056 056 This paper is a product of the Poverty and Equity Global Practice Group. It is part of a larger effort by the World Bank to provide open access to its research and contribute to development policy discussions around the world. The authors may be contacted at jazevedo@worldbank.org or jyang4@worldbank.org or oinan@worldbank.org. The Poverty & Equity Global Practice Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. ‒ Poverty & Equity Global Practice Knowledge Management & Learning Team This paper is co-published with the World Bank Policy Research Working Papers. How Equitable Is Access to Finance in Turkey? Evidence from the Latest Global FINDEX Joao Pedro Azevedo Osman Kaan Inan Judy S. Yang The World Bank1 JEL Codes: D14, D63, E2, G21, I3 Keywords: Access to Finance, Equity, Human Opportunity Index, Tukey, Benchmarking, Poverty This paper has benefitted from comments by participants from presentations made in World Bank Washington, DC and Ankara offices, more specifically from the Poverty and Equity Global Practice’s Europe and Central Asia Team and the Turkey Country Office as well as inputs by our colleagues in the Finance and Markets Global Practice. The team thanks Martin Raiser for his guidance and support. We are also thankful for comments and assistance received from Alper Ahmet Oguz, Ilias Skamnelos, Jose Montes, and Minh C. Nguyen. The team is grateful to Leora Klapper and the Global FINDEX team for their assistance with the FINDEX dataset. The usual disclaimer applies. This paper is a product of the FY2015 Turkey Poverty Team in the World Bank’s Poverty and Equity Global Practice. 1Contacts: Joao Pedro Azevedo (Lead Economist, jazevedo@worldbank.org); Judy S. Yang (ET Consultant, jyang4@worldbank.org); Osman Kaan Inan (Junior Professional Associate, oinan@worldbank.org) This paper has benefitted from comments by participants from presentations made in World Bank Washington, DC and Ankara offices, more specifically from the Poverty and Equity Global Practice’s Europe and Central Asia Team and the Turkey Country Office as well as inputs by our colleagues in the Finance and Markets Global Practice. The team thanks Martin Raiser for his guidance and support. We are also thankful for comments and assistance received from Alper Ahmet Oguz, Ilias Skamnelos, Jose Montes, and Minh C. Nguyen. The team is grateful to Leora Klapper and the Global FINDEX team for their assistance with the FINDEX dataset. The usual disclaimer applies. This paper is a product of the FY2015 Turkey Poverty Team in the World Bank’s Poverty and Equity Global Practice. 1. INTRODUCTION Access to finance is an important tool against poverty, since it allows for the smoothing of consumption, savings, management of money, and loans for purchases. Universal access to finance is also a development goal by 2020. With about 2.5 billion unbanked adults in the world, the World Bank put forward a vision of universal financial inclusion that can be achieved through affordable services and innovation. In the case of Turkey, this note focuses on four financial indicators: Bank Account Use, Savings, Debit and Credit Card Use, and Borrowing. As an upper-middle income country, Turkey has a very low level of savings but high levels of borrowing and credit use. However, it is not only important to understand the coverage of financial indicators in the population, but also the distribution of these characteristics across sub-populations. Is one group of the population just as likely to have a bank account as another? Which groups are the least likely to be financially included? In that sense, achieving rigorous financial inclusion can only be possible with an equitable allocation of financial resources and tools across different economic, social and demographic groups in the country. There is a large existing literature on the role of access to finance in growth generation and inequality reduction. Beck, Demirguc-Kunt, and Levine (2007) find that 60 percent of the income growth of the poorest quintile is due to financial development’s impact on the economy. Claessens and Perotti (2007) explain how financial development can be related to inequality, theorizing that in unequal societies, unequal financial access is a result of skewed political influence and regulatory capture. Johnston and Murdoch (2008) build on the notion of unequal financial access as they empirically demonstrate that only 10 percent of the poor take loans even though around 40 percent of them are creditworthy. They argue that this misallocation is caused because the sizes of loans requested by the poor are not large enough to be profitable for the lenders suggesting that the costs associated with this process must be reduced in order to improve financial access. McKenzie and Woodruff (2008) offer a method to efficiently improve access to finance as they show that an increase in access to finance has the highest returns for the financially “super- constrained” micro-sized firms, those that have never taken formal loans or supplier credit. This paper will analyze the relationship between financial inclusion and poverty reduction in Turkey while examining the distribution of financial coverage across different groups in the society. A brief overview of Turkey’s poverty reduction and economic mobility performance will be followed by an analysis of access to finance across different segments of the population. The paper will conclude with suggestions on how to best address challenges concerning availability of financial resources in Turkey in order to enhance shared prosperity and reduction of poverty. 2 2. POVERTY TRENDS Between 2002 and 2011, Turkey had positive performance in poverty reduction and economic mobility. The strong performance in poverty reduction led to a shrinking of the number of poor from 43 to 22 percent (Figure 1). The reduction in the level of poverty was coupled with a significant expansion of the middle class during the same period. The size of the middle-class doubled from 20 to 40 percent in 9 years. One exception was in the aftermath of the 2007 financial crisis when poverty levels experienced a marginal increase. By 2009, Turkey was able to once again establish a positive trend in poverty reduction. Figure 1. Economic Mobility in Turkey during 2002-2011 100 90 MIDDLE CLASS 80 70 60 50 VULNERABLE 40 30 20 10 POOR 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Notes: The poor live below the $5/day PPP per person poverty line. The vulnerable class live between the $5-$10/day PPP per person line, and the middle-class live above the $10/day line. Source: Azevedo and Atamanov (2014) The main driver of poverty reduction has been the labor markets, as both the share of employees and wage levels increased significantly. Overall income poverty declined by 33.7 percent between 2002 and 2011 (Figure 2). The labor market was responsible for 18.7 percentage points of this reduction with the improved quality of jobs (11.6 percentage points) and the added worker effect (7.1 percentage points). Social protection was also crucial in this process as pensions (7.1 percentage points) and social assistance (3.3 percentage points) facilitated the reduction of poverty. In additional to its direct effect, social protection also reduced the depth and severity of poverty. Figure 2. Components of Poverty Reduction in Turkey during 2002-2011 2 0.1 1.3 poverty reduction, percentage points 0 -2 -0.1 -4 -3.3 -3.3 -6 -8 -7.1 -7.1 -10 -12 -11.6 -14 share of share of wages social pensions remittances agricultural other adults employed assistance income income Notes: The poverty line is $5/day PPP per person Source: Azevedo and Atamanov (2014) 3 In addition to moving people out of poverty, Turkey was successful in preventing downward economic mobility. From 2002 to 2011, 41 percent of the poor moved to the vulnerable group while 40 percent of the vulnerable group moved into the middle class (Figure 3). On the other hand, only 2 percent of the middle class fell into the vulnerable group and 1 percent of the vulnerable group descended to poverty. Turkey was able to establish a stable environment for the middle class and the vulnerable while providing upward mobility for the poor. Figure 3. Protection of the Vulnerable Group, 2002-2011 Origin Percentage moving to 2011 (In 2002) Poor Vulnerable Middle Class Poor 43 58 41 0 100 Vulnerable 37 1 58 40 100 Middle class 20 0 2 98 100 Total 100 22 38 40 100 Source: Azevedo and Atamanov (2014) Notes: transition matrix is based on synthetic panel for 2002-2011. Welfare aggregate is consumption (+health, +durables) per capita and poverty line is 5 and 10 USD PPP 2005. Rsq for consumption model is 0.36. Explanatory variables include year of birth cohort, number of children, education of the head of household, rural/urban dummy and different interactions between these variables. Sample: head of households 25-55 years of age. Values based on lower bound estimates. 3. METHODOLOGY Four financial indicators are examined: Bank Account Usage, Savings, Debit and Credit Card Usage, and Borrowing. While savings rates are low in Turkey, the use of bank accounts, credit, borrowing, and informal lending is high. In terms of the trends in financial inclusion, savings patterns increased while use of credit and debit cards diminished between 2011 and 2014. An Equity Adjusted Coverage Rate2 (EACR) in these four indicators is calculated for Turkey and compared to other countries across the world. The EACR index is a measure of the coverage of financial services adjusted for equity in the use of these services across `circumstances’. The characteristics or circumstances that are accounted for are gender, age, education, income quintile, and urban/rural. Differences in individual and household level data also yield interesting results, since characteristics at the individual-level are more descriptive of the population rather than characteristics of the head of household. Using the EACR index, three types of analyses are presented: 1. The dissimilarity index in financial indicators is decomposed into composition, equalization, and scale effects. The dissimilarity-index is a measure of dissimilar coverage rates across groups (gender, age, education, and income). At the individual level, gender accounts for large dissimilarity differences in Bank Account use as well as Debit and Credit Card usage. However, gender effects disappear when using household level data since characteristics are the head of household. 2. Turkey’s EACR index and its components are compared to nearest neighbor countries in terms of coverage rates. For example, in Bank Account Ownership, Turkey’s coverage rate is most similar to Brazil, Bosnia and Herzegovina, Italy, and the United Arab Emirates out of 145 countries. 2 See the appendix for an explanation of the methods used in this paper: the Equity Adjusted Coverage rate (EACR) calculation, calculation of components in differences in the EAC, and Shapley decomposition. 4 3. The third computation conducted is attributing the differences in EACR between Turkey and nearest neighbor countries to composition, scale, and equalization. For example, bank account usage between Turkey and Italy may differ either because of the characteristics of individuals who hold bank accounts (composition), inequity between groups (equalization), or general coverage (scale). The paper presents only a selection of graphics and readers are encouraged to use the interactive Financial Inclusion Benchmarking Dashboard.3 The dashboard allows the user to explore more financial indicators that have been compiled using other data sets. The interactive feature allows the user to customize comparisons of Turkey’s financial inclusion EACR measures with a larger set of countries. Illustration of Equity Adjusted Coverage Rate – Bank Accounts The figure below illustrates the calculation and intuition for the Equity Adjusted Coverage Rate (EACR) for Bank Accounts in Turkey. The average Bank Account coverage rate is 57.6 percent (the red line). However, coverage rates vary by groups. In this case, there are 51 unique groups based on age, gender, education, and income. The coverage rates for these groups range from 0 to 100 percent and are illustrated by the area in blue. Penalty (P) Coverage Rate (C) Source: FINDEX 2011 The Equity Adjusted Coverage Rate is equal to the average coverage rate (C) minus a penalty for the inequity in distribution. This inequity is equal to the area (P), where certain groups have coverage rates that are much higher than the average and hence illustrates an inequity in the distribution of Bank Account Usage. A dissimilarity index (D) can also be computed to quantify this inequity: 25.0 The Formula for the EACR = ∗ 1 57.6 ∗ 1 0.237 44.0 Note: Mathematical formula based on the Human Opportunities Index construction of Molinas Vega et al. (2010) 3 See Appendix for user notes. URL: http://dataviz.worldbank.org/t/ECA/views/tur_finance_equity_poverty7/Dshnew?:embed=y&:display_count=no#3 5 4. ACCESS TO FINANCE INDICATORS This section profiles four financial indicators: Bank Account Use, Savings, Debit and Credit Card Use, and Borrowing. Each section profiles the level of use across demographic characteristics and then discusses the Equity Adjusted Coverage Rate of each indicator. i. BANK ACCOUNT USE In terms of access to finance, there are four main factors that influence the level of banking in Turkey. These primary factors that affect the rate of banked adults in Turkey are income, gender, education and age. Turkey, overall, has a high proportion of formally banked adults, but account penetration varies significantly across these characteristics. A considerable portion of the poor lack a safe and secure place to store money because of restrictive bank fees and fines. Since transaction costs have substantially larger impact on the poor, these expenses may pressure them to use more channels of informal financing. Informal financing may be more unreliable and have higher interest rates, which would have negative impact on the poor. Moreover, the poor also face challenges in providing collateral for loans, which disincentives participation in the formal banking sector. From the banks’ perspective, low loan to deposit ratios further deteriorate the situation by perpetuating a cycle of low lending. The negative effects of the restrictive measures in the Turkish banking sector can be observed in account usage rates across different income quintiles (Figure 4). Account use in Turkey in 2011 is significantly higher than rest of the ECA region and slightly higher than the BRICS. However, usage rates are lower than the level in the developed world. More importantly, the difference in account use between the lowest and highest income quintiles in Turkey is similar to ECA and BRICS economies, but is considerably larger than the difference in the developed world. This relatively large difference in comparison to high-income countries indicates that Turkey has to not only increase the overall scale of account usage, but also aim to equalize account penetration among different income levels to boost financial inclusion. Between 2011 and 2014, equality in access to bank accounts between different income groups improved in Turkey despite the lack of an overall increase in account usage. Bank account usage remained stable over the 3 year period. On the other hand, BRICS and other developing countries experienced increases in access to bank accounts. As a result, Turkey’s bank account usage rate was less than the BRICS average in 2014. Within Turkey, account penetration among individuals in the lowest income quintile increased substantially from 42 percent to 52 percent. However, account usage in the highest two income quintiles fell considerably, which neutralized the positive trend from the low income individuals. These two opposite trends resulted in a more equalized level of bank account ownership in 2014. 6 Figure 4. Proportion of Adults with an Account at a Formal Financial Institution 2011 2014 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Rest of ECA Turkey BRIC Rest of High Rest of Developing Income Developing Income ECA World World Lowest quintile q2 q3 q4 top quintile Source: FINDEX 2011 and 2014 The largest gaps in access to finance are due to differences by gender in Turkey. Females have a 49 percentage point lower proportion of having an account at a formal institution in Turkey (Figure 5). This disparity is over five times larger than the difference in the BRICS countries. The discrepancy in comparison with rest of ECA is even more alarming, as males and females use accounts at an almost identical rate. This difference in account usage in Turkey may be resulting from historical social norms, low female labor force participation, and a relatively lower female education level in the country. The disparity between males and females in terms of account usage decreased between 2011 and 2014. Male use of bank accounts diminished from 82 percent to 69 percent, while female usage increased from 33 to 44 percent. Despite Turkey’s success in expanding access to accounts among women, the gender discrepancy is still significantly larger when compared with the BRICS, ECA and other developing countries. Going forward, specifying the policies that resulted in the improved access to banking for females and ensuring their sustainability can help Turkey in catching up to benchmark country groups. Figure 5. Account Use in Turkey, by Gender 2011 2014 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High_Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World male female Source: FINDEX 2011 and 2014 The level of education and age groups also affect account use in Turkey. Remarkably, 99 percent of individuals with tertiary education have an account at a formal financial institution, which exceeds the 7 developed world’s rate of 94 percent (Figure 6). However, there is large drop off to 58 percent when individuals with secondary education are considered. This 42 percent fall-off is larger than any other change in the BRICS, ECA or the developed world. However the account use levels of Turkey for both individuals with secondary and primary education is still higher than their counterparts in the BRICS economies and other ECA countries. Between 2011 and 2014, there is a fall in the account use of individuals with tertiary education in Turkey. Usage of accounts falls from 99 percent to 87 percent in 2014 amongst individuals with tertiary education. However, it is important to note that 87% is still a high level of bank usage when compared to the rest of ECA, in which 80 percent of individuals in the same cohort have an account at a financial institution. Figure 6. Account use by Education 2011 2014 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Turkey BRIC Developing High Income Rest of ECA Turkey BRIC Developing High Income Rest of ECA Primary Secondary Tertiary Source: FINDEX 2011 and 2014 Age also plays a role in account use across Turkey. Only 44 percent of 15-24 year olds have an account compared to 62 percent of 25-64 year olds and 67 percent of individuals over 65 (Figure 7). Interestingly, in none of the other compared regions do persons over 65 years old have the highest rate of account use while Turkey seems to have a direct correlation between age and account use. Changes in account ownership between 2011 and 2014 diverge between different age groups in Turkey. Individuals over 65 years of age experience an increase while the two younger cohorts have a lower level of account usage in 2014. When compared to BRICS economies in 2014, younger cohorts use banks at a lower rate while the elderly are more likely to own accounts in Turkey. Figure 7. Account use, by Age Groups 2011 2014 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World 15-24 25-64 65+ 15-24 25-64 65+ Source: FINDEX 2011 and 2014 8 A measure of an Equity Adjusted Coverage Rate (EACR) in bank account usage informs on the coverage of bank account use with an adjustment on the equity of coverage. Using the FINDEX database, 57.6 percent of individuals (aged 15+) in Turkey had a bank account (coverage rate) in 2011. However, when adjusted for equity, the EACR rate is 44.0 percent. Ranked against all countries, Turkey ranks in the top half of countries in terms of bank account use coverage (See Figure 22 for graph of all 145 countries in the FINDEX database). In 2014, Turkey’s coverage rate decreases slightly but the EACR actually increases to 48.3 because of the falling level of dissimilarity. On the other hand three of the four benchmark countries experience significance increases in their coverage rates and substantial falls in their dissimilarity scores. Countries such as Belarus and Brazil are able to make significant advances in three years and move higher than Turkey in terms of their EACR rates in bank account ownership. Compared to countries with a similar coverage of bank account usage, Turkey has a higher dissimilarity index (Figure 8) in 2011. The dissimilarity index can be interpreted as the difference in bank account coverage rates across groups. For example, if the bank account coverage rate is equal across age, gender, education, and income groups, then the dissimilarity index would be equal to zero. The dissimilarity index shows that in 2011, there is a higher level of inequality in economic opportunity in Turkey compared to benchmark countries in terms of bank account ownership across age, gender, education, and income groups. Between 2011 and 2014, Turkey is able to make a significant improvement in its dissimilarity index indicating that the inequality in bank account ownership is decreasing in turkey even if the overall level of ownership slightly decreases. Turkey’s dissimilarity in account ownership is driven by gender (Figure 8). A Shapley decomposition measures which factors contributes to the dissimilarity index in Bank Account ownership. The Shapley decomposition is calculated for Turkey and countries with similar coverage in Bank Account Ownership. For example in 2011, the coverage in Turkey is 57.6 and in Brazil it is 55.9. However, the level of dissimilarity, and what accounts for the dissimilarity is very different between the two countries. Turkey’s dissimilarity is 8 percentage points higher than Italy at 23.7 which also results in a lower equity adjusted coverage rate for bank account ownership in Turkey. Moreover, 58.4 percent of dissimilarity is due to gender in Turkey, compared to only 8.8 percent in Italy. In Italy, the main sources of inequality are age and income. Turkey was able to substantially decrease the impact of gender on its dissimilarity of bank account usage. Contribution of gender on the d-index fell from 58.4 to 39.4 over the three years. The overall decrease in the level of dissimilarity was mainly driven by this improvement in the gender component as the absolute contribution of gender on dissimilarity fell from 13.8 in 2011 to 5.7 in 2014. However, Gender continues to be the largest source of dissimilarity, suggesting that the continuation of this positive trend in gender equity is crucial in increasing the EACR of Turkey further in terms of bank account ownership. 9 Figure 8. Bank Account Ownership – Individual Level Equity Adjust Coverage, (Individual level, FINDEX) 2011 2014 100 100 90 90 83.4 77.7 80 80 72.0 68.2 70 59.8 70 61.5 64.2 55.9 56.2 57.6 58.6 56.5 60 60 52.6 47.1 47.2 48.1 48.7 48.3 50 44.0 50 44.5 40 40 30 23.7 17.9 18.6 30 15.6 16.0 20 20 15.3 14.6 9.7 10.9 10 6.9 10 0 0 Brazil Bosnia and Turkey Belarus United Arab Brazil Bosnia and Turkey Belarus United Arab Herzegovina Emirates Herzegovina Emirates Coverage Dissimilarity EACR Shapley Decomposition of the Dissimilarity Index, (Individual level, FINDEX) 2011 2014 100% 5.3 100% 13.8 15.6 13.2 16.8 14.8 90% 20.2 90% 21.0 0.7 20.8 32.3 0.0 80% 37.8 80% 10.4 70% 18.3 70% 39.4 37.3 33.8 16.5 6.6 37.6 60% 60% 8.8 58.4 50% 33.2 50% 25.8 20.3 33.9 40% 40% 18.2 17.8 30% 30% 57.4 45.6 49.9 20% 20% 13.8 35.4 35.4 29.2 30.1 33.0 27.6 10% 10% 14.0 0% 0% Brazil Bosnia and Turkey Belarus United Arab Brazil Bosnia and Turkey Belarus United Arab Herzegovina Emirates Herzegovina Emirates Age Education Gender Income Notes:. FINDEX is an individual level data base. In FINDEX, 145 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. The variable control list for LITS and FINDEX varies. Sources: FINDEX 2011 and 2014 Using the LITS, which is a household level database, Turkey’s coverage rate of bank account ownership decreases by over 17 percentage points; and the contribution of gender in the dissimilarity index is greatly reduced (Figure 9). The reduction in the role of gender can be explained by the low number of households with female head of household as well as the presence of male household members with bank accounts. In the Shapley decomposition of the dissimilarity index, it is also important to note that the effect of gender, at almost 18 percent, is still an important part of dissimilarity at the household level. 10 Figure 9. Bank Account Ownership – Household Level Equity Adjust Coverage, (Household level, LITS) 50 45.5 47.1 44.2 45 41.1 38.1 40 36.1 35 29.4 30 25 22.8 22.4 22.8 20.3 20 16.3 15 10.7 11.9 10 6.2 5 0 Russian Federation Bulgaria Turkey Albania Montenegro Coverage Dissimilarity EACR Shapley Decomposition (Household level, LITS) 100% 90% 7.8 14.9 80% 43.2 70% 54.6 60% 25.0 50% 47.1 70.1 40% 10.0 18.1 30% 20% 34.2 10% 0% Russian Federation Bulgaria Turkey Albania Montenegro Age Education Gender Income Urban Notes:. LITS is household level. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries.The variable control list for LITS and FINDEX varies. Sources: LITS Differences in income and level of education are the main sources of dissimilarity in having a bank account. The importance of education is a potential reason for the drop in coverage at the household level. This drop could have occurred because households with high levels of income and education have multiple individuals with bank accounts while households with lower income and education have none. Compared to benchmark countries, gender and income seem to account for a high level of dissimilarity in Turkey while the urban and rural populations have more equality in terms of bank account ownership. An analysis of the individuals that do not own an account shows that not having enough money (50 percent of respondents) is the main reason for not having an account in Turkey (Figure 10). A lack of trust in financial institutions, and the high costs associated with banking are the two largest subsequent reasons at 27 percent apiece. This trend is similar to the rest of ECA while in other regions the cost of owning an account seems to be the distinctive second biggest reason of not having one. One interesting aspect of survey answers in Turkey is that a significantly lower portion of individuals listed lack of money when compared to the rest of ECA and the BRICS, and a slightly lower level than even the developed world. Considering the income levels of these different country groups, individuals in Turkey seem to have a different perspective regarding the budget necessary to use a bank account which may explain the higher usage of accounts in the country. 11 Figure 10. Reasons for Not having an Account, 2011 80 72 70 70 66 Percent of Respondents 60 54 50 50 46 40 36 27 27 29 30 22 22 23 22 24 21 19 21 18 17 20 17 20 14 15 14 13 15 17 16 7 6 9 10 4 6 5 0 Turkey BRIC Rest of Developing World Developed World Rest of ECA Too Far Away Too Expensive Lack Documentation Lack Trust Lack of Money Religious Reasons Family Member Already Has One Notes: Multiple responses allowed. Sources: FINDEX 2011 ii. SAVINGS & SAVINGS USING A FINANCIAL INSTITUTION Turkey is among the countries with the lowest savings rates, both at the individual and the household level (Figure 30 and Figure 32). In the European region (which is the coverage of the LITS household level survey), Turkey has the fifth lowest savings rate among 35 countries (10 percent). In the FINDEX database, Turkey also ranks the fifth lowest in the world among 145 countries, with 9.6 percent of individuals reporting that they save. The rate of savings at a financial institution in Turkey is low compared to the rest of the world, including only developing countries (Figure 11). In 2011, rate of savings using a bank in Turkey’s top quintile is similar to the rate of the BRICS’ lowest quintile. Turkey’s overall bank savings rate stands at 4 percent which is less than a fourth of the savings level in the BRICS countries and less than half of other ECA countries. Considering both the account usage and average income in Turkey is higher than the compared groups in 2011, Turkish individuals have a significantly lower propensity to save money. Moreover, the amount of savings differs among income groups in Turkey. Individuals in the highest income quintile are almost three times more likely to save than the members of the lowest quintile. This disparity among the bottom and top quintile groups is parallel with varying account usage levels across Turkey. 12 Figure 11. Percentage of Population (15+) that has Saved Money in a Financial Institution 2011 2014 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Rest of Turkey BRIC Rest of High Rest of Developing Income ECA Developing Income ECA World World Lowest quintile q2 q3 q4 top quintile Source: FINDEX 2011, 2014 Between 2011 and 2014, Turkey succeeded in increasing its savings rate at financial institutions at a faster rate than comparator country groups. However, bank use for savings in Turkey, at 9.1 percent, is still lower than the developing world average of 10.3 percent in 2014. Moreover, decomposing the growth of formal savings in Turkey points to specific cohorts that drive the heterogeneous increase. The increase in the top quintile of income earners was higher in comparison to the lower quintiles. Figure 11 displays the increasing disparity among the income quintiles in Turkey between 2011 and 2014. In addition to the heterogeneity in terms of earnings, individuals older than 65, and the ones who have higher education were able to expand their formal savings rates at a higher pace. Therefore, it will be important for Turkey to build on its success of increasing formal savings by ensuring that this growth is evenly spread across different age, income and level of education cohorts. There is high dissimilarity in savings using financial institutions, but not compared to countries with similar coverage. Compared to all 145 countries in the FINDEX database, the dissimilarity index is high at 33.7 (Figure 30). However, countries with similar coverage also have very high inequity in formal savings, and many at rates higher than in Turkey. The decomposition of the dissimilarity in 2011 shows that level of income and gender are the two main reasons of not being able to save at financial institutions in Turkey. Age stands out as a larger source of dissimilarity among the benchmark countries. Between 2011 and 2014, Turkey’s coverage rate in formal savings increased from 4.2 percent to 9.2 percent. However, this increase is not coupled with an improvement in the d-index as Turkey’s dissimilarity rises slightly during the same period. The Shapley decomposition of the d-index shows that the impact of gender decreased while the contribution of education increased. However Turkey’s gender portion of dissimilarity in 2014 was still higher than the benchmark countries with the exception of Algeria. Going forward, making formal savings more abundant amongst individuals with a lower level of income and education could help Turkey in lowering the d-index and boost equitable financial inclusion. 13 Figure 12. Savings Rate at a Financial Institution Equity Adjust Coverage, (Individual level, FINDEX) 2011 2014 60 60 53.1 49.5 50.5 48.5 50 50 41.5 42.4 40 36.2 40 34.0 33.7 30.6 30 30 20 20 13.8 9.1 9.6 10 10 6.6 6.0 3.7 3.8 4.2 4.3 4.5 2.9 3.3 4.1 2.9 2.2 2.0 2.8 2.0 2.3 1.4 0 0 Senegal Argentina Turkey Algeria Mali Senegal Argentina Turkey Algeria Mali Coverage Dissimilarity EACR Shapley Decomposition of Dissimilarity, (Individual level, FINDEX) 2011 2014 100% 100% 13.0 90% 17.7 90% 25.7 33.2 34.9 35.9 33.9 80% 7.7 80% 41.7 49.3 47.2 70% 9.8 70% 17.2 60% 60% 9.7 60.3 25.7 24.6 50% 1.3 34.9 50% 14.8 9.3 40.9 40% 40% 14.4 13.0 38.2 31.4 30% 57.4 30% 20.1 19.2 27.6 4.1 20% 20% 29.1 30.6 10% 23.5 22.7 21.4 18.0 12.6 10% 17.3 10.8 0% 0% Senegal Argentina Turkey Algeria Mali Senegal Argentina Turkey Algeria Mali Age Education Gender Income Notes:. FINDEX is an individual level data base and LITS is household level. In FINDEX, 145 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: FINDEX, LITS The decomposition of differences in the Equity Adjusted Coverage Rate (EACR) between countries summarizes if differences are driven by composition, equalization, or scale effects (Figure 13). For example, the EACR in Mali is 2.9 compared to 2.8 in Turkey. The net difference between the EACR values is only 0.1, but Figure 13 displays that different three different factors have individually larger impacts that neutralize each other. The higher EACR in Mali is driven by the scale effect, but diminished by equalization and composition effects. Savings has a higher coverage overall in Mali, however the equity of coverage among individuals in Mali contributes to a lower EACR. Compared to Turkey, Argentina has a lower EACR (2.8 vs 2.0 respectively). This lower EACR is explained by composition and equalization effects. In Argentina, equity of coverage in formal savings is poorer than in Turkey and the composition of the individuals have a negative effect on the EACR. 14 Figure 13. Decomposing the EACR Differences Across Countries (2011) Individual (FINDEX 2011) Household (LITS) 2 0.8 0.4 1 0.0 0 ‐0.4 -1 ‐0.8 -2 ‐1.2 -3 ‐1.6 -4 Senegal Argentina Algeria Mali Azerbaijan Bulgaria Tajikistan Moldova Composition Equalization Scale Notes:. FINDEX is an individual level data base and LITS is household level. In FINDEX 2011, 141 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: FINDEX, LITS Turkey’s dissimilarity index in savings falls substantially when measuring at the household-level despite a similar coverage rate. As a result, Turkey has a significantly lower dissimilarity index than the benchmark countries (Figure 28). Part of this effect can be explained by some homogeneity in households. Even with this reduction, the composition of the d-index remains fairly stable as age of the household head and income of the household are the principal sources of dissimilarity. (Figure 29) The decomposition of the difference in the EACR using the LITS survey shows that households save at a more equitable level in Turkey compared to the benchmark countries (Figure 13); the equalization effects explains why the EACR is lower in the comparator countries of Azerbaijan, Bulgaria, Tajikistan, and Moldova. iii. DEBIT AND CREDIT CARD USE Debit and credit card usage in Turkey is notably higher than the BRICS economies and the rest of ECA countries in 2011 (Figure 14). Over 42 percent of the lowest income quintile in Turkey has debit cards compared to around 13 percent of their counterparts in the BRICS and around 29 percent of other ECA countries. The difference is even greater in credit card usage. In Turkey, 30 percent of the lowest income quintile has credit cards. This value is under 10 percent in both the BRICS and rest of ECA. In fact, Turkey’s overall credit card usage is higher than that of the developed world in 2011. The high level of credit card utilization may form a financially dangerous situation especially for the poor who are more prone to having larger debt than their income. Turkey’s gender discrepancy in debit and credit card ownership is parallel to the use of bank accounts. In 2011, debit and credit card usage rates for males are 81 percent and 64 percent. The same rates are 31 and 26 percent for females, respectively. Between 2011 and 2014, both debit and credit card ownership levels decrease for males and females. However, the fall is steeper amongst the males than it is for the females resulting in a reduced gender disparity. 15 Figure 14. Debit and Credit card Ownership in Turkey, 2011 Debit Card 2011 2014 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Developing High Rest of ECA Turkey BRIC Developing High Rest of Income Income ECA Credit Card 2011 2014 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Developing High Rest of Turkey BRIC Developing High Rest of Income ECA Income ECA Lowest quintile q2 q3 q4 top quintile Notes: Developing world excludes Turkey Source: FINDEX 2011 and 2014 The equity adjusted coverage decomposition of debit card ownership in Turkey produces similar results to the analysis of bank account usage at the individual level (Figure 15). However, since the coverage rate of debit card usage in other countries is lower than bank account ownership in general, Turkey’s analogous values across the two indicators place it on a higher ranking in the country distribution in terms of debit card usage (Figure 41). Turkey has a higher dissimilarity index for debit card ownership, which is primarily driven by gender when compared to benchmark countries. The subdivision of the difference in EACR’s across the benchmark countries indicates that debit cards are used at a higher scale in Turkey while the composition effect is more positive across the countries with similar coverage rates (Figure 39). Between 2011 and 2014, debit card ownership at the individual level fell from 56.6 percent to 43.4 percent in Turkey. On the other hand for three of the four benchmark countries, there were significant increases in coverage rates. Despite the falling coverage, Turkey’s dissimilarity level fell from 24.3 percent to 21.6 percent over the three years, mainly driven by the improvement in the gender component. However, the improvement in dissmilarity is not able to neutralize the negative pressure coming from the declining coverage rate as Turkey’s EACR falls form 42.9 to 33.9 between the two survey periods. 16 Figure 15. Debit Card Ownership Equity Adjust Coverage, (Household-level, LITS) 80 70.1 70 64.8 63.8 60.7 55.7 57.5 60 53.0 51.6 50.5 47.0 50 40 30 20 11.4 12.1 7.3 6.5 8.9 10 0 Latvia Czech Republic Turkey France Croatia Coverage Dissimilarity EACR Equity Adjust Coverage, (Individual level, FINDEX) 2011 2014 90 90 77.0 80 80 69.6 67.1 65.7 70 60.7 70 56.6 58.0 58.4 56.5 55.4 60 50.3 52.0 52.1 60 50 45.2 39.9 42.9 50 39.9 43.3 40 40 33.3 33.9 30 24.3 30 20.8 18.5 21.6 14.1 16.5 20 10.4 20 9.6 12.9 14.1 10 10 0 0 Belarus United Arab Turkey Rep. of Korea Mongolia Belarus United Arab Turkey Rep. of Korea Mongolia Emirates Emirates Coverage Dissimilarity EACR Notes:. FINDEX is an individual level data base and LITS is household level. In FINDEX, 145 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors are shown. See Appendix for all countries. Sources: FINDEX, LITS The role of gender in having a debit card is reduced substantially at the hosuehold level from 57.4 percent to 16.8 percent. (Figure 16) The sources of this decline are likely to be parallel to the reduction in the contribution of gender to the dissimilarity index in the usage of bank accounts. Income and education are the main sources of dissimilarity as they collectviely account for almost 75 percent of the d-index. Comapred to benchmark countries, this value is relatively higher as age or the urban/rural divide play a more important role in countries with similar coverage rates at the household level. 17 Figure 16. Shapley Decomposition of Debit Card Dissimilarity Index (Household level, LITS) 100% 2.2 4.6 3.8 90% 15.3 20.9 22.1 80% 4.7 37.3 70% 19.0 10.8 56.7 23.9 60% 10.2 50% 16.8 12.7 40% 10.6 30% 58.8 3.2 25.6 52.3 38.0 20% 15.3 10% 14.3 16.9 0% 4.1 Latvia Czech Republic Turkey France Croatia Age Education Gender Income Urban (Individual level, FINDEX) 2011 2014 100% 6.2 100% 17.1 13.0 15.6 90% 18.9 25.1 90% 21.8 19.0 22.9 26.2 80% 1.5 22.4 5.7 80% 1.3 5.2 7.0 4.5 70% 28.2 8.6 70% 34.2 60% 16.9 60% 58.6 26.4 44.5 33.7 50% 50% 43.6 44.4 16.1 40% 40% 70.2 30% 54.5 30% 53.1 20.9 20% 13.8 20% 44.8 42.1 34.0 22.0 10% 25.7 10% 14.6 15.6 0% 0% Belarus United Arab Turkey Rep. of Mongolia Belarus United Arab Turkey Rep. of Mongolia Emirates Korea Emirates Korea Age Education Gender Income Notes:. LITS is household level. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS, FINDEX Turkey’s credit card usage rate places it among a group of high-income benchmark countries both at the individual and the household levels (Figure 46 and Figure 48). Turkey’s dissimilarity value at the individual level is higher when compared to countries with similar coverage rates indicating that even though credit card ownership is well-spread, the inequality level is relatively high across the measured indicators (Figure 17). The effect of gender continues to be the principal component of the d-index at 44.4 percent despite the relative reduction in comparison to the respective values for debit card and bank account ownership. The decomposition of the EACR difference shows that the scale in credit card usage of Turkey is significantly higher than the benchmark countries (Figure 44). However, the equalization effects confirm the relatively high dissimilarity level in Turkey and suggest that the equity of credit card ownership has room for improvement. Parallel to the trend in debit card usage, the coverage of credit card ownership also fell in Turkey between 2011 and 2014, from 45.1 to 32.8 percent. The dissimilarity remains stable at around 26 percent. Netherlands and Demark experience a similar decline in their coverage and EACR rates which shows that Turkey is not alone in undergoing a decline in credit card ownership. 18 Figure 17. Credit Card Ownership Equity Adjust Coverage, (Individual level, FINDEX) 2011 2014 60 60 54.7 54.9 50 45.1 45.1 46.0 50 46.4 41.6 42.4 42.9 38.9 40 35.3 40 36.6 32.5 33.2 32.7 33.7 32.8 32.7 28.9 29.7 30 26.3 30 26.7 23.7 24.1 21.7 21.8 20 16.6 20 13.6 15.3 10.6 10 10 0 0 Netherlands Spain Turkey Denmark Taiwan, Netherlands Spain Turkey Denmark Taiwan, China China Coverage Dissimilarity EACR Equity Adjust Coverage, (Household-level, LITS) 80 69.4 70 64.4 55.4 55.9 57.2 60 52.2 47.4 48.7 50 45.7 42.3 40 30 23.7 20 15.1 14.8 12.4 7.1 10 0 France Hungary Great Britain Turkey Sweden Coverage Dissimilarity EACR Notes:. FINDEX is an individual level data base and LITS is household level. In FINDEX, 145 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: FINDEX, LITS Coverage of credit card ownership in Turkey is considerably higher at the household level than at the individual level (Figure 17). This difference is noteworthy since a similar change is not observed for the coverage rate in debit card ownership. Compared to the benchmark countries, the effect of age on the d-index seems lower across the Turkish population (Figure 18) in 2011. At the individual level, the primary explanatory variables for dissimilarity in credit card usage are gender and age. However, at the household level, the dissimilarity in credit card usage across groups is driven more by the education of the household and its income level. Between 2011 and 2014, Turkey succeeded in reducing the negative impact of gender disparity in credit card ownership as the contribution of gender fell form 44.4 percent to 31.7 percent. Age became the largest contributor to the d-index by 2014 as the percentage of 15-24 year olds that had a credit card fell from 32 to 12 in three years. This was a sharp reduction, however in comparison to benchmark country groups, 12 percent ownership at the youngest measured cohort was a more expected level for Turkey. 19 Figure 18. Shapley Decomposition of the Dissimiliarity in Credit Card use (Individual level, FINDEX) 2011 2014 100% 100% 12.9 17.6 13.4 15.5 90% 20.8 24.7 90% 20.4 3.1 29.5 26.8 25.8 0.6 80% 17.5 80% 4.9 70% 31.0 70% 5.3 32.8 22.8 16.0 9.3 31.7 60% 27.5 44.4 60% 33.2 50% 42.0 50% 13.6 40% 40% 15.0 36.4 15.7 30% 17.9 30% 58.7 52.4 51.2 46.8 20% 38.9 20% 32.9 27.6 25.3 10% 20.1 10% 18.1 0% 0% Netherlands Spain Turkey Denmark Taiwan Netherlands Spain Turkey Denmark Taiwan Age Education Gender Income (Household level, LITS) 100% 2.2 4.6 3.8 90% 15.3 20.9 22.1 80% 4.7 37.3 70% 56.7 19.0 10.8 23.9 60% 10.2 50% 16.8 12.7 40% 10.6 30% 58.8 3.2 25.6 52.3 38.0 20% 15.3 10% 14.3 16.9 0% 4.1 Latvia Czech Republic Turkey France Croatia Age Education Gender Income Urban Notes:. LITS is household level. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: FINDEX, LITS iv. BORROWING In terms of borrowing, informal sources are more heavily used than formal ones in Turkey (Figure 19). Less than 5 percent of adults borrow from financial institutions in Turkey which is lower than the formal borrowing levels at both the BRICS and the rest of ECA. On the other hand, over 40 percent of adults use store credit and family/friends as an informal borrowing source. These levels indicate that individuals in Turkey are more than 8 times likely to use store credit than a financial institution whereas no other country groups uses store credit more than financial institutions as a lending source.4 Individuals in Turkey do not seem to trust formal sources as viable borrowing source. 4 In fact, Turkey has the highest coverage rate in having store credit amongst the surveyed FINDEX countries. See appendix for further information on the EAC decomposition for borrowing form formal and informal institutions. 20 Figure 19. Rate of Borrowing from Formal and Informal Sources (%), 2011 50 43.1 43.2 45 40 35 30 25 20 15 10 4.6 2.9 5 0.5 0 Financial Institution Store Credit Family/Friends Past Employer Other Turkey BRIC Rest of Developing World Developed World Rest of ECA Source: FINDEX 2011 5. GOING FORWARD Turkey can use different public and private services to increase overall saving levels and reduce transaction costs associated with the banking industry. Deposit-collection services can be used to facilitate the savings process, especially in rural areas where traveling to branches of financial institutions is more burdensome. Moreover, collection boxes and drop-off sites can also be established to further reduce the transaction costs. Mel, McIntosh and Woodruff’s (2013) experiment in Sri Lanka demonstrates how policies to increase savings can be used most efficiently. In the experiment, workers had methods of direct deposit but could only withdraw from the closest bank. The results show that by offering deposit-collection services or collection boxes, banks can form an incentive structure that favors deposits over withdrawals. They find that the frequency of deposit-collection does not substantially change the overall level of savings even though it affects the deposit amounts. The usage of collection boxes also does not reduce the amount of savings, even though it significantly reduces the costs for the service provider. Therefore a community savings lockbox appears as an effective method that can be employed along with deposit-collection services. Summing up, Turkey has been successful in reducing poverty between 2002 and 2011 primarily through the improvements of its labor market. Overall, individuals in Turkey use bank accounts at a high level but the usage is significantly lower amongst especially the females and also the lower income groups, the less educated and the young. The observed trends between 2011 and 2014 indicate that Turkey has been most successful in addressing disparity issues related to gender. This is highly encouraging since financial inclusion of females was the most severe shortcoming of Turkey based on the 2011 results. Despite this improvement, gender disparity continues to be the most significant source of discrepancy in multiple indicators suggesting that Turkey is only at the beginning of the improvement process in gender equity. Moreover, a lack of money and trust in financial institutions along with high costs associated with banking seem to be the common reasons for not holding a formal account. Individuals in Turkey have low savings levels even though usage of debit and credit cards is significantly higher than the rest of ECA. Turkish citizens also notably prefer informal lending rather than formal borrowing sources when they are in need of money. Therefore several challenges must be addressed including: (i) continuing to improve the access to finance for females and targeting the poor more explicitly for financial inclusion, (ii) increasing trust in formal financial institutions, (iii) reducing the costs of banking especially in terms of services that incentivize increased savings such as the use of deposit-collection methods, and (iv) increasing public awareness of the appropriate and informed use of debit and credit cards. 21 6. REFERENCES Azevedo, Joao Pedro; Atamanov, Aziz. 2014. Pathways to the middle class in Turkey : how have reducing poverty and boosting shared prosperity helped?. Policy Research working paper ; no. WPS 6834. World Bank Group. http://documents.worldbank.org/curated/en/2014/04/19354560/pathways- middle-class-turkey-reducing-poverty-boosting-shared-prosperity-helped Beck, Demirguc-Kunt, and Levine 2007. Finance, Inequality and the Poor. Journal of Economic Growth Volume 12, Issue 1 , pp 27-49 Kluwer Academic Publishers-Plenum Publishers http://link.springer.com/article/10.1007%2Fs10887-007-9010-6 Brunori, Paolo, Francisco H.G. Ferreira, Vito Peragine. 2013. Inequality of Opportunity, Income Inequality and Economic Mobility WPS6304 World Bank Group. http://elibrary.worldbank.org/doi/pdf/10.1596/1813-9450-6304 Claessens, Stijn and Perotti, Enrico C. 2007., Finance and Inequality: Channels and Evidence, Available at SSRN: http://ssrn.com/abstract=998468 or http://dx.doi.org/10.2139/ssrn.998468 de Mel, Suresh, Craig McIntosh, and Christopher Woodruff. 2013. "Deposit Collecting: Unbundling the Role of Frequency, Salience, and Habit Formation in Generating Savings." American Economic Review, 103(3): 387-92. Demirguc-Kunt, Asli, Leora Klapper, Dorothe Singer, and Peter Van Oudheusden. 2015. “The Global Findex Database 2014: Measuring Financial Inclusion around the World.” Policy Research Working Paper 7255, World Bank, Washington, DC. Johnston, Don Jr., and Jonathan Morduch. 2008. “The Unbanked: Evidence from Indonesia.” World Bank Economic Review 22 (3): 517-537. http://www.nyudri.org/wp-content/uploads/2011/10/unbanked.pdf McKenzie, David, and Christopher Woodruff . 2008. “Experimental Evidence on Returns to Capital and Access to Finance in Mexico” World Bank Economic Review Volume 22 (3): 457 – 482. http://elibrary.worldbank.org/doi/pdf/10.1093/wber/lhn017 Molinas Vega, José R.; Paes de Barros, Ricardo; Saavedra Chanduvi, Jaime; Giugale, Marcelo; Cord, Louise J.; Pessino, Carola; Hasan, Amer. 2012. Do Our Children Have a Chance? A Human Opportunity Report for Latin America and the Caribbean. World Bank. © World Bank. https://openknowledge.worldbank.org/handle/10986/2374 License: CC BY 3.0 IGO.” Paes de Barros, Ricardo; Ferreira, Francisco H. G.; Molinas Vega, Jose R.; Saavedra Chanduvi, Jaime. 2009. Measuring Inequality of Opportunities in Latin America and the Caribbean. Washington, DC: World Bank; New York: Palgrave Macmillan. © World Bank. https://openknowledge.worldbank.org/handle/10986/2580 License: CC BY 3.0 IGO. 22 7. DATA APPENDIX The Household Budget Survey (HBS, 2009-2012) Variables used form HBS:  Access to Banking (Bank_Access): Accessibility of Banking services according to the location of the dwelling. Original variable name in dataset: ZOR_BANKA Dummy variable equals 1 if accessibility is rated “easy” or “very easy”, equals 0 if accessibility is rated “difficult” or “very difficult” The Survey on Income and Living Conditions (SILC, 2006-2012)  Mortgage and Other Payments (Mortgage): Possession of a mortgage, loan repayment or rent payment in last 12 months preceding the survey. Original variable name in dataset: HE010 Dummy variable equals 1 if respondent possesses the payment, equals 0 if he/she does not have any such payment.  Arrears on Mortgage and Other Payments (Arrear_Mortgage): Possession of an arrear on a mortgage, loan repayment or rent payment in last 12 months preceding the survey. Original variable name in dataset: HE010 Dummy variable equals 1 if respondent has an arrear on the payment, equals 0 if he/she does not have any such arrear.  Credit Card and Other Payments (Credit_Card_Payments): Possession of a hire purchase installment, credit card or other loan payment in last 12 months preceding the survey. Original variable name in dataset: HE030 Dummy variable equals 1 if respondent possesses the payment, equals 0 if he/she does not have any such payment.  Arrears on Mortgage and other Payments (Arrear_ Credit_Card): Possession of an arrear on a hire purchase installment, credit card or other loan payment in last 12 months preceding the survey. Original variable name in dataset: HE030 Dummy variable equals 1 if respondent has an arrear on the payment, equals 0 if he/she does not have any such arrear. The Global Financial Inclusion Database, (FINDEX, 2011, 2014)  Bank Account Ownership (Bank_Account_p): Ownership of an account at a bank or credit union (or another financial institution, where applicable – for example, cooperatives in Latin America) Original variable name in dataset: q1a Dummy variable equals 1 if respondent has a bank account, equals 0 if he/she does not own an account.  Savings Rate (Savings_p): Any saved or set aside money in the 12 months preceding the survey Original variable name in dataset: q11 Dummy variable equals 1 if respondent has saved money, equals 0 if he/she has not.  Savings in Bank (Savings_Bank): Any saved or set aside money in the 12 months preceding the survey using an account at a bank, credit union (or another financial institution) or microfinance institution as supposed to saved money using an informal savings club or a person outside the family. Original variable name in dataset: q13a Dummy variable equals 1 if respondent has saved money using a financial institution, equals 0 if he/she has not.  Debit Card Ownership (Debit_Card_p): Ownership of a debit card (sometimes called a bank card, bank book or salary card) Original variable name in dataset: q3a Dummy variable equals 1 if respondent has a debit card, equals 0 if he/she does not own a card. 23  Credit Card Ownership (Credit_Card_p): Ownership of a credit card Original variable name in dataset: q3ab Dummy variable equals 1 if respondent has a credit card, equals 0 if he/she does not own a card.  Money Borrowed from a Financial Institution (Borrow_Bank): Any money borrowed from a bank, credit union (or another financial institution, where applicable – for example, cooperatives in Latin America), or microfinance institution Original variable name in dataset: q14a Dummy variable equals 1 if respondent has borrowed money from a financial institution, equals 0 if he/she has not.  Money Borrowed from a Store (Borrow_Store): Any money borrowed from a store using installment credit or buying on credit Original variable name in dataset: q14b Dummy variable equals 1 if respondent has borrowed money from a store, equals 0 if he/she has not.  Money Borrowed from Family or Friends (Borrow_Family): Any money borrowed from family or friends Original variable name in dataset: q14c Dummy variable equals 1 if respondent has borrowed money from family or friends, equals 0 if he/she has not.  Reasons for not having a bank account: o Too Far Away (Too_Far) o Too Expensive (Too_Expensive) o Not Possessing the Necessary Documentation (Lack_Doc) o Not Trusting Financial Institutions (Lack_Trust) o Not having Enough Money to Use the Account (Lack_Money) o Religious Reasons (Religious) o Someone Else in Family Has an Account (Family_Has) Only respondents who do not have a bank account are asked to list their reasons. Original variable names in dataset: Too_Far (q10a), Too_Expensive (q10b), Lack_Doc (q10c), Lack_Trust (q10d), Lack_Money (q10e), Religious (q10f), Family_Has (q10g) The Life in Transition Survey (LITS, 2010)  Bank Account Ownership (Bank_Account_h): Ownership of a bank account by any household member Original variable name in dataset: q225c Dummy variable equals 1 if any individual in the household has a bank account, equals 0 if no one has an account.  Savings Rate (Savings_h): Any left over money to put into savings in the household in a typical month. Original variable name in dataset: q223_t1 Dummy variable equals 1 if household has saved money, equals 0 if household has not.  Debit Card Ownership (Debit_Card_h): Ownership of a debit card by any household member Original variable name in dataset: q225d Dummy variable equals 1 if any individual in the household has a debit card, equals 0 if no one has a a card.  Credit Card Ownership (Credit_Card_h): Ownership of a credit card by any household number Original variable name in dataset: q225e Dummy variable equals 1 if any individual in the household has a credit card, equals 0 if no one has a card.  Borrowed Money (Borrow): Money borrowed from anyone (e.g. friend, other person or institution) by any individual in the household. Original variable name in dataset: q805 Dummy variable equals 1 if any individual in the household has borrowed money, equals 0 if no one has borrowed 24  Borrowed Money from Informal Source (Non_Bank_Borrow): Money borrowed from sources other than banks. Original variable names in dataset: q805 and q810_ Dummy variable equals 1 if any individual in the household who has borrowed money borrowed the money from a source other than a bank (relative, friend, private money lender, NGO, other), equals 0 if money was borrowed from banks.  Trust in Banks (Trust_Bank): Level of trust in Banks and the financial system. Original variable name in dataset: q303j Dummy variable equals 1 if there is some or complete trust in the banking system, equals 0 if there is some or complete distrust or neither trust or distrust. The World Bank Development Indicators (WDI, 2014)  “NY.GDP.PCAP.PP.KD” - GDP per capita, PPP (constant 2011 international $)  “NY.GDP.PCAP.PP.KD.ZG” - GDP per capita, PPP annual growth (%)  “NY.GDS.TOTL.ZS” - Gross domestic savings (% of GDP)  “SI.POV.25DAY” - Poverty headcount ratio at $2.5 a day (PPP) (% of population)  “SI.POV.2DAY” - Poverty headcount ratio at $2 a day (PPP) (% of population)  “SI.POV.4DAY” - Poverty headcount ratio at $4 a day (PPP) (% of population)  “SI.POV.5DAY” - Poverty headcount ratio at $5 a day (PPP) (% of population)  “SI.POV.DDAY” - Poverty headcount ratio at $1.25 a day (PPP) (% of population)  “SI.POV.NAHC” - Poverty headcount ratio at national poverty line (% of population) 25 8. FINANCIAL INCLUSION BENCHMARK DASHBOARD –USER NOTES URL: http://dataviz.worldbank.org/t/ECA/views/tur_finance_equity_poverty7/Dshnew?:embed=y&:display_count=no#3 Financial Inclusion and Equity Benchmark Dashboard Access to finance allows for the smoothing of consumption, savings, management of money, and loans for purchases. Therefore it is important as an instrument against poverty and to boost shared prosperity. This dashboard explores the 2011 and 2014 FINDEX surveys to construct a panel of benchmarks accross countries to help decision makers better understand the specificities of several Financial Inclusion indicators in the context of Turkey Step 1. Select the Indicator and the Controls Cross - Country Comparisons Outcome Indicator (Equity Adjusted Coverage Rate) Coverage, Dissimilarity, and EACR Bank Account Ownership 2011 2014 Control List Controllist 2: age, education, gender, urban/rural, income quintiles Take a quick look: Turkey's Bank Account Ownership Turkey coverage: 57.601 Turkey coverage: 56.513 Turkey hoi: 48.276 Turkey hoi: 43.953 2011 2014 Turkey Turkey Turkey dindex: 15.969 Turkey dindex: 14.576 57.60 56.51 43.95 48.28 15.97 14.58 Coverage Dissimilarity EACR Coverage Dissimilarity EACR Step 2. Select Country Comparators Shapley Decomposition of Dissimilarity Index EACR Decomposition (Turkey as base country) 2011 2014 2011 2014 The value of this dashboard is the possibility of the user to select comparison countries on the basis of different criterias such as poverty level, GDP per capita, and national savings 50 as a share of GDP or on the basis of one of the financial inclusion indicators. Choice of filter indicator (WDI, FINDEX) GDP per capita, PPP (constant 2011 international $) 0 Filter: All (highlight countries to select them) ‐50 . . TUR 17,188 . Measures the size of contribution of the selected dimensions to the Three different sources of disparity amongst the equity adjusted dissimilarity index of each country coverage rates of selected countries with respect to Turkey Dimensions Change Age Gender Composition Scale 0.1427 to 127,213.12 Enter Range Education Income Equalization Source: LiTS; Global FINDEX; HICS; and, SILC. Dashboard produced as part of the background note to the Financial Inclusion Strategy for Turkey and Financial Inclusion Conference held in June 2014 . Authors: João Pedro Azevedo , Judy Yang and Osman Kaan Inan in collaboration with Alper Ahmet Oguz. and Ilias Skamnelos. Dashboard for comments. For selecting country comparators: WDI indicators are averaged between 2008 and 2013 for avilable years. FINDEX indicators for selecting comparators represent values from FINDEX 2011. 26 GUIDELINES TO USE DASHBOARD 1. Select the relevant dataset a. FINDEX 2. Select the indicator that is going to be analyzed 3. Select the controllist number with the relevant combination of control groups FINDEX: a. Controllist 31: age, education, gender b. Controllist 32: age, education, gender, income 4. Select appropriate filter criteria to generate the country distribution a. Selected World Development Indicators b. Selected FINDEX variables 5. Adjust the Lower bound and Upper Bound cursors to narrow the values of the filter criterias. 27 9. METHODOLOGY The Equity Adjusted Coverage Index (EACR) is a measure of the availability of services, discounted or “penalized” by how unfairly the services are distributed among the population. The EACR calculations in this paper generate three indicators:  Coverage: Overall level of availablity/access of the indicator across the pouplation  Dissimilarity (D-index): The difference in access rates for a given service for groups defined by circumstance compared with the average access rate for the same service for the population as a whole.  Equity Adjusted Covergae rate (EACR): Average access rate in the population, penalized by the degree of dissimilarity in coverage across different types of indicators. The variation between calculated EACR values for different units of observation or changes across time can be decomposed in to three effects:  Scale: Changes related to the expansion of the overall coverage of a particular good or service  Equalization: Changes on distribution of access  Composition: Changes in the relative importance of different groups in society (i.e. demographic changes) The D-index can be decomposed using the Shapley decompositoin which identifies the contribution of each circumstance to inequality in access to opportunities. The Shapley decompostion decomposes the main explanatory factors of the dissimilarity Constructing the Equity Adjusted Coverage Index5: 1. A separable logistic model is estimated based on whether the individual i has access to a given indicator (etc, bank account usage, savings). Different specifications were chosen for circumstances; categorical for age and income and binary for the other indicators. The coefficent estimations were oibtained as a result of the logistic regression. 2. With the coefficent estimates, the probablity of access to the finanical service was precited for each individual in the sample. Probability (p̂ ) is based on the predicted relationship of , and a vector of the circumstances . ∑ p̂ 1 ∑ 3. The coverage rates, C, were computed p̂ 5 The mathematical framework of the Equity Adjusted Coverage Index is based on “Do Our Children Have A Chance?Index”, Molinas Vega et al. (2010) pg 49 28 where or other sampling weights. 4. The Dissimilarity index D was computed 1 | p̂ | 2 5. The penalty was computed, ∗ 6. The EACR was computed EACR = C – P for each financial service 29 10. TABLES i. BANK ACCOUNTS Figure 20. EACR Decomposition of Individual Bank Account Ownership FINDEX 2011 FINDEX 2014 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 Brazil Bosnia and Belarus United Arab Brazil Bosnia and Belarus United Arab Herzegovina Emirates Herzegovina Emirates Composition Equalization Scale Notes:. FINDEX is an individual level data base. In FINDEX, 141 countries are sorted on Coverage in 2011 and 143 countries are sorted in 2014, and only nearest neighbors in coverage rates are shown. Sources: FINDEX Figure 21. EACR Decomposition of Household Bank Account Ownership Household (LITS) 15 10 5 0 -5 -10 -15 -20 -25 -30 Russian Bulgaria Albania Montenegro Federation Composition Equalization Scale Notes:. LITS is household level dataset. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS 30 Figure 22. Individual (Bank_Account, FINDEX 2011) China, 63.7142 Turkey, 57.6014 Brazil, 55.8604 Russian Federation, 48.201 India, 35.2637 0 20 40 60 80 100 120 EACR dissimilarity Coverage Notes: 141 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 31 Figure 23. Individual (Bank_Account, FINDEX 2014) China, 78.9 Brazil, 68.2 Russian Federation, 67.4 Turkey, 56.5 India, 52.7 0.0 20.0 40.0 60.0 80.0 100.0 120.0 EACR dissimilarity Coverage Notes: 143 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 32 Figure 24. Household (Bank_Account, LITS) France Sweden Germany Slovenia Great Britain Slovak Republic Estonia Czech Republic Italy Lithuania Latvia Croatia Poland Serbia Kosovo Macedonia Hungary All Mongolia Bosnia and Herzegovina Montenegro Albania Turkey 41.1 Bulgaria Russian Federation Romania Belarus Kazakhstan Armenia Ukraine Moldova Georgia Uzbekistan Azerbaijan Tajikistan Kyrgyzstan 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 35 countries sorted by coverage rates. Coverage rates are labeled. Sources: LITS 33 ii. SAVINGS Figure 25. Savings at a Financial Institution across Gender 2011 2014 50% 50% 45% 45% 40% 40% 35% 35% 30% 30% 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World male female Source: FINDEX 2011 and 2014 Figure 26. Savings at a Financial Institution across Education Levels 2011 2014 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World Primary Secondary Tertiary Source: FINDEX 2011 and 2014 34 Figure 27. Savings at a Financial Institution across Age Groups 2011 2014 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Income Developing World World 15‐24 25‐64 65+ Figure 28. Savings EACR (household level, LITS) 30 27.4 25 23.7 21.4 20 15 12.4 10.1 10.4 10.5 10.8 9.0 9.0 9.2 10 8.2 6.8 6.5 5.3 5 0 Azerbaijan Bulgaria Turkey Tajikistan Moldova Coverage Dissimilarity EACR Notes:. LITS is household level. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS 35 Figure 29. Shapley Decomposition of the Dissimilarity in Savings (household level, LITS) 100% 90% 80% 41.1 70% 60% 9.7 50% 40% 15.0 30% 20% 33.8 10% 0% Azerbaijan Bulgaria Turkey Tajikistan Moldova Age Education Gender Income Urban Notes:. LITS is household level. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS 36 Figure 30. Savings at a Financial Institution EACR (Individual level, FINDEX 2011) China, 32.1 India, 11.6 Russian Federation, 10.9 Brazil, 10.3 Turkey, 4.2 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 141 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 37 Figure 31. Savings at a Financial Institution EACR (Individual level, FINDEX 2014) China, 41.2 Russian Federation, 15.4 India, 14.4 Brazil, 12.3 Turkey, 9.1 0 20 40 60 80 100 EACR Dissimilarity Coverage Notes: 143 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 38 Figure 32. Savings EACR (household level, LITS) Sweden Germany Czech Republic Slovak Republic France Italy Ukraine Great Britain Belarus Estonia All Lithuania Kosovo Russian Federation Slovenia Albania Poland Latvia Hungary Kazakhstan Uzbekistan Kyrgyzstan Romania Croatia Bosnia and Herzegovina Mongolia Serbia Macedonia Montenegro Moldova Tajikistan Turkey 10.1 Bulgaria Azerbaijan Georgia Armenia 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage 35 countries sorted by coverage rates. Coverage rates are labeled Sources: LITS 39 iii. DEBIT AND CREDIT CARDS Figure 33. Debit Card Ownership across Gender 2011 2014 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World male female Source: FINDEX 2011 and 2014 Figure 34. Credit Card Ownership across Gender 2011 2014 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Rest of Developing Developing Income ECA World World male female Source: FINDEX 2011 and 2014 40 Figure 35. Debit Card Ownership across Education Levels 2011 2014 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Rest of ECA Developing Developing Income World World Primary Secondary Tertiary Source: FINDEX 2011 and 2014 Figure 36. Debit Card Ownership across Education Levels 2011 2014 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Turkey BRIC Rest of High Income Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Developing World World Primary Secondary Tertiary Source: FINDEX 2011 and 2014 41 Figure 37. Debit Card Ownership across Age Groups 2011 2014 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Turkey BRIC Rest of High Rest of ECA Turkey BRIC Rest of High Income Rest of ECA Developing Income Developing World World 15‐24 25‐64 65+ Figure 38. Credit Card Ownership across Age Groups 2011 2014 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Turkey BRIC Rest of High Rest of Turkey BRIC Rest of High Income Rest of ECA Developing Income ECA Developing World World 15‐24 25‐64 65+ 42 Figure 39. EACR Decomposition of Debit Card Ownership between countries FINDEX 2011 Household (LITS) 40 40 30 30 20 20 10 10 0 0 -10 -10 -20 -20 Belarus United Arab Korea, Rep. Mongolia Belarus United Arab Korea, Rep. Mongolia Emirates Emirates Composition Equalization Scale Notes:. FINDEX is an individual level data base. In FINDEX, 141 countries are sorted on Coverage in 2011 and 143 countries are sorted in 2014, and only nearest neighbors in coverage rates are shown. Sources: FINDEX Figure 40. EACR Decomposition of Debit Card Ownership between countries Household (LITS) 20 15 10 5 0 -5 -10 -15 -20 Latvia Czech France Croatia Republic Composition Equalization Scale Notes:. LITS is a household level data set. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS 43 Figure 41. Debit Card EACR (individual level, FINDEX 2011) Turkey, 56.6 Brazil, 41.2 China, 40.9 Russian Federation, 37.0 India, 8.4 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 141 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 44 Figure 42. Debit Card EACR (Individual level, FINDEX 2014) Turkey, 56.6 Brazil, 41.2 China, 40.9 Russian Federation, 37.0 India, 8.4 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 143 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 45 Figure 43. Debit Card EACR (household level, LITS) Sweden Germany Estonia Great Britain Italy Croatia France Turkey 57.5 Czech Republic Latvia Slovenia All Bulgaria Slovak Republic Macedonia Lithuania Azerbaijan Serbia Romania Poland Russian Federation Montenegro Albania Bosnia and Herzegovina Georgia Kosovo Moldova Hungary Belarus Ukraine Kazakhstan Armenia Uzbekistan Mongolia Kyrgyzstan Tajikistan 0 10 20 30 40 50 60 70 80 90 100 EACR dissimilarity coverage 35 countries sorted by coverage rates. Coverage rates are labeled Sources: LITS 46 Figure 44. EACR Decomposition of Credit Card Ownership between countries FINDEX 2011 FINDEX 2014 25 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 Netherlands Spain Denmark Taiwan, China Netherlands Spain Denmark Taiwan, China Composition Equalization Scale Notes:. FINDEX is an individual level data base. In FINDEX, 141 countries are sorted on Coverage in 2011 and 143 countries are sorted in 2014, and only nearest neighbors in coverage rates are shown. Sources: FINDEX Figure 45. EACR Decomposition of Credit Card Ownership between countries Household (LITS) 20 15 10 5 0 -5 -10 -15 -20 Latvia Czech France Croatia Republic Composition Equalization Scale Notes:. LITS is a household level data set. In LITS, 35 countries are sorted on Coverage, and only nearest neighbors in coverage rates are shown. See Appendix for all countries. Sources: LITS 47 Figure 46. Credit Card EACR (individual level, FINDEX 2011) Turkey, 45.1 Brazil, 29.2 Russian Federation, 9.6 China, 8.0 India, 1.8 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 141 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 48 Figure 47. Credit Card EACR (Individual level, FINDEX 2014) Turkey, 32.8 Brazil, 32.1 Russian Federation, 21.0 China, 15.8 India, 4.1 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage Notes: 143 countries sorted by coverage rates. Coverage rates are labeled. See dashboard for full visualization of all countries. Sources: FINDEX 49 Figure 48. Credit Card EACR (household level, LITS) Sweden Turkey 57.2 Great Britain Hungary France Slovenia Italy Czech Republic Slovak Republic Croatia Germany Latvia Macedonia All Estonia Mongolia Serbia Ukraine Montenegro Poland Kosovo Albania Bulgaria Bosnia and Herzegovina Romania Lithuania Belarus Russian Federation Armenia Georgia Kazakhstan Azerbaijan Moldova Tajikistan Kyrgyzstan Uzbekistan 0 10 20 30 40 50 60 70 80 90 100 EACR Dissimilarity Coverage 35 countries sorted by coverage rates. Coverage rates are labeled Sources: LITS 50 Poverty & Equity Global Practice Working Papers (Since July 2014) The Poverty & Equity Global Practice Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. This series is co‐published with the World Bank Policy Research Working Papers (DECOS). It is part of a larger effort by the World Bank to provide open access to its research and contribute to development policy discussions around the world. For the latest paper, visit our GP’s intranet at http://POVERTY. 1 Estimating poverty in the absence of consumption data: the case of Liberia Dabalen, A. L., Graham, E., Himelein, K., Mungai, R., September 2014 2 Female labor participation in the Arab world: some evidence from panel data in Morocco Barry, A. G., Guennouni, J., Verme, P., September 2014 3 Should income inequality be reduced and who should benefit? redistributive preferences in Europe and Central Asia Cojocaru, A., Diagne, M. F., November 2014 4 Rent imputation for welfare measurement: a review of methodologies and empirical findings Balcazar Salazar, C. F., Ceriani, L., Olivieri, S., Ranzani, M., November 2014 5 Can agricultural households farm their way out of poverty? Oseni, G., McGee, K., Dabalen, A., November 2014 6 Durable goods and poverty measurement Amendola, N., Vecchi, G., November 2014 7 Inequality stagnation in Latin America in the aftermath of the global financial crisis Cord, L., Barriga Cabanillas, O., Lucchetti, L., Rodriguez‐Castelan, C., Sousa, L. D., Valderrama, D. December 2014 8 Born with a silver spoon: inequality in educational achievement across the world Balcazar Salazar, C. F., Narayan, A., Tiwari, S., January 2015 Updated on December 2016 by POV GP KL Team | 1 9 Long‐run effects of democracy on income inequality: evidence from repeated cross‐sections Balcazar Salazar,C. F., January 2015 10 Living on the edge: vulnerability to poverty and public transfers in Mexico Ortiz‐Juarez, E., Rodriguez‐Castelan, C., De La Fuente, A., January 2015 11 Moldova: a story of upward economic mobility Davalos, M. E., Meyer, M., January 2015 12 Broken gears: the value added of higher education on teachers' academic achievement Balcazar Salazar, C. F., Nopo, H., January 2015 13 Can we measure resilience? a proposed method and evidence from countries in the Sahel Alfani, F., Dabalen, A. L., Fisker, P., Molini, V., January 2015 14 Vulnerability to malnutrition in the West African Sahel Alfani, F., Dabalen, A. L., Fisker, P., Molini, V., January 2015 15 Economic mobility in Europe and Central Asia: exploring patterns and uncovering puzzles Cancho, C., Davalos, M. E., Demarchi, G., Meyer, M., Sanchez Paramo, C., January 2015 16 Managing risk with insurance and savings: experimental evidence for male and female farm managers in the Sahel Delavallade, C., Dizon, F., Hill, R., Petraud, J. P., el., January 2015 17 Gone with the storm: rainfall shocks and household well‐being in Guatemala Baez, J. E., Lucchetti, L., Genoni, M. E., Salazar, M., January 2015 18 Handling the weather: insurance, savings, and credit in West Africa De Nicola, F., February 2015 19 The distributional impact of fiscal policy in South Africa Inchauste Comboni, M. G., Lustig, N., Maboshe, M., Purfield, C., Woolard, I., March 2015 20 Interviewer effects in subjective survey questions: evidence from Timor‐Leste Himelein, K., March 2015 21 No condition is permanent: middle class in Nigeria in the last decade Corral Rodas, P. A., Molini, V., Oseni, G. O., March 2015 22 An evaluation of the 2014 subsidy reforms in Morocco and a simulation of further reforms Verme, P., El Massnaoui, K., March 2015 Updated on December 2016 by POV GP KL Team | 2 23 The quest for subsidy reforms in Libya Araar, A., Choueiri, N., Verme, P., March 2015 24 The (non‐) effect of violence on education: evidence from the "war on drugs" in Mexico Márquez‐Padilla, F., Pérez‐Arce, F., Rodriguez Castelan, C., April 2015 25 “Missing girls” in the south Caucasus countries: trends, possible causes, and policy options Das Gupta, M., April 2015 26 Measuring inequality from top to bottom Diaz Bazan, T. V., April 2015 27 Are we confusing poverty with preferences? Van Den Boom, B., Halsema, A., Molini, V., April 2015 28 Socioeconomic impact of the crisis in north Mali on displaced people (Available in French) Etang Ndip, A., Hoogeveen, J. G., Lendorfer, J., June 2015 29 Data deprivation: another deprivation to end Serajuddin, U., Uematsu, H., Wieser, C., Yoshida, N., Dabalen, A., April 2015 30 The local socioeconomic effects of gold mining: evidence from Ghana Chuhan-Pole, P., Dabalen, A., Kotsadam, A., Sanoh, A., Tolonen, A.K., April 2015 31 Inequality of outcomes and inequality of opportunity in Tanzania Belghith, N. B. H., Zeufack, A. G., May 2015 32 How unfair is the inequality of wage earnings in Russia? estimates from panel data Tiwari, S., Lara Ibarra, G., Narayan, A., June 2015 33 Fertility transition in Turkey—who is most at risk of deciding against child arrival? Greulich, A., Dasre, A., Inan, C., June 2015 34 The socioeconomic impacts of energy reform in Tunisia: a simulation approach Cuesta Leiva, J. A., El Lahga, A., Lara Ibarra, G., June 2015 35 Energy subsidies reform in Jordan: welfare implications of different scenarios Atamanov, A., Jellema, J. R., Serajuddin, U., June 2015 36 How costly are labor gender gaps? estimates for the Balkans and Turkey Cuberes, D., Teignier, M., June 2015 37 Subjective well‐being across the lifespan in Europe and Central Asia Bauer, J. M., Munoz Boudet, A. M., Levin, V., Nie, P., Sousa‐Poza, A., July 2015 Updated on December 2016 by POV GP KL Team | 3 38 Lower bounds on inequality of opportunity and measurement error Balcazar Salazar, C. F., July 2015 39 A decade of declining earnings inequality in the Russian Federation Posadas, J., Calvo, P. A., Lopez‐Calva, L.‐F., August 2015 40 Gender gap in pay in the Russian Federation: twenty years later, still a concern Atencio, A., Posadas, J., August 2015 41 Job opportunities along the rural‐urban gradation and female labor force participation in India Chatterjee, U., Rama, M. G., Murgai, R., September 2015 42 Multidimensional poverty in Ethiopia: changes in overlapping deprivations Yigezu, B., Ambel, A. A., Mehta, P. A., September 2015 43 Are public libraries improving quality of education? when the provision of public goods is not enough Rodriguez Lesmes, P. A., Valderrama Gonzalez, D., Trujillo, J. D., September 2015 44 Understanding poverty reduction in Sri Lanka: evidence from 2002 to 2012/13 Inchauste Comboni, M. G., Ceriani, L., Olivieri, S. D., October 2015 45 A global count of the extreme poor in 2012: data issues, methodology and initial results Ferreira, F.H.G., Chen, S., Dabalen, A. L., Dikhanov, Y. M., Hamadeh, N., Jolliffe, D. M., Narayan, A., Prydz, E. B., Revenga, A. L., Sangraula, P., Serajuddin, U., Yoshida, N., October 2015 46 Exploring the sources of downward bias in measuring inequality of opportunity Lara Ibarra, G., Martinez Cruz, A. L., October 2015 47 Women’s police stations and domestic violence: evidence from Brazil Perova, E., Reynolds, S., November 2015 48 From demographic dividend to demographic burden? regional trends of population aging in Russia Matytsin, M., Moorty, L. M., Richter, K., November 2015 49 Hub‐periphery development pattern and inclusive growth: case study of Guangdong province Luo, X., Zhu, N., December 2015 50 Unpacking the MPI: a decomposition approach of changes in multidimensional poverty headcounts Rodriguez Castelan, C., Trujillo, J. D., Pérez Pérez, J. E., Valderrama, D., December 2015 51 The poverty effects of market concentration Rodriguez Castelan, C., December 2015 52 Can a small social pension promote labor force participation? evidence from the Colombia Mayor program Pfutze, T., Rodriguez Castelan, C., December 2015 Updated on December 2016 by POV GP KL Team | 4 53 Why so gloomy? perceptions of economic mobility in Europe and Central Asia Davalos, M. E., Cancho, C. A., Sanchez, C., December 2015 54 Tenure security premium in informal housing markets: a spatial hedonic analysis Nakamura, S., December 2015 55 Earnings premiums and penalties for self‐employment and informal employees around the world Newhouse, D. L., Mossaad, N., Gindling, T. H., January 2016 56 How equitable is access to finance in turkey? evidence from the latest global FINDEX Yang, J., Azevedo, J. P. W. D., Inan, O. K., January 2016 57 What are the impacts of Syrian refugees on host community welfare in Turkey? a subnational poverty analysis Yang, J., Azevedo, J. P. W. D., Inan, O. K., January 2016 58 Declining wages for college‐educated workers in Mexico: are younger or older cohorts hurt the most? Lustig, N., Campos‐Vazquez, R. M., Lopez‐Calva, L.‐F., January 2016 59 Sifting through the Data: labor markets in Haiti through a turbulent decade (2001‐2012) Rodella, A.‐S., Scot, T., February 2016 60 Drought and retribution: evidence from a large‐scale rainfall‐indexed insurance program in Mexico Fuchs Tarlovsky, Alan., Wolff, H., February 2016 61 Prices and welfare Verme, P., Araar, A., February 2016 62 Losing the gains of the past: the welfare and distributional impacts of the twin crises in Iraq 2014 Olivieri, S. D., Krishnan, N., February 2016 63 Growth, urbanization, and poverty reduction in India Ravallion, M., Murgai, R., Datt, G., February 2016 64 Why did poverty decline in India? a nonparametric decomposition exercise Murgai, R., Balcazar Salazar, C. F., Narayan, A., Desai, S., March 2016 65 Robustness of shared prosperity estimates: how different methodological choices matter Uematsu, H., Atamanov, A., Dewina, R., Nguyen, M. C., Azevedo, J. P. W. D., Wieser, C., Yoshida, N., March 2016 66 Is random forest a superior methodology for predicting poverty? an empirical assessment Stender, N., Pave Sohnesen, T., March 2016 67 When do gender wage differences emerge? a study of Azerbaijan's labor market Tiongson, E. H. R., Pastore, F., Sattar, S., March 2016 Updated on December 2016 by POV GP KL Team | 5 68 Second‐stage sampling for conflict areas: methods and implications Eckman, S., Murray, S., Himelein, K., Bauer, J., March 2016 69 Measuring poverty in Latin America and the Caribbean: methodological considerations when estimating an empirical regional poverty line Gasparini, L. C., April 2016 70 Looking back on two decades of poverty and well‐being in India Murgai, R., Narayan, A., April 2016 71 Is living in African cities expensive? Yamanaka, M., Dikhanov, Y. M., Rissanen, M. O., Harati, R., Nakamura, S., Lall, S. V., Hamadeh, N., Vigil Oliver, W., April 2016 72 Ageing and family solidarity in Europe: patterns and driving factors of intergenerational support Albertini, M., Sinha, N., May 2016 73 Crime and persistent punishment: a long‐run perspective on the links between violence and chronic poverty in Mexico Rodriguez Castelan, C., Martinez‐Cruz, A. L., Lucchetti, L. R., Valderrama Gonzalez, D., Castaneda Aguilar, R. A., Garriga, S., June 2016 74 Should I stay or should I go? internal migration and household welfare in Ghana Molini, V., Pavelesku, D., Ranzani, M., July 2016 75 Subsidy reforms in the Middle East and North Africa Region: a review Verme, P., July 2016 76 A comparative analysis of subsidy reforms in the Middle East and North Africa Region Verme, P., Araar, A., July 2016 77 All that glitters is not gold: polarization amid poverty reduction in Ghana Clementi, F., Molini, V., Schettino, F., July 2016 78 Vulnerability to Poverty in rural Malawi Mccarthy, N., Brubaker, J., De La Fuente, A., July 2016 79 The distributional impact of taxes and transfers in Poland Goraus Tanska, K. M., Inchauste Comboni, M. G., August 2016 80 Estimating poverty rates in target populations: an assessment of the simple poverty scorecard and alternative approaches Vinha, K., Rebolledo Dellepiane, M. A., Skoufias, E., Diamond, A., Gill, M., Xu, Y., August 2016 Updated on December 2016 by POV GP KL Team | 6 81 Synergies in child nutrition: interactions of food security, health and environment, and child care Skoufias, E., August 2016 82 Understanding the dynamics of labor income inequality in Latin America Rodriguez Castelan, C., Lustig, N., Valderrama, D., Lopez‐Calva, L.‐F., August 2016 83 Mobility and pathways to the middle class in Nepal Tiwari, S., Balcazar Salazar, C. F., Shidiq, A. R., September 2016 84 Constructing robust poverty trends in the Islamic Republic of Iran: 2008‐14 Salehi Isfahani, D., Atamanov, A., Mostafavi, M.‐H., Vishwanath, T., September 2016 85 Who are the poor in the developing world? Newhouse, D. L., Uematsu, H., Doan, D. T. T., Nguyen, M. C., Azevedo, J. P. W. D., Castaneda Aguilar, R. A., October 2016 86 New estimates of extreme poverty for children Newhouse, D. L., Suarez Becerra, P., Evans, M. C., October 2016 87 Shedding light: understanding energy efficiency and electricity reliability Carranza, E., Meeks, R., November 2016 88 Heterogeneous returns to income diversification: evidence from Nigeria Siwatu, G. O., Corral Rodas, P. A., Bertoni, E., Molini, V., November 2016 89 How liberal is Nepal's liberal grade promotion policy? Sharma, D., November 2016 90 CPI bias and its implications for poverty reduction in Africa Dabalen, A. L., Gaddis, I., Nguyen, N. T. V., December 2016 91 Pro-growth equity: a policy framework for the twin goals Lopez-Calva, L. F., Rodriguez Castelan, C., November 2016 92 Building an ex ante simulation model for estimating the capacity impact, benefit incidence, and cost effectiveness of child care subsidies: an application using provider‐level data from Turkey Aran, M. A., Munoz Boudet, A., Aktakke, N., December 2016 93 Vulnerability to drought and food price shocks: evidence from Ethiopia Porter, C., Hill, R., December 2016 94 Job quality and poverty in Latin America Rodriguez Castelan, C., Mann, C. R., Brummund, P., December 2016 Updated on December 2016 by POV GP KL Team | 7 For the latest and sortable directory, available on the Poverty & Equity GP intranet site. http://POVERTY WWW.WORLDBANK.ORG/POVERTY Updated on December 2016 by POV GP KL Team | 8