global economic outlook 4q13 - bbva research
DESCRIPTION
Global growth will accelerate in 2014, although below previous recoveries and with downward risks.. Emerging markets were resilient to the impact of the crisis, is it an indication of structural change?. The GDP is not the only relevant factor behind the equilibrium level of debt, economic structural factors and the regulation of the banking sector are also important.. http://www.bbvaresearch.com/KETD/ketd/bin/ing/publi/global/novedades/detalle/348_246241.jsp?id=tcm:348-183383-64TRANSCRIPT
Economic OutlookFourth Quarter 2013Economic Analysis
Global
• Global growth will accelerate in 2014, although below previous recoveries and with downward risks.
• Emerging markets were resilient to the impact of the crisis, is it an indication of structural change?
• The GDP is not the only relevant factor behind the equilibrium level of debt, economic structural factors and the regulation of the banking sector are also important.
Global Economic OutlookFourth Quarter 2013
Page 2www.bbvaresearch.com
Closing date: November 4, 2013
REFER TO IMPORTANT DISCLOSURES ON PAGE 31 OF THIS REPORT
Index1. Underway recovery with downward risks if the policies do not support the positive tone in the markets .....................................3
2. A slow global recovery with bearish risks.....................................................4
Box 1. Structural change in growth after the crisis .....................................................................10
Box2.Impactoffinancialturmoilongrowthinemergingmarkets ...............................................13
Box 3. BBVA Global Trade Index (BBVA-GTI) ..............................................................................15
3. An overview of the global process of leverage .......................................18
4. Tables ..............................................................................................................................28
Global Economic OutlookFourth Quarter 2013
Page 3www.bbvaresearch.com
1. Underway recovery with downward risks if the policies do not support the positive tone in the marketsThe economic assessment of the last quarter is positive, with a decline in risk aversion in thefinancialmarkets.Theconfidenceindicatorsformanufacturing,whichareareliablecyclicalsignal,havecontinuedtoimproveintheadvancedeconomiesandhavereversedthepreviousdeterioration seen in some major emerging economies like China. There has also been a decline infinancial stressand inoutflows fromemergingmarkets,bothasa resultof thechange inexpectations about the Fed swiftly ending its policy aimed at increasing liquidity. The recovery of growthinChina,theendoftherecessionintheeurozoneandthe(temporary)fiscalagreementintheU.S.completethesuccessionofeventsthathavekeptriskaversioninthefinancialmarketsat relatively low levels.
The global economic cycle continues to recover, although moderately compared with previouscycles.Theglobaleconomywillgrownearly3% in2013,somewhat less thanweexpectedpreviously,given thedownwardrevisionsofgrowth in theU.S.andMexicoand insomeeconomiesof emergingAsia.Nevertheless,wemaintainour forecastsof anupturn ingrowthin2014,nowtoaround3.5%.Thisisascenariooflimitedrecovery,supportedmorebythe advanced economies (which except for Japan should grow more strongly in 2014) and with a sustained contribution from the emerging economies in both Asia and Latin America.
The risks surrounding the global economic landscape are bearish, although the probability of menace events strongly enough to derail the ongoing recovery is lower today than three months ago.Firstly,a “disorderlyexit” fromtheFed’sQEprogram, i.e.an “excessive” increasein long-term rates in the U.S. (as a result of the market’s loss of confidence in the exit pacesoughtbytheFed,notofstrongergrowth),wouldplaceastrainonglobalfinancialconditions.Thiswouldslowdownanalreadymoderateglobalrecovery,especiallyintheeurozone,wherefinancialfragmentation,thereformspendinginthebankingsector,andthelengthanddepthofthe recession that is now coming to an end have resulted in a situation of weakness. Also related totheU.S.,anotherriskistheshort-termresolutionofthebudgetnegotiationandthedebtceiling,amatterthatwillbetakenupagaininafewweeks.Secondly,anadjustmentofgrowthinChinaandotheremergingeconomies,asaresultofeitheridiosyncraticfactorsorofthedilemmasfacedbythedomesticpoliciesoftheeconomiesmostvulnerabletotheexitfromtheFed’sQEprogram.InthecaseoftheChineseeconomy,althoughthegrowthoutlookhasnotchanged,thereisalackofguidancethatwouldeasetherisksbeyond2013:thefinancialvulnerabilityofsomebusinesssectorsthatresortintensivelytocredit,theliberalizationofthebankingsector,regionaldebtandthe opening of the services sector to refocus growth on domestic demand.
Lastly, thepossibilityofa resurgenceof theeurocrisis isaglobally relevant risk. One of themostpositiveelementsofthecurrentsituationisthemarkets’changingperceptionaboutthe European currency area due to the end of the recession and the advances on the road to banking union. The return to growth has been the consequence of the effective short-term easing offiscalpolicyandtheongoingconfidenceintheECB’sroleasaguarantorofthecurrency’sstability. The authorities need to sustain the positive markets mood with decisive progress in strengthening the monetary union, and in particular banking union (where the planned advancesfor2014arekey). All of them must be supported by measures that reinforce the sustainability of public debt in the medium term.
Inshort, intheglobalscenariothatweconsidermost likely,the improvedconfidenceshouldbereflected instrongerGDPgrowthfromthesecondhalfof2013.Nevertheless,weshouldpoint out that improvingtheconfidenceofhouseholdsandbusinessesandloweringfinancialmarkets stress depend on how the economic policies are implemented and coordinated in theU.S.,EuropeandChinainordertoreduceuncertainty,facetheirlong-termchallengesand,atthesametime,shoreuptherecovery.
Global Economic OutlookFourth Quarter 2013
Page 4www.bbvaresearch.com
2. A slow global recovery with bearish risks
The economic cycle is improving, above all in advanced economies, although it is still far from a strong recovery Twogeneralfeatureshavecharacterizedthelastquarterfortheglobaleconomicscenario.First,theconfidenceindicatorsofbusinessesandthevolatilityofthefinancialmarketshavecontinuedto reflect the lowprobability of tail risk events, those that couldbedisruptive for theglobalsituation. Thus economicrecoveryimprovesandthereislessriskofitderailing.However,some events have contributed to a scenario of, at the end, a feeble global recovery within aoneortwo-yearhorizon. They are events with a current impact (the partial closure of the US government)butalsoafutureone(thetighteningoffinancialconditionsduetoexpectedendofthe exceptional support of monetary policy).
Overall,wehavereviseddownby0.2pptheexpectedgrowthfortheglobaleconomyin2013to2.9%andin2014to3.6%.Therevisionof2013growthisduetotheworsefiguresrecordedintheU.S.andtheslowdowninsomeofthecountriesindevelopingAsia,whicharealsoaffectedbyfinancialturbulenceinthewakeofmarketsexpectationofanimminenttaperingofQEfollowingFED’s lastMayannouncement.Growth in2014hasalsobeenreviseddownto3.6%.Theemergingmarketsarebehindthisdownwardrevision,(exceptforChina,wherewesticktoourforecasts),althoughtheywillcontinuetobethebiggestcontributorstoglobalgrowth (Chart 1). The higher rate of global growth in 2014 is backed by an acceleration of the economyinallgeographicalareas,exceptforAsia,wheregrowthisexpectedtoremainatthesamelevels.ParticularlyworthnotingistheimprovementexpectedintheEurozoneaftertwoyearsofrecession,andthesignificantaccelerationinLatinAmericaaftertheblipin2013(seetables in the Appendix for more details).
Chart 1
Global growth
Chart 2
U.S.:Privatenon-agriculturalemploymentgrowth(monthly changes in thousands, 3-month moving average)
2.8
-0.4
5.2
4.03.3
2.9
3.6
-2
-1
0
1
2
3
4
5
6
2008 2009 2010 2011 2012 2013(f)
2014(f)
Emerging economiesAdvanced economiesGlobal growth
-1000
-800
-600
-400
-200
0
200
400
Jan-0
7
Nov-
07
Sep-08
Jul-09
May
-10
Mar
-11
Jan-1
2
Nov-
12
Sep-13
Source:BBVAResearchandIMF Source: BBVA Research and BLS
Global Economic OutlookFourth Quarter 2013
Page 5www.bbvaresearch.com
The tension in the financial markets caused by the announcement oftheFed’staperingeases,providingaboostnotonlyintheU.S.TheFedcausedsurprisewheninSeptemberitdecidednottostarttheprocessoftaperingin itsquantitativeeasing (QE)program.Bydelaying thestart, itunderpinned thenatureofthe program as data-dependent. It appears that thedatahavenotbeenasexpectedsincethetimeinMay2013whentheFedbegantooutlineitexitplans. The growth acceleration is still expected in thesecondhalfof theyear,butHousehold’sconsumption isweaker thanexpected,whilethereal-estatemarket,whichhadbeengainingstrength,hassufferedfromtheinitialreactiontotapering.Atthesametime,thelabormarketcontinuestobeweak(Chart2),and there is uncertainty inherent to the prolonged negotiations on the budget and the public debtceiling,whichhavetoberepeatedinafewweeks.The lack of long-term solutions and therepetitionofabrinkmanshipstrategyinfiscalnegotiationincreasetheprobabilityofaslowdownindecisionsonexpenditureandinvestment,aswellasthedirectimpactofthepartialclosureofgovernmentactivity.
Theclarificationson theprocessof tapering,which theFed’smembersarepreparing in thelightof theunexpectedreactionof themarkettotheirfirstannouncementand itsdelayuntil(possibly)thestartof2014,havereducedtherisksofaderailmentintherecovery.The initial marketresponsetothetaperingannouncementtightenedfinancialconditionsinadvancedeconomies to over-restrictive level for their cyclical moment, as well as putting a sudden brake onfinanceinsomeemergingmarkets,particularlythosewiththeweakestfundamentalsandthatareatthesametimefinanciallymostintegrated.
However,much of the rise in long-term interest rates recorded since May has been reversed (Chart3).ThemarketsdonotnowanticipaterisesinFedFundsratesuntil2015,inlinewithwhat was discounted by the market immediately before Bernanke suggested that he would initiate the tapering process (Chart 4).
Chart 3
U.S.:yieldon10-yeargovernmentdebtChart 4
U.S.:impliedFedfundrates
1.50
1.75
2.00
2.25
2.50
2.75
3.00
3.25
1-J
an-1
3
26-Jan-13
20-F
eb-1
3
17-M
ar-1
3
11-A
pr-
13
6-May-13
31-M
ay-1
3
25-J
un-1
3
20-J
ul-1
3
14-A
ug-1
3
8-Sep-13
3-O
ct-1
3
28-Oct-13
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
Oct
-13
Jan-1
4
Apr-
14
Jul-1
4
Oct
-14
Jan-1
5
Apr-
15
Jul-1
5
Oct
-15
Jan-16
Apr-16
22-Oct-2013
June 25th (one week after June FOMC meeting)
May 21st (one day before Bernake´scongressional hearing)
Source: BBVA Research Source: BBVA Research
Inaddition,volatilityandfinancialtensionshaveeasedatagloballevel,particularlyregardingemerging marketsinAsiaandLatinAmerica,whicharealsoaffectedbymajorcapitalflight.ThefirstsignsthattheFedcouldbeconsideringputtinganendtoitsmonetaryexpansionprogram(with all the reservations and steadiness adopted) led to a major depreciation of the currencies ofemergingmarkets,aswellasmajorcapitalflight(Chart5).Thesefinancialtensionscoincidedwith doubts about the performance of these economies during a slowdown that was becoming more marked.
Global Economic OutlookFourth Quarter 2013
Page 6www.bbvaresearch.com
Chart 5
Flowstoemergingeconomies(overtotalportfolioassets,June2013,%)
-4
-3
-2
-1
0
1
USA
Japan
Can
ada
UK
Sweeden
Norw
ayD
enm
ark
Finla
nd
Ger
man
yA
ust
ria
Net
her
lands
Fran
ceBel
giu
mItal
ySpain
Irel
and
Port
ugal
Gre
ece
Pola
nd
Cze
ch R
epHungary
Turk
eyR
uss
iaMexico
Bra
zil
Chile
Colo
mbia
Per
uA
rgen
tina
Chin
aIn
dia
Kore
aThai
land
Indones
iaPhili
ppin
esHongKong
Singapore
Source: BBVA Research and EPFR
Theseverityofcapitalflightand thedepreciationofcurrencies followingexpectationsofan imminent tapering raised fears of a “sudden stop” of emerging markets funding and thenasteepfall ineconomicactivity.However,thesell-offprocesshasgraduallybeenlosingintensity,andascanbeseeninChart6,wearefarfromthelevelofseverityofcapitaloutflowsobservedafter thecollapseofLehmanBrothers.At the same time,emergingmarkets showsomeindicationsofarecoveryinconfidence,afterthecheckinthemiddleoftheyear(Chart7).
Chart 6
Portfolioflowstoemergingmarkets(%assetsundermanagement, compared with pre-crisis levels)
Chart7
Emerging markets: manufacturing PMI
-20%-18%-16%-14%-12%-10%-8%-6%-4%-2%0%
1 4 7 10 13161922 252831 34 37 40 43
Current(SinceMay2013)
LehmanBrothers(sinceAug-2008)
weeks
44
46
48
50
52
54
56
58
60
Jan-1
1M
ar-1
1M
ay-1
1Ju
l-11
Sep-11
Nov-
11
Jan-1
2M
ar-1
2M
ay-1
2Ju
l-12
Sep-12
Nov-
12
Jan-1
3M
ar-1
3M
ay-1
3Ju
l-13
Sep-13
China Mexico Brazil India
Source: BBVA Research and EPFR Source:BBVAResearchandHaver
In any event, tapering will end up arriving, and change the global scenario of liquidity injections that favored indiscriminate flows to emergingmarkets. The impact of tapering,onceitiseffectivelyunderway,willprobablybeagreaterdiscriminationinflowstowardemergingmarketsaccordingtothefundamentalsofeachofthem(current-accountdeficit,foreign-currencydebtlevels,greaterorlessermaturityofshort-termdebt,etc.).Inanyevent,theextratimeallowsareductionintheriskofasuddenfallineconomicactivity,atleastintheshortterm.
Global Economic OutlookFourth Quarter 2013
Page7www.bbvaresearch.com
China once more stimulates its growth, but in a more limited fashion; its increased demand drives other economiesIn China, the doubts at the start of the year on the possibility of a sharp adjustment in its economyhavealsodissipated,at least intheshortterm.Overtheyear,theeconomyhasmaintainedahighrateofgrowth,andthemostrecentdata(thirdquarter)suggestthattheGDPispickingup(Charts8and9).Thebetterthanexpectedfiguresin2013meanthattheannualgrowthoutlookhasbeenrevisedupwardslightlyfrom7.6%to7.7%.Even so, doubts remain on the sustainability of growth in the medium and long term, as the recent upturn in growth hasbeentheresultoftheimprovementinforeigndemand,butalsoofone-offmeasuresoftaxpolicyandpublicspendingwithareneweduseofcredit.Thismeansthatfinancialvulnerabilities havebeenallowedtoincrease;theyarestillmanageable,buttheyhavetobeaddressed.Priorities include reducing excessive reliance of credit in some sectors of the economy; advancing domesticfinancialliberalizationandreducingtheroleofshadowbanking;andreformingfiscalrelations to address high local government debt. The appropriate management of these aspects must ensure a steady transition, a re-balancing of growth toward more weight of domesticdemand and household consumption.
Chart 8
GDP growth in emerging markets (q/q%,seasonallyadjusted)
Chart 9
China:indexofindustrialoutput(y/y%)
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Mar
-10
Jun-1
0
Sep-10
Dec
-10
Mar
-11
Jun-1
1
Sep-11
Dec
-11
Mar
-12
Jun-1
2
Sep-12
Dec
-12
Mar
-13
Jun-1
3
Sep-13
Mexico Brazil India China
8
9
10
11
12
13
14
15
16
Jan-1
1M
ar-1
1M
ay-1
1Ju
l-11
Sep-11
Nov-
11
Jan-1
2M
ar-1
2M
ay-1
2Ju
l-12
Sep-12
Nov-
12
Jan-1
3M
ar-1
3M
ay-1
3Ju
l-13
Sep-13
Source:BBVAResearchandHaver Source:BBVAResearchandHaver
TheperceptiononEuropeimprovesandthemostextremerisksare dissipated. The reforms geared toward better governancecontinue and growth returnsInEuropetheforecastshavebeenconfirmedandtheeconomicsituationhascontinuedtoimprove,tothepointthattheEurozoneemergedfromtherecessionwithgrowthof0.3%inthesecondquarterof2013,after6quartersofrecession.Recenteconomicdatasuggestthatthetrendwillhavebeenmaintainedinthethirdquarter,whentheEurozonewouldhaveoncemore recorded a few tenths of a percentage point of growth. The reading of the data is positive intworespects.First,astherecent upturn is based not only on foreign demand, but also on animprovementindomesticdemand.Second,theimprovementinactivityisnotonlyboostedby countries within the center of Europe (with Germany recording strong growth, but alsoFrancesurprisingrecently),buttheimprovementalsoextendstotheperiphery (with Spain and Portugalemergingfromrecession,andItalyreducingitslevelofdecline),contributing(togetherwith the progress made to improve governance and the reforms) to eliminate the systemic risks characterizingpreviousquarters.The recoveryofactivityhasbeenhelpedbya reduction infinancialtensionsintheareaandbyarelaxation(defacto)ofthemoreshort-termtargetsoffiscalconsolidation,implicitlytoleratedbytheEuropeanauthorities.
Global Economic OutlookFourth Quarter 2013
Page 8www.bbvaresearch.com
GDP growth in this part of the year formed part of our scenario, and there has been no additionalelementtomakeuschangeourexpectationsofaweakrecovery.In2013Europe’sGDPwill fallby0.4%andgrowby1.1% in2014.Theweakrecovery isconsistentwith thedeleveraging process underway in the private sector in some economies in the area and the financial fragmentation that is still inplace,which affects the capacityof bank credit supply.Monetarypolicywill continue tobe loose,offsetting inpart thecontinuednegativeeffectongrowthoffiscalpolicy.Additionally,farfrominflationarypressures,theECBhasshownitselfprepared to act if necessary,eitherwithanewroundoflong-termliquidityforthebanks,orevenwithanothercut in rates (whichcannotbe ruledout).At thesame time, thenext fewmonths will be decisive in progress toward banking union, with the entry into operation of asinglesupervisor,theECB,andthedefinitionofthemechanismsforbankresolution,themodelforimplementingwhichisstillunderdiscussion.
Chart 10
BBVAResearchFinancialStressIndicatorChart 11
Europe:GDPgrowth(q/q%)
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Jun-08
Feb-09
Oct-09
Jun-1
0
Feb-1
1
Oct
-11
Jun-1
2
Feb-1
3
Oct
-13
UnitedStates Euro Area
-1.25
-1.00
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
EM
U
Ger
man
y
Fran
ce
Ital
y
Spain
4Q12 1Q13 2Q13 3Q13(f)
Source: BBVA Research Source:BBVAResearchandHaver
ThefiscalagreementintheU.S.hasbeenanotherpatchthatdoesnot address long-term fiscal sustainability and does not avoid a contractive short-term impactOn October 16 agreement was reached between the two parties in Congress and the Senate allowingthereopeningoftheFederalGovernmentaftertwoandahalfweeksofclosure,andelevatingthedebtceiling.However,theagreementreachedisasimpleextensionofthecurrentsituation,asitonlyincludesthattheGovernmentwillhavefinanceuntilJanuary15,whilethenewdebtceilingwillbereachedonFebruary7(thoughitcouldbeextendeduntilMarch,with“ingenious”measuresbytheTreasury).Inaddition,theagreementcreatesacommitteeof29membersofCongressandtheSenatewhowillhavetoproposeaplanbeforeDecember13ona10-yearbudgethorizon.
Intensenegotiationsaredrawingnearoncutsindiscretionaryexpenditureandincreasesintaxation. The U.S. thus once more has to address an uncertain process that it has already passed throughinthesemonthsonpreviousoccasions,andthiscanonlyhavenegativeconsequences.First,thereistheperceptionthatthepoliticalconfrontationhasbeentoobitter,andthatitcouldhave an impact on the electoral outlook. This suggests that a more moderate stance that is morepronetomakingpactswhentheFebruarydeadlineapproaches.Second,theclosureoftheGovernmentwillalsohaveeconomicconsequences,andnotonlydirectones.Itistruethatthe markets have so far remained relatively immune to the imminent possibility of an event as disruptiveasadefaultontherisk-freeassetoftheworldeconomy.Itseemsself-convincedthat,
Global Economic OutlookFourth Quarter 2013
Page 9www.bbvaresearch.com
inonewayoranother,aneventof thismagnitudewouldbeavoided, througha last-minutepact,asinfacthappened.Thatisnottosaythattherehasbeennoimpactontheeconomy.Mostlikelyisthatthepartialclosureofgovernmentfor16dayshashadarelativelymarginaldirecteffectontheGDPforthequarter,perhapsafewtenthsofapp.However,thethreatofthisprocesscontinuingmayhaveanadditionalimpact.Thiseffect,moredifficulttomeasure,would be through the drain on consumption and investment in the face of increased political uncertaintyrelatedtothegovernment’scapacitytomeet itsspendingcommitments.Overall,we consider that the decision that the Fed has to take on when to start the reduction in its asset purchaseprogram(betweenDecemberandthefirstquarterof2014)willonlydependonthestateoftheeconomy.Inanyevent,the situation in which economic policies push in opposing directions will continue, with a loose monetary policy that will continue to be so for a long period,andanunnecessarilycontractivefiscalpolicyintheshortterm.Thus the U.S. public deficitwillhavefallenwithoutmarketpressure(unlikeinEurope)from6.8%in2012to4%in2013,whichcanbeconsideredadrainof1.3percentagepointsofGDPgrowthin2013.Andthelong-termchallengesforthefiscalsustainabilityoftheU.S.economyhavenotbeentackled.
Risks in the forecast: downward biased but with less probability and lower impactThe risks to the moderate recovery scenario with a growing contribution from advanced economiesandasustainedcontributionfromemergingmarketshavebeenreduced. This does not take away the fact that the balance of risks continues to be downward. It is worth pointingoutfirstduetoitscharacterthepossibilityofa“disorderlyexit”fromtheFed’sQE,whichcouldgenerateanexcessiveincreaseininterestrates(intheU.S.andinothercountries),notasaresultofimprovedgrowthprospectsorhigherinflation,butduetouncertaintyregardingthe rate of exit planned by the Fed. Financial conditions that are too tight for the rest of the world could terminate a global recovery if it is not especially dynamic, as it is particularly in theEurozone.Inaddition,itisalsoworthnotingasarisktheresolutionofthefiscalquestionsintheU.S.,thebudgetandthedebtceiling,whichhavenowbeenpostponeduntilthefirstquarterof 2014. The negotiations that the parties have to carry out until then are a potential source of uncertaintyandmayleadtoanadditionaldragifthefiscaldrainincreases.
Second,itisworthidentifyingasariskfactortheadjustment in growth in China and in other emergingmarkets.Thiscouldbe the resultof idiosyncratic factors,butalsoofdilemmas towhichdomesticpolicieshavetoaddressinamoreacuteglobalfinancialenvironment.Althoughashasbeenseenrecently,thedifferencesbetweeneconomiesarerelevant,andaninterruptionintherecoveryunderwayisnottobeexpectedunlesstherearefinancialscenariosthatareasadverse as those registered between the end of 2008 and 2009.
Lastly, theresurgenceof theeurocrisis isagloballyrelevantrisk. The authorities have to support the positive perception of the markets with decisive progress to strengthen monetary union,inparticularbankingunion.Inall,thereareanumberofelementsthatcouldleadthebetterperception to change. Some peripheral countries are in a situation of lack of political consensus that could check the necessary reforms. Negotiations on the programs of Portugal and Greece maybethesourceofuncertainty.Inaddition,shortlyworkwillbeginonthereviewofthebankbalancesheetsandtheteststresstoriskscenarios,neededfortheimplementationofasinglebankingsupervisor,theECB.Finally,ashasbeenshownbypastexperience,disagreementsonthedefinitionofpoliciesthatstrengthentheeuroarea,inthiscasebankresolutionmechanismsmayproducetensionsandvolatilityinthefinancialmarkets.
Global Economic OutlookFourth Quarter 2013
Page 10www.bbvaresearch.com
Box1.Structuralchangeingrowthafterthecrisis
The global crisis that officially broke out in September2008 with the collapse of Lehman Brothers1 and continued with various episodes in the euro zone2 until last year generatedaworldwidedeclineinGDPintheorderof-0,4percentage points in 2009, compared with an averagegrowthofaround4,0%inpreviousyears(seeChart12).
Despitebeingacrisisinadvancedeconomies,its impact varied across regions. In the year of greatest impact of thecrisis(2009),economicactivityindevelopedcountriesdeclined by -3.4% (with falls of -4.1% in the EMU and-3.1%intheU.S.),whileemergingcountriesgrewby3.0%(due mainly to growth in emerging Asia, led by China,which counteracted the slight decline in Latin America). The slowdown in emerging economies during the years of the crisis was considerably lower than in developed countries. One approach to this point is the one followed bytheIMFandrecentlypublishedbytheIIF3. It calculates thepercentageofGDPlost(orgained)duringtheyearsofthe crisis in different countries and regions comparing the currentGDPlevel(2013)relativetothelevelthatwouldhave been achieved if the growth trend observed in the years prior to the crisis continued4 (i.e., the EMU –seeChart13).Inourcase,wereproducethiscalculations,butusing our growth estimates for 20135.
Asshown inChart14,developed countries have been worse off by the crisis in terms of lost GDP, particularly inUnitedStatesandtheeurozone.Ontheotherside,emergingcountries,whichonaveragegainedongrowthduringthecrisis,speciallyinLatinAmericaandemergingAsia6 (thanks to strong economic activity in China).
Chart 12
World growth
-1
0
1
2
3
4
5
6
7
8
1986 1989 1992 1995 1998 2001 2004 2007 2010 2013(e)
SourceBBVAResearchandIMF
Chart 13
EMU:realGDPand98-05trend
80
85
90
95
100
105
110
115
2000 2002 2004 2006 2008 2010 2012
Trend98-05 Actual
SourceBBVAResearchandHaver
1:Thecrisisbuiltsince2005asaresultofthelackofconfidenceinsubprimemortgages. 2:FirstGreekbailout(April2010),sovereignandbankingriskcontagionamongtheperipheraleurozonecountries(June2011),riskofbreakdownoftheeuro(May2012). 3:‘FiveyearsaftertheCrisis’,InstituteofInternationalFinance,September2013. 4:TheIMFestablishestheaveragegrowthbetween1998and2005asthepre-crisistrend.Inthisway,thecalculationdoesnotconsider2006and2007,asitattempts to reduce the distortions of the possible effects of the pre-crisis bubble. 5:Inthefewcasesforwhichwedonothaveforecasts,weusethoseoftheIMF’sWorldEconomicOutlook(WEO)ofOctober2013. 6: The performance of emerging economies as a whole is somewhat worse than that registered in emerging Asia and Latin America due to the poor performance of emergingEurope,whichsufferthedeclineoftheeurozone.
Global Economic OutlookFourth Quarter 2013
Page 11www.bbvaresearch.com
Chart 14
GDPobservedin2013vs.pre-crisistrend
-20%
-15%
-10%
-5%
0%
5%
10%
EM
U US
Adva
nce
d
Worl
d
Japan
Em
ergin
g
Em
ergin
g A
sia
Lata
m
Source: BBVA Research
Thisfirstsimpleapproachtothetrendgrowthseemstoshow that the areas at the end of Chart 14 would have atfirst registered “structural” changes in theirgrowthcapacity, actually Latin America upward and the euro zoneandtheUSdownward.
A more refined analysis can be conducted using astatistical test of structural growth change. A structural change test determines whether changes in growth rates are sufficiently important (significant in statistical terms)to break previous dynamics. There is a wide range of structuralchangetestsintheliterature.However,becauseof its advantages and given that its characteristics adapt to thekindoftestweareconducting,wehaveselectedtheBai and Perron test7,whichwehaveruntoasampleof35 countries that include developed and emerging from different areas for the period 1980 to 2013, quarterlyfrequency.
Thetest’sresults8(seeTable2)confirmthatnone of the emerging economies in the sample has undergone a structuralchangeasaresultofthecrisis.However,thetest cannot verify the existence of a structural change in theUSandJapaninrecentyears.Ontheotherhand,theresultsaredifferentwhenweanalyzethe euro zone. In this case,thetestdoesnotrejecttheexistenceofstructuralchangebeginningin2008innearlyallmembersexceptforGermany.Moreover,it is importanttopointoutthatthe test also notices a structural change in the growth rates of the United Kingdom.
A preliminary conclusion of our findings after runningthe test is that the current crisis can be interpreted in aggregate terms for almost all countries, as the negative phase of an economic cycle that has not changed the capacity to return to the average growth rates prior the crisis. Thus, although the depth of the downturnwasmuchgreaterthanthatobservedinpreviouscycles,we should expect the crisis to have transitory effects on growth.
In contrast, when the euro zone is analyzed, the test reveals that the impact of the crisis could go beyond the current cycle and showmore permanent effects.Thisbehaviour iscoherentwith thespecificgovernancevulnerabilitiesoftheeurozoneinbankingandfiscalunion.
Table 1
PotentialGDPgrowth(%)
91-00 01-10 11-20
US 3.3 1.9 2.1
EMU 2.2 1.4 1.2
China 8.7 9.3 7.9
LatAm 2.7 3.2 3.6
Source: BBVA Research
7:J.BaiandP.Perron,2003.“Computationandanalysisofmultiplestructuralchangemodels”,JournalofAppliedEconometrics,vol.18(1),pages1-22.Insum,thetestidentifiestheperiodsinwhichachangeinthegrowthaverageenablesasignificantimprovementintheadjustmentofasimplelinearregression.Amajorfeatureofthetestisthatitprovesrobusttotheexistenceofserialcorrelationandchangesintheerror’svariance(heteroscedasticity). 8: The test is implemented by imposing a minimum period of 5 years between each structural changes (trimming). In the trade-off between sensitivity to changes and certaintyinsuchchangesasbeing“structural”orpermanent,a5yearperiodseemstobeappropriate,asitmakesitpossibletostartcapturingwhathappenedintheGreatRecessionoflate-2008andthefirsthalfof2009.
Global Economic OutlookFourth Quarter 2013
Page 12www.bbvaresearch.com
Insum,thisisinlinewiththeestimatesofpotentialGDPpublished recently by BBVA Research9. The estimates (see Table 1) suggest that all the geographical areas wouldincreasetheirpotentialgrowthduringthisdecade,particularly in the case of Latin America, which would
improve by 4 tenths (from 3.2% in the last decade to 3.6%thisdecade).Theonlyexceptionistheeurozone,which overall would lose 2 tenths of potential growth during this decade.
Table 2
StructuralchangeingrowthfollowingBai&Perron’stest(selectedcountries)
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
US
China
Japan
Euro
pe
France
Germany
Italy
Spain
UK
Lata
m
Argentina
Brazil
Chile
Colombia
Mexico
Peru
Venezuela
Em
ergin
g A
sia
India
Indonesia
Korea
Malasya
Philippines
Singapore
Thailand
Source: BBVA Research
9:Seepoint3,“EstimatingPotentialGDPatBBVAResearch”,inthethirdquarter2013GlobalOutlook,availableat: http://www.bbvaresearch.com/KETD/fbin/mult/1308_Globaloutlook_tcm348-398668.pdf?ts=21102013
Global Economic OutlookFourth Quarter 2013
Page 13www.bbvaresearch.com
Box2.Impactoffinancialturmoilongrowthinemergingmarkets
Growth in emerging economies has lost its momentum in recent months as lending conditions worsened in some ofthemandvolatilityintheirfinancialmarketsincreased.Since last June, financial tensions have increasedsignificantlycomparedwiththelevelsseenduringthelastpartof2012andthefirsthalfofthisyear,withthelatterbeing similar to those registered before the collapse of Lehman Brothers (see Chart 15).
Thisincreasedfinancialturmoilasaresultofthemarket’sview of the growing likelihood that the Federal Reserve would start to reduce its third program of purchases of financialassets in theUS,suchassovereignbondsandmortgagesecuritizations,whichjustlikethetwopreviousprograms has enabled liquidity to be injected into the market since the start of the crisis. The probability of lower liquidity on the global market increased the fear of reduced capital flows to emergingmarkets in searchofprofitability.Inpractice,theseeffectsdifferedacrossassetsand countries, since emerging Asia has been affectedthemost,mainlyIndonesiaandIndia,whiletheeffectinTurkeyandBrazilisnotnegligible.
Althoughthetensionshaveeasedrecently,thelevelsseenatthestartoftheyearhavenotbeenreachedagain,asthemarket continues to discount the reduction in monetary stimulus packages by the Federal Reserve in the mid-term.
Thus, it seems appropriate to quantify the impact ofrecent market volatility on emerging economies as a resultoftheuncertaintysurroundingtheFed’spolicy.Inaddition,weareinterestedinanalyzinghowgrowthwouldbe affected by scenarios that involve even greater volatility in the emerging markets as a result of the growing lack of confidenceabouthowtheFederalReserveisplanningtogradually reverse its current stimulus policy.
Tothisend,weproposeastructuralautoregressivevectorsmodel(SVAR),imposingsignrestrictionsontheanswersof the original model in order to identify the shock we
are interested in, i.e.ashockofvolatility inthefinancialmarket10. The model used incorporates as variables the GDP growth in the regions in which the analysis isconducted, namely Latin America and emerging Asia,whileitintroducesglobalGDPgrowthinordertocapturethe global economic cycle using this variable. As for the measurementofvolatilityweintroduceintothemodel,weproposeourfinancialstressindex11.
ThefirstexerciseproposedistotrytoquantifytheimpactofthevolatilityregisteredsincetheendofMay,intermsof lost growth in the emerging markets. To do so, wecompare the answer in terms of economic activity in both regions resulting from conditioning a series of shocks that replicate the slight increase in tensions registered since last June in a scenario where this increase in tensions wouldnothavebeenverified.12
Chart 15
BBVAResearchFinancialStressIndexinEmergingMarkets
-0.1
0.0
0.1
0.2
0.3
0.4
2006 2007 2008 2009 2010 2011 2012 2013
USTapering Uncertainty
EuropeanPeripheryCrisis
Lehman Brothers
Source: BBVA Research
10:AVARisasimultaneousequationsmodelwhereeachequationonlyincludeslaggedvaluesofthemodel’svariablesasexplanatoryvariables.Inpractice,VARmodelsdonotverifyrestrictions,andthisiswhytheirshocksarenotidentified,althoughtheycanbeidentifiedbyimposingcertaincriteria.Inthiscase,theshockidentificationstrategywehaveusedinvolvesimposingsignrestrictionsontheoriginalVARmodel’simpulse-reactionfunctionsfollowingUhlig(see:HaraldUhlig,”Whataretheeffectsofmonetarypolicyonoutput?Resultsfromanagnosticidentificationprocedure”.JournalofMonetaryEconomicsNo.52,2005,pages381-419). 11:TheFinancialStressIndex(FSI)summarizesinonesingleindicatorthedynamicsofasetofvariablesthatmeasurethevolatilityincapitalmarkets,interestratesandforeigncurrencies,creditrisk(includingsovereignrisk)andliquiditytensions.Forfurtherdetails,pleaserefertotheDecember2012QuarterlyCountryRiskReport,page23,availableon:http://www.bbvaresearch.com/KETD/fbin/mult/121200_Country_risk_Q4_2012_EN_tcm348-363546.pdf?ts=1110201312:InordertobuildthescenariowehaveassumedthatfinancialtensionswilleaseinthenexttwomonthsandonJanuary2014theywillreachtheminimumvalue,registered last June.
Global Economic OutlookFourth Quarter 2013
Page 14www.bbvaresearch.com
Chart 16
BBVAResearchFSIinemergingmarkets,alternativeshocks
-0.1
0.0
0.1
0.2
0.3
0.4
Jan-13Apr-13 Jul-13 Oct-13Jan-14Apr-14 Jul-14 Oct-14Jan-15
Shock1 Shock2 Shock3 NoShock
Source: BBVA Research
Inaddition,andinordertobuildasomehowmoreextremescenario,weassumeaseriesoffinancialtensionshockssimilar to those seen in emerging markets as a result of thecrisisintheeurozonebeginninginSeptember2011.In order to be able to compare both exercises, we willmakethemomentwhentensionsincreasedcoincide,i.e.lastJune.Asinthepreviouscase,wecomparetheanswerin terms of economic activity in relation to a scenario with noincreaseinfinancialtensions.
As seen inCharts17 and18, according to themodel,both shocks would have a reduced impact on the activity of emerging economies, with emerging Asia displayinga slightly greater effect. It is important to point out that theseresultsshouldbetakenasminimumassumptions,given the difficulty that this family of linear models hasin capturing the non-linearities present in this kind of exercise.
Weproposeathirdscenariowherethesimulatedshockis even more extreme. The aim of this new exercise is to try to establish the level of shock needed for the impact in economic activity terms to be relevant. In building this newscenariowehaveassumedthatthefinancialtensionsin emerging markets repeat the dynamics seen during the Lehman Brothers crisis13. In this case the answer producedbythemodelisconsiderable,asseeninCharts17and18.
Chart17
ImpactofFSIshocksongrowthinLatinAmerica
-1.2%
-1.0%
-0.8%
-0.6%
-0.4%
-0.2%
0.0%S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3
2013 2014 2015 2016 2017
Source: BBVA Research
Chart 18
ImpactofFSIshocksongrowthinEmergingAsia
-1.2%
-1.0%
-0.8%
-0.6%
-0.4%
-0.2%
0.0%S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3
2013 2014 2015 2016 2017
Source: BBVA Research
The conclusion from the exercise is that shocks in the financialmarketsequaltotheoneobservedrecently,orsomehowmorepronounced,shouldhavenosignificantimpact on the level of economic activity in emerging markets. It would thus be necessary to observe a shock on a scale similar to the one registered during the Lehman Brotherscrisis tonoticesignificantdeclines ineconomicactivity. However, it is unlikely that this kind of shockswill be generated in emerging economies, as it seemsnecessary for them to arise not only from an isolated deterioration in theirconditions,butalso incombinationwith a highly negative global environment as seen during the recent crisis.
13:Onceagain,inordertomakecomparabletheexercises,wesupposethattheincreaseinfinancialtensionsisverifiablesincelastJune.
Global Economic OutlookFourth Quarter 2013
Page 15www.bbvaresearch.com
Box3.BBVAGlobalTradeIndex(BBVA-GTI)14
At times of great uncertainty, having the most up-to-date information on the changes in the economy becomesparamount. The unexpected collapse of world trade in the end of 2008 and in early 2009 came as a huge economic shock. It clearly underlined the need for new tools to monitor economic developments in high frequency. In particular, to better monitor and forecastglobaltradeflows.15 Trade data face important delays in theirpublication,makingitextremelydifficulttotracktheevolution inreal time.So, for instance,preliminarymonthly data from the CPB (Netherlands Bureau for Economic Policy Analysis) on world trade of merchandise isavailablewithalagofclosetwomonths.However,asfaras growing attention is claimed by world trade of services (travel,transportation,insurance,financialservices,andsoon),itcomesasnosurprisethatwe focus our attention on world trade in goods and also services published by the Organization for Economic Co-operation and Development (OECD) (Chart 19). Such series is published onlyonaquarterlybasisandfirstestimatesarereleasedaboutonequarteraftertheendofthequarter,reflectinglate publication of national accounts breakdown in some countries.16Takingeverythingintoconsideration,thesizeand volatility of cyclical movements in recent years have once again raised the issue for the need to develop tools thatwilldefinethestateoftheeconomyinrealtime.
Chart 19
Observedworldtradeofgoodsandservicesgrowth(%,q/q)
-10
-8
-6
-4
-2
0
2
4
6
8
Mar-91
Mar-93
Mar-95
Mar-97
Mar-99
Mar
-01
Mar
-03
Mar
-05
Mar
-07
Mar-09
Mar
-11
Mar
-13
Source:OECD
Chart 20
Observedworldtradeofgoodsandservicesgrowth(%,q/q)andUSBofAMerrillLynchHighYieldSpread(%)
-3
2
7
12
17
22-10
-8
-6
-4
-2
0
2
4
6
8
Jan-97
Jan-99
Jan-0
1
Jan-0
3
Jan-0
5
Jan-0
7
Jan-09
Jan-1
1
Jan-1
3
World trade
USHighYieldSpread(rightandinvertedaxis)
Source: BBVA Research
BBVAGlobalTradeIndex(BBVA-GTI)isamonthlyindexdesignedtogaugeoveralltradeofgoodsandservices.It is based on the notion that co-movements among economicvariablesarereflectinganunderlyingcommonfactor which represents global trade cycle dynamics, a non-observed latent variable. As such, BBVA-GTI hasbeen built upon a single-index dynamic factor model framework to produce high-frequency measurement of theglobaltradeinasystematic,replicable,andstatisticallyoptimal manner from merchandise exports (CPB),global industrial production, world retail sales, global purchasingmanager’sindex(PMI)andglobalnewexportordersindex.OurextensionofAruobaandDiebold(2010)allows us to examine the information content of additional real trade data and survey indexes to produce accurate short-term forecasts of global trade growth.17
14:ThisboxsummarizesthoseresultsofourforthcomingEconomicWatch(Martínez-Martín,2013). 15:Thecollapsewascausedbythesudden,severeandgloballysynchronisedpostponementofpurchases,especiallyofdurableconsumerandinvestmentgoods(andtheir parts and components) On the world trade collapse and the debate on whether world trade was just a victim of the crisis or contributed importantly to it see for instance Baldwin (2009). 16:Worldtradeiscalculatedastheaverageofworldimportsandexportsofgoodsandservicesinvolumeof2005. 17:Seemethodologicalsectionforfurtherdetails.
Global Economic OutlookFourth Quarter 2013
Page 16www.bbvaresearch.com
According to World Trade Organization (WTO) figures,between 80 to 90 per cent of world trade relies on trade finance (trade credit and insurance/guarantees), mostlyof a short-term nature. As a result, further extensionsto examine the extent to which its accuracy might be improvedwithtradefinanceindicatorsaretobetackled.Accordingly, we have to take into consideration anindicator of the risk premium paid by risky borrowers which would capture both the global impact of credit conditionsonactivityaswellasviaglobal tradefinanceconditions. The quite high communality associated with worldtradeanditshighcountercyclicalmovement’smakethe US high yield spread19 an excellent potential proxy to be examined (Chart 2).
Evaluation and preliminary results of BBVA-GTI
The correlation of global trade growth with respect to BBVA-GTI is higher than 0.8, indicating the highpotential of the indicators used to capture global trade cycleturningpoints.20Asaresult,itallowsustoproduceshort-term forecasts of global trade growth.
Inbrief,ourpreliminaryempiricalresultsaresummarizedas follows; global trade growth backcast estimate for the second quarter has been slightly revised upwards (from 1.0%to1.3%,q/q)whileforthethirdquarterestimatefromthreemonthsagohas improvedataround1.4% (q/q) (Chart 21 and Chart 22).
Allinall,weconsiderthatour BBVA-GTI is a valid tool to be used for short-term analysis on world trade of goods andservices.Yet,wealsoconsiderthattheworkbegunhere could be further extended to examine the extent to which the single-factor dynamic factor model accuracy mightbe improvedby includingtradefinance indicatorsalong with leading indicators.
Chart 21
Worldtradeofgoodsandservicesgrowth(%,q/q)andforecasts based on BBVA-GTI
-1
0
1
2
3
4
5
6
-1
0
1
2
3
4
5
6
2Q09
3Q09
4Q09
1Q
10
2Q
10
3Q
10
4Q
10
1Q
11
2Q
11
3Q
11
4Q
11
1Q
12
2Q
12
3Q
12
4Q
12
1Q
13
2Q
13
3Q
13
ICal20% ICal40%
ICal60% Point estimates
Source:BBVAResearchandOECD
Chart 22
Worldtradeofgoodsandservicesgrowth(%,q/q)andforecasts based on BBVA-GTI
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Q3-12 Q4-12 Q1-13 Q2-13(30-Jul)
Q2-13(30-Oct)
Q3-13(30-Jul)
Q3-13(30-Oct)
Actual Estimates
Source: BBVA Research
19:BofAMerrillLynchUSHighYieldMasterIIOption-AdjustedSpread(BAMLH0A0HYM2),Percent,Monthly,NotSeasonallyAdjusted. 20:Whenusingthisindex,trenddirectionisthemostimportantelement–notnecessarilythevaluewhentheindexisabove/belowacertainfigure.
Global Economic OutlookFourth Quarter 2013
Page17www.bbvaresearch.com
Methodological description
Accordingly,BBVA-GTI comprises several high-frequency economic indicators that share a common cycle component and exhibit high statistical correlation with the global trade growth rate. In addition to the correlation criteria, those indicators should use the published dataeachquarterbeforethecorrespondingworldtradefigurebecomes available, and they must be relevant in themodel from both a theoretical and empirical point of view. Thus the evolution of each of the indicators i for the period t, z
ti can be broken down into the sum of two stochastic
unobservable components. The first component, xt,
usually called “common factor”, includes the combineddynamicsofall the indicatorsandcanbe identifiedwiththeglobaltradecycle.Thesecondcomponent,u
ti,known
as the idiosyncratic component, refers to the particulardynamics of indicator i during period t.
zti = ß
i x
t + u
ti
The movement of the common and idiosyncratic components is established by autoregressive models of order p and q.
xt = ρ
1x
t-1 + ... + ρ
px
t-p + e
t
uti = d
1i ui
t-1 + ... + d
qi ui
t-q+ ε
ti
In this case, et and ε
ti are non-observable error terms
that are assumed to be independent and not serially correlated.MarianoandMurasawa(2003)proposethatifwe consider the quarterly series as the weighted sum of its monthlyexpressions,theabove-specifiedmodelmightberepresentedinstate-spaceformandeventuallyestimated,bymaximumlikelihoodusingKalmanfiltering.
The methodology used is in line with the seminal proposal ofStockandWatson(1991),sinceweuseasmall-scalesingle-index dynamic factor model to produce an accurate index of global trade conditions in real time. As in the Stock-Watsonproposal,themodelbenefitsfromtheinformationprovided by several monthly coincident indicators. In addition,weusetheapproachproposedbyAruobaandDiebold(2010)onhowtoadjustafactormodeltohandlethe different start and finish dates of the indicators, asthey are typically available in real-time forecasting due to differingreleasetimeliness.Inshort,webelievethatsuch
an extension is extremely useful to deal with monthly and quarterly indicators,which allowus to include quarterlyestimation of global trade of goods and services as an additional coincident indicator to the constituent set of indicators.
Chart 23
BBVA-GTIandWorldTradeofgoodsandservices(%,q/q)inmonthlybasis(Updated@30Oct,2013).Shadingcorresponds with forecasts
-20
-15
-10
-5
0
5
10
-10
-8
-6
-4
-2
0
2
4
6
8
Mar-91
Mar-92
Mar-93
Mar-94
Mar-95
Mar-96
Mar-97
Mar-98
Mar-99
Mar
-00
Mar
-01
Mar
-02
Mar
-03
Mar
-04
Mar
-05
Mar-06
Mar
-07
Mar-08
Mar-09
Mar
-10
Mar
-11
Mar
-12
Mar
-13
World trade GlobalTradeIndex
Source: BBVA Research
References
Aruoba, B., and Diebold, F. 2010. Real-timemacroeconomic monitoring: Real activity, inflation, andinteractions. American Economic Review: Papers and Proceedings 100: 20-24.
Baldwin,R. (2009), “TheGreat TradeCollapse:Causes,ConsequencesandProspects”,VoxEU.org.
Mariano,R.,andMurasawa,Y.2003.Anewcoincidentindex of business cycles based on monthly and quarterly series. Journal of Applied Econometrics18:427-443.
Stock,J.,andWatson,M.1991.Aprobabilitymodelofthe coincident economic indicators. In Kajal Lahiri and Geoffrey Moore editors, Leading economic indicators,new approaches and forecasting records. Cambridge University Press, Cambridge.
Global Economic OutlookFourth Quarter 2013
Page 18www.bbvaresearch.com
3. An overview of the global process of leverage
1.Leverageandeconomicgrowth:amultidimensionalrelationshipThe increasing level of leverageinmosteconomies,particularlythemostadvanced,duringthedecade before the crisis has often been put forward as an argument from two different points ofview:asacauseofthecrisisandaconditionforrecovery.Mostanalystsacceptthatgiventhe high and unsustainable levels of debt20duringtheexpansionaryperiod,itisnecessary to start aprocessofdeleveragingtomeetmoresustainablelevels.Moreover,theneedfordeleveragingto achieve greater economic growth is related to two alternative but not exclusive arguments. Thefirstpointisthatpartofthemajorincreaseindebtinthedecadeofexpansionbeforethestart of the 2008 crisis was the result of abundant liquidity of the kind that does not appear likely torecur.Thesecondarguesthat,aboveallintheeconomieswiththehighestdebtlevels(oratleast higher than the average of their peers) and with many other structural problems (as in the caseofsomeoftheeconomiesintheEuropeanperiphery),thesesignificantstartinglevelsofdebt will oblige households and companies to allocate a major proportion of their resources to servicetheirdebt,thuschannelingthemfromexpenditureonconsumptionandinvestment.The economycouldthereforeonlyrecovertoasignificantextentoncethedeleveragingprocesswas at an advanced stage and the debt burden on households and corporates had been sufficientlyreducedtoleaveroomfornewexpenditure.
At the most aggregate level, data show that in principle there is a clear relationship between economic development (the real GDP level) and the weight of debt in the economy (for example,asmeasuredbytheratioofdebttonominalGDP).Chart24showsboththesevariablesinaggregateformfortheG7economies(theirmedianvalue).AclearrelationshipcanbeseenbetweenGDPandtheleverageratioduringthelastthreedecades,whichisnotbroken(partially)untilthemid-2000s.DebtbegantoincreasefasterthanGDPin2005.Thisprocesslastedataggregateleveluntil2011,althoughthegrowthindebtbegantomoderatein2009.After the outbreak of the crisis, the speed of debt growth gave rise to concerns about its high levels insomecountries(Chart25)anditssustainability,asithadexceededequilibriumlevels,oratleastall-timehighs.
20:Unlessotherwisespecified, theanalysis in thissectioncovers theprivatenon-financial sectorsof theeconomy; i.e. it specificallyexcludessustainabilityanalysisofthepublicdebtandthedebtofthefinancialsector,wherethedynamicsaredifferentfromtheprivatenon-financialsector.
Inaddition,theterm“debt”isusedhereinthebroadsensethatisnotrestrictedtobankcredit,inaccordancewiththestatisticsoftheBank for International Settlements-Insectionthree,dedicatedtotheU.S.andSpain,theanalysisislimitedtothecreditprovidedbythebanking system.
Finally,theanalysisisfocusedherebasicallyonadvancedeconomies,giventhattheseeconomiesarewheretheleveragingprocesshasbeenmostnotable.Thedynamicsofdebtareverydifferentinadvancedandemergingordevelopingeconomies,asaneconomy’sleveloffinancialdepthisadifferentialcharacteristicforthetwogroups.
Global Economic OutlookFourth Quarter 2013
Page 19www.bbvaresearch.com
Chart 24
G7 economies, GDP and debt to GDP ratio (median)
Chart 25
Degree of leverage: debt to GDP ratio (selected countries)*
90109128147166185204223242
8090
100110120130140150160
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Total credit to private sector, G7 median,%GDP(LHS)
GDPLevelG7,1980=100(RHS)
0
50
100
150
200
250
300
Irel
and
Den
mar
kSw
eden
Port
ugal
Spain
Bel
giu
mN
ether
lands
Norw
ay UK
Kore
aHongKong
Switzerland
Aust
ralia
Japan US
Finla
nd
Euro
zone
Aust
ria
Hungary
Fran
ceG
7 m
edia
nC
anad
aG
erm
any
Ital
yC
hin
aSouthAfrica
CzechRep.
Pola
nd
Russ
iaIn
dia
Bra
zil
Turk
eyIn
dones
iaMexico
Arg
entina
Dec-08 Lasta data
Source:OEDC,BISandBBVAResearch Some economies register a high in their leverage between the end of 2008 and the latest available data Source:OEDCandBIS
Then, a period of deleveraging appeared inevitable, forced by both supply and demand factors.Onthesupplyside,thepreviousexpansionperiodhadbeencharacterizedbyconditionsof abundant liquidity that were clearly not going to return in the new post-Lehman era. The current period is featured by a “re-regulatory” process of the financial system (adaptation tothenewBaselIIIregulationsamongotherissues),aperiodwhereprovisionoffundingmaybeless abundant or made under less favorable conditions. To this has to be added the process of bank restructuring that is taking place in some countries and the greater risk aversion of the financialinstitutions.Tothesesupplyfactorshavetobeaddedothersonthedemandside.Somecountries (for example Spain) have had to deal with a deep recession and a restructuring of their productivemodel,whichwillmake itdifficult fordemand forcredit to return tosomeof thesectorsthatweremostbuoyantduringthepreviousexpansionperiod.However,althoughinthedecade of expansion prior to the crisis most advanced economies responded in a fairly similar fashion(moreGDPwasequivalenttomoredebt),inthefivepost-crisisyearstherelationshipbetweendebtandGDPhasbeenverydifferent,soitisdifficulttoextractageneralconclusionin this respect, as can be seen in Chart 26.
Chart 26
ChangeintheleverageratioandGDPsince4Q08Chart27
DebtandGDPinSweden
Austra
Aust
Bel
CanCzechR.Den
FinFra
Ger
Hun
Ire
Ita
Jap
KorMex
Net
Nor
PolPor
Spa
Swe
Swi
Tur
UK US
EuroArg
Bra
Indi
Indo
Rus
S.Af.
-30
-20
-10
0
10
20
30
-20 0 20 40
Chan
ge
in d
ebt
to G
DP r
atio
(pp
ChangeinrealGDP,%
900
1400
1900
2400
2900
3400
100
120
140
160
180
200
220
240
260
280
Mar
-70
Mar-76
Mar-82
Mar-88
Mar-94
Mar
-00
Mar-06
Mar
-12
Mill
ions
Credit/GDP,%(LHS) GDP(RHS)
Source:OEDC,BISandBBVAResearch Source:OEDC,BISandBBVAResearch
Global Economic OutlookFourth Quarter 2013
Page 20www.bbvaresearch.com
Firstof all, for the majority of economies included in this analysis21 increases in leverage ratiosdidnotpreventGDPlevelsfrombeinghigherthanin2008(upper right-hand section ofChart26).Mostoftheseeconomiesareemergingmarkets,butadvancedeconomiessuchasBelgium,FranceandSwedenarealsoincluded.ThecaseofSwedenisparticularlyillustrativeinthisrespect(seeChart27).Itisaneconomythatsuffereditsownbankingcrisisinthe1990s(withsomefeaturesthataresharedbytheglobalcrisisof2008),andafteritthelevelofprivate-sectordebtreachedaround160%ofGDP.Theratiofellby25pointsofGDPinthedeleveragingprocess that went hand-in-hand with growth; although Sweden is also one of the countries where debt increased quickest in the pre-crisis years.
Currently Sweden is one of the countries with the highest level of private-sector debt in the world (in fourthplace inoursampleof35countries). In fact,despite themorerecentreduction, itcontinuesatabovepre-crisislevels.Evenso,in2012itsGDPgrewby1.3%(comparedwithafallof0.5%intheEurozoneasawhole),althoughin2013growthhasonlyjustremainedpositive.
Another substantial group would be made up of the economies that have been able to combine economic growth with deleveraging (bottom right-hand section of the chart). Germany (Chart 28) and the United States are good examples of this situation. Next there are economies where there hasbeenaprocessofdeleveraging,butwherepost-recessioneconomicgrowthorrecoveryhasnot kicked in. Spain is the most representative case here (Chart 29).
Chart 28
Debt and GDP in GermanyChart 29
DebtandGDPinSpain
500
800
1100
1400
1700
2000
2300
2600
50
60
70
80
90
100
110
120
130
140
Mar
-70
Mar-76
Mar-82
Mar-88
Mar-94
Mar
-00
Mar-06
Mar
-12
Mill
ions
Credit/GDP,%(LHS) GDP(RHS)
300
400
500
600
700
800
900
1000
1100
70
90
110
130
150
170
190
210
230
Mar
-70
Mar-76
Mar-82
Mar-88
Mar-94
Mar
-00
Mar-06
Mar
-12
Mill
ions
Credit/GDP,%(LHS) GDP(RHS)
Source:OEDC,BISandBBVAResearch Source:OEDC,BISandBBVAResearch
Finally there is a large group of economies that combine increases in debt with falls in economic activity(topleft-handsection).ThisgroupismadeupbasicallyofEuropeancountries,butitisveryvaried(Nordic,centralEuropeanandperipheral).Forourpurposes,ofnoteisthepresenceinthisgroupofperipheralEuropeaneconomiessuchasPortugal,IrelandandItaly,asacontrastwiththeSpanisheconomy,wherethelevelofdebthasfallensignificantly.
Thus a simple comparison of GDP and debt in the private sector is not enough to obtain a one-to-onerelationshipbetweendeleveragingandgrowth.Thisrelationshipmustthereforebe dependent on other variables not considered so far,specifictoeachcountry,andwhichcould even help identify a level of equilibrium for credit.
21: 35 economies at a different level of development.
Global Economic OutlookFourth Quarter 2013
Page 21www.bbvaresearch.com
2. Mind the gap: estimating the private credit equilibrium.Economic growth, structural and regulatory variables are relevant to determine long term equilibrium levelInthelongtermthereisapositivecorrelationbetweenfinancialdevelopment,measuredbytheprivatecredit-to-GDPratio,andthelevelofincomepercapitaofagivencountry.Sametime,therecentglobalfinancialcrisis,withverydifferentevolutionofcreditratiosandeconomicgrowth,has reminded us of the necessity of having a measure of whether the current level of credit in a country is excessive or sustainable. This assessment allow us to know whether the feasibility ofmaintainincomegrowthandcredit-to-GDPratioaroundcurrentlevelsishighorislow,givingsomecluesaboutafutureeconomicand/orfinancialcrisis.
Weproposeanempiricalpaneldatamethodology22 based on the idea that the long-run relationship betweentheprivatecredit-to-GDPratioandincomepercapitafollowsalogistic(s-shaped)typeofrelationship,withasaturationlevelatthehighestlevelsofincome.The saturation levels and the shapeoftherelationshipbetweenfinancialdeepeningandincomedependalsoonalargesetofinstitutionalandregulatorydeterminants.Additionally,theobservedlevelofthecreditratiocoulddeviate from its long-term structural level due to the effect of mid-term and short-term deviations ofincome,investment,interestratesandtheothervariablesofthemodel.This kind of deviations couldinducetheappearanceofcreditboomsorbusts,whichwecallcountry’s“creditgap”.
Alternative ways to estimate credit gapsAmong the commonly used indicators of an excessive credit growth is the change in the credit-to-GDP ratio or estimating credit gaps as the difference of the credit-to-GDP ratio from its own trend.
Analysts and international institutions who opt for the former usually consider that if such ratio grows more than 5 points in a year, the country is going through a credit boom. But, is it equally risky that the ratio grows 5 points in an emerging country with an initial ratio of 20 that if it does in an advanced economy with an initial ratio of 200? How do we compare such annual change in credit ratio if the GDP is growing at a 5% rate in the emerging economy but is only growing at a 2% rate in the advanced economy? How do we account for the effect of a recent regulatory change that favours financial development?
Estimating credit gaps as the difference of credit-to-GDP from trend have also some problems. The gap can be very different depending on the technique to estimate the long-run trend. Besides, credit gaps can also be different depending on whether the trend is linear or stochastic (normally a Hodrick-Prescott filter which is subject to end point bias).
Chart 30
Goumpertz Curve (Non Linear Approach)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
0
50
100
150
200
250
0
10,0
00
20,0
00
30,0
00
40,0
00
50,0
00
60,000
70,0
00
80,000
90,000
100,0
00
110,0
00
120,0
00
130,0
00
140,0
00
150,0
00
Ela
stic
ity
of
Cre
dit R
atio
CredittoGDPratio(%GDP)
Real GDP per capita in PPP-adj.USD
Estimated Gompertz Curve Elasticity (rhs)
Source: BBVA Research
22:Forthcoming,forfurtherdetails:AlvaroOrtizVidal-Abarca,[email protected];AlfonsoUgarteRuiz,[email protected] .
Global Economic OutlookFourth Quarter 2013
Page 22www.bbvaresearch.com
Wecansummarizethemainadvantagesofourmethodologywithrespectthealternativewaysto estimate credit gaps in the following three assumptions: i) we assume a more realistic type of relationshipbetweenthecreditratioandincomepercapita(Gompertz-curve,chart30);ii)weallowfordifferentsensitivitiestoindependentvariablesofthemodel(macroeconomic,structuralandregulatoryvariables)dependingonthetimehorizon.
Weapply thismethodology to a largepanel of83 countries between1990and2012.Wecontrol for 13 structural and regulatory variables (Table 3).
Table 3
Long Run Determinants (Financial Deepening Variables)
Macroeconomic Structural Variables Regulatory
GDPpercapita Population density CreditBureau,qualityofInformation
GDPpcxInvestmentRatio Rule of Law Private Bureau Coverage
InvestmenttoGDPRatio GiniCoefficient(inequality) Public Registry Coverage
Inflation Financial Openness StrengthofCreditors’Protection
Real Interest Rate (ST) Banking Concentration PCACostofEnforcingContracts,Registering Property
Banks Real Spread Regulatory Capital to Assets Ratio PCA Restrictions to Banking Activity
Capital Requirements PCA Restrictions to Entry in Banking sector
Source: BBVA Research
It is worth noting that we allow for different sensitivities of the credit ratio to the other variables dependingonthetime-horizoncomponent(long,mediumandshortrun).23
Through our methodology,we find clear evidence in support of a Gompertz-curve kindof relationship and in favour of different sensitivities to income per capita depending on the time-horizon that is considered. First, long-term increases of income have a positive effect on the credit ratio but with an upper-limit;second,mid-termdeviationsseemtobethemaindriversofcreditbooms;andfinally,short-termdeviationshaveanegativeimpactonthecreditratio.Thisisquiteimportantintermsofdeterminingwhatthestructuralleveloffinancialdevelopmentis,i.e.whatlevelofcreditissupportedbythefundamentalvaluesofaneconomyandtodeterminewhatlevelsofthecreditratioshouldbeconsidered“excessive”.24
The different sensitivities allow us to estimate a “structural” level of the credit ratio related to the long-term components of the independent variables and to the saturation level determinedby the institutional frameworkof each country. Then, themain results of themodel are:
i. Therearesignificantandnegativeeffectsfrominformalitybutnotsignificanteffectsfrominequality.
ii. FinancialopennessispositiverelatedwithcredittoGDP.
iii. Highercapitalrequirementshaveastrongnegativeeffectonfinancialdeepening,and
iv. Restriction on the entry of new participants and higher concentration appears to have a negativeeffectonfinancialdeepening.Ahigheravailabilityandqualityofinformationhaveaclearpositivesignonfinancialdeepening.
v. Thecombinationofhighercreditors’protection,easierenforcingofcontractsandaneasierprocess of registering property has a clear positive effect on credit.
23:Allinall,weconsideralong-termvalueofavariableequaltoits15-yearmovingaverage;mid-termdeviationisthegapbetween5-yearsmovingaverageandlong-termvalue;and,finally,short-termdeviationisthegapbetweenthevalueandthe5-yearsmovingaverage.Additionally,toleadwithmulticollinearityproblemsbetweenGDPpcandothervariablesweusePrincipalComponentsAnalysis(PCA) and residual to regressions.24:Forinstance,somerecentstudiestrytoestimatethe“benchmark”leveloffinancialdevelopmentastheleveldeterminedbyincomepercapitaandothervariables.However,ifthesensitivityofthecreditratiotochangesinincomepercapitawashigherinthemediumorshorttermthaninthelongterm,wewouldestimateamuchhigherstructuralleveliftheeconomywasgoingthroughatemporaryperiodofstronggrowth,eventhoughatemporaryincreaseinincomeshouldnotwarrantahigherlevelofcredit.Forinstance,thiswouldbethecase of an economy going through a bubble experience
Global Economic OutlookFourth Quarter 2013
Page 23www.bbvaresearch.com
The estimated “credit gap” appears to be a predictor of banking crises and,moreover, itsperformance is as good as (or even superior) to the most commonly used indicators to measure creditboomssuchastheannualchangeinthecredit-to-GDPratioandthedeviationofthecredit-to-GDPratiofromitsowntrend(eitherlinearorstochastic).Asaillustration,thefollowinggraphsshowthesizeofthecreditgaps(deviationofcredit-to-GDPratiosfromequilibrium)duringtheNordicbankingcrisis(1991),theemergingmarketcrisis(1997),thelastfinancialcrisis(2007)and the current situation (2013).
The excess credit growth during the Nordic Banking Crisis (1991) was concentrated mostly in developedcountriessuchasSweden,FinlandandDenmarkbutalsoinCanadaandJapan.TherewerealsosomesignificantcreditpressuresinBrazilandtoalesserextentinChile(Chart31).Thiscontrastswiththesituationprevioustotheemergingmarket(EMs)crisis(1997)wheremostofthesizeablecreditgaps(representedbydarkerbluecolours)wereconcentratedintheEMs.ThiswasparticularlyrelevantinAsiaandLatinAmerica,withonlyafewcountriesinemergingEurope showing excess of credit (partly due to very low starting levels after the Communism fall). Incontrast,justasmallnumberofdevelopedcountrieswereshowingexcesscreditgrowthasJapanandSweden,bothcountriesfacingalonglastingde-leveragingprocess(Chart32.
Chart 31
CreditGapMap(1991) (deviation of Credit- to-GDP from equilibrium, pp)
Chart 32
CreditGapMap(1997) (deviation of Credit- to-GDP from equilibrium, pp)
Credit Gap<6.5
Credit Gap>25
6.5<Credit Gap<10
10<Credit Gap<15 15<Credit Gap<25
Credit Gap<6.5
Credit Gap>25
6.5<Credit Gap<10
10<Credit Gap<15 15<Credit Gap<25
Source: BBVA Research Source: BBVA Research
Theprivate-creditmapof2007(previous to the latestfinancialcrisis)showsmostof thedevelopedcountriesconcentratingthecreditvulnerability.Inthiscase,theUSA,theperipheryof Europe and Australia and New Zealand concentrated most of the pressure. Emerging Europe wastheonlyregionwhereexcessprivatecreditwassomehowgeneralwhiletherestoftheEMswere isolated from the ensuing private credit slumps (Chart 33).The actual credit map (2013) shows how theUSorSpainhaveadvancedinthedeleveragingprocessandthebluecolourshave becomemore neutral. However, in some of the EU peripheral countries (still in darkblue) the de-leveraging process is still on going. One emerging signal is the increasing credit pressuresinsomeEMsinAsia,ChinaandIndia.
Global Economic OutlookFourth Quarter 2013
Page 24www.bbvaresearch.com
Chart 33
Credit Gap Map (2007) (deviation of Credit- to-GDP from equilibrium, pp)
Chart 34
Credit Gap Map (2013) (deviation of Credit- to-GDP from equilibrium, pp)
Credit Gap<6.5
Credit Gap>25
6.5<Credit Gap<10
10<Credit Gap<15 15<Credit Gap<25
Credit Gap<6.5
Credit Gap>25
6.5<Credit Gap<10
10<Credit Gap<15 15<Credit Gap<25
Fuente:BBVAResearchyBancoMundial Fuente:BBVAResearchyBancoMundial
3.U.S.andSpain.Differentratesof(de)leveraginganddifferentrelevant factorsAswesawintheearliersectionsofthisarticle,thesituationinwhichtheeconomiesaredealingwithdeleveragingprocessesareverydifferent,bothintermsofthecharacteristicsoftheprocessand the economic environment in which it takes place. To illustrate this we have taken the cases oftheU.S.andSpain,whichareanalyzedinmoredetailinthissection.Tomakecomparisoneasier, andgiven that it ismucheasier for companies in theU.S. toobtain finance throughcorporatedebt issues, the analysis is restricted tobank lending to theprivate sector,whichcoverspracticallyallcorporateandhouseholdfinanceinSpain.
Ascanbeseeninchart35,thelevelofcreditwassimilarinthetwocountriesinaround2000,atabout90%ofGDP.However,inthesubsequentdecadetherewasamajorincreaseintheSpanishfigurestoahighof175%,comparedwithamoremoderatehighof130%intheU.S.ThedeleveragingprocessstartedlaterinSpain,andtodayisclosetothelevelof2006,whilethe U.S. began to reduce credit earlier and has now reached more moderate levels of around 100%ofGDP.
Chart 35
Credittonon-financialcompaniesandhouseholds(%GDP)
40%
60%
80%
100%
120%
140%
160%
180%
Jun-80
Dec-81
Jun-83
Dec-84
Jun-86
Dec-87
Jun-89
Dec-90
Jun-92
Dec-93
Jun-95
Dec-96
Jun-98
Dec-99
Jun-0
1
Dec
-02
Jun-0
4
Dec
-05
Jun-0
7
Dec-08
Jun-1
0
Dec
-11
Jun-1
3
Spain US
Source:BBVAResearch,basedonECB,FDICandEurostatdata
Global Economic OutlookFourth Quarter 2013
Page 25www.bbvaresearch.com
Tobetterunderstandhowlendinghasevolvedinthetwocountries,itisessentialtoanalyzethelevelofdebtofthedifferenteconomicagents,householdsandcompanies.
Realestatelending:thesubprimeepisodeintheU.S.andtheboominSpainIn the U.S. the itemmost responsible for the growth in bank lending during the boomwasmortgagelending,andspecificallytheso-called“subprime”loans. These were mortgage loanswithgenerallyveryeasyconditionsfortheinitialquartersoryears,issuedbymortgagebrokers (banking institutions) to low-income families. These loans were then sold on to traditional banks,andrepackaged intoasset-backedsecurities thatweresold ingoodconditionsontheinternationalfinancialmarketsthankstofavorablecreditratings.Intheend,itwasdifficultforthefinalholdertovaluetheriskassumed.ItcouldthereforebesaidthatintheU.S.thefactorbehindtheincreaseinmortgagelendingwastheregulationofthefinancialsystem,whichgeneratedincentivestotakerisksthatintheendresultedexcessive.Currently new lending forhomeshasstillnotrecovered,mainlyduetoweaksupply.Publicpolicies,frequentlyfocusedonprotectingconsumers,havepartiallydelayed theadjustment, sobanks still holdproblemloans on their balance sheets and they will have to continue their deleveraging process. Recently somesymptomsofimprovementintherealestatemarkethavebeennoted,butnotinhomepurchases.
InSpaintheproblemswerealsoderivedfromthereal-estateboom,whichledtoamajorincreasein lending to constructors and developers (41% of GDP in 2008) and residential mortgagelending(63%ofGDPinthesamedate).However,inthiscasethefactorsbehindthislendingbehavior were not only linked to regulation; tax incentives also rewarded home ownership,and increasing competition in the financial system led togrowth in themarket shareof theentitiesthatweremostexposedtotherealestate,thesavingsbanks,followingtheeliminationofrestrictionstotheiractivity.Lastbutnotleast,anotherimportantfactoristhefallininterestratesresulting from the entry into the euro and pro-cyclical lending standards.
Chart 36
U.S.:Outstandingloansbydebtor(%GDP)Chart37
Spain:Outstandingloansbydebtor(%GDP)
0%
20%
40%
60%
80%
100%
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
Consumer credit to GDPMortgage to GDPCorporate credit to GDP
0%
20%
40%
60%
80%
100%
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
Consumer credit to GDPMortgage to GDPCorporate credit to GDP
Source:BBVAResearchbasedonFDICdata Source:BBVAResearch,basedonECBandEurostatdata
The other major part of households credit, consumer lending, increased moderately during thecrisisandhasalreadystartedtheroadtorecoveryinbothcountries. The shorter terms and lower amounts explain why it is an item with less inertia.
No particularly notable behavior has been observed on the corporate lending side in the U.S. during the crisis, as major firms tend to use the corporate debt markets for finance, whichmakes thesectorasawhole lessdependentonfinancial institutions. Ingeneral, thebiggestcompanieshavealreadycompleted theirdeleveragingprocess,while thesmallandmedium-
Global Economic OutlookFourth Quarter 2013
Page 26www.bbvaresearch.com
sizedenterprises(SMEs)haveonlydonesopartly.Theeconomicandregulatoryuncertaintyismakingitmoredifficultforbankstoincreasetheirflowofnewlendingtothesector.
InthecaseofSpain,mostofthedifficulties lie inthe loansaccumulatedbyconstructionandrealestatecompanies.Eventoday,aftersixyearsofcrisis,withthenewregulationsincreasingthespecificprovisionsforthesectorandthetransferofloansfromlesssolventbankstoareal-estateassetcompanycreatedbythegovernment(Sareb),companiesintherealestatesectorhaveastockofcreditthataccountsfor23%ofGDP,ahighfigureconsideringtheirsituationandoutlook in terms of their contribution to the Spanish economy.
In all, for thebusiness sector as awhole thenumberof newcredit operations continues todecline.
Deleveraging: different speed, different causes Part of the explanation why the reduction in the outstanding balance of credit began before andhasbeenswifter intheU.S.than inSpain lies intheaccountingrules,asthe procedure fortransitionfromdelinquencytodefault isquickerintheU.S. Inaddition, insomeoftheAmerican states mortgages are granted without universal liability (or the judge tends to decree this a posteriori). This means that customers with negative equity (a volume of debt higher than thevalueofthedwelling)mayhandovertheirhometothebankandthuscanceltheirdebt,withoutthebankhavinganyrightsontherestofthedebtor’sassetsorincome.Asaresult,theoutstanding mortgage balance is reduced quicker.
Themanagementofthefinancialcrisishasalsobeendifferentinthetwocases. From the timeoftheinitialturbulence,theU.S.authoritiesreactedbyusingmonetary,fiscalandspecificmeasuresonthefinancialsystem,whichspedupthestabilizationoftheeconomy.Inparticular,thefirststresstestonthemajorbankswascarriedoutinMay2009,whiletheSpanishexercisewith international supervision that managed to convince the markets did not take place until September2012. Inaddition,banks that receivedaid in theU.S.wereadequatelymanagedand they couldbenefit from the economic recovery,whichbeganquickly.The institutional difficulties inEuropeand lackofadequatediagnosisof thecrisis in its initialphaseshasdelayedandmademoredifficulttherestructuringoftheSpanishfinancialsystem.
WhatdoesthefutureholdforSpain?Forecastssuggestthatthedeleveragingwillcontinueforthenextfewquarters.Thismustbevaluedpositively,giventhatthecurrentlevelofroughly130%ofGDPisstillverymuchhigherthantheaverageintheEurozone(90%)andnotsurprisinglyalsohigherthanitslong-termequilibriumlevel.Inaddition,credithastoberebalancedbysectors,forexample,reducingtheshareoftherealestatesector.
The challenge is to make this deleveraging compatible with the provision of new lending tosolventdemand.Specifically,thesectorslinkedtoexportsarethosethathavethebiggestpotentialatthepresenttime,giventhatdemandinthisareais improving,justasinpreviouscyclical recoveries. Overall, in order to boost the flow of new lending operations again it isimportanttodiagnosewhetherthedifficultiesonthesupplysidearecausedbylackofliquidity(as occurred during other phases of the crisis) or the risk of applicants (as is the case now). To easetherisks,measuressuchasguarantees,risksharingorthereductionofcapitalconsumptioncould be adequate.
Global Economic OutlookFourth Quarter 2013
Page27www.bbvaresearch.com
4.ConclusionThroughout this section we have looked at the changes in debt and growth, showing howpreliminary analyses that suggest a direct relationship between the two variables could be simplistic. Different economies are registering (de)leveraging and (negative) growth in their four possible combinations, as in fact is observed now when reviewing the behavior of both variables in 35 economies. This is because of the complexrelationshipbetweenleverageandgrowth, which does not involve the two alone, but that also depends on more structural variables suchaschanges inthepopulation, informality, investmentandthestructureofthefinancialsystem;aswellas purely regulatory variables, such as requirements relating to capital adequacy,thequalityofthefinancialreportingorcreditbureausandthelevelofprotectiontoconsumersoffinancialservices.
One example of the importance of the economic environment and the rules of the game as they stand can be found in the performance of bank credit in two economies such as the U.S. andSpain,bothinthegroupofadvancedeconomies.Adeleveragingprocessofbankcreditisunderwayinboth,butwhereastheU.S.isgrowingSpainisbarelyemergingfromadeepandlastingrecession.Overall, if the fall in the stock of credit is not to delay economic recovery, it has to take place while new credit operations pick up at the rhythm of solvent credit demand, withoutbeingheldbackbysupplyfactors.
Global Economic OutlookFourth Quarter 2013
Page 28www.bbvaresearch.com
4. TablesTable 4
Macroeconomic Forecasts: Gross Domestic Product
(YoY growth rate) 2010 2011 2012 2013 2014
UnitedStates 2.5 1.8 2.8 1.6 2.3
Eurozone 1.9 1.6 -0.6 -0.4 1.1
Germany 3.9 3.4 0.9 0.6 1.8
France 1.6 2.0 0.0 0.2 1.2
Italy 1.7 0.6 -2.6 -1.9 0.7
Spain -0.3 0.4 -1.6 -1.3 0.9
UK 1.7 1.1 0.1 1.4 2.3
Latin America * 6.0 4.0 2.5 2.4 3.1
Mexico 5.1 4.0 3.6 1.2 3.1
Brazil 7.5 2.7 0.9 2.6 2.8
EAGLES** 8.4 6.6 5.0 4.8 5.2
Turkey 9.2 8.5 2.2 3.7 3.6
AsiaPacific 8.2 6.0 5.3 5.2 5.3
Japan 4.7 -0.6 2.0 1.9 1.5
China 10.4 9.3 7.7 7.7 7.6
Asia(exc.China) 6.8 3.6 3.7 3.4 3.6
World 5.1 4.0 3.3 2.9 3.6
*Argentina,Brazil,Chile,Colombia,Mexico,Peru,Venezuela. **Brazil,China,India,Indonesia,Korea,Mexico,Russia,Taiwan,Turkey. Forecastclosingdate:November4,2013. Source: BBVA Research
Table 5
MacroeconomicForecasts:Inflation(Avg.)
(YoY growth rate) 2010 2011 2012 2013 2014
UnitedStates 1.6 3.1 2.1 1.7 2.2
Eurozone 1.6 2.7 2.5 1.5 1.4
Germany 1.2 2.5 2.1 1.7 1.8
France 1.7 2.3 2.2 1.1 1.3
Italy 1.6 2.9 3.3 1.4 1.5
Spain 1.8 3.2 2.4 1.5 1.1
UK 3.3 4.5 2.8 2.7 2.2
Latin America * 7.6 8.0 7.5 8.8 9.4
Mexico 4.2 3.4 4.1 3.8 3.4
Brazil 5.0 6.6 5.4 6.2 5.9
EAGLES** 5.3 6.0 4.2 4.2 4.3
Turkey 8.6 6.2 8.9 7.5 6.4
AsiaPacific 3.7 4.9 3.1 2.9 3.6
Japan -0.7 -0.3 0.0 0.1 2.0
China 3.3 5.4 2.6 2.8 3.5
Asia(exc.China) 3.9 4.5 3.4 3.0 3.6
World 3.7 5.1 4.1 3.8 4.0
*Argentina,Brazil,Chile,Colombia,Mexico,Peru,Venezuela. **Brazil,China,India,Indonesia,Korea,Mexico,Russia,Taiwan,Turkey. Forecastclosingdate:November4,2013. Source: BBVA Research
Global Economic OutlookFourth Quarter 2013
Page 29www.bbvaresearch.com
Table 6
MacroeconomicForecasts:CurrentAccount(%GDP)
2010 2011 2012 2013 2014
UnitedStates -3.1 -3.0 -3.0 -2.5 -2.8
Eurozone 0.0 0.1 1.2 2.1 2.1
Germany 6.0 5.7 7.0 6.5 5.6
France -1.6 -2.0 -2.3 -1.6 -1.7
Italy -3.5 -3.1 -0.7 0.1 0.2
Spain -4.5 -3.8 -1.1 1.2 1.8
UK -3.9 -1.6 -4.6 -3.0 -2.8
Latin America * -0.7 -0.9 -1.6 -2.3 -2.1
Mexico -0.2 -0.9 -1.2 -1.3 -1.4
Brazil -2.2 -2.1 -2.4 -3.5 -3.1
EAGLES** 1.6 0.5 0.3 0.1 0.4
Turkey -6.4 -9.9 -5.9 -6.8 -6.6
AsiaPacific 3.3 1.5 0.9 1.0 1.3
Japan 3.7 2.0 1.0 1.2 1.7
China 4.0 1.9 2.3 2.4 2.8
Asia(exc.China) 2.0 1.3 0.0 0.1 0.4
*Argentina,Brazil,Chile,Colombia,Mexico,Peru,Venezuela. **Brazil,China,India,Indonesia,Korea,Mexico,Russia,Taiwan,Turkey. Forecastclosingdate:November4,2013. Source: BBVA Research
Table7
MacroeconomicForecasts:GovernmentDeficit(%GDP)
2010 2011 2012 2013 2014
UnitedStates -8.9 -8.7 -6.8 -4.0 -3.4
EMU -6.2 -4.1 -3.7 -2.8 -2.4
Germany -4.1 -0.8 0.2 0.0 0.0
France -7.1 -5.3 -4.8 -4.1 -3.6
Italy -4.3 -3.8 -2.8 -3.0 -2.5
Spain* -9.6 -9.1 -6.8 -6.8 -5.8
UK ** -10.2 -7.8 -6.3 -6.0 -5.9
Latin America *** -2.5 -2.3 -2.5 -2.6 -2.6
Mexico -3.4 -2.6 -3.1 -2.4 -2.3
Brasil -2.5 -2.6 -2.5 -3.2 -3.7
EAGLES**** -2.5 -1.9 -2.3 -2.2 -2.0
Turkey -3.6 -1.4 -2.1 -1.2 -2.1
AsiaPacific -3.6 -3.7 -3.7 -3.7 -3.1
Japan -9.5 -10.0 -9.5 -10.0 -8.0
China -2.5 -1.1 -2.1 -2.0 -1.8
Asia(exc.China) -4.5 -5.4 -4.8 -4.8 -3.9
*Excludingaidtofinancialsector. **Fiscalyearfrom1Aprilto31March. ***Argentina,Brazil,Chile,Colombia,Mexico,Peru,Venezuela. ****Brazil,China,India,Indonesia,Korea,Mexico,Russia,Taiwan,Turkey. Forecastclosingdate:November4,2013. Source: BBVA Research
Global Economic OutlookFourth Quarter 2013
Page 30www.bbvaresearch.com
Table 8
MacroeconomicForecasts:10-yearInterestRates(Avg.)
2010 2011 2012 2013 2014
UnitedStates 3.2 2.8 1.8 2.3 3.2
Eurozone 2.8 2.6 1.6 1.6 2.1
Forecastclosingdate:November4,2013 .Source: BBVA Research
Table 9
MacroeconomicForecasts:ExchangeRates(Avg.)
US Dollar per national currency 2010 2011 2012 2013 2014
UnitedStates(EURperUSD) 0.76 0.72 0.78 0.75 0.77
Eurozone 1.33 1.39 1.29 1.33 1.31
UK 0.65 0.62 0.63 0.64 0.64
Japan(JPYperUSD) 87.8 79.7 79.8 97.3 109.9
China(RMBperUSD) 6.77 6.46 6.31 6.19 6.02
Forecastclosingdate:November4,2013. Source: BBVA Research
Table 10
MacroeconomicForecasts:OfficialInterestRates(Endperiod)
2010 2011 2012 2013 2014
UnitedStates 0.25 0.25 0.25 0.25 0.25
Eurozone 1.00 1.00 0.75 0.50 0.50
China 5.81 6.56 5.75 6.00 6.00
Forecastclosingdate:November4,2013. Source: BBVA Research
Global Economic OutlookFourth Quarter 2013
DISCLAIMER
ThisdocumenthasbeenpreparedbyBBVAResearchDepartment,itisprovidedforinformationpurposesonlyandexpressesdata,opinionsorestimationsregarding the date of issue of the report, prepared byBBVAor obtained fromor based on sourceswe consider to be reliable, and have not beenindependentlyverifiedbyBBVA.Therefore,BBVAoffersnowarranty,eitherexpressorimplicit,regardingitsaccuracy,integrityorcorrectness.
Estimations this document may contain have been undertaken according to generally accepted methodologies and should be considered as forecasts or projections.Resultsobtainedinthepast,eitherpositiveornegative,arenoguaranteeoffutureperformance.
Thisdocumentanditscontentsaresubjecttochangeswithoutpriornoticedependingonvariablessuchastheeconomiccontextormarketfluctuations.BBVA is not responsible for updating these contents or for giving notice of such changes.
BBVAacceptsnoliabilityforanyloss,directorindirect,thatmayresultfromtheuseofthisdocumentoritscontents.
Thisdocumentand itscontentsdonotconstituteanoffer, invitationorsolicitation topurchase,divestorenter intoany interest infinancialassetsorinstruments.Neithershallthisdocumentnoritscontentsformthebasisofanycontract,commitmentordecisionofanykind.
Inregardtoinvestmentinfinancialassetsrelatedtoeconomicvariablesthisdocumentmaycover,readersshouldbeawarethatundernocircumstancesshould they base their investment decisions in the information contained in this document. Those persons or entities offering investment products to these potential investors are legally required to provide the information needed for them to take an appropriate investment decision.
Thecontentofthisdocumentisprotectedbyintellectualpropertylaws.Itisforbiddenitsreproduction,transformation,distribution,publiccommunication,makingavailable,extraction,reuse,forwardingoruseofanynaturebyanymeansorprocess,exceptincaseswhereitislegallypermittedorexpresslyauthorizedbyBBVA.
ThisreporthasbeenproducedbytheEconomicScenariosUnit:
Chief Econmist for Economic ScenariosJulián Cubero [email protected]
Rodrigo [email protected]
BBVA Research
Group Chief EconomistJorgeSicilia
Emerging Markets:AliciaGarcí[email protected]
Cross-CountryEmergingMarketsAnalysis Álvaro Ortiz Vidal-Abarca [email protected]
Asia StephenSchwartz [email protected]
Mexico CarlosSerrano [email protected]
Latam Coordination Juan Ruiz [email protected]
Argentina GloriaSorensen [email protected]
Chile JorgeSelaive [email protected]
Colombia Juana Téllez [email protected]
Peru HugoPerea [email protected]
Venezuela Oswaldo López [email protected]
Developed Economies: Rafael Domé[email protected]
Spain Miguel Cardoso [email protected]
Europe Miguel Jiménez [email protected]
United States Nathaniel Karp [email protected]
Global Areas:
Economic Scenarios Julián Cubero [email protected]
Financial Scenarios SonsolesCastillo [email protected]
Innovation & Processes Clara Barrabés [email protected]
Financial Systems & Regulation:SantiagoFerná[email protected]
Financial Systems Ana Rubio [email protected]
Financial Inclusion David Tuesta [email protected]
Regulation and Public Policy María Abascal [email protected]
Recovery and Resolution Policy José Carlos Pardo [email protected]
Global Regulatory Coordination Matías Viola [email protected]
Contact details:BBVA ResearchPaseoCastellana,81–7thfloor28046Madrid(Spain)Tel.:+34913746000and+34915377000Fax:+34913743025bbvaresearch@bbva.comwww.bbvaresearch.comLegalDeposit:M-31256-2000
Alberto González de [email protected]
With the contribution of:
SpainMiguel [email protected]
Jaime Martínez-Martí[email protected]
Jorge Rodríguez-Vá[email protected]
Financial SystemsAna [email protected]
Jaime [email protected]
Cross-CountryEmergingMarketsAnalysisÁlvaro [email protected]
Alfonso [email protected]