This article was downloaded by: [Memorial University of Newfoundland] On: 02 August 2014, At: 11:01 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Journal of Comparative Social Welfare Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rjcs20 Income inequality and its driving forces in transitional countries: evidence from Armenia, Azerbaijan and Georgia Nazim Habibov a a School of Social Work , University of Windsor , Windsor , Ontario , Canada Published online: 17 Dec 2012. To cite this article: Nazim Habibov (2012) Income inequality and its driving forces in transitional countries: evidence from Armenia, Azerbaijan and Georgia, Journal of Comparative Social Welfare, 28:3, 209-221, DOI: 10.1080/17486831.2012.749504 To link to this article: http://dx.doi.org/10.1080/17486831.2012.749504 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions
Transcript
This article was downloaded by: [Memorial University of Newfoundland]On: 02 August 2014, At: 11:01Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Comparative Social WelfarePublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/rjcs20
Income inequality and its driving forcesin transitional countries: evidence fromArmenia, Azerbaijan and GeorgiaNazim Habibov aa School of Social Work , University of Windsor , Windsor ,Ontario , CanadaPublished online: 17 Dec 2012.
To cite this article: Nazim Habibov (2012) Income inequality and its driving forces in transitionalcountries: evidence from Armenia, Azerbaijan and Georgia, Journal of Comparative Social Welfare,28:3, 209-221, DOI: 10.1080/17486831.2012.749504
To link to this article: http://dx.doi.org/10.1080/17486831.2012.749504
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.
This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Income inequality and its driving forces in transitional countries:evidence from Armenia, Azerbaijan and Georgia
Nazim Habibov*
School of Social Work, University of Windsor, Windsor, Ontario, Canada
(Received 7 August 2012; final version received 11 November 2012)
The purpose of this study is to measure and compare income inequality and its drivingforces in the low-income countries of the Caucasus by drawing on micro-data fromnationally representative household surveys in Armenia, Georgia, and Azerbaijan.Inequality in the region of the Caucasus is very high. The Gini coefficient for theregions as a whole reached 55%. Azerbaijan has the lowest income inequality,followed by Armenia and Georgia. Among predictors, graduate and postgraduateeducation has the strongest positive effect on income in all countries. By contrast, thepositive effect of technical vocational education is relatively smaller and can beobserved only in Azerbaijan and Georgia. In addition to formal education, knowledgeof English and computers also has a separate positive effect in all countries. Anincrease in age, and therefore an increase in years of experience, has a low positiveimpact on the increase in income in all countries. By contrast, being a female has thestrongest negative effect on income across the region. Living in rural areas andreporting poor health is associated with having lower income.
Keywords: poverty; living standards; comparative social policy; social welfare
Introduction
The collapse of the communist system was one of the most important events of thesecond half of the twentieth century (Mohan, 2009). A spectacular rise in incomeinequality has become one of the most important features of the dramatic process ofpolitical, economic, and social transition experienced by the countries of the EasternUnion and the former Soviet Union since the end of the 1980s (Heyns, 2005; Milanovic,1999). Before transition, the state had an explicit objective of limiting inequality thatwas achieved through compressing labor income, using taxes and transfers to suppressincome differentials, providing universal social services, and setting the prices ofsocially important goods such as food, transportation, and housing below productioncosts to make them universally affordable (Habibov, 2010a, 2010b). As a result, inequalityin Soviet society was fairly low before transition (Falkingham, 2004; Kislitsyna, 2003).
The dismantling of the centrally planned economy in favor of a market economy led toabolishment of labor income control, generous transfers, universal service, and price regu-lation (Atkinson & Micklewright, 1992; Habibov, 2012a). Consequently, the income gapbetween the rich and the poor rose in all transitional countries since the 1990s. Since
Journal of Comparative Social WelfareVol. 28, No. 3, October 2012, 209–221
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
inequality was fairly low at the commencement of transition, a modest raise in income gapduring transition could have been expected (Aghion & Commander, 1999). By sharp con-trast, the raise in inequality in low-income transitional countries of the Caucasus, Armenia,Azerbaijan, and Georgia exceeded the worst expectation, with inequality increased dramati-cally for a short period of time (Habibov, 2011). To better appreciate the magnitude of therise, let us consider the overtime evolution of the Gini coefficient in these countries. TheGini coefficient is the most widely used measure of income inequality, which takes valueof zero in a situation of complete equality and takes a value of one in a situation of completeinequality (Cowell, 2000). Since the start of transition, the Gini coefficient jumped from0.30 to 0.50 in Georgia, from 0.27 to 0.51 in Azerbaijan, and from 0.26 to 0.49 inArmenia (Falkingham, 2005). In just a decade of transition, inequality doubled from acomparatively low level to a level higher than that in Latin America (Kislitsyna, 2003).
According to the World Bank estimations (Ersado, 2006), inequality continued togrow in the first half of the twentieth century, but decreased and stabilized by2006. Thus inequality in Azerbaijan peaked in 2002 with a Gini coefficient exceeding0.55, but decreased to 0.47 in 2003 (Ersado, 2006). Estimations in various surveys con-ducted between 2006 and 2008 with the Gini coefficient based on a consumption per-capita indicator, rather than income, demonstrated that the Gini coefficient stabilized atabout 0.30 in Azerbaijan, 0.40 in Georgia, and about 0.28 in Armenia (World Bank, 2010a).
Despite the magnitude of the problem, most studies on income inequality in transitionhave been conducted on the high-income and middle-income transitional countries (Heyns,2005). The problem of income inequality in the low-income countries of Central Asiaand the Caucasus has been given only cursory attention (Habibov, 2011). The gap inunderstanding of inequality in low-income transitional countries has been created by thereluctance of governments to provide data about income distribution and a lack of indepen-dent high-quality survey data on income inequality (Habibov & Fan, 2007; Micklewright &Marnie, 2005). For instance, government surveys are still not open for independentresearchers in Azerbaijan. As a result, the above-mentioned data about inequality havebeen gleaned from the so-called “grey literature”, the World Bank’s Poverty Reports con-ducted in the countries of the Caucasus (Falkingham, 2005).
These reports, however, have multiple limitations. They concentrated on poverty andnot on inequality per se and no determinants of inequality were fully explained. Second,the survey methodology such as sampling and questionnaire designs, as well as the meth-odologies of inequality estimations, varied between the countries. In addition, the surveyswere conducted in different countries at different points in time. The results therefore couldnot be robust across countries.
Given the scattered body of available research, the topic of income inequality in thelow-income transitional countries of the Caucasus region deemed further investigation inas much as reducing inequality being one of the main objectives of development(Mohan, 2008). Moreover, uncovering the driving forces of inequality is vital to identifyand prevent causes of social exclusion, which is another objective of development(Mohan, 2007). With this in mind, the objectives of this study are to estimate the magnitudeof inequality, identify its driving sources, and provide policy implications for inequalityreduction in Armenia, Azerbaijan, and Georgia.
Context of socio-economic development in the Caucasus during transition
The beginning of the transition happened against a background of the dissolution of theformer Soviet Union and the collapse of economic cooperation between its former
210 N. Habibov
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
republics. In the countries of the Caucasus, the beginning of the transition was furtheraggravated by political instability, civil unrests, ethnic conflicts, and wars. All togetherthese negative processes lead to a deep and prolonged economic crisis. The economicoutput fell sharply by the mid-1990s in all countries (Table 1, panel A). For instance,per-capita gross domestic product (GDP) of Georgia plummeted from US$4103 in 1990to US$1328 in 1994.
The situation begin to gradually improve after 1996, and by the beginning of the 2000sall countries of the Caucasus enjoyed accelerated rates of economic growth. In 2006,Georgia almost reached pre-transitional level of economic output, while Armenia and Azer-baijan exceeded it. The increased economic output allowed maintaining a relatively steadylevel of spending in real terms on public health and education, even though the level ofthese expenditures may decrease in nominal terms (Table 1, panels B and C). In fairness,it must be noted that the level of expenditure on public health and education in Armenia,Azerbaijan, and Georgia is still much lower than in other countries of the former SovietUnion (World Bank, 2009, 2010a, 2010b).
The rising economy and government expenditures in social areas led to significantpoverty reduction in all countries (Table 1, panel D). For example, the poverty rate inArmenia went down from 20% in 2001 to 3% in 2006. Non-monetary indicators of well-being also improved in all countries, as demonstrated by increased female and male lifeexpectancy at birth (Table 1, panels E and F). Let us now to turn to discussion of the meth-odology for our study.
Methodology
We commence with inequality measurement in the region as a whole and by countries. Tomeasure inequality we use total household income, which is reported in the CaucasusBarometer Survey (CBS) of 2007. The CBS is a cross-sectional household surveyconducted annually in Armenia, Azerbaijan, and Georgia to collect information aboutsociodemographics and income of population. Weights provided in the CBS render theresults nationally representative for each participating country.1 The survey questionnaire,data collection and archiving is standardized to ensure full comparability of CBS resultsacross the participating countries. Due to its high quality, the various annual rounds ofthe CBS have been actively used for comparative studies in the region (Habibov, 2011;Habibov & Afandi, 2009, 2011).
The 2006 round of the CBS reported household income in the form of a continuousvariable that allows Gini coefficient estimation. We adjusted it to the number of householdmembers using an equivalence scale. To create the adult equivalent income, we dividedtotal household income by the square root of the number of household members (Atkinson,Rainwater, & Smeeding, 1995; Gottschalk & Smeeding, 2000). Having the adult equivalentincome, we estimate the Gini coefficient for the region as a whole and each country separately.
The next step is to use multivariate regression to identify and measure the effect ofvarious individual, household, and community characteristics on income. The 2008round of the CBS is used for this step since this round collected information about the per-sonal income of respondents. To gauge the effect of various these characteristics on income,we follow several previous studies in the region (Anderson & Pomfret, 2002, 2003;Habibov, 2011; Habibov, 2012a) to estimate a human capital model in which individualincome is determined by respondent age, gender, education, foreign language and computerskills, marital and health status, and place of living. The descriptive statistics for indepen-dent variables by country are presented in Table 2.
Journal of Comparative Social Welfare 211
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
Table 1. Selected indicators of socioeconomic development.
Note: Data are rounded up. Max., maximum; min., minimum; SD, standard deviation.
Journalof
Com
parativeSocial
Welfare
213
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
Since the 2007 round of the CBS reported the household income, it is not possible toestimate what effect individual characteristics of a respondent, such as age, gender and edu-cation, might have on total household income. Hence, we have to use the 2008 round of theCBS that reports individual income of the respondent. This round asked respondents toindicate whether their monthly income excluding taxes belongs to one of seven incomebrackets (1 = “up to US$50”, 2 = “US$51–100”, 3 = “US$101–250”, 4 = “US$251–400”,5 = “US$401–800”, 6 = “US$801–1200”, and 7 = “more than US$1200”).
As shown, our outcome variable is categorical and ordered. We therefore used anordered logit regression model, where the outcome variable is the seven income bracketsand predictors are the human capital characteristics of the respondent. We report coeffi-cients, standard errors, and percentage change in odds that demonstrates the extent towhich the odds of being in higher income brackets change as a result of change in oneunit of a predictor variable.
Findings
Results of inequality measurement in the region and by country are reported in Table 3. Theinequality is high in the region as a whole and in all three countries under investigation. Theinequality in the region as a whole is high in as much as the Gini coefficient exceeds 54%.Across the countries, the highest level of inequality is recorded in Georgia where the Ginicoefficient is approximately 60%, while in equality in Armenia is relatively lower – theGini coefficient is about 51%. By contrast, Azerbaijan has the lowest inequality with aGini coefficient of about 41%.
Results of descriptive statistics on the income distribution brackets are reported inTable 4. The significant number of individual income fell in the range of “up to US$50”and “US$51–100” per month in the region as a whole and in each of the countries underinvestigation. Considerably fewer individuals reported their monthly income in the range
Table 3. Inequality measures.
Gini coefficient Standard error p value95% confidence
of US$101–400. Finally, very few respondents reported income of more than US$800 permonth.
Results of the ordered logit analysis on probability to be in one of the income bracketsare reported in Table 5. As can be observed, graduate and postgraduate education has thestrongest positive effect on income, holding all other variables constant. In Armenia andGeorgia, having graduate and postgraduate education increases the odds of being inhigher income brackets by 136% and 104%, respectively. The highest positive effect ofgraduate education can be observed in Azerbaijan – 261%.
By contrast, a positive effect of technical vocational education can be observed only inAzerbaijan and Georgia, where having this type of education increase increases the odds ofbeing in higher income brackets by 59% and 39%, respectively. No effect of technicalvocational education was found in Armenia. In comparison, completing secondaryschool has no effect on income in all three countries under investigation.
In addition to formal education, knowledge of English and computers also has a separ-ate positive effect. Being fluent in English increases the odds of being in higher incomebrackets by 37% in Georgia, 28% in Azerbaijan, and 100% in Georgia. Likewise, knowl-edge of computers increases the odds of being in higher income brackets by 58% inGeorgia, 63% in Azerbaijan, and 90% in Georgia.
Age also has a positive impact on the propensity of having higher income. In Armenia,the odds of having higher income increases by 2% with a one-year increase in the date ofbirth (being younger). The similar result can be observed for Azerbaijan (2%) and Georgia(3%). In fairness, it must be noted that the positive effect of age is relatively low.
By contrast, being female has the strongest negative effect on income, holding all othervariables constant. The odds of having higher income are 59% smaller for women than formen in Georgia, 67% in Armenia, and 75% in Azerbaijan.
Similarly, poor health negatively affects income. The odds of having higher income are37% smaller for people with poor health in Azerbaijan and 40% smaller for people withpoor health in Georgia, with no significant effect in Armenia. Likewise, living in ruralareas reduces the odds of having higher income in Armenia by 36%, with no significanteffect in Azerbaijan and Georgia. Lastly, being married has no significant effect onincome in all three countries under consideration.
At the end of Table 5, we report several measures of fit. The large negative value forpseudo-log likelihood demonstrates that a full model with independent variables hasmuch better fit as compared with a null model, which is the intercept-only model, in allcountries under investigation. Likewise, the significant likelihood ratio chi-square valueindicates that our models as a whole are statistically significant, as compared with thenull model with no predictors. In other words, taken together, our independent variableshave a significant effect on the outcome variable. Finally, we report McKelvey andZavoina’s R2 value that is the best available approximation for the classic R2 in themodels with ordered outcomes (Long & Freese, 2006). The reported R2 value demonstratesthat our model explains about 23% of variation in income in Azerbaijan, 17% in Armeniaand 14% in Georgia. These results are in line with the previous study in transitionalUkraine, Kazakhstan, and Tajikistan (Anderson & Pomfret, 2002, 2003, Brück, Danzer,Muravyev, & Wisshaar, 2010).
Discussion
The purpose of this study was to measure and compare income inequality and its drivingforces in the low-income countries of the Caucasus by drawing on micro-data from two
Journal of Comparative Social Welfare 215
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
Table 5. Results of multivariate analysis.
Armenia Azerbaijan Georgia
Independent variableCoefficient
(standard error)Percentage change
in oddsCoefficient
(standard error)Percentage
change in oddsCoefficient
(standard error)Percentage
change in odds
Date of birth 0.978*** –2.2 0.980*** –2.0 0.973*** –2.7(0.00311) (0.00295) (0.00296)
Pseudo-log likelihood –2643.57 –2848.32 –2526.55Likelihood ratio chi-square (10) 302.24*** 453.59*** 233.14***McKelvey and Zavoina’s R2 0.170 0.225 0.140
Note: Data are rounded up. Percentage change in odds for unit increase is an independent variable. Significance: *p < 0.05, **p < 0.01, ***p < 0.001.
216N.Habibov
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
rounds of nationally representative household surveys in Armenia, Georgia, and Azerbai-jan. The Gini coefficient was used to measure and compare the level of inequality, whileordered probit regression was used to identify the driving forces of gaining higherincome. The empirical findings presented in this paper allow us to draw several interestingconclusions.
First, we found that inequality in the Caucasus region is very high. The Gini coefficientfor the regions as a whole reached 55%. This finding is in line with results reported by pre-vious studies conducted in the 1990s, as summarized by Falkingham (2005). Hence, despitesignificant economic growth experienced by the countries in the region at the beginning of1990s, the level of income inequality continues to remain high. This finding demonstratesthat the results of rapid economic development were not equally shared by all groups of thepopulation.
Second, the income inequality varies considerably between countries of the region.Azerbaijan has significantly lower income inequality than Armenia and Georgia. Ersado(2006) demonstrated that the more generous and better administrated income securitysystem in oil-reach Azerbaijan considerably reduced inequality in the country. Indeed,Habibov and Fan (2007) found that approximately 82% of poor households receivedsocial transfers and the share of social transfers in total consumption of poor householdsexceeds 80%. Likewise, Habibov and Fan (2006) showed that, even without social assist-ance, for one of the smallest social programs in the country the Gini coefficient will increaseby 2.5%, while Ersado (2006) estimated that if all social transfers were eliminated theninequality would skyrocket by 59–67%.
Third, increase in age, and therefore increase in years of experience, has very lowimpact on the increase in income, whereas graduate and postgraduate education has thestrongest positive effect on income. It appears that younger individuals who just graduatefrom colleges and universities are able to obtain higher income. At the same time, the yearsof experience of the older workers are not valued very much. This phenomenon can beexplained by the interplay of two factors. On the one hand, younger and older householdsuse distinct income-generation strategies to adapt to transitional shocks (Alexeev & Kaga-novich, 1998). Older individuals who received their education and experienced longer yearsof experience during the Soviet period are less likely to drastically switch to more profitablebut riskier strategies such as opening own small business or becoming a private farmer. Incomparison, younger individuals who usually lack long years of occupation-specific experi-ence are more likely to be engaged in recently emerging riskier but more profitable strat-egies. An alternative explanation of the negative effect of age might be that pensionersexperience a drop in income since their pension could be lower than their labor-marketincome. On the other hand, the low impact of years of experience on income is causedby the rapid depreciation of human capital after graduation from educational institutionsand an inability to update it in older age. Although most of employees would like toreceive further professional development and training while remaining on the job, nosystem of effective continuing education has been developed in the region (Habibov,2010b). The strong positive effect that continuing education would have on income is sup-ported by our findings which indicate that short-term programs outside multiyear formaleducation, such as foreign-language or computer-literacy courses, are associated withobtaining higher income.
Fourth, the impact of graduate and postgraduate education should not be underesti-mated. Having higher education has the strongest positive effect on income in all countriesunder investigation. Hence, ensuring access to education for the poorer should be instru-mental from the perspectives of poverty and inequality reduction. However, the transitional
Journal of Comparative Social Welfare 217
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
countries suffer from a widening gap in access to education under which the poorer havemuch less access to educational programs than the wealthier. This gap begins with theunequal access to early childhood care and education programs (Habibov, 2012b) and con-tinues into higher education (Micklewright, 1999).
Fifth, our study revealed a strong gender gap in income. In all countries under investi-gation, women receive lower income than men. These phenomena in transitional countriesof the region can be explained by the interplay of several factors, as discussed by Habibov(2012a). The lack of labor-market regulation enforcement leads to a situation whereemployers explicitly avoid hiring a female who is more likely to go on maternity leaveor sick leave to care for children. The lack of affordable early childhood care and educationinstitutions forces many women to stay at home with children. If a woman works, she ismost likely to be employed in government sectors such as education or healthcare wherewages are low. However, even in these sectors the number of women in better-paid leader-ship positions is currently decreasing. Likewise, women are mostly excluded from partici-pation in the more profitable new economic activities. In Azerbaijan, for instance, onlyabout 7% of newly privatized units belong to women (Asian Development Bank, 2005).
Sixth, the rural–urban gap in income, which is observed in Armenia, is also important.This can be explained by the fact that agriculture in Armenia was reduced to a subsistencelevel. Most farmers were allocated very small parcels of land that were barely sufficient tofeed the family, especially without access to seeds, fertilizers, irrigation, and agriculturalmachinery and research. Often, having very limited surplus, farmers have to be involvedin barter without using cash or another type of wage income. This is not to say that subsis-tence farming does not exist in Azerbaijan or Georgia, but that subsistence farming appearsto be more pronounced in Armenia, where living in rural areas has a significant negativeeffect on income. The difference between Armenia and neighboring countries could becaused by different strategies of rural reforms undertaken after the collapse of the collectivefarms of the Soviet era. For example, Azerbaijan opted out for a more drastic and completeagricultural reform. The country privatized 95% of all arable land with only 5% remainingin state land reserve, while in Armenia the state land reserve reached about 15% of all arableland (Giovarelli & Bledsoe, 2004). Similarly, more households in Azerbaijan benefitedfrom the land reform process than in other transitional countries, and the level of satisfac-tion with agricultural reform is the highest among other transitional countries (Dudwick,Fock, & Sedik, 2005).
In general, however, the origins of the rural–urban gap are associated with the lessdeveloped infrastructure and hence the lack of higher paid jobs in the rural regions. In thecountries of the Caucasus, the lack of infrastructure such as a lack of high-quality roadsand interruption of electricity and natural gas supplies are particularly important in explainingthe rural–urban gap in income (Lokshin & Yemtsov, 2004; World Bank, 2005). On the otherhand, the example of transitional China shows that public investments in rural education andagricultural development are essential in reducing this gap (Zhang & Fan, 2004).
Finally, the economic recession of 2008/09, which happened after the data used in thispaper were collected, could affect the inequality in the region. The available data indicatethat Georgia experienced sharp economic shocks caused by war with Russia for the SouthOssetia and global economic recession. According to the World Bank (2011a), between2003 and 2007 Georgian GDP grew by average 9% per annum. In 2007, GDP growthexceeded 12%. However, in 2008, GDP growth plummeted to 2.3% and became negative(–3.8%) in 2009. These shocks reversed the gains made during the years preceding thecrises, with the poverty rate increasing from 22.7% in 2008 to 24.7% in 2009 (WorldBank, 2011a).
218 N. Habibov
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
Armenia was hit even harder. In fact, according to the World Bank: “Armenia was oneof the hardest hit countries by the 2008–09 global economic crisis” (2010b, p. vi). Arme-nian GDP growth became negative in 2008 and exceeded –18% in 2009. About 40% ofhouseholds in the country suffered the direct impact of the crisis through loss of income,with poverty and extreme poverty increasing by about three percentage point from 2008to 2009 (World Bank, 2010b). Moreover, consumption inequality increased by 6%(World Bank, 2011b).
By sharp contrast, Azerbaijan was much less affected by the crisis due to high oil prices.Nevertheless, a World Bank (2010a) simulation projected that poverty incidence couldincrease by approximately one percentage point, meaning about 86,000 people couldbecome poor as a result of the global economic crisis.
Conclusion and implications
In this study, we have demonstrated that inequality remained high in low-income transi-tional countries of the Caucasus. From a practical perspective, the findings of this studyallow us to suggest several practical measures to counterbalance growing income inequalityin low-income transitional countries. Equal access to education is the key to ensure that thelow-income individuals will be able to increase their human capital and gain higher income.Special attention should be paid to access for short-term education (e.g. foreign-languagecourses and computer classes), which could be a relatively faster and inexpensive way toupgrade human capital. Gender mainstreaming is urgently required for the current govern-ment socioeconomic policies to reduce a gender income gap. The strengthening socialsecurity system will also significantly reduce the existing inequality. Finally, the invest-ments in rural development will help to reduce the rural–urban gradient in income.
From a theoretical perspective, the findings of this study suggest that the increase ininequality in low-income transitional countries in 2006 is mainly driven by a mixture ofhuman capital factors (e.g. access to higher education) and institutional factors (e.g.social security or lack of gender mainstreaming).
Note1. Other available sources of comparable information include the Life in Transition Survey and the
European Values Studies, but their country samples are smaller than those of the CBS.
Notes on contributor
Nazim Habibov is Associate Professor in School of Social Work University of Windsor. He focuses oncomparative and cross-national social welfare policies and practice; emerging trends and issues. Inparticular, he is interested in issues of poverty and inequality including determinants of povertyand inequality, income and subjective measures of poverty and inequality, as well as public perceptionof causes of poverty and public attitude to welfare state. He also focuses on effectiveness of publicservices such as health care and early childhood care and education. Dr. Habibov is author of morethan 30 peer-reviewed research articles on these topics.
AcknowledgementsThis research was funded by the Social Sciences and Humanities Research Council of CanadaStandard Research Grant. Usual disclaimer is applied.
Journal of Comparative Social Welfare 219
Dow
nloa
ded
by [
Mem
oria
l Uni
vers
ity o
f N
ewfo
undl
and]
at 1
1:01
02
Aug
ust 2
014
ReferencesAghion, P., & Commander, S. (1999). On the dynamics of inequality in transition. Economics in
Transition, 7, 275–298.Alexeev, P., & Kaganovich, M. (1998). Return of human capital under uncertain reform: Good guys
finish last. Journal of Economic Behavior and Organization, 37, 53–70.Anderson, K., & Pomfret, R. (2002). Relative living standards in new market economies: Evidence
from Central Asian household surveys. Journal of Comparative Economics, 30, 683–708.Anderson, K., & Pomfret, R. (2003). Consequences of creating a market economy. Evidence from
household surveys in Central Asia. Northampton, MA: Edward Elgar.Asian Development Bank. (2005). Azerbaijan. Country gender assessment. Manila: Author.Atkinson, A., & Micklewright, J. (1992). Economic transformation in Eastern Europe and the distri-
bution of income. Cambridge: Cambridge University Press.Atkinson, A., Rainwater, L., & Smeeding, T. (1995). Income distribution in OECD countries: the evi-
dence from Luxemburg Income Study. Paris: Organization of Economic Cooperation andDevelopment.
Brück, T., Danzer, A., Muravyev, A., & Weisshaar, N. (2010). Poverty during transition: Householdsurvey evidence from Ukraine. Journal of Comparative Economics, 38, 123–145.
Cowell, F. (2000). Measurement of inequality. In A. Atkinson & F. Bourguignon (Eds.), Handbook ofincome distribution (pp. 87–150). Amsterdam: Elsevier Science.
Dudwick, N., Fock, K., & Sedik, D. (2005). A stocktaking of land reform and farm restructuring inBulgaria, Moldova, Azerbaijan and Kazakhstan. Washington, DC: World Bank.
Ersado, L. (2006). Azerbaijan’s household survey data: Explaining why inequality is so low? (WorldBank Policy Research Working Paper 4009). Washington, DC: The World Bank.
Falkingham, J. (2004). Inequality and poverty in the CIS-7 countries, 1989–2002. In C. Shiells & S.Sattar (Eds.), The low-income countries of Commonwealth of Independent States. Progress andchallenges in transition (pp. 141–170). Washington, DC: The International Monetary Fund.
Falkingham, J. (2005). The end of the rollercoaster? Growth, inequality and poverty in Central Asiaand the Caucasus. Social Policy and Administration, 39(4), 340–360.
Giovarelli, R., & Bledsoe, R. (2004). Land reform in Eastern Europe–Western CIS, Transcaucus,Balkans, and EU accession countries. Food and Agriculture Organization of the UN.Retrieved from www.fao.org/DOCREP/007/AD878E/AD878E03.htm#P322_56641/
Gottschalk, P., & Smeeding, T. (2000). Empirical evidence on income inequality inindustrialized countries. In A. Atkinson & F. Bourguignon (Eds.), Handbook of income distri-bution (pp. 261–307). Amsterdam: Elsevier Science.
Habibov, N. (2010a). An intertemporal evolution of inequality in Azerbaijan: 1995–2002. Problemsof Economic Transition, 52(9), 51–77.
Habibov, N. (2010b). Understanding the over-time evolution of living standard determinants intransitional countries: Evidence from Azerbaijan. Journal of Comparative Social Welfare, 26(1), 43–63.
Habibov, N. (2011). Self-perceived social stratification in low-income transitional countries: Micro-data evidence from Armenia, Azerbaijan and Georgia. International Journal of SocialEconomics, 38(1), 5–22.
Habibov, N. (2012a). How and why determinants of household welfare changed in Azerbaijan duringthe transition: Lessons from national surveys. Problems of Economic Transition, 54(11), 3–52.
Habibov, N. (2012b). Early childhood care and education attendance in Central Asia. Children &Youth Services Review, 34(4), 798–806.
Habibov, N., & Afandi, E. (2009). Analysis of subjective well-being in low-income transitionalcountries: Evidence from Armenia, Azerbaijan and Georgia. Journal of Comparative SocialWelfare, 25(3), 203–219.
Habibov, N., & Afandi, E. (2011). Self-rated health and social capital in transitional countries:Multilevel analysis of comparative surveys in Armenia, Azerbaijan, and Georgia. SocialScience & Medicine, 72(7), 1193–1204.
Habibov. N., & Fan, L. (2006). Social assistance and the challenges of poverty and inequalityin Azerbaijan, a low-income country in transition. Journal of Sociology and Social Welfare, 33(1), 203–226.
Habibov, N., & Fan, L. (2007). Social protection and poverty in Azerbaijan, a low-income country intransition: Implications of a household survey. International Social Security Review, 60(4),47–68.
Heyns, B. (2005). Emerging inequalities in Central and Eastern Europe. Annual Review of Sociology,31, 163–197.
Kislitsyna, O. (2003). Income inequality in Russia during transition. How it can be explained?(Economic Education and Research Consortium Working Paper Series No. 03/08). EERC:Moscow.
Lokshin, M., & Yemtsov, R. (2004b). Combining longitudinal household and community surveys forevaluation of social transfers: Infrastructure rehabilitation projects in rural Georgia. Journal ofHuman Development, 5(2), 265–277.
Long, S., & Freese, J. (2006). Regression models for categorical dependent variables using stata(2nd ed.). College Station, TX: Stata Press.
Micklewright, J. (1999). Education, inequality and transition. Economics of Transition, 7(2),343–376.
Micklewright, J., & Marnie, S. (2005). Targeting social assistance in a transition economy: TheMahallas in Uzbekistan. Social Policy and Administration, 39, 431–447.
Milanovic, B. (1999). Explaining the increase in inequality during the transition. Economics ofTransition, 7(2), 299–341.
Mohan, B. (2007). Social exclusions: Challenges for new social development. Journal ofComparative Social Welfare, 23(1), 69–79.
Mohan, B. (2008). The development delusion. Journal of Comparative Social Welfare, 24(1), 83–88.Mohan, B. (2009). The fall and rise of the Third World. Journal of Comparative Social Welfare, 25(1),
August 5, 2012, from http://www.transmonee.orgWorld Bank. (n.d.). World development indicators. Washington, DC: The World Bank.World Bank. (2005). Azerbaijan. Issues and options associated with energy sector reforms.
Washington, DC: The World Bank.World Bank. (2009). Georgia poverty assessment. Washington, DC: World Bank.World Bank. (2010a). Azerbaijan living conditions assessment report. Washington, DC: World Bank.World Bank. (2010b). Armenia. The 2008–09 global economic crisis, policy responses, and house-
hold coping strategies. Washington, DC: World Bank.World Bank. (2011a). Georgia poverty dynamics, 2003–2010. Washington, DC: World Bank.World Bank. (2011b). Armenia Poverty update using revised poverty lines. Washington, DC: World
Bank.Zhang, X., & Fan, S. (2004). Public investment and regional inequality in rural China. Agricultural
