poverty and inequality

1

PPoovveerrttyy aanndd iinneeqquuaalliittyyCentral and Eastern Europe and the Commonwealth ofIndependent States are viewed by much of the develop-ment community as countries without serious povertyproblems. Such views are reinforced by the relativelyhigh levels of human development that some of thesecountries experienced under socialism, the strong eco-nomic recoveries that have taken hold in some coun-tries since the late-1990s, and the May 2004 accession ofeight of these countries to the European Union (EU).Poverty and inequality are often seen as residual prob-lems, to be addressed by European integration andmore effective social policies.

Such views are not without basis. Development levels inthe new EU member states and candidate countries arewell above most of Africa and the Caribbean. Europeanintegration processes do confer important advantageson much of the post-communist world. However, as sev-eral of the articles in this issue demonstrate, the issues ofpoverty and inequality are pressing concerns — even inEU members or candidates like Poland [page 5], Turkey[page 10], or Serbia [page 7], where social inclusion chal-lenges are closely linked to high unemployment ratesand regional, gender, and ethnic disparities. In countrieslike Tajikistan [page 19] — where per-capita GDP levelsare below those of Sudan — and Uzbekistan [page 18],efforts to align the Millennium Development Goals

(MDGs) with poverty reduction strategies, and to assessthe capital and financial needs of meeting the MDGs, areat the heart of development debates. The region alsodisplays a variety of social policy models, ranging fromthe introduction of flat taxes (in Russia, Estonia, Slovakia,and elsewhere [page 2]) on the one hand to the consol-idation of Soviet-era social policy frameworks (in Belarus[page 12]) on the other.

The variation of needs across the region means that pol-icy makers face difficulties in combining broad access tosocial services with prices that cover the costs of provid-ing these services [page 3]. The credibility and accuracyof statistical methods associated with measuring pover-ty and inequality in transition economies pose addition-al challenges [page 18]. Moreover, the heterogeneity ofthe region has not prevented the emergence of somecommon — and much lamentable — features of pover-ty across the region, particularly relatively high povertyrates among children [page 21], and residents of ruralareas [page 14].

Poverty is a salient issue for the region. Under closerexamination, it is clear that poverty reduction strategiesand the MDGs have a major role to play in the develop-ment of Central and Eastern Europe and the Common-wealth of Independent States.

Ben Slay and James Hughes

NOVEMBER 2005 | issue 2

DECEMBER 2006

Published by the United Nations Development Programme and the London School of Economics and Political Science

www.developmentandtransition.net

A l b a n i a A r m e n i a A z e r b a i j a n B e l a r u s B o s n i a a n d H e r z e g o v i n a B u l g a r i a C r o a t i a C y p r u s C z e c h R e p u b l i c F Y R M a c e d o n i a G e o r g i a H u n g a r y K a z a k h s t a n K o s o v o ( S e r b i a ) Ky r g y z s t a nL a t v i a L i t h u a n i a M a l t a M o l d o v a M o n t e n e g r o P o l a n d R o m a n i a R u s s i a n F e d e r a t i o n S e r b i a S l o v a k i a S l o v e n i a T a j i k i s t a n T u r k e y T u r k m e n i s t a n U k r a i n e U z b e k i s t a n

5DEVELOPMENTTRANSITION&&

Interview: Mart Laar on flat taxes Waltraud Schelkle 2

Utility tariffs and the political economy of social policy Deborah Mabbett 3

Poland’s very difficult labour market Janusz Czamarski, Ben Slay 5

Serbian regional inequalities in access to jobs Mihail Arandarenko 7

Inequality in poor households in Turkey: Does only gender matter? Leyla Sen 10

Poverty reduction In Belarus: Experience and lessons Lyudmila Istomina 12

Central Asia: Spatial disparities in poverty Jacek Cukrowski 14

Poverty and inequality in Uzbekistan Uktam Abdurakhmanov, Sheila Marnie 16

Poverty and inequality in Armenia Aghassi Mkrtchyan 18

Fighting poverty in post-Soviet Tajikistan Cheickna Diawara 19

Putting the focus back on children during a time of economic growth Gordon Alexander, Petra Hoelscher 21

Microfinance Jonathan Brooks 24

http://www.developmentandtransition.net

2

DEVELOPMENT &TRANSITION

FFllaatt rraattee ttaaxx rreeffoorrmmss ffoorr ttrraann--ssiittiioonn aanndd ddeevveellooppiinngg ccoouunn--ttrriieess

Waltraud Schelkle interviews Mart Laar

What especially commends flat rate taxes to transition anddeveloping economies?

Flat taxes work so well because they are simple, fair andsupport the motivation of people to work more and toplan for their own future. The issue of motivation wasparticularly important after the collapse of Soviet Unionnations in Central and Eastern Europe.

In December 1993, the ruling coalition in Estoniapushed through parliament a radical reform of personalincome taxes, which became a flat 26 per cent level,against massive opposition from the ‘economic experts’.The effects were very positive. Budget revenues did notfall but instead increased significantly, personal incometax revenues doubled within two years. Black markets inEstonia disappeared and tax compliance improved. Theintroduction of the flat rate, proportional income taxhelped to boost economic activity and create new jobs,helping Estonia to avoid massive unemployment.

The success of the Estonian tax reform encouragedEstonian neighbours to follow its example. In 1995,Latvia introduced a flat rate personal income tax and in1996, Lithuania followed with a flat rate tax of 32 percent. Comparing the unweighted GDP growth rates ofeconomies in Central Eastern Europe with flat rate per-sonal income taxation to growth rates of economies inthe same region with progressive income taxation, wecan see that countries with flat rates grew significantlyfaster.

The success of the Baltic tax reforms encouraged Russia,Slovakia and Romania to follow suit. The most radical taxreform was undertaken on 1 January 2004 in Slovakiawhere value added tax (VAT), the corporate income taxand the personal income tax were all fixed at a flat rateof 19 per cent.

So flat rate taxes have improved growth and creatednew jobs. This form of taxation has been very effective infighting the black economy and increasing tax obedi-

ence. It is easy to pay but hard to avoid. Therefore, it isvery good for the budget.

Do you consider flat rate tax systems to be the first-best fis-cal institution at all stages of development or primarily forthe catching-up process?

A flat tax system was first proposed by Milton Friedmanand Alan Rabuschka for the ‘old’ capitalist countries tooverhaul their traditional tax structures. They proposedit because it supports individual freedom, promoteseconomic activity, thus more jobs and growth. They alsoconsidered it to be fairer because flat rate taxes reducebureaucracy and tax avoidance, although these latteraspects were not emphasized then. In the process oftransition, these latter aspects of a flat rate tax systembecame much more important. Over time, the classicadvantages of a flat tax become more important by fos-tering growth and activity.

So I think that the flat taxes will work well at all stages ofdevelopment, because they are fairer and support eco-nomic activity.

Lump sum taxes (the same absolute amount instead of thesame rate) would be economically more efficient —whynot go for that?

Such taxes were often used in the ancient and medievalworld and there are some scholars who are advocatingthem even today. In practice, such lump sum taxes haveserious drawbacks. First of all, they tax poor people pro-portionally more. That is not a good way of collectingdirect taxes from the poorest people: their tax share inrevenue is small but the costs to collect them are high.This is the reason why in a flat rate tax system the poor-er section of households is exempt from paying taxes,thanks to an income threshold that is too high for them.

Secondly, even in the most stable countries there is atleast some inflation. Therefore, governments will haveto change the lump sum quite often which is not practi-cal. I think the flat rate tax is a better solution.

Progressive taxes are seen as being fairer to low incomehouseholds —why should governments of transition anddeveloping countries ignore that?

Progressive taxes ‘are seen’ as being fairer to lowincome households, but when you study what actuallyhappens to these households under a progressive taxsystem, you find that it is not fairer. Flat rate tax systemshave two rates – zero per cent for low income familiesand one fixed rate for all others – with exemptions pri-marily for the number of children. In reality, mostexemptions in existing progressive systems are not somuch helping low income families but middle income

Average GDP Growth

YEARFLAT RATE COUNTRIES :Estonia, Latvia, Lithuania

PROGRESSIVE RATE COUNTRIES:Czech Rep. Hungary, Poland, Slovakia

1996 6.8 per cent 4.75 per cent



3

DECEMBER 2006 | issue 5

and high income families who actually do not needsocial support. Progressive systems make taxation socomplicated that in many countries you need a special-ist to fill in your tax declarations – it is quite clearly aneconomic burden. And at the same time it providesincentives for avoiding taxes.

Some argue that flat rate income taxes lead to moreinequality of disposable income. The evidence fromCentral and Eastern Europe (CEE) proves the opposite: inthe flat rate countries usually the inequality as measuredby the GINI coefficient is not increasing but decreasing.

Low proportional tax rates weaken the automatic stabiliz-ers (revenues varying in line with the business cycle, reduc-ing disposable income in a boom and increasing it in arecession) of the government budget – can transition anddeveloping countries afford that?

I don’t think so. First of all it is clearly seen that transitionand developing countries can afford low proportionaltaxes – their revenues will be higher and financeshealthier thanks to this tax structure.

As I said before, if we compare budgets of the flat ratecountries with the budgets of non-flat rate countries inthe same region, we find that the former are financiallyhealthier. Of course, this is not due only to the flat ratesystem, but flat rates help. In theory, the picture may bedifferent for so-called developed societies, where theautomatic stabilizers can play a more important role. Buteven then I do not think that one should solve othereconomic problems such as stability with differenttaxes. Governments should only care about keeping thebudget balanced. When they fail in this regard, auto-matic stabilizers will not help.

How do you respond to allegations that flat rate taxreforms introduce a form of tax competition that is notcompatible with the solidarity principle of the EuropeanUnion?

I do not think that the solidarity principle has anythingto do with tax competition. This would only be the caseif countries lowered taxes to a level that would jeopard-ize their financial stability. But CEE countries with lowflat rate taxes have in fact collected more revenue. Theyare developing faster and are thus moving to a levelwhere they will become net payers to the EU budget.

In this context, the example of Ireland is telling: by usinglower taxes, one of the poorest countries in Europebecame one of the richest. Solidarity in Europe means tomove new member states as quickly as possible to thesame level of development as ‘old Europe’.

So, on the contrary, I am convinced that tax competition– which will push down taxes everywhere in Europe –will make Europe more competitive. For the new mem-ber states it is not only important to rejuvenate theircountries, we want to rejuvenate entire Europe.

Mart Laar was Prime Minister of Estonia in 1992-1994and 1999-2002. He is widely credited with leading Esto-nia through a period of rapid economic reforms thatwon Western praise and led to rapid economic growthand entry to the EU.

Dr Waltraud Schelkle is Lecturer in Political Economy atthe European Institute, LSE.

UUttiilliittyy ttaarriiffffss aanndd tthhee ppoolliittiiccaall

eeccoonnoommyy ooff ssoocciiaall ppoolliiccyyDeborah Mabbett

In many developing and transition countries, the provi-sion of utilities such as water, gas and electricity isrestricted by low investment and marked by inefficien-cies arising from network losses (leakages and theft),lack of metering and problems collecting payments.These sectors also often make losses which are borne bythe government budget. Many economists advocate

raising prices to cost-recovery levels, eliminating subsi-dies and cross-subsidies. An alternative is to implementa ‘social’ tariff to provide households with a basic level ofsupply of the utility at a low price. The policy advice frominternational financial organisations is often to elimi-nate social tariffs and replace them with a social assis-tance scheme to help poorer households pay their bills(Velody et al 2003). Researchers have drawn attention toa number of problems with the workings of social assis-tance schemes in practice (Lovei et al 2000). This articleexplains why the superiority of social assistance cannotbe established a priori, and instead depends on a coun-try’s specific economic and political conditions.

4


Both social tariffs and social assistance pursue distribu-tional objectives at some cost in allocative efficiency. Wecan think of the economy as comprising a number ofmarkets in which households participate as buyers andsellers. A general social assistance scheme may be seenas promoting allocative efficiency in consumer markets,but it is liable to have distortionary effects in labour mar-kets. Once we frame the intervention problem in thisway, it is clear that focusing intervention on income sup-port rather than utilities prices will be less distortionaryonly under certain conditions.

A key condition is that substitution between employ-ment and informal or domestic activity (or leisure)should be limited (Stern 1990). The superiorproductivity of formal employment supports this condi-tion in developed countries, although informal workand leisure are a problem for benefit administration. Butthe problems are much greater in transition economieswhere the informal economy accounts for 30-40 percent of GDP and a similar percentage of employment,compared with less than 20 per cent in most OECDcountries (Schneider 2002). Obviously it is difficult toadminister means-tested benefits if people are derivinga lot of income informally (and this might not only becash income, but goods and services of value too). Fur-thermore, a means-tested benefit system creates incen-tives for more informality, encouraging people to substi-tute other activities for earning a formal income.

Utility bills in many transition countries account for highproportions of poor households’ expenditure (EBRD2004). A social assistance scheme to address affordabili-ty problems of this magnitude will have significantincentive effects for many households. It is widelyrecognised that it is difficult to target social assistanceeffectively in transition economies, but policy makerstend to interpret this as an administrative rather than anefficiency problem. Furthermore, the high cost of socialassistance expenditure, which is directly related to utili-ties price increases, creates pressure for inefficientlydesigned assistance schemes, such as burden limitschemes, where households receive assistance with util-ities bills that exceed a certain percentage of theirincome. Such ‘earmarked’ social assistance distorts con-sumption decisions (for those who qualify, there is noreason to economize on usage) and may also create aclose administrative relationship between social assis-tance and bill collection, bringing the risk of politicalinteraction between social administrations (often localgovernments) and utilities providers.

The inefficiencies of social assistance have to beweighed against the inefficiencies of social tariffs. Low

tariffs may contribute to wasteful usage, although otherdeficiencies in market mechanisms, such as lack ofmetering and non-collection of bills, may have a greaterimpact. The inefficiency of social tariffs has to be evalu-ated against the indeterminacy of efficient prices forutilities, where short-run marginal cost (SRMC) is gener-ally below long-run marginal cost (the cost of producinga unit of output when additional capital investment isrequired), and also below the cost-recovery level whichwould enable a supplier to cover its sunk costs, notablyinterest payments. Strictly speaking, the price is waste-fully low only if it is below SRMC.

If the argument based on economic distortions is notstrongly against social tariffs, what other reasons arethere for recommending cost recovery? One problem isthat social tariffs attract support for the ‘wrong’ reasons.Defenders of inefficient utilities suppliers or industrialgiants with huge utilities arrears may find that invokingsocial policy considerations is strategically useful. Min-istries of Finance sometimes support social tariffsfinanced by cross-subsidies for budgetary reasons, con-tributing to the perception that these are the third-bestpolicies of fiscally weak governments. For internationaladvisers drawing their models from developed coun-tries with established welfare states, this perception isheightened by the assumption that distributional poli-tics should be assigned to the income side of house-holds’ budgets.

Another reason for focusing on income rather thanaccess to services is that social tariffs do not always ben-efit the poorest households, which may not be connect-ed to utilities. In some countries, social tariffs are a ‘mid-dle class’ benefit (Komives et al 2006). In transition coun-tries, which have higher connection rates than develop-ing countries with similar incomes, social tariffs favoururban over rural populations. This is a problem if thefocus is on addressing income poverty, but it is notdeterminative for the choice between social tariffs andsocial assistance, as assistance schemes also tend not toreach rural areas or the poorest of the poor.

Arguably, the pressure against social tariffs comes fromthose with most interest in promoting private financialparticipation in utilities infrastructure projects. For theEBRD, for example, tariffs must move to cost-recoverylevels if private participation is to occur (Kennedy 2003;EBRD 2004). Ongoing subsidies carry too much politicalrisk. A cost-recovery rule can, by contrast, provide a sus-tainable basis for tariff-setting by an independent regu-lator. Given the importance of investment in thedecrepit infrastructures of many transition countries,this is an important argument, but is it correct?

5


Note that it is not an argument of economic efficiencybut of political economy, emphasizing the importanceof securing property rights and protecting them frompolitical contestation. But the problems of achievingand maintaining regulatory independence in utilitiessuggest that security for private investors will comefrom political stability rather than from political disen-gagement. Social tariffs could have a place in establish-ing the necessary political support for a stable regulato-ry regime. It is sometimes asserted that social tariffs pre-clude competition, and this would jeopardize regulato-ry independence by reducing efficiency gains from pri-vatization and contributing to regulatory capture. Butregulators can impose social obligations in a competi-tive market, although they must guard against cream-skimming and ensure that the costs apply to all com-petitors (Irwin 1997).

It has become the norm to argue that social purposesshould be kept out of sectors seeking private invest-ment, but it is possible that this position is too doctri-

naire. We can ask whether an overburdened social assis-tance scheme, possibly closely tied to utilities prices, willreally protect an independent regulator from politicalinterference. A social tariff might not raise political riskcompared with a realistic assessment of the alternatives.

Deborah Mabbett is Reader in the School of Politics andSociology, Birkbeck College, University of London.EBRD (2004) Transition Report 2004: Infrastructure London: EBRD.

Irwin, T (1997) Price Structures, Cross-Subsidies, and Competition in Infrastructure’ PublicPolicy Journal No. 107.

Kennedy D (2003) ‘Power sector regulatory reform in transition economies: Progress andlessons learned’ EBRD Working Paper No 78.

Komives K, J Halpern, V Foster, Q Wodon with R Abdullah (2006) ‘The Distributional Inci-dence of Residential Water and Electricity Subsidies’ World Bank Policy Research WorkingPaper 3878.

Lovei, L, E Gurenko, M Haney, P O’Keefe and M Shkaratan (2000) ‘Scorecard for Subsidies:How Utility Subsidies Perform in Transition Economies’ Public Policy Journal No. 218.

Schneider F (2002) ‘The Size and Development of the Shadow Economies of 22 Transi-tion and 21 OECD Countries’ IZA Discussion Paper No 514.

Stern N (1990) ‘Uniformity versus Selectivity in Indirect Taxation’ Economics and PoliticsVol 2 No 1 pp.83-108.

Velody M, M Cain and M Philips (2003) A Regional Review of Social Safety Net Approachesin Support of Energy Sector Reform, USAID.

PPoollaanndd’’ss vveerryy ddiiffffiiccuulltt llaabboouurr

mmaarrkkeettJanusz Czamarski

Ben Slay

Poland is widely regarded as one of the most successfultransition economies, with GDP per-capita in 2005 some50 per cent above 1990 levels. On the other hand,according to the World Bank’s November 2005 regionalpoverty study, Poland is one of two countries in theEurope and CIS region in which absolute poverty rateshave risen in recent years.1 Poland’s labour markettrends may well have something to do with this: Polishunemployment rates have been among Europe’s high-est since the mid-1990s. Improvements in the areas ofpoverty and social inclusion in Poland are unlikely if theunemployment situation is not addressed.

What do the statistics show?

As the data in Chart 1 show, Poland’s overall registeredunemployment rate has been in the 18-20 per centrange ever since 2001. This is roughly twice the EU aver-age. Moreover, hidden unemployment (mainly in ruralareas) is estimated at approximately 1 million people,and informal employment is estimated at slightly lessthan a million. Labour market trends are particularly

unfavourable for women and young people. The long-term unemployed (those without work for at least 12

months) are also quite numerous, accounting for morethan 50 per cent of total unemployment. Poland’semployment rate is not surprisingly quite low: at only51.5 per cent in 2003, it was well below the EU15’s aver-age employment rate of 64 per cent, and 71 per cent inthe United States.

According to Poland’s 2002 census, work was the mainsource of income for some 12.4 million Polish workers(out of 16 million individuals aged 15-64). Some 10.7million people listed old-age and disability pensions,and other benefits as their main source of income; while14.5 million people (9 million of whom were below 15years of age) were supported by other workers. These

60

50

40

30

20

10

0 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06

long term unemployment: unemployedfor over 12 months

unemployment among peopleaged 15 - 24

unemployment (proportion of registered unemployedin the economically active civil population)

per centChart 1: Rates of registered unemployment, unemployment

among young people, and long-term unemployment, 1991-2005

6


data indicate that, on average, one Polish worker mustsupport two non-working Poles. The burdens theseratios pose on Poland’s social protection system sloweconomic growth in the short term and may not be sus-tainable in the long term.

Why is it so bad?

Although high unemployment rates are recorded in vir-tually all transition economies, Poland stands out assomething of a special case. The factors responsible forPoland’s unfavourable labour market trends include thefollowing: 2

Demographic conditions. Polish labour markets areabsorbing large numbers of new workers created by thepopulation boom of 1979-1985. New labour forceentrants will average some 650,000 annually through2010; many of these young workers are not finding jobsor face poor prospects for career advancement.

Weaknesses in education. Polish schools often do notprovide education of the desired quality, particularly interms of skill development. Polish students consistentlyscore at the bottom of the OECD’s Programme for Inter-national Student Assessment (PISA) tests; the numbersof vocational training schools are inadequate; and life-long learning systems have not yet been developed inPoland, reducing the flexibility and adaptability of thelabour. Meanwhile, Poland’s education system is unableto address problems of growing shortages of skilledworkers, particularly in the booming electrical engineer-ing and construction industries, as well as in medicine(nurses and doctors). In sum, the Polish educational sys-tem too often serves as a factory for creating unem-ployed workers, rather than preparing for jobs of thefuture. 3

Inappropriate social protection policies. During the early1990s some 2 million working-aged individuals tookadvantage of early retirement and rather lenient disabil-ity schemes to leave the labour market. While notincreasing unemployment directly, these measuresintroduced large disproportions between the workingand the occupationally passive populations. High pay-roll taxes, reduced numbers of jobs in the formal sector,and large fiscal burdens resulted. Expenditures on dis-ability pensions, retirement and other benefits consumealmost 44 per cent of Poland’s national budget!

Structural changes in the labour market. Growth in thedemand for labour has been slowed by enterpriserestructuring and heightened competition (which hasaccelerated the introduction of labour-saving technolo-gies), and by the relatively rapid development of mod-ern, capital-intensive sectors like finance and electrical

engineering. In these circumstances, annual GDPgrowth needs to exceed 5 per cent in order to signifi-cantly reduce unemployment in Poland — a figure thatwas last reached during 1995-1997.

Poland’s still small service sector. At present, Poland’sservice sector accounts for only about 50 per cent of Pol-ish employment; in Germany, the figure is more than 66per cent and in the UK it is 76 per cent. Much of the dif-ference can be explained by the absence of changes inPolish agriculture: the majority of Poland’s 1.4 millionfarms do not produce for the market, and contain largeamounts of hidden unemployment and low-productivity labour (which is also a major cause of ruralpoverty in Poland).

Poland’s relatively unfavourable business and investmentclimates. In surveys conducted by the Union of Industri-al and Employers’ Confederations of Europe, significantbureaucratic barriers to entrepreneurial activities werenoted by 88 per cent of surveyed foreign investors inPoland. Corruption was a concern for 60 per cent ofrespondents, as were poor infrastructure (73 per cent)and unstable legal and fiscal regulations. According toEconomist Intelligence Unit data, Poland’s $2,471 in per-capita foreign direct investment recorded in 2005 wasone of the lowest in the EU — and less than half the lev-els recorded in the Czech Republic and Hungary.

Ineffective decentralization of labour market regulation. Reforms introduced in the late 1990s transferred muchof the responsibility for conducting active labour marketpolicies to local governments. Many localities still do nothave the institutional capacity needed to dischargethese responsibilities. Matters are further complicatedby the fact that the decentralization was introduced inconditions of rapid growth in unemployment and ofinadequate public funding for active labour marketmeasures.

High taxes on labour. Taxes and other levies needed tofinance Poland’s social welfare system account for astunning 80 per cent of gross wages in Poland, leavingonly 20 per cent as net take-home pay. This large ‘taxwedge’ means that increases in net wages have a multi-plicative effect on enterprise labour costs. This weakensemployers’ incentives to create new jobs, or to raisewages for those currently employed.

The absence of effective dialogue. Despite the severity ofthese problems, Poland’s labour relations institutionsseem unable to find a common language. Trade unionsfocus on protecting existing (relatively rigid) labour reg-ulations and pay levels for those workers who areemployed. Employers who seek to reduce labour costs

7


and boost competitiveness are often unable to do sowithin the framework of Poland’s relatively inflexiblelabour law and collective bargaining system, and toooften, therefore, go outside the law. Whereas stateagencies in the past were Poland’s main employer, thestate is now increasingly focusing on macroeconomicand regulatory problems. In such circumstances, oppor-tunities for dialogue too often turn into confrontation,with obvious implications for prospects for addressingPoland’s labour market challenges.

Possible solutions

UNDP’s 2004 national human development reportWorking out Employment calls for the development andimplementation of a comprehensive national employ-ment programme that would be consistent with theEuropean Employment Strategy. Key elements wouldinclude:

1. According absolute priority in Poland’s developmentstrategy to measures to increase the demand for

labour, with a special focus on supporting entrepre-neurship and reducing the tax burden on labour.

2. Introducing more supply-side flexibility into thelabour market by, inter alia, improving the educationsystem (especially vocational education institutions) inorder to realize the principles of life-long learning.

3. Modernizing Poland’s active labour market policy

framework, in order to strengthen work incentives whilestill providing appropriate social protection.

4. Supporting a broad, more effective social dialogue

both within and outside the framework of Poland’s tri-partite labour relations system.

Prospects for change

Facing up to these employment challenges requires aresponse from Poland’s political elites, central and localgovernment bodies, entrepreneurs, and social partners,

including trade unions and non-governmental organi-zations. While the minority government that came topower in late 2005 has yet to seek to forge such aresponse, it is working on legislation to improve thebusiness climate and reform Poland’s public finances.Some (rather modest) proposals to reduce the taxwedge on Polish labour have also been put forth. Fundsfrom the EU’s European Social Fund, EQUAL, and (to alesser extent) INTERREG III are financing measures toaddress Poland’s labour market problems. The WorldBank’s ‘Post Accession Rural Support Project’ is directedat reducing long-term unemployment in Poland’s ruralareas.

Fortunately, EU membership has allowed (according tovarious estimates) some 800,000 to 1,200,000 Polishworkers to find legal temporary employment in otherEU countries, mostly in Ireland and the UK. Labour short-ages in construction and other sectors are increasinglyfilled by economic migrants arriving (not always legally)both from neighbouring states (Ukraine, Belarus) andfrom such countries as Vietnam and North Korea. Thisphenomenon, which is growing very rapidly and is notyet well understood in Poland, is likely to become partic-ularly important once the waves of young labour marketentrants begin to subside after 2010.

Despite these current difficulties, the future of thelabour market in Poland can perhaps be viewed withmoderate optimism.

Janusz Czamarski is the national human developmentspecialist in UNDP’s country office in Poland.([email protected])

Ben Slay is director of UNDP’s Bratislava Regional Centre.

1 Georgia is the other country. See World Bank , Growth, Poverty and Inequality: EasternEurope and the Former Soviet Union (2005: Washington D.C.), especially p.240.

2. For more detail, see UNDP’s National Human Development Report Poland 2004 - Work-ing out Employment (2004, Warsaw); www.undp.org.pl/nhdr_en/nhdr2004_en.pdf .

3. These problems are discussed in UNDP’s ‘NHDR 2006 – Education for Work’ which willcome out at the end of this year.

SSeerrbbiiaann rreeggiioonnaall iinneeqquuaalliittiieess

iinn aacccceessss ttoo jjoobbss Mihail Arandarenko

Regional labour markets and transition in Serbia

Regional labour market differences often deepen andpersist during the process of economic transition. Typi-cally the capital city and a few privileged regions experi-ence fast growth and a tight labour market, while muchof the rest of the country sinks into prolonged recession,

with high unemployment and low wages. Serbia is noexception to this general rule. We argue that Serbiashould adopt differentiated approaches to regionallabour markets, which would lead to regionally specificemployment policies.

Creating effective employment policies to combat per-sistent and rising unemployment has been a major chal-lenge for Serbian economic policy makers for manyyears. However, until recently little attention has beenpaid to the fact that workers face very different risks ofunemployment in different regions of the country. Itwas only in 2004 that regional labour market inequali-

mailto:[email protected]

http://www.undp.org.pl/nhdr_en/nhdr2004_en.pdf

8


districts, roughly comparable to NUTS 3 regions), andthen at lower levels (municipalities).

Our final list includes 21 indicators. In order to assesslabour market risks (which can be regarded both as aresult of past developments and a probability of futureimprovement), two main criteria have been used: (1)inherited conditions and the current situation (com-bined static factors); and (2) potential for positivechanges (dynamic conditions). We employed 10 staticand 11 dynamic indicators. We also decided to use threegroups of alternative indicators, giving evidence of (a)the general economic situation of the region (5 indica-tors); (b) regional labour market conditions (9 indica-tors); and (c) progress in economic restructuring andreforms (7 indicators).

After standardizing and normalizing the collected data,we developed a simple two-dimensional taxonomy ofregions with regard to their labour market situation andprospects. While regions with positive compositeindices of both their situation and their prospects ingeneral do not require additional intervention, theintroduction of employment policy measures and activelabour market programmes tailored to regional charac-teristics is recommended for regions in the other threequadrants. This methodology enables computing: (a)individual indicators, showing the relative position ofthe observation concerning a specific measure in rela-tion to other regions, or simply by its sign and magni-tude to the average value for Serbia; (b) the total com-posite index, combining all the indicators suitable forcomparison of the regions; and (c) group indices, forcomparisons regarding both the static and dynamicaspects, and the general situation/labourmarket/reforms aspects.Chart 1. Two-dimensional classification of regions in Serbia

po

ten

tial fo

r cha

ng

e

current situation

10

5

-5

-15 -10 -5 0 5 10 15 20 25

Beograd

Južno-bački

NišavskiRaški

Branič

Pomoravski

Rasinski

JablaničskiToplički

Pčinski

Pirot Zaječarski

Severno-bački

ŠumadijskiMačvanski

ties became one of the priority goals of Serbia’s Nation-al Employment Strategy. This recognition has inspired agroup of researchers to develop a specific methodologyfor regional risk assessment on the Serbian labour mar-ket, and to draw regionally-specific employment policyrecommendations.1

Workers are geographically rather immobile in Serbiadue to low wages, shortages of affordable housing, andreliance on kinship and social networks. Since these fea-tures are quite persistent and even intensify in times ofcrisis, the more promising way to capture regionalunemployment differences is to create more jobs in theregions that need them most. But, for the regionalemployment policy intervention to really work, policymakers need more information on which regions areaffected the most, and in what respects.

Therefore we have designed a methodology that woulduse all available relevant determinants to measure therelative labour market position for each of Serbia’sregions, as well as develop criteria for classifying theregions according to their prevailing problems. Theresults of this effort can be useful to policy makers atboth central and local levels, in designing developmentpriorities and regional aid programmes. They could alsobe useful in tailoring active labour market programmes,supporting programme design, monitoring the per-formance of the National Employment Service, and inadvertising investment opportunities for foreign firms.Our research also aims to offer a useful analytical andaction framework to local employment councils, whichare composed of local authorities, trade unions, busi-ness organizations, employment services, schools anduniversities and non-governmental organizations. Final-ly, the results could be of interest to international devel-opment agencies in providing more reliable guidanceon the overall needs of the regions and specific areas inwhich their intervention is most needed.

Regional labour market risk assessment

In developing the regional labour market risk assess-ment methodology, two main principles have beenemployed. First, large numbers of labour market charac-teristics must be taken into account. This is due not onlyto imperfections in the relevant data, but also due tovery complex interrelationships among the determi-nants of current labour market trends, their rapidchanges, and prospects for the future. Secondly, in orderto compare and rank different regions, a compositeindex of observable variables has been developed. Geo-graphically, we partitioned the Serbian labour marketfirst at a higher level of aggregation (25 administrative

9


The computed indices for all the observed regions canindicate: the individual position of each region withrespect to every indicator, group indices or the com-pound index, according to the average value of Serbia;the ordering of the regions by individual indicators orgroup indices, by interval with the extreme values ofindicators or indices for all regions; frequency distribu-tions of indicators or indices; or a mapping of Serbia bythe intensity of the compound index for each region.These different methods of presentation can serve dif-ferent research or practical purposes. For instance, itmay be interesting to analyze whether the same regionshows a negative discrepancy from the average meas-ured by one group index, but a positivedivergence for another, and whichregions in Serbia are especially vulnerablein more than one aspect. Likewise, it maybe of interest to note which group indicesshow much larger asymmetry than theothers, etc.

The analysis of the 21 individual indicatorsshows that such indicators as the share ofyoung population, or the ratio of jobvacancies to unemployment, have thelargest intervals of variation. On the other hand, thesmallest variances are found in the differences in theextreme values for the standardized indicators thatmeasure reforms, such as the share of employment inthe state sector. This means that indicators representingreforms and structural changes are much more alikethroughout Serbia than indicators that measure actualeconomic activity or inherited characteristics.

What does the model say?

A practical categorization of local labour markets can becreated, so that typical groups of regions can be identi-fied, based on the estimated partial assessments for thegroups of indicators. This taxonomy can help in identify-ing common characteristics, and in designing policiesthat are best suited to overcome unfavourable situa-tions of a certain type. For instance, according to ‘inher-ited conditions and current situation’ (1) and ‘potentialfor positive changes’ (2), a classification of regions intofour characteristic subgroups can be made by Descartes'quadrants, as illustrated in the Chart 1. The first quad-rant is characterized by plus-plus signs for the currentsituation and potential for change, the second byminus-plus signs, the third by minus-minus, and thefourth by plus-minus signs.

This taxonomy facilitates a simple segmentation intofour types of regions, suitable for regionally specific sug-

gestions about the labour market policy actions. Forinstance, regions in the first quadrant enjoy a relativelyfavourable situation on both counts, and thus shouldnot count on much additional support beyond thenationally designed employment policies.

Regions in the second quadrant reveal poor current cir-cumstances but above-average future prospects. Therecommended policy mix includes improvements insocial service delivery; strengthening of employmentassistance; support to SMEs and self-employment pro-grammes; the transformation of severance paymentsinto re-employment incentives; wage subsidies; retrain-

ing programmes; public works; and tempo-rary mobility programmes.

Regions in the third quadrant have belowaverage index values for both current andfuture criteria. These regions are obviously inthe most unfavorable situation and probablyneed massive assistance in solving theirproblems. As they tend to be traditionallyunderdeveloped, the recommended policymix should consist of: investment in physicalinfrastructure; building of industrial parks;

support to tourism development, crafts and healthyfood production; improvement in education coverageand establishment of new programmes; training pro-grammes; adult literacy programmes; public works pro-grammes; and workfare programmes.

Regions in the fourth quadrant show above-averageindicators for the current situation, but negative indica-tors of future potential. As these regions are obviouslylagging in reforms, accelerated restructuring and transi-tional changes would seem to be particularly important.Appropriate policy mixes here include: an emphasis oncompleting enterprise restructuring and privatization;SME development and self-employment programmes;transformation of severance payments into re-employ-ment incentives; strengthening of public employmentservices; training and retraining programmes; andstrengthening of educational networks, especiallyincreasing the number of university graduates.

While there is a clear and expected geographical dividebetween (traditionally developed) predominantlynorthern and central-north regions concentrated in thefirst quadrant and (traditionally underdeveloped) most-ly southern and eastern regions concentrated in thethird quadrant, such a pattern cannot be established forthe regions belonging to the second and fourth quad-rants. The former finding seems to confirm the domi-

Slobodan Lalović Employment Minister

10


IInneeqquuaalliittyy iinn ppoooorr hhoouussee--hhoollddss iinn TTuurrkkeeyy:: DDooeess oonnllyy

ggeennddeerr mmaatttteerr??Leyla Sen

Policies that promote greater social equality are expect-ed to lower poverty; however, if such policies fail to takeinto account internal household dynamics and hetero-geneity, they may actually reinforce inequalities withinthe same household. Although gender is a key determi-nant in the distribution of rights and resources withinthe household, it is not the only factor, as seen in thefindings of various field studies.

Interviews1 conducted with 150 women from poor fam-ilies in five provinces in eastern and southeasternTurkey, namely, Kars, Sivas, Batman, Diyarbakir andGaziantep, in the scope of various development projectsand initiatives, highlight inequalities in resource distri-bution in all poor households, with women having lessaccess to resources than men, regardless of the degreeof household poverty. However, resource allocation alsovaries among women themselves, and is influenced bysuch factors as age, marital status, employment statusand location (rural/urban).

The data show that poor women’s access to fundamen-tal resources varies. For example, access to nutritionalresources varies in households, but in all cases womenare the last to eat. Equally, hunger is a differentiatingfactor between rural and urban women, as rural womenpossess more coping mechanisms for securing meals.Poor women’s access to household monetary resourcesalso varies. In villages in Sivas and Kars, where seasonalmigration of male household members is a widespread

coping strategy, women take over the householdresponsibilities, including financial responsibilities,while the men are absent. However, while they may actas the ‘household safety valve,’ women require permis-sion from the men of the household before makingactual expenditures, and young women require permis-sion from older women.

In terms of labour force participation, poor urbanwomen are differentiated from poor rural women bytheir status as wage earners; however, poor workingwomen in Gaziantep, Diyarbakir and Batman are not ahomogeneous group. For example, whereas regularwage-earners feel they have a certain amount of controlover how their earnings are spent, women who arecasual employees have no such control, with their dailyincome, in some cases, paid directly to their husbands.Women’s control over their own earnings also variesaccording to their marital status. Relying on the data, itis possible to conclude that married women have morespace to enjoy control over their earnings. Half of themarried women indicated that they can spend theirearnings on the needs of their children without askingtheir husbands. Young working women, on the otherhand, to a great extent are denied control over theirearnings without much questioning. Many of theunmarried women respondents indicated that they donot want to enjoy control over their earnings while theirfamilies are suffering from poverty: “We are aware of thefact that we are working because we are poor.” Only afew asked for some of their earnings when they felt itnecessary.

In Diyarbakir and Batman, with few exceptions in thecases of married working women, male householdmembers, either husbands, fathers or brothers, con-trolled female labour and female earnings and denied

nant role of ‘inertia’ factors if the regional differences arevery pronounced; but the latter also shows that per-formance during the transition can significantlyimprove a region’s relative position. It seems that somerecent government’s economic and labour market poli-cy decisions recognize the importance of a simultane-ous attack on both ‘inertia’ and ‘transitional’ sources ofregional inequalities. ‘Inertia’ differences are beingaddressed by the positive discrimination of underdevel-oped regions within a very ambitious National Invest-ment Plan, aimed at large infrastructure projects, worthsome 1.5 billion Euros, to be implemented in the forth-coming four years. ‘Transitional’ differences, on theother hand, are being addressed by the repeatedlydeclared political will to complete, by now unevenlyregionally advanced, processes of privatization of social-

ly owned firms and restructuring of public sector asplanned, by the end of 2007 and 2008, respectively. Thereal challenge, still, remains to tailor and sequence thecompletion of privatization, large development projectsand specific policy measures recommended above in amanner which would decisively contribute to the clos-ing of the regional labour market gaps.

Mihail Arandarenko is Associate Professor of LabourEconomics at Belgrade University’s Faculty of Economicsand Employment Programme Adviser for UNDP Serbia.1 ‘Mapping Serbian Labor Market’, 2005 - the Foundation for Advancement of Econom-ic Science from Belgrade. It was followed by the book ‘Mapping Serbia’s Labour Market:Assessing Regional Risk and Potential’, edited by M. Arandarenko (2006) by the Centre forAdvanced Economic Studies, Belgrade. Methodological aspects of the research havebeen elaborated in ‘Assessment of regional labour market differences: methodology andresults’ M. Jovicic and M. Arandarenko presented at the 9th bi-annual conference of theEuropean Association for Comparative Economic Studies in Brighton in September 2006.http://www.brighton.ac.uk/bbs/eaces/papers/2d3.pdf.

http://www.brighton.ac.uk/bbs/eaces/papers/2d3.pdf

11


young women access to professional courses on thegrounds that their working is only on a temporary basis“until male family members find jobs or increase theirearnings”. By contrast, young working women inGaziantep were permitted to attend professional cours-es that could improve their labour market position. Theyalso indicated that it was possible for them to controltheir own earnings, although their sense of family obli-gation led them to pool their wages for the good of thehousehold. These different trends can be explained bythe length of household residence in urban areas.Whereas women interviewed in Gaziantep have beenliving in the city for more than five years, women inter-viewed in Diyarbakir and Batman were from householdsliving in the city less than two years. Their five to sixyears of experience in the city have helped Gaziantephouseholds to better develop their survival strategies. Incontrast, for households in Diyarbakir and Batman, uti-lization of female household labour seems to be less ofa strategy and more of a spontaneous developmentarising from poverty and misery.

Whereas young working women in Gaziantep receivedopen support from their mothers, young women’slabour force participation in Diyarbakir and Batmanincreased differentiation between female householdmembers and exacerbated intra-household conflict.Mothers in Diyarbakir and Batman attempted to placerestrictions upon their daughters in response to the newpower dynamics, which transformed young, unmarriedwomen from ‘silent servants’ of the village into ‘bread-winners’ in the urban labour market — and olderwomen into non-productive household members.Daughters, in turn, viewed their mothers as “wastingtime in the house” and as the hidden force behind theincreased aggressiveness of male household members.

Field research confirms Sen’s conclusion that the mobi-lization of the female labour force can be seen as a coop-erative act on the part of the household to ‘add to thetotal availabilities of the household’ (Sen: 1990). But italso indicated that, when women work outside thehousehold, this cooperation can coexist with extensiveconflict. While few households escape from this conflict,its nature depends on the household dynamics. Thesefindings support Sen’s bargaining model concerningintra-household relations and women’s empowerment,particularly in terms of the extent to which they candeploy their bargaining power and negotiate theirrights. If the conditions for female economic initiativesare restrained, women’s intra-household bargainingpower and opportunities will also be restricted (Row-lands: 1997). In households that have recently migrated

to cities, household members are generally not support-ive of working women; this reduces working women’sopportunities and limits their bargaining power.

These findings also indicate that relations betweenwomen can either mitigate or exacerbate these effects.When mothers regard their daughters’ work as a chal-lenge to their own position within the household (as inDiyarbakir and Batman), they tend to cooperate withmales to reinforce the subordinate position of youngerwomen. In addition to worsening younger women’sdaily lives, this also hinders their future in the labourforce by denying them access to training opportunities.When mothers cooperate with their daughters (as inGaziantep), the older women’s mediation between themale and young female household members canincrease the latter’s bargaining and negotiating power. Statements of working women suggest that their partic-ipation in the labour market, in terms of being a ‘bread-winner’ and contributing to the well-being of the house-hold, also improves their personal welfare. The mostchallenging aspect of perceptions about workingwomen is the notion of ‘breadwinner’. This is partlyrelated to the fact that ‘women’s labour – whether paidor unpaid – is conflated with social and gender identityand with membership in social groups such as the fami-ly’ (White: 1994).

Fieldwork shows that women in Turkey, regardless oftheir age, marital and employment status, or location,suffer from increasing inequality in access to food andbasic services such as healthcare. However, these com-mon experiences associated with the feminization ofpoverty do not mean that women are a homogeneousgroup; they instead remain a heterogeneous group,with conflicting interests. Current poverty line defini-tions and poverty mitigation/alleviation schemes inTurkey to a great extent ignore household dynamicsand their implications for differentiated experiences ofpoverty within the same household. Although a widerange of work on intra-household poverty dynamicsexists, their interactions with gender issues have beenlargely neglected; ineffective use of available resourcesand policies that do not touch the main causes of pover-ty and deprivation are the result. The multi-dimensionalnature of poverty requires revisions in these definitionsand policy approaches, in order to better reflect thecomplex dynamics between and within genders andage groups in the household.

A wider, in-depth examination of the dynamics withinpoor households is very important in this respect.Instead of treating poor households as homogeneous

12


units, differences in the needs and expectations ofhousehold members should be reflected, not in suchbroad categorical terms as ‘male’ and ‘female’, butrather, according to the structured hierarchy shaped bythe prevailing values within the household. The fieldstudies also indicate that values are crucial determinantsof the structured hierarchy within the household, gov-erning relationships between household members anddetermining the status of members and the allocationof resources within the household. If household dynam-ics are not taken into consideration, poverty alleviationschemes may further add to ‘wealth and power gaps’between household members, which will in turn lead toincreased conflict within the family and add further bur-

dens to households seeking to escape from the povertytrap.

Leyla Sen is a Project Manager at the UNDP TurkeyOffice. [email protected]

1 Interviews took place in the scope of various unpublished baseline surveys undertak-en for various development interventions, e.g., the Southeastern Anatolia RegionalDevelopment Programme, or the Baku-Tblisi-Ceyhan Community Investment Pro-gramme ranging, from 1998 to 2005.

References

Jo Rowlands, Questioning Empowerment: Working with Women in Honduras, Oxford:Oxfam, 1997.

Amartya Sen, ‘Gender and Cooperative Conflicts’ in Persistent Inequalities, ed. IreneTinker, New York, Oxford: Oxford University Press, 1990, pp. 123-150.

Jenny B. White, Money Makes Us Relatives: Women’s Labour in Urban Turkey, Texas:University of Texas Press, 1994.

PPoovveerrttyy rreedduuccttiioonn iinn BBeellaarruuss::

EExxppeerriieennccee aanndd lleessssoonnss

Lyudmila Istomina

According to official statistics during the past few yearsBelarus has shown significant success in poverty reduc-tion whilst maintaining a rather low level of income dif-ferentiation. The share of the population below thepoverty line has fallen from 41.9 per cent (2000) to 12.7per cent (2005), with a reduction of the absolute num-ber by 2.9 million people. The ratio between the percapita income of the most well-off 10 per cent and theleast well-off 10 per cent of the population is 5.4 (in Rus-sia it is 14.8). Efforts to reduce extreme poverty1 wereeven more remarkable – it fell from 3.2 per cent in 1995to 0.9 per cent in 2005. Since 2000 the average wage ofthose working in the public sector has risen eightfoldand, in real terms, doubled. During the past few yearsthe fast pace of income growth among the least well-offgroups of the population is one of the key factors inpoverty reduction. Thus, if in 2004 the real income of thepopulation increased by 14.4 per cent, in the pooresttenth it increased by 17.7 per cent. The data is by andlarge confirmed by the research of international organi-zations. In particular, the World Bank has recognizedthat the benefits of economic growth are shared by allgroups of the population, including the poorest.2 TheHuman Development Index ranks Belarus at 67 out of177 countries and places it in the group of medium-leveldeveloped countries, and thus ahead of all the other CIScountries except Russia.

However, it must be pointed out that the aforemen-tioned data reflect only trends. In terms of per capitaincome Belarus considerably lags behind the developed

countries in the world: per capita GDP (at purchasingpower parity, PPP) is $6644.

What is the basis of the socio-economic model?

The socio-economic policy of any state is based on theoptimal combination of two components – charity andefficiency. The Belarussian model of developmentemphasizes charitable traits. The emphasis on socialpolicies is declared at a political level, it is demonstratedby budget allocations for social affairs and it is translat-ed into state social programmes. During 1995-2005expenditure on social affairs comprised 48-50 per centof the state budget, amounting to 13-14 per cent ofGDP. According to experts, state social security pro-grammes account for 12-13 per cent of budget expendi-ture, which roughly corresponds to the share of poor inthe population in the country. This is one of the highestrates among the transition economies. Social securityprogrammes cover the majority of the population of thecountry (95 per cent of those working in the economy);and a number of programmes are implemented to pro-tect socially vulnerable groups (one-parent or large fam-ilies, handicapped people, veterans of war). Belarus hasa system of state social benefits and privileges availablefor different groups of the population and they aregranted based on more than 50 criteria. Owing to theLaw on State Social Standards adopted in 1999, the gov-ernment has worked out normative budget allocationsfor public healthcare services per person; also a numberof local social security programmes are run to meet setsocial standards in public healthcare. In the future simi-lar systems of social security standards will be devel-oped for transport and utility services. It must beacknowledged that the centrally determined mandato-ry wage target levels and various social programmescontribute to the decrease in social project expensesand thus play a role in reducing poverty.


13


The micro-level and the structure of poverty

A comprehensive assessment of poverty in Belarus3

shows that as in other transition economies, the socialgroups most likely to be at risk of poverty are children(according to 2005 data, 20.4 per cent of children under18 live below the poverty line); people living in ruralareas and small towns; unemployed; handicapped per-sons; and high-risk social groups. The phenomenon of‘working poor’ is rather widespread. Gender differencesare not a significant factor in poverty; food expensesdominate in the consumption structure (68.1 per cent);life expectancy, especially for men (62 years), is one ofthe least favourable indices amongst the countries inthe region. Ownership of houses, land and other materi-al assets serves as an important resource to sustain liv-ing standards and to secure additional income. There isan average of 22 square metres living space per personin the country. The produce (production of naturalgoods) from part-time farms and summer house landallotments still makes up a significant part of the totalincome of an average household. In rural areas this fig-ure stands at 18.1 per cent of the total householdincome and in urban areas seven per cent. Without theproduction of natural products, the poverty level in ruralareas could be double the current level. Sociologicalstudies show that groups with low incomes normallyhave a structure where paternal domination is pro-nounced.

The price of the Belarussian social model

The implementation of social programmes and projectsrequires resources which can be acquired by a redistrib-ution of GDP. In recent years the share of centralizedresources in the consolidated budget is 46-48 per centof GDP. This requires high levels of taxation and alloca-tion of resources to various budgetary and unappropri-ated funds. The high level of taxation is one of the diffi-culties facing the real economy. This is suggested by lowindices of profitability and by the significant share ofenterprises running at a loss (20 per cent).

The concentration of unprofitable and low-profit enter-prises in certain regions and sectors poses a serious riskof unemployment and leads to income differentials,especially in certain small towns and rural areas. Thenecessity of urgent restructuring and modernization ofindustry in Belarus is on the government agenda. How-ever, industrial reforms are stalled by the unsolved prob-lem of how to deal with the excess labour that wouldbecome redundant after the restructuring of industry iscarried out. Small businesses, especially in the regions,will not be able to absorb the aforementioned excess

labour due to the lack of a favourable environment forprivate business. According to IMF experts, price rises ofenergy imported from Russia would threaten to force areduction of state subsidies and other forms of redistri-bution, such as credits and transfers to unprofitablebusinesses.

The clear emphasis on social policies in certain waysrestricts development, including the development ofthe social security system itself, and decreases the effi-ciency of socio-economic policies aimed at curtailingthe spread of poverty. Here are some of the limitationsthat the emphasis on social policies brings:

� Insufficient financial stability of the nationalpension system which, against the backdrop ofunfavourable demographics, in the future may lead to adecrease in the level of pension security;

� Overloading the pension system with numer-ous pension benefits that are not of a social securitynature, and thus decreasing the system’s ability to influ-ence the labour market;

� Underdeveloped state of social security mecha-nisms such as unemployment insurance, medical andvoluntary insurance;

� Domination of the categorical principle of socialsecurity leading to unsubstantiated redistribution ofstate funds and to a distorted distribution, which bene-fits the relatively well-off groups most.

Preventive measures

During recent years there has been significant progressin terms of acknowledging that poverty is a multi-faceted problem, and in improving the methods ofanalysis and assessment of poverty by complying withexisting international standards. In particular, themethod of studying households has been improved bythe introduction of a number of modules used to assessavailable living space, access to education and health-care, as well as to certain social services and consumergoods. Within the framework of the joint project of theUNDP and the government of Belarus, the first stepshave been taken to assess the level of poverty and towork out a comprehensive strategy to prevent poverty.It is important that not only poverty reduction targetswere indicated — from 12.7 per cent in 2005 to 10 percent in 2010, and an estimated 8.9 per cent in 2015 —but also that preventive measures and concrete socialpolicy tools were recommended that will be used acrosstarget groups (children, families raising children, popu-lation of small towns and rural areas, low-paid workers,unemployed, unaccompanied elderly, handicapped

14


persons, high-risk social groups). Recently, the govern-ment has made a decision to include a poverty preven-tion strategy in the list of planned measures to imple-ment the Programme for Socio-economic Developmentin 2006-2010. The government has also outlined themost important parts for implementation in social poli-cies: strengthening the role of wages as the primary fac-tor in stimulating economic development and increas-ing the efficiency of the economy; supporting thegrowth of real income as a means to improve the quali-ty of life, reducing the number of people who are notwell-off; preventing excessive differences in theincomes of different groups across the regions and eco-nomic sectors.

Conclusions

The Belarussian government’s socially oriented policieshave contributed to rapid growth of real income, toalmost full employment and the reduction of poverty toone of the lowest levels registered in the CIS. The expe-rience of Belarus in sustaining existing state supportmechanisms and offering state social security is impor-tant for countries in transition, the duration of whichdepends on the state of the economy and the potentialof the country. External changes such as increases inprices of raw materials and energy may cause the situa-tion to deteriorate but cannot lead to immediate crisis.Moreover, the policies of redistribution of resources andcross-sector subsidizing help to dampen the adverseeffects of an economic shock, but they also decrease theoverall efficiency of the economy by extractingresources from the most profitable sectors and compa-nies to fund the implementation of social projects.

The policy to overcome poverty is anchored in statepaternalism and cannot be considered as a strategy fordevelopment. This policy puts certain limits on the long-

term development of an effective economy and curbsthe development of private entrepreneurship.

The need for sustaining state social security minimumstandards must be supplemented by the creation of anenvironment conducive to effective entrepreneurship.The structural economic reforms must be accompaniedby important tools/institutions that would provide pri-vate entrepreneurs with credits, finance and materialsupport. In particular, excessive administrative barriersto the opening and running of businesses must beremoved, and the tax burden must be reduced, in orderto increase investment. Against the background of anestimated excess labour force that will become redun-dant after the restructuring of companies is carried out,it is important to create conditions for encouragingmicro- and family businesses, especially in small townsand rural areas, and thus prevent redundant workersfrom slipping below the poverty line.

Lyudmila Istomina is Coordinator of projects on econo-my, social affairs and business at the UNDP office inBelarus.

Since 1993 in the Republic of Belarus the share of the population, which has a per capi-ta income less than the state minimum subsistence wage is considered poor (not well-off). MSG is a social standard, which is used for analysis and making projections on theliving standards of the population; providing social security to poor people (families);laying grounds for minimum state social and labour security (minimum wage, minimumpension depending on age, stipends, benefits).

Assessment of poverty is carried out by gathering information from studies of selectivehouseholds, and the results of those studies are published quarterly by the Ministry ofStatistics and Analysis of the Republic of Belarus. 6000 households were covered by theresearch.

The Article was prepared using the materials from the Joint Project of the UNDP and theBelarussian Ministry of Labour and Social Security ‘Assistance in Preparation of MainDirections for National Strategy to Prevent Poverty in the Republic of Belarus’ (2002-2006).

1 Households are considered extremely poor if their per capita resources are twicebelow the minimum subsistence wage.

2 ‘Benefits from growth were broadly shared by the population. Poverty rates declinedsubstantially, while inequality remained rather stable and moderate.’ Summary of theWorld Bank’s Country Economic Memorandum for Belarus (November 2005).

3 The Report on comprehensive assessment of poverty in the Republic of Belarus wasprepared within the framework of the UNDP Project on ‘Assistance in Preparation ofMain Directions for National Strategy to Prevent Poverty in the Republic of Belarus’(2002-2006).

CCeennttrraall AAssiiaa::

SSppaattiiaall ddiissppaarriittiieess iinn ppoovveerrttyyJacek Cukrowski

The economies of the Central Asian republics were cen-trally planned during the Soviet era and followed devel-opment strategies determined in Moscow. Their eco-nomic structures were broadly similar, although naturalresource endowments varied from country to country.Significant transfers from the central budget reduceddifferences across the region, so that Gini coefficients(measures of inequality) for all the Central Asianrepublics were close to the average value for the whole

USSR (Table 1). Poverty rates, however, were much high-er than in other Soviet republics.

After the dissolution of the Soviet Union, the newlyindependent Central Asian countries became more dis-tinct from one another as their governments followeddifferent national development paths. All five sufferedsharp drops in real output during the first half of the1990s, the impact of which was exacerbated by the ces-sation of transfers from Moscow. Inequalities rose inmost of the Central Asian states; today, poverty rates inthese countries are among the highest in the Common-wealth of Independent States (Table 1).

Increases in poverty did not affect all the countries andtheir constituent regions uniformly, however. The 15

15


Table 1. Poverty and inequality in Central Asia

Initial conditions Recent data

Per capitaGNP (1990)*

Gini coefficient(1989)

Poverty, per cent ofpop. (1989)**

Per capitaGNP (2003)*

Gini coefficient(2003)

Poverty, per cent ofpopulation*** (2003)

USSR $2870 0.289 11% -- -- --

Kazakhstan $2600 0.289 16% $1780 0.32 21%

Kyrgyzstan $1570 0.287 33% $340 0.28 70%

Tajikistan $1130 0.308 51% $210 0.33 74%

Turkmenistan $1690 0.307 35% $1120 -- 44%

Uzbekistan $1340 0.304 44% $420 0.35 47%

* GNP per capita, Atlas method (current US dollars)** Individuals in households with gross per capita monthly income of less than 75 roubles*** Percent of population with expenditures below $ 2.15 (PPP) per daySources: Pomfret, Anderson (2001), World Bank (2005a and 2005b)

years of economic transition show that the most suc-cessful economic regions are those with or in the vicini-ty of large urban agglomerations, and especially capitalcities. In fact, residence in a capital city is the mostimportant positive determinant of high living standardsin all Central Asian countries (Table 2), while povertyrates in rural regions are significantly higher.

The root causes of high rural poverty lie in the large-scale collective and state farms that dominated socio-economic life in rural Central Asia during the Soviet peri-od. In addition to agriculture and food products, thesefarms provided such social services as schools, healthservices and housing. Following independence, many ofthese farms collapsed or underwent dramatic restruc-

turing. Their social responsibilities were often trans-ferred to local governments that did not have the finan-cial or human resources to discharge them. The ruralpoor often lack formal property rights (in terms of hous-ing, land and other assets), without which they cannoteasily borrow money or engage in entrepreneurial activ-ities. Access to piped and running water, heat, health-care and kindergartens in rural areas has thereforedeclined and the quality deteriorated (see Table 2). All ofthis reduced living standards in rural areas.

More than 60 per cent of Central Asia’s poor live in ruralareas. (In Kyrgyzstan, Tajikistan and Uzbekistan, theshare is between 72 and 76 per cent). The highest pover-ty rates are observed in mountainous areas — which arealso prone to flooding, landslides and other natural dis-asters. In Kyrgyzstan, differences in poverty ratesbetween the richest Chuy region (the province sur-rounding the capital, Bishkek) and the poorest region(mountainous Naryn) exceed 50 percentage points (Fig-ure 1). In Uzbekistan, the regional disparity in povertyrates between Tashkent city and the Syrdarya regionexceeds 65 percentage points.

The spatial variation of poverty rates in Central Asia (Fig-ure 2) shows lower levels of poverty in capital cities andcentral regions and much higher poverty levels in

Table 2. Differences within countries

Poverty rate $PPP 2.15/ day (per cent, 2003) Household access to water (per cent, 2003)

Capital Other urban Rural Capital Other Urban Rural

Kazakhstan 2% 14% 31% 100% 96% 87%

Kyrgyzstan 42% 68% 77% 87% 97% 77%

Tajikistan 54% 73% 76% 100% 93% 56%

Turkmenistan -- -- -- -- -- --

Uzbekistan 4% 43% 55% 100% 90% 60%

Source: World Bank 2005

Bishkek (50-59 per cent)

ChuyYsyk-Kol

Naryn

OshBatken

Jalal-Abad

Talas

40-49 per cent

70-79 per cent

80-89 per cent

90-100 per cent

Figure 1. Poverty rates in Kyrgyzstan (according to a $2.15 PPP/day poverty line)

Source: UNDP estimates

Figure 2. Poverty rate in Central Asia(poverty line at $2.15 PPP)

no data

up to 20 %

20 - 40 %

40 - 60 %

60 - 80 %

above 80 %

KAZAKHSTAN

UZBEKISTAN

TURKMENISTANTAJIKISTAN

KYRGYZSTAN

Source: UNDP estimates

remote border areas. This, together with regional clus-tering of ethnic minorities in Central Asia, suggests astrong link between spatial distribution of poverty andethnicity. Capitals and large urban agglomerations aretypically populated by large Russian (and other Euro-pean) communities, as well as by the wealthier membersof these countries’ titular nationalities—groups with rel-atively high levels of human capital. The poor rural andborder areas, by contrast, are more likely to be home tonon-Russian (or European) ethnic minorities. These pat-terns of ethnicity and poverty, combined with the pres-ence of extra-territorial enclaves in most Central Asianrepublics, leave much scope for disruption and conflict.

Reductions in regional disparities in Central Asia couldimprove regional stability, human security and overalldevelopment prospects in Central Asia. Althoughresponsibilities for equalizing regional disparities andensuring balanced development within these countriesmust rest with their governments, reducing high pover-ty levels in border areas requires cooperation betweenneighbouring states. As indicated in UNDP’s Central AsiaHuman Development Report,1 regional cooperation,open borders, and the free movement of goods andpeople are essential for poverty reduction and develop-ment in Central Asia.

Jacek Cukrowski is regional adviser on the MillenniumDevelopment Goals, based in UNDP’s Bratislava Region-al Centre. ([email protected])

1. Bringing down barriers: Regional Cooperation for human development and humansecurity. Central Asia Human Development Report. UNDP RBEC Bratislava Regional Cen-tre, Bratislava, 2005.

Pomfret, R., and Anderson K., (2001) Economic Development Strategies in Central Asiasince 1991. Asian Studies Review, Vol 25, No.2, pp. 185-200.

World Bank (2005a) Growth, Poverty and Inequality. Eastern and the Former SovietUnion. World Bank, Washington D.C., 2005.

World Bank (2005b) World Development Indicators 2005, World Bank, Washington D.C.,2005.

16


PPoovveerrttyy aanndd iinneeqquuaalliittyy iinn

UUzzbbeekkiissttaannUktam Abdurakhmanov

Sheila Marnie

The Millennium Development Goals (MDGs) and Pover-ty Reduction Strategy Papers (PRSPs) discussionprocesses have provided a new focus for opening up thepoverty debate in Uzbekistan. In the 1990s, poverty wasnot officially discussed or recognized as a problem.Since then the government, with support from UNDPand other international development partners, hasbecome more actively engaged in global efforts toreduce poverty. In 2006, the government produced itsfirst national MDG Report. At present, UNDP is taking thelead in coordinating work with the government andother stakeholders in formulating the country’s first fullPRSP, while also supporting macroeoconomic and sec-toral studies, as well as projects aimed at focusing MDGsat the regional level, and using them as a basis for theformulation of both regional and, more recently, area-based development strategies.

The profile of poverty in Uzbekistan

It is estimated that 27.5 per cent of the population or 6.8million people were living below the poverty line in

2001 and that 9 per cent were living in extreme pover-ty.1 Within the framework of the PRSP, the governmenthas updated the poverty estimates for 2003 using thesame methodology, and estimates that the overall levelof poverty fell slightly to 26.2 per cent in 2003. TheHousehold Budget Survey (HBS) results suggest thataround 70 per cent of the poor population lives in ruralareas. However, other surveys have also suggested thatthe poverty risk is high for residents of small towns,where employment opportunities have decreased (dueto, for example, the non-functioning of large employ-ers), and where the access to land is more limited.2 In2001, households with an unemployed head of house-hold were more likely to be poor, but over 50 per cent ofthe poor live in households where the head is actuallyemployed. This suggests that poverty risk is not justlinked to unemployment, but also to underemploymentand low wages. Low salaries in the public sector (wherewages are 60 per cent of the national average) and agri-culture (50 per cent of the national average) contributeto poverty among the employed. Poorer householdswere larger than average, and had both more children(four or more) and adults.

The Gini coefficient for Uzbekistan is estimated at 0.353

and has decreased since the mid 1990s. This indicatesthat, as in other countries in the region, there are aver-


17


age-to-high levels of income inequality. However, thereare doubts about the household budget survey’s abilityto capture the upper income population. This wouldlead to the under-reporting of the largest incomes, andthe under-estimation of inequality. Differentials in aver-age wages between sectors also suggest high and grow-ing disparities, particularly between agriculture andother sectors.

Poverty, inequality and the character of economicgrowth

Independence and the onset of the transition in Uzbek-istan coincided with a number of severe shocks. Torespond to these shocks and to transform its economicsystem, Uzbekistan followed a policy model differentfrom that adopted in most other transition economies.This approach has been ‘unorthodox’,4 both during themacroeconomic stabilization phase of 1991-1995 andduring the recovery phase of 1996-2003. However onthe basis of this ‘home-grown’ approach, the countrymanaged to outperform the rest of the region duringthe stabilization phase; and in the recovery phase sus-tained moderate but acceptable annual GDP growthrates of 3.5-4.0 per cent. By 2001 it was the only countryof the former Soviet Union to have surpassed its esti-mated 1989 level of GDP.5

These relatively encouraging outcomes have, however,been marred by a number of persistent problems, whichhave prompted analysts and more recently the govern-ment to recognize that some policy changes might be inorder. While poverty increased sharply – and, to someextent, unavoidably – during the contraction-stabiliza-tion years of 1991-1995, it then stagnated during thesubsequent recovery and was accompanied by growinginequality. Indeed, income inequality has escalatedmarkedly since 1995-1996, due largely to the policy biasthat favoured urban-based, capital-intensive, medium-and large-size enterprises. The capital-intensive natureof the import-substitution model followed since 1995has meant that both urban and rural labour marketswere unable to absorb the rapidly growing working-agepopulation,6 as well as the labour being shed in agricul-ture and the state-owned enterprises. As a result, whileopen unemployment has remained low, the number ofunderemployed engaged in low-productivity, low-wageor part-time jobs is high. This can largely explain thecontinued high poverty levels despite relatively goodgrowth rates. Other factors have also contributed to are-examination of the country’s economic developmentmodel, not least the fact that world cotton prices havebeen cut in half during the last few years. In short, theimport substitution strategy pursued from 1995

onwards has shown itself to be increasingly sub-optimalboth in terms of pace of economic growth and incomedistribution.

The last two or three years have seen signs of movementaway from the strategy of the late 1990s. This is appar-ent in the unification of the exchange rate (which hasled to substantial increases in exports), and measuresaimed at promoting private sector development, includ-ing the lightening of the corporate tax rate (from 30 percent to almost 15 per cent), simplification of registrationprocedures for small and microenterprises, andimproved access to markets through reductions inexport restrictions. As a result, annual GDP growth dur-ing 2004-2005 exceeded 7 per cent.

Underemployment, not unemployment, is associat-ed with income poverty

Despite this accelerating growth, there are as yet fewsigns of falling rural poverty rates. Instead, reformsaimed at improving agricultural productivity may havefurther reduced employment opportunities in ruralareas.

The unemployment rate for Uzbekistan, based on ILOmethodology, was just 3.6 per cent of the economicallyactive population in 2003, and has generally remainedlow. However, there is also a high level of hidden unem-ployment in the formal sector. As was shown above, lowwages in the public sector and low incomes in agricul-ture (in particular from dekhan farming, see below)mean that being employed does not necessarily protectfrom poverty. Moreover, an estimated 29 per cent of thetotal number of employed are working in the informalsector, where they have no social security or otherrights.

In the 1990s, rural households were allocated land plotson the basis of what had been personal plots on stateand collective farms during the Soviet period. Some 82per cent of Uzbekistan’s households benefited fromthese plots, which allowed for a minimal standard of liv-ing. These plots (average size of which is 0.12 ha) havesince become the basis for small family or dekhan farms.During the same period, the vast majority of state andcollective farms were converted into shirkats (coopera-tive farms based on household contracting). Since thelate 1990s, private commercial farms have also beenintroduced. In contrast to the dekhan farmers, the pri-vate farmers are allowed to use hired labour and to havelarger plots (up to 10 hectares), but are subject to com-pulsory procurement at state-determined prices (partic-ularly for cotton and wheat). Since 2003, the shirkatshave been gradually transformed into private commer-

18


cial farms, and in the near future, the share of commer-cial private farms is forecast to increase to up to 85 percent of total sown land, and dekhan farms to 12 per cent.Land will continue to be owned by the state and, for thetime being, commercial private farms will continue to besubject to compulsory procurement of cotton and grainat state-determined prices. However, during the last fewyears the government procurement prices for cottonand wheat have reached world levels.

Still, without strong private sector development in thecountryside, much of the rural population still relies onemployment in dekhan farms, which in most casesamounts to little more than subsistence agriculture.While these workers are classified as employed, theiractual income generating opportunities are very limit-ed, due to poor access to markets, credit, equipment,and unreliable irrigation systems. The transformation ofshirkats into private farms is also further reducingemployment, since the more efficient private farms havehigher labour productivity rates and therefore employfewer workers.

Rural areas in Uzbekistan were always disadvantagedcompared to cities even in the Soviet period. However,the post-1995 economic policies have probably rein-forced these disadvantages, producing rising povertylevels in the countryside and more disparities betweenrural and urban areas. Demographic trends have meantrapid growth in the supply of (mainly low-skilled) labour,and poor employment opportunities in rural areas andsmall towns. These trends are pushing growing num-bers of rural residents into small subsistence farming, onto mardikhors (informal labour markets), or into emigra-tion. In human terms this translates into more vulnera-bility for rural residents, with growing numbers of chil-dren being brought up in incomplete families.

Conclusions

Data constraints make the determination of clear trendsdifficult, but it seems that Uzbekistan’s post-1995 eco-nomic policies have achieved economic growth rateswithout necessarily contributing to poverty reduction orincome equality. This has prompted the government tolook for ways of adapting policies to ensure that eco-nomic growth becomes more pro-poor. UNDP is cur-rently helping the government to formulate povertyreduction measures, by supporting both the formula-tion of a national poverty reduction strategy and region-al and district-level initiatives, based on participatory,grass-roots approaches. The goal is to find area-specificsolutions to poverty problems, especially but not only inrural areas, which can in turn support the implementa-tion of the national strategy at the local level.

Uktam Abdurakhmanov is PRSP Project Manager andSheila Marnie is International Consultant on PRSP,UNDP, Uzbekistan

1. The poverty line used here is based on the cost of a minimum food basket whichwould provide 2100 calories per day. In Uzbekistan a food poverty line is used, as tech-nical factors prevent the construction of a reliable consumption welfare aggregate usinghousehold budget survey data on household expenditures on non-food products.Therefore, food consumption expenditure is measured against a food-based povertyline. Extreme poverty rates were calculated using a more restrictive poverty line, whichin theory would only guarantee 1500 calories per day. For more details, see World Bank,Uzbekistan Living Standards Assessment, 2003.

2. See for example the survey of Dzhizak oblast carried out by Tahlil SociologicalResearch. 2003.

3. See Growth, Poverty, and Inequality. Eastern Europe and the Former Soviet Union.World Bank, 2005

4. The ‘orthodox’ approach, taken for example by neighbouring Kazakhstan and Kyr-gyzstan, and supported by most international organizations, can be characterized byrapid price liberalization, removal of all consumer and producer subsidies, budgetaryausterity, and trade liberalization and privatization. Uzbekistan, on the other hand,adopted a policy characterized by gradual price liberalization, continuing enterprisecontrols and subsidies, tight monetary supply control, limited trade liberalization, andslow privatization. For more detailed analysis see Giovanni A. Cornia, ‘MacroeconomicPolicies, Income Inequality and Poverty: Uzbekistan 1991 – 2004’, in Pro-Poor Macroeco-nomics. London, Palgrave, 2006.

5. ‘Linking Macroeconomic Policy to Poverty Reduction in Uzbekistan’, UNDP 2004.

6. Every year more than 250,000 young persons enter the labour market (1.7% of work-ing age population).

PPoovveerrttyy aanndd iinneeqquuaalliittyy iinnAArrmmeenniiaa

Aghassi Mkrtchyan

Recent trends in growth, poverty, and inequality

Armenia’s overall economic performance in recent yearshas been outstanding, with real GDP growing on aver-age at more than 10 per cent annually since 2000. Signif-icant progress has been made in reducing absoluteincome poverty: poverty incidence vis-à-vis the nationalpoverty line (which is nearly $ 4 per day per person inpurchasing power parity terms) dropped from 55 percent in 1999 to 35 per cent in 2004. Progress in terms ofreducing food poverty is particularly impressive: theincidence of food poverty fell to around 6 per cent in2004 from 22 per cent in 1999.

Poverty reduction in Armenia has resulted not only fromrapid economic growth, but also from reductions inincome inequality. In less than five years, Armenia’s Ginicoefficient (for household incomes) is measured to havedropped from 0.6 to 0.4. This is an impressive trend,especially since very few developing countries have suc-ceeded in reducing income inequality during periods ofdouble-digit economic growth. Unfortunately, concernsabout the quality of Armenia’s household survey datastrongly suggest that the reported declines in inequalityare exaggerated.

Possible problems in income distribution statistics

Estimates of income equality come from household sur-vey data conducted by Armenia’s National Statisticalservice. A comparison of data from 1999 and 2004household surveys with national account data for those

19


years suggests that data deficiencies could be affectingthe inequality estimates. At issue is the differencebetween the increases in total household incomes/con-sumption reported in the household surveys, versus theincome data that is reported in Armenia’s nationalaccounts (particularly for gross national disposableincome (GNDI) and private consumption).

For example, in the 1999-2004 period, Armenia’s realGNDI1 posted 63 per cent cumulative growth. Butaccording to household survey data, total householdincomes have increased by only 14 per cent. In terms ofregional breakdown, the household survey data indi-cate that the real household income growth was verylow in the capital Yerevan (less than 4 per cent). Thiscontrasts not only with GNDI dynamics but also withconventional perceptions that in recent years growth inYerevan was much higher than in other parts of thecountry. Discrepancies in Armenia’s consumption dataalso seem to be present, although somewhat smaller.While household survey data indicate only 20 per centreal consumption growth during 1999-2004, the nation-al accounts data show private consumption increasing47 per cent during this time. The smaller gap in con-sumption data from these two sources might alsoexplain the less dramatic change reported in consump-tion inequality, shown by Gini coefficients of 0.26 in2004 versus 0.3 in 1999.

Under-reporting poverty and inequality estimates

Under-reporting is a perpetual concern in householdsurveys, particularly in transition economies whereupper-income survey respondents may fear the tax con-sequences of accurately reporting their incomes. Suchunder-reporting might have prevented the householdsurveys from capturing the full increase in householdincomes and consumption. If so, this would bias esti-mates of poverty and inequality in Armenia. The bias inthis under-reporting might have a larger effect on theinequality figures than on the poverty data, as povertyassessments are based on consumption aggregates andsmaller differences are apparent in the consumptiondata derived from household survey and nationalaccount sources.

An examination of Armenia’s income distribution dataacross income groups indicates that under-reportingmight have been skewed toward higher income groups.Reported growth in incomes and consumption of thelower quintile generally mirrors trends in the economicexpansion (the incomes of the lowest two quintilesincreased 56 per cent in 1999-2004). However, thewealthiest two quintiles of Armenia’s population report-ed a 5 per cent decline in their real incomes for the peri-od of 1999-2004. Regionally, the two highest quintiles inYerevan during 1999-2004 reported a 14 per cent realincome decline. Taken at face value, these data suggestthat the wealthiest households did not benefit at allfrom Armenia’s macroeconomic boom — a claim that isdifficult to accept at face value. These discrepanciesstrongly suggest that the under-reporting of householdincomes has been concentrated in Armenia’s wealthiestpopulation groups, and in the capital area.

Almost 90 per cent of the growth in the incomes of theincome group in the lowest 40 percent is associatedwith expanded economic opportunities such as higherincome from salaried employment, self-employment,and agricultural incomes. The role of social assistanceseems to be rather small compared to labour incomes,underscoring the importance of economic growth(rather than redistribution) as the major source of pover-ty reduction. This argument is strengthened by the factthat the relative size of the government during 1999-2004 did not increase. In fact, both tax/GDP and expen-ditures/GDP ratios declined during this time, implyingreductions in income redistribution via social policy.

Conclusion

Armenia’s rapid economic growth since 1999 has cer-tainly increased incomes and consumption, therebyreducing absolute poverty (measured in income terms).However, a comparison of the household survey data(which are used to measure income poverty andinequality) with Armenia’s macroeconomic data strong-ly suggest that the large reported declines in incomeinequality during 1999-2004 are overstated.

Aghassi Mkrtchyan is a Country Economist at the UNDPCountry Office in Yerevan, Armenia.1 As data on household disposable incomes are not available for 2004, GNDI (whose1999-2003 trend is similar to that of household disposable income) can be used as aproxy for household disposable income.

FFiigghhttiinngg ppoovveerrttyy iinn ppoosstt--

SSoovviieett TTaajjiikkiissttaannCheickna Diawara

Poverty and inequality in Tajikistan

Tajikistan’s transition from a centrally planned to a mar-ket economy coincided with the country’s 1992-1997civil war, and the fierce political, social and economicshocks of the Soviet break-up. Industrial complexes,

roads, power plants, universities and research centres,water and sanitation systems had been maintained withtechnical and financial support from Moscow. Theexpensive and collapsing Soviet-made infrastructuresubsequently proved difficult to maintain, and many ofthe skilled workers needed to service it emigrated. Inrural areas, much basic social infrastructure and manyincome sources were linked to the collective farms orother Soviet-era institutions that did not survive the tur-bulence of the 1990s. By the late 1990s, Tajikistan (and

20


some other Central Asian countries) was reportingincome poverty levels that were comparable to those ofAfrica’s poorest countries—despite far better access toeducation and basic social services inherited from theSoviet Union.

Transition in Tajikistan and neighbouring countriestherefore became a matter of reconciling the need tomodernize the out-of-date infrastructure (and the asso-ciated Soviet-era institutions) with the need for emer-gency rehabilitation of the existing infrastructure, inorder to reverse the sharp increases in poverty andinequality throughout the country. For policy makersand development practitioners in Central Asia, thedilemma centers on how to overcome the Soviet legacyand convince local communities to subscribe to reform.These efforts are complicated by the sometimes disap-pointing results in countries like Ukraine and Georgiawhere, despite promises from “new democratic leaders”,there is still much to be desired in terms of transparen-cy, accountability, individual liberties, democracy, andbetter living conditions for the vast majority of citizens.

Fortunately, Tajikistan’s economy has been growingsince 1997, with GDP increasing at average annual ratesof 9-10 per cent since 2000. This reflects the Govern-ment’s commitment to reforming the economy andtightening monetary policy, liberalizing prices, and the‘voluntary privatization’ of land, housing and state-owned small- and medium-sized enterprises (SMEs), andrevising the tax codes and trade tariffs. Prices and theexchange rate have remained stable, particularly since2004. According to World Bank and national data, theshare of the population living in absolute poverty (onless than $ 2.15 per day in purchasing power parityterms) decreased by almost 23 per cent, while the shareof the population living in extreme poverty (on less than$ 1 per day) was halved, from 36 per cent to 18 per centby 2003. Nonetheless, Tajikistan remains the poorestcountry amongst the former Soviet states, with a GDPper capita of about $340 in 2005. Significant regionaland demographic disparities exist within Tajikistan, andinequality has significantly increased despite smallimprovements reported in consumption levels for mostincome groups. Rural poverty is exacerbated by badgovernance, corruption, poor land management andusurious short-term farm loans, producing farm debtsthat had risen above $300 million in 2006). Rural com-munities are being depopulated by migration becauseof the collapse of farms, unemployment and constraintson entrepreneurship. Remittances account for morethan a third of GDP. Households headed by women, orwith five or more children (which are common in bothrural and urban areas) are more likely to be poor.

Despite this GDP growth, the economy remains heavilyreliant on a few export goods (cotton and aluminium),and is thus vulnerable to world price fluctuations.Growth has likewise not been reflected in public financ-ing for basic social services. For example, state budgetallocations to the water sector remain at 6-7 per cent oftheir 1990 level. Although they represent significantshares of the state budget, allocations for health andeducation, for basic social infrastructure, and for socialsecurity and pensions also remain quite insufficient. Incontrast to many poor countries, literacy rates in Tajik-istan remain relatively high, at 95 per cent (compared to99 per cent in 1991). Nonetheless, the number of chil-dren not attending school is growing, and secondaryschool attendance rates are declining. Annual popula-tion growth rates in excess of 2 per cent add furtherpressures to basic social services, as well as hinderingthe government’s ability to prepare for and respond tonatural disasters.

The MDGs and poverty reduction in Tajikistan

The United Nations Millennium Summit in 2000 adopt-ed the Millennium Development Goals (MDGs), callingfor significant reductions in global poverty by 2015. Atthe September 2005 follow up summit, the UN called ondeveloping countries to align poverty reduction andother national development strategies with the MDGs.Work in Tajikistan in this area actually preceded the2005 summit declaration: the government and UNDPpublished the MDG Needs Assessment,1 linking project-ed economic reforms, growth trends, and external assis-tance to the prospects of halving poverty in Tajikistan by2015. Following the September 2005 summit, workbegan on preparing an MDG-based national develop-ment strategy (NDS).

The NDS, which is supported by all of Tajikistan’s donors,is meant to provide a development vision to 2015, onethat builds both on the MDG needs assessment andwork done within the framework of Tajikistan’s PovertyReduction Strategy Paper (PRSP).2 It focuses on ques-tions of government leadership, public participation,and broad national ownership. The NDS recognizes agri-culture, energy and aluminium production as the coun-try’s ‘growth engines’, but also calls for sectoral diversifi-cation and SME development, more funding for socialservices, and better public expenditure management.

If successful, the NDS, together with Tajikistan’s secondPRSP (for 2007-2009), will improve access to safe drink-ing water, education, and healthcare, and lead to lowerchild mortality rates. However, a number of problemswill need to be surmounted. For one thing, the PRSP andthe sectoral development strategies developed by vari-

21


ous ministries and donors have yet to be aligned withthe NDS or finalized by the government (with supportfrom UNDP in consensus with the country’s main part-ners in development). Donors’ commitment to Tajik-istan, both in terms of the long-term vision of the NDSand the medium-term vision of the second PRSP,remains to be seen. Human and institutional capacitiesto design, formulate, and implement policy reformsremain weak, and Tajikistan’s external balance is verysensitive to fluctuations in the world prices for alumini-um and cotton (the country’s leading exports). Despitenearly a decade of economic growth, Tajikistan’s per-capita GDP remains 40 per cent below its pre-transitionlevel. Even if the NDS’s projected 7 per cent annual base-line growth rates are realized, the country’s pre-transi-tion level of GDP per capita would be achieved only by2025. Therefore together with the reforms, the debtrelief and the commitments by donors to substantialincreases in development assistance are extremelyimportant to allow Tajikistan to reach most of the MDGsby 2015. Tajikistan benefited from a $250 million debt-for-equity swap with the Russian Federation in 2004,and continues to enjoy concessional financing from theIMF and World Bank. However, the borrowing costsassociated with large new investments—particularly forwater, energy and other infrastructure projects—couldsignificantly increase the strains on Tajikistan’s externalbalance.

Will it be enough?

The characteristics of the poverty and developmentchallenges facing Tajikistan are clearly very differentfrom those of African and other low income countries,particularly in terms of causes and historical legacies.

Still, income levels in Tajikistan are similar to thosereported in many poorer African countries, and thispoverty has similar devastating effects on vulnerablecommunities and human security. The institutionalcapacity needed to assess poverty, formulate and imple-ment pro-poor policies is in some cases even lower inthe poor Central Asian countries, because of their shorthistories and young institutional cultures (as independ-ent countries). In the case of Tajikistan’s weak state,institutional capacity was further damaged by the civilwar. Civil society and community-based organisations,parliaments, independent mass media institutions, andtrade unions are just nascent institutions. For these andother reasons, the poorest Central Asian countries needthe kind of assistance and attention from the interna-tional community that is accorded to African countries.Indeed, the relatively good access to health, education,and other social services in Tajikistan, and the rapid eco-nomic growth now being experienced, suggest thatpoverty and other development challenges in CentralAsia have a more shallow character, and can perhaps bemore easily reversed than in other developing countries.Adapting the MDGs to national circumstances can helppolicy makers to more quickly identify the steps neededfor this reversal.

Cheickna Diawara is the National Development Strategy/Poverty Reduction Strategy team leader at the UNDPoffice in Tajikistan. [email protected]

1. See http://www.undp.tj/home/MDG_NA_Full_Eng.pdf

2. PRSPs ideally align donor assistance and national macroeconomic and social policyframeworks with the imperatives of poverty reduction, during a 3-5 year time period. Formore on Tajikistan’s PRSP, see www.imf.org/external/pubs/ft/scr/2006/cr0601.pdf.

PPuuttttiinngg tthhee ffooccuuss bbaacckkoonn cchhiillddrreenn dduurriinngg aattiimmee ooff eeccoonnoommiiccggrroowwtthh

Gordon AlexanderPetra Hoelscher

Child poverty: have we learned from the 1990s?

During the 1990s, the region saw a strong emphasis onradical market policies and moves to privatize public

goods and utilities. Social reforms tended to be neglect-ed. The architects of transition, preoccupied with press-ing problems on the economic and political fronts,viewed social costs as temporary or at least reversible.Children paid the price of that neglect, with sharpincreases in poverty and unprecedented reversals insocial indicators affecting children. Since 2000, and thereturn of economic growth, the situation of children inthe region has improved and there is a sense that pover-ty among children is a thing of the past. There is a risk,however, that in the economic upswing, benefits forchildren will be assumed rather than identified, and thatonce again children will find themselves missing out.


http://www.undp.tj/home/MDG_NA_Full_Eng.pdf

http://www.imf.org/external/pubs/ft/scr/2006/cr0601.pdf

22


Bosnia and H

erzegovina

FYR Mace

donia

Serbia a

nd Monte

negro

Bulgaria

Russia

Romania

Kazakhsta

n

Albania

Uzbekist

an

Mold

ova

Armenia

Georgia

Tajikist

an

Kyrgyz R

epublic

Overall popualtion and children living underPPP $2.15 / day 2002-2003 (per cent)

80

70

60

50

40

30

20

10

0

all persons

children

Indeed, poverty among children has come down signif-icantly in the recent period. In 2002-3, an estimated 18million children were living in households where con-sumption was less than $2.15 per person per day, downfrom 32 million in 1998. Yet the problem of child pover-ty is far from solved. One in four children is currentlygrowing up in poverty. In all countries in the region, chil-dren are now the age group most at risk of poverty.Poverty experienced by children is increasingly happen-ing early in life. In a number of countries, poverty is con-centrating in certain groups of children. Furthermore, awider set of outcome indicators for children suggestthat gains on the income front are not being translatedinto child wellbeing.

Yet few countries have taken child poverty as a distincttarget of national policy. Most of the anti-poverty poli-cies in place are indirect and have not been developedto address those dimensions of poverty and deprivationthat are specific to children. Data on different dimen-sions of child poverty are rarely gathered systematically.

There is a need for a new focus on children, with betterunderstanding of the different dimensions of materialpoverty and deprivation as they affect children in thislatest phase of transition. There is also a need for specif-ic actions on child poverty to prevent such deprivationfrom becoming entrenched, passed on from one gener-ation to another.

The picture across the region

The 2006 Innocenti Social Monitor highlights the pat-tern of persisting child poverty across the region despite3-4 years of sustained economic growth. The findingsare striking:

� Child poverty has come down – but not uniformlyacross the region. Reductions are greatest in the richercountries of the region, The Russian Federation andUkraine, mirroring significant economic recovery. InCentral Asia and the Caucasus, poverty levels amongchildren remain very high (four out of five children inTajikistan live below the poverty line).

� Disparities within countries are marked. Capitalcities consistently have lower levels of child povertythan secondary cities or rural areas. In rural areas andsmall towns, the percentage of children in poverty canbe up to seven times that of the capital.

� In every country of the region, child poverty is high-er than that of poverty among adults. This is not a stan-dard relationship, however. Some countries have muchwider ‘gaps’ between adult and child poverty than oth-

ers. In Romania, the adult poverty level is 12 per cent;poverty among children is 9 per cent higher

� For the first time, we are seeing the emergence ofchronic poverty affecting children (over 50 per cent ofchildren in Kyrgyzstan experienced poverty for threeconsecutive years or more between 2002-2003).

One of the main characteristics of this poverty is the wayheightened vulnerability is associated with family struc-ture:

� Children in large families, or in single parent-headedfamilies, are at significantly higher risk of poverty.

� The relative risk of poverty is highest among youngchildren (under three years). This is a relative risk thathas increased in the most recent period.

Two factors are at work here. Demographic change isreducing the cohort of children, especially in theWestern CIS. Poverty outcomes, however, depend onboth the distribution of children across families and onthe effectiveness of state support for families. It is in thislast area that there are wide variations across the region.Benefits for children, child allowances and maternitybenefits differ across countries but are converging, los-ing ground in real terms, and have seen a shrinking oftheir coverage.

Children do not only experience poverty through theirfamily though. Children are dependent on services suchas education and health. Despite government expendi-ture ratios in the social sectors being maintained as apercentage of GDP, overall levels of investment inhealth and education have been low by internationalcomparison. While indicators of physical access and cov-erage remain high, quality is faltering. In education,TIMSS1 data suggest declining learning outcomes andincreasing disparities in learning outcomes for children.New barriers to health and other services for the poorare emerging, coinciding with the withdrawal of thestate as sole provider.


22000066 NNoobbeell PPeeaaccee LLaauurreeaattee MMuuhhaammmmaadd YYuunnuuss..MMiiccrrooffiinnaannccee hhaass aa lloonngg wwaayy ttoo ggoo iinn EEaasstteerrnnEEuurrooppee aanndd tthhee CCIISS.. SSeeee bbaacckk ppaaggee bbooxx..

Perhaps most disturbing are the numbers of childrengrowing up in public care. Rather than coming down ina period of economic growth, the trend is upward. Ratesof institutionalization of children (0-17 yrs) haveincreased between 1990 and 2004 in as many as 14 outof the 24 countries for which figures are available (threecountries with no figures). For Central and EasternEurope and the Commonwealth of Independent Statesoverall, the institutionalization rate has increased from422 to 448 per 100,000 children.

Economic growth in principle should expand opportuni-ties for households and families and in turn children. Ifthis is not happening and children are not benefitingfrom these opportunities, then something is amiss.

Bringing children into focus

There is little doubt that poverty reduction on the scalestill required in the region requires broad-based eco-nomic growth. This will require macroeconomic stabili-ty, tax reform to ensure revenue flows are sufficient, fis-cal prudence and reform of still overstaffed social servic-es.

Will this be enough to ensure much better outcomes forchildren? Continued economic growth will certainlycontribute, but is unlikely to be sufficient in itself.

Recovery has allowed social reform to come back ongovernment agendas. Too often, however, reform hasbeen limited to the easiest solutions rather than themore complex system change demanding the design ofnew structures and services. One of the risks as econom-ic growth continues is the reincarnation of earlier sys-tems and approaches without fresh ideas.

Far from being a marginal issue, tackling poverty anddeprivation among children needs to be given a muchsharper focus in macroeconomic and social policy.

There is no single policy action that alone will bringabout reductions in poverty and deprivation amongchildren. What is needed is a concerted mix of measuresthat address the adverse conditions that children expe-rience, especially early in life. These include tackling lowwages, and ensuring access to an agreed set of qualitybasic services. State support to the family ensuring fam-ily and maternity benefits are adequate to protect fami-lies from dire poverty need to be part of such a package.Interventions need to be linked to systemic reform ofthe social sectors. Data collection that can monitor dif-ferent dimensions of deprivation among children needsto guide policy.

The real battlefield is that of ideas. With economicrecovery, there is a need for a fresh debate on policiesfor children, what the data is saying, and where invest-ments for the future can be best made. Without suchrenewed discussion and dialogue, the risk of a furtherbacksliding in social achievements in the region shouldnot be discounted.

The costs of inaction are high.

Gordon Alexander ([email protected]) and PetraHoelscher ([email protected]) work for the UNICEFRegional Office for CEE, CIS and the Baltics in Geneva.

1. Trends in International Maths and Science Study.

Cornia, Giovanna Andrea ' The Case for a pro child/ pro youth growth and social policyin the European Economies in Transition' mimeo 2006.

Ovcharova, L.N. and Popova, D.O. ‘Child Poverty in the Russian Federation . AlarmingTrends and Policy Options ’ 2005 Independent Institute of Economic Policy, Moscow.

Social Monitor 2006 ‘Understanding Child Poverty in SEE and CIS Countries’ UNICEFInnocenti Research Centre, Florence, October 2006.

‘Understanding the Dynamics of Poverty and Development Risks of Children in Romania’2005 Institute for Quality of Life, Department of Sociology, University of Bucharest.

World Bank 2005 ‘Growth, Poverty and Inequality in Eastern Europe and the Former Sovi-et Union’.

Hoelscher P 2006 ‘”I want us to be a family again” Impact of migration on children in theCEE/CIS countries’ UNICEF Regional Office for CEE/CIS Social and Economic Policy forChildren Discussion Paper #5, Geneva

23



24

DECEMBER 2006 | issue 5DEVELOPMENT &TRANSITION

Development & Transition is published by the UnitedNations Development Programme and the LondonSchool of Economics and Political Sciencewww.developmentandtransition.net

EEddiittoorr: James Hughes: [email protected] EExxeeccuuttiivveeEEddiittoorr: Ben Slay DDeeppuuttyy EEddiittoorr Gwendolyn Sasse,MMaannaaggiinngg EEddiittoorr Horatio Mortimer AAddvviissoorryy BBooaarrdd :Nick Barr, Willem Buiter (chair), Stanislaw Gomulka, MaryKaldor, Dominic Lieven, Margot Light, Kate Mortimer

Houghton Street, London WC2A 2AE, UKwww.lse.ac.uk

UNDP Bratislava Regional CentreGrosslingova 35Bratislava 81109Slovakia

Tel: +421 2 59337 111Fax: +421 2 59337 450www.undp.org/europeandcis

Microfinance Compared with South Asia and Latin America, Europe and the CISis a latecomer to microfinance. The first microfinance institutions(MFIs) in the region appeared in the early 1990s when rising unem-ployment forced people into entrepreneurship. Data produced bythe Consultative Group to Assist the Poor (CGAP) show that micro-finance in Europe and CIS is now dominated by credit unions andNGOs offering loans. Loan sizes are typically larger than in otherregions, and partly for this reason a higher proportion of microfi-nance institutions have achieved self-sustainability. The region alsohas the highest level of per capita private sector investment inmicrofinance, accounting for 46 per cent of total private invest-ment in the global industry.

As CGAP acknowledges, microfinance globally tends to benefit thevulnerable non-poor and moderate poor, rather than the extreme-ly poor and destitute. Higher than average loan sizes combinedwith uneven geographical coverage suggest that this pattern iseven more pronounced in Europe and the CIS. A survey undertak-en by the Warsaw-based Microfinance Centre in 2004, for example,indicates that over half of all borrowers in the region are located inthe relatively prosperous upper-middle-income countries of Cen-tral and Eastern Europe, whereas Russia, the Western CIS and Cen-tral Asia achieve far lower levels of market penetration. 75 per centof borrowers throughout the region are located in relatively pros-perous urban areas, with the result that the rural poor remainunder-served. And to make matters worse, MFIs in the region areless cost-efficient, with loan officers managing much smaller port-folios.

Evidence from other regions suggests that the impact of microfi-nance on the poor is greatest where it is integrated into ‘main-stream’ financial services markets and growing competition leadsto innovation, product diversification and increased outreach. Forthis to happen in Europe and the CIS, governments — particularlyin the lagging transition countries — need to do far more to createstable and transparent regulatory and fiscal environments for MFIsand the wider emerging business communities that they aim toservice. This means, amongst other things, creating enabling legis-lation for the operation of MFIs, allowing them to borrow funds foron-lending and freeing potential borrowers from the burden ofcomplex — and frequently punitive — legal and administrativerequirements that increase their costs and undermine their prof-itability. Meanwhile, MFIs themselves need to take steps toimprove their efficiency and begin cooperating to build the infra-structure – the payments and clearing systems, information sys-tems and technical support services — that characterize moremature financial services markets.

Microfinance is not a panacea for poverty reduction, but in Europeand the CIS at least, its full potential has yet to be realized.

Jonathan Brooks [email protected]

FFOORRTTHHCCOOMMIINNGG CCOONNFFEERREENNCCEESSSub-Regional Workshop on the Anti-CorruptionPractitioners Network, Skopje, 21 – 22 November 2006The workshop will gather anti-corruption practitionersin Eastern European and CIS countries and inform themabout the newly created Anti-Corruption PractitionersNetwork (ACPN) and the related ACPN website. Mem-bers will be able to share information and practicalexperiences, look for solutions to similar problems in dif-ferent countries and solicit advice on specific topics.Contact: [email protected]

Regional Workshop on Social Enterprises in CEE andthe CIS, Brussels, December 11th and 12th, 2006.TheUNDP Bratislava Regional Center (UNDP-BRC) togetherwith the EMES European Research Network will convenethis workshop on completion of an ambitious projectentitled Study on Promoting the Role of Social Enterprisesin CEE and the CIS. Focus is on the capacity of socialenterprise to encounter an unmet demand through theproduction of a spectrum of social and community serv-ices. Attention is paid to their capacity to generate newemployment in the sectors where they specifically oper-ate, and through the integration of disadvantagedworkers. The aim on the one hand is to share the find-ings of the studies and discuss the practices that pro-mote the role of social enterprises in the target coun-tries, and on the other hand, to identify follow-up activ-ities and commitments. For more information, contactGeoff Prewitt [email protected].

‘Ensuring border security in the context of modernchallenges and threats to the system of internation-al security’ 28-30 November, Minsk. Part of theEU/UNDP joint programme ‘Enhancing Border Manage-ment in the Republic of Belarus’ (BOMBEL). The confer-ence examines the role of border security in internation-al security and the tasks facing the border protectionagencies in the changing world. Contact: AnnaChernyshova [email protected] next issue of Development and Transition willfocus on Conflict and post-conflict societies.

The editors welcome contributions. If you wish to sub-mit an article please follow the guidelines atwww.developmentandtransition.net




http://www.lse.ac.uk

http://www.undp.org/europeandcis





poverty and inequality

Documents

o s o v o s e r b

g e o r g

r u s b o s n

z e r b

n d h e r z e g o v

c e d o n

i t h u

i c f y r