international markets. Core inflation remainedstable throughout theyear,hoveringarounda low levelof0.9%.Withnostrengtheningofwagedynamicsintheeuro area, headline inflation shows no convincing signsofconvergingontheEurosystem’starget,namelyalevelbelow,butcloseto,2%overthemediumterm.

ExceptforGreece,allMemberStatescontributedtotheincreaseintheeuroarea’sGDP.Amongthemainecono‑mies, Ireland recorded the strongest growth. In Spain,growth remained robust while it gained momentum inGermany.Conversely,inPortugal,FinlandandItaly,eco‑nomicexpansionwaswellbelowtheeuroareaaverage.InGreece,GDPdeclined,butlesssteeplythanduringthegreat recession.

Economicactivityintheeuroareawaspredominantlysup‑portedbyprivateconsumption,whichwasunderpinnedbysubstantialjobcreation,aidedinsomecountriesbytheeffectsofpastreforms.Togetherwithlowinflation,thesedevelopments increased the real disposable income of householdsdespitethecontinuingsluggishnessofwages.The household savings rate remained stable overall, in‑dicating that the gains in purchasing power were usedmainlyforconsumerspending.

Publicconsumption likewisemadeanon-negligiblecon‑tribution to growth in the euro area. Still, there aremarkeddifferencesfromonecountrytoanother.Insomecountries the growth of public consumption resultedfromadditionalexpenditureonaccommodatingrefugeesortighteningsecuritymeasures.InGermany,publiccon‑sumptiongrewtwiceasfastasGDP.

The increase in gross fixed capital formation was alsorelatively strong at 3.3%, similar to the 2015 figure. Itwas supported not only by favourable financing condi‑tions,butalsobyarecoveryinthegrossoperatingsurplusoffirms,whiletheuseofproductioncapacity improved.

CHART 5 LOWINFLATIONINTHECONTEXTOFAMODESTRECOVERYINTHEEUROAREA

2011 2012 2013 2014 2015 2016

–3

–2

–1

0

1

2

3

2011 2012 2013 2014 2015 2016

GDP (1)

Change in inventories

Net exports

Gross fixed capital formation

Public consumption

Private consumption

CONTRIBUTION TO ANNUAL GDP GROWTH(volume data ; percentage points, unless otherwise stated)

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

INFLATION(HICP, percentage annual change)

Total excluding energy and food

Total

source : Ec.(1) Percentagechangecomparedtothecorrespondingquarterofthepreviousyear.


However,investmentwasheldbackbythestillmutedanduncertain demand outlook, particularly in regard to theglobaleconomyandtheemergingeconomiesatthestartoftheyear,butalsobytheprocessofdeleveragingintheprivate sector of some countries and the uncertainties that persistedin 2016.

ExpressedasapercentageofGDP,investmentintheeuroareaasawhole in 2016wasstillwellbelowtheaveragelevelrecordedinthepre-crisisyears.That investmentgapprimarily reflects the sharpdecline in theMember Statesseriously affected by the crisis. In Ireland, where invest‑menthadalreadybouncedback in2012, the investmentshortfall has now clearly been made up, and the ECestimates Ireland’s investment growth at 16% in 2016.Investment gave a substantial boost to the economy ofthe Netherlands and Spain, but also to slower-growingeconomiessuchasFinland,andevenGreece.Inmostcases,the largestcontributioncamefromcorporate investment,although in some countries – including the Netherlands,FinlandandGermany–investmentinhousingalsoplayedamajorpart.Conversely,inPortugal,thefallininvestmentinhousingandpublicinvestmentseriouslyhamperedgrowth.

The Investment Plan for Europe, and more particu‑larly theEuropeanFund forStrategic Investments (EFSI),have already produced concrete results. Proposed inNovember 2014bytheECPresident,Jean-Claude Juncker,thatplanwasintendedtotriggerinvestment.TheEFSIwascreated inorderto increasethecapacityforrisk-bearing

finance in the EU by providing, over a period of threeyears, guarantees forprojectfinancingby the EuropeanInvestmentBank(EIB)andtheEuropeanInvestmentFund(EIF).Duringthatperiod,the investmentplanwouldpo‑tentiallygenerateinvestmentamountingto€315 billion.The resourcesmadeavailableby theEFSIup to theendof 2016 could finance new investment totalling around€164 billion in theEUMemberStatesasawhole, thusmaking a significant contribution to themobilisation ofprivate investment. In December 2016, the Council de‑cidedtoextendthecapacityoftheEFSIintermsofbothdurationandfinancialresources.

Whiledomesticdemandwasquitebuoyantin 2016,asithadbeeninthepreviousyear,exportgrowthweakened.The sluggishness of external demand confronting theeuroareasincetheendof 2015persistedin 2016,inlinewiththeweaknessofglobaltrade.WhiletradewithintheEUwas relatively resilient, itbarelycompensatedfor thedecline inexportsoutsidetheEUduringthefirsthalfoftheyear,especiallyasthenominaleffectiveexchangerateoftheeuroappreciatedslightlyatthattime.However,theeurolostgroundfromNovemberonwards.

In parallel with modest activity growth, the situation on the euro area labour markets improved further in 2016

Job creation in the euro area and in most of the Member Stateswas fairly robust,which isnoteworthy in viewof

CHART 6 RECOVERYCONTINUESINTHEEUROAREABUTTHEREISSTILLASUBSTANTIALINVESTMENTGAP

EL PT ES IT IE FI AT NL DE FR BE EA–14

–12

–10

–8

–6

–4

–2

0

2

4

IE ES DE NL AT FR BE PT FI IT EL EA–1

0

1

2

3

4

5

2014 2015 2016

26.3

INVESTMENT AS A PERCENTAGE OF GDP (in percentage points)

GDP AT CONSTANT PRICES(percentage change compared to the previous year)

8.5

Gap in 2016 compared to the average in 1999-2006

Largest gap since 2009

source : Ec.


thesluggishexpansionofeconomicactivity.Onereasonfor that contrast may be a shift in the composition oflabour towards sectors such as services,which are rela‑tively labour-intensive and use more part-time workers.Theproportionofpart-timersincreasedconsiderablyafterthe2008 crisis,andcontinuedrisinguntil2013 when itlevelled out. It was driven up by involuntary part-timework and flexible employment contracts against thebackground of the poor and uncertain demand outlook. Thus,thenumberofhoursworkedperworkerfellsharplyduringtherecessionyears.

The unemployment rate for the euro area as a wholeedged slightly downwards to an annual average of10.1%,itslowestlevelsincetheendof2011.AlthoughthedivergencesinunemploymentratesbetweenMemberStateshavediminished,theyremainsubstantial,ascondi‑tions remaindifficult in the countrieshardesthit by thecrisis.Forinstance,inGreeceandSpain,the 2016unem‑ploymentrateswerestillmuchhigherthanin2007.Thereiscontinuingunder-employmentonthelabourmarketinmostothereuroarea countries. That isduenotonly totheunemployedbutalsotopeoplewhoareabsentfromthe labourmarket because of the limited job prospectsbut might enter the market if conditions improved. Itseems,though,thattheupwardtrendintheparticipationrate evidentbefore the crisis has recently been restoredto someextent.At theendof 2016, the labourmarket

participation rate for the population of working agereached73%.

Further correction of the macroeconomic imbalances in the euro area, but the vulnerabilities have not entirely disappeared

Themostvulnerablecountrieshavemademajorconsoli‑dationeffortssincethestartofthedecade,enablingthemto reinforce the sustainability of developments in theireconomies.Nonetheless,many countries still facedhighexternalorinternaldebtlevels,andtheprogressachievedonthatfrontwasrathermeagrein 2016.

Externally, after the substantial corrections alreadymadeinprecedingyears,theadjustmentofcurrentac‑countbalancescontinuedin 2016,yetitwasasymmet‑ricbetweentheMemberStates.Theeuroarea’scurrentaccountsurpluswiththerestoftheworld,whichhadbeenmarginalin2010,thusincreasedrapidlyto3.3%ofGDPin 2015and3.7%in 2016.Apartfromthere‑ductioninthedeficitsrelatingtonetimportsofpetro‑leumproducts,which benefited virtually all euro areaMember States following the fall in prices, that trendisduetotheeconomicadjustmentsincountrieswhichhadsubstantialdeficitsbeforethecrisis,combinedwiththe persistent or even increasing sizeable surpluses in other countries.

CHART 7 LABOURMARKETSITUATIONIMPROVES,BUTTHEREISSTILLSCOPEFORABSORBINGUNDER-EMPLOYMENT

2007

2016

2008 2009 2010 2011 2012 2013 2014 2015 201695

96

97

98

99

100

101

J

JJ

J

J

JJ

J

J

JJ

J

DE AT NL BE IE FI FR PT IT ES EL EA

0

5

10

15

20

25

30

Hours worked per person

Employment in persons

UNEMPLOYMENT RATE IN SEVERAL EURO AREA COUNTRIES(in % of the labour force, annual averages)

EMPLOYMENT IN THE EURO AREA(indices first quarter of 2008 = 100)

Peak since 2007

source : Ec.


Inthecountriesthatinitiallyhadadeficit,theadjustmentsresulted both from a fall in their domestic demand – and hencetheirimports–andstrongerexportgrowththankstomajoreffortsmade inpastyears to restore theircostcompetitiveness. Some formerly very large current ac‑countdeficitswerethusgraduallyabsorbed,givingwaytoabalancedaccount inGreeceandsurpluses inPortugal,Spain and Ireland. Nevertheless, taking account of theaccumulation of substantial deficits in the past, leadingto increased financial liabilities towards the rest of theworld, the total net international investment positionat mid-2016 was still very negative in Ireland (–190%ofGDP),Greece (–133%ofGDP), Portugal (–106%ofGDP)andtoalesserdegreeSpain(–88%ofGDP).Thosecountries must therefore continue to achieve positive cur‑rent account balances in the future so as to bring their net external liabilities down to more sustainable levels.The environment of modest growth and low inflationmakes it harder to reduce those liabilities, and explainstherelativelystaticoutstandingpositions,despitepositivedevelopmentsintermsofflows.

Theadjustmentonthepartofthedebtorcountrieswasnot accompanied by an adjustment in the opposite di‑rection by the creditor countries. In 2016, large currentaccount surpluses persisted, or even expanded. In the

Netherlands,thesurplushasstabilisedsince 2014–dueto,amongstotherfactors,areductioninitsgasproductssurplus following cuts in domestic output –, but it stillremained substantial at 8.5%ofGDP. InGermany, thesurplus continued to grow, reaching a record 9% ofGDP in 2016.Apartfromoil-balance-relatedeffects, thehigh and persistent surpluses point to excessive savingscompared to investment, accentuated by the continueddeleveraging in those economies. More generally, theincrease in the surpluses is a symptomof thepersistentweaknessofaggregatedemand,reflectedsince2009 ina negative output gap in the euro area in general, andweakpotentialgrowth,inGermanyamongstothers.Thisunder-utilisationofresourcesisafactorintheparticularlylow core inflation, a situationwhich has hampered theeffortsofthedeficitcountries.

In parallelwith themovement in the net external posi‑tion,theprivatesectorcontinuedtoreduceitsdebtlevels,thoughprogresswas slowandpatchy. Thedebt dimin‑ishednotonlyinsomecountrieswherethelevelhadriseninthepre-crisisyears,butalsoinotherswhereitismuchless of a problem.

Inregardtohouseholds,thereductionsinthedebt/GDPratiouptothesecondquarterof 2016werebiggestinthecountrieswiththehighestdebtburdenattheoutbreakofthe sovereigndebt crisis. Thatwasprimarily the case inIreland,where–bymid-2016–thedebtwascutbyalmosthalfinrelationtoitsmaximumlevelthen(over100%ofGDP), but also to a lesser extent in Spain and Portugal(cutbylessthanaquarter)andtoasmallerdegreeintheNetherlands. Some countrieswhere thehouseholddebtratiowasalreadybelowtheaveragealsorecordedaslightfall,e.g.Austria, ItalyandGermany. Inthecaseofnon-financialcorporations,thepicturewasmuchmorevariedin the countries which had the highest (consolidated)debt/GDP ratio in 2012. While all those countries re‑cordedadownwardtrend,someofthemmaintainedthatcorrectionuntil 2016,suchasSpain(tojustoveraquarterof the2012 level)andPortugal (toover10%),while incontrast, others such as Belgium and the Netherlandsreversedthattrendin2013 or 2014.Nevertheless,inthelattercountries,substantialcapitalflowsandintra-grouploanscomeintoplay,duetotheoperatingandfinancingmethodsofthemanymultinationals locatedintheirter‑ritory. Inothercountrieswherethedebtratiowascloseto,orevenbelow,theeuroareaaverage,thechangesincorporatedebtlevelswereminimal.

Despitethisnewreduction,thedebtlevelremainedhighin some cases in 2016.Although it is difficult todefineanequilibriumvalue that thedebt level should tend to‑wards,somecountrieshaveprivatedebtlevelswhichare

CHART 8 THEEUROAREA’SCURRENTACCOUNTSURPLUSONTHEBALANCEOFPAYMENTSHASINCREASED

(in€billion)

2010 2011 2012 2013 2014 2015–300

–200

–100

0

100

200

300

400

500

DE

NL

AT BE FI

FR

IT

ES PT IE

EL

2016

p.m. Euro area

Other euro area countries (1)

source : Ec.(1) Cyprus,Estonia,Latvia,Lithuania,Luxembourg,Malta,SlovakiaandSlovenia.


stillsubstantialfromtheirownhistoricalperspective.Thedownward trend should continue, and could thereforedepresseconomicactivityintheshortterm.Finally,thosecountries sometimes also have a high outstanding public debtwhich,overall,makestheireconomyveryvulnerableto shocks.

Fiscal policy was slightly expansionary

The general government deficit in the euro area as awhole continued to fall, from2.1%ofGDP in 2015 to1.8%ofGDP in 2016. In addition, thepublic debt de‑clinedforthesecondsuccessiveyear,tosettleataround90%ofGDP.Theimprovementinthebudgetbalanceisduemainlytothecontinuingeconomicrecovery,butalsotothelowerinterestchargesandone-offfactors.Evenso,thefiscalpolicystancewasslightlyexpansionary;thatledtoafallinthestructuralprimarysurplus,largelyoffsettingthe said positive factors.

The lower interest charges on the public debt had afavourable impact on public finances in all euro areacountries. The economic climate likewise had a benefi‑cial influence,except inafewsmallMemberStatesandin Belgium. In 2015, Greece, Portugal and Ireland hadconducted one‑off operations for the purpose of bank recapitalisation, thus placing a heavy burden on theirpublicfinances. In 2016, thatceased tohavean impact

on the budget, so that one-off factorsmade a positivecontribution, on average, to the improvement in thebudgetbalance. Judgingby themovement in thestruc‑turalprimarybalance,manyeuroareacountries–includ‑ingSpain,Austria,Italy,BelgiumandGermany–pursuedanexpansionaryfiscalpolicytoalargerorlesserdegree.Thatwas often accompanied by tax cuts. In Spain, de‑spitetheverypositiveinfluenceoftheeconomicclimateandthefallininterestcharges,thestrongfiscalstimulusmeantonlyasmallreductioninthebudgetdeficit,whichstillcameto4.6%ofGDP.InItaly,too,thepublicdeficitcontractedonly slightly,while inAustria andBelgium itactuallygrewbigger.

According to the EC’s autumn forecasts, apart fromSpain,Francealsostillhadabudgetdeficitinexcessofthereferencevalueof3%ofGDP.Thesetwocountries,likeGreeceandPortugal,arestillsubjecttoanexcessivedeficitprocedureunderthecorrectivearmoftheStabilityandGrowthPact(SGP).InNovember 2016,afterhavingexaminedwhetherSpainandPortugalhadtakeneffec‑tive action to complywith the Council’s recommenda‑tions on the subject, the EC felt that the procedureshouldbekeptinabeyanceinthosetwocountries.TheprocedureforFrancewasadjournedinJuly 2015.InthecaseofIreland,CyprusandSlovenia,theexcessivedeficitprocedure was terminated in June 2016. EU MemberStates also have to respect a medium-term objective

CHART 9 DELEVERAGINGCONTINUED,BUTTHEPACEWASSLOWANDUNEVENBETWEENCOUNTRIES

NL FI DE IE AT FR ES BE PT IT EL EA0

20

40

60

80

100

120

140

160

180

200

2016

DE IT EL AT FR ES FI BE PT NL IE EA0

50

100

150

200

250

300

350

400

450

500

Households (2016 Q2)

Non-financial corporations (2016 Q2)

Peak level of total private debt (1)

DEBT OF THE NON-FINANCIAL PRIVATE SECTOR(in % of GDP)

PUBLIC DEBT(in % of GDP)

Peak (2)

Sources:ECB,EC.(1) Since1999 orthefirstavailabledata,dependingonthecountry.(2) Since1999.


under the preventive arm of the sGP. This concerns a referencevalueforthebudgetbalancespecifictoeachcountryandexpressedinstructuralterms.Thebalancesestimated by the EC for 2016 indicate that, in someMember States such as Austria, the structural budgetbalanceisapproachingthemedium-termobjective,andin others such asGermany that goal has already beensurpassed.Conversely,manyeuroareacountries,includ‑ingBelgiumandItaly,stillneedtomakeaconsiderableefforttoattainthatobjective.

The fragility of the euro area’s banking sector is impeding the growth revival

Intheeuroarea,thebankingsectorisanessentialsourceoffundingforprivatesectoractivity.However,thebankswereseriouslyaffectedby thecrisesof2008 and2012.Sincethen,greatprogresshasbeenmadeinstrengthen‑ingtheirbalancesheet,inparticularasregardsreinforcingtheir capital ratios following the comprehensive assess‑ment that preceded the introduction of the single super‑visorymechanism(SSM)inNovember 2014andthephas‑ing in of the new prudential requirements under BaselIII. Increased resilience in termsof solvencyand liquidityand an accommodativemonetary policy havehelped to

restorethebankcreditchannel.Withtheadditionalsup‑portofthedemandrevival,thegrowthofbanklendingtotheprivatesectortookoffagainrecently,butremainsveryfragileinsomecountries.Yetcapitalisationlevelsstillvarywidely betweenMember States,with the lowest Tier 1capital ratios in mid-2016 being recorded in Portugal,Italy,LatviaandSpain.

The problems still confronting the banking sector caused episodesofnervousnesson thefinancialmarketsat theendof 2015andduring 2016.Thechallengesessentiallyconcern theweakoutlook for the returnoncapitalandthe persistence of large outstanding amounts of non‑performingloansinsomecountries,alegacyofthecrisiswhichcontinuestoencumberbanks’balancesheetsandcouldfurtherimpairtheirprofitability.

While the low level of interest rates attenuated thecyclical slowdown and supported economic activity,it tended to be detrimental to the profitability of thefinancial sector. In such an environment, the accumu‑lated assets could only generate ameagre return, andthatwas sometimes combinedwith the persistence ofoperationalandstrategicmodelswhichwereoutdated,inappropriateorcostly.Thismodestprofitabilitywhich,

CHART 10 THEFURTHERREDUCTIONOFGOVERNMENTBUDGETDEFICITSISACCOMPANIEDBYASLIGHTEASINGOFFISCALPOLICY

EL PT NL IE ES FI IT FR DE ATBE EA

–2

–1

0

1

2

3

4

5

6

DE NL IE AT FI IT EL PT BE FR ES EA–5

–4

–3

–2

–1

0

1

Structural primary balance

Cyclical component

Interest charges

One-off factors

Budget balance

BUDGET BALANCES IN 2016(in % of GDP)

COMPONENTS OF THE BUDGET BALANCES(change in 2016 compared to 2015, percentage points of GDP)

source : Ec.


inmid-2016,was lowest inGreece,Portugal,GermanyandItaly,hadanimpactonthevaluationofthebanks’capital,affectingtheirabilitytoraisenewfundsonthecapitalmarkets;ultimately,thatthreatenedtolimittheirscopeforlendingtotheprivatesectorortoimplyasub-optimum allocation of the capital.

Inadditiontotheseissues,thereareproblemsconcerningassetqualityandcreditrisk.Thebalancesheetsofsomebanks still often have massive outstanding amounts of non‑performing loans as a result of past crises. ParticularlyhighratioswereseeninCyprusandGreeceand to a lesser degree in Italy, Ireland and Portugal.In some of those countries, the substantial volume ofnon-performingloanswascombinedwithrelativelylowcapital ratios. A problematic level of private sector debt can cause additional vulnerabilities, owing to the pos‑sibleriskofdefaultthathastobeadequatelycoveredbyprovisions. Although bank balance sheets are still being cleaned up, various factors prevent that type of loanfrombeingliquidatedmorequicklyandefficiently,suchas bankruptcy legislation and procedures, or practicesconcerning loan loss provisions which vary greatly be‑tweenMemberStates.Anin-depthexaminationofthisissueandtheinitiativestakenrecentlytoresolveitisde‑scribedindetailinthepartofthisReporton“Prudentialregulationandsupervision”.Finally,sincebanksinmanycountries still have a very substantial exposure to debtinstruments issued by national governments, renewedpoliticalturmoilcouldfurtherexacerbatethesaidweak‑nesses in the banking sector.

On the road to completion of Economic and Monetary Union

Despite the progress achieved since the sovereign debt crisis in the euro area – particularly the establishmentof the banking union and the improvement in eco‑nomicgovernance– theEconomicandMonetaryUnion(EMU) isstill imperfect.ThatpromptedthePresidentsofthe European Commission, the European Council, theEurogroup,theEuropeanCentralBankandtheEuropeanParliament topublish a report in June 2015 settingouttheir plans for strengthening EMu. The intention is to finaliseEMUbynolaterthan2025.InOctober 2015,inaccordancewiththatreport,theEChadalreadyapprovedaninitialsetofmeasuresandmaderecommendations.InSeptember 2016,inlinewiththatpackageofmeasures,theEcofinCouncilcalledontheeuroareacountriestosetupnationalproductivityboards.Thoseboardsaretoex‑aminedevelopmentsinproductivityandcompetitiveness,and conduct an independent analysis of the challengesthattheypresentforpolicy.TheMemberStatesandtheECcanusethatindependentexpertiseintheannualmon‑itoringofeconomicpolicyinthecontextoftheEuropeanSemester.Asregardsfiscalpolicy,atitsDecembermeet‑ing, theEcofinCouncildecidedtogiveagreaterroletotheexpenditureruleindeterminingandassessingtheSGPobjectives.Nevertheless,thestructuralbudgetbalancere‑mainsanessentialelementofthebudgetaryframework.Also,attheNovember 2016launchoftheannualcycleoftheEuropeanSemesterofeconomicpolicycoordination,theECadvocatedafiscalpolicystanceaimedmoreatsup‑portingtherecoveryoftheeuroareaasawhole. Inad‑dition,inOctober 2016,theECappointedtheChairmanand the four members of the European fiscal Board set upinOctober 2015.

During the year under review, thanks to the progressachieved, thefinancial union –oneof the cornerstonesof the Five Presidents’ Report –was consolidated as re‑gardstheEuropeanCapitalMarketsUnion(CMU).Sincethe 2015 action plan,which defined the prioritymeas‑urestoenabletheCMUtotakeeffectby 2019,thefirstfollow-upreportwasproducedinApril 2016.Tofacilitatethe development of new types of funding, the EC alsoproposedrenewingthelegislativeframeworkconcerningventurecapitalbyamending, inJuly,theRegulationsonEuropean venture capital funds (EuVECA) andEuropeansocialentrepreneurshipfunds(EuSEF),bothofwhichaimto support young, innovative businesses and firms thatwishtomakeapositivesocialimpact.ThosechangeswereendorsedbytheCouncilattheendoftheyear. Inaddi‑tion,thesecuritisationmarketwillbestrengthenedbytheproposal submitted in September 2015 for establishinganewregulatoryframeworkforsimple,transparentand

CHART 11 FRAGILITIESPERSISTINTHEEUROAREA’SBANKINGSECTOR

(non-performingloans,in%oftotalloansandadvances(1))

CY EL IT IE PT MT SI ES LT SK FR LV BE AT NL DE EE LU0

10

20

30

40

50

2016 Q2 (or latest available data)

Euro area

Source:IMF.(1) NodataavailableforFinland.Comparabilityofthedataacrosscountriesmaybe

limitedduetotheuseofdifferentdefinitionsandconsolidationperimetersfornon‑performing loans in domestic banking sectors.


standardised(STS)securitisationoperations(SecuritisationRegulation). Finally, in November 2016, in order to re‑movenationalbarrierstocapitalmarket integration,theEc tabled a proposal on business restructuring and in‑solvencywiththeaimofspeedingupassetrecoveryandoffering a second chance to entrepreneurs suffering their firstbankruptcy.

1.3 TheEurosystem’slowinterestratepolicyisintendedtosupporttherecoveryintheeuroarea

Increased risks to price stability required supplementary measures

Inrecentyears,theECBGoverningCouncilhasadoptedawiderangeofstimulusmeasures.Thattookplaceagainstthe backdrop of persistently low inflation in the euroarea:sincemid-2013, inflationhasbeen lowerthanthelevelcompatiblewithpricestability,i.e.aratebelow,butcloseto,2%.Moreover,theeconomicrecoveryremainedweak: after eight years, thegap in relation topotentialoutputhasstillnotbeenclosed,sothatunderusedpro‑ductioncapacityisexertingdownwardpressureonprices.

The Governing Council began by further reducing thekey interest rates, and in June 2014 the deposit facilityrate was cut below zero for the first time. In view ofthe lack of scope for reducing short‑term interest rates anyfurther, itthenextendeditsrangeof instrumentstocontinue steering the economy’s financing conditions.One of those exceptional but crucialmeasureswas theJanuary 2015decisiontolaunchanexpandedassetpur‑chaseprogramme(APP).Thatmeasuresupplementedtheexisting programmes for the purchase of covered bankbonds and asset-backed securities by adding a massiveprogramme for purchases of government bonds. The APP wasoriginallyintendedtorununtilSeptember 2016,butitwasemphasised that itwould certainly continueuntiltheGoverningCouncilsawasustainedadjustmentintheinflationpathconsistentwith itspricestabilityobjective.The Governing council thus added a further dimension to theforwardguidancewhich ithadadopted inthesum‑merof2013. lnfact, it indicatedthat itwouldcontinuetopursueanexpansionarymonetarypolicyuntilitspricestabilitymandatewasfulfilled.

However, unexpected financial and economic eventsexerting downward pressure on economic activity andinflationnecessitatedfurthermonetaryeasingandhenceadjustmentoftheAPP.InDecember 2015,theGoverningCouncilthusdecided,inthelightofthedownwardrevi‑sionoftheinflationforecast,tocutthedepositfacilityrate

from–0.20 to–0.30%andtoextendtheplannedperiodforassetpurchasesfromtheendofSeptember 2016tothe end ofMarch 2017, or even longer if necessary. Itlikewise decided to extend the list of assets eligible fortheAPPtoincludesecuritiesissuedbyregionalandlocalgovernments in theeuroarea, and to reinvest theprin‑cipalrepaidonmaturingbondsforaslongasnecessary.Finally, the Governing Council also stressed that it wasreadytotakeadditionalmeasuresinthefuture.Insodo‑ing, itemphasised that theAPPwasflexible so that thesize,compositionanddurationof theprogrammecouldbemodified.

InMarch 2016,theGoverningCouncilonceagaindem‑onstrateditswillingnesstoact.Atthebeginningofthatyeartheoutlookforworldgrowthdeteriorated,financialmarketvolatility increasedand inflationexpectationsbe‑ganfallingagain.TheinflationforecastsproducedbytheEurosystem’sstaffpointedtoheightenedriskstopricesta‑bility;theMarch 2016projectionsassumedinflationratesofonly0.1%and1.3% in 2016and 2017 respectively(inDecember 2015therespectiveforecastrateswerestill1%and1.6%),whereasfor 2018theyforesawinflationat1.6%,a levelwhich isadmittedlybelow,butstillnotcloseto,2%.

CHART 12 THERENEWEDFALLININFLATIONEXPECTATIONSATTHEBEGINNINGOF 2016REQUIREDMONETARYEASING

H

H

H HH H

H

H

H

HH

HH

H

HH

H

HHH

H

2012 2013 2014 2015 2016

Derived from financial data (1)

(left-hand scale)

0.8

1.3

1.8

2.3

2.8

1.6

1.7

1.8

1.9

2.0

Based on survey data (2)

(right-hand scale)

March meeting of the Governing Council

LONG-TERM INFLATION EXPECTATIONS(in %)

Sources:ECB,Bloomberg.(1) Implicitinflationratederivedfromswapscoveringtheinflationriskinthe

euroareaforafive-yearperiodcommencingfiveyearsafterconclusionofthecontract.

(2) Averageoftheaggregateprobabilitydistributionofinflationexpectationsinfiveyears.DataobtainedfromtheECB’squarterlysurveyofprofessionalforecasters.


Atitsmeetingon10 March 2016,theGoverningCounciltherefore announced a comprehensive package of newmeasures. As regards the APP, the monthly asset pur‑chaseswere increased from €60 billion to €80 billion.In addition, itwas decided that from June 2016 bondsissued by non-financial corporations(1) would also bepurchased under the APP (corporate sector purchase programme – CSPP). Moreover, the Governing Councildecided to make further cuts to its key interest rates:the rates on themarginal lending facility and themainrefinancing operations were reduced by 5 basis points,to0.25%and0%respectively,whilethedepositfacilityratewascutby10 basispoints to–0.40%. Inaddition,the Governing council announced that it intended to holdthekeyinterestratesatorbelowtheirpresentlevelsforanextendedperiodoftime,wellpastthehorizonforthe net asset purchases,which it had confirmedwouldcontinueuntilMarch 2017orbeyond,ifnecessary.Finally,inMarch 2016, itwas alsodecided to launch a secondseriesoffourtargetedlonger-termrefinancingoperations(TLTROII),withatermoffouryearsandafixed interest

rate. Those operations provide cheap and stable funding for the banks in return for more lending to the private sector.Byanchoringtheinterestrateontheselong-termoperations,theGoverningCouncilalsoindicatesthatthekey interest rateswill remain low for a long time, thusfurtherreinforcingitsforwardguidance.

All these measures are designed to hold down all theinterestratesthatarerelevantintheeconomy.Themorefavourableborrowingconditionsaremeanttoencouragehouseholds,businessesandpublicauthoritiestoconsumeandinvest,soastotakefulladvantageoftheeconomy’sproductionpotentialanddriveupinflation.

Practical implementation of the monetary policy measures

At theendof 2016, the securitiespurchasedunder theAPPamountedtoaround€1500 billion,accountingforapproximately 40% of the Eurosystem’s balance sheettotal.Governmentbondsmadeupthebulkofthat,with€1259 billion.Thebalancesheetalsocontainedcoveredbank bonds amounting to €204 billion, bonds issuedby non-financial corporations totalling €51 billion, and

(1) Morespecifically,thisconcernshigh-quality(investment-grade)euro-denominatedbondsissuedbynon-bankbusinessesestablishedintheeuroarea.

CHART 13 WITHTHEKEYINTERESTRATESCLOSETOTHEFLOOR,THEEUROSYSTEM’SBALANCESHEETMEASURESAREVERYIMPORTANTFORPROVIDINGAMONETARYSTIMULUS

20121 500

1 750

2 000

2 250

2 500

2 750

3 000

3 250

3 500

3 750

2013 2014 201520162012 2013 2014 2015 2016

EUROSYSTEM CONSOLIDATED BALANCE SHEET(in € billion, assets)

–0.5

0.0

0.5

1.0

1.5

2.0

KEY INTEREST RATES AND OVERNIGHT MONEY MARKET RATES(in %)

Deposit facility rate

Marginal lending facility rate

Main refinancing rate

Eonia rate

Other balance sheet items

Covered bank bonds

Asset-backed securities

Government bonds

TLTRO

APP

Bonds issued by non-financial corporations

Sources:ECB,ThomsonReutersDatastream.


asset-backedsecuritiesamountingto€23 billion.Thetar‑getpurchasevolumesweremet,whiletheAPP’sflexibilitylimitedthepotentialmarketdisruptionthatcouldresult,forexample,fromashortageofsecureassetsonthepri‑vatemarket.Thus,duringthesummer,takingaccountoftheseasonalreductioninfinancialmarketliquidity,small‑ervolumeswerepurchased,andthatwasoffsetinothermonths.MostoftheEurosystemcentralbanksalsomadearrangements for lending securities,whereby the assetsheldundertheAPP–particularlygovernmentbonds–areloaned to market participants. Those securities are in fact highlyprized,partlybecausetheyconstitutecollateralforrepo transactions.

As regards the TLTRO II, demand for liquidity at thetime of the three operations in June, September andDecembertotalled€399 billion,€45 billionand€62 bil‑lionrespectively.However,thenetliquidityinjectedintothebankingsystem(€115 billion)waslessthanthesumofthoseamounts.Infact,atthetimeofthethreeopera‑tions,butmainlythefirstone,thebanksrepaidpartoftheTLTROIfundsandborrowedthemoneybackintheformof TLTRO II funds. Three characteristicsmake thelatterfinanciallymoreattractive.First,theyarecheaper:under the TLTROs, the banks pay a fixed interest ratecorresponding to the rateon themain refinancingop‑erationsprevailingatthetimeoftheoperation.Inviewof thecut in thekey interest rates inMarch 2016, thenewTLTROsarethereforecheaperthanthoseconcludedduring thefirst seriesofoperations.What ismore, thecostoffinancing theTLTRO II could fallex post to the levelofthedepositfacilityrate,providedthatthebanksgrant sufficient credit to theprivate sector(1).Next, theTLTROIIoffersamorestablesourceoffunding:incon‑trast towhathappenedwith thefirst series, thebanksare not obliged to effect early repayment of the totalamountborrowediftheygrantinsufficientcredit.Finally,the longer term of the TLTRO II alsomakes the seriesmoreattractivethantheTLTROIforwhichtheduedate(September 2018)isapproaching.

InfluencedbytheAPPandtheTLTROs,theconsolidatedbalance sheet of the Eurosystem continued to expand,exceeding its mid-2012 peak. In accordance with thegrowthof thecentralbank’sbalancesheet, the liquiditythateuroareacreditinstitutionsholdwiththeEurosystemalsoincreases.Thus,attheendof 2016,theexcessliquid‑ity–i.e.thereservesthatcreditinstitutionsholdwiththecentral bank over and above their reserve requirement,eitherontheircurrentaccountorviathedepositfacility–cametoaround€1200 billion,comparedto€655 billion

at thebeginningof the year. Sincebanks endeavour toplace theirexcess reservesonthemarket, theabundantsurplus liquidity stabilised the overnight money marketrate– theEonia–ata level close to thedeposit facilityrate.Throughout 2016,theovernightmoneymarketratethereforeremainednegative,sothatinterbankfinancingcostswereatunprecedentedlylowlevels.

The measures produced results : lower nominal interest rates and improved bank lending

Themonetarypolicymeasuresresultedinsignificanteas‑ing on many financial market segments. Thus, the for‑wardguidanceandthesignalgivenbytheAPPregardingthemonetarypolicy stance–namely that thekey inter‑est rateswill remain low for an extended period – alsobrought about a further fall in the longer‑term risk‑free interestrates,leadingtoflatteningoftheyieldcurve.

Similarly, interest rates on government securities fellfurther in most euro area countries in 2016, decliningtoall-time lows. Itwas the longest-dated securities thatrecorded the largest falls. Apart from the higher demand resultingfromthepurchasesundertheAPP,otherfactorsalsodrovedownyieldsongovernmentbonds,particularlythose of the strongest sovereign issuers. The nervous‑nesson thefinancialmarkets, triggeredpartlybyBrexit,andtheneedtoprovidegood-qualitycollateral for repotransactionsboosteddemand for this typeof securities,andparticularlyGermangovernmentbonds.Therelativelylimited outstanding volume of government securities is‑sued by some largeMember States combinedwith thecharacteristicsoftheAPP,whichdistributesthepurchasesof government bonds among the euro area countries according to the ECB’s capital key, also led to a biggerdecline in interest rates on certain government bonds. As the capital key reflects the size of a country’s economyandpopulation,aconsiderableproportionofthepurchas‑es concerned the German Bund in particular. However,the outstanding amount of German government bonds andtheprospectsforissuanceofthosesecuritiesarefairlylimited,sotheAPPexertsevenstrongerdownwardpres‑sure on the interest rates on German securities than on thoseofcountrieswitharelativelylargeoutstandingvol‑umeofgovernmentpaper,butwitharathersmallshareinthepurchasesundertheAPP(seealsobox1).Finally,theEurosystem’soriginaldecisionnottopurchasesecuri‑tiesundertheAPPofferinganinterestratelowerthanthedeposit facility rate also reduced the range of securitiesactuallyeligiblefortheAPP.ThatcertainlyappliestotheGerman Bund, as the short- and medium-term interestratesonthosebondsweregenerallybelowthatthresholdin 2016.Tomeetitstargetforpurchases,theEurosystemtherefore bought government securities with longer

(1) FormoreinformationonTLTROII,seehttps://www.ecb.europa.eu/press/pr/date/2016/html/pr160310_1.en.html.


maturities. Consequently, the longer-term interest ratesofstrongeuroareacountriesalsodeclined,approachingthe deposit facility rate, or even falling below it,whichsometimespresentedchallengesfortheEurosystem’sim‑plementationofthe APP.

Although these factors mainly drove down the bondyieldsof strong issuers, the spreadson thegovernmentbondsofmosteuroareacountriesinrelationtotheBund,forexample,werefairlystableafternarrowingsharplypri‑ortotheprogramme’slaunchandatitsstart.Thatcouldindicate the active operation of the portfolio rebalancing channeloftheAPP:byrearrangingtheirportfolioofeuroarea government bonds to the detriment of securities pur‑chasedona largescalebytheEurosystembut infavourofthoseofferinghigheryields,investorsalsodrovedowninterest rates on the latter securities.

Apart from thebanks andpublic authorities, businessesalso benefited from the lower financing costs. TheAPPpurchasesofbonds issuedbynon-financial corporationsled to a considerable decline in the interest rates on those securities. Yields on bonds issued by both financial andnon-financialcorporationsnoteligibleforpurchaseunderthe CSPP also declined, once again indicating portfoliorebalancing in thewake of the APP. At the end of theyear,while those interest ratesbegan to creepupwardsagaintheyremainedbelowthelevelprevailingatthebe‑ginningoftheyear.Itispossiblethatthemorefavourablefinancingconditions in 2016encouragedthefundingofbusinessesviathemarket,astherewasinfactanincreasein net issuance of private sector bonds.

Thebankingsector–whichperformsakeyroleinmon‑etary transmission within the euro area – continued topassonthemonetarypolicystimulustotherealeconomy.The monetary policy measures thus further supportedlending to the private sector. On the one hand, bankfunding costs for households and businesses continued to fall. Debit interest rates on bank loans to businesses in the various euro area countries therefore converged again tosomeextent,reflectingafurtherreductioninfinancialfragmentation.Also,thepositivedynamicsofthevolumeofbanklendingtotheprivatesectorpersisted,butinthevulnerableMemberStatesotherthanItaly,thegrowthofcredit remained negative.

Thebanklendingsurvey(BLS)revealsthatboththecreditsupplyanddemandforcreditcontributedtothestrongercreditexpansionintheeuroarea.Loancriteriaforbusiness‑esandhouseholdswereeased,usuallyasaresultofkeenercompetitionandattenuationofthebanks’riskperception.ThebanksalsoannouncedthattheliquiditygeneratedbytheAPPwouldbeusedinthefirstinstancetoexpandtheir

CHART 14 NOMINALINTERESTRATESFELLSHARPLY

(in%)

–1

0

1

2

3

4

201620152014

–1

0

1

2

3

4

201620152014

–0.6

–0.3

0.0

0.3

0.6

0.9

1.2

YIELD ON GOVERNMENT BONDS

YIELDS ON EURO-DENOMINATED PRIVATE SECTOR BONDS

RISK-FREE YIELD CURVE(OIS ; average over the period)

(1)

2015

January 2016

2014 March 2016

October 2016

Deposit facility rate (level since March 2016)

December 2016

7-year German bonds2-year Spanish bonds2-year German bonds

7-year Spanish bondsDeposit facility rate

Financial corporations BBB

Non-financial corporations AAA

Financial corporations AAA

Non-financial corporations BBB

Deposit facility rate

1y 2y 3y 4y 5y 6y 7y 8y 9y 10y

Sources:BarclaysCapital,ThomsonReutersDatastream.(1) InterestratesonEoniaswapswithdifferentmaturities.


lending.Inaddition,theyindicatedthatthenegativeinter‑estrateonthedepositfacilityhadapositiveinfluenceonthecreditvolume,althoughitcouldatthesametimeputpressure on their interestmargin. That said, demand forloans on the part of businesses and households also con‑tinuedtoriseasaresultofthelowinterestrates,mergersandacquisitions,arevivalinconsumerconfidence,aswellas the improved outlook on the housing market.

The recovery of credit growth and the implementationof the APP – so long as the asset seller is a non‑bank resident of the euro area (1)–engenderedastrongexpan‑sionofthemoneysupply.ThebroadmonetaryaggregateM3 recordedrelativelystableyear-on-yeargrowthof5%

throughout 2016.Thedrivingforceherewastheexpan‑sion of themost liquid components ofM3, particularlysightdeposits (M1).The lowandflatprofileoftheyieldcurveinfactimpliesthattheopportunitycostsofholdingthemostliquidM3 assetsareminimal.

The favourable funding conditions in turn help to restore economic activity, inflation and inflation expectations.As a result, the monetary policy measures slowed thedownwardtrendininflationexpectationsandsupporteddomestic demand in the euro area.

CHART 15 BANKFINANCINGOFTHEPRIVATESECTORINTHEEUROAREACONTINUESTOIMPROVE(1)

2012 2013 2014 2015 2016–6

–4

–2

0

2

4

6

2012 2013 2014 2015 2016

BANK LOANS (2) TO THE NON-FINANCIAL PRIVATE SECTOR

AND M3(year-on-year percentage change)

1.5

2.0

2.5

3.0

3.5

4.0

4.5

DEBIT INTEREST RATES ON BANK LOANS TO HOUSEHOLDS (dotted line) AND NON-FINANCIAL CORPORATIONS (solid line) (in %)

M3Euro area Vulnerable countries Other countries

source : EcB.(1) ThevulnerablecountriesareCyprus,Spain,Greece,Ireland,ItalyandPortugal.TheothercountriesaretheremainingeuroareaMemberStates.Thetwoseriesofdata

concernaGDP-weightedaverage.(2) Takingallmaturitiestogether.Thedataareadjustedforsecuritisation.

Box1 –AssetpurchaseprogrammesintheUnitedStatesandintheeuroarea:impactongovernmentpaperheldbytheprivatesector

Centralbankassetpurchaseprogrammescandepresslong-terminterestratesviavariouschannels.Forinstance,bypurchasingcertain long-termsecuritiesona substantial scale, thecentralbank reduces the supplyof thosesecuritiesontheprivatemarketsothattheirpricerisesandtheiryielddeclines.Certaininvestorsprefertoholdthesesecuritiesandarethereforepreparedtopayahigherpricetoacquirethese“scarce”assets.Inaddition,thecentralbankcanreducetheaggregatelevelofthedurationrisk(i.e.theexposuretounexpectedchangesinthekeyinterestrate)intheprivatesectorportfoliobytakinglong-term–andhenceriskier–securitiesoffthemarket,

4

(1) Formoreinformation,seebox2intheReport2015.


4

replacing themwith safe, short-termsecurities (centralbank reserves).As the totalduration risk in themarketdiminishes, private investors should demand less compensation for that risk, thus lowering the termpremiumincluded in the return on long‑term securities.

Ofcourse,purchasesby thecentralbankarenot theonly factoraffecting thesupplyofsecuritiesavailableontheprivatemarket:debtsecurityissuersalsoplayanimportantrole.Ascentralbankpurchasesareconcentratedmainlyongovernmentpaper–onwhichtheyieldisanimportantbenchmarkforfixingthepriceofawiderangeoffinancialassets–theyindicateasubtleinteractionbetweenmonetaryandfiscalpolicy.Whilethecentralbanktendstoputdownwardpressureonlonger-termsovereignbondyieldsviaitspurchasepolicy,governmentscanpartlyoffsetthateffectbyissuingmoreofthosedebtsecurities,e.g.forthespecificpurposeoftakingadvantageofthelowinterestrates.ThisboxaimstoexaminetheextenttowhichmonetaryandfiscalpolicyhavechangedtheavailabilityofgovernmentpaperontheprivatemarketintheUnitedStatesandintheeuroarea.

Between December 2008 and October 2014, the Federal Reserve launched three programmes of large-scaleassetpurchases (LSAPs)(1) and thus removed from the private market long‑term debt securities totalling around $3900 billion. The purchases comprised US government bonds and debt securities issued or guaranteed bygovernment-sponsoredenterprises(GSEs).Thus,attheendof 2014,theFed’sportfolioofsecuritieshadexpandedby22%ofGDPcomparedtothethirdquarterof2008,whennopurchaseprogrammeshadyetbeenintroduced.Over the same period, the US Treasury issued additional sovereign bonds for the equivalent of around 50%ofGDP,whilethenet issuanceofsecuritiesbytheGSEsremainedratherflat.DespitetheLSAPs, thesupplyofsovereigninstrumentsontheprivatemarketthereforeincreasedconsiderablyinnetterms.Theadditionalissuanceof sovereign securities thusoffset thedownward effect ofmonetarypolicyon interest rates; from the central

HOLDERSOFGOVERNMENTPAPER(1)

(changesinoutstandingamounts,in%ofGDPin2014,comparedtothequarterprecedingtheannouncementofthefirstpurchaseprogramme)

0

10

20

30

40

50

–12–9–6–30369

12

–12–9–6–30369

12

–12–9–6–30369

12

LSAP1 LSAP2 LSAP3 PSPP PSPP PSPP

2008

Q3

2009

Q3

2010

Q3

2011

Q3

2012

Q3

2013

Q3

2014

Q3

UNITED STATES

2014

Q4

2015

Q1

2015

Q2

2015

Q3

2015

Q4

2016

Q1

2016

Q2

2016

Q3

EURO AREA20

14 Q

4

2015

Q1

2015

Q2

2015

Q3

2015

Q4

2016

Q1

2016

Q2

2016

Q3

2014

Q4

2015

Q1

2015

Q2

2015

Q3

2015

Q4

2016

Q1

2016

Q2

2016

Q3

GERMANY BELGIUM

Public debt purchased by the central bank for monetary policy purposes (b)

Public debt held by the private sector (a – b)

Outstanding public debt (a)

Sources:ECB,BureauofEconomicAnalysis,EC,FederalReserve.(1) Theconceptof“publicdebt”usedinthechartvariesslightlybetweentheUnitedStatesandtheeuroarea.FortheUnitedStates,itcoversdebtsecuritiesissued

bothbytheAmericangovernmentandbytheGSEs.ThesetwoassetcategoriescanbeconsideredclosesubstitutesandwerepurchasedinmoreorlessequalproportionsbytheFederalReserveundertheLSAPs.Intheeuroarea,theoutstandingamountofthe“publicdebt”onlycomprisessecuritiesissuedbynationalgovernments;consequently,thepurchasesbytheEurosystem,representedbytheredlineinthechart,alsoconcernonlythetypeofsecuritiespurchasedunderthePSPP(inotherwords,excludingsecuritiesissuedbysupranationalinstitutions).Thepublicdebtconceptusedinthechartconcernsgovernmentpaperatmarketprices:consequently,itdiffersfromtheMaastrichtTreatydefinitionofthepublicdebt,whichrelatestotheconsolidatedtotalpublicdebtatnominalvalue.Themovementsshownonthechartthereforedonotnecessarilycorrespondtothoseconcerningtheusualconceptsofpublicdebt.

(1) LSAP1 ranfromNovember2008 totheendofMarch2010.LSAP2 –whichwasconfinedtopurchasesofUSgovernmentbonds–ranfromNovember2010 toJune 2011,andLSAP3 ranfromSeptember2012 totheendofOctober 2014.


bank’spointofview,thatisnotnecessarilynegative,asthefiscalstimulusprovidesadditionalsupportfordomesticdemand,whichmaybeusefulwhenthemonetarypolicyinstrumentscomeupagainstlimits,e.g.whenshort-termnominal interest ratesareat their lowerbound.Moreover,a steeperyieldcurve implies lower risks tofinancialstability,andreducesthetransformationriskfacingthecentralbank’sbalancesheetviapurchasesoflong-termsecuritiesfinancedbymeansofshort-termliabilities.

Intheeuroarea,thesituationisdifferent.Inviewofthecurrentratherneutralfiscalstance,relativelyfewsovereignbondsareissued.Theadditionalgovernmentpaperissuedsincetheendof 2014(totallingaround6%ofGDP)wasentirelyabsorbedby thepublic sectorpurchaseprogramme (PSPP), so that the supplyof sovereignbondsfortheprivatesectorremainedvirtuallyunchanged.Theeuroarea’sfiscalpolicy,unlikethatintheUnitedStates,thereforedidnotoffsetthepressureexertedbythecentralbankonlonger-terminterestrates.

However,theimpactofthePSPPonyieldsongovernmentsecuritiesvariesfromoneeuroareacountrytoanother,dependingon the issuance of newdebt. InGermany, the virtual absence of any net issuance of governmentpapercombinedwiththepurchasesunderthePSPPsignificantlyreducedtheavailabilityofthoseinstrumentsontheprivatemarkettobelowtheend-2014level.Drivenbyaflighttosafety,yieldsonGermangovernmentpaperslumpedtoanall-timelow,andactuallybecamenegativeuptoamaturityof13 yearsinthesummerof 2016.InBelgium,theissuanceofgovernmentpaperasapercentageofGDPwasinlinewiththatintheeuroareain 2015,butitincreasedsubstantiallyin 2016.Consequently,thevolumeavailablefortheprivatemarketrecentlyexceededtheend-2014level.

InparallelwiththeimplementationofthePSPP,thebanksreducedtheirexposuretothepublicsector.Inthefinalquarterof 2014,theeuroareabanksstillheld,onaverage,€1826 billionineuroareagovernmentbonds,butbyNovember 2016thatvolumehaddroppedbyalmost10%,anditsshareinthebalancesheettotalwasdownfrom6%to5.5%.Theaccumulationofeuroareagovernmentbondsontheassetssideofthebanks’balancesheetsthereforeseemstobeending.Itisworthnotingthatthisgradualreductionisduemoretoashrinkingportfolio

EUROAREASOVEREIGNBONDSHELDBYBANKS(1) FROM THE EURO AREA (2)

(in€billion)

2006 2008 2010 2012 2014 20160

300

600

900

1 200

1 500

Domestic government bonds

Bonds issued by governments of other Member States

Launch of the PSPP

source : EcB.(1) ThisspecificallyconcernsmonetaryfinancialinstitutionsexceptfortheEurosystem.(2) Toavoiddistortionofthemovementintheoutstandingamountsofsecuritiesbythereclassificationorrevaluationofsecurities,theoutstandingamountforJanuary

2005 wasincreasedcumulativelybythemonthlyvolumeoftransactionsinsuchsecurities.

4


There are limits to the reduction of the key interest rates

The transmission of the low interest rate policy tocredit interest rates and hence to the real economy

continuedin 2016,butinthefutureitcouldbeham‑pered by the downward rigidity of interest rates onbankdeposits.Althoughthebanksdidtosomeextentpass on the reduction in the deposit facility rate intheirratesoncustomerdeposits,theformerstoodat

ofbondsissuedbytheirowngovernment,whiletheportfolioofbondsofothergovernmentsintheeuroareaismorestable.Asthesovereigndebtcrisisintheeuroarearevealedtherisksinherentincloselinksbetweennationalgovernmentsandthedomesticbankingsector,theEurosystem’spurchaseprogrammeisthusindirectlyhelpingtosupportfinancialstability.Asmallerconcentrationofsovereignbonds–andinparticularthoseofthedomesticmarket–onthebanks’balancesheetsisinfactagoodthinginthatrespect.

CHART 16 LIMITSTOFURTHERCUTSINTHEKEYINTERESTRATES

–1 0 1 2 3 4 5 6

L

L

L

L

L

L

L

L

L L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L 0

10

20

30

40

50

60

70

80

90

100

Sept. 2016

June 2016

Jan. 2016

March 2015

June 2014

Dec. 2013

Sept. 2012

May 2010

Jan. 2008

DISTRIBUTION OF THE INTEREST RATES OFFERED BY A SAMPLE OF 283 EURO AREA BANKS ON HOUSEHOLD AND BUSINESS DEPOSITS

(1)

(in %)

Households Non-financial corporations

COMPOSITION OF THE BALANCE SHEET OF EURO AREA BANKS(in % of the balance sheet total, liabilities)

September 2016

Relatively fixed costs

Capital and reserves

Other deposits

Securities issued : long-term and fixed interest rates

Securities issued : short-term or variable interest rates

Interbank deposits

Eurosystem financing

Other liabilities

source : EcB.(1) Therectanglecontainsthevaluesbetweenthelowerandupperquartiles.Thehorizontallinesincludethevaluesbetweenthelowerquartileless1.5 timestheinterquartile

rangeandtheupperquartileplus1.5 timestheinterquartilerange.Thedotsindicatetheextremevalues.


at –0.40% at the end of 2016while the latterwasbarelynegative.

One reason for thedownward rigidityofdeposit interestrates is that retail customers’ deposits are an important,stable,andthereforevaluablesourceoffundingforbanks.However,iftheremunerationondepositsweretobecomelowerthanthecostofholdingbanknotes,householdsandbusinessesmightdecidetowithdrawtheirdepositedfundsandholdbanknotes instead.As large-scaledepositwith‑drawalscouldcausemajorproblemsforthebankingsector,banksarereluctanttocutthedepositratesbelowzero.Inviewofthesubstantialliquidassetsofsomefirms,thecostofholdingtheseassetsinbanknotesissignificant,sothattheinterestrateontheirbankdepositscanbecomeslightlynegative, and that did happen recently at some banks.Furthermore,somecountriessetlegallimitsonthereduc‑tionofcertain interest rates.That is thecase inBelgium,forexample,wheretheinterestrateonregulatedsavingsdepositscannotbelessthan0.11%(0.01%forthebasicrateand0.10%fortheloyaltybonus).

Thedifficultyofcuttingtheremunerationondepositsbelowzero,combinedwiththeimportanceofdepositsasasourceof funding, implies that – despite the low policy interestrates–thebanksfacerelativelyhighandrigidfundingcosts.Whendepositinterestratesdroptotheirlowerbound,banksmaydecidenottoreducetheircreditinterestratesanyfur‑ther,thuspreservingtheirinterestmargin.Furthercutsinthecentralbank’spolicyinterestrateswillthusceasetoinfluencecreditrates,andthatwillseriouslyimpedethetransmissionoftheaccommodativemonetarypolicythroughthebankingsector. The excess capacity in some national banking sys‑temsandtheresultingfiercecompetitionmayneverthelesspromptbankstocontinuereducingtheircreditinterestrates,eveniftheyleavetheirdepositratesunchanged.Moreover,the large proportion of loans granted at a variable interest rate–whichautomaticallytrackstheEuribororothershort-term interest rates – undermines the interest income of some banks.Cutsinthepolicyinterestratecanthusleadtolowercreditinterestrates,butatthesametimetheydepressthebanking sector’s profitability which has been under stresssince the outbreak of the crisis. In the long run, that cancauseacontractionofthecreditsupply,thwartingmonetarytransmission.Awareofthelimitsoftheinterestrateinstru‑ment,theECBGoverningCouncilthereforeindicatedthatitcannotcontinueindefinitelycuttingtheshort-termnominalinterestratesbelowzero.

Less conventional measures extend the boundaries of monetary policy

The Eurosystem therefore adopted non-standard instru‑ments such as balance sheet measures and forward

guidance,whichexertdownwardpressureonlonger-terminterest rates. The central bank thus continues to create morefavourablefinancingconditions,notonlyforbanksbut also for businesses and public authorities. Decisions on spending in the economyarenot influencedbyonespecific interest ratebut by thewhole rangeof interestratesapplicableintheeconomy.

Thesenon-conventionalmeasuresmayalsohaveunwel‑comesideeffects.For instance,thecontinuedflatteningof theyield curve in 2016 limits the scope forbanks toearnmoneyfrommaturitytransformation.Inthatregard,thebanksfundloansandlong-terminvestment(onwhichthey receive interest) mainly by means of short-termliabilities, including deposits (on which they pay inter‑est). Low long-term interest ratesmean that the banksprobablyhave to reinvest securities reachingmaturity inlower-yielding securities, thus risking further erosion oftheir interest margin.

Allthesame,theimpactofthelowinterestratepolicyonthebanking sector’sprofitability isnotnecessarilynega‑tive,atleastintheshortterm.Thebanks’netinterestin‑comeisdepressedbytheshrinkinginterestmargin,butatthesametimeitissupportedbythegrowthinthevolumeoflendingandtheimprovementinitsquality,whichalsoresult from the monetary measures. Those measures infactencouragetheeconomicrecovery,boostingdemandfor loans and placing borrowers in a better position tomeet their obligations. The APP also raises the value of thebanks’investmentportfolio,whiletheTLTROsreducetheir fundingcosts.But the longer the low interest rateenvironmentpersists, thegreatertheriskofpressureonthebankingsector’sprofitability,andhence itsability totransmitthemonetarystimulustotherealeconomy.

Thelowinterestrateenvironmentmayalsoimplyriskstofinancialstabilityinthebroadsense.Thus,itmaythreatenthefinancialhealthoflifeinsurersandpensionfunds,orleadtoanexcessivesearchforyield.

Finally, the non-conventional measures also come upagainstoperational limits.For instance,thespecificrulesof the APP – such as the original decision not to purchase assets offering an interest rate below the deposit facil‑ity rate,or theobligation toeffectpurchases inaccord‑ancewith the ECB capital subscription key –make themeasuresmoredifficult to implement.Theprogramme’sflexibility nevertheless allows the APP parameters to beadjustedsothatanyscarcityproblemscanbeovercome.In March 2016, for example, the Eurosystem raisedthe limits on purchases of securities issued by interna‑tional institutions and multilateral development banks. It is now permissible to hold 50% – instead of the


previous33% –perissuerandperissueofthosesecuri‑ties.Nonetheless,thescopeforadjustingtheAPPparam‑eters is still limited. For example, the Eurosystemwantsto prevent anymarket distortion and avoid acquiring adominant market position in securities.

The December decisions reflected those limits

In December 2016, the ECB Governing Council con‑sidered that the very accommodative monetary policystance should be maintained in order to strengthen the economic recovery and steer inflation back towards itstarget. In fact, according to the December projectionsproducedbytheEurosystem’sstaff,inflationwouldcometonomorethan1.7%in 2019.

The key interest rates were kept unchanged, althoughthe Governing Council confirmed that they would re‑main at or below their present levels for an extendedperiod of time, and well past the horizon for the netasset purchases. Regarding theAPP, theplannedperiodwas extended from the endofMarch 2017 to the endofDecember 2017, or beyond if necessary. That exten‑sionimpliesamorelastingpresenceonthemarket,andconsequentlymoresustainedtransmissionofthesupportmeasures. Nevertheless, the intensity of the measureswasreducedas,inthelightofebbingdeflationrisks,theGoverning Council also decided to reduce the monthly

netassetpurchasesfrom€80 billionto€60 billionfromApril 2017, thepaceadopted in the initialphaseof thepurchase programme. The reduction in the volume of monthly purchases and extension of the period of theAPP should alleviate anyproblems concerninga scarcityof securities and reduce the downward pressure on in‑terestrateswhilemaintainingthestrongstimulus. If theoutlook deteriorates or if financial conditions becomeinconsistent with further progress towards the inflationobjective, theGoverningCouncilwould consider raisingthe volume and/or extending the duration of the APP.Finally,toensurethesmoothimplementationoftheAPP,the Governing Council decided to relax certain param‑eterswitheffect fromJanuary 2017.Thus, therangeofmaturities for securities eligible for the PSPP, currentlyfrom two to thirty years,will extend fromone to thirtyyears.Moreover,theEurosystemwillifnecessarybeable,undertheAPP,topurchasesecuritiesyieldinganinterestratebelowthedepositfacilityrate.

The decline in real interest rates was weaker …

Whilenominalinterestratesfelltoahistoricalfloor,thatwaslesstrueofrealinterestrates.Infact,lowerinflationexpec‑tations,whichwerepartof thereasonfor theEurosystemmeasures, exerted upward pressure on real interest rates.Thus,thefive-yearrisk-freenominalinterestrateandtherateonnewcorporateloansforatermoffiveyearsorlesshave

CHART 17 REALINTERESTRATESDECLINEDLESSSHARPLYTHANNOMINALINTERESTRATES

(in%)

–3

–2

–1

0

1

2

3

4

2013 2014 2015 20162012

–3

–2

–1

0

1

2

3

4

2013 2014 2015 20162012

Nominal interest rate (a)

Inflation compensation (reverse sign, b) (1)

Real interest rate (a + b)

FIVE-YEAR RISK-FREE INTEREST RATEINTEREST RATE ON BANK LOANS TO NON-FINANCIAL CORPORATIONS FOR A FIVE-YEAR TERM

source : Thomson Reuters Datastream.(1) Measuredonthebasisofswapshedgingtheinflationriskintheeuroareaforafive-yearperiod.


exhibitedacleardownwardtrendinrecentyears,whilerealinterestratescalculatedonthebasisofinflationexpectationsderivedfromfinancialdatahavetendedtoremainstable.Inpractice, inflationexpectationsdeclinedtomuchthesamedegree as the nominal interest rate.

Since economic agents’ decisions on saving and invest‑mentareinfluencedbyrealinterestrates,fundingcondi‑tionsintheeconomy–atleastthosemeasuredbythesevery simple indicators – have eased less in recent yearsthanthedeclineinnominalinterestrateswouldsuggest,despite the significant monetary policy measures. It istherefore vital to bring inflation back towards 2% andanchor inflationexpectations around that level again sothatrealinterestratescanbesteeredappropriately.Hencethe importance of the announcement that theAPPwillremain in force until the Governing council is sure that inflationistendingtowardsitsobjective.

… while the real equilibrium interest rate is too low …

The low interest rate environment is not solely the re‑sultofmonetarypolicy,but is alsodue tonegative realdevelopments in theeconomywhicharedepressing theequilibriuminterestrate,i.e.therateatwhichsavingandinvestmentareinbalance,ortherateatwhicheconomicactivityachieves itspotential levelandinflation isstable.Structural forces, suchas the increase inexcess savings,lowerproductivitygrowthandgreater incomeinequalityintheeuroarea,butalsocyclicalfactors,particularlythedeleveraging initiated since the crisis and the reneweduncertainty, account for the current extremely low levelof theequilibrium interest rateboth in theEMUand inotheradvancedcountries.Accordingtosomeestimates,theequilibriumrateisactuallynegativeintheeuroarea.

Thelowequilibriuminterestrateobligesthecentralbanktopursuea low interest ratepolicy,because inorder toachievepricestability(andmacroeconomicstabilityinthebroad sense) monetary policy always tries to bring realinterestratesintheeconomyappropriatelyinlinewiththerealequilibriuminterestrate.Itispreciselybecauseoftheextremelylowlevelofthatrateandtheriskofanexces‑sivedeclineininflationexpectationsthattheEurosystemhas had to take non‑conventional measures to ensure thatthemonetarypolicystanceissufficientlyexpansion‑ary.Thelowinterestratethusreflectsadeeperproblem,namelyeconomicweakness.Atthesametime,thispolicyhelps to remedy thatweakness since it aims to rekindleeconomicactivityandhenceinflation.

In these circumstances, the economic recovery in theeuro area benefits not only from the low real interest

rates but also from the increase in the equilibrium in‑terest rate.TheEurosystemcan takemeasures todrivedown the former but has little or no influence overthe atter.

… which highlights the importance of fiscal policy and structural reforms in reinforcing the monetary stimulus

The Governing Council has already repeatedly stressedinitscommunicationthatotherpolicyspheresalsoneedto be used to ensure the structural continuation of the economicrecovery,supportedbythemonetarystimulus.Monetarypolicyisnottheonlyfactor;agoodpolicymixwhichcanactivatethevariousspheresiscrucial.

Aroleisthusassignedtoafiscalstancemoreconducivetogrowthintheeuroarea,withmaximumuseofthescopeofferedbytheSGPrules.Inthatregard,thelimitedfiscalleewayavailabletomosteuroareacountriesrequirescare‑fullyconsideredmeasurestoachieveabalancebetweensupport for aggregate demand and the maintenance of confidenceinpublicfinances.Forexample,aneutralshiftin public expenditure to the detriment of unproductiveitemsandinfavourofgrowth-friendlypublicinvestmentmeets that requirement. In addition, such a measureoptimises the beneficial effects of the low interest rate,whichrendersinvestmentprojectsmoreattractive.Higherpublicinvestmentinturnboostspotentialgrowthandtheequilibriuminterestrate.

Newstructuralreformsaimedatthefundamentalforcesbehindthelowequilibriuminterestratearelikewiseveryimportant. But the pace of reform was only moderatein 2016,whiletherearestillmanychallengesfacingtheeuroareacountries:continuingEMUintegration,raisingtheemploymentrate,stimulatingentrepreneurship,guar‑anteeingthesustainabilityofpensions,reducingthepro‑portionofnon-performingloansonbankbalancesheets,etc.Astructuralpolicy thataddresses theseproblems inasustainedmannerwillhelptorestoreconfidence,raiseproductivity, drive up the equilibrium interest rate andboostpotentialgrowth.

Inaddition,ariseintheequilibriuminterestratereducesthe risks that the low interest rate policy may implyfor financial stability, as it allows the central bank tonormalise its interest rates more quickly. In the euroarea,thereislittlesignsofarofwidespreadriskscausedby the low interest rate environment. In the FinancialStabilityReviewitpublishedinNovember 2016,theECBconsidersthattheimpactofthelowinterestratepolicyonbanks’profitability is fairlyneutral.Nor is thereanywidespread evidence of an excessive search for yield,


inordinate overvaluation of assets, or excessive creditexpansion. If such risks materialise, they are generallyspecific to certain countries or branches of activity, sothatprudentialpolicyisthebestwayofcontainingthemin thefirst instance.TheEuropeanSystemicRiskBoard

(ESRB) plays a key coordinating role here. Monetarypolicy can thus concentrate fully on maintaining pricestability.Nonetheless,theECBstilltakesaccountofanyadverse effects of its extremely accommodative policy,andkeepsaclosewatchonthem.

Economicactivity,demandandemploymentinBelgium

71Economicandfinancialdevelopments ❙ ECONOMICACTIVITY,DEMANDANDEMPLOYMENTINBELGIUM ❙

2. Economicactivity,demandandemploymentinBelgium

2.1 Belgium’seconomyrecordssubduedgrowthin 2016butcreatesmanyjobs

Followingtwoyearsofalreadyquitemodestgrowth, 2016sawafurthermoderationinthewakeofarelativelylack‑lustrestarttotheyear.Involumeterms,year-on-yearGDProseby1.2%.Yet, theeconomicexpansion,whichhadfluctuatedaround1.5%inthetwopreviousyears,provedquiterobust.

Growthisalsoincreasinglycreatingjobs–anormalpat‑ternatalaterstageofeconomicrecovery.Thatsaid,therising labour intensity of economic growth also reflects

policies regardingwagemoderationandcost reductionsimplementedoverthepastfewyearswhichwererecentlyaccelerated, as well as recent years’ structural reformsthat supported both the demand for and the effective supplyoflabour.Robustjobcreationmakesaneconomymore resilient as it broadens the fundamentals of domes‑tic demand.

Thisresiliencewasputtothetest in 2016,startingwiththe terrorist attacks. After the Paris attacks in the autumn of 2015, Belgium imposed strict security measures. On22 March 2016, Belgium was itself hit. The aftermathoftheattackswasparticularlynoticeable intourismandbusinesstravel,aswellasinthehotels,cafésandrestau‑rants industry. In June, the unexpected outcomeof theBrexitvoteintheUnitedKingdomgreatlyaddedtopoliti‑caluncertainty,whiletheautumnsawawaveofadversecorporatenewsgrabbingtheheadlines.Thisdentedcon‑sumerconfidence–albeittemporarily–asunemploymentexpectationsroseandgeneraleconomicprospectsweak‑ened. As it turned out, the economic recovery provedsolidly based and this series of shockswould appear tohavehadonlylimitedconsequencesforgrowth.Businessconfidencetrendsremainedpointingclearlyupwards.

Growth of economic activity chiefly driven by private sector branches

In manufacturing, which had been a key contributorto economic expansion in 2015, weak foreign de‑mand – particularly from economies outside the euroarea – halved the increase in added value in 2016.Construction also failed to sustain 2015 growth rates,with added value peaking onmildweather conditionsin the first quarter, then edging down but still endingthe year higher than the average rise recorded since

CHART 18 ECONOMICGROWTHCREATESMANYJOBS

(averageannualgrowthin%)

2000-2007 2008-2013 2014 2015

0.0

0.5

1.0

1.5

2.0

2.5

GDP

Employment

2016 e

Sources:NAI,NBB.


2000. Non-market services, which include public ad‑ministration, education, health care and social work,wereaffectedbyfiscalconsolidationeffortsand infactdetractedfromgrowth.Onlymarketservices,whichac‑countforover50%ofthetotal,continuedtogrowatthesamerate,butthisprovedinsufficienttooffsettheslowdowninotherbranches.

Belgian growth below most neighbouring countries for past few years

Despite resilient economic activity and a robustly recov‑ering labour market, Belgian growth does not appeartobe really kickingoff: havingnotchedupaveragean‑nual growthof0.7% in the2008-14 period, economicactivityrosebynearly1.4%onaverageinbothin 2015and 2016.Thetotalgrowthfigureof2.7%forthepast

twoyearsiscomparabletoFrance’s,butclearlybelowthatoftheNetherlandsandGermany.

Belgium’s relatively low growth numbers are primarilycausedbyasmallergrowthcontributionbyprivatecon‑sumption,asthewagemoderationofthepastfewyearshasdepressedhouseholds’purchasingpower.Inaddition,general government consumption made a smaller contri‑butionbecauseofcontinuedgovernmentsavings.Intheshort term, then, these government-imposed measuresslowdowngrowthalthough theywouldalreadyappeartobebenefitingthelabourmarket,andshouldtranslateinto healthier economic fundamentals in the longer run. ThisispreciselywhathappenedintheNetherlands,wherethe economy, after a longer and deeper recession, iscurrentlybouncingbackmorestrongly thanks to rapidlyimplementedreformsandausteritymeasures.

CHART 19 MODERATEGDPGROWTHANDSTEADILYGROWINGBUSINESSCONFIDENCEINANEVENTFULYEAR

–1

0

1

2

3

2012 2013 2014 2015 2016–15

–10

–5

0

5

2012 2013 2014 2015 2016

–30

–20

–10

0

10

20

30

2012 2013 2014 2015 2016

–70

–60

–50

–40

–30

–20

–10

0

10

20GDP AND BUSINESS CONFIDENCE

EXPORT OF TOURIST SERVICES(Percentage changes compared with the corresponding month of the previous year)

Percentage changes compared with the corresponding quarter of the previous year

Overall synthetic business survey curve (1)

(right-hand scale)

Smoothed series

Gross series

GDP (left-hand scale)

Percentage changes compared with the previous quarter

Consumer confidence indicator

Sub-indicator of unemployment expectations (–)

CONSUMER CONFIDENCE (1)

Sub-indicator of the outlook for general economic conditions

Sources:EC,NAI,NBB.(1) Balanceofrepliestomonthlysurveys,noncalendaradjusteddata


CHART 20 PRIVATESECTORBRANCHESMAINCONTRIBUTORSTOGDPGROWTH

(contributions to the change in GDP relative to the previous year,inpercentagepointsunlessotherwisestated;dataadjustedforseasonalandcalendareffects)

2012 2013 2014 2015 2016–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

Manufacturing industry

Construction

Market services

Non-market services

Other (1)

GDP (2)

Source:NAI.(1) Particularly“Agriculture,forestryandfishing”andproduct-relatedtaxesnetof

subsidies.(2) Percentagechangescomparedwiththepreviousyear.

CHART 21 PUBLICSPENDINGCUTSANDWAGEMODERATIONASHORT-TERMDRAGONBELGIANGROWTH

(contributionstogrowthin 2014-16 period,expressedinpercentagepointsunlessotherwisestated)

FR BE NLDE

–2

–1

0

1

2

3

4

5

6

J

Private consumption

General government consumption

Investment

Net exports

Change in inventories

GDP growth in %

Sources:BdF,DBB,DNB,NBB.

Box2 – BrexithaslittleimpactonBelgium–fornow

Oneofthemainconfidenceshocksof2016wastheunexpectedoutcomeoftheBrexitreferendum.On23June2016,theUnitedKingdomaskedtheBritishpeoplewhetherthecountryshouldremainamemberoftheEuropeanUnion.Confoundingallexpectations,a slightmajorityof51.9%ofvoterschose to leave theUnion.Althoughthiswasanadvisoryreferendumandmanypoliticalandlegalhurdlesstillneedtobecleared,theUnitedKingdomseemscertaintowithdrawfromtheEU.ThisboxbrieflyoutlinestheUK’seconomicimportanceforBelgiumanddiscussesthepotentialrepercussionsofBrexitforeconomicactivity,bothintheUKandinBelgium.

The leave vote and eventual Brexitwill be felt in theUnited Kingdomfirst and foremost. In economic terms,demand channels are likely be affected anduncertainty and confidence effects couldhave adirect impact onhouseholdconsumption,corporate investmentandemploymentdecisionsevenbefore thecountrypullsoutoftheEuropeanUnion.Theseeffectsareliabletotriggerresponsesinthefinancialandcurrencymarketsthatmayintheirturninfluenceconsumption,capitalspending,andimportandexportflows.

Inhindsight,theUKeconomyprovedunexpectedlyresilientinthemonthsfollowingthereferendum.Amongotherfactors,thiswasprobablyaresultoffurthereasingofmonetarypolicyasannouncedbytheBankofEnglandinAugustandofnewsthatfiscalpolicieswouldbeputinplacetocushionanyeconomicslowdown.Afteraverybriefperiodoftensioninthefinancialmarkets,theinitialshockworeoff.Sincethereferendumannouncement,sterling

4


hadweakenedanditsdeclinecontinuedpost-referendum.Thecurrencydepreciatedbynearly15%againsttheeuroin2016.Intheshortterm,aweakerpoundgivesUKmanufacturersacompetitiveedgeovertheiroverseasrivalsandissupportiveofindustrialactivity.Attheendoftheday,GDPstagedthird-andfourth-quartergrowthmatchingthespringfiguresof0.6%.Atthesametime,though,sterlingdepreciationmakesformoreexpensiveimportsofgoodsandservices, raisingthecostof imported inputs formanufacturersandpushingup inflation,whicherodedhouseholdpurchasingpower.

TheBrexiteffectontheUKeconomyandonthesterlingexchangerateischieflypercolatingthroughtoitstradingpartnersviaglobaltrade.TheriseinthevolumeofexportstotheUnitedKingdomfromBelgium–whichhadgotunderwayatthebeginningof2016–initiallyremainedunchangedinthesummer,butrecentforeigntradedatasuggest a drop in september and october.

In themedium to long term,aUKdeparture from theEU is likely to increaseobstacles to theflowofgoods,services,peopleandcapital, including intangibleassets,whichwoulddepressproductionpotential,both intheBritisheconomyandthatofitstradingpartners,includingBelgium.Thepotentialreintroductionofcustomsandexcisedutiesongoodsisunlikelytobetheonlyeffect;non-tariffbarriersmightalsobite,astechnicalstandardsorcertificationrequirementsfortheUnitedKingdomandtheEUcouldgraduallystarttodivergeandcompanieswillincuradditionalcoststoservethesetwomarkets.Theseaspectsareparticularlypertinentinservices,includingfinancialservices,inwhichtheCityofLondonisacentralplayer.Brexitwillcomplicatethemovementofcapitalandpeople,andmightputabrakeonforeigndirectinvestment,akeydisseminatorofinnovation.

AnyassessmentofBrexit’s long-term impactwillbe largely influencedbyassumptions regarding thenatureoftheeconomictiesbetweentheUnitedKingdomandtheEUafter thecountry leaves theUnion.TheoptimisticscenarioofasoftBrexitseestheBritishcontinuingasmembersoftheEuropeanEconomicArea(EEA),implyingfreemovementofpeople.A‘hard’Brexit,afterwhichtherelationshipbetweentheUnitedKingdomandtheEU

BREXITCOULDDEPRESSBELGIANEXPORTSTOTHEUNITEDKINGDOM

(exportsofgoodsfromBelgium;byvolume,index2013 =100)

2014 2015 201696

98

100

102

104

106

108

110

112

114

Monthly data

Average in the last three months

Average in the past three months

Monthly data

Exports to the United Kingdom

Total exports

Source:NAI.

4


wouldbecoveredonlybytherulesoftheWorldTradeOrganisation(WTO),wouldhavemuchgreatereconomicrepercussions.Atthispoint,allscenariosremainpossible,includinghybridsthatincludefreetradeagreements,astheUKgovernmenthasyettoformallynotifytheEuropeanCouncilofitsintentiontoleave.NotuntilithasdonesowillnegotiationsgetunderwaytoworkoutthepracticaldetailsoftheBritishexit,toformalisefuturetiesandperhaps to hammer out a transition agreement.

Oncurrentinformation,theextentoftheimpactishardtogauge.StudiesbytheBritishauthoritiesandinternationalinstitutionsputthenegativeimpactonBritishGDPatbetween2.7%and6.3%by2030intheeventofa‘soft’Brexit,andatbetween7.5%and9.5%ina‘hard’Brexitscenario.FormostofitsEuropeanpartners,thiseffectwouldbequiteabitsmaller,butstillsignificant.GivenBelgium’scloseeconomictieswiththeUnitedKingdom,thiscountry is likelytosee itsgrowthsomewhatdepressed,astheUKisakeyeconomicpartner inarangeofways.Onthemostrecentavailabledata(for2011),around2.5%ofBelgianvalueaddedderivesfromBritishfinaldemand;inBelgium’sneighbouringcountries,theUKaccountsforasmallerproportion.Mostimportantinthiscontextintheindustrialarenaarechemicalsandthemanufactureofcarsandothermeansoftransport,alongsidetravel,transportandbusinessservices.

Inthe2013-15period,theUnitedKingdomaccountedforanaverage5.3%ofBelgium’simportsand7.4%ofits exports (national concept),making it the fourth biggestmarket for Belgian exporters. Irrespective of theseproducts’relativeweightintheBelgianeconomy,thismarketisexceptionallyimportantforsuchgoodsascarpetsandfurnishings,carsandvariousfoods.Tradeinservicesisalsorelativelyimportant,cominginat8.7%forimportsand8.5%forexports. Inaddition,theUKisamajor investmentpartner:thebalanceofpayments inthesameperiodshowsthat6.7%ofBelgium’sforeigndirectinvestment(FDI)endedupintheUnitedKingdom,while3.5%ofcapitalspendinginBelgiumcomesfromtheUK.

BELGIUM’SCLOSEECONOMICTIESWITHTHEUNITEDKINGDOM

0 1 2 3 4 5 6 7 8 10 8 6 4 2 0 1086420

Italy

Finland

France

Germany

Sweden

Netherlands

Belgium

Luxembourg

Cyprus

Malta

Ireland

Foreign directinvestment

Services transactions

Trade in goods

UK FINAL DEMAND AS A PROPORTION OF VALUE ADDED (in % of GDP, 2011)

BELGIUM AND UK DIRECT ECONOMIC TIES(in % of total, 2013-2015)

EXPORTS IMPORTS

Sources:Eurostat,NAI,OECD(TiVA),NBB.


2.2 Growthofeconomicactivitydrivenbyinvestmentandnetexports

The contribution to GDP growth of domestic demand,excludingthechangeininventories,wasalmostthesameasthatofnetexports in 2016,at1 and0.8 percentagepoints respectively. However, the change in inventorieshadanegativeimpactof0.5 percentagepoints.

General government consumption inched ahead by amere0.2%in 2016,yetanotherdeclineonthepreviousyear. This near-stagnation reflects continuous fiscal con‑solidationbythegovernment,whichcausedaslowdownin thegrowth rateofcurrentexpenditure.Public invest‑ment,however,wasuponcemore,toitshighestshowinginfiveyearsat3.6%.

Subdued growth in private consumption

Household consumption added 0.7% in real termsin 2016, 0.4 percentage points below the figurefor 2015,butcomparabletopreviousyears.Weakcon‑sumption figures were also reflected in, for instance,retail sales: after an all-time high in mid-2015, theserapidly revertedback to their levelsof2010, in theaf‑termath of the great recession. Disappointing tourism numbers and modest spending on leisure and hotels,cafésandrestaurantsaftertheParisandBrusselsattacksundoubtedlyplayedapart.

To an extent, the year’s slow consumption growth wasattributabletoweakerfiguresinthesecondhalfof 2015,which spilledover into the 2016annualgrowth in con‑sumerspending.Ineffect,evenatthebeginningof 2016,consumption growth remained constrained, as the sup‑portive effects of lower oil prices on real purchasingpowerpeteredoutandhouseholdssteppeduptheirpre‑cautionarysavings.Notuntilthesecondquarterof 2016was there any upward trend in the quarter-on-quarterchange in consumption.

Although the consumption figures have not been par‑ticularly dynamic, confidence indicators are suggestingfairlystableunderlyingconfidenceinhouseholdincomes:although themonthly consumer confidence gaugewasfairlyvolatileas 2016progressed,everydipwasfollowedbyarobustrecovery.Oneratherstrikingcaseinpointwasthesub-indicatorforunemploymentexpectations,whichpeakedintheautumnof 2016onawaveofnegativeandhigh-profilecorporatenews.

Resilience of confidence is also captured by the sub-indicator gauging consumers’ views of the presentas the right time for big‑ticket purchases. This has been steadily recovering since the great recessionand peaked provisionally in the first half of 2016. Itis clear from risingnewcar registrations in 2016 thatitsrecoveryhaseffectivelygonehandinhandwithanincrease in consumption. It would appear, then, that

Table 2 GDP AND MAIN EXPENDITURE CATEGORIES

(calendar adjusted volume data ; percentage changes compared to the previous year, unless otherwise stated)

2012

2013

2014

2015

2016 e

Private consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.6 0.7 0.6 1.1 0.7

General government consumption . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 0.1 1.4 0.5 0.2

Gross fixed capital formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.2 −1.5 5.1 2.4 2.4

Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.1 −3.9 4.7 0.8 5.4

Enterprises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0 −0.3 5.6 3.1 1.2

General government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 −4.2 3.1 1.9 3.6

p.m. Final domestic expenditure (1) (2) . . . . . . . . . . . . . . . . . . . . . . . . 0.7 0.1 1.8 1.2 1.0

Change in inventories (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.9 −0.6 0.4 0.3 −0.5

Net exports of goods and services (2) . . . . . . . . . . . . . . . . . . . . . . . . 0.3 0.4 −0.6 0.0 0.8

Exports of goods and services . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8 0.8 5.1 4.3 5.8

Imports of goods and services . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 0.3 5.9 4.3 5.0

GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.1 −0.1 1.7 1.5 1.2

Sources : NAI, NBB.(1) Excluding the change in inventories.(2) Contributions to the change in GDP compared with the previous year, percentage points.


confidencewasresilientenoughforconsumerstomakemajorpurchases.

Spendingonresidentialpropertyalsoaddedahefty5.4%in 2016andhouseholdswouldappear fairly imperviousto shocks and temporary blips and dips caused by un‑certainty. Besides, persistently lowmortgage rates havereducedbarriers tobuildingorpurchasinghomes,whileaccommodatingmonetarypoliciesaresqueezingreturnsonfinancialassetsand thereforemaking themarket fornewbuildsrelativelymoreattractiveto investors lookingforyields.Thatsaid, investment inresidentialproperty isstillover7%belowitsend-2007 high.

Household gross disposable income surges

Privateconsumptionrecordedmuchslowergrowththanhouseholds’ gross disposable income, which saw arise of 1.3% in real terms in 2016, its biggest increasesince 2009. Moreover, gross disposable income grewfasterthanBelgium’spopulation,whose 2016growththeFederal PlanningBureau (FPB) puts at around0.8%ac‑cordingtothemostrecentdemographicdata.Asaresult,realaverageincomepercapitawasupbyabout0.5%.

Theincreaseingrossprimaryincomewaschieflyattribut‑abletoemployees’highertotalgrosswages(up2.3%at

currentprices),astheseaccountforaroundtwo-thirdsofgross primary income. These higher grosswages reflecta sharp rise in thevolumeof labour,whilegrosswagesper hour worked (excluding employers’ contributions)bouncedbackby1.1%.In 2015,thislatteritemrosebyamere0.5% in thewakeofcontinuouswage restraintpolicies.Althoughthewageindexationmechanismswerereactivated fromApril 2016, their temporary suspension(the so-called index jump) still constituted a significantdragonaverageannualfiguresfor 2016.

The sum total of the gross operating surplus and gross mixedincomegrewby2.2%innominalterms,mainlyonthebackofincomeearnedbytheself-employed,whosenumbers kept rising, while favourable economic condi‑tions also contributed.

After steep falls in previous years, income fromwealthwasvirtuallystablein 2016,ashigherdividendsreceivedoffset the ongoing decline in net interest income and otherincomefromwealth.

Thesharpriseindisposableincomewasalsounderpinnedbythedistributionofsecondaryincome,i.e.thebalanceof current transfers received from and paid to other sec‑tors.Currenttransfersreceived,mainlysocialbenefitpay‑ments,rosemorethaninthetwopreviousyears,by2.5%

CHART 22 CONTINUEDSUBDUEDGROWTHINPRIVATECONSUMPTION

2012 2013 2014 2015 2016–25

–20

–15

–10

–5

0

5

10

15

2012 2013 2014 2015 201616

18

20

22

24

2012 2013 2014 2015 201695

97

99

101

103

105

Appraisal of intentions of making major purchases

Outlook for major purchases of households (next 12 months)

DURABLE GOODS(Balance of replies, seasonally adjusted data)

Number of registrations

Average since 2000

NEW CAR REGISTRATIONS(in thousands, seasonally adjusted data)

RETAIL SALES(2010 index = 100, volume data adjusted for seasonal and calendar effects)

Sources:EC,FEBIAC,NBB.


atcurrentprices.Theindexjumpwasinfactphasedoutin the course of 2016 and benefits rose as the springprogressed, after the threshold index had been passed.All that said, transfers paidwere the key driver behindthesurgeinnettransfers,astheyweresqueezedbylowerpersonalincometaxandreducedsocialsecuritycontribu‑tions by employees under the tax shift, adding up to amere0.2%increase.However,thetaxshiftandaraftofothergovernmentmeasuresalsohadanimpactoninfla‑tion(up0.8 percentagepoints)andrealpurchasingpowerwasnegativelyaffected,whichmayproveakeyfactorforhouseholdswithahighratioofconsumptiontoincome.

Household savings ratio up, after earlier steady declines

Withprivateconsumptionclearlylaggingbehindtheincreaseindisposableincome,householdsstartedsavingmore.Thesavings ratio – i.e. the proportion of disposable income that householdssave–shotupto12.2%in 2016,endingthelongandsteadydeclineofpreviousyears.Thesavingsratio

hasbeenfluctuatingaround12%for threeyears runningbutisstillwellbehindlevelsseenbeforethegreatrecession.

Althoughcautionisrequiredwheninferringcausalrelation‑ships,lowersavingsbetween2010 and 2015andtheirlowlevelsin 2016wouldappeartoberelatedtodecliningincomefromwealth,fromwhichhouseholdstypicallysavemorethantheydofrompayorbenefits.Duringthistime,households’private consumption relative to non-wealth income (i.e. allincomeexcluding incomefromwealth)hasremainedmorestable than its relationship to their disposable income. This wouldseemtoshowthatthelowinterestrateenvironment,whichbasicallymakessavinglessattractive,hasnotprompt‑ed households to consume more of their incomes.

Perhaps itwasnot somuch theirhigherwealth incomethat prompted consumers to save more in 2016; theymay well have started saving more because of mount‑inguncertaintyandasaprecautionarymeasure.Whatismore,individualsmaywelltakesometimetoadjusttheir

Table 3 DETERMINANTS OF HOUSEHOLD GROSS DISPOSABLE INCOME, AT CURRENT PRICES

(percentage changes compared to the previous year, unless otherwise stated)

p.m. In € billion

2012

2013

2014

2015

2016 e

2016 e

Gross primary income (1) . . . . . . . . . . . . . . . . . . . . . . 1.6 1.0 0.8 0.7 2.0 235.0

Gross wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 1.6 1.4 1.1 2.3 157.7

Volume of labour of employees . . . . . . . . . . . 0.4 −0.7 0.1 0.6 1.2

Gross wages per hour worked (1) . . . . . . . . . . 2.8 2.3 1.3 0.5 1.1

Gross operating surplus and gross mixed income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 −0.2 1.8 2.7 2.2 50.6

Capital income (2) . . . . . . . . . . . . . . . . . . . . . . . . . −5.4 0.0 −4.2 −4.6 0.2 26.7

Interest (net) . . . . . . . . . . . . . . . . . . . . . . . . . . . −20.9 −32.0 −57.3 −60.3 −74.8 0.1

Dividends received . . . . . . . . . . . . . . . . . . . . . . −1.8 11.4 4.3 0.9 4.3 15.9

Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −1.4 0.4 0.3 −5.0 −1.9 10.7

Net current transfers (1) . . . . . . . . . . . . . . . . . . . . . . . 30.9 −8.7 15.6 18.2 48.0 6.6

Current transfers received . . . . . . . . . . . . . . . . . . 5.2 3.3 1.5 1.2 2.5 95.4

Current transfers paid (1) . . . . . . . . . . . . . . . . . . . 4.3 3.8 1.0 0.5 0.2 88.8

Gross disposable income . . . . . . . . . . . . . . . . . . . . . 1.9 0.9 1.0 1.0 2.9 241.6

p.m. In real terms (3) . . . . . . . . . . . . . . . . . . . . . . . . . −0.1 0.0 0.3 0.7 1.3

p.m. In real terms per person (4) . . . . . . . . . . . . . . . −0.6 −0.4 −0.3 0.1 0.5

Savings ratio (5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.9 12.3 12.1 11.7 12.2

Sources : FPB, NAI, NBB.(1) Wages and salaries received and current transfers paid, not including contributions paid in by employers.(2) These are net amounts, i.e. the difference between income or transfers from other sectors and those paid to other sectors.(3) Data deflated by the household final consumption expenditure deflator.(4) Gross disposable income in real terms divided by annual average number of people.(5) In % of disposable income in the broad sense, i.e. including changes in households’ supplementary pension entitlements accruing in the context of an occupational activity.


consumption habits to the way their incomes develop,particularlywhenhitbysuddenshocks–e.g.measurestoboostpurchasingpoweraspartofthetaxshift.

Specificsub-indicatorsforhouseholdsavingsasmeasuredbytheNBB’smonthlyconsumersurveysuggest thatpri‑vate individualsexpecttokeeptheirsavingsunchanged,eveniftheyrealisethatconditionsdonotcurrentlyfavoursavers.Thetwotrendshaveclearlydivergedsincethestartof 2015andsuggestthatpeoplehaveoptedtocontinuesavingdespitefallingsavingsyields.

Businesses invested heavily

In 2016,businessconfidenceshruggedoffemergingun‑certaintyandimprovedsteadily,asborneoutbyastronginvestment performance. Granted, business investmentshowedamere1.2%advancein 2016,butthispercent‑agewasheldbackbyafewlargepurchasesmadeabroadin the previous year. If these individual transactions are

factoredout,businessinvestmentgrewby5.2%in 2016,arapidaccelerationon 2015.CorporatecapitalspendingmadeasignificantcontributiontoGDPgrowth.

The levelofgrossfixedcapitalformationwassupportedby an improveddemandoutlook for themanufacturingindustry, encouraging companies to invest in expan‑sion.What ismore, production capacity utilisation rateshavebeenconsistentlyabovetheaverageobservedsince1980 androseevenfurtherinthecourseof 2016.

Accommodating funding options also promote investment growth.Andithasnotjustbeenexternalfinancialcondi‑tionsthatarehighly favourableonthebackofmonetarystimulus–companies’internalresourceshavekeptgrowingandareincreasinglyheldincash.In 2016,companiessawtheirgrossoperatingsurplus,i.e.theirrevenuesfromoper‑atingactivities,goupby5.7%innominalterms,whichiscomparabletothefigurefor 2015.Thisimpliesthat,onceagain,thesurplusrosefasterthanaddedvalue.

CHART 23 FALLINGINCOMEFROMWEALTHALONG-TERMDRAGONHOUSEHOLDSAVINGS

2000

2002

2004

2006

2008

2010

2012

2014

10

12

14

16

18

20

2000

2002

2004

2006

2008

2010

2012

2014

20

25

30

35

40

45

2000

2002

2004

2006

2008

2010

2012

2014

96

98

100

102

104

106

2012

2013

2014

2015

2016

–40

–20

0

20

40

SAVINGS RATIO(in % of gross disposable income)

Capital income

CAPITAL INCOME AND SAVINGS(in € billion)

2016

e

2016

e

Savings

CONSUMPTION RELATIVE TO NON-CAPITAL INCOME(in %)

Household savings capacity outlook

SAVINGS OUTLOOK AND EXPEDIENCY(balance of replies, deviations from average 2005-15 period)

2016

e

Appraisal of expediency of saving

Sources:NAI,NBB.


CHART 24 SOLIDFOUNDATIONSUNDERPINBUSINESSINVESTMENT

–20

–15

–10

–5

0

5

10

15

20

2008 2009 2010 2011 2012 2013 2014 2015 2016–40

–35

–30

–25

–20

–15

–10

–5

0

5

10

15

20

2008 2009 2010 2011 2012 2013 2014 2015 201660

65

70

75

80

85

Gross fixed capital formation of enterprises (1)

(percentage changes compared with the same quarter of the previous year)(left-hand scale)

Smoothed series

Gross series

Demand outlook in manufacturing industry (balance of replies to the monthly surveys) (right-hand scale)

BUSINESS INVESTMENT AND DEMAND OUTLOOK(volume data adjusted for seasonal and calendar effects)

CAPACITY UTILISATION IN MANUFACTURING INDUSTRY(in %, quarterly data adjusted for seasonal effects)

Capacity utilisation rate

Average since 1980

Sources:NAI,NBB.(1) Adjustedforexceptionallylargeindividualtransactions.

Table 4 DETERMINANTS OF THE GROSS OPERATING SURPLUS OF COMPANIES (1), AT CURRENT PRICES

(percentage changes compared to the previous year, unless otherwise stated)

2012

2013

2014

2015

2016 e

Gross operating margin per unit of sales (2) . . . . . . . . . . . . . . . . . . −2.8 0.1 0.9 2.6 2.3

Unit selling price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 0.1 −0.4 −1.2 −0.6

On the domestic market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 0.9 0.5 0.3 1.4

Exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 −0.3 −1.0 −2.1 −1.5

Unit sales costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 0.1 −0.6 −1.9 −1.1

Imported goods and services . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 −0.5 −1.4 −3.1 −2.5

Costs of domestic origin per unit of output (2) (3) . . . . . . . . . . 3.8 1.4 −0.2 −0.7 0.6

of which :

Unit labour costs (4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 1.8 −0.5 −1.1 −0.2

Unit net indirect taxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 −0.1 0.3 −0.6 3.2

Final sales at constant prices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 0.3 3.9 3.1 3.3

Gross operating surplus of companies . . . . . . . . . . . . . . . . . . . . . . −1.7 0.4 4.8 5.8 5.7

Sources : NAI, NBB.(1) Private and public companies.(2) Including change in inventories.(3) In addition to wages, this item includes indirect taxes less subsidies, and gross mixed income of self‑employed people.(4) Unit labour costs are expressed in units of value added of the business sector and are not calendar adjusted.


Gross operating surplus as a share of companies’ valueaddedreachedhistoricallyhighlevelsin 2016,asthewagebillhas increased little in thepast fewyears. Typically re‑spondingtochangesintheirfinancialsituationaftersomedelay,companiesaretakingtheopportunitytoinvestandthe investment ratio – business investment as a ratio of their value added – appears to have been following theupwardtrajectoryoftheoperatingsurplussincemid-2015.

Much as in 2015, both wider margins and unit salestrends contributed to companies’ higher operating sur‑plus. Companies notched up a 2.3% advance in grossoperatingmargin,astheydidnotfullypassonthereduc‑tioninunitsalescosts–of1.1%–toprices,whichonlycontracted by 0.6%. As in previous years, these lowerunit sales costs reflected cheaper goods and servicesimports, the main factor being low commodity prices,andoilpricesinparticular.Unlikethepreviousyear, 2016didseecostsofadomesticorigingobackup,by0.6%,mostlybecauseofhigherunitnetindirecttaxes.Unit la‑bourcostsweredownonlymodestlyin 2016,comparedwithmoresignificantfallsinthetwopreviousyears.Salesvolumes,bycontrast,wereupby3.3%.

CHART 25 GROWINGIMPORTANCEOFGROSSOPERATINGSURPLUSAPPEARSTOFUELBUSINESSINVESTMENTRATIO

(in%ofgrossaddedvalueofcompanies)

2008

2010

2012

2014

2016

34

36

38

40

42

44

22

24

26

28

30

32

Share of gross operating surplus(left‑hand scale)

Investment ratio (1)

(right‑hand scale)

Sources:NAI,NBB.(1) Adjustedforexceptionallylargeindividualtransactions.

CHART 26 HIGHERCORPORATEANDINDIVIDUALFINANCINGCAPACITYANDAPOSITIVECURRENTACCOUNTBALANCE

(in%ofGDP)

2000

2002

2004

2006

2008

2010

2012

2014

–6

–4

–2

0

2

4

6

8

10

2008

2009

2010

2011

2012

2013

2014

2015

–4

–3

–2

–1

0

1

2

3

4

5

Companies

General government

Individuals

p.m. Current account (according to the balance of payments)

Financing requirement (–) or capacity (+) of domestic sectors as a whole

Goods and services

Primary income

Secondary income

Current account

FINANCING BALANCE OF ALL DOMESTIC SECTORSCURRENT ACCOUNT (ACCORDING TO THE BALANCE OF PAYMENTS)

2016

e

2016

e

Sources:NAI,NBB.


Current account continues to improve

Confirming the recovery that had started in the previ‑ousyear,Belgium’sfinancingcapacitycontinuedtogrowin 2016 and, as estimated in the national accounts, itreached nearly 0.9% of GDP, compared with 0.2%in 2015.Thecountry’simprovedfinancingcapacitymostlyreflectedthatofcompanieswhichrecordedasignificantincrease in their operating income and a reduction of theirinventories.Privateindividualslikewiseboostedtheirfinancingcapacityin 2016astheirsavingsoutpacedtheirinvestmentspending.Together,thesetrendsamplyoffsetthegeneralgovernment’shigherborrowingrequirements.

Balanceofpaymentsfigures show this favourable trendin the financing capacity of the Belgian economy as awholetohavepercolatedthroughintothecountry’scur‑rent account balance with the rest of the world. Afterfour yearsofminordeficits, the current account turnedpositive again in 2015and recordeda surplusof nearly1%ofGDPin 2016.

The current account improvement was underpinned byhigherresultsfromgoodsandservicestransactions,whichsaw a surplus of nearly €9.3 billion in 2015 widen to€15.9 billion in 2016,and towhichboth foreign tradeprices and volume trends have contributed.

Ontheonehand, justas in 2015,theBelgianeconomybenefitedfromanimprovementinthetermsoftradeofaround1%in 2016,mainlybecauseoftheenergypricefallsthatcontinueduntilthebeginningof 2016andtherelativelylowlevelatwhichthesesubsequentlystabilised.Netenergybillsfellfromanaverageofnearly€17.2 bil‑lion per year between 2010 and 2014 to €12.4 billionin 2015, and to around €6.2 billion in the first ninemonthsof 2016.

On theotherhand, in volume terms,exportsofgoodsandservices(5.8%)increasedquiteabitmorethandidimports (5%). However, both flows were significantlyand quite similarly boosted by the reorganisation ofa large pharmaceuticals company’s activities. Ignoringthis,marketsharewouldhavegrownby0.9%(insteadof3.3%) in 2016,while lossesofmarket sharewouldhave been recorded in 2015 (instead of a small gain).Evenallowingforthisspecificfactor,theimprovementsin exports since mid-2014 are not a typically Belgianphenomenon.ThesamewashappeninginBelgium’skeyEuropean partners, reflecting the favourable effects ofthedepreciation intheeuroeffectiveexchangerate, inparticular the fall against theUSdollar.Wagemodera‑tion in the various countriesmay alsohavehad aparttoplay.

CHART 27 FURTHERRECOVERYINTHETRADEBALANCEASARESULTOFSUSTAINEDEXPORTGROWTHANDPOSITIVETERMSOFTRADE

(inpercentagepointsofGDP,unlessotherwisestated)

2010 2011 2012 2013 2014 2015

–2

–1

0

1

2

3

4

2010 2011 2012 2013 2014 2015–30

–20

–10

0

10

20

30

2015 2016

–30

–20

–10

0

10

20

30

2016 e

First nine months

TRADE BALANCE ACCORDING TO THE NATIONAL ACCOUNTS

Volume effect

Price effect

Change in the balance of goods and services according to the national accounts

p.m. Balance of goods and services according to the balance of payments

(1)

p.m. Balance of goods and services according to the national accounts

(1)

GOODS AND SERVICES BALANCE ACCORDING TO THE BALANCE OF PAYMENTS

(2)

Goods balance (excluding energy products) (3)

Services balance

Energy products balance (3)

Goods and services balance

Sources:NAI,NBB.(1) In%ofGDP.(2) In€billion.(3) Accordingtoforeigntradestatistics–mineralfuels,lubricantsandrelated

materials(SITC-3).

Incontrasttoitsgoodsandservicesbalance,Belgiumsawitsprimaryincomesurplusdeterioratefurtherin 2016inthewakeofinvestmentincomecontinuingitsdownwardtrendatworksince2012.Variousincomecategoriesmak‑ingupthisaggregatehavedeteriorated, inparticular in‑come from foreign direct investment and portfolio invest‑ment.Indeed,whilethecountry’snetexternalpositionis


Income from foreign direct investment, in particular, isdeeplyindeficitinnetterms.In 2015,around€7.6 bil‑lion net flowed abroad, some 1.9% of GDP, whilethis net amount paid to the rest of theworld rose tonearly €9.8 billion in the first nine months of 2016.This outcomemay be explained both by the relativelyweak – and even negative – net external FDI positionand the lacklustre returns that Belgium achieves on its direct investment abroad compared with those gener‑ated by foreign countries on investment in Belgium.Asecond explanation for the country’s subdued invest‑ment income is its international investment position,whichmostly comprises assets in portfolio investmentsand “other investment”, made by the financial sectorin particular. Against the broader backdrop of declin‑ing returnson investmentproductsacross thefinancialmarkets,netincomeonthesetwoinvestmentcategorieshas been falling, partly because Belgium’s paymentsabroadwouldappearlesssensitivetomarkettrendsthanincome received from abroad. Tax considerations andcompositioneffects–includingduration–mightexplainwhyreturnsontheseassetshavedeterioratedmorethanreturns on liabilities.

Secondary income, which includes net governmentpayments to the rest of the world and more specifi‑cally Belgium’s contribution to the general budget ofthe European union and other budgets for international cooperation, isalsoshowingashortfall,which in 2016was slightly greater than in 2015due to a larger con‑tribution to the European budget. Belgium’s higher

CHART 28 INCREASEDEXPORTMARKETSHARE(1)

(volumedataadjustedforseasonalandcalendareffects,averageannualchanges,in%)

2010 2011 2012 2013 2014 2015–4

–2

0

2

4

6

8

BE DE ES FR IT NL

H

2016 e

MARKET SHARE : BELGIUM

Intra‑ euro area

Total

Extra‑euro area

MARKET SHARE : INTERNATIONAL COMPARISON

Average 2010‑2013

Average 2014‑2016 e

Estimate excluding the effect of the reorganisation of a large pharmaceuticals company’s activities

–0.5

0.0

0.5

1.0

1.5

2.0

Sources:EC,ECB,NAI,NBB.(1) Basedonthemostrecentprojectionsforimportdemandfromtradingpartners.

CHART 29 NEGATIVENETINCOMEFROMINVESTMENT

(balancesaccordingtothebalanceofpayments,in€billion,unlessotherwisestated)

2010

2011

2012

2013

2014

2015

–20

–15

–10

–5

0

5

10

15

2015

2016

–20

–15

–10

–5

0

5

10

15

First nine months

Income from direct investment

Income from other investment and reserve assets

Income from portfolio investment

Primary income

Investment income

2016

e

source : NBB.

still exceedinglypositiveatover60%ofGDP, 2016netincomeswerenegativebyaround2.3%ofGDP.


contribution reflects a regularisation of amounts duefor thefiscal years from 2014 in keepingwith the im‑plementation,withretroactiveeffect,oftheEUCouncilofMinisters’agreementonthe 2014-2020multiannualfinancialframework.

2.3 Significantlabourmarketimprovementin 2016

Robust job creation in face of uncertainty

Despite subdued economic growth and uncertainty re‑sulting from the terrorist attacks, Brexit and collectiveredundanciesannouncedinthethirdquarter,the labourmarkethadanotherdynamicyear.Employment intermsof numbers of people and, even more so, in terms oflabour volume grew more than GDP. Average workingtimeperworkerincreasedafterthreeconsecutiveyearsofdeclines.Thismaybeattributable to the introductionofflexiblejobsallowingemployeestodosomeextraworkinhotels,cafésandrestaurants,effectivelyboosting labourvolumes without increasing the number of employedpeople.Apparentlabourproductivity,bycontrast,decel‑eratedin 2016afterhavinggraduallyrevivedasisusualinthefirststageofarecovery.

Table 5 BALANCE OF PAYMENTS AND NET LENDING TO THE REST OF THE WORLD

(balances ; in € billion, unless otherwise stated)

2012

2013

2014

2015

2016 e

1. Current account

Goods and services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −3.1 0.2 1.7 9.3 15.9

Goods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −9.6 −6.6 −4.6 2.6 n.

Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5 6.8 6.2 6.8 n.

Primary income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9 5.4 2.1 −0.9 −4.5

Compensation of employees . . . . . . . . . . . . . . . . . . . . . . . . 5.4 5.5 5.7 5.8 5.9

Investment income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 0.5 −3.1 −6.1 −9.8

Other primary income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.6 −0.6 −0.5 −0.6 −0.6

Secondary income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −6.0 −6.9 −6.4 −6.7 −7.0

General government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −3.1 −3.9 −3.6 −3.9 −4.0

Other sectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.9 −3.0 −2.8 −2.8 −3.0

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.2 −1.2 −2.7 1.8 4.4

p.m. Idem, in % of GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.1 −0.3 −0.7 0.4 1.0

2. Capital account . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 −0.4 −1.1 0.1 −0.7

3. Net lending to the rest of the world (1 + 2) . . . . . . . . . . . . . . . 2.1 −1.7 −3.8 1.9 3.7

p.m. Idem, in % of GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.5 −0.4 −0.9 0.5 0.9

Sources : NAI, NBB.

CHART 30 EMPLOYMENTRISESWHILEPRODUCTIVITYFALLS

(percentagechangescomparedwiththecorrespondingquarterofthepreviousyear,calendaradjusteddata)

2011 2012 2013 2014 2015–2

–1

0

1

2

3

4

Hourly productivity

Domestic employment

Average working hours

p.m. Real GDP

2016

e

Sources:NAI,NBB.


The labour market remained generally lively through‑out the year, albeit that the March terrorist attacksslowed the demand for agency workers, particularlyin Brussels. However, business leader sentiment onemployment prospects subsequently picked up, particu‑larly in Flanders, in both manufacturing and services.

Restructuring programmes announced in september and October had no impact on the year’s employmentfigures: due to Belgium’s lengthy collective redundancyprocedures,theirpotentialeffectsareunlikelytobecomevisible before 2017. In fact, robust employment growthcombinedwithemergingshortagesinsomepartsofthe

CHART 31 TERRORISTATTACKS’TEMPORARYEFFECTSONLABOURMARKETDYNAMICSATTHEBEGINNINGOFTHEYEAR

JJJ

J

JJJ

J

JJJJ

JJJ

J

2012 2013 2014 2015 2016–25

–20

–15

–10

–5

0

5

J

JJJ

J

JJJ

J

J

J

JJJJ

J

2012 2013 2014 2015 2016–10

–5

0

5

10

15

20

2012 2013 2014 2015 2016–20

–15

–10

–5

0

5

10

15

20

25

EMPLOYMENT FORECASTS ACCORDING TO BUSINESS SURVEYS (2)TEMPORARY AGENCY WORK (1)

Flanders

Wallonia

Brussels

MANUFACTURING INDUSTRY

Smoothed series

Gross series

Flanders Wallonia

BUSINESS SERVICES

Sources:Federgon,NBB.(1) Year-on-yeargrowthofhoursworkedperday;monthlydata,three-monthmovingaverage.(2) Balanceofreplies,seasonallyadjusteddata.

Table 6 LABOUR SUPPLY AND DEMAND

(in thousands of people, unless otherwise stated)

2012

2013

2014

2015

2016 e

Level 2016 e

Working‑age population (15‑64 years) . . . . . . . . . . . . . . . . . . . . . . 22 12 9 16 18 7 302

Labour force (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 9 32 22 34 5 291

Domestic employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 −15 19 42 59 4 661

Self‑employed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 6 7 11 15 782

Employees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 −21 12 31 45 3 879

Sectors sensitive to business cycle (1) . . . . . . . . . . . . . . . . . . . . −4 −25 −1 16 28 2 398

Public administration and education . . . . . . . . . . . . . . . . . . . . 2 3 6 1 1 807

Other services (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1 7 14 16 674

Unemployment (3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 25 14 −19 −26 553

Sources : DGS, FPB, NAI, NEO, NBB.(1) All activities not included in the other items, the most important of which are agriculture, industry, construction, trade, accommodation and food service activities,

financial services and business services.(2) Health care and social work ; art, entertainment and leisure ; other services ; and households as employer.(3) Unemployed job‑seekers.


labour market should keep the net effect limited as these arehighly-skilledemployeeswhoshouldfindemploymentelsewhere.

The demand for labour received fundamental support fromcontinuedeconomicactivitygrowth,wagemodera‑tionandreducedwagecostsresultingfromthetaxshift.At the same time, the supplyof labourwasboostedbythe tax shift in the shape of lower fiscal and parafiscalpressureforthose inwork,unemployment insurancere‑forms and end‑of‑career measures.

In 2016,atotalof59000 additionaljobswerecreated,as against 42000 in 2015. Salaried employment inbusiness-cycle-sensitive sectors (including manufactur‑ingandbusinessservices)andin“otherservices”madesignificant advances, confirming the recovery that gotunderwayin 2014.Stabilisationofthenumberofmanu‑facturing jobs in 2016 constituted a clear break withthe downward trend previously seen in the sector, asinagricultureandthefinancialsectortoo.Theemploy‑mentgainsof thepast twoyearshavebeenfuelledbywage-earners inbusiness-cycle-sensitive sectors (43%),by other services, particularly health care and socialservices (30%), and also by the self-employed (25%).Public administration and education, by contrast, haveseenvirtuallyno increase inemployeenumbers,mainlybecause of government cost‑cutting.

Self-employment dynamics and their determinants arebest studiedusingdetailed information collectedby thecountry’sNational Institute for theSocialSecurityof theSelf-employed(NISSE),asitregisterseveryoneinBelgiumwithself-employedstatus,whetheritbeastheirprincipaloccupation, secondary occupation or even after retire‑ment.Beingself-employedturnsouttohavegreaterap‑pealforwomenandyoungerpeople,andtherelaxationof the rulesallowinga combinationofpensionbenefitsand professional income has also had its effect, as thismakes it easier for retired people to continueworking.Beingself-employedinasecondaryoccupationislikewiseverypopular,enablingemployeestolockinalltheadvan‑tagesoftheiremployeestatus–e.g.timecredit,parentalleave–whilemakingextramoneyontheside.Combiningthetworeducesincomeriskcomparedwithentrepreneur‑shipasaprincipalactivity.Self-employmentalsoprovidesdirectaccesstothelabourmarketformanyfindingithardtofindjobsasemployees.

Unemployment decline spreads

Whileemploymenthasgrown,thenumberofunemployedjob-seekersfell in 2016forthesecondyearinarow,by26000.Yet thenumberofunemployed job-seekers still

CHART 32 UNEMPLOYEDJOB-SEEKERNUMBERSDOWNBYREGION,AGEANDUNEMPLOYMENTDURATION

(changesinthousandsofpeoplecomparedwiththesamemonthofthepreviousyear)

2008 2009 2010 2011 2012 2013 2014 2015 2016–60

–40

–20

0

20

40

60

80

2008 2009 2010 2011 2012 2013 2014 2015 2016–60

–40

–20

0

20

40

60

80

2008 2009 2010 2011 2012 2013 2014 2015 2016–60

–40

–20

0

20

40

60

80

BY REGION

Brussels

Flanders

Wallonia

BY AGE

Under 25

Between 25 and 49

50 and over

BY UNEMPLOYMENT DURATION

Under three months

Three months to less than one year

One year to less than two years

Two years and over

source : NEo.


peopleafterthreeyearsoutofwork.Thosepeoplethenhave the possibility of claiming social benefits from apublicsocialassistancecentre.SinceJanuary 2015,whenthe effects of the reformmeasures first became visible,there has been a clear increase in the number of subsist‑encebenefitclaimants.Thosewhowishtoqualifyforthisschememustre-registerasjob-seekers,butaproportionoftheunemployedwhoserightslapsedisappearfromtheworkforce:theysimplyceasetobeamongthoseinworkorseekingwork.

Also,allagebracketshaveseen joblessness fall,butwhilefor people under the age of 24 it started to come downfrom 2014, the next bracket up – i.e. unemployed job-seekersbetween25 and49 –didnot join the trenduntilayear later.Atthebeginningof 2016,thenumberofun‑employedjob-seekersinthe50+bracketclearlyfellforthefirst time since 2000: even when unemployed job-seekernumbersveeredsharplydownin2008 or,toalesserextentalsoin2011,unemploymentamongpeopleovertheageof50 continuedtoswell.Thiswastheoutcomeofaseriesofconsecutivechangesintherulesgoverningearlyretirement.From now on, all jobless people, including thosewith anemployertop-upandirrespectiveoftheirage,arerequiredtobeavailableforworkandactivelylookforemployment.However, some exemptions, transition measures and an“adjustedavailability”optionwereimplemented.Conditions

averaged553000 intheyear,significantlyhigherthanthepre-great-recessionnumbers(500000 in2008).

In 2016,thenumberofunemployedjob-seekersdeclinedin all threeof the country’s Regions, in all agebracketsandforallunemploymentdurations.

InWallonia,thisfigurehasbeenfallingmorerapidlythanintheothertwoRegionsforthepasttwoyears,buttheunemploymentrateremainsalotlowerinFlanders,whereitaveraged5.1%inthefirstthreequartersof 2016,com‑paredwith10.8%inWalloniaand17.2%inBrussels.

The decline in numbers of unemployed job-seekers ismostly the result of lowermedium-term and long-termunemployment, i.e. periods of between three monthsand two years.Very long-termunemployment,which isdefinedas lasting twoyearsor longer, also camedownin 2016, a trend that started at the end of 2015 andlasted throughout the year. The long-term unemployedaretypicallytheleastconnectedtothelabourmarketandthehardest to re-employ,andmeasures takenbypublicemploymentservicestoactivateandeducatethisparticu‑lar groupof unemployedpeople have helped somegetback intowork.Yet, thefall in theirnumberschiefly re‑flectsthethree-yearcaponunemploymentbenefitssince1 January2012,leadingtotheexclusionofunemployed

CHART 33 SIGNIFICANTDROPINBENEFITCLAIMANTSEXEMPTFROMLOOKINGFORWORK

(inthousandsofpeople)

2000

2002

2004

2006

2008

2010

2012

2014

2016

0

100

200

300

400

500

600

2000

2002

2004

2006

2008

2010

2012

2014

2016

0

50

100

150

200

250

300

Job‑seekers

Non‑job‑seekers

Social and family reasons

Older unemployed people

Unemployment with employer top‑up

After voluntary part‑time work

BENEFIT CLAIMANTS NON‑JOB‑SEEKING BENEFIT CLAIMANTS

sources : NEo.


werealsotightenedforthoseseekingexemptionsforsocialandfamilyreasons.Thereformshavecontributedtoadropinthenumberofunemployedexemptedfromseekingwork,andhaveexpandedthesupplyoflabour.

Thedeclinesinjoblessnessreflecttheeffectivenessofthemeasures that were needed to boost the labour force,as the working-age population was beginning to showcontractinggrowth.Infact,themostrecentdemographicprojectionssuggestthattheworking-agepopulationwillactuallyshrinkfrom 2019.

Migration flows in the second half of 2015 temporar‑ily curbed the slowdown in theworking-agepopulationgrowth: in that year, Belgium received 44800 asy‑lum applications, compared with 17000 in 2014 and18700 in 2016.In 2016,therefugeeeffectonthelabourforcewasveryminorindeed,asthereisatimelagbeforethis group enters the labour market.

Limited increase in employment rate despite faster job creation

While the number of people in work rose by nearly550000 between 2000 and 2016, the employment ratehardlymoved (from65.8%to67.4%)owing to thecon‑comitantincreaseintheworking-agepopulation.Also,whilejob creationhas clearly improved in thepast twoyears, ithasnotquitehealedthelabourmarketpainsthatfollowedthe great recession and the slump Europe suffered in its af‑termath:in2008,theemploymentratehadstoodat68%.

The employment rate’s modest but persistent improve‑ment is primarily attributable to theongoing rise in thenumberofwomenenteringtheworkforce,whereasmenhavebeen facingadrop inemployment since2008.Atthe same time, 55-64-year-olds have also seen a rise intheir employment rate in the past 15 years, from26%in 2000to44%in 2015.

CHART 34 EMPLOYMENTRATEBYGENDER,AGE,EDUCATIONANDNATIONALITY

(in%ofthecorrespondingpopulationaged20 to64)

2000

2002

2004

2006

2008

2010

2012

2014

201

6 (1

)

201

6 (1

)

20

30

40

50

60

70

80

90

2000

2002

2004

2006

2008

2010

2012

2014

20

30

40

50

60

70

80

90

2000

2002

2004

2006

2008

2010

2012

2014

2016

(1)

20

30

40

50

60

70

80

90

2000

2002

2004

2006

2008

2010

2012

2014

20

30

40

50

60

70

80

90

201

6 (1

)

BY GENDER

WomenMen

BY EDUCATION

Medium‑skilledLow‑skilled High‑skilled

BY AGE

30‑54 55‑6420‑29

BY NATIONALITY

Other EUBelgian nationals

Non‑EU

source : Ec.(1) Averageofthefirstthreequarters.


Inthe55-plusbracket,themanychangesintroducedsince2000 toBelgium’snumerousearlyretirementmechanismshaveconspiredwiththeraisingofthestatutoryretirementageforwomentocauseastepchangeinlabourmarketexits. Thenumbersof theseare rapidly fallingacross allage backets and under all schemes, disability excluded.Indeed,employeesovertheageof50 showasurgeinthenumberofdisabilityclaimants,ofaround33%between2010 and 2015,while the self-employed report half asmany. In 2015, therewere a total 218000 50-64-year-oldsinthisscheme,comparedwith112000 in2000.

In the youngest age bracket, employment rates havefallensince2008,asstudyperiodshavegenerallylength‑ened. The financial crisis may have persuaded mostyoungstersthatfurthereducationiskeytogainingentryto the labourmarket,but itmaywellhavediscouragedothers from looking forwork in thefirst place. By con‑tinuingwiththeireducation,youngpeopleinvestintheirfuture,whichisagoodthingprovidedtheypickcoursesthatreflectwhatthelabourmarketneeds.

Theproblemsnon-Europeanworkersfacewhenenteringthe labour market require special attention, with barri‑ers including insufficientknowledgeofoneofBelgium’snational languages,problemswithgettingqualificationsand skills recognised, administrative hurdles, discrimina‑tionandoftenlowerlevelsofeducation.In 2015,45.2%ofthecountry’snon-EUnationalswereinwork,comparedwith68.5%ofBelgians.Unlikeinneighbouringcountries,second-generationworkersinBelgiumwouldappearnottodoanybetterthanfirst-generationimmigrants.

People’s level of education is key both in this case andmoregenerallyforallcategoriesofthepopulation,withtheemploymentrategapbetweenthehighly-skilledandthe low-skilled working out at 36.2 percentage pointsin 2016. On the one hand, stiffer competition withmedium-skilled workers as the labour market polarises,and, on the other hand, the higher skill requirementson the labour market have done nothing to improve the position of the lowest-skilled, who have seen theiremploymentratefallby3.5 percentagepointssincethegreatrecession.Thisgroupoftenfindsithardertoadapttochangesandnewrequirements,frequentlyresultinginunstablejobsandrelatedlossofincome.

The sixth State reform shifted responsibility for a largeproportion of employment support measures to theRegions,whicharenowabletodefinetheirtargetgroupsonthebasisofthespecificfeaturesoftheirown labourmarkets.Chapter3 willdescribethereformscarriedoutby the Flemish Region, which is targeting younger andolderworkers,whileWalloniaandBrusselshaveretained

their focus on the long-term unemployed, older unem‑ployedpeopleandlow-skilledyoungsters.

In December 2016, the government approved a draftLaw on Feasible and Manageable Work with the aimof injectingmoreflexibility intothelabourmarket.Thislarge-scalereformprojectallowsemployers,aftersecto‑ralconsultations,tocalculateaverageworkingtimeperweekonanannualratherthanquarterlybasis,providesforsimplifiedpart-timeworkarrangements,andallowsemployees to savedaysoff ina“careeraccount”.Thenewlegislationenvisagesanextensionofpalliativeleaveandmotivatedtimecredit,andintroducesthepossibilityofemployeesdonatingdaysofftocolleagueswithasickchildwhohaveexhaustedallleaveschemes.Temporaryworkers have the possibility of signing permanentcontracts, thus ensuring continued payment betweentwo assignments. With regard to continuous training,employees will be entitled to five days of training perfull-time equivalent on average, while individual train‑ing accounts could be implemented. The reforms target both employers, enabling them to bolster their com‑petitiveness, andemployees,whose careerswill gradu‑allylengthenonthebackofthesemeasuresaddressinglabourconditionsandthework-lifebalance.

CHART 35 REDUCEDEARLYRETIREMENTFROMTHELABOURMARKETAMONG50-64-YEAR-OLDS

(in%ofthecorrespondingpopulation)

2000 2015 2000 2015 2000 2015 2000 20150

10

20

30

40

50

60

70

Retired

Unemployed with employer top‑up,not seeking workOlder unemployed people not seeking work

50‑54 year

Time credit and career breaks,full‑time

Disabled

55‑59 year 60‑64 year 50‑64 year

Sources:DGS,FPB,NIHDI-INAMI-RIZIV,NEO,PSPS.

labour costs and prices

93Economicandfinancialdevelopments ❙ LABOURCOSTSANDPRICES ❙

3.labour costs and prices

ThemarkedincreaseinBelgianemploymentwenthandinhandwithanextremelymodestadvanceinlabourcosts,with the private sector even seeing hourly wage costscontractslightlyin 2016.At1.8%,however,inflationre‑mainedhighandwasinsharpcontrastwithonlysubduedprice increases in other euro area countries.

3.1 Wagemoderationandtaxshifthave an effect

Awhole rangeofmeasureshavebeentaken in the lastfew years to curb labour cost growth and reduce taxchargeson labour, benefiting employers and employees

Table 7 PRIVATE SECTOR HOURLY LABOUR COSTS DOWN IN 2016

(calendar adjusted data. percentage changes compared with the previous year, unless otherwise stated)

2011

2012

2013

2014

2015

2016 e

Labour costs in the private sector

Gross hourly wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 2.8 2.4 1.0 0.4 1.0

Collectively agreed wages (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 3.0 2.0 0.8 0.1 0.6

Real agreed adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0 0.2 0.1 0.0 0.0 0.0

Indexation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 2.8 1.9 0.8 0.1 0.6

Wage drift and other factors (2) . . . . . . . . . . . . . . . . . . . . . . . . . −0.1 −0.2 0.4 0.2 0.3 0.4

Employers’ social contributions (3) . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 0.1 0.1 0.1 −0.5 −1.3

Social security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.1 0.0 0.0 0.0 −0.2 −1.1

Other contributions (4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 0.1 0.1 0.1 −0.2 −0.3

Hourly labour costs in the private sector . . . . . . . . . . . . . . . . . . . . 2.1 3.0 2.4 1.1 0.0 −0.4

Hourly labour costs. economic concept (5) . . . . . . . . . . . . . . . . . . . . 2.1 3.0 2.4 1.0 −0.1 −0.3

p.m. Unit labour costs in the private sector . . . . . . . . . . . . . . . . . . 2.3 3.5 2.3 −0.2 −1.1 −0.2

Unit labour costs. economic concept (5) . . . . . . . . . . . . . . . . . 2.3 3.5 2.3 −0.3 −1.1 −0.1

Hourly labour costs in the public sector . . . . . . . . . . . . . . . . . . . . . 3.7 3.4 3.1 1.7 1.2 1.7

Of which : indexation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 2.5 2.3 0.0 0.0 1.0

Hourly labour costs in the economy as a whole . . . . . . . . . . . . . . 2.4 3.1 2.6 1.2 0.2 0.1

Sources : FPS ELSD, NAI, NSSO, NBB.(1) Wage increases fixed by joint committees.(2) Increases and bonuses granted by companies over and above those under interprofessional and sectoral collective agreements ; wage drift resulting from changes in the

structure of employment, and errors and omissions ; contribution to the change in labour costs, in percentage points.(3) Contribution to the change in labour costs resulting from changes in the implicit social security contribution rates, percentage points.(4) Actual social contributions not paid to the government, including premiums for group insurance, pension funds or occupational pension institutions, and imputed

contributions, including redundancy pay.(5) Labour costs based on the economic concept do not match national accounts data, as they also include reductions in social security contributions for the relevant target

groups as well as wage subsidies, resulting in a more accurate reflection of the labour costs that companies face.


indexationmechanismsinthegovernmentandprivatesectors until April 2016. Although reactivated after‑wards,index-linkingsystemsdifferfromonejointcom‑mitteetothenextandtheindexjump’seffectscontin‑uedtoshowintheremainderoftheyear.At0.6%inthe private sector, the indexation effect in 2016 waswellbelowtheincreaseinthesmoothedhealthindex,whichstoodat2.1%.

Employers’ social securitycontributionsexertedstrongnegative pressure on labour cost growth in 2016 ofapproximately 1.3 percentage points. This is due tothe fact that a large number of government meas‑ures with an impact on employers’ social securitycontributions came into force under the tax shift andCompetitivenessPact.On1 April 2016,employerssawtheir social contributions’ basic rate cut to 22.65%from 24.92%, while the flat-rate reduction was alsolowered, to €438 from €462.60. Lastly, the govern‑ment raised the lower limit for the structural reduc‑tion in contributions for lowwages and adapted thecalculation parameter so as to widen the scope ofthemeasure.Thegeneralexemptionfrompaying1%of payroll tax for employers in the private sectorwasconverted into lower basic rates for employers’ social

CHART 36 IMPACTOFTEMPORARYSUSPENSIONOFINDEXATIONMECHANISMSONREFERENCEINDEXFORWAGES

(index,2013 =100)

2014 2015 2016

98

99

100

101

102

103

104

105

Health index

Smoothed health index reflecting index jump

Smoothed health index (1)

April 2015 April 2016

Sources:DGS,NBB.(1) Four-monthmovingaverageofthehealthindex.

CHART 37 TAXSHIFTREDUCESSOCIALSECURITYCONTRIBUTIONSATALLWAGELEVELS

(actualemployers’contributionstosocialsecuritybywagelevel,as%ofmonthlygrosswage)

1 50

01

800

2 10

02

400

2 70

03

000

3 30

03

600

3 90

04

200

4 50

04

800

5 10

05

400

5 70

06

000

6 30

06

600

6 90

07

200

7 50

0

9

11

13

15

17

19

21

23

25

27

29

2016

2018

2015

2019

2017

source : NBB estimate.

aswell as the self-employed.Thegovernment’saimhasbeen to retain cost competitiveness relative to the coun‑try’s three main trading partners and narrow the gapbetweengross andnetwages for employees, as highernetpayhelpssupportlaboursupply.

In the private sector, the increase in gross wages perhour worked was held back by the index jump andgovernment-imposed constraints on rises in collectivelyagreed wages, as the government set the maximumincrease at 0.5% of the gross wage bill (including allcharges) and on top of that 0.3% of the wage billin net terms. In practice, however, collectively agreedwages barely moved in 2016, suggesting unconven‑tional take-ups of thismargin after the previous year’swage freeze. In around one-fifth of joint committees,forinstance,luncheonvoucherssawanincreaseintheirnominalvalues,whileemployeecontributionswerekeptunchanged, implying higher net wages for the work‑force. The wage drift reflects such wage adjustments,and this is also influenced by fundamental changes intheemploymentcomposition,suchasanageinglabourforce,higheraveragelevelsofeducationandtheshifttoservicesintheeconomy.

Theindex jumpcameintoforce inApril 2015. Incon‑creteterms,thegovernment’sofficialgaugeforindex-linkingwages, i.e. the average of the health index inthepast fourmonths,wasblockeduntil ithadgrownby 2%, resulting in a suspension of most automatic


contributionsfrom1 April 2016,andtheexemptionofbusinesses from a proportion of payroll tax for nightandshiftworkwasraisedto22.8%from15.6%from1 January 2016,whileitchangedfrom17.8%to25%forcontinuouswork.

The combined measures reduced the actual employercontributionpercentagesignificantly in theyear,kicking inatallwage levelsbutmostparticularly lowwages. Infact,the contribution rate looks set to decline further in 2018and 2019 on the back of additional measures under theCompetitivenessPactandthetaxshift.On 2014structure-of-earningsdata,3.6%ofemployees inBelgiumwereongrossmonthlypaybelow€2000 while63%receivedmorethan € 2 750.

In addition to the tax shift, the sixth State reformalsohad an impact on labour costs. competence for tar‑geted reductions in social security contributions wasshiftedtotheRegionsasearlyas 2014buttheydidnotchange their policies in this area until 2016. Flandersbrought new employers’ contribution reductions intoeffectfrom1 July 2016,focusingparticularlyon55-plusemployees, under-25s and those with disabilities. TheBrussels‑capital Region adapted the targeted reductions forolderemployeeson1 Octoberandwillputtherestoftheadjustmentsinplacein 2017,withmeasuresmostlypertaining to the young, long-term job-seekers andthe low-skilled.Walloniaplans tomoveonthereformsin 2017.Heretoo,reductionswillfocusonthelong-termunemployed,butalsoonemployeesaged55 andover,aswellaslow-skilledyoungpeople.

Factoringinallsocialsecuritycontributionreductionsandwage subsidies affecting labour costs (according to theeconomicconcept),hourlywagecostsintheprivatesec‑torshouldhavecomedownbyatotalof0.3%in 2016,a little more than in 2015. In fact, using the nationalaccounts concept, the drop would have been greaterstill, in particular because of the April 2016 conversionofthegeneralexemptionfrompaying1%ofpayrolltaxinto lower basic rates. Previously, this exemption wasregistered as “other subsidies on production” and notincludedinthewagebill,whereas it isnow.Thiscausesadownwarddistortioninyear-on-yearchangesinlabourcosts,althoughthismethodologicalchangedoesnotalterthe actual labour costs facing employers. Despite theseslightdifferences inmethodologies,bothsetsofcalcula‑tions – exceptionally – show the private sector’s annuallabourcostsslightlydown.

With labour costs up in the public sector, hourly wagecosts across the economy inched up by a slight 0.1%yearonyear.

However,lowerlabourcostsforemployersdidnotmeanthat employees also saw a drop in their net pay. Infact, 2016 netwages rose on the back of a change inthe fiscal work bonus, a higher deductible professionalexpenseallowanceandtax-freeincomethreshold,aswellas tax band reforms. From the 2016 income year (taxyear 2017), the lower threshold for the 30% tax bandwill gradually be raised,with theband set todisappearentirely from incomeyear 2018.Theflat-rateallowancefor professional expenses, already put up in 2015, wasratchetedupagainat thebeginningof 2016,while thefiscal work bonus, a refundable tax credit for low-paidemployees, was also raised again after an earlier risein 2015.

The taxwedge, equal to thedifferencebetween labourcostspaidbyemployersandnetearningsreceivedbyem‑ployees,hasfurthernarrowedonthebackofthesemeas‑ures.Thatsaid,itwouldhaveremainedlargerinBelgiumthaninitsneighbouringcountriesin 2016.

The wage gap accumulated since 1996 was closed in 2016

For Belgium, an open economy whose three largestneighbouring countries also happen to be its three

CHART 38 BELGIUM’STAXWEDGE(1) STILL EXCEEDS THATOFOTHEREUMEMBERSTATESANDITSNEIGHBOURINGCOUNTRIES

(leviesonincomefromlabour,in%oflabourcostsforemployers)

2001 2015 2016 (2) 2001 2015 2001 2015 2001 2015

0

10

20

30

40

50

60

70

be DE FR NL

Employers’ social security contributions

Employees’ social security contributions

Income tax

Source:EC,OECD,NBB.(1) Businesseswithatleasttenemployees.Taxwedgeforasinglepersonwithout

childrenearninganaveragewage.Thiscomparisondoesnottakeaccountofwagesubsidies.

(2) PreliminaryresultforBelgiumin2016.


increases in Germany, France and the Netherlands.According to the Central Economic Council (CEC)Secretariat, hourly wage cost increases in Belgiumwere generally comparable with average trends inthe threeneighbouring countries between1996 and2005. In subsequent years, a wage gap opened up,peakingat4.8%in2008 andgraduallycomingdownbefore disappearing completely in 2016: on a rangeofdifferentCECestimates,itworkedoutatbetween–0.2%and0.0%,mainlythankstowagemoderationmeasures imposedby the government. Belgium’s so‑cialpartnershavingbeenunabletoagreeonawagenorm to guide developments in collectively agreedwagessince2010,thegovernmentsteppedinandseta benchmark itself.

This average reference fixed by law masks the factthat the country has built up a labour cost edge onFrance and theNetherlands over this period,whereasBelgiancompaniesstillshowaclearwagegaprelativetoGermany.

To prevent any future labour costs divergences relativeto Belgium’s three neighbouring countries, the federalgovernmentdrafted a Law in the autumnof 2016 thatproposesanumberof amendments to theexisting Lawon the Promotion of Employment and the PreventiveSafeguarding of Competitiveness dating from 1996(see box 3).

CHART 39 BELGIUM’SWAGEGAPDISAPPEAREDIN 2016

(cumulativedifferencessince1996 intermsofhourlywagecostsintheprivatesectorcomparedwiththethreemainneighbouringcountries,in%)

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

–15

–10

–5

0

5

10

15

Three main neighbouring countries (1)

Germany

France

Netherlands

2016

e

Sources:CEC,NBB.(1) WeightedaveragebasedonrelativesizeofGDP.

maintradingpartners,itisimportantthatwagegrowthshould not deviate too much from average wage


Box3 – DraftLawtoamendthe26July1996LawonthePromotionofEmploymentandthePreventiveSafeguardingofCompetitiveness

TheLawonthePromotionofEmploymentandthePreventiveSafeguardingofCompetitiveness,adoptedin1996,wascreatedbyBelgium’sfederalgovernmenttohelpalignlabourcosttrendswiththoseofthecountry’sthreemain tradingpartners,namelyGermany, Franceand theNetherlands.TheLaw failed topreventBelgium fromseeingawagegapopenupinthe2005-08period,whichitdidnotmanagetocloseuntil2016.Inpractice,the1996Lawdidnotalwaysproveadequateforthetaskofpreventinglabourcostsfromderailing.Forthatreason,thefederalgovernmentproposedaseriesofadjustmentsin2016,definedinmoredetailindraftlegislationputto parliament.

The new Law follows the same basic tenets as its 1996 predecessor. As such, the expected average nominalwageincreaseinthethreemainneighbouringcountrieswillcontinuetobeabenchmarkforthemaximumwageincrease in Belgium; the social partnerswill continue to negotiate thewage norm, and automatic indexationmechanismswillstayinplace.However,ifaccordingtoavailableprojectionsthewagegapthreatenstowidentotheextentthatitcannolongerbeclearedwithinthespaceoftwoyears,thegovernmentwouldbeabletotakeaction–afterseekingtheadviceofthesocialpartners–andrestrictindexation.

Thebiggestchangeinvolvesthemethodologyusedforcalculatingthetwo-yearmaximummarginforrealwagecostgrowth.Henceforth,theCentralEconomicCouncil(CEC)isrequiredtoallowforanadjustmentfactorandasafetymargininadditiontoexpectednominallabourcostdevelopmentsinBelgium’sneighbouringcountriesandprojected indexation inBelgium.Theadjustment factor shouldgivemore roomfornegotiatingwage increaseswheneverthesafetymargin–oraproportionthereof–frompreviousnegotiationshasprovenunnecessaryorifalabourcostadvantagehasbeenfound.If,bycontrast,awagegapisfoundtoexist,theadjustmentfactorwillconstraintheavailablescopefornegotiation.ThenewmethodologyallowsforasafetymargintotakeaccountofanyforecastingerrorsintheexpectednominallabourcoststrendsinBelgium’sneighbouringcountriesaswellasindexationinthecountryitself.Thissafetymarginisputataminimumof0.5percentagepoint.TheseadjustmentswillautomaticallytriggeracorrectioninthemaximummarginwheneverlabourcostdevelopmentsinBelgium’sneighbouringcountriesareoverestimatedand/orinflationunderestimated.Consequently,Belgium’slabourcostsshouldbeunabletodivergefromitsthreeneighbouringcountriesforanylengthytime.

Whencalculatingthemaximummarginavailable,themethodologydoesnotfactor inwagesubsidies,nortax-shift-derivedreductionsofsocialsecuritycontributionsinexcessofthoseresultingfromtheCompetitivenessPact.Thenewset-upwillseetheCECnotjustcalculatingthehourlywagegapsince1996,butalsothehistoricalwagegap,i.e.thewagegaprelativetoBelgium’sthreeneighbouringcountriesbeforethe1996Lawcameintoforce.Aslongasthishistoricalwagegapisnotatnil,aproportionofanyreductionsinsocialsecuritycontributionswillgotowardsclosingthedifference.

Theamendmentstothe1996Lawhavemajorconsequencesforwagegapcalculationsandthereforealsoforthebargainingscopebetweenthesocialpartners.Forinstance,theriseinhourlywagecostsin2016isbiggerwhensocialsecuritycontributioncutsunderthetaxshiftarenotfactoredin.Moreover,theproposedlegislation stipulates that theCEC’s technical reports shouldalso include calculationsof anabsolutewagegap,whichcomparesBelgium’swagelevelswiththoseinitsneighbouringcountries,alongsideanabsolutewagegapfactoringinanydifferencesinproductivity.Inaddition,theCECalsohastocalculateawagegapthat takes on board all reductions in social security contributions and wage subsidies in Belgium and itsneighbouringcountriessince1996,andthesocialpartnerswillberequiredtofactorintheseparticularwagedifferentialsaswellwhenengagingincollectivewagebargaining.Allthatsaid,thekeyreferencewagegapwillremainthesame:thedifferenceinhourlywagecostgrowthsince1996.Incidentally,thewagenormasagreedbythesocialpartnerswillbebindinginthecollectivelabouragreementoftheNationalLabourCouncil(NLC),whichshouldgiveitstronglegalforceandmakeanybreachesanoffencepunishablebylaw.


doesnottakeaccountofproductivitytrends.Butthisisakeyfactorinanyassessmentoflabourcostgrowth.Fromaneconomicperspective,afterall,amorerapidincreasein hourly wage costs may well be justified if it reflectsgreaterproductivitygains.Unitlabourcosts,bycontrast,capture both sets of developments, and these fell inBelgiumin 2014andagainin 2015,beforelevellingoutin 2016.Infact,forunitlabourcosts,thegapnarrowedfurther as its neighbouring countries recorded steeper risesthandidBelgium.Thatsaid,Belgiumhasnotman‑agedtoclearthewagegapithasbuiltupsince1996,butitisentirelydowntounfavourabledevelopmentsrelativetoGermany.

Movements in hourly wage costs nevertheless remainakey indicator, asbusinessesmay try tooffsethourlywage costs that get out of hand through investmentthat raises capital intensity and consequently reducesthenumberofworkersneeded intheproductionpro‑cess. Of course, such measures could sharply reduceunitlabourcosts,butatthesametimewouldhaveanadverseeffectonemployment.

If trends in labour costsperhourworkedandperunitproduced in the Belgian business sector during the re‑cent period of strongwagemoderation are comparedwith the samefigures for the 1997-2008 period – thefirst set of wage moderation measures were taken in2009 and immediately cut social security contributionssignificantly, possibly causing comparative distortionsfromthatyear–itcanbeseenhoweffectivewagemod‑erationpolicieshavebeenforlabourcostgrowthacrossallsectors.Between1997 and2008,bothhourlywagecosts and unit labour costs recorded average annual increasesacrossallsectorswellinexcessoftheaveragegrowthfor 2014and 2015.FortheBelgianbusinesssec‑torasawhole,unitlabourcostsaddedanaverage1.3%perannuminthe1997-2008 periodwhiletheycontract‑edby anannual1.2%onaverage in 2014and 2015.This contraction is due to declining unit labour costs in manufacturing industry and construction, aswell as tothemarkedslowdownofthesecostsinmarketservices,thankstovirtuallyunchangedhourlywagecosts.

CHART 40 WAGEGAPFORUNITLABOURCOSTS(1) NARROWINGSINCE 2014

(cumulativedifferencesince1996 forthebusinesssectorinBelgium (2),in%)

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

–20

–10

0

10

20

30

Germany

France

Three main neighbouring countries

Netherlands

source : Ec.(1) Apositivesignimpliesthatunitlabourcostsandhourlylabourcostsarerising

faster in Belgium than the average for its three main neighbouring countries.(2) ThebusinesssectorcomprisesNACEcategoriesBtoNandincludesindustry,

constructionandmarketservices,servingasaproxyfortheprivatesector.

In January 2017, the CEC drew on these new provi‑sions topegat1.1% themaximumavailablemarginfor real wage increases in the 2017-18 period. Thismargin was taken up by the social partners in thedraft interprofessional agreement. Thedraft Lawalsotasks the CEC with calculating the historical differ‑ence for labour costs fromabaseline in1996. Thesecalculations were not yet available when this Reportwenttopress.

Thewagegap,whichhasupuntilnowservedasarefer‑ence benchmark in collective bargaining negotiations,


3.2 PersistentlyhighinflationinBelgium

Services inflation remained high in Belgium, in contrast with elsewhere in euro area

Formanycompanies,thecostoflabouraccountsforasig‑nificantproportionoftotalproductioncostsandcanmakea key difference in pricing. Thewagemoderation policiesof the recent past should therefore curb price trends in Belgium,inparticularservicesprices,aslabourcoststypicallytake up a greater proportion of production costs for services thanforothercomponentsofthepriceindex.Contrarytoassumptions,however,priceshavecontinuedtostageunin‑terruptedhighgrowthinBelgium’sservicessector.

In 2016, services inflation worked out at 2.2% andcontributed 0.9 percentage point to headline inflation,whichstoodat1.8%.The twopreviousyearshadseen

servicespricesmoveaheadatalmostthesamepace,by2.4%in 2015and2.2%in 2014.Bycontrast,Belgium’sneighbouring countries have been reporting systemati‑callylowerservicespricesandevenafurtherslowdown:1.6%and1.3%in 2014and 2015respectively,andthendownto1% in 2016.Averageservices inflationfor theeuroareawasthesame:around1%in 2016.Againstthisbackground,a taskforceofFPB,FPSEconomyandNBBexperts was established in mid-2016 to investigate theunderlyingcausesofthecountry’shighservicesinflation.Itwill report itsfindingstotheMinisteroftheEconomyin 2017.

Services sector companies find their costs not drivenjust by labour, but also – as in other sectors – by in‑termediate consumptiongoods,whether importedornot. Examples include fuels, a key input in transportservices,andfoodinaccommodationandfoodserviceactivities.Companies’profitmargins are alsoamajor

CHART 41 LABOURCOSTSFELLINMOSTBRANCHESACROSSBUSINESS

(compoundannualgrowthrate,in%)

–6

–4

–2

0

2

4

6

–6

–4

–2

0

2

4

61997-2008

Busi

ness

sec

tor (1

)

Of

whi

ch :

Indu

stry

and

ene

rgy

Con

stru

ctio

n

Mar

ket

serv

ices

Of

whi

ch :

Trad

e, r

epai

rs, t

rans

port

, acc

omm

odat

ion

and

food

ser

vice

act

iviti

es

Info

rmat

ion

and

com

mun

icat

ion

Fina

ncia

l and

insu

ranc

e ac

tiviti

es

Oth

er m

arke

t se

rvic

es

Of

whi

ch :

Real

est

ate

activ

ities

Inde

pend

ent

prof

essi

ons,

aca

dem

ic a

nd t

echn

ical

activ

ities

, adm

inis

trat

ive

and

supp

ort

serv

ices

Hourly labour costs

Hourly labour productivity (2)

Unit labour costs

2014-2015

Busi

ness

sec

tor (1

)

Of

whi

ch :

Indu

stry

and

ene

rgy

Con

stru

ctio

n

Mar

ket

serv

ices

Of

whi

ch :

Trad

e, r

epai

rs, t

rans

port

, acc

omm

odat

ion

and

food

ser

vice

act

iviti

es

Info

rmat

ion

and

com

mun

icat

ion

Fina

ncia

l and

insu

ranc

e ac

tiviti

es

Oth

er m

arke

t se

rvic

es

Of

whi

ch :

Real

est

ate

activ

ities

Inde

pend

ent

prof

essi

ons,

aca

dem

ic a

nd t

echn

ical

activ

ities

, adm

inis

trat

ive

and

supp

ort

serv

ices

Sources:EC,NAI,NBB.(1) ThebusinesssectorcomprisesNACEcategoriesBtoNandsoincludesindustry,constructionandmarketservices,servingasaproxyfortheprivatesector.(2) ApositivesignimpliesthatlabourproductivityinBelgiumisfalling.


factor influencing sales prices. They will notably de‑pend on how intense the competition is: the fiercerthe competition, the lower the sales price must be,reducingprofitmargins forbusiness.Thestateof theeconomyalsohasaparttoplay:inflationtypicallyriseswhen theeconomy isat the topof its cycle,boostedby greater demand for consumer goods as well asstrongupwardpressureonwages.Lastly,governmentdecisions also have amajor impact on the pricing ofcertain services.

Belgium’s high services inflation suggests that a de‑tailedreviewofpricetrendsforthisparticularcategoryis in order. Phillips curves capture the correlation be‑tween macroeconomic variables and price increases.Ouranalysisestimatedservicesinflation,eachtimeusinga model of three macroeconomic variables. Its condi‑tional forecasts for services inflation from 2012 – i.e.theyear inwhichBelgium’sservices inflationstartedtodeviate from that of its neighbouring countries and the euro area – suggest that recorded services inflation inBelgiumhasbeensignificantlyhigherthanpredictedonthe basis of macroeconomic conditions since the end of 2014. It would appear that determinants that usedtopredicta largeproportionofservices inflation inthepastnolongerapply.Thesameconclusionemergesfor

CHART 42 SERVICESINFLATIONREMAINSHIGHINBELGIUM,WHILESLOWINGINITSTHREENEIGHBOURINGCOUNTRIESANDTHEEUROAREA

(annualchanges,in%)

2010 2011 2012 2013 2014 2015 20160.0

0.5

1.0

1.5

2.0

2.5

3.0

p.m. Belgium – services inflation disregarding government decisions

Euro area

Belgium

Three neighbouring countries

Sources:EC,NBB.

CHART 43 UNLIKETHEEUROAREA,SERVICESINFLATIONINBELGIUMEXCEEDSPROJECTIONSBASEDONMACROECONOMICVARIABLES

(annualchanges,in%)

2012 2013 2014 2015 20162012 2013 2014 2015 20160.0

0.5

1.0

1.5

2.0

2.5

3.0EURO AREA

0.0

0.5

1.0

1.5

2.0

2.5

3.0BELGIUM

Services inflation excluding government measures (1)

Services inflation projected by Phillips curve models (2)

Recorded services inflation

Sources:ECB,NBB.(1) Projectionofservicesinflationtrend,notincludingthefollowinggovernmentmeasures:theincreaseinsewagechargesinJanuary2012,January 2014andJanuary 2015;

theincreaseinwastecollectionchargesinMay 2014andMay 2015;theVATriseforcableTVsubscriptionsinJanuary2012;theimpositionofVATonnotaries’feesfromJanuary2012;higherratesformedicalservicesresultingfromaharmonisationofpatientfeespaidtomedicalspecialistsinJanuary 2015;andsteepertuitionfeesforhighereducationinFlandersinOctober 2015.

(2) ServicesinflationcalculatedusingPhillipscurvemodelsonarangeofspecificationsandincorporatingthefollowingvariables:unitlabourcosts–which,inadditiontowagecostsperhour,alsotakeintoaccountproductivity–intheservicessectors;realGDP;importpricesforcompetitorsfromoutsidetheeuroarea;pastservicesinflation;consumerconfidence;oilprices;unemploymentrate;andinvestmentvolume.Thedatarelatestotheperiodfromthefirstquarterof1998 tothesecondquarterof 2016.Conditionalforecastsofservicesinflationstartfromthefirstquarterof2012.


aconditionalforecastbeginningin2009,atthestartoftheperiodofwagemoderation.Factorsotherthantheeconomicenvironmentmusthavepushedservicesinfla‑tiontoexceptionalhighsinrecentyears.

Theservicesinflationgapthathasopenedupsince2012between Belgium and its neighbouring countries andwiththeeuroareaisnotmerelyattributabletoBelgianpriceincreasessimplybeing“toobig”.Asimilaranalysisfor theeuroareasuggests thatactual services inflationworks outwell below predictions based onmacroeco‑nomicmodels. That said, forecast spreads for theeuroareaareratherwide,makingforlessreliableoutcomes.

Belgium’s higher services inflation is due, among otherfactors, to the indexation of a range of services prices.Drawingon the consumerprice indexorotherprice in‑dicators, thesepricesareadjustedonceor several timesa year, frequently at fixed intervals – e.g. rent, publictransport rates and fire insurance premiums. This typeof indexationbasedon recentprice trends typically cre‑atesinflationpersistence.Thesemechanismsshouldhave

causedservicesinflationtofallinthewakeoflowerinfla‑tion in 2014 and 2015. However, the declinewas onlyverysmall.

Meanwhile, a few key telecommunications operatorsraisedtheirratesinFebruary 2015andatthebeginningof 2016, pushing up 2016 inflation in this categoryto4.2%onaverage.As thiscategoryclaims6.5%ofservices,itcontributed0.3 ofapercentagepointtoser‑vices inflation. Ithadhelpedslow itdown in thepast,albeit toa lesserextent than inBelgium’s threeneigh‑bouring countries:between2012 and 2016, telecomsshavedanaverage0.04 percentagepointsofftherateofincreaseinservicespricesinBelgium,comparedwith0.19 percentagepointsforitsneighbouringcountries.

Governmentmeasures also had a significant impact onservicesinflation.In 2016,highertuitionfeesinFlanders(0.6%of thehouseholdconsumerbasket)addeda sig‑nificant0.4 ofapercentagepointtoservicesinflationupto and including September. Increased excise duties onalcoholandtheintroductionoftheso-called“sodatax”on soft drinks containing sugar – measures taken as part ofthetaxshift–pusheduppricesincafésandrestaurantsevenfurther,to3.2%from2.7%in 2015.Thelegalob‑ligation tokeepa registered till fromJanuary 2016maywellhavehadsomethingtodowithpricerisesinthehos‑pitalityindustry,butitisdifficulttofactoroutthismeas‑ure fromallotherpricechanges.Restaurantsandcafés,whichaccountfor6.4%oftheconsumptionbasket,werethebiggestcontributorstoservicesinflationin 2016,asinpreviousyears.Theircontributionworkedoutataroundhalf a percentagepoint, comparedwith0.4 percentagepointin 2015.

Inrecentyears,arangeofgovernmentmeasureshadal‑readybumpedupservicesinflation:in2012,thegovern‑ment raised indirect taxes on certain services, increasedsewagechargesin 2014and 2015andputupchargesformedicalservicesin 2015.

Ifallthesemeasuresarefactoredout,pricetrendsintheservices sectors are a much closer match to the estimated Phillips curves, but still higher than might be expectedon themajorityof the relevant factors. Itwouldappearthat services inflation in Belgium, excluding the impactof government decisions, is constantly at around 2%,independently of the macroeconomic context. To anextent, thismightbeexplainedby the fact that servicescompanies face less foreign and domestic competition than businesses in other sectors.

The report to be submitted by the task force to theMinisteroftheEconomyshouldprovideamorein-depth

Table 8 ESTIMATED IMPACT OF SOME GOVERNMENT MEASURES ON HEADLINE INFLATION

(in percentage points)

2014

2015

2016

Tax shift-impacted categories (1) . . −0.1 0.1 0.4

Food . . . . . . . . . . . . . . . . . . . . . . 0.1 0.1 0.2

Tobacco . . . . . . . . . . . . . . . . . 0.1 0.1 0.1

Soft drinks . . . . . . . . . . . . . . . 0.0 0.0 0.0

Alcoholic drinks . . . . . . . . . . 0.0 0.0 0.1

Energy . . . . . . . . . . . . . . . . . . . . −0.3 0.0 0.3

VAT on electricity . . . . . . . . . −0.3 0.0 0.2

Motor fuels . . . . . . . . . . . . . . 0.0 0.0 0.1

Inflation impact of other government measures (2) . . . . . . . 0.1 0.3 0.5

Electricity . . . . . . . . . . . . . . . . . . 0.0 0.1 0.4

Higher education tuition fees in Flanders . . . . . . . . . . . . . . . . . 0.0 0.0 0.1

Harmonisation of patient fees paid to specialists . . . . . . . . . . . 0.0 0.1 0.0

Higher charges for waste collection and waste water treatment . . . . . . . . . . . . . . . . . 0.0 0.1 0.0

Total . . . . . . . . . . . . . . . . . . . . . . . 0.1 0.4 1.0

Sources : EC. NBB.(1) The tax shift was announced at the end of 2015 and concerned all categories

listed. Some of these had faced earlier measures. which are also included in this table.

(2) Non-exhaustive list of measures. The table only includes measures with headline inflation impacts of at least 0.1 percentage point.


analysisofthecausesthatunderpinhigherservicesinfla‑tion in Belgium. It will discuss the telecoms industry aswell as restaurants and cafés in greater detail, as thesehave contributed most to the inflation gap relative toBelgium’sneighbouringcountries.

Other inflation components also affected by measures

The government has taken a series of measures to cush‑ion the adverse effects of the tax shift on its budgetandthesehavefedintoinflationsincetheendof 2015.First, inNovember 2015,excisedutiesonalcoholwereraised significantly and from that same month thegovernmentgradually increased exciseduties ondieselusingtheratchetsystemwhileatthesametimecuttingback excise duties on petrol, albeit to a lesser extent.InJanuary 2016, it raisedexcisedutiesontobaccoandintroduceda“soda tax”.Exciseamountsare set togoupevenfurther intheyearsaheadandthesemeasuresarelikelytohaveaninflationimpactatleastuntil 2019.Lastly, VAT on electricity was put back up to 21%in September 2015 after having been cut to 6% inApril 2014.

Yet other measures triggered a surge in energy prices– morespecificallyofelectricity–asBelgium’sutilitycom‑panies became subject to corporation tax in mid-2015and passedmost of the extra charges onto consumers.In Flanders, meanwhile, a so-called “prosumer rate”wasputintoplace,imposingachargefortheuseofthepowergridonconsumersgeneratingaproportionoftheir

own(green)energy.Andthen, inMarch 2016,Flandersmassivelystokeduptheenergy levy fromaround€3 to€100 per annum per average household. The FlemishRegionalsogot ridof the freebasicelectricitypackage,whichshowedupintheindexfromMay 2016onwards.Thesemeasures,coupledwiththeVATriseunderthetaxshift, pushed electricity prices up by an average 28%in 2016.

InNovember 2015,anewEuropeanemissions standardfor vehicles came into force (switchover fromEuro5 toEuro6).Thetechnicalchangesrequiredtomeetthenewstandardwerealsopassedoninthepricesthatenduserspayfortheirnewcars.Whatismore,FlandersraisedthevehicleregistrationandlicensingtaxfromJanuary 2016,adding 0.1 of a percentage point to price increases fornon-energyindustrialgoods.

Lastly, partners in the Belgian agri-food chain platformagreed to support the country’s dairy farmers: betweenSeptember 2015andtheendofMay 2016,thedistribu‑tionsectorput14 eurocentsper litreofmilksold intoaspecial fund. Thiswas passed through to the consumerpriceofmilkinfullandthuswentstraighttotheinflationfigures.Itwasnotuntilafewmonthsaftertheagreementhadendedthatmilkpricescamebackdown.

And so, Belgium’s inflation gap with its three neighbouring countries has widened again

Headline inflation in Belgium worked out at 1.8%in 2016,upfrom0.6%ayearearlier.Itsaccelerationwas

Table 9 HARMONISED INDEX OF CONSUMER PRICES

(percentage changes compared with previous year)

Three neighbouring

countries

2013

2014

2015

2016

2016

HICP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 0.5 0.6 1.8 0.3

Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −4.6 −6.0 −8.0 −0.6 −4.4

Food . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 0.8 1.8 3.1 0.9

Underlying inflation (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 1.5 1.6 1.8 0.9

Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 2.2 2.4 2.2 1.0

Non energy industrial goods . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.0 0.3 0.5 1.0 0.6

p.m. Health index (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 0.4 1.0 2.1 −

Source : DGS, EC.(1) Total harmonised index of consumer prices, excluding the volatile food and energy components.(2) National consumer price index, excluding products considered to damage health, i.e. tobacco, alcohol and motor fuels.


largelydowntoalessrapidfallinenergypricesbecauseof the measures we have reviewed – and the higherelectricitypricetheyresultedin–aswellastohigheroilprices (Brent)making inflation fromoil-derivedproductslessnegativethanin 2015.Underthetaxshiftmeasuresand themilk agreement, food also helped to speed upinflation,andproductpricesrecordedaverageupticksof3.1%,from1.8%in 2015.AwideruseofscannerdataforpriceindexcalculationssinceJanuary 2016,andprinci‑pallyforfoodproducts,alsoseemstohavehadtheeffectof pushing up food prices.

Underlying inflation – i.e. headline inflation excludingthe volatile components foodandenergy–workedoutat1.8%in 2016,nolongerexceedingheadlineinflationfor the first time since 2012. As noted, services pricescontinuedtorisefast. Inflationfornon-energy industrialgoods,whichaccountsforasmallerweightinunderlyinginflation, stood at 1%, compared with 0.5% in 2015,and largely tied in with the measures making vehiclepurchasesmore expensive (Euro 6 standard and vehicleregistrationandlicensing).

The range of developments and measures as described here also explain the significant 1.4 percentage point

inflationgapthatexistedbetweenBelgiumandthethreeneighbouringcountriesin 2016.Thekeydifferenceswereattributable to electricity, processed food and services,three categories whose prices were heavily influencedbygovernmentmeasures. Pricesofoil-derivedproducts,by contrast, served to narrow the inflation gap withBelgium’s three neighbouring countries, as changes inBrent oil traditionally translate more strongly into con‑sumer prices than in other countries. In Belgium, com‑modities account for a much larger proportion of the price that final consumers pay, in part because of verylowexcisedutiesonheatingoil.Asa result,Brentpriceupsanddownshaveamuchbiggerimpactoninflation.Recentyears’historicallylowBrentprices,whichreachedtheirnadiratthebeginningof 2016,mitigatedtheimpactofthestarkincreasesinelectricityprices.

Bytheendoftheyear,theinflationgaphadnarrowed,asseveralkeymeasures–suchastheelectricityVAThikeandhighertuitionfeesinFlanders–werenolongerreflectedintotalpricetrends.However,theincreaseinBrentpriceskept Belgium’s inflation figure high. In its neighbouringcountries,moreexpensiveoilpushedupinflationto0.8%inthefinalquarterof 2016from0.1%inthefirst.Asimi‑lartrendwasreportedfortheeuroareaatlarge.

CHART 44 ELECTRICITY,PROCESSEDFOODANDSERVICESCONTRIBUTEDTOTHEINFLATIONGAPBETWEENBELGIUMANDITSTHREENEIGHBOURINGCOUNTRIES

(inpercentagepoints)

2010 2011 2012 2013 2014 2015 2016

–2.0

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

–2.0

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

Gas and electricity

Oil-derived products

Unprocessed food

Processed food

Services and non-energy industrial goods

Total

source : Ec.


Thehealth index, i.e. thenational consumerprice indexexcludingalcohol,tobaccoandmotorfuels,roseby2.1%in 2016,asagainst1%inthepreviousyear.Thisindicatorisusedforcalculatingwageindexation.

One-off factors exerting a temporary and direct influ‑enceoncertaincomponentsof inflationcombinedwithmorestructuralfactorshavekeptinflationhigh.Itshouldbenoted,however, thatnon-recurringfactorsmayhavewhat are called second-round effects: services whoseprice trends are somehow linked to a price indicatormaypushupinflationfurther.Andautomaticindexation

mechanisms imply that higher prices may drive wageincreases,whichinturncanpushpricesupevenfurther.Andso,an initial, temporaryanddirect impactcanturnstructuralafterall.Suchwage-pricespiralsarelesslikelyinBelgium’sthreemainneighbouringcountries,asthesedonot apply automatic indexation. Consequently, govern‑mentmeasuresthatservetoraisepricesmayenduperod‑ingthecountry’scompetitiveness.Byintroducinganauto‑maticadjustmentintheeventofunexpectedderailments,theamended1996 LawonthePromotionofEmploymentand the Preventive safeguarding of competitiveness might help to prevent this from happening.

SavingsandfinancingtheBelgianeconomy

107Economicandfinancialdevelopments ❙ SAVINGSANDFINANCINGTHEBELGIANECONOMY ❙

4. savings and financing the Belgian economy

4.1 financing conditions support transactionsbynon-financialcorporations

Belgian businesses expanded their assets again in 2016

The year 2016 saw Belgium’s corporations draw onenhancedprofitabilityandlowerfundingcoststoboostproductioncapacity,andtheirfixedcapital investment

remained particularly high in the first ninemonths ofthe year. This worked out at €41.6 billion after al‑ready touchinghigh levels in2015.Meanwhile, somecorporationswentdownthemergersandacquisitionsroutetogrowtheiractivities,insomecasesalsobytak‑ing stakes in the assets of non‑resident corporations. Cross-border transactions have bolstered outwarddirect investment since 2015 : recent developments are the subject of box 4. SABMiller’s acquisition byAB InBevinvolvedexceptionallyhighamountsin2016,

Table 10 NET TRANSACTIONS BY NON‑FINANCIAL CORPORATIONS

(consolidated data, in € billion)

First nine months

2011

2012

2013

2014

2015

2015

2016

Non‑financial assets . . . . . . . . . . . . . . . . . . . . . . . . . 56.3 54.4 53.3 58.4 59.5 45.2 41.5

Gross fixed capital formation . . . . . . . . . . . . . . . 50.9 52.5 52.7 55.9 57.8 41.3 41.6

Other non‑financial assets (1) . . . . . . . . . . . . . . . . 5.5 1.9 0.5 2.4 1.7 3.9 −0.1

Financial assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.5 63.3 0.9 42.9 18.1 24.8 47.7

Currency and deposits . . . . . . . . . . . . . . . . . . . . . 1.4 −8.8 5.9 −7.5 −1.5 1.8 11.0

Intra‑group assets (2) . . . . . . . . . . . . . . . . . . . . . . . 102.9 67.4 −0.8 57.4 18.9 22.9 −4.4

Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 4.6 −4.3 −7.1 0.7 0.1 41.0

Financial liabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . 114.1 52.3 1.0 45.9 24.7 28.7 28.7

Bank loans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 −1.7 5.7 −2.7 0.2 1.1 2.1

Debt securities . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 6.3 5.6 4.8 5.4 2.9 12.9

Intra‑group liabilities (2) . . . . . . . . . . . . . . . . . . . . . 92.3 38.4 −6.3 34.5 17.8 23.8 20.0

Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.0 9.3 −4.0 9.2 1.3 0.9 −6.3

p.m. Gross operating surplus . . . . . . . . . . . . . . . . . 86.4 85.2 85.8 88.9 94.8 72.2 76.4

Sources : NAI, NBB.(1) Comprise changes in inventories and purchases of non‑financial, non‑manufactured assets.(2) Transactions with foreign non‑banking sector, as well as captive financial institutions and money lenders, excluding those involving debt securities.


makingupthebulkofthe€47.7billioninnewfinan‑cialassetsacquiredbynon-financialcorporationsinthefirstninemonths(1).

Investment by non-financial corporations in 2016was bolstered by internal resources released throughearningsgrowth.Thegrossoperatingsurplustheygen‑eratedinthefirstninemonthsoftheyearactuallyrosefrom€72.2 billion in 2015 to €76.4 billion in 2016.some corporations did not spend all their surplus on grossfixed capital formationordirect investment and

putittowardsfurtherstrengtheningtheircashreserves.Allinall,Belgium'snon-financialcorporationsamassed€11 billion in cash or bank deposits in the first ninemonthsof2016.Bytheendofthethirdquarter,thesereserveshadgrownto25.2%ofGDP,comparedwith24.2% a year earlier. Their healthier cash positionsmay point to something of a wait-and-see attitudeonthepartofcorporations,notbeingableto identifyinvestmentopportunitiesor sufficiently assured returnprospects in the current climate. At the same time,their cash reserves serve as a guarantee for the future in the sense that these should enable corporations to at least partly finance future projects, while greasingthewheelsofcorporateoperationsbyboostingwork‑ing capital.

(1) Theacquisitionhadnotbeencompletedbytheendofthethirdquarterof2016andtheearmarkedfinancialresourceswerelargelykeptinliquidassets.Thefinancialaccountsdonotthereforeincludethetransactionasastakeinforeign companies.

Box4 –Mergersandacquisitions:multinationalcorporationsandtheBelgianeconomy

Withcheapfundingandamplecashattheirdisposal,somecorporationshavehittheacquisitiontrailandM&Aactivity has been rising across theworld, after slumping in the aftermath of the financial crisis. Cross-bordertransactions are included in statistics on foreign direct investment (FDI) and reflect the increasing weight ofmultinational corporations.

Across theworld, theannual ratioofdirect investment toGDPreached itshighest level sincetheonsetof thefinancial crisis in2008,on thebackofhighernumbersof cross-bordermergersandacquisitionsamongotherfactors.Manufacturingindustry,inparticular,experiencedawaveofconsolidationin2015 thatappearstohavecontinued into2016,even if theoutcomeof theBrexitvoteandtheUSpresidentialelections later in theyearmayputthebrakesoninternationalinvestment.Directinvestmentintheshapeofstakesincorporations’equitycapital suggests that, inBelgium too,mergersandacquisitionswere relativelyplentiful from2015,particularlythoseeffectedbyBelgiancorporationsabroad.Nostatisticsareavailablefor2016justyet,buttheseareboundtobehighlyinfluencedbytheacquisitionofSABMillerbyABInBev.

Againstthisbackdrop,thisboxwillexaminethedegreeofinternationalisationofBelgium’scorporationsaswellastheconsequencesofthisforthewiderBelgianeconomy.Itwillcoverbothfinancialaspects–theeconomy’snetincomefrominwardandoutwardFDI–andrealconsequences,e.g.multinationals’contributionstovalueadded,capitalspendingandemployment.

FDIinflowsandoutflowscorroborateBelgium’spositionasanopeneconomy.OutstandingvolumesofincomingandoutgoingFDIasaratioofGDP–at215%and213%ofGDPbyend-2015 respectively–aremuchhigherthantheaveragesrecordedbyeuroareacountries(132%and156%ofGDP).However,alargeproportioninvolvescapitalintransitandreflectspoliciestoattractFDI,inparticularfiscalincentivessuchasthenotionalinterestdeduction(1).

Belgiumreceivesmoredirect investment than it sendsabroad. Itsnetoutward investmentpositionworkedoutatanegative–2%ofGDPbytheendof2015 –atypicalforadevelopedeconomywithsizeabletotalnetexternalassets,asisclearfrompositivenetdirectinvestmentoutflowsinBelgium’sneighbouringcountries.

4

(1) Formoreinformation,seeDuprezC.andCh.VanNieuwenhuyze(2016),“Belgium’sinwardandoutwardforeigndirectinvestment”,NBB,Economic Review,September,pp.45-63.


Unlike its neighbours, Belgium is currently losing income from its direct investment relationships abroad(–1.9%ofGDPin2015),weighingdownitscurrentaccount.ThisunfavourablefinancialperformanceisthenetoutcomeofbothanegativenetoutwarddirectinvestmentpositionandrelativelylowreturnsonBelgium’sdirectinvestmentabroad.FDIreturnsreflectawholehostoffactors,includingcurrencyfluctuations,economicgrowth inforeignmarketsandtaxation.Comparedwith itsneighbouringcountries,Belgium’soutwardFDIreflectsanunder-representationofmarketsoutsidetheeuroareawhiletheactualinvestmentmasksalargeproportionofintra-grouplendingaimedattaxoptimisationunderthenotionalinterestdeductionschemeandrelativelyfewequitystakesofthesortthatarecommoninmergersandacquisitions.

4

GREATERMERGERSANDACQUISITIONSACTIVITYSUPPORTSDIRECTINVESTMENT

(in%ofGDP,year-on-yearflows)

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

2008

2009

2010

2011

2012

2013

2014

2015

0

2

4

6

8

10

12

14

16

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Cross‑border mergers and acquisitions, industry

Cross‑border mergers and acquisitions, services

Direct investment, flows

WORLDWIDE (1)

Outward direct investment in the shape of equity capital

BELGIUM

Sources:UNCTAD,NBB.(1) Averageofinwardandoutwardcapitalflows.

FOREIGN DIRECT INVESTMENT AND RETURNS

(in % of GDP, 2015, unless otherwise stated)

Belgium

Germany

France

Netherlands

Net outward direct investment (outstanding amounts) . . . . . . . . −2.0 17.8 22.7 100.6

Income from net direct investment . . . . . . . . . . . . . . . . . . . . . . . . . −1.9 1.3 1.9 3.6

Return on outward direct investment (in %, 2013‑2015 averages) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 5.0 4.9 4.6

Sources : ECB, NBB.


Multinational corporations – both parent companies and subsidiaries – make a large contribution to economic activity inBelgium.Althoughrelatively few innumber (1%ofthetotalnumberofprivatesectorcompanies in2014), theyaccount forasubstantialproportionofvalueadded (38%).However, theyhaveasmallernumberofemployees (29%)and therefore recordhigher-than-averageproductivity.Theyalsoplaya significantpart inBelgium’s external competitiveness as theyaremajor contributors to the country’snet exports (1.3 percentagepointsofGDPin2014).Lastly,multinationalsenhancegrowthpotential:between2008and2014,theyaccountedfor33%ofrealprivatesectorinvestmentandfornolessthan63%oftheR&Dspend.

Themultinationalpresencedidnotbringonlyadvantages:inthe2008-14 period,employmentinsubsidiariesofforeigngroupsprovedlessstablethanatBelgianparentcorporations.Althoughadmittedlyanunusualtime–i.e.theaftermathofthefinancialcrisiswithsomecorporationsfacingsuddenexcesscapacity–makingithardtodrawanygeneralconclusions,aseriesofrestructuringplansannouncedbykeymultinationalsin2016wouldappeartoconfirmtheseobservations.

EMPLOYMENT:MULTINATIONALSVSDOMESTICCOMPANIES(1)

(in%,changesbetween2008and2014)

–6

–4

–2

0

2

4

6

8

10

12

Purely domestic companies

Parent corporations and affiliated companies

Subsidiaries and affiliated companies

source : NBB.(1) Thisanalysisonlyincludescorporationswithpaidemployeesthatdidnotchange

theirstatusintheperiod,i.e.fromdomestictomultinationalorviceversa.


Companies increasingly tapped non-bank financing resources in 2016

Despite growing cash reserves, a few corporationsstill rely on sources of external funding to shapetheir investmentprojects,particularly large-scaleones.AB InBevwasacase inpoint in2016: its takeoverofSABMillerwas part-financed through a bond issue. Infact, this accounted for the largest proportion of the€12.9 billion in new net liabilities recorded by non-financialcorporationsinthefirstninemonthsof2016.Although exceptional in terms of the actual amountsinvolved, the bond issue is a close match to similaroperations carriedoutbyfirms since2012. From thatyear, significant falls in long-term returns made thisresource even cheaper for companies able to tap the bond market.

Inthefirstninemonthsof2016,non-financialcorpora‑tions also enjoyed intra-group funding to the tune of€20 billion,which they either received from affiliatedentities established abroad or from financial holding

companies or treasury companies in Belgium itself(1). Once again, SABMiller’s acquisition by AB InBev ac‑counts for much of the size of the resources tapped via this channel in 2016. The rest of these transactions capture financial services provided by some Belgium-based corporationswithin theirowngroups. This typeofactivityhasbecomeverypopularsinceBelgiumintro‑ducedtaxallowancesforriskcapital in2005,sparkinga significant inflowof foreign direct investment,morespecificallyintheshapeofinvestmentinequitycapital.But this taxbreakhasgraduallybecome lessattractivesince2012,mostlybecauseofloweryieldsonten-yearlinearbondsknownasOLOs.Theseserveas therefer‑ence rate for calculating notional interest that Belgium‑basedcompaniescanoffsetagainstcorporationtax.

Bank lending grows on loan demand from SMEs and increasingly favourable loan conditions

Having less access to market funding or resources pro‑vided by affiliated companies, SMEs turned to banksfor funds more keenly than they had in the recentpast. Lending to SMEs by resident banks betweenthe fourth quarter of 2015 and the third quarter of2016 rose by an average 3.6% on an annual basis.Loans to large corporationswere also up, albeit to alesserdegree,notchingupanaverageriseof1.7%inthe same period. credit provided to all categories of Belgiancompaniestogethershowedrobustgrowthin2016:anannualised5.5%inthemonthofNovembercomparedwith2.4%attheendof2015.IgnoringthespecificeffectoftheAB InBevtransactionsattheendoftheyear,loangrowthwasstillsupportedbygrowingdemand–althoughlessintensethanin2015,accord‑ingtothebanks–andbyfavourableloanconditions.Thebank lending survey (BLS) revealedademand forcredit supported by inventories and working capital,mergers and acquisitions and restructuring. Fixedcapitalinvestmentwasanotherdriverin2016,butlessthan in 2015.

Thehighly accommodatingpolicies of the Eurosystemhavemademoneymarket funding very cheap indeedfor euro area banks. They also had access to theexpanded asset purchase programme (APP) and thetargeted longer-termfinancing operations (TLTROs) toboosttheirlendingcapacity,andatveryattractivelend‑ing rates too. InBelgium, thesemeasuresbroughtondownwardrevisionsofcredit riskand intensifiedcom‑petition between banks, translating into a fresh nar‑rowingofcommercialmarginsonstandardloansandto

CHART 45 FUNDINGVIABONDISSUANCESTILLCHEAPEST

(monthlydata,in%)

2008

2010

2012

2014

2016

0

2

4

6

8

10

12

Total weighted financing cost (1)

Listed shares (2)

Corporate bonds (3)

Bank loans (4)

Sources:BarclaysCapital,ThomsonReutersDatastream,NBB.(1) Obtainedbyweightingthecostoffundingbylistedshareissuance,bondissues

and bank loans according to their respective shares in the total outstanding amountofthesefinancialliabilities.

(2) Estimatedonthebasisofadividenddiscountmodel(seebox19inthe 2005 AnnualReport).

(3) Returnonanindexofeuro-denominatedbondsissuedbyBelgiannon-financialcorporations,withmaturitiesofmorethanoneyearandwithratingsinexcessofBaa:theindexisweightedtoreflecttheoutstandingamounts.

(4) Weightedaveragerateappliedbyresidentbankstobusinessloans.Theweightingisbasedontheoutstandingamountofthevarioustypesofcredit.

(1) InBelgium’sfinancialaccounts,holdingsandtreasurycompaniesareincludedinthe“captivefinancialinstitutionsandmoneylenders”sector.


a lesser degree also on riskier loans. long‑term interest rates on loans to resident non-financial corporationseventually plumbed historical lows in 2016, similar toshort-termrates,whileBelgianbanksalsoeasedsome

other loanconditionsunrelated to interest rates, suchas restrictions on volumes and terms of loans and / or other covenants specified in loan arrangements withcompanies.

CHART 46 SIGNIFICANTINCREASEINBANKLENDING,DESPITESLOWINGDEMANDGROWTH

2012 2013 2014 2015 2016–150

–100

–50

0

50

100

150

2012 2013 2014 2015 2016–4

–3

–2

–1

0

1

2

3

4

5

6


Inventories and working capital

Interest rate level (2)

Other borrowing requirements

Internal funding

Total demand

Loans from competitors

Market funding

LOAN DEMAND AND EXPLANATORY FACTORS(quarterly data, weighted net percentages

(1))

Euro area

Belgium

GROWTH IN BANK LENDING TO RESIDENT NON‑FINANCIAL COMPANIES

(3)

(monthly data, annual percentage changes)

Sources:ECB,NBB.(1) Apositive(negative)netpercentagecorrespondstoafactorcontributingtoborrowingdemandgoingup(down).(2) Afactorfirstincludedinthesurveyfromthefirstquarterof2015.(3) Residentbanklendingonly(includingsecuritisedloans).

Table 11 LENDING BY RESIDENT BANKS TO NON‑FINANCIAL CORPORATIONS : GROWTH BY COMPANY SIZE

(average annualised gowth rates (1), in %)

2005Q1‑2016Q3

2005Q1‑2008Q4

2009Q1‑2016Q3

2015Q4‑2016Q3

SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 9.1 2.9 3.6

Small businesses . . . . . . . . . . . . . . . . . . . . . . . . . . 4.9 9.3 2.6 3.8

Medium‑sized businesses . . . . . . . . . . . . . . . . . . 5.4 8.8 3.6 3.3

Large corporations . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 10.9 −2.4 1.7

Source : NBB (Central Corporate Credit Register).(1) Annualised average of quarterly growth. Not included are the second quarter of 2012 and the fourth quarter of 2014, which saw a break in the statistical series.


Non-financial corporations’ debt ratio remains high, but funding structure is robust

Non-financial corporations’ consolidated gross debt ra‑tios rose to 120.6% of GDP in September 2016 from116.1%attheendof2015,anincreasewhollyattribut‑able to AB InBev’s financial transactions as it turned todebt financing to prepare for its takeover of SABMiller.Disregarding these latter transactions,debt ratioswouldinfacthavefallenslightly.

Belgiannon-financialcorporations’levelsofdebtremainedhigh (80.5% of GDP) when compared with euro areacorporations,but it shouldbe remembered thatdebtsatnon-financialcorporations inBelgiumlargelyreflect intra-groupfunding.Ratherthanmeetingarealfundingneed,suchfinanceisoftenmoreofaredistributionofresourcesacrossagroup,possiblyfortaxreasons,andaliabilitythatcomeswithanequalclaimonagroupentity.Ifthistypeoffundingisexcluded,non-financialcorporations’debtratiosinBelgiumworkedoutat55.3%ofGDP.

In the euro area at large, non-financial corporations’debt ratios peaked in 2010, but Belgium has seen a

continued slight rise on the back of recovering bank lending(partofotherdebt)anddebtsecuritiesissuance(corporatebonds).Thislattercategoryrosetoarecord15%ofGDPatthebeginningof2016,comparedwith7.1%inearly2010.AlthoughinfluencedontheupsidebyABInBev’stransactionsin2016,bondmarketshavebeenincreasinglytappedinrecentyears,withthelowinterestrateenvironmentplayingamajorpart.

Belgiannon-financialcorporations’relativelyhigh–andrising – debt ratios do not spell immediate danger to debt sustainability, as is clear from their relativelyhealthy funding structure. For one thing, the debt-to-capital ratio is relatively low, both compared withits own long-term average and compared with euroarea corporations. That said, the ongoing fall in thisratio – supported by the notional interest deductionsince2006–wouldappeartohaverunitscourse.From2012 on, this has been partly attributable to this no‑tional interest deduction scheme constituting less of a taxadvantage.Secondly, corporationsare increasinglyattracting long-term finance to limit refinancing risk,and the proportion of long‑term debt – i.e. over one year–rosetoahighof72.2%ofconsolidateddebtin

CHART 47 EVENBETTERLOANCONDITIONS

2012 2013 2014 2015 2016

–150

–100

–50

0

50

100

150

2012 2013 2014 2015 2016

0

1

2

3

4

5

MAIN LENDING CONDITIONS (quarterly data, weighted net percentages

(1))

Margins on standard loans

Margins on riskier loans

Other credit conditions (averages)

Medium‑term rate (4)

Short‑term rate (3)

Long‑term rate (5)

INTEREST RATES ON NEW BUSINESS LOANS (2)

(monthly data, in %)

Source:NBB(BLSandMIRsurvey).(1) Apositive(negative)percentagecorrespondstoatightening(easing)ofloanconditions.(2) Loansofupto€1 million.(3) Variablerates,initiallyfixedforuptooneyear.(4) Rateinitiallyfixedatbetweenoneandfiveyears.(5) Rateinitiallyfixedformorethanfiveyears.


the thirdquarterof2016.Thiscoincidedwithgreateruptakeofnon-bankfinancingresources,i.e.corporatebondsthat typicallyhave longermaturities.Lastly,de‑spitehigherdebt levels and longer terms tomaturity,interestratecharges(interestpaymentsasapercentageofgrossoperatingsurplus)continuedtodeclinefurtherthanksboth toaccommodatingmonetarypoliciesandthe higher operating surplus.

4.2 Positive and negative effects of the lowinterestrateenvironmentonhouseholds’assetsandliabilities

Sharply lower money market rates since 2009 – aresult of the ECB’s highly accommodating monetarypolicies–havebeenlargelypassedontofinancialinsti‑tutions’ customers and clients, particularly households,in the shape of retail rates on both deposits and loans. Whiletransmissionwasonlypartialuntil2013,itbecamemuchfuller from2014 andbanks’commercialmarginshavestabilisedsince.Inadditiontoretailrates,monetary

policy also affected other types of interest-bearing as‑sets,suchasgovernmentandcorporatebonds.

Lower returns on households’ interest-bearing assets– the bulk of their savings – inevitably suffered fromthe drop in income from financial assets invested inthese types of instruments. Interest received in thefirstninemonthsof2016again contracted toaverage€0.7 billion a quarter, compared with €1 billion in2015. In 2009,when interest rateswere higher, thesequarterlyaveragesstillcameinat€2.4 billion.

For private individuals and their financial situations thereis some good to be had from the low interest rateenvironmentaswell,astheyattractfreshliabilitiesatlowerinterestrates.Existingloansmayalsobenefit iftheyhavevariable interest ratesorareupfor refinancing.However,asignificantincreaseinliabilitieshassomewhatweakenedtheperceptibleeffectoninterestpaidbyhouseholds.

Lower interest rates also have an impact on assetprices. Between 2009 and 2016, households enjoyed

CHART 48 HIGHCORPORATEDEBTLEVELS,BUTRELATIVELYROBUSTFUNDINGSTRUCTURE

0

20

40

60

80

100

120

140

2002

2004

2006

2008

2010

2012

2014

2016

40

60

80

100

120

140

160

Share of long‑term debt in total debt

Debt‑to‑capital ratio (2)Belgium of which

Intra‑group debt (1)

Debt securities

Other debt

Euro area

CONSOLIDATED DEBT(in % of GDP)

FUNDING STRUCTURE(in %)

Long‑term average for Belgium, 1999‑2006

Belgium

Euro area

Belgium

2010Q1 2016Q3

Sources:ECB,NBB.(1) Loansprovidedbyforeignnon-bankingcorporations,captivefinancialinstitutionsandmoneylenders.(2) Non‑consolidated data.


favourable valuation effects on their financial wealthin keeping with generally positive developments inthefinancialmarkets, cushioning the repercussionsoflowerinterestincome.Gainsweremostlylockedinonequityportfoliosandinvestmentfundsintheperiod.In2016,by contrast, volatile financialmarketsmade forratherlesssubstantialvaluationeffects:whereasequitypriceshadnotchedupincreasesofatotal€7.1 billionin thefirst threequartersof2016, theoverallportfo‑lio of investment funds as held by households facednegative valuation effects to the tune of € 2.6 billion. Overall,though,lossesincurredinthefinancialmarkets– caused by the interest rate hike by the US FederalReserve, weaker economic activity in China and thereferendum in theUnitedKingdom–were short-livedand limited in size.

Households’interestearningsmayhavecontractedsincetheonsetofthefinancialcrisis,buttheirfinancialwealthhasgenerallycontinueditsuninterruptedupwardtrajec‑tory.Portfolioshavegrownonthebackofassetrevalua‑tions(€174 billion)andfreshinvestment(€203 billion).

Nominal house prices have more than doubled since 2000 andhouseprice falls during thegreat recessionwereveryminorboth insizeanddurationwhencom‑pared with quite a few euro area countries. Propertyprice increasesconsistentlyslowedbetween2011 and2014 butpickedup sharply in2015,notwithstandingpropertytaxreforms, intheFlemishRegioninparticu‑lar. In2016, the year-on-yeargrowth rateofpropertyprices slowedsignificantly,amounting tobarely0.7%inthefirstthreequarters.Inrealterms,propertypricesdeclinedby0.9%.

Deviations in house prices from their equilibriumvalue – as determined by econometricmodels factor‑ing in fundamental determinants such as households’disposable income, mortgage rates and demographicpressures–wereestimatedat6%in2016.Thisislessthan in2015,whenmarketovervaluationwassharplyup, to 10.9% in the fourth quarter, particularly asmortgage interest relief reforms in the flemish Region did not cause any negative pressure on growth inpropertyprices.

CHART 49 LOWINTERESTRATEENVIRONMENT’SIMPACTONINTERESTPAIDANDRECEIVED

2007 2008 2009 2010 2011 2012 2013 2014 2015 20160

1

2

3

4

5

6

2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 (1)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

HOUSEHOLD LENDING AND DEPOSIT RATES(in %)

Average interest on deposits with maturities of under two years

Interest on new deposits at notice of less than three months

Main refinancing rate

Average interest on mortgage loans with a remaining maturity of over five years

Interest on new mortgage loans initially fixed at over ten years

INTEREST RATES PAID AND RECEIVED BY HOUSEHOLDS (quarterly averages, € billion)

Interest paid

Interest received

source : NBB.(1) Averagesforthefirstthreequarters.


Activityintheresidentialpropertymarketwasstronglyaffectedbydevolutiontotheregionsaswellasbytaxrelief reforms for mortgage loans that came into ef‑fect on 1 January 2015. This was particularly true intheFlemishRegion,whichhadseensecondarymarkettransactionsrocketinthelastquarterof2014 andthenveersharplydowntheyearafter.In2016,propertyac‑tivityperkedbackupandendedupcloserto itsusuallevels. The total number of property transactions inBelgiumgrew11.3%inthefirstthreequartersofthereportingyear.

Astheboxbelowoutlines,theactualmixofbothnega‑tiveandpositiveeffectsof lowinterestratesonhouse‑hold income and financial assets is rather complex. Itdependsonhouseholds’financialsituations(indebtorincredit)aswellasthecompositionoftheirassets,whichtendstovaryaccordingtowealth.

CHART 50 POSITIVEVALUATIONEFFECTSOFHOUSEHOLDFINANCIALASSETS

(in€billion:annualdata,unlessotherwisestated)

2011

2012

2013

2014

2015

2016

(1)

–15

–10

–5

0

5

10

15

20

25

30

35

40

Debt securities

Equities

Insurance products

Investment funds

Total

Other

source : NBB.(1) first nine months.

Box5 –Lowinterestratesandassetvaluations:distributioneffects

Persistentlylowinterestrateshaveawiderangeofeffectsonhouseholds’financialpositions.Togaugetheactualimpactofthelowinterestrateenvironment,ahighlysimplifieddistinctionissometimesmadebetweenhouseholdsthatsaveandhouseholdsthatborrow,withsaverslosingoutoninterestearningswhileborrowersfinditeasiertorepaytheirloans.Monetarypolicy’stransmissionmechanismsingeneraland,morespecifically,theeventualeffectsoflowinterestratesaresocomplexthattoosimpleananalysiscangiverisetoahighlydistortedpicture.Foronething, current low interest rate policies alsomake formore stablemacroeconomic andfinancial environmentssupportiveofgrowthandemployment,andsoalsoaffecthouseholds’earnedincomesandthesecurityofthese.Lowinterestratesdoindeedhavenegativehouseholdincomeeffects,astheydepressincomefromfinancialassets–whereastheypushupthevalueofbonds,equitiesandproperty,andsocausepositivevaluationeffects.Butalowinterestratepolicyisalsoapositiveforincomeasitmakesiteasiertopaydowndebt.Asthecompositionandsizeoftheirnetwealthwillvaryfromonehouseholdtothenext(orbetweengroupsofhouseholds),lowinterestearningsorchargesandpositivevaluationeffectswillbedifferentforeachandeveryoneofthem.

ThesheercomplexityoftheissuecanbeillustratedbydatatakenfromtheHouseholdFinanceandConsumptionSurvey(HFCS)(1),with2014 themostrecentyearforwhichsurveyresultsareavailable.Theserevealthatninein

4

(1) Foradescriptionandanalysisofthe2014 outcomes,seeDuCajuPh.(2016),“HouseholdwealthinBelgium:initialfindingsofthesecondwaveoftheHouseholdFinanceandConsumptionSurvey(HFCS)”,NBB,Economic Review,September,pp.27-43.


everytenhouseholdshavelabourincomesfromregularemployment,enjoyincomesfromaportfoliooffinancialassets(andsoreceive interestordividends)orarerepayingdebt(andsopay interest).One-thirdofhouseholdssurveyedfallsintoallthreecategoriesatthesametime:theyearnwages,havefinancialincomesandrepaydebt.Besides,thevastmajorityofhouseholdsrepayingdebtalsoreceiveincomefromfinancialassetsand,viceversa,nearlyhalfofhouseholdswithincomesfromfinancialassetsalsorepaydebts.ThisgoestoshowthattheBelgianpopulationcannotbefittedintoanysimplebreakdownofspecificgroupsthatdoordonotbenefitfromspecificinterest rate developments.

Thesizeandcompositionofnetwealthvarygreatlyfromonehouseholdtothenext.Householdsinthelowestwealthquintile–i.e.the20%leastwealthyhouseholds–ownlittleinthewayofassets,whichtypicallycomprisedepositsandotherrealassets(vehiclesorothervaluables).Wealthinthethreeintermediatequintiles,whichwemightrefertoasthemiddleclassesforthesakeofconvenience,primarilytakestheshapeofhomeownership,plusdepositsand–toalesserdegree–individualvoluntarypensionsandlifeinsurance.Thisgrouptypicallyalsohasthelargestamountofoutstandingmortgagedebt.Incontrast,wealthierhouseholdstendtoownpropertieswithhigheraveragevaluesthanmiddle-classhouseholds,buttheseaccountforlessthanhalfoftheirtotalwealth.Richerhouseholds’wealthisalsomadeupofotherpropertyandarangeoffinancialassets.

Distributiondatasuggestthatpositivevaluationeffectsasaresultoflowinterestratepoliciesarechieflyfeltinthehighestwealthquintileasequities,bondsandinvestmentfundsarealmostexclusivelyownedbytheseparticularhouseholds.Voluntarypensionsandlifeinsurance–whichtypicallyfocusontheratherlongertermandtendtobecomeworthmorewhen interest rates fall – aremoreequallydistributedacross thehigherwealthquintiles.Property, andmore specifically themain residence, is themost importantwealth component for all groupsof

COMPLEXDISTRIBUTIONASPECTSOFLOWINTERESTRATESANDASSETVALUATIONS

(2014)

I II III IV V

–200

0

200

400

600

800

1 000

1 200

57.5% earn a wage income

78.3% receive income from financial assets

40.2% pay off debt

Debt

Real assets

Financial assets

Mortgage loans for main residence

Mortgage loans for other property

Non-mortgage loans

Other financial assets

Voluntary pensions and life insurance

Listed shares

Bonds and savings notes

Investment funds

Deposits

Other real assets

Self-owned businesses

Other property

Main residence

TYPOLOGY OF BELGIAN HOUSEHOLDS(in % of total number of households)

COMPOSITION AND DISTRIBUTION OF NET WEALTH(average value of assets in €1 000, by wealth quintile)

4.8

4.0

3.8

31.0

17.7

25.8

1.4

Source:NBB(HFCS2014).

4


New financial assets : search for yield grinds to a halt

Despite the low interest rate environment, Belgianhouseholds continued to prefer savings in the shape of liquid assets, e.g. savings deposits or long-term, low-risk assets such as capital‑protected insurance products. In the first three quarters of 2016, households put€11.5 billionintodepositsintotal,morethantheyhadin 2015. Insurance products excluding those withoutcapitalguarantee(class23)andpensionfundsattracted€8.4 billion. So the bulk of household savings werechannelledmostly intodepositsandinsuranceproductsinthefirstninemonthsoftheyear,whiletheremainderwas spent on investment fund units (€3.8 billion, sig‑nificantly less than the year-earlier €13 billion). Lastly,householdscontinuedsellingoutoftheirdebtsecurities,to the tune of €8.9 billion. Although involving only amoderatecounterparty risk, interest rate riskmayhavemade these assets less attractive. Households also jet‑tisoned €0.5 billion worth of equities. The search foryieldthathadstartedin2013 appearedtohavegroundto a halt in 2016.

Theinvestmentchoicesmadebyhouseholdsin2016wereinfluencedbyanumberoffactors,rangingfromlackofconfidence and uncertainty – fostering a wait-and-seeattitudeandencouragingprecautionarysavingagainstabackdropofvolatilefinancialmarketsandconcernsovergeo-politicalconditions–tothetaxtreatmentofthevari‑ousfinancialassets.

Theyearsawhouseholdsloseheartonseveraloccasions.Thedeteriorationatthebeginningof2016reflectedtheirperceptionofgeneraleconomicconditionsandtheirown

households.Allthesequintilesbenefitfromcapitalgainsbutthewealthiesthouseholdsmoresoinabsoluteterms,astheirrealestatetendstobethemostvaluable.Cheaperpropertyfinanceduetolowinterestratesendsupinthepocketsofpeopleinthemiddlenetwealthquintiles,asthoseinthelowestquintilearenotusuallyhomeownersanddon’t havemortgage loans. Both in absolute terms and relative to their assets, households in themiddlequintilestypicallyhavelargermortgagedebtsthandothoseinthewealthiestquintile.

Evenafterweighingupallaspects,thedistributioneffectsofthecurrentlowinterestrateenvironmentremainhardtoestablish.Forthewealthiesthouseholds, low interest ratesspell lower incomefromfinancialassets,butthisgroupalsobenefitsmostfromcapitalgains.Householdsinthemiddlequintilesfinditeasiertofinancetheirhomesthankstoalowerinterestburden:whileinterestontheirdepositsfalls,thevalueoftheirpropertyincreases.Theleastaffluenthouseholds,intheirturn,donotusuallyprofitfromcapitalgains:thosewithsavingsloseincome,thosewithconsumercreditandthefewwhoowntheirownhomespaylessinterestontheirloans.Itisalsohardtomeasuretheindirecteffectsofalowinterestratepolicyonhouseholdwealthandincomes.Afterall,lowinterestratepoliciesbringmacrofinancialstability,anecessitybothforasmoothlyoperatingeconomyandforsupportingandgrowinghouseholdlabourincomesandfinancialwealth.

CHART 51 PREFERENCEFORLIQUIDORLONG-TERM,BUTNOTVERYRISKYPRODUCTS

(in€billion)

2007

2008

2009

2010

2011

2012

2013

2014

2015

–20

–10

0

10

20

30

40

50

2015

2016

–20

–10

0

10

20

30

40

50

Notes, coins and deposits

Debt securities

Investment funds

Insurance products, excluding class 23

(1)

Total

First nine months

Shares

Less risky instruments

Riskier instruments

Class 23 insurance products (1)

Other (2)

source : NBB.(1) These items comprise the net claims of households on technical insurance

reserves,onpensionfundsandonstandardisedguaranteeschemes.(2) Thisitemcomprises,insofarastheyhavebeenrecorded,tradecreditaswellas

miscellaneousassetsofgeneralgovernmentandfinancialinstitutions.


financial situations,while theirassessmentof the labourmarketunderpinnedtheirchoicesbytheautumn.

Changeswerealsoafootintaxesleviedonincomefrommovableassets.Thewithholdingtaxoninterest,dividendsandroyaltieswasbumpedupagainon1 January2016,to27%from25%–andraisedto30%inearly2017.Special tax deals still apply to some products, such asregulatedsavingsdeposits(exemptiononthefirstinterestincomebandof€1880,whichishardlyusedgivenverylowinterestrates:15%foramountsover),guaranteed-return insuranceproducts (undercertainconditions)andspecificStatenotes(withholdingtaxof15%onLetermeStatenoteyields).

Alongsidestructuraltaxesonfinancialwealthincome,thegovernment introduceda33%taxonfinancialspecula‑tion from 1 January 2016: this affected gains realisedwithinsixmonthsofpurchase,appliedtohouseholdsandlisted assets only (equities, equity certificates, warrants,optionsorotherequity-basedfinancialinstruments).Thistaxwasscrappedatthestartof2017.

Higherwithholdingtaxandtheintroductionofthespecu‑lationtaxmayhavekepthouseholdsfrominvestingintheequitymarketsandpushedthemtowardsmorepropertyinvestment.Between2013 and2016,householdactivityin the equitymarkets slumped comparedwith the four

previous years: cumulative net transactions in equitiesstood at €30 billion between the first quarter of 2009andend-2012,whilethesewerevirtuallynil(€0.5 billion)between 2013 – when the withholding tax was raisedfrom21%to25%–and2016,theyearinwhichitwentupto27%.

Household debt ratio still rising

Lowinterestratescontinuedtoshoreuphouseholdloandemand,particularlyformortgageloans.Althoughbanksactually tightened their non-interest-rate-related loanconditions slightly, the pace of mortgage loan growthstabilised at around 5.4%.With outstandingmortgageloanscontinuingtogrowfast(to€205 billion),thegrosshouseholddebtratioincreasedtrendwise,workingoutat59.1%ofGDPinthethirdquarterof2016.Sincethebe‑ginningof2016,thedebtratioofBelgianhouseholdshasbeenabovetheaveragefortheeuroarea,whichhasbeenmarkedbyadeleveragingprocesssince theendof 2009and which has seen the household debt ratio graduallycomedown from63.9% to 58.7%ofGDP in the thirdquarterof2016.

Microeconomic data from the Hfcs point up important vulnerabilities or pockets of risk in the debt structure. some households are taking on large debts as a propor‑tionoftheirincomesortheirliquidassets–asisapparent

CHART 52 HOUSEHOLDDEBTLEVELSKEEPRISING,BUTDEFAULTSREMAINRELATIVELYFEW

2002

2004

2006

2008

2010

2012

2014

2016

0

10

20

30

40

50

60

70

2008 2009

2015

2010 2011

2012 2013 2014

JJ

J

J J

Euro area

BelgiumDebt / GDP Debt / financial assets

DEBT RATIOS(in %)

M+6

M+1

2

M+2

4

M+3

6

M+4

8

M+6

0

M+7

2

M+8

4ARREARS ON MORTGAGE LOANS

(1) BY YEAR OF GRANTING

(2)

(in % of the number of loans)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Sources:ECB,NBB.(1) Aloanisregisteredasindefaultwhenaduesumhasnotbeenpaideitherinpartorinfullwithinthreemonthsfollowingitsduedateorwithinonemonthafterformal

noticehasbeenservedbyrecordeddeliveryletter.(2) Loansaregroupedbytheyeartheyweregranted,withthecurvesshowingthenumberofloanspastdueforeachyearasapercentageofthetotalnumberoforiginal

loans,afterasetnumberofmonthsfollowingtheirissue.Noaccountistakenofanyregularisationoftheloans.


from the loan-to-value (LTV) ratio, i.e. the valueof theamount borrowed relative to that of theproperty – orare displaying limited repayment capacity. 2014 surveydata show that 13% of outstanding mortgage debtis to be repaid by households that need40%of theirdisposable incomestodoso.SuchhouseholdstypicallyhavelittleinthewayofassetsandrelativelyoftenagreeloanswithhigherLTVs.

For these reasons, the European Systemic Risk Board(ESRB)formallyflaggedupheightenedrisksintheBelgianmortgagemarket inNovember2016,as itdidforsevenother countries. A less supportive macroeconomic envi‑ronment–withrisinginterestrates,forinstance,orade‑terioration of labour market conditions and the business cycle–wouldthreatenhouseholdrepaymentcapacity.Inanticipationofsuchascenario,theBankhastakenvari‑ousmeasuresinitscapacityasmacroprudentialauthoritytoaddressandmitigatetherisksrelatedtothepropertymarket and a rising household debt ratio.

Atthispoint,relativelylowandstablelevelsofarrearsonloansbyhouseholds–asapercentageoftotalloans–asregisteredby theCentral IndividualCreditRegister (CICR)suggest that the current debt service burden is still manage‑able formosthouseholds. In fact,mortgage loanarrearsareevenfallingslightlyonthebackofanimprovedmacro-economic climateand,more specifically, reduced jobless‑ness. In addition, accommodating monetary policy hasreducedthetotalinterestburdenfacinghouseholds,evenif their overall debt levels have increased – thanks in part toloanrefinancing.Quitepossibly,inthesecases,non-in‑terest-rate-relatedloanconditionshavealsobeenchanged,furtheralleviatingtherepaymentburden.Theserefinancingtransactions,whichoftenentailfixed-ratemortgages,helphouseholdstoretainbeneficialinterestrateconditionsforthe remaining term of their loans, with the risks of anyupturn in interest rates devolved to their banks.

Despite the furtheradvance in thedebt ratio,Belgianhouseholds continue to display healthy aggregatefinancialpositions,asdemonstratedamongother fac‑torsbytheratioofdebttofinancialassets,whichhasremained stable and is lower in Belgium than in thebroader euro area. Incidentally, Belgian householdshavethe largestnetfinancialwealth in theeuroarea,as a percentage of GDP.

Thishighfinancialwealthalsocontributestothefavour‑able financial position of the broader Belgian economy,theso-callednetexternalposition.Afterall,thedifferencebetween the financial assets and liabilities of domes‑tic sectors translates into either a net creditor or a net debtorpositionrelativetoothercountries.Itshouseholds’

sizeablenetwealthputsBelgiuminaselectgroupofnetcreditor countries in the euro area, together with theNetherlands,Germany,AustriaandFinland.Onnationalfinancialaccountsstatistics,Belgium’snetclaimsonothercountriesamountedto€201 billionor48.3%ofGDPbyend‑June 2016

4.3 2016anothergoodyearforBelgianbanks

Inthefirstninemonthsof2016,Belgium’sbanksachievedrobust results, as theyhaddone in2015.Annualised re‑turn on equity neared 10%,while assets yielded 0.6%.The sector at large recordedprofits totalling€4.8billionin the period, comparedwith a year-earlier €4.4 billion.Despite a refocusing of activities back to the Belgian mar‑ket,thecountry’sfinancialinstitutionsgeneratedone-thirdof theirnetprofitsoutsidenationalborders,as theykeptup a strong presence in a number of foreign “strategicmarkets” such as East and South-Eastern Europe, theNetherlands,Ireland,SwitzerlandandLuxembourg.Unlikein2015,however,Belgium’sbankssawafewfundamentalsourcesofincomedipdownandwereunabletoleverageseveral temporary factors that had previously benefitedthem. Consequently, gross operating result shrank bynearly€200 Emilliononthecorrespondingperiodof2015.

CHART 53 NETFINANCIALWEALTH(1)

(in%ofGDP,end-June 2016)

N

L

DE

be

AT FI EA FR IT EE SI LT SK LV ES PT CY EL IE

–300

–200

–100

0

100

200

300

F

General government

Households

Enterprises

Total economy (net external position)

Sources:ECB,NBB.(1) Differencebetweentheoutstandingamountsoffinancialassetsandliabilities.

LuxembourgandMaltaarenotincludedinviewofthehighvolatilityoftheirdata.


Low interest rate environment leads to slight fall in net interest income

Gross operating result in the first ninemonths of 2016contracted,amongother factors,asnet interest incomeedged down by nearly €100 million, despite a robustincreasein lendingtoBelgium’snon-bankprivatesector.The2016interestmarginendedupatroughlythesamelevel as in 2015, averaging 168 basis points comparedwith 169 in thefirst ninemonths of 2015, bringing anendtotheiruninterruptedadvancesince2008.

The various components of the interest margin came underpressurein2016.Foronething,intensifiedcompe‑tition in the Belgian market caused commercial margins oncertainproductstoshrink.Commercialmarginsreflectthedifferencebetweeninterestrateschargedbybanksonloans(orpaidondeposits)andtheircorrespondingmar‑ketrates.Theintermediationmargin,anothercomponentoftotalinterestmargin,isdeterminedbytheyieldcurve.Itwasalsosqueezedfornewtransactions,asevidencedbyafurthernarrowingofthespreadbetweenten-yearandone-monthrates.However, theeffectshavesofarbeen

mitigatedbytheBelgianbanks’balancesheetstructure,i.e.liabilitieswithshortertermstomaturitythanassets.

In 2016, the country’s banks were still able to cushionthe subdued returnsonassetsbycutting fundingcosts,withmost of them reducing rates on savings and sightaccounts for Belgian households to 0.11% (includingfidelitypremiums)and0.00%respectively.Mostlytakenin the second quarter, these measures caused interestmargins to widen temporarily, but they had started tocontractagainbythethirdquarter.

With interest rates on deposits having reached a lowerbound, there is very little room for further reductions,which is likely todepress interestmargins in the future.Meanwhile,marginsarealsoslowinginthewakeofthesaleofhigh-yieldingdebtsecurities,possiblyfacilitatedbytheECB’sexpandedassetpurchaseprogramme(APP).Themanyrefinancingsofmortgageloanssince2014 –whichhave speeded up the decline in returns on this portfolio – havea similareffect, even thoughnew loanscommandhigher commercial margins than the original loans. Inthis context, the early redemption fees banks earned

Table 12 INCOME STATEMENT OF BELGIAN CREDIT INSTITUTIONS

(consolidated data ; in € billion, unless otherwise stated)

First nine months

In % of operating income

2012

2013

2014

2015

2015

2016

Net interest income . . . . . . . . . . . . . . . . . . . . . . . . . 13.6 13.3 14.5 14.9 11.2 11.2 67.2

Non-interest income . . . . . . . . . . . . . . . . . . . . . . . . 4.5 7.0 6.2 7.1 5.6 5.5 32.8

Net fee and commission income (including commission paid to agents) . . . . . . . 4.5 5.0 5.3 5.9 4.5 4.2 25.3

(Un)realised gains or losses on financial instruments (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0 0.8 −0.1 1.2 1.2 0.8

Other non-interest income . . . . . . . . . . . . . . . . . 0.0 1.3 0.9 0.1 −0.1 0.5

Operating income . . . . . . . . . . . . . . . . . . . . . . . . . . 18.1 20.3 20.7 22.0 16.8 16.6 100.0

Operating expenses . . . . . . . . . . . . . . . . . . . . . . . . . −13.0 −12.4 −12.7 −12.9 −9.9 −9.9 59.5 (2)

Gross operating result . . . . . . . . . . . . . . . . . . . . . . . 5.0 8.0 8.0 9.1 6.9 6.7

Impairments and provisions . . . . . . . . . . . . . . . . . . −2.6 −3.0 −1.3 −1.3 −0.6 −0.7

Impairments on loans and receivables . . . . . . . . −2.0 −2.3 −1.3 −1.1 −0.9 −0.7

Impairments on other financial assets . . . . . . . . 0.8 0.0 0.0 0.0 0.0 0.0

Other impairments and provisions . . . . . . . . . . . −1.5 −0.6 0.1 −0.1 0.3 0.0

Other components of the income statement . . . . −0.8 −1.8 −2.2 −1.7 −1.9 −1.3

Net profit or loss . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 3.3 4.5 6.1 4.4 4.8

Source : NBB.(1) This item also includes the net realised gains (losses) on financial assets and liabilities not measured at fair value through profit or loss, the net gains (losses) on financial

assets and liabilities held for trading and designated at fair value through profit or loss, and the net gains (losses) from hedge accounting.(2) Cost / income ratio of the Belgian banking sector.


on refinancing transactions were also lower in the firstninemonthsof 2016, providing less support to interestmarginsandinterestincome.Internalaswellasexternalrefinancingsfellto€18billion,stillarespectableamountbutwellbelowthefigureforthecorrespondingperiodof2015 (€31 billion).Thesemayalsohave involvedlower-interest loansasmanyofthemhadbeenrefinancedbe‑fore,alsodepressingearlyredemptionfeesin2016.

Various components of non-interest income declined sharply

Non-interestincomealsofell,from€5.6billioninthefirstninemonthsof2015 to€5.5 billion in2016. Thiswasmainlydue to thecontractionofaround€0.3 billion innet commission income – caused by falling fees on thesaleof investment fundsand, toa largerextent,ontheissueand transferof securities.Volatility in thefinancialmarketscombinedwitharangeoftaxfactorstoreducethe number of transactions. Gains on financial instru‑mentswerealsodown, asunrealisedgains –whicharerecognisedwhenthemarketvaluesofsecuritiesincrease–were less substantial.Conversely,gainson thesaleofsecuritieswerehigher–albeittoalesserdegree–onthebackofsalesofhigh-yieldingequitiesanddebtsecurities.Non-interestincomewasalsosupportedbyothersourcesofincome,suchasdividendsreceived.

Operating expenses remain relatively stable but loan loss provisions fall again

Operating expenses at Belgian banks worked out at€9.9billion for thefirst ninemonthsof 2016, about thesame level as in 2015. Unlike in previous years, spendingonstaffremunerationacrossthesectorisinchingdownasa number of restructuring programmes are starting to bear fruit. Other operating expenses remained stable, despitehigherbanktaxesandcontributions–particularlythehighercontributionstotheEuropeanSingleResolutionFund(ESRF).

Lower banking income combinedwith stable operatingexpenses to cause a slight deterioration in the Belgian

CHART 54 BELGIANBANKS’INTERESTMARGINSPERSISTENTLYHIGHIN2016

(quarterlyconsolidateddata)

2008 2009 2010 2011 2012 2013 2014 2015 2016 2008 2009 2010 2011 2012 2013 2014 2015 20161.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

NET INTEREST INCOME(annualised average percentages of interest-bearing assets)

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

DIFFERENCE BETWEEN TEN-YEAR SWAP RATE AND MONTHLY INTERBANK RATES(in %)

Sources:ThomsonReutersDatastream,NBB.

CHART 55 SIGNIFICANTREFINANCINGOFMORTGAGELOANSIN 2016,BUTLESSTHANIN 2015

(newmortgageloans,ninemonthscumulative;in€billion)

2008

2009

2010

2011

2012

2013

2014

2015

2016

0

10

20

30

40

50

60

Mortgage loans, excluding refinancing

Mortgage loans, excluding internal refinancing

All mortgage loans

source : NBB.


bankingsector’scost/incomeratio,whichgrewto59.5%inthefirstninemonthsof2016from58.9%in2015.

Impairmentsandprovisions,whichwerealreadybackatrespectablelevelsduetoreductionsthatgotunderwayin2011,continuedto fall throughout2016.Abouthalfofthe€0.7billionfigurerecordedforthefirstthreequartersderived from foreign portfolios. subdued net provisions reflectedbothaqualityimprovementinbankassetsandreversalsofexcessprovisionspreviouslymade.

Banks gird up for the future and take range of measures to keep profits up

Relativetotheirshareholders’equityaswellastototalas‑sets,returnsintheBelgianbankingsectorweretwicethatoftheeuroareaaveragein2016.Thebankingindustry’srelativelyhighprofitability levelswerepartly attributableto exceptional developments, such as tax losses carriedforwardbyoneofthecountry’sbiggestbanks.Returnonequityat10%slightlydistortsthesector’srealprofitabilityanditshouldnotinanywaybeassumedthatprofitabilitywill return topre-crisis levels. That said, despite themi‑norslowdowninthereportingperiod,Belgianbanksdoenjoyhealthyunderlyingprofitabilityandfindthemselvesinarobuststartingpositionfromwhichtofaceatimeinwhichtheircapacitytoturnaprofitwillbetested.

Income from mark-to-market gains on portfolios offinancial instruments or the sale of securities will nolongerbeashighasinthepast,whilemoneymadefrom

earlyredemptionfeesonmortgageloansareatemporaryboon. Interestmarginsmaywellnarrowifthelowinter‑estrateenvironmentintheeuroareapersists.However,asudden sharp rate hike might also hit institutions that fail tohedgeproperly againsthigher short-term rates: theirfunding costs would rise whereas the returns onmanyof theirassetsarefixedfor themediumor longer term.Lastly, the extremely low levels of loan loss provisionslookunsustainable,ifonlybecausesomeofthisisduetoimpairment reversals.At the same time,digitalisationofbankserviceswilleventuallystokeupcompetitioninkeysegments of the market – e.g. asset management – and mayadverselyaffectBelgianbanks’capacitytogenerateprofitsonsuchbusiness.

Digitisation of bank product offerings may present awhole host of challenges to incumbent players, but italsoprovidesmanyopportunitiestomaketheiractivitiesmoreefficient.Andso,numerousbankshaveannouncedlarge‑scale restructuring plans and savings in areas such as branchofficenetworksandworkforce,both in thepastandtoday.As it turnedout,spendingcutsprovedquitemoderatein2016butthesearelikelytogaintractioninthefuture.Ofcourse,anysuchmeasuresentailpotentialrisks,notonlyattheoperationallevel,butalsointermsofcorporategovernanceandpossiblelossofknow-how,butthese are an inevitable part of the process of banks ad‑justingtheircoststructurestotheirnewbusinessmodels.Cost/income ratios of Belgian banksmay recently haveslottedback into their pre-crisis levels, but this recoveryhas in part been enabled by temporary factors, whichsuggests that the sector’s cost structure isdue formoresweepingreforms.Thetransitiontofreshbusinessmodelsgoes hand in hand with a focus on more diversifiedsourcesof income,preferably fromactivities thatgener‑atefeesandarelesstiedinwithinterestrates.Infact,the2016fallinothertypesofincomesoundsacautiousnoteontheabilityofbankstofindstructuralreplacementsforinterest income.

If, despite all their measures, banks find they cannotachieve sufficient profitability, they might be temptedto seekyieldsbyallocatinga largerproportionof theirportfolios to riskier, but higher-yielding sub-segments.Todate,therehasbeenlittlesignofthis intheBelgianbankingsector,somethingthatiscorroboratedbyasig‑nificantincreaseinthemoneyBelgianbanksplacewithcentral banks.

Belgian banks continue to grant a lot of mortgage loans

More than ever, Belgium’s banks would appear to betargeting the domestic mortgagemarket. In 2016, low

CHART 56 BANKINGSECTORINBELGIUMCOMMANDSHIGHERRETURNONEQUITYTHANINMOSTEUROAREACOUNTRIES

(first-halfdata 2016,in%)

–20

–15

–10

–5

0

5

10

15

20

J

LV MT EE SK SI BE FI AT NL IE FR LU ES CY IT DE PT EL

Euro area average

Return on equity

source : EcB.


interest rates led to a further increase in the mortgage loan portfolio and the total outstanding amount of mortgage loanstoBelgianhouseholdsroseto€184.2 billionattheendofSeptember2016,from€174.9billionatend-Sep‑tember2015,i.e.year-on-yeargrowthofover5%.Grossnew business in mortgage loans worked out at €50.6billionintheperiod,including€14.7billionininternalre‑financing,withexternalrefinancing–i.e.whereborrowersturntoadifferentbank–amountingto€6.9billion.

In2015 and2016, theupswing inmortgage loanswasno longeraccompaniedbya further tighteningof lend‑ing criteria, as it had been between 2012 and 2014.Whilemortgageloanswithtermstomaturityinexcessof25 yearsasaproportionofnewmortgagebusinessonlygrewby2%in2016,therewasnofurtherdeclineinthepercentage of new loanswith a debt-service-to-income(DSTI)ratioinexcessof50%,despitelowinterestrates.Onefindingthatstandsoutevenmoreisthatlowinterest

rates also triggered a further rise in average amounts con‑tracted innew loans. This tooknewmortgagebusinesstohigheraverageloan-to-valueratios(LTVs),i.e.thesizeofthemortgageinrelationtothevalueoftheproperty.Although these trendswerewellwithin the boundariesagreed inbanks’ lendingpolicies, thisdoes imply thatanotinsubstantialportionofnewbusinessstillcommandshighLTVandDSTIratios.

Lending criteria have recently been identified as one ofthe vulnerabilities of the Belgian mortgage market. The surge in mortgage loans further increased household debt burdens, while prices again grew and the Belgianhousingmarketbecamemoreovervaluedin2015,beforeboth slowed down somewhat. These observations havepromptedtheBank,aspartofitsremitasmacropruden‑tial authority in Belgium, to extend a macroprudentialmeasure applied since 2013 and to announce its inten‑tion to implementanewmeasure thatmorespecifically

CHART 57 BELGIANBANKSCOMPLETEDELEVERAGINGPROCESS

(balancesheetstructureofBelgiancreditinstitutions;end-of-perioddata,onaconsolidatedbasis(1);in€billion)

2000

2001

2002

2003

2004

2005

0

200

400

600

800

1 000

1 200

1 400

1 600

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

0

200

400

600

800

1 000

1 200

1 400

1 600

2000

2001

2002

2003

2004

2005

0

200

400

600

800

1 000

1 200

1 400

1 600

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

0

200

400

600

800

1 000

1 200

1 400

1 600

Interbank claims

Loans

Debt securities

Derivatives (3)

STRUCTURE OF ASSETS

2016

Q3

Interbank debts (including central banks)

Deposits and savings certificates (4)

Own funds, minority interests and subordinated debts

Certificates of deposit, bonds and other debt instruments

STRUCTURE OF LIABILITIES

2016

Q3

Other liabilities (2)

Other assets (2)

Derivatives (3)

source : NBB.(1) DatacompiledaccordingtoBelgianaccountingrules(BelgianGAAP)until2005 andaccordingtoIAS/IFRSstandardsfrom2006.(2) “Otherassets”mainlycomprisebalanceswithcentralbanks,shares,tangibleandintangibleassets,anddeferredtaxassets.“Otherliabilities”areprimarilyshortpositions,

liabilitiesexcludingdepositsanddebtinstruments,provisionsandliabilitiesfordefinedbenefitobligations.(3) Derivativesarerecognisedatmarketvalues,including–from2007 –incomereceivableandchargespayable(whicharenotincludedinthedatarelatingto2006).(4) Fromthethirdquarterof 2014,savingscertificatesarenolongerincludedin“depositsandsavingscertificates“,butrankunder“certificatesofdeposit,bondsandother

debtinstruments”.Liabilitieslinkedtotransferredassetsarenolongerrecognisedunder“otherliabilities“,butareincludedunderdifferentitemsontheliabilitiesside.


targetsloanswithhighLTVs.SectionAofthe“Prudentialregulation and supervision” part of this Report offers amore detailed discussion of the various measures.

Exposures to non-financial and financial sectors up again in 2016

This expansion of the domestic mortgage loan portfoliofits intoabroaderprocess that is seeingBelgium’sbanksrefocusontheirtraditionalintermediationactivitiesaswellason their strategicmarkets,andon theBelgianmarketfirstandforemost.Between2008and2013,Belgianbanksdownsizedtheiractivities,especiallythecross-bordertrans‑actionsthathadreachedhighlevelsbeforethefinancialcri‑sis.Theirtotalassetsshrankfrom€1578billionattheendof2007–i.e.458%ofGDP–to€961 billionattheendof2013.TheaggregatebalancesheetoftheBelgianbankingsectorhasbeenataround€1000 billionsincethen:totalassets stood at €1030 billion by the end of September2016,or246%ofGDP.

For the first nine months of 2016, most key asset items recorded a rise on end-2015 figures.

Loans to households,mainlymortgage loans, agreed inBelgium and in some strategic markets outside Belgium rosebyatotal€9billionto€230 billion,andaccountedfor22%oftheBelgianbankingsector’stotalassets.Theportfolio of loans to Belgian non-financial corporationsgrewto€120 billionattheendofSeptember2016–i.e.12%oftotalassets–from€115 billionatend-2015.

Exposures to thepublic sector in the formof loansanddebt securitieswhich, because of the unrest in govern‑mentbondmarketsin2011 and2012,hadlargelyshiftedto Belgian public authorities, fell by €2 billion. This fallparticularlyreflectslowerclaimsontheBelgianpublicsec‑torthankstosalesundertheECB’spublicsectorpurchaseprogramme(PSPP),althoughBelgium’sshareofoutstand‑ingclaimsonthepublicsectorwasstillcloseto50%bythe end september 2016.

Having fallen sharply in 2007, outstanding loans tobanking financial corporations (€94.5 billion at end-September2016)andtonon-bankfinancialcorporations(€27.1 billion) were back on the rise in the first threequarters,upby€12.7billion.Morespecifically,thecoun‑try’sbanksinvestedinthewholesalemarkettheexcessofdeposits garnered from the non-financial private sector,particularlyintheformofreverserepos(fundinggrantedsecuredbycollateral).Aproportionoftherequiredfund‑ingforthisbusinesswasalsotappedintheformofrepos(funding received secured by collateral), which mightpoint toapotential searchforyield–apossibilitywhen

counterparties or collateral differ in quality. Wholesaletransactionsthusalsoincreasedontheliabilitiesside,al‑thoughlessmarkedlysothanontheassetsside.Belgium’sbanks also attracted more ample funding from central banks and locked in the cheap funding provided under theTLTROIIprogramme.

In terms of funding of the Belgian banking sector, themost notable driver in 2016 turned out to be a sharp increaseindepositsofhouseholdsandnon-financialcor‑porations,anotherfeatureoftheBelgianbanks’returntoamore traditional businessmodel. This typeof fundinghasgenerallyincreasedasaproportionofthesector’sbal‑ancesheettotalsince2007andchippedin48%bytheend of september 2016. outstanding household deposits expandedby€19billion inthefirstninemonthsoftheyear,to€359billion,andmainlyinvolvedsightaccounts,assavingsdepositsremainedvirtuallystable.

Balance sheet trends have left quality of assets untouched

The various trends affecting the balance sheets of Belgium’sbanks in2016–e.g. theshift innewmort‑gagebusinesstohigherLTVratios–didnotcauseanydeterioration in thequalityof theassets.At3.4%byend-September2016,thepercentageofimpairedloanswasdownyetagain–from3.6%attheendof2015 –as a lower total of non-performing loans combinedwithanincreaseinthetotalportfolio.Notethattheseconsolidated statistics cover both Belgian portfolios and theforeignportfoliosheldbyBelgianbanks.Generallyspeaking, this latter setofportfolios still hasabiggershare of non-performing loans. More specifically, theratiosharplyimprovedforloanstonon-financialcorpo‑rations.IntermsofBelgianmortgageloans,thequalityofBelgianbanks’portfoliosimprovedslightly–despiteafewremainingpocketsofrisk–asthepercentageofmortgageloansthatwereinarrearsdeclinedto1.1%at the end of October 2016 from 1.2% at the endof 2015.

Belgian banks continue to enjoy comfortable solvency and liquidity positions

Allotherthingsbeingequal,better-qualitydebtorsen‑surea lowerprobabilityofdefault forbankassets, asdeterminedbybanks’internalmodels.However,inrisk-weightedassetcalculations,thispositiveeffectismorethanwipedoutbytheriseinexposuressubjecttocreditrisk.Risk-weightedassetsrosefrom€345 billionattheendof2015 to€358billionattheendofSeptember2016.Atthesametime,commonequityTier1 capital(CET1)went from€53 billion to€55 billion so that


Belgianbanks’CET1 ratiowasunchangedat anaver‑age 15.4%. This puts them, on average at least, ina comfortable solvency position that is amply aheadof regulatory requirements, as EBA stress tests alsoestablished (1).

Asof2016,theCapitalRequirementsDirective(CRDIV)prescribes the gradual implementation of various add‑on buffers. The capital conservationbuffer,which consistsof a fixed margin on top of the required minimum,was set at0.625% for2016andwill gradually rise to2.5%by2019.Acountercyclicalbufferwillneedtobeactivatedintheeventofexcessivelendinggrowthintheeconomy–thiswassetat0%inBelgiumin2016.Lastly,add-onbufferswereimposedoneightbankswithintheframework provided by CRD IV, as these banks havebeen designated as systemically important in Belgium.The Directive also envisaged a gradual phasing‑out of transitionmeasures by 2018. If CRD IV had been fullyinplacein2016,CET1 ratioswouldhavemerelyedgeddownto15.2%,whichisstillarespectablelevel.

Although Belgium’s banks saw the proportion of en‑cumbered assets rise in 2016 due in part to increased repo transactions, the sector continued to enjoy acomfortable liquidity position. On average, the short-termliquiditycoverageratio (LCR)fellbacka little–to135%attheendofSeptember2016from137%attheendof2015 –butwasstillwell inexcessofregulatoryrequirements.

4.4 Insurancesectorprofitabilitydeclinedinboth2015 and2016

Belgium’sinsurancecompanieshaveseentheirnetresulton a downward trajectory since 2012. In 2015, it con‑tracted to €1.2 billion,while return on equity –whichhadaveraged20%beforethefinancialcrisis–shrankto8.2%.Partialdataavailable for thefirstninemonthsofthe year suggest that this negative trend persisted into2016.AsetofdatacollectedforasampleoffiveBelgianinsurance groups shows annualised return on equity tohaveedgedbackagain,to7.4%.However,thesefiguresneed to be treated with some caution as they do notincludethefullsector.Thatsaid,thesearetheonlymean‑ingful data available, as detailed three-month figuresbasedonindividualstatutoryaccountsofallofBelgium’sinsurance companies have ceased to be available since thenewSolvencyIIregulatoryframeworkcameintoforcein 2016.

While the sector’s non-technical result –mainly incomefrom investment not imputed to life and non‑life insur‑ance activities, aswell as exceptional items and taxes –remained negative in 2016 (-€0.6 billion in 2015), thedecline in profitability was chiefly attributable to de‑teriorating results in the life insurance sector. Despite the locking-in of significant capital gains on sales, thedownturnto€0.2 billionin2015 reflectedfallingincomefrom investment – due topersistently low interest ratesandagraduallymoreconcretemanifestationofreinvest‑mentrisk,aswellas lowerpremiumincome. Inthefirstninemonths of 2016, income from gross life insurancepremiums continued to slow as the low interest rate

(1) Formoreinformationaboutthesetests,seeSectionC.2 ofthisReport’s“Prudentialregulationandsupervision”part.

Table 13 BREAKDOWN OF TIER 1 CAPITAL AND RISK‑WEIGHTED ASSETS, SOLVENCY AND LIQUIDITY RATIOS

(end‑of‑period data, on a consolidated basis, in € billion, unless otherwise stated)

2012

2013

2014

2015

September 2016

Tier 1 capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55.9 55.6 53.4 55.1 57.0

of which :

Common equity Tier 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . − − 51.5 53.3 55.2

Risk‑weighted assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.7 339.4 349.8 345.4 357.7

of which :

Credit risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301.0 287.7 290.1 282.8 298.3

Tier 1 ratio (in %) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.9 16.4 15.3 16.0 15.9

Common Equity Tier 1 ratio (in %) . . . . . . . . . . . . . . . . . . . . . . . . . − − 14.7 15.4 15.4

Liquidity coverage ratio (in %) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137.0 135.0

Source : NBB.


environmentmadetheseproductslessattractiveinsavers’eyes.Comparedwiththeyear-earlierperiod,thisincomecontractedby3.2%to€10.7billion.

Bycontrast,non-lifeprofitabilitykickedsharplyahead in2015 on the back of higher premium income. Non-lifeinsurance activities returned profits of €1.6 billion intheyear.However,2016sawareversalof fortunes:netpremium income fell by 3.6% in the first ninemonthscomparedwiththeyear-earlierperiod.Butthenetcom‑binedratio,whichcomparesthetotalcostofclaimsandoperating expenses to net premium income, remainedclose to 100%, showing that insurance companies aremaintaining a sound balance between insurance costsand premium income. It is worth noting, though, thatthesegenerally solid resultswerenot achieved in all in‑suranceclassesofnon-lifebusiness:insome,costswerehigher than premiums.

Insurance companies’ profitability squeezed by lower premiums and high guaranteed returns

Investment income is being kept down by falling pre‑miums but also by the low interest rate environment,particularly in life insurance,wheretheaverageduration

ofliabilitiesexceedsthatonassetsandwhereguaranteedreturns to be honoured remain high.

Totalinventoryreservesforguaranteed-returncontractsfellfrom€168.2 billionto€164.3 billionbetweentheendof2014 andend-2015,thelatestperiodforwhichdetailed annual data are available. This contraction in reserves was caused by individual insurance, wherereservesnarrowedbynearly5%.Group insurance,bycontrast, recorded an increase of 3%. Although theshare of contracts with guaranteed returns of over2% is steadily decreasing in favour of lower-returnagreements,contractsforwhichtheguaranteedreturnon accrued and/or yet to be accrued reserves (basedon future premiums) exceeded 4.5% amounted to€24.5 billion,or15%ofinventoryreserves.Thecom‑parablefigurefor2014 was€27.4 billionandfor2011€31.3 billion.

Ifcurrentlowinterestratesareheretostay,significantamountsofhigh-rated securities (AAAorAA) comingtomaturitywillhavetobereinvestedinlower-yieldingassets. There is a real risk, then, that the effectivereturnonassetswillnotbeenoughtocovertheguar‑anteed interest rates on contracts entered into earlier.

CHART 58 FALLINLIFEANDNON-LIFEPREMIUMINCOME

(non-consolidateddata;in€billion,unlessotherwisestated)

2001

2003

2005

2007

2009

2011

2013

2015

0

5

10

15

20

25

30

2001

2003

2005

2007

2009

2011

2013

2015

0

2

4

6

8

10

12

14

90

95

100

105

110

115

120

0

2

4

6

8

10

12

14

0

2

4

6

8

10

12

14

0

5

10

15

20

25

30

2015

2016

2015

2016

LIFE INSURANCE PREMIUMSNON‑LIFE INSURANCE PREMIUMS AND COMBINED RATIO

(1)

Nine first months

(2)Nine first months

(2)

Class 21 (individual)

Total

Class 21 (group)

Class 23

Other classes

Non‑life insurance premiums (left‑hand scale)

Combined ratio (in %) (right‑hand scale)

source : NBB.(1) Thecombinedratiorelatesthesumofthecostofclaimsplusoperatingexpensestonetpremiumincome.


The outstanding total of life insurance contractswithguaranteed returns and the actual rates paid on them arethereforevery important riskparameters for insur‑ance companies at a time of falling interest rates on risk‑free investments.

Insurers respond by adapting their products and lowering costs

Persistently low interest rates are forcing insurancecompaniestooffercontractsmore in linewithmarketconditions, taking the average guaranteed return ofclass 21 agreements down from 2.91% in 2014 to2.82% in 2015 – or, more specifically, from 2.72%to2.64%forindividualinsuranceandfrom3.25%to3.19%forgroupinsurance.Atthesametime,insurersimposed time limits on guarantees and promoted sales ofclass23 agreementslinkedtoinvestmentfundsandwithout any guaranteed returns. In addition, hybridproductsweredeveloped,combiningfeaturesofclasses21 and23.

Toensuretheirprofitabilitypotential,insurancecompa‑nies have been further adapting their activities and cost structures. Business models guaranteeing sustainable profitability, even in adverseeconomic conditions, area necessary precondition for maintaining the stabilityof the broader sector. In this context and using theresults of transversal analyses of the insurance sector,the Bank decided to take specifically targeted actionstobolster, inasustainableway,theprofitabilityofse‑lected institutions.

In addition, it may prove necessary for insurers toslash operating expenses further, possibly throughconsolidations, in order to align cost structures withtheshrinkageinactivityvolumes.Variousinsurershaveannounced restructuring plans,which should be seenagainstabackdropoffallingemploymentintheinsur‑anceindustry.Assuraliastatisticsputaveragenumbersof employees at insurance companies at 22800 in2015,downfrom23300 in2008.Theaveragenumberof intermediaries – i.e. agents and brokers – shrank to 9600 from10900.

CHART 59 CONTINUEDHIGHPROPORTIONOFCONTRACTSWITHHIGHGUARANTEEDRETURNS

(breakdownoflifeinsuranceinventoryreservesbyguaranteedreturnsperindividualcontract;end-of-perioddata,in€billion)

1050 15 20 25 30 35 40 45 50

[ 0.00 % ; 0.50 % ]

] 0.50 % ; 1.00 % ]

] 1.00 % ; 1.50 % ]

] 1.50 % ; 2.00 % ]

] 2.00 % ; 2.50 % ]

] 2.50 % ; 3.00 % ]

] 3.00 % ; 3.50 % ]

] 3.50 % ; 4.00 % ]

] 4.00 % ; 4.50 % ]

> 4.50 %

Group insurance

Individual insurance

2014 20112015

source : NBB.

Table 14 GUARANTEED RETURNS AND ADDITIONAL PROVISIONS

(non‑consolidated data ; in %, unless otherwise stated)

2011

2012

2013

2014

2015

Long‑term yields (OLO – ten years) . . . . . . . . . . . . . . . . . . . . . . . . . 4.01 2.06 2.56 0.83 0.98

Average guaranteed returns existing contracts . . . . . . . . . . . . . . . 3.17 3.12 3.04 2.91 2.82

Group insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.59 3.54 3.41 3.25 3.19

Individual insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.01 2.95 2.88 2.72 2.64

Additional provisions (in € billion) . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 3.0 4.1 5.5 6.6

Source : NBB.


Range of regulatory measures put in place

Previous prudential rules oblige insurance companies to book additional annual provisions in their accounts to ensure they can meet their liabilities despite lowinterest rates. For reasons of caution and continuity,ithasbeenagreedthattheseprovisions,forwhichnoexemptions havebeengranted since2013 andwhichstoodat€6.6billionby theendof2015,will stay inplace over the period of the transition to Solvency II.Infact,theseadditionalprovisionswillalsoneedtobetopped up unless a conditional exemption is granted(renewableannually).

The maximum reference rate for long-term life insur‑ance contractswas cut to 2% in early February, from3.75%,andwillbekeptatthesamelevelin2017.From1 January 2016, employer-guaranteed returns on sup‑plementarypensionsweresetat1.75%formemberandemployer contributions.Without these cuts, employersmight well have scrapped the supplementary pensionsystemaltogetherastheywouldhavebeeninthesameposition as insurers and unable to pay the guaranteedreturns.

TheBankhasrecommendedthatinsurerscurbanydis‑tribution of profits to shareholders and policy-holdersin order to safeguard their fundamental resilience in the longer term,while also suggesting that they con‑sider their interestmargin levels. In this context, newregulatoryprovisionsgoverningprofit-sharingwereputinplace,specifyingthattheBankmayconstrainprofit-sharing in some instances, factoring in insurer profit‑abilityandsolvency.Improvingthesolvencyoffinancialinstitutionsbyretainingearningsand/orraisingcapitalwillbethekeytoinsurersbeingabletomeetthenewchallengesoftheeconomiccycleandevermorerigor‑ous regulatory requirements. Given these constraints,the Bank has cautioned reticence on the part of insur‑erswhenrealisingcapitalgains,as this shouldfit into

a preventive strategy primarily focused on meetingcontractual obligations.

SolvencyIIhasbeeninplacesince1 January2016.Thisnew regulatory framework,whichhasbrought radicalchangetothesector,obligesinsurerstomaintaineligi‑bleownfundstocoversolvencyrequirements,knownas the solvency capital requirement (SCR). Portfoliosnow being valued on the basis ofmarket pricesmaybe causinggreaterfluctuations in the equity amountsclassifiedbutshouldalsomakeiteasiertoassessthefi‑nancialrisksfacinginsurers,whichinturnshouldmakeit easier to anticipate the impactof low interest ratesontheirfuturesolvency.

Solvency of Belgian insurance is generally robust, but remains varied

The sector’s solvency ratio, calculated as eligible ownfunds relative to SCR, stood at 165.4% by the end ofSeptember2016.Ratiofluctuations in thecourseof theyearmaybeexplainedbothbychanges inshareholders’equity–whichgoesupanddownwiththevolatilityofthefinancialmarkets–andtrendsintheSCRamountoftheindividualcompanies.Averagesolvencylevels inthesec‑torremaincomfortabledespitethedownturninthethirdquarter.Currentlevelsappearrobustenoughforinsurerstofacedownalengthyperiodoflowinterestrates,asiscorroboratedbyEIOPAstresstests(1). This general observa‑tion nevertheless masks a degree of divergence : although quitea few insurersare verywell capitalised indeed,ananalysis of solvency margins under Solvency II suggeststhat some are much more vulnerable. The Bank has urged the relevant institutions to take specificmeasurestomeettherequirements.Thelowinterestrateenviron‑mentinwhichthenewprudentialregulatoryframeworkis put into force might well prompt a restructuring at

Table 15 OWN FUNDS, CAPITAL REQUIREMENTS AND SCR RATIO

(in € billion, unless otherwise stated)

01‑01‑2016

31‑03‑2016

30‑06‑2016

30‑09‑2016

Solvency requirements . . . . . . . . . . . . . . . . . . . . . . . 17.9 18.3 18.0 19.2

Total eligible own funds . . . . . . . . . . . . . . . . . . . . . 32.9 31.5 32.8 31.8

SCR ratio (%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.7 171.8 182.0 165.4

Source : NBB.

(1) Formoreinformation,seeSectionDofthisReport’s“Prudentialregulationandsupervision”part.


thesector’sweakestentitiesortakeoversbyother,morerobust, players. The Bank is watching events closely astheyunfold.

Insurers’ balance sheet composition does not suggest general search for yield

Inviewofthelowreturnsontheirtraditionalportfolios,some insurance companies are attempting to shift their investmenttohigher-yieldingassetsinordertomeettheirobligations arising from life insurance contracts. for sev‑eralyearsnow,governmentbondshavebeenfallingasaproportionoftheseportfolios,despitedevelopingslowlyinstepwithlonger-termassets.BytheendofSeptember

2016,bondsstillaccountedforthemajorityofportfolioholdings,at€209.8billion,withgovernmentpapermak‑ingup€147.5 billionandcorporatebonds€62.4 billion,and over 60% of these securities commanding highratings(AAAorAA).Theloanportfolio,whichhasbeengrowingrobustlyforanumberofyearsnow,wasworth€27billionbytheendofthethirdquarterof2016,ac‑counting for 10% of investment. The rise of this typeof portfolio reflects how insurers are diversifying theirinvestmentandunderscoreshowsomearecontinuingtosearchforyield.Thevalueofassetsheldtocovertechnicalprovisionsforclass23 contractsamountedto€31 billion,withtheseassetsmostlyheldinunitsofundertakingsforcollectiveinvestment(UCIs).

Table 16 COMPOSITION OF INSURANCE COMPANIES’ BALANCE SHEETS

(2016Q3, market value, in € billion)

Assets

Liabilities

Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279.4 Technical insurance reserves, excluding class 23 . . 245.5

Government bonds . . . . . . . . . . . . . . . . . . . . . . . 147.5 Life insurance reserves, excluding health insurance . . . . . . . . . . . . . . . . . 211.7

Corporate bonds . . . . . . . . . . . . . . . . . . . . . . . . . 62.4 Non‑life insurance reserves, excluding health insurance . . . . . . . . . . . . . . . . . 17.6

Lending and mortgage loans . . . . . . . . . . . . . . . 27.0 Health insurance reserves . . . . . . . . . . . . . . . . . . 16.2

Equities and participating interests . . . . . . . . . . 16.1

Investment funds . . . . . . . . . . . . . . . . . . . . . . . . . 13.4

Property . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8

Other investment . . . . . . . . . . . . . . . . . . . . . . . . . 5.3

Other assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28.5 Other liabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33.7

Assets linked to class 23 activities . . . . . . . . . . . . . 31.0 Class 23 technical insurance reserves . . . . . . . . . . 30.5

Total assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338.9 Total liabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309.7

Surplus of assets over liabilities . . . . . . . . . . . . . 29.1

Source : NBB.

Public finances

133Economicandfinancialdevelopments ❙ PUBLICFINANCES ❙

5. Public finances

5.1 Deterioration in Belgian public finances

Belgiumended 2016onageneralgovernmentdeficitof2.8% of GDP. This was a substantial deterioration com‑paredwith 2015,withanincreaseof0.3 percentagepointsofGDP,takingitonceagainveryclosetothethresholdof3%ofGDP set by the European fiscal framework as itsreferencevaluefordeterminingexcessivedeficits.Generalgovernmentdebtalsoroseagain,to106.6%ofGDP.

Thehigherdeficitwastheresultofadropinrevenuecom‑binedwithaslightriseinprimaryexpenditureexpressedasapercentageofGDP;thisinturncausedtheprimarybal‑ancetofallby0.6 percentagepointofGDP,takingitbackintonegativeterritoryforthefirsttimesince2012.Interestchargesfellfurther,by0.3 percentagepointofGDP.

The structural overall balance, which is obtained by cor‑rectingthebudgetfortheeffectsofcyclicalandtemporary

factors, also deteriorated in 2016 compared with theprevious year. The structural budget deficit widened by0.2 percentagepointofGDP,slightlylessthanthenominaloveralldeficit.Non-recurrentfactorshadapositiveimpactof0.1 percentagepointofGDPonthenominalbalance,lessthanthe0.3 percentagepointboostin 2015.Bycon‑trast, the impactof thebusinesscycleonpublicfinanceswas virtually unchanged. The structural primary balance,whichisabetterreflectionofdiscretionaryfiscalpolicybe‑causeitisimmunetotheinfluenceofchangesininterestrates,deterioratedby0.5 percentagepointofGDP.All inall,then,Belgium’spublicfinancesworsenedin 2016.Thismeans that there has been no structural improvement in theoverallbalanceforthelastthreeyearsandthatthereisaneedforfurtherconsolidationofpublicfinances.

Fiscal targets missed again in 2016

The fiscal policy of Member States has to conformwith Europeanfiscal rules as laiddown in the Treaty of

Table 17 GENERAL GOVERNMENT OVERALL BALANCE AND DEBT

(in % of GDP)

2000

2012

2013

2014

2015

2016 e

Revenue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49.0 51.6 52.7 52.0 51.3 50.9

Primary expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.4 52.2 52.4 51.8 50.9 51.0

Primary balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6 −0.6 0.3 0.2 0.5 −0.1

Interest charges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.7 3.6 3.3 3.3 3.0 2.7

Overall balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.1 −4.2 −3.0 −3.1 −2.5 −2.8

General government debt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108.8 104.1 105.4 106.5 105.8 106.6

Sources : NAI, NBB.


MaastrichtandintheStabilityandGrowthPact.BelgiumcurrentlyfindsitselfinthepreventivearmofthePact,butmustofcoursealsomeettheconditionslaiddowninthecorrective arm. Thismeans that the budget deficitmay

notexceed the referencevalueof3%ofGDPand thatpublicdebtmustbebroughtdownto60%ofGDPatasatisfactorypace.

In the event, however, Belgium failed to lift its struc‑turaloverallbalancein 2016bytherequiredminimumof0.6 percentagepointofGDP,eventhoughtheApril 2016stabilityprogrammehadassumedthatthis improvementwouldbeachieved.Thestructural improvement in 2015wasalsowellbelowthethreshold.

Basedonitsautumnforecast,theECdecidedthattherewas a risk of a significant deviation from the requiredadjustment path towards the medium-term budgetaryobjective (MTO), a key requirement of the Stability andGrowthPact’spreventivearm.Thecurrentobjectiveistoachieveastructurallybalancedbudget.TheECtakestheviewthattheriskofasignificantdeviationin 2016meansasimilar riskexistsover 2016and 2017taken together.Thisconclusionremainsunchangedifthedirectbudgetaryimpactoftherefugeecrisisandadditionalsecuritymeas‑uresarefactoredin,whichrequireadditionalexpenditureestimatedat0.2%ofGDPin 2016.

Until end-2016, Belgiumwas subject to the transitionalruleforthedebtcriterion,whichsetsaspecifiedminimumannual improvement in the structural balance aimed at creating sufficient momentum in reducing the level ofdebt. According to an analysis by the EC based on its

CHART 60 DETERIORATIONINBOTHSTRUCTURALPRIMARYBALANCEANDSTRUCTURALOVERALLBALANCEIN 2016

(in%ofGDP)

2000

2002

2004

2006

2008

2010

2012

2014

–6

–4

–2

0

2

4

6

8

2016

e

Interest charges

Structural primary balance (1)

Structural overall balance (1)

Sources:EC,NAI,NBB.(1) Thecyclicalcomponentofthestructuralprimarybalanceandthestructural

overallbalanceisdeterminedbasedonECmethodology.

Table 18 TARGETS FOR THE OVERALL BALANCE OF BELGIAN GENERAL GOVERNMENT

(stability programme targets; unless otherwise stated ; in % of GDP)

2013

2014

2015

2016

2017

2018

2019

Nominal balance

April 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.5 −2.0 −0.5 0.4

April 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.1 −1.4 −0.4 0.6

April 2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.5 −2.0 −1.0 −0.2

April 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.5 −1.4 −0.4 −0.2

November 2016 (draft budget) . . . . . . . . . . . . . −2.9 −1.7 −0.4 −0.3

p.m. Actual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −3.0 −3.1 −2.5 −2.8 e

Structural balance

April 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −1.8 −1.2 0.0 0.75

April 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −1.4 −0.7 0.0 0.75

April 2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.0 −1.3 −0.6 0.0

April 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −1.7 −0.8 0.0 0.0

November 2016 (draft budget) . . . . . . . . . . . . . −2.4 −1.1 0.0 0.0

p.m. Actual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.7 −2.9 −2.6 −2.8 e

Sources : EC, FPS Budget and Management Control, FPS Finance, NAI, NBB.


autumnforecast,itlookedasifBelgiumwouldnotmakesufficientprogressin 2016tocomplywiththedebtrule.Consequently, the ECwill prepare a new report pursu‑ant to Article 126, paragraph 3 of the Treaty on theFunctioningoftheEuropeanUniontodeterminewhetherBelgium,takingintoaccountallrelevantfactors, iscom‑plyingwiththedebtcriterion.

Given the absence of a structural improvement in Belgium’s public finances in 2016, implementation ofthe measures needed to achieve a structural balance was once again pushed into future years, continuingthetrend inrecentyearsofconsistentlydeferringtargetachievement.

Structural budget balance to be achieved in the medium term

The government must achieve and maintain a healthybudgetbytakingdeterminedactiontoreducethebudgetdeficitresolutelyandsustainablytoapositionofstructuralbalance.Thiswouldlowergovernmentdebt,whichiscur‑rentlywellabovethethresholdof60%ofGDPsetinthe

TreatyofMaastricht,allowingthenecessarybufferstobeestablished for the future and creating thepolicy scopeneeded to meet the costs of an ageing society. Thosecostsremainconsiderable,thoughtheyaremitigatedbypension reform measures.

If Belgium adheres to the European fiscal framework,itspublicfinancescouldbemadesustainable.Asnoted,achieving the MTo is at the heart of the preventive arm oftheStabilityandGrowthPact.InitsApril 2016stabilityprogramme, Belgium revised this objective downwards,fromastructuralsurplusof0.75%ofGDPtoapositionofstructuralbalance.Thisobjectiveapplies from 2017andisinlinewithminimumMTOsascalculatedbytheEC,asit is more ambitious. Revising Belgium’sMTOwas pos‑sible largely thanks topension reformswhich lower theprojectedcostsofpopulationageing.Ifitsteadilyfollowstheroutetowardsastructurallybalancedbudget,Belgiumshouldcreateabudgetarymarginthatcouldbeusedtoaccommodate the costs of population ageing. The funds releasedbyfalling interestchargesfollowingaccelerateddebtreductioncouldbeusedtocoverthesecosts,whilethestructuralbalanceobjectivecouldbeeasedassoonas

CHART 61 BELGIUMNEEDSTOMAKERELATIVELYMARKEDIMPROVEMENTINSTRUCTURALBALANCETOACHIEVEMEDIUM-TERMOBJECTIVE (1)

(in%ofGDP)

ES BE PT SI FR IT SK IE MT FI LV AT NL LT CY EE DE LU–4

–3

–2

–1

0

1

2

–4

–3

–2

–1

0

1

2

Medium-term objective (MTO)

Structural budget balance in 2016 e

Consolidation effort needed to meet MTO

Budgetary margin

Sources:EC,NBB.(1) ThechartcontainsdataforcountriesintheeuroareaexcludingGreece,forwhichnonewMTOhasyetbeendefinedbecauseitissubjecttoanadaptationprogramme.

Countriesarerankedbysizeoftherequiredconsolidationeffort.


thedebthasbeenreducedtoasubstantially lower levelthan at present.

Belgium’s structural balance for 2016 is projected toshow a deficit of 2.8% of GDP. Based on the Bank’sDecember 2016economicprojections,thedeficitshouldfallto2.1%ofGDPin 2017andremainaroundthatlevelinthefollowingtwoyears.ToachieveitsMTO,therefore,Belgium will need to improve its balance substantially.Basedondatafor 2016, itwillneedtomakemorepro‑gressthaneveryothercountryintheeuroareawiththeexception of Spain. Some countries – the Netherlands,Lithuania,Cyprus, Estonia,Germany and Luxembourg –havealreadyachievedorexceededtheirMTOsandcon‑sequentlyhavesomebudgetarymargin.

Worsening federal government deficit

Sound management of public finances is the essentialduty of all governments and government sub-sectors.However, the sixth State reform, the budget part ofwhichlargelycameintoeffectin 2015,ledtoanumberof transfers of powers and related resources in 2016.For example, responsibility for hospital infrastructurewas transferred from the social security sector to theCommunitiesandRegions.Anewstructuralreductionof€1.25 billioninfiscaltransfersfromfederalgovernmenttotheCommunitiesandRegionswasalsoappliedpursu‑anttotheSpecialFinanceActof6 January 2014,takingthecontributionmadebytheCommunitiesandRegionstotheausteritymeasuresto€2.5 billion.

Against this institutional backdrop, the increase in thenominal deficit of the Belgian government can be

attributedentirelytothedeteriorationoftheEntity Iac‑counts,comprisingthefederalgovernmentandsocialse‑curity.Giventhereductionintheborrowingrequirementof Entity II,which ismade up of theCommunities andRegionsaswellaslocalgovernment,EntityIaccountsforvirtuallytheentiregeneralgovernmentdeficit.

The federal government deficit widened from 2.4% ofGDP in 2015 to 2.7% in 2016. Social security closedthe financial year on a balanced budget. Despite theircontributiontotheausteritymeasuresreferredtoabove,theCommunitiesandRegionsmanagedtocuttheirdefi‑cit from0.3 to 0.2 percentage points ofGDP. The fourmain federated entities, namely the Flemish and FrenchCommunities and the Walloon and Brussels-CapitalRegions, ended 2016 in deficit,while local governmentwasabletoconsolidatethesurplusof0.1%ofGDPre‑cordedin 2015.

Solid internal fiscal coordination essential

Thecomplexbudgetaryinteractionsbetweenthedifferentpolicylevelsrequireefficientoperationalcoordination.

This coordination is organised and formalised in the co‑operation agreement signedon13 December 2013 be‑tween the federal government, the Communities, theRegions and the Community Commissions. The agree‑ment stipulates that when the stability programmesare drawn up, nominal and structural annual budgettargets will be allocated between the various levels ofgovernment on the basis of a recommendation by thePublicSectorBorrowingRequirementsectionoftheHighCouncilofFinance.Thisallocationhastobeapprovedby

Table 19 OVERALL BALANCE OF GENERAL GOVERNMENT AND BY SUB‑SECTOR

(in % of GDP)

2013

2014

2015

2016 e

Entity I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.6 −2.5 −2.3 −2.7

Federal government (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −2.5 −2.5 −2.4 −2.7

Social security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0 0.0 0.1 0.0

Entity II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.5 −0.5 −0.2 −0.1

Communities and Regions (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.2 −0.4 −0.3 −0.2

Local government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.2 −0.2 0.1 0.1

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −3.0 −3.1 −2.5 −2.8

Sources : NAI, NBB.(1) With effect from 2015, these figures include the advances on the regional additional percentages on personal income tax although, according to the methodology of

ESA 2010, those advances are regarded as purely financial transactions and the regional additional percentages are only taken into account at the time of collection. This approach is in line with that observed in developing fiscal targets in the recommendations of the Public Sector Borrowing Requirement section of Belgium’s High Council of Finance, as well as in stability programmes.


theConsultativeCommittee,abodycomprisingthePrimeMinister and the Minister‑Presidents of the communities and Regions.

Asin 2015,theConsultativeCommitteewasagainunabletoagreeontheallocationofthebudgettargetsin 2016.TheCommunitiesandRegionsonlytooknoteofthedistributionproposed by the federal government, which prescribes astructural balance for the general government by 2018,matchingBelgium’sMTO.Thestabilityprogrammefor thesub-sectorsalsoreferstoastructuralbalance,butdoesnotstipulateaprecisephasingorquantifiedobjectives for theindividualCommunitiesandRegionsfor 2016.

Consultationswerehindered in 2015and 2016byun‑certaintyresultingfromtheimplementationofthesixthStatereform,relatingtoaspectssuchasthecalculationof the advances on the regional additional percentages onpersonal income taxand the fundingof transferredpowers in relation to hospital infrastructure. Oncethese issues have been resolved, the members of theConsultativeCommitteewillneedtoforgecooperationagreements during the discussions on future versions of the stability programmes. Moreover, this concern wasincluded in the latest recommendations to Belgium bythe Eu council.

5.2 further reduction in revenues

Governmentrevenuesfellby0.4 percentagepointofGDPin 2016,continuing thedownward trendseen in recentyears.Thedeclinewastheresultofpolicymeasurescom‑binedwithunforeseenfactors.

Revenuefromleviesonearnedincomeshrankby0.9 per‑centagepointofGDP,almostentirelyduetothemeasuresreferredto intable21 inrelationtothetaxshift,whichgreatlyreducedleviesonemploymenttoimprovebusinesscompetitiveness,promoteemploymentandboosthouse‑hold purchasing power. Changes to income tax bandsintended tobolster purchasingpower, especially of em‑ployeesonlowandmiddleincomes,combinedwithanin‑creaseintax-deductibleprofessionalexpenseallowances,resulted in a reduction in revenue from personal income taxof0.3 percentagepointsofGDP.Socialsecuritycon‑tributionsfellby0.5 percentagepointofGDPduetothecutintheemployer’scontributionratefrom1 April 2016,0.2 percentagepointofwhichwasfundedbyrestrictingwagesubsidiesthroughpartialremittanceofpayrolltax.

Thetaxshiftmeasureswereintendedtoleadtofurtherre‑ductionsinrevenuefrompersonaltaxesandsocialsecuritycontributionsoverthecomingyears.Thisloweringofthe

Table 20 GENERAL GOVERNMENT REVENUE (1)

(in % of GDP)

2012

2013

2014

2015

2016 e

Fiscal and parafiscal revenue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44.3 45.2 45.0 44.5 44.0

Levies weighing chiefly on earned income . . . . . . . . . . . . . . . . . 26.0 26.5 26.3 26.0 25.2

Personal income tax (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5 11.8 11.7 11.5 11.1

Social contributions (3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.5 14.7 14.5 14.5 14.0

Taxes on company profits (4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.0 3.1 3.2 3.4 3.6

Levies on other incomes and on assets (5) . . . . . . . . . . . . . . . . . . 4.0 4.4 4.4 4.2 4.2

Taxes on goods and services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 11.2 11.1 10.9 11.1

of which :

VAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.9 7.0 6.9 6.7 6.8

Excise duties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 2.1 2.1 2.1 2.2

Non-fiscal and non-parafiscal revenue (6) . . . . . . . . . . . . . . . . . . . . . 7.3 7.5 7.1 6.8 6.9

Total revenue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.6 52.7 52.0 51.3 50.9

Sources : NAI, NBB.(1) In line with ESA 2010, total revenue of general government does not include the proceeds of customs duties transferred to the EU nor the revenues levied directly by the EU.(2) Mainly payroll tax, advance payments, assessments and additional percentages on personal income tax.(3) Including the special social security contribution and the contributions of people not in work.(4) Mainly advance payments, assessments and withholding tax.(5) Mainly withholding tax on income of individuals, withholding tax on income from immovable property (including the proceeds of additional percentages),

inheritance taxes and registration fees.(6) Income from assets, imputed social contributions, current transfers and capital transfers from other sectors, plus sales of goods and services produced, including revenues on

guarantees granted by the State on interbank loans.


taxonemployment is intended toaddress the fact thatthese levies in Belgium are among the highest in the euro area:onthemostrecentECstatistics,theimplicitlevyonemploymentinBelgiumwas8.3 percentagepointshigherin 2014thantheaverageacrosstheeuroarea.

Lower taxonemploymentas implemented in 2016waspartiallyoffsetbyamovetowardsataxbasewithalesssevereimpactoneconomicactivity.Anexamplewastheincrease in taxesongoods and servicesby0.1 percent‑agepointofGDP.VAT receiptswentup slightly,mainlyduetothe liftingoftheVATrateondomesticelectricityconsumption on 1 September 2015, the full impact ofwhichwasfelt in 2016.Revenuefromexcisedutiesalsorosemoderatelyfollowingincreasesinthedutyondiesel,tobaccoandalcohol.However,themeasurestakensofartoraiseincomesarenotenoughtofundthetaxshiftfully.Consequently,thetaxshiftlargelyexplainsthesharpfallin the revenue ratio recorded in 2016 and projected tocontinueinthecomingyears.

Taxes on company profits increased by 0.2 percentagepointofGDP,dueamongotherthingstotheclaw-backofthetaxbreaksgrantedtocompaniesbythegovernmentviathesystemofexcessprofitrulings,whichwerejudgedbythe Ec to constitute illegal state aid. The reduction in the reference rates fornotional interestdeductions, linked to

thesharpdeclineinyieldsonBelgianten-yeargovernmentbonds,pusheduptheamountofcorporationtaxpayable.

Leviesonother incomeandonassets remainedvirtuallyunchanged. The impact of the rise in the standard rate of withholdingtax,from25%to27%,whichcameintoef‑fecton1 January 2016,wasoffsetbyareductioninyieldsduetolowerinterestrates.TheschemeenablingSMEstoformcashreserves inexchangeforataxrateof10%isalsoleadingtoshrinkingrevenuesfromwithholdingtax,asthisisnolongerpayablewhenacompanyiswoundup.

MeasurestakenbytheCommunitiesandRegionsboostedtax revenuesby0.1 percentagepointsofGDP.Thiswaslargely achieved through reform of the contribution tothe Energy Fund in the Flemish Region (essentially anannual levywhenconnectingtotheelectricitygrid).Theproceedsmainlygotowardsreducingthedebtmountainof the transmissionsystemoperator thathasbuiltup inrecentyearsthroughfinancingrenewableenergyprojects.Thechanges togift taxes inallRegions,coupledwithalessgeneroussystemofmortgageinterestrelief(housingbonus) in theWalloonandFlemishRegions,alsohelpedpushuptaxrevenues.

Non-fiscalandparafiscalrevenuesroseby0.1 percentagepointsofGDP,principallyduetotheintroductionbytheRegions of a kilometre‑based charge for lorries.

5.3 spending cuts have ground to a halt

The moderation of primary expenditure seen in recentyearsweakened in 2016, leaving expenditure as a pro‑portionofGDPvirtuallyunchanged.Nominally,spendingtherefore rose at a comparable rate to the growth ineconomicactivity.

Thegovernments formed following theMay 2014elec‑tions implementednumerousausteritymeasurestocurbthe growth in primary expenditure. The federal govern‑ment has committed to reducing its operating costs sub‑stantiallybydownsizingthecivilserviceandbuyingfewergoodsandservices.Theindexjumpimplementedin 2015fitsinwiththisaim.Highersocialsecurityspendingisbe‑ingaddressedbyaseriesofmeasuresaimedamongotherthings at slowing the rise in health care costs. Like thefederal government, theCommunities andRegions alsodecidedtointroduceausteritymeasures,partlybycuttingsubsidies.Finally,localgovernmentalsoneededtomakesavingstoremainfinanciallyhealthy.

The consolidation of public finances through spend‑ingmeasureswas impededby a number of unforeseen

CHART 62 HIGHLEVIESONEMPLOYMENTINBELGIUM

(implicittaxrateonemployment(1),in%ofwagecosts)

2002

2004

2006

2008

2010

2012

2014

32

34

36

38

40

42

44

2016

e

Three main neighbouring countries (2) (3)

Belgium

Euro area (2)

8.3

6.5

Sources:EC,NBB.(1) Definedasthetotalleviesonincomefromemploymentpaidtothegovernment,

dividedbythewagebill.Calculatedonthebasisofthenationalaccounts.(2) Unweightedaverages.(3) France,GermanyandtheNetherlands.


Table 21 MAIN FISCAL AND PARAFISCAL MEASURES (1)

(in € million, differences compared to the previous year)

2014

2015

2016 e

Total

of which : Taxshift

Total fiscal measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 -434 984 356

Structural fiscal measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 −21 989 356

Federal government and social security . . . . . . . . . . . . . . . . . . . 298 −103 551 356

Personal income tax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −56 −492 −1 643 −1 680

Corporation tax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 82 226 339

Levies on other incomes and on assets . . . . . . . . . . . . . . . . . . 131 50 493 460

Taxes on goods and services 123 257 1 475 1 237

Communities and Regions and local authorities . . . . . . . . . . 20 82 438 n.

Non-recurrent measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −109 −413 −5 n.

of which :

Tax regularisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 −837 −276 n.

Tax agreements and court decisions . . . . . . . . . . . . . . . . . . . . −159 174 −25 n.

Energy sector contribution for unutilised sites . . . . . . . . . . . . 0 0 120 n.

Liquidation levy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 −367 −124 n.

Registration fees : anticipatory effect of mortgage interest relief reform . . . . . . 150 −150 0 n.

Excess profit rulings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 0 500 n.

Measures on social contributions (2) . . . . . . . . . . . . . . . . . . . . . . . . . −285 −174 −2 011 −1 872

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −76 −608 −1 027 −1 516

p.m. In % of GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0 −0.1 −0.2 −0.4

Sources : Budget documents, NBB.(1) This generally concerns the presumed influence of measures shown in budget documents. The final impact may be different.(2) Of which the reduction in social contributions amounting to € 764 million was funded by lowering wage subsidies through incomplete transfer of withholding tax.

Table 22 GENERAL GOVERNMENT PRIMARY EXPENDITURE

(deflated by the GDP deflator, percentage changes compared to the previous year, unless otherwise stated)

2012

2013

2014

2015

2016 e

Average 2000‑2015

Level recorded (1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.2 52.4 51.8 50.9 51.0 47.4

1. Real recorded growth . . . . . . . . . . . . . . . . . . . . . 2.9 0.3 0.5 −0.4 1.6 2.6

2. Influence of non‑recurrent or fiscally neutral factors (2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 −1.4 0.3 −0.3 0.3 0.0

3. Influence of cyclical factors (2) . . . . . . . . . . . . . . 0.1 0.1 −0.1 −0.1 −0.1 0.0

4. Indexation effect (2), (3) . . . . . . . . . . . . . . . . . . . . . 0.3 0.6 −0.4 0.0 0.3 0.0

5. Adjusted real growth (1 − 2 − 3 − 4) . . . . . . . 1.4 1.0 0.6 0.0 1.0 2.6

Sources : DGS, NAI, NBB.(1) In % of GDP.(2) Contribution to real recorded growth of primary expenditure.(3) Effect caused by the difference between the actual indexation (or the theoretical indexation for 2015 and 2016 in view of the index jump) of public sector wages and

social security benefits and the rise in the GDP deflator. The other effects due to differences between inflation measured by the GDP deflator and the movement in price factors influencing other expenditure categories – whether these are attributable to the indexation mechanisms or to divergent patterns in the prices of certain expenditure categories – are not adjusted, owing to the absence of sufficient information.


factorsin 2016,includinganexceptionaleffort,totallingaround0.2%ofGDP, insecuritymeasuresandcombat‑ing terrorism, and in accommodating asylum-seekers.A number of non-recurrent factors also pushed up ex‑penditure,includingcourtordersfortaxrefundstoEthiasandEliaandtherecapitalisationbyDexiaofitssubsidiaryDexiaCréditLocal.Thetimelagbetweentheindexationofsocialsecuritybenefitsandpublicsectorwagesontheonehandandinflationontheotherwerealsobehindthegrowth inprimary expenditure.Conversely, therewas aslightdownwardcyclicaleffectonprimaryexpenditureintheformofareductioninunemploymentbenefits.

Adjusted for the influenceof temporary factorsand thecyclical effect, aswell as the time lagbetween inflationandindexation,realprimaryexpenditurewentupby1%in 2016.Thiswasinlinewiththemodestrateofincreaseinrecentyears,andwaslowerthanthegrowthinGDPinvolume terms.

Public sector pay accounted for a quarter of primaryexpenditurein 2016,or12.4%ofGDP.Thisreflectstherelatively high employment level in the Belgian publicsector: almost 850,000 people are employed in thegeneralgovernmentsector,45%ofwhomworkfortheCommunities and Regions, 37% in local government

andjust18%inthefederalgovernmentandsocialse‑curity. This distribution has undergone radical changesin the past due to transfers of powers between gov‑ernment sub-sectors, particularly at the time of the1988 Statereform,andtoalesserextentfollowingthepolice reforms in 2001. The other state reforms involved fewer transfers of civil servants. Drawing all strandstogether,thenumberofpeopleemployedinthepublicsectorincreasedsharplybetween2000 and2010,anditisonlysince2011 thatthetotalnumberofcivilservantshasremainedrelativelystable.

Besides employment, other determinants of the publicsectorwagebillalsobarelymovedin 2016,forexampleconventionalpayrisesandthewagedrift,bothofwhichwere virtually zero. By contrast, wage indexation wasintroduced midway through the year, after the thresh‑old index figure was passed in May. Due to the 2015index jump,theprevious indexationofpublicsectorpaydatedfromearly2013.Overall,publicsectorwagesfellby0.1 percentagepointofGDPcomparedwith 2015.

Other current expenditure categories rose in 2016.Purchasesofgoodsandservices, likewages, turnedouthigherduetoadditionalspendingonsecurityandmeas‑urestocombatterrorism.Theseexceptionalexpenditureitems cancelled out the effect of the earlier austerity

CHART 63 GENERALGOVERNMENTPRIMARYEXPENDITUREANDGDP

(percentagechangescomparedwithpreviousyear)

2007

2008

2009

2010

2011

2012

2013

2014

2015

–3

–2

–1

0

1

2

3

4

5

–3

–2

–1

0

1

2

3

4

5

Adjusted primary expenditure (1)

2016

e

GDP by volume (2)

Sources:NAI,NBB.(1) PrimaryexpendituredeflatedbytheGDPdeflatorandadjustedforcyclicaland

non-recurrentorbudget-neutralfactors,andfortheindexationeffect.Thelatteriscausedbythedifferencebetweentheactualindexation(orthetheoreticalindexationfor 2015and 2016inviewoftheagreedindexjump)ofpublicsectorwagesandsocialsecuritybenefitsandtheriseintheGDPdeflator.

(2) Calendaradjusteddata.

CHART 64 PUBLICSECTOREMPLOYMENTBYSUB-SECTOR

(Annualaverages,indexfigure1995 =100)

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

85

90

95

100

105

110

115

120

125

130

135

140

2016

e

Total

Federal government and social security

Communities and Regions

Local government

Sources:NAI,NBB.


measures takenby the federalgovernment,mainlycon‑sisting of a structural reduction in operating budgets. Business subsidies also rose substantially comparedwiththepreviousyear,amongother thingsdue topayments

by the Energy Fund to transmission systemoperators intheFlemishRegion.Currenttransfersalsowentup,partlythanks to the temporary increase in Belgium’s contribu‑tion to the Eu budget.

Box6–Makinggovernmentmoreefficient

Efficiencycanbedescribedastheabilitytoachievethebestpossibleresultswiththefewestpossibleresources.Thepublicsectorneedstostriveforefficiency.Todothis,governmentcouldimproveitsservicestothecommunity,cappublicspendingorseektoachievebothgoalssimultaneously.

Publicsectorefficiencyisdifficulttoanalyse,andmeasuringoutputisnoteasyduetothefrequentlackofdataonitsquality.Sincepublicservicesare,bydefinition,notmarketservices,thereisalsonopricingmechanismtoenable their value tobequantifiedobjectively.Nonetheless, relevant indicators canbedesigned for anumberoffunctions,suchashealthcare,education,securityandmobility.Theseindividual indicatorscanbecombinedto forma composite index for each function. Examples of indicators that canbeused for health care are lifeexpectancy,childmortalityandtheaccessibilityofmedicalcare.ExamplesinthefieldofeducationincludeschoolstudentscoresintheOECD’sPISAtests,aswellastheshareofthepopulationwithasecondaryorhighereducationqualification.Whenitcomestosecurity,theindexisbasedonlyonsurveyresults,forexampleontheperceivedimpartialityofthelegalsystemorthepercentageofrespondentswhofeeltheyhavebeenvictimsofaggressionor theft.Theoutputof themobility function isapproximatedusingacomposite indexpublishedby theWorldEconomicForum(WEF)fortransportinfrastructure(roads,rail,water,air).Here,anattemptismadetoassesstherelativeefficiencyofBelgiumonthesefourfunctionscomparedwithotherWesternEuropeancountries.

Unlikeoutput,inputcanbemeasuredrathersimplyonthebasisofexpenditure.ThecomparisonisbasedontheCOFOGclassification(ClassificationoftheFunctionsofGovernment)drawnfromthenationalaccounts,inwhichpublicexpenditureisbrokendownacrossthedifferentfunctions.Insomeareas,however,householdsalsoaccountforanotinsubstantialportionoftheexpenditure.Sinceitisnotpossibletoseparateouttherespectiveshareofgovernmentandprivatehouseholdsinthefinalresult,toavoiddistortingtheanalysis,theoutputobtainedthenneedstobecomparedagainstthetotalexpenditure.Todothis,informationonhouseholdexpenditurefromtheCOICOPclassification(ClassificationofIndividualConsumptionbyPurpose)isaddedtototalpublicexpenditure.Averageexpenditureoverthelast15 yearsisalsotakenintoaccount,sincecurrentoutcomesareoftentheresultofexpenditurespreadoverarelativelylongperiod.Eachrecentchangeintheexpendituretrend–forexampleduetoEurope’sgreatrecessionandsovereigndebtcrisis,aswellasthemeasurestakenbygovernmentinthewakeoftheseevents–isthereforeweakenedsomewhatinthemeasuredinput.

Belgiumscores fairlywellonhealth care.That ispartlydue to thequalityof thehealth infrastructureand thegeneral satisfaction of recipients of health care, but also to the short waiting times when arrangingmedicalappointments.However,thisgoodresultdoescomeatveryhighcosts–onlyonecountryspendsmoreonhealthcare than Belgium.

Whenitcomestoeducation,theanalysisshowsGermanyandFinlandtobethemostefficientofallthecountriesstudied.ExpenditureinthecountriesofSouthernEuropevarieswidely,butoverallthisregionachievestheweakestresults. Belgiumoccupies an intermediateposition, aligning fairly closelywith the relatively efficient countries:its expenditure and results are below those of Finlandbut better thanGermany’s.An above-average share oftheBelgianpopulationhasahighereducationqualification,butBelgium’s score is average for thepercentageofpeoplewithasecondaryeducationqualification.ThereiswidedisparityintheresultsinthedifferentBelgiancommunities;inthePISAsurvey,whichteststheknowledgeandskillsofschoolstudents,studentsintheFlemishCommunityachieveahigheraveragescorethantheirpeersintheFrenchCommunity.

4


Belgium also occupies amiddle-of-the-road position on the security front,with seven countries provingmoreefficient,i.e.achievingbetterresultswithlowerexpenditure.DenmarkandFinlandareamongthemostefficientcountries for public order and security, closely followed by Austria and Luxembourg. France and the UnitedKingdomarecomparabletoBelgiumintermsofefficiency,thougheachtacklesthisissueinverydifferentways,withFrancespendinglessandachievingworseresults,andtheUKspendingmoreandachievingbetterresults.ThecountriesofSouthernEurope,finally,bringuptherear.

EFFICIENCYOFTHEPUBLICSECTOR

5 6 7 8 5 6 7 89 10 11 4

DK

DE

IE

EL

ESFR

IT

LU

NL AT

PT

FI

SE

UK

BE

DK

DE

IE

EL

ES

FR

IT

LU

NL

AT PT

FI

SEUK

BE

DK DEIE

EL

ES

FR

IT

LU NL

AT

PT

FI

SEUK

BE

DK

DE

IE

EL

ES

FR

ITLU

NL

AT

PT

FI

SE

UK

BE

–1.0

–0.5

0.0

0.5

1.0

Out

put

(com

posi

te in

dex)

(1)

Expenditure by the public and private sector in % of GDP, average 2010‑14

HEALTH CARE

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

Out

put

(com

posi

te in

dex)

(1)

Out

put

(com

posi

te in

dex)

(1)

Expenditure by the public and private sector in % of GDP, average 2010‑14

EDUCATION

–2.0

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

0.5 1.5 2.5

Expenditure by the public sector in % of GDP, average 2010‑14

SECURITY

4.0

4.5

5.0

5.5

6.0

6.5

1.5 2.5 3.5

Expenditure by the public sector in % of GDP, average 2010‑2014

MOBILITY

French Community (2)

Flemish Community (2)

Out

put

(com

posi

te in

dex)

(1)

1.0 2.0 2.0 3.0

Sources:EC,WEF,NBB.(1) Withtheexceptionofthe“mobility”function,forwhichtheoutputisderivedfromaWEFsurvey,thecompositeindicesdevelopedtomeasureoutputarebased

onthosedescribedinEugène,B.(2008),“Theefficiencyfrontierasamethodforgaugingtheperformanceofpublicexpenditure:ABelgiancasestudy”,NBBWorkingPaper138,September.

(2) NotalltheindicatorsusedfortheinternationalcomparisonwereavailablefortheFlemishandFrenchCommunities.Intermsofoutput,theirresultswerethereforeapproximatedonthebasisoftheresultsofthe 2015PISAsurvey,whiletheinputmeasurementwasbasedonanestimateofthespendingperstudentin2012.

4


Public spending onmobility is fairly inefficient in Belgium,with the considerable resources invested deliveringonlymodest results.AustriaandLuxembourgareeven lessefficient thanBelgium,whileGermany,France, theNetherlands,Spain,theUnitedKingdomandFinlandallperformbetterintermsofbothexpenditureandquality.The other countries spend less than Belgium but also achieve less good results.

Belgium isnotamongthemostefficientcountries foranyof the four functionsmeasured,butalsonever fallsintothe leastefficientgroup. Inotherwords,Belgiumachievesanaveragescoreforefficiency.ThisconclusionlargelymatchestheresultsofaWEFsurveyin 2015,inwhichBelgiumachievedacomparablescoretoFranceforefficiency,aheadoftheSouthernEuropeancountriesbutlaggingbehindtherestoftheEU15.

Thereisnodoubt,therefore,thattheBelgianpublicsectorcouldmakemajorefficiencygainsineachofthefourfunctionsstudied.Theprecisenatureofthepotentialefficiencygainsdiffersdependingonthefunction.

Thereisaneedtocontrolcostsinhealthcare,anareainwhichBelgiumachievesgoodresults.Morespecifically,the challenge lies in curbing the upward trend in spending without adversely impacting the quality of care.Unfortunately,theanalysisoftheailmenttellsusnothingaboutthecure.Atbest,itcanbeobservedthat,relativelyspeaking,spendingonhospitalsishighestinBelgium.

Itshouldbepossibletoimprovepublicsectorperformanceonfunctionssuchaseducationandmobilitywithoutincreasingtheirbudgets.Thequestionofwhichmightbethebestmethodsofachievingthisfallsoutsidethescopeofthisanalysis,thoughanumberofpossibleroutescanbeoutlinedbasedontheexpenditureprofilesofthemostefficientcountries.Seenfromthisperspective,thedataappeartosuggestthat,withregardtomobility,Belgiummightconsiderboostinginvestmentwhileatthesametimereducingoperatingcostsandassociatedsubsidies.

Turningtopublicorderandsecurity,Belgium ispossiblyatadisadvantageduetofactorsoverwhich ithasnocontrol,suchasitscentrallocation,whichfacilitatesallkindsoftrafficking.Againstthisbackdrop,itisdifficulttocomparetheresultswiththoseofwidelydifferingcountries.Therecanbenodoubt,however,thatthisisanotherareawherethereisscopeforefficiencygains.

GENERALEFFICIENCYOFTHEPUBLICSECTOR

(Assessedbyrespondentsonascalefrom1 to7;surveyconductedin 2015)

0

1

2

3

4

5

6

FI LU SE NL UK DE IE AT DK BE FR PT ES EL IT

Source:WEF(GlobalCompetitivenessReport).

4


ImprovingtheefficiencyofthepublicsectorshouldinanyeventbeacoreobjectiveforalllayersoftheBelgiangovernmentover the coming years. First off, a structural approachwill beneeded to analyse all public sectortasks and enable choices to bemade. The next stepwill be to investigatewhich policy level would bemostsuitedforcarryingoutthesetasks,withaviewtoeliminatingoverlapsandseekingbenefitsofscale.Lastly,thevarious government departments need to be organised as effectively as possible, partly by making optimumuseofIT applicationsandbysimplifyingprocedures.Anynecessaryreformsshouldbeviewedfromalong-termperspectiveandrigorouslyimplemented.

At2.4%ofGDP, investment in 2016stillaccountedforonly a small part of public spending. This is below theaverageacrosstheEU,andmoreoverbarelycompensatesfor the depreciation of earlier public investment. As a consequence, net investment – which reflects develop‑ments in thecapital stock– isstuckatvirtuallyzero.Asgrossfixedcapital formationcan substantiallyboost theeconomy’s growth potential, initiatives to raise itwouldbedecidedlywelcome.

Social benefits, accounting for almost half of primaryexpenditure, remained unchanged as a proportion ofGDP.However,thismasksdivergingtrendsinthevariousbranchesofsocialsecurity.Therise inspendingonsick‑nessandincapacitybenefitsremainsaconcern,thoughitmoderated in 2016comparedwith theprecedingyears.Conversely, unemployment benefits fell for the third

successiveyearthankstopositivelabourmarketdynamicsaswellasarangeofausteritymeasuresaimedatrestrict‑ingentitlementtobenefits.

Rising pension expenditure due to population ageing

Populationageinghasbeen reflected inpublic expendi‑tureforseveralyearsnow,andthistrendissettocontinueinthecomingdecades.TheStudyCommitteeonAgeing,forexample,forecastinits 2016reportthattotalspend‑ingonpensionsandhealthcarewillrisebyfourpercent‑agepointsofGDPinthe 2016-60 period.Thatcompareswith a reduction of 1.8 percentage points of GDP inaggregate spending on other social benefits, especiallyexpenditureonunemploymentbenefitsandfamilyallow‑ances. The rise in spending on pensions and health care therefore needs to be kept under control to guarantee the sustainabilityofpublicfinancesandassuretheviabilityofthesocialprotectionsystemforthelongterm.

Thenumberofpensionershasrisensharplysince2011,bybetween1.5%andjustover2%peryear,comparedwithaverylimitedincreaseinthefirstdecadeofthenewmil‑lennium. Public spending on pensions has consequentlyrisenfrom9.3%ofGDP in2010 to10.3%in 2016. Innominalterms,thisexpenditureroseby3.7%inthelatteryear.Populationageingnaturallymakesitmoredifficulttowipeoutthebudgetdeficit.

Against the backdrop of population ageing, when thefederal government took office inOctober 2014, it an‑nounced a raft of measures aimed at reforming the pensionsystemfurther.Thesereforms,which inessencecomedowntokeepingemployeesinworkforlonger,areintendedtoreducespendingonpensions,boostemploy‑mentandsupporteconomicactivity.Thegovernmenthadalreadytakenvariousimportantmeasuresin 2015,raisingthestatutoryretirementagewhileimposingfurthercon‑straintsonearlyretirement,limitingthenumberofpeopleentitled to survivor pensions and abolishing the pension

CHART 65 PUBLICINVESTMENTBARELYOFFSETSDEPRECIATION

(in%ofGDP)

1971

1976

1981

1986

1991

1996

2001

2006

2011

–3

–2

–1

0

1

2

3

4

5

6

2016

e

Net fixed capital formation

Fixed capital consumption (–)


Sources:NAI,NBB.


job can be taken into account in calculating pensionentitlements, whether a system of part-time pensionscouldbe createdandhowapoints-basedpension sys‑tem could be introduced.

Health care spending must be kept under control

Publicspendingonhealthcarewillbeinfluencedinthecom‑ingdecadesbynewneedsonthepartofthegrowingolderpopulation.Demandforhealthcarewillincrease,partlydueto the need for special facilities for the diagnosis and treat‑mentofcertainage-specificdisordersanddiseases.However,itisimportanttonotethatpopulationageingisjustoneofthefactorsbehindgrowthinhealthcareexpenditure.Otherfactorsrelatetorisingaffluenceandtheconcomitantgrow‑ingdemand for health care, aswell as the rising costs oftreatments,partlyduetorapidtechnologicalprogress.

CHART 66 HEALTHCARE,SICKNESS/DISABILITY,UNEMPLOYMENTANDLIVINGWAGE

(datadeflatedbytheGDPdeflator;percentagechangescomparedwiththepreviousyear,unlessotherwisestated)

2010 2011 2012 2013 2014 20150

1

2

3

4

5

6

7

8

2010 2011 2012 2013 2014 2015–12

–10

–8

–6

–4

–2

0

2

4

6

8

2010 2011 2012 2013 2014 20150

1

2

3

4

5

6

7

8

2010 2011 2012 2013 2014 2015–2

0

2

4

6

8

10

12

Average GDP growth by volume 2000‑15

Real growth

Average growth 2000‑15

HEALTH CARE

UNEMPLOYMENT (1)

SICKNESS/DISABILITY (1)

LIVING WAGE (1)

% of GDP (right‑hand scale)

(left‑hand scale)

2016 e

2016 e

2016 e6.4

6.6

6.8

7.0

7.2

1.2

1.4

1.6

1.8

2.0

1.2

1.4

1.6

1.8

2.0

2016 e0.0

0.2

0.4

0.6

0.8

Sources:Budgetdocuments,NAI,NBB.(1) Expenditureadjustedfortheindexationeffect,causedbythedifferencebetweentheactualindexation(ortheoreticalindexationfor 2015and 2016inviewoftheagreed

indexjump)ofpublicsectorwagesandsocialsecuritybenefitsandtheriseintheGDPdeflator.

bonus.Italsoapprovedanumberofchangestocivilser‑vice pensions and introduced stricter conditions governing unemploymentbenefitwithemployertop-up.

The government introduced a further series of pension reforms in 2016, includingmeasures to harmonise the‘diplomabonus’ system for pension calculations in thethree main pension systems, abolishing the preferen‑tial tantièmes and schemes in the civil service pension scheme, introduction of amixed pension in the publicsector (contractagent/civil servant),and reforming thedisabilitypensionsforcivilservants.However,thecondi‑tionsunderwhichtheseproposalswouldapplyhavenotyetbeenworkedoutandarestillbeingdiscussedbythevarious stakeholders. In addition, the National PensionCommitteewaschargedwithpavingthewayforarangeof reformsby lookingathowthestrenuousnessof the


modest comparedwith the rapid growth still recorded afew years ago. Spendingwas kept under control thankstonewausteritymeasures,suchasthenon-indexationofmedicalfeesin 2016,andtheintroductionofthe‘patentcliff’mechanism,bywhichreimbursementfordrugsdropsoffsharplyassoonasapatentexpires.

Onprojectionsby theSCA,averagespendingonhealthcareinabroadsenseissettogrowby2.1%perannumin real terms in the 2017-60 period,while thepotentialeconomic growth over the same period should average1.6%.Asaresult,theSCAprojectsanincreaseinhealthspending of 1.9 percentage points of GDP by the endofthisperiod. Ifhealthcarespendingweretorisemorequicklythananticipated,ageingcostswouldgoupsharp‑ly.Forexample,expendituregrowthof2.6%or3.1%perannumwouldboosttheimpactto4.2 and7 percentagepointsofGDP,respectively.Itisthusessentialtomaintainstrictcontrolovertheriseinhealthcareexpenditure.

Further decline in interest charges thanks to very low or even negative interest rates

Interest charges fell further in 2016, by 0.3 percent‑agepointsofGDP.As thedebt ratio remainedvirtuallyunchanged,thisdecline–asin 2015–wasdueentirelyto the reduction in the implicit interest rate. This rate,which is the computed interest chargeas aproportion

CHART 67 SHARPINCREASEINPENSIONERNUMBERSOVERSEVERALYEARS

(percentagechangesonanannualbasis,unlessotherwise)stated

2000

2002

2004

2006

2008

2010

2012

2014

6

7

8

9

10

11

12

2016

e

–0.5

0.0

0.5

1.0

1.5

2.0

2.5

Number of pensioners (left‑hand scale)

Expenditure on pensions in % of GDP (right‑hand scale)

Sources:NationalPensionsOffice,NAI,NBB.

CHART 68 PUBLICEXPENDITUREONHEALTHCAREINABROADSENSE(1)

2016

2020

2025

2030

2035

2040

2045

2050

2055

2060

7

8

9

10

11

12

13

14

15

16

7

8

9

10

11

12

13

14

15

16

2017

2020

2025

2030

2035

2040

2045

2050

2055

2060

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

+7.0%

+4.2%

+1.9%

REAL GROWTH IN EXPENDITURE ON HEALTH CARE, ON SCA PROJECTION (in %)

average 2017‑60 : 2.1%

EXPENDITURE ON HEALTH CARE,ALTERNATIVE SCENARIOS (in % of GDP)

0.5 % additional real growth

SCA projection

1 % additional real growth

Sources:SCA,NBB.(1) DefinitionusedforpublicexpenditureonhealthcarebytheSCAinits 2016annualreport.Aswellasspendingonmedicalcareasshowninthenationalaccounts,this

definitionincludesotherspendingitems,includingmostothersocialbenefitsinkindandtheFlemishhealthcareinsurancefund(Zorgfonds).

Publicspendingonhealthcareinanarrowsense(accordingtothedefinitionusedinthenationalaccounts)amountedto7%ofGDPin 2016.Theincreaseinthisspendingwas


coupon. Yields on ten-year reference bonds hoveredaroundthe1%markthroughout 2016,slumpingtoanaveragelowof0.15%inAugustbeforerisingagaininthefourthquarter.

5.4 Debtratiotrendyettoreachturning point

Belgium’spublicdebthadrisento106.6%ofGDPbytheend of 2016, indicating that Belgiumhas not yetman‑agedtoreversetheupwardtrendinthedebtratiowhichbeganfollowingtheonsetofthefinancialcrisisin2008.Intherestoftheeuroarea,bycontrast,thedebtratiohasbeenfallingsince 2015.

Thisisdespitethefavourableinterest-rate-growthdynam‑icsin 2016,when,forthefirsttimesince2011,nominalGDP growth exceeded the implicit interest rate on thepublicdebt,automatically reducing thepublicdebtasapercentageofGDP–thoughthiseffectwaspartiallycan‑celledoutbyasmallprimarybudgetdeficit.

The increase in the debt ratio can be ascribed entirelytoexogenousfactors,whichimpactondebtbutnotontheoverallbalance.Forexample,governmentdebtwaspushedupbyanincreaseinloansgrantedundersocialhousingpolicy,whiletaxesassessedbutnotyetcollectedand state pension liabilities related to pension funds taken over from corporations in the past also had an up‑wardeffect.Thedebtmanagementprocessalsopushedup debt due to costs of interest rate swaps and otherderivativescontractsagreed intheautumnof 2014forthe issue of government bonds at then applicable rates in 2015 and 2016. Interest payments on a cash basisalsoexceededthoseonatransactionbasis,whichserveas the reference value for interest charges in the gen‑eralgovernmentaccounts.Thereasonwastheissuanceof bonds in earlier years with nominal coupons abovemarket interest rates; this has a downward effect ondebtintheyearofissueuptotheamountoftheissuepremiums,becausetheissuepriceofsecuritiesishigherthan their nominal values. However, this effect gradu‑ally peters out as higher coupons are paid. Substantialissuepremiumswereagainrealised in 2016,mitigatingthe upward effect of other debt management-relatedfactors.

Bringingaboutastructuralreductioninthedebtratiowillrequire substantial and sustainedprimary surpluses. TheECcalculatedthat,tomeetthedebtcriterion–i.e.low‑eringthedebtratioby1/20th ofthedifferencebetweenthepresentdebtratioandthe60%-of-GDPthreshold –Belgium would have needed to increase the structural

CHART 69 DECLINEININTERESTCHARGESATTRIBUTABLETOFURTHERREDUCTIONINIMPLICITINTERESTRATE

(in%,unlessotherwisestated)

2000

2002

2004

2006

2008

2010

2012

2014

2000

2002

2004

2006

2008

2010

2012

2014

–1

0

1

2

3

4

5

6

7

8

2016

e

Change in interest charges

2016

e

CHANGE IN INTEREST CHARGES(in % of GDP)

IMPLICIT INTEREST RATE (1) ON TREASURY CERTIFICATES

AND ON LONG‑TERM TREASURY DEBT

Short‑term debt in €

Long‑term debt in €

p.m. Market interest rate (ten-year OLO)

Effect of the implicit interest rate (1)

Of which :

Effect of the debt ratio

–0.8

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

Sources:FPSFinance,NAI,NBB.(1) Interestchargesinthecurrentyearasaproportionofthedebtattheendofthe

precedingyear.

of the outstanding public sector debt at the end of the previousyear,willcontinuetotrenddownwardsaslongasmarketinterestratesonnewissuesremainlowerthanyieldsonmature securitiesandgovernment loans.Thiswasonceagainthecasein 2016.Intheshortterm,thenegativeyields thatpersisted throughout theyearena‑bled the government to generate some funds through the issuance of Treasury bills with maturities of lessthanoneyear. For themediumterm, too, theTreasurywasabletoissuefive-yearlinearbondswithanegative


balanceby2.6%ofGDPin 2016.Sofromthisperspec‑tivetoo,achievingastructuralbalanceisessential.

Further increase in average maturity of government debt

The federal government accounts for by far the larg‑est share of Belgium’s total general government debt:around85%. The gross federal government balance tobefinancedtotalled€43.5 billionin 2016,slightlyuponthepreviousyear.Althoughmediumandlong-termdebtmaturingin 2016waslower,thebudgetdeficitwidenedsubstantiallyincashterms.

Toagreaterextentthanintheprevioustwoyears,financ‑ing requirements were met in 2016 primarily throughthe issue of linear bonds (OLOs). Nearly 90% of OLOsissuedin 2016haveamaturityofatleasttenyears,whilesome have 50-yearmaturities, enabling the Treasury tofix the low yield rate on government bonds for a longperiod. The volume of Treasury certificates – short-termdebtinstruments–roseby€1.2 billion,butwasstilllowcompared with the volumes issued during the financialcrisis.Allinall,theaveragematurityofthedebtportfolioincreasedfortheseventhyearinsuccession,to8.7 years.Thatcompareswithlessthansixyearsin2009.

CHART 70 DEBTRATIOUPAGAININ2016

2000

2002

2004

2006

2008

2010

2012

2014

60

70

80

90

100

110

2008 2009 2010 2011 2012 2013 2014 2015–4

–2

0

2

4

6

8

10

Euro area

Belgium Change in debt

2016 e

CONSOLIDATED GROSS DEBT OF GENERAL GOVERNMENT(in % of GDP)

CHANGE IN CONSOLIDATED GROSS DEBT OF GENERAL GOVERNMENT(in percentage points of GDP)

Primary balance

Interest‑rate‑growth dynamics (1)

Exogenous change

Of which :

2016

e

Sources:EC,NAI,NBB.(1) ThedifferencebetweentheimplicitinterestondebtandnominalGDPgrowth,multipliedbytheratioofdebtattheendoftheprecedingyeartotheGDPovertheperiod

under consideration.


CHART 71 FURTHERINCREASEINAVERAGEMATURITYOFFEDERALGOVERNMENTDEBT

(averagematurityoffederalgovernmentdebt,inyears)

2000

2002

2004

2006

2008

2010

2012

2014

0

1

2

3

4

5

6

7

8

9

2016

source : fPs finance.

Table 23 FINANCING REQUIREMENTS AND RESOURCES OF THE FEDERAL GOVERNMENT

(in € billion)

2014

2015

2016

Gross balance to be financed . . . 38.2 41.0 43.5

Gross financing requirements 32.9 37.0 38.5

Budget deficit (+) or surplus (−) (1) . . . . . . . . . . . . . 10.5 9.0 12.6

Medium- and long-term debt maturing during the year . . . . . . . . . . . . . . . . . 22.4 28.0 25.9

In euro . . . . . . . . . . . . . . . . 22.4 28.0 25.9

In foreign currencies . . . . 0.0 0.0 0.0

Buy-backs (securities maturing the next year or beyond) . . . . . . . . . . . . 5.3 4.0 5.0

Other financing requirements . . 0.0 0.3 0.0

Funding resources . . . . . . . . . . . . 35.7 40.7 42.3

Linear bonds (OLOs) . . . . . . . 31.8 35.6 37.6

State notes and others . . . . 3.8 5.1 4.7

Net change in the short-term debt in foreign currencies . . . . 0.0 0.0 0.0

Change in the outstanding amount of Treasury Certificates 1.8 −1.1 1.2

Net change in other short-term debts in € and in financial assets . . . . . . . . . . . . . . . . . . . . . 1.4 1.4 0.1

Source : FPS Finance.(1) The overall balance is calculated on a cash basis and takes account of financial

transactions which are not included in the overall balance of general government which, in accordance with ESA 2010, is calculated on a transaction basis.

Adynamicandinclusiveeconomy

153Economicandfinancialdevelopments ❙ ADYNAMICANDINCLUSIVEECONOMY ❙

6. Adynamicandinclusiveeconomy

6.1 Belgium:achangingeconomy

Globalisation and new technologies are fundamentally altering the structure of the advanced economies...

Theglobaleconomyhasundergonefar-reachingchangesin recent decades. As well as reducing transport costsanddismantlingbarrierstodataexchange,technologicalprogress has transformed modes of distribution and value creation,andevencustomerrelationships.Thishasbeenaccompaniedbyfundamentalrestructuringofproduction

systems,both locally, through the increaseduseofout‑sourcing,andworldwide,viathereorganisationofglobalvalue chains.

Thesechangesclearlypresenteconomicgrowthopportu‑nitiesfortheadvancedeconomiesbycreatingnewneedsand through increased demand from the emerging na‑tions.Atthesametime,however,theyaredrivingafun‑damental reorganisation of economic structures : entire segmentsof activityhavebeen shiftedworldwide–notonly within industrial production chains, but to a largeextentintheservicessectortoo.

CHART 72 SHARPFALLINEMPLOYMENTSINCE2000 ININDUSTRY,AGRICULTUREANDFINANCIALSERVICESANDMARKEDINCREASEINNETEMPLOYMENTCREATIONINBUSINESSSERVICES,HEALTHCAREANDGOVERNMENT

(changesinthousandsofpeoplebetween2000 and 2015)

–150 –100 –50 500 100 150 200 250 300 350

Business services

Health and social work

Public administration

Construction

Information and communication

Real estate activities

Trade, transport, accommodation and food service activities

Other services

Financial and insurance activities

Agriculture

Industry

Source:NAI.


… resulting in a substantial reallocation of employment between sectors and businesses

In Belgium, these changes have translated over recentdecades into considerable job destruction in a range ofsectors, includingindustry,agriculture,andfinancialandinsurance activities. On balance, just over 140000 jobswere lost between 2000 and 2015 in industry and22000 inagriculture. In industry,thiscontinuedatrendthat began in the mid-1970s, while the process hadbeen under way for much longer in agriculture. Some19000 jobswerelostinfinancialandinsuranceactivitiesin the same period.

These changes are brought home to the population bythe corporate restructuring that accompanies them. There werenumerousexamplesin 2016oftheimpactsuchre‑structuringcanhaveonemployment.Throughout 2016,125 businessesembarkedon collective redundancypro‑cedures, affecting over 12000 employees; the loss of5500 jobs in this way was announced in Septemberalone,morethaninthewholeof 2015.

Reallocation of resources promotes economic growth

These developments took place, however, against thebackdropofanemploymentmarket inwhichhalfamil‑lionnetjobswereaddedbetween2000 and 2015.Afur‑ther59000 newjobswerecreatedin 2016.Thispatternisinkeepingwithaneffectivelyfunctioningemploymentmarket, inwhich jobs aredestroyedand createdall thetime. According to DynaM, statistics drawn up on thebasisofNSSOdata,justover205000 jobswerecreatedin 2014alone,with180000 beinglost.

Viewed more generally, the reallocation of resourceswithin an economy is a reflection of economic dyna‑mism: the transfer of employment and capital fromwaningsectorsandbusinessestoonesthataregrowingdeliversproductivitygains,theonlysourceofsustainablegrowth.

Other thanduringtheeconomicandfinancialcrisis, thereallocation of resources from the least efficient to themostefficientbusinesses–especiallythoseclosesttothefrontiersoftechnologicalefficiency–hasbeenshowntoraiseaggregateproductivityinBelgium.Butthissourceofgrowthhasremainedvery limitedsince2010.The influ‑ence of a relatively small number of businesses on thepatternofaggregateproductivityisdecisive:theprimarysource of growth comprises productivity gains recordedbybusinesses close to the frontierofefficiency,definedasthe5%mostproductivebusinesses.Bycontrast,otherfirms that to some degree lag behind technologically

contributenegativelytoproductivitygains,whichmeanstheirhandicapisgenerallygettingworse.

Theinternalcomponentofproductivitygrowthisexcep‑tionally important in manufacturing industry, telecom‑munications and IT services sectors. The reallocation ofresources to the best businesses makes the biggest con‑tribution to strengthening economic results in trade and other market services.

Public increasingly concerned

changes to the economic fabric and the organisation ofproductionhavean impacton theentirepopulation,althoughnoteveryoneisaffectedinthesameway.

Globalisationandtechnologicaldevelopments,forinstance,arebringingaboutmajorchangesinthedemandforem‑ployment in Belgium, as in other advanced economies.Demand for highly-skilled employees has grown at theexpense of that formedium-skilled positions. Low-skilledoccupationshave,inturn,beenlessseverelyaffected:thekindofworkperformedinthiscategoryisdifficulttotrans‑fer,sinceitfrequentlydisplaysastronglyrelationalcontentand is associatedwith geographical proximity. There is agenderdifferencehere:thepercentageoflow-skilledjobshas increased amongwomen,while it has fallen amongmen. Employment demand has become more stronglypolarisedasaresult.Apartialshifttoflexibleemployment

CHART 73 GROWTHINTOTALFACTORPRODUCTIVITYANDREALLOCATIONOFRESOURCES

(averagespersub-period,inpercentagepoints)

1997-2000 2001-2005 2006-2010 2011-2013–2

–1

0

1

2

3

4

5

Internal growth

Reallocation

Frontier businesses Other businesses

p.m.: Growth in total factor productivity

source : NBB.Basedonindividualdataforbusinesseswithatleastoneemployee,registeredwiththeCentralBalanceSheetOfficebetween1996 and2013.Afrontierbusinessisamongthe5%mostproductiveenterprisesinitssector.


contracts(fixed-termandinterim)attheexpenseofperma‑nent contracts has also been observed.

Sinceallchangeisasourceofuncertainty,thelattermayhave been intensified by the less than rosy economicenvironment and the perception that the benefits of

prosperity arenotbeing sharedequally. This appears tohave led toquestioningof the social contractonwhichoursocietyisbased,especiallybypeoplemostvulnerableto these developments.

Average per capita purchasing power in Belgium roseby0.1%between2005 and 2015–yearsthatsawthegreatrecessionandtheeuroareacrisis,butalsopartofthe preceding phase of rapid expansion and the sub‑sequentperiodofweakgrowth.Thisfigurewas0.5 ofa percentage point less than per capita GDP growth.This difference results from a transfer of income from Belgium to the rest of the world due to deterioratingterms of trade, combined with a contraction in theproportion of net income from investment flowing toprivate individuals.Theshareofwages inGDP,bycon‑trast,returnedtoitsearly-2005 levelin 2015.Ithadrisenduringthecrisis,butcontractedin 2014and 2015duetowagemoderation.Theshareaccountedforbysocialbenefitsrose.

The incomespread inBelgiumnarrowedslightlyduringthisperiod,incontrastwithmanyothercountries,whereit widened. Inequality is measured as the ratio of thetotaldisposableincomeofthe20%ofpeoplewiththehighestincometothatofthe20%withthelowest.Thefigures are adjusted to take into account the composi‑tionofthehouseholdtowhichanindividualbelongs.

TheEuropeanpovertyindicatorusedfortheEurope 2020strategy shows the percentage of people exposed to

CHART 74 TRENDINEMPLOYMENTBYQUALIFICATIONREQUIREMENTINTHE2000-2013 PERIOD

(shareoftotal,changesinpercentagepoints)

BE

–8

–6

–4

–2

0

2

4

6

8

Highly-skilled jobs

Medium-skilled jobs

Low-skilled jobs

EU15

Source:EC(LFS).

CHART 75 AVERAGEPURCHASINGPOWERANDINCOMEDISTRIBUTIONSTABLEINBELGIUMINTHEPASTTENYEARS

IT ES NL UK BE FR DE FI DK SE FI BE NL SE DK FR DE UK IT ES0

1

2

3

4

5

6

7

8

2015

2005

–1.5

–1.0

–0.5

0.0

0.5

1.0

1.5

2.0

2.5

Gross disposable income at constant prices per capita

GDP by volume per capita

REAL DISPOSABLE INCOME AND GDP BY VOLUME(average annual percentage change, 2005 to 2015)

INCOME INEQUALITY(inter-quintile income ratio S80 / S20)

EU27EU

source : Ec.


at least one of the following risks: monetary poverty,material deprivation and low work intensity. One adultin five in Belgium faces poverty or social exclusion. Theproportion in Sweden, France and the Netherlands issignificantly lower. Belgium’s position in the middle ofthe ranking chiefly reflects the less favourable situationofunemployedpeopleandthelatter’sheavierweightingamongthepopulationaged18 andabove.

Brokendownbyagegroup,youngpeopleareatgreatestriskofpovertyorsocialexclusion,andthatpercentageissteadilyrising.Thisgrouphasbeenadverselyaffectedbythefinancialcrisis,whereasforpeopleaged65 andoverthe riskofpovertyor social exclusionhas reduced since2013.

In response to thesevariouschanges, thereare signsofa growing temptation of reverting to protectionism inthe advanced nations. But protectionism does not offer a long-termsolution.Aneconomyrunningatfullpotentialisessentialtotheabilitytoabsorbfundamentalchangesintheeconomicfabric.Thisrequiresavarietyofinstrumentsthat allow rapid and efficient reallocation of productiveresources, and above all more inclusive growth. Theseinstruments must help create the appropriate conditions for rapid entry into the labour market, especially foryoungpeople,butalsoforthedevelopmentofeconomicactivityandemployment.Thiscallsformeasuresaimedatpromoting innovationandbusinesscreation,while limit‑ingfactorsthatstandinthewayofadaptation–allwithaviewtostimulatingtotalfactorproductivity.

6.2 Raisingtheeconomy’sdevelopmentpotentialbyfacilitatingreallocationof resources

Heterogeneous businesses and low entrepreneurial dynamism

As is thecase inmostotheradvancedeconomies,busi‑nessproductivityinBelgiumvariessharply.WhileBelgiancompaniesare,onaverage,amongthemostproductiveinEurope,thelevelofapparentlabourproductivityworksoutbelowthisaverageatalmost70%ofbusinesses.

The productivity league table is generally topped by largecorporationsthataredirectlyinvolvedinglobalvaluechains,whetherasexporters,importersormultinationals.Thishand‑fulofcompaniescontributesubstantiallytothecreationofaddedvalueandemploymentinBelgium.Theyalsoaccountforaverylargeproportionofcapitalspending,bothtangi‑ble,inbuildingsandequipment,andaboveallintangible,insuchareasasR&D.In 2014,forinstance,the2%ofBelgianbusinesses that belong to international groups generated 47%oftheprivatesector’saddedvalueand38%ofitsem‑ployment.Theywerealsoresponsiblefor42%oftangibleand 74%of intangible investment. In otherwords, thesebusinesses,whichalreadyboastasubstantialtechnologicallead,investthemostinimprovingtheirefficiency.

Less efficient companies chiefly include firms that takenopartwhatsoever intheprocessofglobalisation.Thesebusinesses,ofwhich85%areactivesolelyinthedomestic

CHART 76 RISKOFPOVERTYORSOCIALEXCLUSIONINBELGIUMBELOWEUROPEANAVERAGE,BUTHIGHERTHANINNEIGHBOURINGANDNORDICCOUNTRIES

0

5

10

15

20

25

30

2005

2007

2009

2011

2013

2015

10

12

14

16

18

20

22

24

26

28

30

SE FR NL FI AT DK DE BE UK EU ES IT

RISK OF POVERTY OR SOCIAL EXCLUSION : INTERNATIONAL COMPARISON(in % of the population aged 18 and over, 2015)

RISK OF POVERTY OR SOCIAL EXCLUSION IN BELGIUM, BY AGE GROUP(in % of the corresponding age group)

25-49 50-6418-24 65and over

source : Ec.


economy,deliverjust30 to35%ofaddedvalueandem‑ployment,40%oftangibleinvestmentandbarely11%ofintangibleinvestment.Theyareliabletofallfurtherbehindtechnologicallyandtoseetheirfinancialresultsstagnate.

oEcD research (1) suggests that agrowingproportionofbusinesses are managing to survive in the market even though their results are such that they ought to ceasetheiractivities,becausetheiraddedvalueorearningsarenegative,forinstance,orbecausetheirrevenuesarenotsufficient to cover their financial costs over a given pe‑riod.Zombiefirmslikethisaresaidtoaccountforalmost10% of businesses active in Belgium, compared to amuchlowerproportioninothercountries(justunder2%inFrance,4%intheUnitedKingdomand5%in Italy).Researchshowsthat,bycontinuingtooperate,businessesofthiskindlaidclaimtoalmost15%ofjobsandcapitalstock inBelgium in2013.Thisphenomenon reveals theless-than-optimumoperationofthecountry’sgoodsandservices markets.

The closure of the poorest‑performing businesses com‑bined with the growth of the best performers or thecreation of new businesses contributes to the ongoingreallocation of inputs. Eurostat data show that businesscreation and closure rates have both been very low in

CHART 77 DISTRIBUTIONOFAPPARENTLABOURPRODUCTIVITYINBELGIUM(1)

(in€thousandsperFTE, 2014)

500 100 150 200 250 300 350 400

10090

80706050403020100

Average

Cumulative distribution

Productivity

Perc

enta

ge o

f co

mpa

nies

source : NBB.(1) Sampleofbusinesseswithatleast20 employees.

CHART 78 GROSSBUSINESSCREATIONINBELGIUMREMAINSLOWESTAMONGEU15(1)

(in%, 2014)

J

J

J

JJ

J

J JJ

JJ

J

J

J

J

PT UK DK NL LU FR ES FI SE DE IT AT IE BE

–4

0

4

8

12

16

20

–4

0

4

8

12

16

20

Gross business creation rate (2)

Business closure rate (3)

Net business creation rate (4)

EU15

source : Ec.(1) EU15,excludingGreece.(2) Businesses created in tdividedbyactivebusinessesint.(3) Business closures in tdividedbyactivebusinessesint.(4) Differencebetweengrossbusinesscreationint and business closures in t.

(1) SeeAdaletMcGowanM.,D.AndrewsandV.Millot(2017),The Walking Dead ? Zombie Firms and Productivity Performance in OECD Countries,OECDEconomicsDepartmentWP1372.


Belgium for a very long time, comparedwithneighbour‑ing countries as well as the average for the EU15. This,combined with the greater prevalence of zombie firms,indicatesastructural lackofdynamisminthereallocationofresources.Thenumberofnewbusinesses–bothlargeand small – needs to be raised and the resources taken up bythepoorest-performingfirmsneedtobereleasedifthemanyjob-seekersmaderedundantbyrestructuringaretoberedeployedintheemploymentmarketandsustainableeconomic development is to be supported. More recent data,althoughonlyforBelgium,neverthelesssuggestthatbusinesscreationacceleratedin 2016.

One way for individuals to return to the employmentmarket is to set up their ownbusinesses.At themacro-economiclevel,businesscreationbasedoninnovativepro‑jectsisanothersourceofhigh-qualityjobcreation.Analysisof data from the Global Entrepreneurship Monitor suggests that despite the low rate of business creation, entrepre‑neurship in Belgium is more strongly growth-oriented,thankstofactorssuchasthedevelopmentofnewproductsand the internationalisation of activities.

Creating the conditions for a more robust economy and more efficient allocation of resources

strengthening the different aspects of non‑cost competitive‑nessisessentialwhenaneconomyisintransitiontoraiseitsgrowthpotentialandallocateresourcesmoreefficiently.Forexample,theinnovativecharacterofaneconomycanonlybebolsteredwhenitsenvironmentisfavourable.Giventhemanyfactorsthatplayarolehere,particularattentionmustbepaidtothewaythesefactorscomplementeachanother.

Aligning regulations with current and future needs

The regulatory framework is crucial to the optimumoperation of the wider economy, as it determines thegeneral environment inwhichbusinesses and individu‑alscandevelopandadapt.Anoverlystrictframeworkisthought to hinder the optimum allocation of resources byundulyshieldinginefficientbusinessesfrompotentialcompetition from newcomers. Where existing compa‑nies are less subject to competition, it is argued thatthey alsohave less incentive to innovate andhence tocatchuptechnologicallywherenecessary.Theoptimumlevel of regulation is, of course, open to discussion: abalancemustbefoundbetweenessentialprotectionforconsumers, employees and businesses, and the degreeofflexibilityneededformarketstofunctionsmoothly.Aclear regulatory frameworkmustalsoprevent thecrea‑tionorpersistenceofaclimateofuncertaintythatdetersinvestmentandhencepotentialgrowth,whilealsocor‑rectingflawsinthemarketandavoidingcertaintypesofdetrimental behaviour.

The various aspects and patterns of regulation can be assessed from an international perspective using a range of indicators. Generally speaking, the regulatory frame‑workinBelgiumseemstohavedevelopedpositivelysince2009. Belgium performed better than the EU averageand France in 2015, but less well than Germany, theNetherlandsandtheNordiccountries,whichtoppedtheinternational rankings. This improvement occurred against thebackdropoffar-reachinginstitutionalchanges,withawholeseriesofpowersrecentlytransferredtothefeder‑atedentitiesinthewakeofthesixthStatereform.

CHART 79 RELATIVELYFAVOURABLEDEVELOPMENTOFREGULATORYFRAMEWORKSINCE2009…

(syntheticindexforregulation;highervaluesreflectasituationmoreconducivetoeconomicactivity;EUaverage(1)in2009 =0)

JJ JJJJJJ J J J J JJ JJ J J

J JJ J J J JJJJ J J J J J J J J

J

J J J

J

J

J

–7 –6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6 7 8

HR

BG

SI

HU IT EL

RO SK

ES CZ PL AT PT EE UK FI

SE

DK LU IE

HR RO

EL

IT BG PL ES SI SK HU CZ PT

EE

UK AT IE LU SE FI DK

FR DE NL

FR

DE NL

BE

BE

2015

2009

J JJ

J JEU

EU

Sources:IMD,WEF,NBBcalculations.(1) EU,excludingCyprus,Malta,LatviaandLithuania.


The Regions have taken various initiatives to align the regu‑latoryframeworkwiththenewneedsoftheeconomy.TheBrussels-CapitalRegion’s2025 strategicplan,forinstance,provides for the gradual introduction of various measures such as the small Business Act and reducing the admin‑istrative burden for businesses, especially SMEs, in areassuchasthetransferofbusinessownership.Aplantocutred tape has also been introduced inWallonia, focusingon initiatives such as developing a range of online services forbusinesses.InFlanders,lastly,thegovernmentaimstoensurethatalladministrativetransactionswithbusinessesareoffereddigitallyby 2020.AnumberofexistingDecreeswillalsobecombinedinasingleAdministrativeDecree.

This positive general trend has the effect, however, ofdisguising contrary developments in several elementsofBelgium’sregulatoryframework.

WEF and IMD surveys of business leaders reveal, forinstance, that Belgian regulation is still viewed asmorebinding than elsewhere, especially in terms of admin‑istrative obligations. What is more, the legislative andregulatoryframeworkisgenerallyseenasnotadequatelystimulating the creation of commercial activities and as disadvantageoustobusinesscompetitiveness.Lastly,gov‑ernmentpolicyisoftenviewedasinsufficientlyadaptableinresponsetosignificantchangesintheeconomy.

These real or perceived limitations can adversely affectthe adaptability of the economic agents in the currentenvironment. Measures are therefore needed to align the legalandregulatory frameworkwithnewdevelopmentsintheeconomy,especiallyininnovativesectorswithsub‑stantialgrowthpotential.

Various reforms have also been approved recently atfederallevel.Inthisregard,administrativesimplificationisakeyelementofthefederalgovernment’sSMEplan.Anotherobjective is the introductionofamorefavour‑able framework for the development of e-commerce,by amending regulations on night work and the shar‑ing economy and setting out an appropriate fiscalframework.

In other respects, the situation in Belgium is more fa‑vourable than elsewhere. Belgian regulation is such, forinstance, that the economy has been extremely opento international tradeformanyyears. Ithasalsoprovedhighlyconducivetoinvestmentandthelocationofmulti‑nationals,bothofwhicharecrucialtothelocaldissemina‑tionoftechnology.

Improving public governance

In addition to an appropriate regulatory framework, ef‑ficientgovernmentandeffectivelyperforminginstitutionscontribute to an environment favourable to sustainable and robust economic development. The World Bank’sgovernanceanalysisfocusesonfivedimensions:govern‑ment effectiveness (1),politicalstability,ruleoflaw,controlofcorruption,andvoiceandaccountability(2).

(1) Governmenteffectivenessmeasuresperceptionofthequalityofgovernmentdepartments,thequalityofthecivilserviceanditsdegreeofindependencefrompoliticalpressure,theimplementationofpolicy,andthecredibilityofgovernmentcommitment.

(2) Voiceandaccountabilityrelatetothedegreetowhichthecitizensofacountrycanchoosetheirgovernment,andtofreedomofexpression,associationandthe press.

CHART 80 …BUTNOIMPROVEMENTINGOVERNANCERESULTS

(syntheticindexforgovernance;highervaluesreflectasituationmoreconducivetoeconomicactivity;EUaverage(1)in2009 =0)

J JJ J JJ J J J JJ JJ JJ J JJJJ

J J J J J JJJ JJ J JJ J J J J JJJ

J J J

J J J

J

J

J

J

–5 –4 –3 –2 –1 0 1 2 3 4

2015

2009

BG

RO

EL

HR HU IT

SK ES

PL SI CZ PT

EE UK

IE AT

DK

LUSE FI

RO BG

HR

EL IT

HU

SK PL ES CZ EE SI

PT UK

IE AT LU SE DK FI

FR DE

NL

FR DE NL

BE

BE

EU

EU

Sources:WorldBank,NBBcalculations.(1) EU,excludingCyprus,Malta,LatviaandLithuania.


The indicator aggregating these five aspects has notchangedmuchforBelgiumsince2009,andisabovetheEUaverage.Belgiumscoreslowerthantwoofitsneigh‑bours,GermanyandtheNetherlands,butalsolowerthantheScandinaviancountries,whichrecordthehighestrat‑ings. Analysis of the individual components shows thatthescopeforimprovementisgreatestintheruleoflaw,possiblyduetoBelgium’scumbersome legalprocedures.ItisalsointhisregardthatBelgiumdeviatesmostsharplyfrom its neighbours.

High-quality infrastructure to support economic activity

The presence of high-quality transport and communica‑tioninfrastructuregoesalongwaytowardsdetermininga country’s economic attractiveness. It also enhancesproductivity. Infrastructure investment benefits labourmobility, the physical flow of goods and the exchangeofdata,makingitimportanttothesmoothandefficientadaptation of the economy and, more generally, to anoptimum allocation of resources.

AlthoughBelgiumboastsa verygoodharbourandair‑port infrastructure, enabling it to play a central role inEuropeanvaluechains,andscoresabove theEuropeanaveragefor infrastructurequality,certaintransportnet‑work links have deteriorated in recent years, especiallycomparedwith its threeneighbouringcountries.This isparticularly the case for the railwayand roadnetwork,which regularly reaches saturation point – especiallyaroundthecountry’smajorcities–andhasgrownworseinrecentyears.

certain parts of the infrastructure have reached their op‑erationallimitsandasolutionneedstobefound.Indoingso, the emphasismust be on the essential complemen‑tarity of any investmentmade. Past policies – especially

taxation – must be adjusted to make public transportmore attractive, reduce traffic congestion and protecttheenvironment.Newtechnologiescanalsobedeployedto make more effective use of existing infrastructure.Mobility issuesdonotendatadministrativeboundaries,so initiatives in this area must also be coordinated to some degreetomaximisetheirimpact.

The authorities, meanwhile, need to create a stableregulatory framework that will enable the relevantpartiesto investasrequired inan infrastructurethat isalignedwithdevelopmentsintheenergymarketsandisdeliveredonanadequatescaleandwithintherequiredtime-frames. Examples include converting Belgium’slow-calorific gas network (L gas) into a high-calorificnetwork (H gas) by 2030. This will mean physicallyconverting some 1.5 million connections, while alsoensuringthatlow-calorificDutchgasisreplacedbygasofadifferentorigin, toguaranteecontinuityofsupplyfor affected customers.

Turning to the electricity market, the ambition is forBelgium to become a low-carbon economy, thanks tohigher renewable energy production andmore efficientdemand,forinstance.Thiswillrequireinvestmentintheelectricitymarkettoadapttheinfrastructureviawiderdig‑italisation to achieve a more decentralised and responsive distributionnetwork.Greatereffortmustalsobemadetoadjustnetworksatcross-borderlevel,tosecurethesupplyofelectricitywithintheEuropeanmarket.

Lack of entrepreneurship culture chief obstacle to reallocation of resources via creation of new businesses

A flexible and efficient regulatory framework, adequategovernance and good-quality infrastructure are neces‑sary, though not sufficient, conditions for the optimum

CHART 81 INFRASTRUCTURERESULTSCONTINUETOLAGBEHINDNEIGHBOURINGCOUNTRIES

(syntheticinfrastructureindicator;highervaluesreflectasituationmoreconducivetoeconomicactivity,averages2012-2015,EUaverage=0)

–7 –6 –5 –4 –3 –2 –1 0 1 2 3 4 5

RO

BG

PL

IT

HR

HU

SK

EE

SI LT CZ IE ES PT UK AT LU SE

FI

DK

DE FR

NL

BE

Sources:EC,IMD,WEF,NBBcalculations.


allocationofresources.Entrepreneurshaveaparttoplayinthistoo,althoughtheyarerelativelythinonthegroundin Belgium.

Entrepreneurial dynamism is clearly influenced by agreat many factors, over and above the innovativenatureof theeconomy.These factors canbebrokendown into five categories: the regulatory frame‑work,marketconditions,accesstofinance,entrepre‑neurial skills and entrepreneurship culture. A recent study(1)setouttoascertainhowBelgiumperformsinthesefive categories comparedwith the EU15 coun‑tries. Analysis of the results shows that, between2009 and 2015, Belgium scored lowest among theEU15 onentrepreneurshipculture.

Belgium rated above the European average for the other fourdeterminants. Its best relativeperformancewas in market conditions, for which Belgium camefourth out of 15, after the Netherlands, the UnitedKingdom and Ireland. The country performed slightlybetter thanaverage in termsof regulatory frameworkand entrepreneurial skills.

There is scope for improvement in each determinant,and so in 2015and 2016 thegovernment approved aseries of measures to promote entrepreneurial dyna‑mism.Theseincludetheintroductionofataxshelterforstart-upsandthesimplificationofvariousadministrativeburdens as part of the federal sME plan. Progress has been made, moreover, on harmonising employee andself-employed status. The country’s three Regions arealso pursuing various strategies to stimulate business creation.These includeencouragingsynergiesbetweenthe different bodies that support business start‑ups.

Given the large number of factors that weigh in here,however,itisimportanttoensurethatstimulationmeas‑ures complement each other and that actions in the dif‑ferentpolicyareasarealigned.

Since the weakness of the country’s entrepreneurshipculture appears to be the principal obstacle to entre‑preneurshipinBelgium,itisessentialtopresent“beinganentrepreneur”inapositive light.Thisshouldreducethe fear and stigmatisation of failure, foster creativ‑ity and risk-taking, and ensure that startingup abusi‑ness isviewedasanexceptionallyattractiveandusefulchoiceofoccupation, for individuals and for society atlarge. Kindling an entrepreneurship culture requires adeterminedapproachviaarangeofchannels,including

(1) SeeDeMulderJ.andH.Godefroid(2016),“HowtostimulateentrepreneurshipinBelgium?”,NBB,Economic Review,September,pp.65-83.

CHART 82 THEWEAKESTENTREPRENEURSHIPCULTUREINEUROPEBETWEEN2009 AND 2015

(Belgium’spositionandthatoftheEU15 countriesinthefivecategoriesofentrepreneurshipdeterminants:highervaluesreflectasituationmoreconducivetoeconomicactivity,averages2009-2015;EU15 average=0)

–6 –5 –4 –3 –2 –1 1 2 3 40

JJJJJJJJJJJ

JJJJJJJJJJJ

JJJJJJJJJJJ

JJJJJJJJJJJ

JJJJJJJJJJJ

JJJ

JJJ

JJJ

JJJ

JJJ

J

J

J

J

J

FI

UK

IE

SE

DK

LU

ATPTESITEL

UKIELUFIDKSEITPTESELAT

SEFILUDK

UK

ATIEESITPTEL

DKFISELUIEUKATPTESELIT

FIIE

DK

UKSEATLUESPTITEL

NL

DE

FR

NL

FRDE

NLDEFR

NLDEFR

NLFRDE

BE

BE

BE

BE

BE

Regulatory framework

Market conditions

Access to finance

Entrepreneurial skills

Entrepreneurship culture

Sources:EC,ECB,GEM,IMD,IMF,OECD,WEF,WorldBank,NBBcalculations.


schools and the media. Regional action plans have also been developed recently with the goal of introducingsecondary school students to entrepreneurship. Theintroductionof“student-entrepreneur“ status is a stepin this direction.

On top of R&D spending, the innovation ecosystem needs to be encouraged

Ifthepositivedynamismgeneratedbythecreationofnewbusinessesistobemaintainedandstrengthened,avarietyofconditionsmustalsobemettoensureitssubsequentdevelopmentthroughinnovation.Innovationenablesnewbusinessesand,moregenerally,establishedcompaniestodistinguish themselves and move closer to the frontiers of technology.

Innovation contributes to more efficient productionprocesses, is likely to bring down production costs andenhancesthequalityofthegoodsandservicesoffered.Italsohelpstocreatenewproductsandhencetorespondto newneeds in newmarkets. Investment in intangibleassetsandespeciallyspendingonR&Dareamongthekeydriving forces behind this process.

The upward trend in Belgian R&D spending that begansome years ago has persisted, reaching about 2.5%ofGDP in 2014. This is above the EU15 average for R&Dinvestment, and is also higher than in neighbouringcountries,withtheexceptionofGermany.However,the

figure still falls short of the 3% of GDP set out in theEurope 2020strategy.

The proportion of R&D spending accounted for bythe public sector is smaller in Belgium (28.5%) than

CHART 83 BELGIANR&DSPENDINGABOVETHEEU15 AVERAGE…

(R&Dspendingin%ofGDP; 2014,unlessotherwisestated)

SE FI DK AT DE BE FR NL UK IE IT PT LU ES EL

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Private sector

Public sector

p.m. 2006 total

EU15

source : Ec.

CHART 84 …BUTR&DSPENDINGBYTHEPUBLICSECTORREMAINSRELATIVELYLOW…

(R&Dspendingpersectorin%oftotalspending,2013)

0

10

20

30

40

50

60

70

80

90

100

Private sector

Public sector

EU15 BE Brussels Flanders Wallonia

source : Ec.


the EU15 average (35.7%) and, moreover, conceals asharp divergence between the country’s three Regions,sincethepublicsectorshareofR&Dinvestmentis lowerintheWalloonRegion.

Aside from these two findings, the performance ofBelgianbusinesses in the areasof innovation andR&D

is relatively heterogeneous. This reflects the fact thatR&D spending is chiefly carried out by large corpora‑tions, including subsidiaries of multinationals, and isconcentrated in two specific sectors: pharmaceuticalsand chemicals.

This investment is stimulatedbymeasures includinganattractive tax framework. While the aggregate level isencouraging,morebusinesses,especiallypurelydomes‑ticfirms,oughttobeencouragedtoinvestinR&Dandinintangibleassetsgenerally.EffortsinR&Dinparticularshouldbeexpanded,withaspecificfocusonSMEs.Theway that research efforts are currently concentratedmeans that the majority of Belgian businesses do notprofitdirectlyfromthepositiveeffectsofan innovativebehaviour;theyonlybenefitthroughtheirrelationshipswith corporations that do innovate. This concentrationdissipates the spillover impact of innovation.

R&Dspendingisimportantbutdoesnotinitselfpaintafullpictureoftheinnovationecosystem.TheEuropeanInnovation Scoreboard published by the EC annuallyassesses the aggregate results of Member states using a synthetic indicatormade up of eight fields: humanresources, researchsystems,financeandsupport,firminvestments, linkages and entrepreneurship, intel‑lectual assets, innovators, and the economic effectsof innovation.

The synthetic indicator shows that Belgium’s positionstrengthened overall between 2009 and 2015, and thatthe innovation ecosystem is more strongly developedin this country than for the EU15 average. The score‑boardshowsBelgiumtohavebeena“stronginnovator”

CHART 85 …ANDINVESTMENTININTANGIBLEASSETSSTILLCOMESCHIEFLYFROMLARGECORPORATIONS,OFTENMULTINATIONALS

(in%oftotalinvestmentinintangibleassets,averages2008-2014)

28%

9%

63%

Companies with foreign direct investment (parent company and subsidiaries)

Companies without foreign direct investment but linked to companies with foreign direct investment

Purely domestic companies

source : NBB.

CHART 86 BELGIANINNOVATIONRESULTSOUTPACETHEEU15

(innovationindex,scalefrom0 =weaklyinnovativeto1 =stronglyinnovative)

ES

EL

PT

IT

AT LU

UK IE FI DK SE

EL ES

IT

PT UK IE AT

DK

LU FI SE

FR NL

DE

FR NL DE

BE

BE

2015

2009

0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75

EU15

EU15

Source:EC(EuropeanInnovationScoreboarddataset).


in 2015,althoughitsresultsfellshortofthoseof“innova‑tionleaders”suchasSweden,Denmark,Finlandandtwoof our neighbours (Germany and the Netherlands). TheNetherlandsovertookBelgiumbetween2009 and 2015.It is therefore essential to maintain and foster the im‑provement currently detected in the Belgian innovationecosystem.

Variousconclusionsmaybedrawnfromthepatternofthedifferentgroupsofdeterminantssince2009 forBelgium,itsimmediateneighbours,theEUaverageandthecoun‑tries that were most advanced in terms of innovationin 2015.

Belgium’sresultshaveimprovedsince2009 inthehumanresources, research systems and linkages categories andalsoforfirminvestments,whichareestimatedbasedondatasuchasprivatesectorR&Dspending.Althoughtheeconomicimpactofinnovationalsoincreasedsomewhat,Belgiumscorespoorlyinthiscategorycomparedwiththeotherdeterminants.Allthesame,itiscruciallyimportantto turn an innovative attitude into new products andproductionprocesses,sothatthewidereconomystandsto gain.

Financeand support,measuredby factors including theshareofR&Daccountedforbythegovernmentsector,isthe determinant for which Belgium’s score deteriorated

most sharply between 2009 and 2015. “Intellectual as‑sets”–ayardstickforaseriesofaspectsrelatingtothedevelopmentofpatents–and“innovators”–theestimat‑ed percentage of corporations that achieve innovations in termsof products, processes,marketingor organisation–alsoshowedanunfavourabletrend.

This scoreboard can be used to draw a general pictureofBelgium’sstrengthsintheareaofinnovation.Belgiumoutperforms the EU15 and its neighbours on humancapital and research systems, but also in terms of col‑laboration.Thelatterrelatestocollaborationbetweenin‑novativeSMEs,forinstance,andtopublic-privateresearchpartnerships. The greatest scope for improvement, bycontrast, isfoundintheeconomiceffectsof innovation,patentsandfinance.

Like Belgium as a whole, the country’s three Regionsare rated as “strong innovators”. Although the dataare not available in full for all the determinants of in‑novation,avarietyofdifferentdevelopmentshavebeenidentified.

More specifically, the overall result for theBrussels-Capital Region declined by 11% between2013 and 2015. The Innovation Scoreboard showsthattheRegionneverthelessscoreswellforcollabora‑tionbetween SMEs, human capital and the export of

CHART 87 DETERMINANTSOFINNOVATION

(scalefrom0 =unconduciveto1 =highlyconducivetothedevelopmentofinnovation, 2015)

1

1

1

1

1

1

1

1

0

Human resources

Research systems

Finance and support

Firm investments

Linkages

Intellectual assets

Innovators

Economic effects

Belgium Neighbouring countries Nordic countries EU15

Source:EC(EuropeanInnovationScoreboarddataset).


medium‑ and high‑tech products. There is still scope for progress,however,forR&Dspendingbyboththepublicand private sectors. The aggregate result for innovation alsodroppedby11%inWallonia inthesameperiod.PrivateR&Dexpenditureandtheexportofaverage-andhigh-quality goodswere higher than the EU average,but organisational and marketing innovations tended tobelesswelldeveloped–whichwasalsothecaseforgovernmentR&Dspending.Flanders,bycontrast,keptupits2013 results.Innovationinproductsandproduc‑tionprocesseswasratedparticularlywellthere,aswascollaborationbetweenSMEs.Nevertheless,governmentR&D spending ought to be greater, as in Wallonia,along with innovation in the areas of marketing andorganisation.

WhilebothR&Dexpenditureandthesynthetic indicatorarehigher inBelgium than theEU15 average, thisdoesnot diminish the fact that progress on innovation is still possible.Withtheexceptionofcertainfiscalaspects,mostpowersinthisrespecthavebeentransferredtothecoun‑try’s three Regions, which therefore need to take freshmeasures topromoteR&Dspendingandspread innova‑tionmoreevenlyacrossthewidereconomy.Aninnovativeeconomy is,afterall,betterable toreapthebenefitsofchangeandbetterplaced todeploy the largestpossiblenumberofemployeesinthelabourmarket.

Seizing growth opportunities in foreign markets

Globalisation ought to be viewed not only as a threat,but also as an opportunity. For a small, open economylikeBelgium,takinggreateradvantageofincreaseddyna‑mismincertainforeignmarkets–especiallythoseintheemerging countries – is an important lever for adapting andexpandingthecountry’seconomy.

The development of these markets provides an oppor‑tunity forBelgianexporters,whether theyareactive inthem directly or benefit from them indirectly throughtheir integration in global value chains. Globalisation thusoffersBelgianexportersagrowthopportunitythatis both extensive, if able to enter new markets, andintensive, through higher exports to their traditionalmarkets.

Analysisofmicroeconomicdataallowsustobreakexportgrowthdownintothesedifferentmargins.Threesourcesofextensivegrowthareidentified:theappearanceofnewexporters,enteringnewdestinations,andexportingnewproductstoexistingmarkets.

Theintensivemarginhasplayedthemostimportantpartin Belgian export trends since 1995, delivering 60%ofexport growthbetween1995 and 2015. This sharehasactually risenashighasalmost80% in recentyears. Inother words, products already sold on foreign marketshave accounted for the majority of growth in Belgianexports.Newproductsaccountedforthelargestshareoftheextensivemarginsfrom2011 to 2015.Thefigureforthecontributionofcompaniesexportingforthefirsttimewasbarelypositiveinthisperiod.

6.3 smoother and faster transition into workforamoreinclusivelabourmarket

Job insecurity, income inequality and risk of poverty limited in Belgium for workers...

Thebreakdownofpovertyriskshighlightstheimportanceoflong-termlabourmarketparticipation,accompaniedbyhigh-qualityjobsandgoodpay.

WorkingpeopleinBelgiumareatverylowriskofpoverty,wherepovertyisdefinedasanincomebelowthepovertylevel. In 2015,4.5%werecategorisedaspoor.TheonlyEUcountrywithaneven lowerpercentagewasFinland,at3.5%.The threatofpoverty ismuchgreater for job-seekers(40.4%)andinactiveindividuals(31.5%).Despitethegreatrecession,therelevantindicatorhasbeenfairly

CHART 88 INTENSIVEMARGINSDELIVERLARGESTCONTRIBUTIONTOBELGIANEXPORTS

(growthinexportsandcontributionoftheintensiveandextensivemargins–goodsbyvalue,cumulativecontributiontogrowth,unlessotherwisestated)

1995-2000 2001-2005 2006-2010 2011-2015

–20

0

20

40

60

80

100

120

–2

0

2

4

6

8

10

12

New businesses

New destinations

New products

Intensive margin

Growth in exports (average year-on-year, 1995-2015)

Growth in exports(average year-on-year for each sub-period)

(left-hand scale)

(right-hand scale)

Sources:NAI,NBB.


stableinBelgiumoverthepasttenyears.Onlyamongtheunemployed has the risk of poverty increased since thebeginningofthecurrentdecade,duetofactorsincludingthefinancialcrisis,morestringentfollow-upofjob-seek‑ingbehaviourandreformstounemploymentinsurance.

Elements generally factored into any assessment of thequalityofajobincludenotonlythepayreceived,butalsotheperceivedpaylevelsofotherworkers.Belgiumscoreswellinthisrespecttoo,asincomesfromlabourarehighinthiscountry–grossmedianhourlypayintheprivatesec‑torwas€17 in 2014–andtherangeislimited.Theratioofhourlypaybetweentheninthandfirstdecilesstoodat2.4 in 2014,whichiscomparablewiththefiguresfortheNordic countriesand slightly lower than those inneigh‑bouring countries.

The proportion of low-paid employees – those whosepay is less than two-thirds of the national median grosshourlywage–is,incidentally,oneofthelowestinEuropeat less than 4%. This figure is virtually identical formenandwomen,whereas theEUaverageshowsawidergapto thedisadvantageofwomen.The sharpdecline in thatsharewhichoccurredinBelgiumbetween2006 and 2014reflectstheabolitionofthedegressiveminimumwageforemployeesunder21,whoareoverrepresentedinthelow-paid group. The government considered reintroducing the degressiveelementin 2016.

Moreover,labourmarketinsecurity–whichtheOECDesti‑matesbasedonacombinationoftheriskofunemployment,

the anticipated duration of unemployment and the re‑placement rateofunemploymentbenefits relative to thepreviouswage – is low in Belgium, at 3.6%. This partlyreflects the relatively limited loss of income in the eventof unemployment, other than for highly-skilled people.Jobinsecurityismuchgreater,however,foryoungpeople,menandthelow-skilled.Educationandtrainingthereforehelpnotonlysmootherbutalsolonger-termentryintothelabour market.

… but traditional organisation and functioning of the labour market have been altered by growing need for flexibility

Theorganisationoflabour,however,seemstobebecom‑ingmore flexible, due inter alia to the development ofnewtechnologies.Thefinancialcrisisandtheassociateddifficultyofenteringthetraditionallabourmarket,alongwith the unemployment insurance reforms, have alsocontributed to this increased flexibility. This flexibility issupportedforworkers through, forexample,permissiontocombineearnedincomeandpension,andalsothroughtheopportunitiesofferedby“self-employedinsecondaryoccupation”status.Workers’needtobalance theirpro‑fessional and private lives is another contributory factorhere, as is the feminisationof the labourmarket, giventhatwomenaremorelikelytoworkpart-time.Althoughtraditional, salariedemployment–definedasa full-timejobwithapermanentcontract–remainsthemostcom‑monpattern,anincreasingnumberofnewformsofem‑ploymentareemerging.Theseeitherfollowonfromeach

CHART 89 MUCHLOWERRISKOFPOVERTYFORBELGIANWORKERS

(in%ofthepopulationaged18 to64)

2005

2007

2009

2011

2013

2015

0

5

10

15

20

25

30

35

40

45

50

2005

2007

2009

2011

2013

2015

0

5

10

15

20

25

30

35

40

45

50EU

(1)BELGIUM

Employment UnemploymentTotal Inactivity

source : Ec(1) ExcludingCroatia.


otherduringaperson’scareerorareactuallycombined.More and more people are combining different activities in this way, possibly with different statuses: employeewithatemporaryemploymentcontract,full-timeorpart-timetemporaryworker,student,student-entrepreneur,orself-employedinaprimaryorsecondaryoccupation.

According to figures from theNational Institute for theSocialSecurityoftheSelf-employed(NISSE),thenumberof self-employed inaprimaryoccupation rose11%be‑tween2005 and 2015.Thenumberofself-employedinasecondaryoccupationrose39%and55%morepeopleremained active after retirement.

CHART 90 PROPORTIONOFLOW-PAIDEMPLOYEESAMONGLOWESTINEU

(employeesearninglessthantwo-thirdsofthenationalmediangrosshourlywage,in%, 2014)

DE NL AT FR DK FI BE SE0

5

10

15

20

25

0

5

10

15

20

25

2014 2006 Men Women

EU EU BE

source : Ec

CHART 91 LABOURMARKETINSECURITY(1)ISLOWINBELGIUMANDDECLINESASEDUCATIONALLEVELANDAGEINCREASE

(in%,2013)

AT DK FI DE NL FR SEBE

0

2

4

6

8

10

12

14

16

0

10

20

30

40

50

60

70

80

0

1

2

3

4

5

6

7

B

J

15-2

9

30-4

9

50-6

4

INTERNATIONAL COMPARISON BELGIUM

Risk of unemployment(left-hand scale)

Labour-market insecurity

Unemployment insurance (right-hand scale)

Hig

h

Med

ium

Low

Wom

en

Men

Educational level Age Gender

source : oEcD.(1) Definedasanticipatedlossofincomeintheeventofunemployment.Thislossdependsontheriskofunemploymentandthegenerosityoftheunemploymentbenefit.(2) Definedastheriskofbecomingunemployedandtheanticipateddurationofunemployment.(3) Definedbasedonthereplacementandcoverageratios,namelythepercentageofunemployedjob-seekersentitledtounemploymentbenefit.


Some workers who have difficulty securing permanentemployment–youngpeopleandthoseofforeignorigin,forinstance–canaccessthelabourmarketmorereadilythrough contracts of this kind. others opt for them be‑causeofthehighdegreeofflexibilityandautonomytheyoffer.Theyentailgreater risk,however, thanpermanentsalaried employment. Self-employed in secondary occu‑pation status canmitigate that risk,which undoubtedlyhelpsexplainitssuccess.

Although the proportion of temporary employmentcontracts remained fairly stable (anaverageof8.5%ofsalaried employment between 2000 and 2015), that ofshort-termemploymentcontracts(lessthanoneyear)rosefrom75%ofthetotalnumberoftemporarycontractsin2005 to82%in 2015.Theproportionofveryshort-termcontracts(threemonthsorless)roseinturnfrom31%to36%. Temporary employment ismore common amongyoung people: in 2015, the figure for 15-24-year-oldswas 37% of salaried employment, compared to 9%for 15-64-year-olds. These numbers back up the datafor labour market uncertainty, which is much greaterforyoungpeople.

These types of employment contract, along with self-employedstatus,canoffersmootheraccesstothelabourmarket than traditional salaried employment, wherecompetitionwithexperiencedemployeesismoreintense.All the same, the growth in temporary employment

contracts chiefly affects the low-skilled, complicatingtheirworkingcareersand limiting their socialprotectionrights.Householdswithmodestor irregular incomesarealsoexposedtoagreatercreditrisk.Variousgovernmentschemesprovideeasieraccesstofinance,particularlyforthe purchase of a home.

Thesenseof jobsecuritydependsonnumerous factors.The risk of losing one’s job, the time needed to findanother and the future terms of employment all needto be factored in. Belgium scoreswell in terms of bothemploymentstabilityandemploymentconditions,butthelikelihoodofre-enteringthelabourmarketislowerthanin other countries.

A feature of the Belgian unemployment insurance sys‑temisthat it isnottime-limited,unlesstheunemployedperson receives a sanction. Long-term unemployment(over36 months)isproportionatelyhigherinBelgiumonastructuralbasisthaninothercountrieswhereseparaterulesapplytothe long-termunemployed.Sincetherateoftransitionamongthe long-termunemployed is lower,internationalcomparisonisunfavourableforBelgium,asno account can be takenof the duration of unemploy‑mentbasedontheharmoniseddata.Allthesame,transi‑tionintoworkmustbefacilitatedusingthevariouslevers

CHART 92 GROWINGSUCCESSOFSELF-EMPLOYEDSTATUS

(inthousandsofpeople)

2005 2007 2009 2011 2013 20150

200

400

600

800

1 000

1 200

Primary occupation

Secondary occupation

Activity after retirement

Source:NISSE.

CHART 93 GROWINGNUMBEROFTEMPORARYEMPLOYMENTCONTRACTSOFSHORTORVERYSHORTDURATION

(in%ofthetotalnumberoftemporaryemploymentcontracts)

2005 2007 2009 2011 2013 20150

10

20

30

40

50

60

70

80

90

100

Maximum 3 months

From 4 to 6 months

From 7 to 12 months

Over 12 months

source : Ec.


– initial and lifelong learning, guidance for job-seekersand creating opportunities for them thanks to new jobopenings infirms,whichwillgohand inhandwithsus‑tainableeconomicgrowth.

Initial training : the cornerstone of young people’s entry into the labour market

Entry intothe labourmarket foryoungpeoplewhodonot yet have any professional experience depends pri‑marilyontheirbasictraining.Some12%of15-24-year-olds in Belgium are not in employment, education ortraining, a higher percentage than in neighbouringcountries. The Netherlands, Denmark, Germany andSweden are the best performers in this respect, withrateshalf as low. Substantial differences exist betweentheRegions:Flanders,withafigureof less than10%,is doing better than the two other Regions (15% inWalloniaand17.5%inBrussels).Partoftheclearpro‑gress that was achieved between 2000 and 2008 hassince been lost. These young people do not have suf‑ficientwherewithaltoenterthelabourmarketonasus‑tainablebasis.Theyexperiencegreaterdifficultyduringeconomic slowdowns, when theymust competemorethaninnormaltimeswithother,betterqualifiedyoungpeople or with experienced employees. This situationentails high individual and social costs in the short term and,asworkingcareersbecomelonger,aboveallinthelong term.

Certain groups are over-represented among youngpeopleat riskofbeingexcluded fromtheemployment

CHART 94 TRANSITIONFROMUNEMPLOYMENTTOWORK:INTERNATIONALCOMPARISON(1)

(in%ofthenumberofpeoplewhowereunemployedoneyearearlier, 2014)

DK SE UK FR DE BE ES20

25

30

35

40

45

50

55

source : Ec.(1) Estimatedtransitionrateforasinglemanaged35-39 ofaverageeducation.

CHART 95 12%OFYOUNGPEOPLEINBELGIUMARENOTINEDUCATION,EMPLOYMENTORTRAINING(1)


2001

2003

2005

2007

2009

2011

2013

2015 N

L

DK DE SE FI FR

0

5

10

15

20

25

30

0

2

4

6

8

10

12

14

16

18

20

BE

Flan

ders

Wal

loni

a

Brus

selsEU

BE EU

Flanders

Wallonia

Brussels

EVOLUTIONINTERNATIONAL AND REGIONAL COMPARISON (2015)

source : Ec.(1) NEET.


market,particularlynon-Europeanforeigners.ThisisnotatypicallyBelgianphenomenon,butitisespeciallypro‑nouncedinthiscountry.Youngnon-EUcitizensarelesslikely to have followed nursery education, their schoolresultsaresubsequentlylessgoodandtheireducationallevel remains lower on average than that of Belgians,eveniftheywereborninBelgium.Thesedifferencesareattributable to factors such as language barriers and cul‑turalobstacles,aswellasthefactthatteachersarenotadequately trained to dealwith an increasingly diversepopulation.

Generally speaking, Belgium’s education system tendsto perpetuate socio-economic inequalities. Despite itslow tuition fees and the availability of student grantsforlow-incomehouseholds,segmentationexistsamongeducational institutions. The most disadvantaged schools havetoofewresourcestoattractandretainexperiencedteachers.What ismore, students are channelled at anearlystagetoschoolsforgeneral,technicalorvocationaleducation.Inviewofthesechallenges,theCommunitiesrecently began to encourage participation in nurseryeducation–especiallyfortheunder-representedgroups– to improve the social and culturalmix in schools, topostpone streaming and to provide better guidance at thatpoint,andtolimitthenumberofstudentsretakingasecondaryschoolyear.

Technical and vocational education are also being increas‑inglyvaluedandsupportedbygovernmentandbusiness.These shorter streamsallow faster entry into the labourmarket. They can also respond more effectively to the

constantlychangingneedsofbusinesses.Apprenticeships,forinstance,offerconsiderablecareeropportunities.Thisform of education seems, moreover, to be spreadingto all industries and levels of education. Apprenticeship contracts have yet to establish themselves in Belgiumas firmly as in countries such as Germany, Austria orDenmark, where the formula has proved very success‑ful.Vocationallyorientedstudiesarestilloftenasecondchoice in Belgium.

CHART 96 NON-EUCITIZENSNOTSUFFICIENTLYINEDUCATION

0

5

10

15

20

25

0

10

20

30

40

50

60

70

80

90

100

Belgians Other EU Non-EU

NEET BY NATIONALITY (2015)

Nat

iona

l citi

zens

Oth

er E

U

Non

-EU

Belg

ians

Oth

er E

U

Non

-EU

EU BEBE

BELGIUM : EDUCATIONAL LEVEL (population aged 30-34, in %, 2015)

Low-skilled

Medium-skilled

Highly-skilled

source : Ec.

CHART 97 APPRENTICESHIPSPOORLYESTABLISHEDINBELGIUM

(in%ofthepopulationaged15 to29, 2014)

DE AT DK FR FI BE SE NL0

2

4

6

8

10

12

14

source : Ec.


At the same time, more and more young people aretaking up higher education. In 2015, almost 43% ofthepopulationaged30 to34 heldahighereducationqualification,8 percentagepointshigher than in2000.Amongthemostrecentcohorts,morewomenthanmenholdadiplomaofthiskind:foreverytwomalegradu‑ates there are now three female ones. Consequently,their career prospects are better too. Sufficient avail‑able and affordable childcare and personal services also contribute to a balanced distribution of employmentprospectsbygender.

Fewer higher education graduates in Belgium holdscientific or engineering qualifications than the EUaverage. Yet these are rewarding specialisations thatarecloselyalignedwithmany“hard-to-fill vacancies”.DespitetherequirementsoftheeconomyandtheneedtofostertheinnovativepotentialofBelgianbusinesses,thepercentageofgraduatesinthesesubjectsdeclinedbetween2005 and 2015.Thismismatchbetweencho‑sen subject fields and the needs of themarketmightexplain why the employment rate for highly-skilledpeople fell from 84.4% in 2000 to 81.8% in 2015.Nowadays,holdingadegree isno longer sufficient tofindajob:thechoiceofcourseisalsoakeyfactor.

Thatchoicecontinuestovarybygender:womenareover-representedoncoursesrelatingtoeducationandhealth,but under‑represented in engineering studies and the sciences. These choices are also expressed inwagepat‑ternsandexplainsomeofthepaygapbetweenmenand

women.ThedifferenceingrosshourlypayinBelgiumwas6.6%in 2014,oneofthelowestintheEU.Yet,womenstillmovelessreadilyintomanagementpositions,inboththe private and public sectors.

CHART 98 BELGIUM’SGROWINGNUMBEROFYOUNGPEOPLEWITHHIGHEREDUCATIONQUALIFICATION


2001

2003

2005

2007

2009

2011

2013

2015

0

5

10

15

20

25

30

35

40

45

50

SE DK NL FI FR BE

DE

0

10

20

30

40

50

60

EU

BE

Brus

sels

Flan

ders

Wal

loni

a

EU

EVOLUTIONINTERNATIONAL AND REGIONAL COMPARISON (2015)

source : Ec.

CHART 99 TOOFEWSTUDENTSOPTFORSCIENCEORENGINEERINGCOURSES

(highereducationgraduatesbysubjectareaandgender,in%oftotalnumberofgraduates, 2014)

BE BE0

5

10

15

20

25

30

H

Sciences, maths and IT Engineers, process industry and construction

Men

Women

Total

EU EU

source : Ec.


Smoother transition from unemployment to work thanks to active policy and efficient unemployment benefits

Training is not only important for young people: longerworkingcareersandtechnologicalchangemeanthatwork‑ersmustbeable to acquire the skills needed for them toconsiderswitchingjobsorsectorsthroughouttheirworkinglives.Lifelonglearningisessential.Allthesame,barely7%ofworking people between the ages of 25 and 64 tookanytrainingin 2015.ThisfigureplacesBelgiumwellbehindDenmarkandSweden,wheretheproportionwasjustunderathird,butalsosubstantiallybelowtheEUaverage.

Certain groups, most notably the low-skilled and over-55s,undertakeverylittletraining.Participationintrainingbyjob-seekers,bycontrast,isclosetotheEuropeanaver‑age in Belgium, reflecting the numerousmeasures thathavebeentakenbypublicemploymentservices.

Analysisofthetransitionfromunemploymenttoworkre‑veals several important factors in addition to educational level:thebusinesscycle,theperson’sageandhowlongthey have been unemployed. The longer a person hasbeen jobless, the less likely they are to findwork. Thisnegativecorrelationisgenerallyattributabletothegradualdeteriorationofunemployedpeople’shumancapitalortoa sense of demoralisation resulting in a less active search forwork. Itcanalsobetheresultofagradualselectionprocess:peoplewhoremainoutofworkforlongermaybeviewedonaverageaslessemployablethanthosewhohavequickly foundnewwork. These explanationsneednot,however,bemutuallyexclusive.Sinceitisharderforthe long-term unemployed to find another job, efforts

aremade toassist job-seekers at anearlier stage,whilestill maintaining a degree of selectiveness. To this end,publicemploymentservicescandrawonprofilingmodels,whichuseobjectivecriteriatodeterminewhichindividualsrecentlyregisteredasunemployedareatgreatestriskofremainingjobless.Themostvulnerablejob-seekersarethelow-skilled,theelderlyandnon-EUcitizens.Employmenthistoryandhealthmustalsobetakenintoaccounttosup‑portunemployedpeopleefficiently.

CHART 100 LIFELONGLEARNING:EDUCATIONLEVEL,AGEANDSOCIO-ECONOMICSTATUSINFLUENCEPARTICIPATIONRATE

(in%,25-64-year-oldswhohaveundertakentraininginpastfourweeks, 2015)

0

5

10

15

20

BE

25-34 35-44 45-54 55-640

5

10

15

20

0

2

4

6

8

10

12

14

Low Medium High

EDUCATIONAL LEVEL AGE

Working Unemployed Inactive

SOCIO-ECONOMIC STATUS

EU

source : Ec.

CHART 101 OUTFLOWINTOWORK(1)REDUCESWITHAGEANDEDUCATIONLEVEL

(in%ofpeoplewhowereunemployedoneyearearlier, 2014)

20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-640

10

20

30

40

50

60

Highly-skilled

Low-skilled

Medium-skilled

source : Ec.(1) Estimated transition rates based on Belgian data according to a multivariate

model,forasingleman.


4

Keeping over-50s in work and providing jobs for a larger proportion of unemployed in this age group

The likelihood of finding a new job falls off sharply for50-64-year-oldscomparedtoallyoungeragegroups.Thisageeffectispartlyattributabletothecorrelationbetweenageandlengthofunemployment:thereare,onaverage,more long-term unemployed among older people thanyoungerpeople.Sincepayriseswithexperience,olderjob-seekersfrequentlyearnmorebeforelosingtheirjobsthanyoungpeopledo.Thispushesuptheirreservationwage,sothattheyrespondlesstovacanciesandremainunemployedfor longer.Replacement incomeinallunemploymentsys‑temsisbasedonpreviouspay.Olderpeoplearethereforeentitledtohigherbenefits,atleastinitially.Somecountries,moreover, have introduced rules that make the systemevenmoregenerous,withafixedsupplementaryallowancefor over-50s, for instance, or by extending the durationof the benefit. Belgiumhad a systemof this kind in theshapeof“seniorityincrements”,whichwereabolishedon1 January 2015. Higher unemployment benefits like thistendtoresultinaloweraverageoutflowintowork.

Other institutional features besides unemployment ben‑efit parameters also affect the outflow of older peopleintowork.Inthepast,forexample,itwascustomaryatatimeofbothrobustgrowthintheworkingagepopulationandeconomicweaknesstopromoteearlyretirement.Thiswasdoneeitherby introducingaspecific regime–suchasthe“pre-pension” inBelgium–orbyadjustingexist‑ing regimes (unemploymentor disability insurance). Theindividualsinquestionmightnolongerbeobligedtoseek

work,forexample,whilestillreceivingbenefitsequivalenttothoseofunemployedjob-seekers.Thesemeasureshavebeengraduallytighteneduporabolished,asmentionedinchapter2 ofthispartoftheReport.

Demand for experienced employees is another decisivefactor, in addition to those relating to the supply of la‑bour, institutions and policymeasures to promote earlyretirement. In Belgium in particular, it can be expensiveforabusinesstorecruitolderemployees–evenallowingfor productivity – because the seniority element plays asignificantpartinpayincreases.Whatismore,apropor‑tionoftheexperiencebuiltupduringtheperson’scareerwillbeofaspecificcharacterthatmightnotbeusefultonew employers. The latter could also be put off by theindividual’srelativeproximitytoretirement,whichmeansthat theperiod inwhich the employer can recoup theirinvestmentmightbeshorterthaninthecaseofayoungeremployee.Lastly,agediscriminationinrecruitmentcannotbe ruled out in some cases.

Thanks to legislation to promote “workable and flex‑iblework”,asdiscussed inchapter2 of thispartof theReport., employers and employees ought to be able toadapt their employment conditions more smoothly, forinstancebytakingaccountofthecareerdimension.

Lastly, thesevarious instruments toboostentry into thelabourmarketought tobe viewedas supplementary tothemeasurestakeninrespectoflabourcosts,sincereduc‑ing the fiscal and parafiscalwedge boosts the financialstimulitothesupplyofanddemandforlabour.

Box7–SupplementaryindicatorsforBelgiuminadditiontoGDP

Thisboxsummarisesthemainresultsofthesupplementary indicators inadditiontoGDP,ofwhichtheFederalPlanningBureau(FPB)publishedthefirstannualupdateatthebeginningofFebruary2017.InaccordancewiththeLawof14March2014,thisseriesofindicatorsisintendedtomeasurequalityoflife,humandevelopment,socialprogressandthesustainabilityoftheBelgianeconomy.

Theindicatorswerecalculatedusingasimilarapproachtothemethodadoptedforthefirstpublicationin2016.67indicatorswereselected,whichprovideinformationonthreedimensionsofsustainabledevelopment:“Hereandnow”,“Later”and“Elsewhere”.Whererelevant,theindicatorswerebrokendownbygender,age,incomeandeducationallevel.Internationalcomparisonsweremadefor56indicatorsforwhichthenecessarydatawereavailable.Fordetailed informationregardingthemethodologyusedtoselect thethemesand indicatorsandtoperformthebreakdowns,seetheFPB’sreportofFebruary2016(1).

(1) Seehttp://www.plan.be/publications/publication-1541-fr-indicateurs+complementaires+naast+pib


Trends in the indicators

Thediversenatureofthesupplementary indicators inadditiontoGDPmeanstheyofferrelevant informationona largenumberof themes. For thepurposesof thisReport, they complementanalysisof thepositionofindividualsinsocietyandinthelabourmarket,aswellasthatofthepublicdebtorthefinancialpositionoftheeconomy.

Theindicatorsalsoenableafewpreliminarygeneralobservationstobemadeaboutthepatternandsustainabilityofwell-beingandhowsociety isdeveloping.Basedon theprinciples setout in theFPB report, thisevaluationexclusivelyexaminesthedevelopmentofthe indicatorsand,morespecifically,whethertheyhavemovedclosertoordeviatedfromthetargetssetformostofthem.Theseobjectivesaredeterminedbygovernment,eitherinBelgium or at European or global level.

Thisevaluationshouldbeinterpretedwithcaution,asithasnothingtosayaboutthecurrentleveloftheindicatorsortheirpaceofdevelopment.Norhasitbeeninvestigatedwhethertheobjectivesarescientificallyrelevant.Whatismore,thedataarelimitedforroughlyhalfoftheindicators,duetoalowfrequencyofobservationorbecausetheinformationhasonlyrecentlybecomeavailable.

The“Hereandnow”dimensioncomprises38indicators,whichcananswerquestionsaboutBelgians’well-beingandhowthisisdeveloping.Nosystematictrendcouldbeidentified.

– Subjectivewell-beingremainedstablebetween2002and2014.– Most of the indicators relating to the “Health” and “Education and training” themes showed a positivedevelopment.Theonlydeteriorationwasintheindicatorforpostponementorcancellationofmedicaltreatmentforfinancialreasons.

SELECTIONOFSUPPLEMENTARYINDICATORSINADDITIONTOGDP(1)

2001

2003

2005

2007

2009

2011

2013

2015

0

10

20

30

40

2001

2003

2005

2007

2009

2011

2013

2015

0

40

80

120

160

2001

2003

2005

2007

2009

2011

2013

2015

1

2

3

4

5

6

7.1

7.2

7.3

7.4

7.5HERE AND NOW

70

74

78

82

86

0.2

0.3

0.4

0.5

0.6

0.7

Trust in institutions (score at least 6 out of 10, in % of the population over 15 ; right-hand scale)

Satisfaction with life (average score out of 10 ;left-hand scale)

Life expectancy, women (years at birth ; left-hand scale)

Greenhouse gas emissions (in million tonnes carbonequivalent ; right-hand scale)

Life expectancy, men (years at birth ; left-hand scale)

Domestic use of fossil fuels (tonnes per capita ; right-hand scale)

Official development aid (in % of GNP ; left-hand scale)

LATER ELSEWHERE

source : fPB.(1) TheindicatorsconsideredhereforeachofthethreedimensionsofthesupplementaryindicatorsinadditiontoGDPwereselectedtoavoidoverlapwithseriesused

elsewhereinthisReport.

4


4

– Incomeinequalitydevelopedfavourably,buttheotherindicatorsforthe“Standardoflivingandpoverty”themedeviated further from their targets.

– Formostofthe indicatorsrelatingtothe“Workand leisure”theme,the improvementsdetectedsince1990werepartiallyreversedbycontrarydevelopmentsinthepastfiveyears.

– Threeofthefivefinalindicatorsforthe“Hereandnow”dimension(PM2.5 and Noxemissionsandroaddeaths)movednearertheirtargets,whilethoserelatingtoenergydependencedevelopednegatively.Theindicatorforvictimsofnaturaldisastersistoovariableforanysignificantconclusionstobedrawnfromit.

The“Later”dimensionismadeupof35indicatorsrelatingtothecapacityofBelgiansandofsocietytomaintainandeven improvewell-being in the future.Most indicators for this dimensionmoved closer to their targetsduringboththeperiodsince1990andthefivemostrecentyears.Themostnoteworthyexceptionsaresetoutbelow:

– Lifeexpectancyingoodhealthhasdeclinedinthepastfiveyears.– Anegativepatternisalsofoundintheproportionofyoungpeoplenotineducation,employmentortraining,forstressatwork,andthepercentageofpeoplewithhighereducationqualifications.

– Theindicatorsforthemeadowbirdpopulationandsustainablefisheries(“Landandecosystems”)–twooftherareindicatorsforbiologicaldiversitythathavebeenavailableoveralongperiodoftime–deviatedfromtheirtargets.

– Withinthe“Economiccapital”theme,thepublicdebtindicatorhasbeendeviatingfromitstargetsince2007.– Withinthe“Mobilityandtransport”theme,theshareofroadtransportinfreighttransportdeviatedfromitstargetfrom1990onwards,buthasbeguntomoveclosertoitagaininthepastfiveyears.

The “Elsewhere” dimension incorporates eight indicators that help answer the question: “How does thedevelopmentofsocietyinBelgiuminfluencethecapacityofothercountriestodevelopandthewell-beingoftheirpeople?”.

– The indicators for the“Natural resources”and“Climateandenergy” themes relate toglobalenvironmentalcapital,suchastheclimatesystemandrawmaterialsresources.Theseindicatorsconvergedwiththeirtargets.

– TheshareofGDPdevotedtoofficialdevelopmentaidwashigheronaverageinthe2000-10periodthaninthe1990s.Ithasdeclinedsince2010,dippingbackbelowthe1990levelin2015.

Takentogether,theindicatorsforenvironmentalaspectshavegenerallydevelopedmorepositivelysincetheeconomicandfinancialcrisisof2008thanhavethoserelatingtosocialthemes.Thisreflectsthefactthatthegreatrecessionandthesubsequentprolongedperiodofweakgrowthadverselyaffectednumeroussocialindicators,includingtheriskofpovertyorsocialexclusion,whiletheeconomicslowdownsomewhateasedpressureontheenvironment.

Internationalcomparison

Comparisonof the indicatorswith theEuropeanaverageorwithneighbouring countriesgenerally shows thatBelgiumhas performedwell on the social indicators, but poorly on those relating to the environment. This isattributable,ontheonehand,tothefactthatBelgium’ssocialsecuritysystemismoreadvancedthanthatofmanyotherEuropeancountrieswhile,ontheother,pressureonthelivingenvironmentinBelgiumishighduetoitshighpopulationdensityandthespecialisationofBelgianindustryinintermediaryactivities,whicharemorepollutingthan the average.


Breakdownoftheindicators

27ofthe67supplementaryindicatorsinadditiontoGDPwerebrokendownintorelevantpopulationcategories.Breakdownbygendershowsthatmanyofthedifferencesbetweenmenandwomenarenarrowing,despiteafewpersistentinequalities.Thedetecteddifferencesintermsofeducationallevelremainandhaveactuallywidened.Lastly,asdescribedatthebeginningofthissection,thesituationofyoungpeoplehasdeteriorated,whilethatofolder people has improved.

Breakdownbyincomelevelrelatestoindicatorsforfourthemes:“Subjectivewell-being”,“Standardoflivingandpoverty”,“Health”and“Society”.Analysisoftheseshowsthatthesituationismorefavourableforthehighestincome categories :

– The“Subjectivewell-being”theme,for instance,showsthatsatisfactionwith life increaseswith income.Theconfidenceintervalscalculatedinthisregardarerelativelylarge,however,andthistrendhasyettobeconfirmedbyotherobservations.

– Withinthe“Standardoflivingandpoverty”theme–theindicatorsforpovertyorsocialexclusion–verylowworkintensityandseriousmaterialdeprivationareveryhighamongthequintilewiththelowestincome,andthreetofivetimeslowerinthesecondquintile,whereastheyareextremelysmalloreffectivelynon-existentinthethreeotherquintiles.

– When the indicators for “Health” were broken down by income, this too showed that the proportion ofsmokersandpeoplesufferingfromobesityordepressiondeclineswithrisingincomelevels,asdidpostponingorcancellingmedicalcareforfinancialreasons.

– Within the “Society” theme, general confidence, sense of security in public spaces and trust in institutionsincreasewithincomelevel.Theindicatorforcontactwithfriendsandrelatives,bycontrast,doesnotchangebyincome.Thevictimofcrimeindicatorgenerallyincreaseswithincome,albeitnottoaverypronounceddegree.

Thoughts for the future

Thesupplementary indicators inadditiontoGDPwillbeupdatedannuallyandmaybeadjusted inresponsetonewideasanddebatesinsociety.

TheFederalPlanningBureauwillcontinuetoworkonthesupplementaryindicatorsinadditiontoGDPintheyearsaheadwithaviewtoaddingnewindicatorstotheexistingseriestoimprovetheirconvergencewithsustainablede‑velopmentindicators–withaparticularfocusonthefollow-upoftheUN’sSustainableDevelopmentGoals –andresearchingthecompositeindicatorsingreaterdepth.Tothisend,theFPBwillexplorethepossibilityofdevelopingspecificcompositeindicatorsforBelgiumforthethreemaindimensionsofsustainabledevelopment.

report 2016 - economic and financial developments · 2017-03-17 · by 25 basis points in december...

Documents