Abstract
Selected Issues
Dimensions of Inequality in Estonia1
This paper examines the different dimensions of income inequality in Estonia for a basis of policy discussions. While income inequality in Estonia has decreased over recent years, it remains high compared to the Euro area average and close to those of other Baltic countries. Using mainly quantitative approaches applied to household survey data and other sources, the paper assesses the main factors behind overall inequality and the gender pay gap (GPG). The paper also explores the potential role of institutions in explaining the large unexplained part of the GPG in Estonia. Then, the paper discusses policy options to address inequality based on an empirical analysis.
A. Income Inequality in Estonia: Current State and Recent Trend
1. Despite progress over the last decade, income inequality and relative poverty remain elevated in Estonia. After a surge over the period 2008 to 2014, income inequality has come to a declining trend, albeit close to other Baltic countries but above the EU average (Figure 1).2 Indeed, income inequality, as measured by the Gini coefficient was among the highest in Europe in 2017. Eurostat estimates the Gini index for disposable income at 31.6 points in 2017, slightly above the EA19 average of 30.6 points.3 Relative poverty rate, defined as the proportion of the population with a disposable income lower than 60 percent of the median disposable equivalized income, despite a small decline since 2016, was also among the highest in Europe. 2017.4
Income Inequality vs. Peers
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
This is the at-risk-of-poverty rate which Eurostat estimates in 2017 at 21 percent (higher than the EU27 average of 17 percent). Over the last decade, the at-risk-of-poverty rate has significantly been above EU average except in 2010 mainly because of a decline in income.
2. Inequality has been particularly elevated on gender terms. First, the gender pay gap has decreased since 2008 but remains elevated across most sectors in the economy, especially in sectors with high female concentration.5 Sectors such as human health and social work activities, manufacturing and wholesale and retail trade have high gender pay gap and employ largely women in 2016.6 This pervasive gender pay gap seems to put women at a high degree of vulnerability to poverty. The persistent at-risk-of-poverty has increased over the last decade in Estonia.7 In 2017, the proportion of the population that was persistently at risk of poverty was 16 percent compared to the EU28 average of 11 percent. However, this proportion is higher for women: in 2017 this proportion represents 18 percent, well above the EU27 women average of 11.6 percent.8
Overall Gender Pay Gap
(In Percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Source: Eurostat.3. Old-age population has been also vulnerable to poverty. Despite the rise in income across generations, the elderly has become particularly vulnerable. Real median equivalized disposable income has increased over the last decade for all age groups but the proportion of the elderly that are at-risk of poverty, has reached an historical high of 41 percent. In addition, elderly poverty has diverged from other groups particularly since 2013.
At-Risk-of-Poverty Rate by Age Groups
(In Percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Source: Eurostat.4. Despite the level of inequality, a large part of the Estonians has experienced real income growth over the last two decades. Real GDP per capita almost doubled between 2000 and 2018 (from euros 7,600 to euros 15,100). In addition, average monthly wages have grown on average by 5.9 percent over the period 2010–18. Also, the population in poverty as measured by the headcount ratio decreased from 2.5 percent to only 0.5 percent in 2015, equivalent to 80 percent reduction.9 However an analysis of household income distribution shows that the largest gains accrued to the upper shares as shown in the growth incidence curve. Households’ incomes have grown on average by 6.1 percent over the period 2010–17. But, this overall average growth rate hides some disparities. Income has grown by 5.6 percent for the bottom 10 percent (below the average), while the seventh decile has registered the highest growth (7.2 percent).
Estonia: Growth Incidence Curve (2010–2017)
(In percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Statistics Estonia and IMF Staff calculations5. Against this backdrop, we will now look into the drivers shaping the trend and dimensions of inequality.
B. Drivers of the Trend in Inequality in Estonia
6. Using household level data over 2010–17, we analyze the dimensions of income inequality in Estonia. First, we calculate the Theil index, which is primarily used to measure economic inequality and other economic phenomena, to examine the contribution of education, age cohort, economic status, the geographical gap (rural/ urban and county of residence) and gender. The Theil index is an index calculated as follows:
Where yi and
7. We then decompose income inequality by different subgroups. We exploit the desirable property of the Theil index to separate total inequality into a component that is due to within group inequality and the component related to differences between groups. If the sample of households consist of G subgroups and that the Theil index of each group is Tg , the index for the sample can be formulated as:
with sg and pg are respectively the income share and the share in total number of households in the sample of subgroup g. The first summation term is the average of the Theil indexes of all subgroups weighted by their respective income shares sg and captures the component of the overall inequality that stems from within group inequality. The second summation term represents the calculation of the Theil index on the mean income of each subgroup and captures the part of overall inequality that is due to between groups inequality.
8. We consider seven types of decomposition over 2010–17. We split the sample into different subgroups according to residence of the household (the county and rural-urban location), the gender, the educational attainment, the age cohort, the economic status and the economic sector of the head of the household. Sixteen counties are considered while educational attainment is classified into six subcategories (no formal education or below ISCED1,10 primary education -ISCED1, lower secondary- ISCED2, upper secondary- ISCED3, post-secondary but not tertiary-ISCED4, tertiary education first stage-ISCED5 and tertiary education second stage-ISCED6). We consider four age groups (15–24 years old, 25–54 years old, 55–64 years old and over 65 years old). Next, four economic status are examined (employee, self-employed, retired and other nonactive). Finally, we account for the economic sector of the head of the household through six categories.
9. Differences in economic sector, economic status, age cohorts, education, the county of residence and gender have been important drivers of income inequality in Estonia. The rise in inequality between 2010 and 2013/2014, and the modest decline in 2017 are captured by these five dimensions with a prominent role for the economic status, the economic sector, the level of education and the age cohort.
Economic sector and the economic status of the head of the household. Difference in earnings between the six groups of economic sectors (agriculture, forestry and fishing, industry including energy, construction, wholesale and retail trade, financial and real-estate, and, public sector and other services) accounts for 26.7 percent of the recent trend in inequality. Between-group inequality as captured by the economic status has grown by 43 percent between 2013 and 2010 while the pace was down to 19 percent between 2010 and 2017. Earnings difference between employee, self-employed, retired and other nonactive contributed to the recent trend at the same magnitude as the components of the economic sector. Education and age cohort of the head of the household. Education accounts for the second largest factor by explaining about 14.7 percent of the recent trend in inequality in Estonia. Income difference across age cohorts (15 to 24, 25 to 54, 55 to 64 an over 65 years old) accounts for 13.5 percent of this trend.
County of residence, gender of the head of the household. They respectively account for the recent trend in inequality by 8.4 percent and 5 percent on average. The difference in income, based on the residency in rural versus urban areas is very insignificant while it has slightly declined below its 2010 level regarding the county of residence.
Estonia: Theil Index – Income Inequality Decomposition
(2010–2017)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Statistics Estonia and IMF staff calculationsContribution to the Trend in Inequality
(In percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Statistics Estonia and IMF 5 taff calculationsC The Analysis of the Gender Pay Gap in Estonia: Recent Trend and Explanation
How Large is the GPG in Estonia?
10. We investigate the gender pay gap (GPG) dynamics using household level data from Estonia Social Survey.11 Our approach follows Blau and Kahn (2017). Table 1 shows both the evolution of the female to male ratio of net average monthly wage and the unadjusted GPG at the mean but also at the tenth, the fiftieth and the ninetieth percentiles over the period 2010–17.
Estonia: Unadjusted GPG and Female to Male Log Net Monthly Wage Ratio, Full-time Workers
We exclude from the sample households reporting average net monthly wage below the minimum wage for a given year. With this restriction, we lose the year 2011 which features very low-income levels.
Estonia: Unadjusted GPG and Female to Male Log Net Monthly Wage Ratio, Full-time Workers
| Sample size | ||||||
|---|---|---|---|---|---|---|
| Year | Men | Women | Mean | 10th percentile | 50th percentile | 90th percentile |
| Panel A: Unadjusted Raw GPG | ||||||
| 2010 | 1582 | 1603 | 25.6% | 11.3% | 28.8% | 28.8% |
| 2012 | 1717 | 1686 | 30.0% | 22.3% | 33.6% | 40.5% |
| 2013 | 1870 | 1749 | 31.5% | 21.2% | 31.0% | 43.8% |
| 2014 | 1939 | 1788 | 26.1% | 22.3% | 28.8% | 37.5% |
| 2015 | 1838 | 1735 | 25.2% | 17.4% | 19.4% | 33.2% |
| 2016 | 1876 | 1711 | 24.2% | 9.5% | 28.8% | 37.9% |
| 2017 | 1969 | 1887 | 18.5% | 18.2% | 22.3% | 26.0% |
| Panel B: Unadjusted Female/Male log net Monthly Wage | ||||||
| 2010 | 1582 | 1603 | 77.4% | 89.3% | 75.0% | 75.0% |
| 2012 | 1717 | 1686 | 74.1% | 80.0% | 71.4% | 66.7% |
| 2013 | 1870 | 1749 | 73.0% | 80.9% | 73.3% | 64.5% |
| 2014 | 1939 | 1788 | 77.1% | 80.0% | 75.0% | 68.8% |
| 2015 | 1838 | 1735 | 77.7% | 84.0% | 82.4% | 71.8% |
| 2016 | 1876 | 1711 | 78.5% | 90.9% | 75.0% | 68.4% |
| 2017 | 1969 | 1887 | 83.1% | 83.3% | 80.0% | 77.1% |
We exclude from the sample households reporting average net monthly wage below the minimum wage for a given year. With this restriction, we lose the year 2011 which features very low-income levels.
Estonia: Unadjusted GPG and Female to Male Log Net Monthly Wage Ratio, Full-time Workers
| Sample size | ||||||
|---|---|---|---|---|---|---|
| Year | Men | Women | Mean | 10th percentile | 50th percentile | 90th percentile |
| Panel A: Unadjusted Raw GPG | ||||||
| 2010 | 1582 | 1603 | 25.6% | 11.3% | 28.8% | 28.8% |
| 2012 | 1717 | 1686 | 30.0% | 22.3% | 33.6% | 40.5% |
| 2013 | 1870 | 1749 | 31.5% | 21.2% | 31.0% | 43.8% |
| 2014 | 1939 | 1788 | 26.1% | 22.3% | 28.8% | 37.5% |
| 2015 | 1838 | 1735 | 25.2% | 17.4% | 19.4% | 33.2% |
| 2016 | 1876 | 1711 | 24.2% | 9.5% | 28.8% | 37.9% |
| 2017 | 1969 | 1887 | 18.5% | 18.2% | 22.3% | 26.0% |
| Panel B: Unadjusted Female/Male log net Monthly Wage | ||||||
| 2010 | 1582 | 1603 | 77.4% | 89.3% | 75.0% | 75.0% |
| 2012 | 1717 | 1686 | 74.1% | 80.0% | 71.4% | 66.7% |
| 2013 | 1870 | 1749 | 73.0% | 80.9% | 73.3% | 64.5% |
| 2014 | 1939 | 1788 | 77.1% | 80.0% | 75.0% | 68.8% |
| 2015 | 1838 | 1735 | 77.7% | 84.0% | 82.4% | 71.8% |
| 2016 | 1876 | 1711 | 78.5% | 90.9% | 75.0% | 68.4% |
| 2017 | 1969 | 1887 | 83.1% | 83.3% | 80.0% | 77.1% |
We exclude from the sample households reporting average net monthly wage below the minimum wage for a given year. With this restriction, we lose the year 2011 which features very low-income levels.
11. The unadjusted GPG (at the mean) has decreased over the recent years. The GPG has decreased by 28 percent between 2010 and 2017 at the mean. After the GPG has reached its pic, rising by 23 percent in 2010, it started to decline. This declining trend translates into a rapid increase in female to male wage ratio from 77 percent to 83 percent. In 2017, thus, women earned a net monthly wage representing 83 percent of the earning of men while this ratio was 77 percent in 2010. Further, the GPG is on average lower at the bottom of the distribution compared to its top.12 The rise in the minimum income by 9 percent on average since 2012 might have contributed to the reduction in the GPG at the bottom.13
The Gender Pay Gap and Female to Male Earning Ratio, Full time workers
(In percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Statistics Estonia end IMF staff calculationsWhat Drives the Pay Gap between Men and Women in Estonia?
12. Estonia Social Survey data shed some light on the GPG level. Using the survey, we look into the impact of differences in human capital, employment location, industry or sector and occupation in employment on the evolution of the gender pay gap over the period 2010–16 following Blau and Kahn (2017).14 We estimate Mincer-type wage equation for male and females separately where human capital is measured by the level of education, experience and health. We also control for employment location by accounting for the county and rural/urban location, whether the individual has a managerial or supervisory position. Control variables also include disability status to account for any potential discrimination.15
13. The adjusted GPG and the female to male log wage ratio are respectively lower and higher than the unadjusted metrics. We use the standard Oaxaca-Blinder decomposition of males/female differences in characteristics and an unexplained component which can be used as a proxy of the extent of biased practices. This unexplained component could also be interpreted as capturing the extent to which females and men are unequally paid while they are equally qualified; it could be viewed as including also compensation scheme differentials or unmeasured productivity. However, including the assessment of health condition may help reduce the portion of unmeasured productivity in the unexplained portion. Once adjusted, the female to male earnings ratio increased by 3 percent while the GGP decreased by 11 percent on average over 2010–16.
Gender Pay Gap
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
14. The unexplained part of the GPG remain large in Estonia. Consistently with previous studies, the unexplained part of the GPG is high.16 On average, about 86 percent of the GPG cannot be explained over the period.17 Accounting for more structural factors such as biased practices against women and/or a better measurement of productivity differences would contribute to better understand the GPG in Estonia. 18
Unexplained Versus Explained Part of the Gender Pay Gap
(In percent)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Statistics Estonia and IMF Staff estimates.15. The large GPG may reflect labor market institutions in Estonia. Given the large unexplained part of the GPG, we explore a potential role for labor market institutions in Estonia in line with previous studies (Kahn, 2015). High minimum wages may contribute to reduce the GPG as women are often disproportionately represented in the bottom of the wage distribution.19 Collective wage bargaining agreement could also contribute to reduce the GPG by allowing to set wage floors and thus raising the wages of the low-paid workers (thus female wages). Using the most recent data available, Figure 3 above shows that Estonia labor market is characterized by relatively low minimum wage and collective bargaining coverage which may have contributed to the large GPG. In addition to these labor market institutions, social protection institutions such as long parental leave for may also have an influence on the GPG by keeping women a long time away from work.20
Labor Market Institutions and the Gender Pay Gap in Europe
(most recent data)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: Eurostat;OECD;and Employment Protection Legislation: Strictness of employment protection legislation: regular employment”, OECD Employment and Labour Market Statistics (database), https://doi.org/10.1787/data-00318-en (accessedon 25 July 2019). 1/2014 data is used for Croatia, lreland,Greece,and Hungary.D. Policies to Reduce Inequality in Estonia
16. The role of fiscal policy to reduce inequality can be explored using an empirical approach. We estimate a cross-country fixed effect model using an approach similar to Jain-Chandra et al (2018) and Chen et al (2018). The model specification is as follows:
Where the dependent variable (Ginict) is the Gini index of equivalized disposable income of country c in year t, Policyct-1 is a vector of lagged policy variables focused on the role of fiscal policy (public social protection expenditure as share of GDP and property tax revenues as share of GDP), Xct is a set of control variables capturing some structural characteristics and their nonlinear forms, αc is a country fixed effect and τt is a time fixed effect21 We use data on 21 European countries over the period 2004–17. 22
17. We estimate the impact of policies on inequality in Estonia using the specification including nonlinear forms of structural variables which capture well the trend of inequality in Estonia. We estimate two versions of equation (3) (See Appendix II) and select the model that captures better the trend in the Gini index (model 2).23 Model 2 accounts for nonlinearities of control variables.
Gini Index, Data and Fitted Values
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: IMF staff estimates18. Our estimates suggest large gains from increasing social protection spending. We use the previous econometric model to predict the Gini index and estimate the impact of desirable level of the policy variables, keeping all the other variables constant. We define desirables policies as the values of the policy variables at the EU28 average. The results show that increasing the property tax revenues and social protection spending at the EU28 average over the period would have led to Gini index lower by 3 percent and 11.3 percent respectively.24
Policy Impact on Gini Index (2010–17)
(Average, in points)
Citation: IMF Staff Country Reports 2020, 013; 10.5089/9781513526911.002.A002
Sources: IMF Staff estimatesE. Conclusions and Policy Discussion
19. Over the last years Estonian households have experienced a considerable income growth especially at the upper part of the income distribution. Income inequality remains elevated with two main aspects related to the gender pay gap and old age poverty.
20. Income difference between the economic sector of activity, the economic status, age cohort and the level of education account mainly for the recent trend in inequality. At the same time, income inequality between rural and urban areas appear to have reduced recently.
21. The gender pay gap remains elevated and is mainly driven by the top of the wage distribution, while a significant portion cannot be explained by the data. The large GPG may also reflect some institutional characteristics of Estonia’s labor market such as the low collective bargaining power and the low relatively minimum wage but also long parental leave that may keep women durably from the labor market. Policies for more transparency and for stronger reduction of biases should be enhanced: (i) a transparent reporting policy of gender pay gap, as already applied by the authorities in some public entities, extended to other sectors would be a significant step; and (ii) less occupational gender biases and career interruptions by women may help reduce the GPG without raising female unemployment.
22. Other policies could be seen to contrast with the conservative optional policies prevailing in Estonia, however they could be given some considerations as increasing social protection spending and broadening the tax base through an intelligent system of property tax could contribute to reduce income inequality. Given its relatively low level with peers, increasing social protection spending bears the potential of reducing income inequality. More specifically, social protection spending (including pensions) on old age population would help to widen the social safety net.
References
Addison, J. T., and Ozturk, O. D., 2012, “Minimum Wages, Labor Market Institutions, and Female Employment: A Cross-country Analysis,” ILR Review, Vol. 65, Iss. No. 4, pp.779–809 (New York: Cornell University).
Anspal, S., 2015, “Gender Wage Gap in Estonia: A Non-Parametric Decomposition,” Baltic Journal of Economics, Vol. 15, No. 1, pp. 1–16,
Anspal, S., Rõõm, T., Anspal, S., Kraut, L. and Rõõm, T., 2011, “Gender Pay Gap in Estonia: Empirical Analysis,” Report for the Estonian Ministry of Social Affairs (Tallinn).
Bertola, G., Blau, F. D., and Kahn, L. M., 2007, “Labor Market Institutions and Demographic Employment Patterns,” Journal of Population Economics, Vol. 20, Issue No. 4, pp. 833–867.
Blau, F. D., and Kahn, L. M., 2017, “The Gender Wage Gap: Extent, Trends, and Explanations,” Journal of Economic Literature, Vol. 55, Issue No. 3, pp. 789–865.
Chen, T., Hallaert, M. J. J., Pitt, M. A., Qu, M. H., Queyranne, M. M., Rhee, A., and Yackovlev, I., 2018, “Inequality and Poverty Across Generations in the European Union,” IMF Staff Discussion Notes 18/01, (Washington: International Monetary Fund.)
Ferraro, S., Meriküll, J., and Staehr, K., 2018, “Minimum Wages and the Wage Distribution in Estonia,” Applied Economics, Vol. 50, Issue No. 49, pp. 5253–5268.
Jain-Chandra, M. S., Khor, N., Mano, R., Schauer, J., Wingender, M.P. and Zhuang, J., 2018, “Inequality in China–Trends, Drivers and Policy Remedies,” IMF Working Paper 18/127 (Washington: International Monetary Fund).
Kahn, L. M., 2007, “The Impact of Employment Protection Mandates on Demographic Temporary Employment Patterns: International Microeconomic Evidence,” The Economic Journal, Vol. 117, Issue 521, pp. F333–F356 (Hoboken: Blackwell Publishing).
Ñopo, H., 2008, “Matching as a Tool to Decompose Wage Gaps,” The Review of Economics and Statistics, Vol. 90, Issue 2, pp. 290–299 (Cambridge: MIT Press).
Majchrowska, A., and Strawiński, P., 2018, “Impact of Minimum Wage Increase on Gender Wage Gap: Case of Poland,” Economic Modelling, Vol. 70, pp. 174–185 (Philadelphia: Elsevier B.V.).
Meriküll, J., and Mõtsmees, P., 2017, “Do You Get What You Ask? The Gender Gap in Desired and Realised Wages,” International Journal of Manpower, Vol. 38, Issue 6, pp. 893–908 (Bingley: Emerald Publishing Limited).
Meriküll, J., Kukk, M. and Rõõm, T., (2019). “What Explains the Gender Gap in Wealth? Evidence from Administrative Data,” Working Papers of Eesti Pank 4/2019 (Tallinn: Bank of Estonia).
Appendix I. Theil Index Calculation
Agriculture, hunting and forestry; fishing and operation of fish hatcheries and fish farms;
Industry, including energy;
| Education | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| No formal education | 0.0874 | 0.0567 | 0.0445 | 0.0374 | 0.0265 | 0.0178 | 0.0172 | 0.0146 |
| Primary | 0.0494 | 0.0451 | 0.0540 | 0.0535 | 0.0658 | 0.0541 | 0.0596 | 0.0386 |
| Lower secondary | 0.1186 | 0.1220 | 0.1339 | 0.1495 | 0.1919 | 0.1499 | 0.1385 | 0.1291 |
| Upper secondary | 0.1366 | 0.1397 | 0.1843 | 0.1967 | 0.1873 | 0.1756 | 0.1761 | 0.1514 |
| Post-secondary | 0.1161 | 0.1182 | 0.1275 | 0.1731 | 0.2124 | 0.1665 | 0.1667 | 0.1459 |
| Tertiary education (1) | 0.1203 | 0.1292 | 0.1668 | 0.2171 | 0.2049 | 0.1949 | 0.1645 | 0.1510 |
| Tertiary education (2) | 0.0665 | 0.0302 | 0.0884 | 0.1459 | 0.1785 | 0.1618 | 0.1236 | 0.1086 |
| Within | 0.1250 | 0.1300 | 0.1680 | 0.1990 | 0.1950 | 0.1810 | 0.1650 | 0.1480 |
| Between | 0.0234 | 0.0240 | 0.0247 | 0.0325 | 0.0284 | 0.0349 | 0.0264 | 0.0249 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Economic status | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Employee | 0.1095 | 0.1137 | 0.1556 | 0.1843 | 0.1773 | 0.1646 | 0.1425 | 0.1188 |
| Self employed | 0.2282 | 0.3866 | 0.2927 | 0.3258 | 0.3595 | 0.3725 | 0.3309 | 0.3504 |
| Retired | 0.0460 | 0.0418 | 0.0463 | 0.0700 | 0.0900 | 0.0821 | 0.0756 | 0.0896 |
| Other nonactive | 0.2189 | 0.2125 | 0.2302 | 0.3248 | 0.3098 | 0.3191 | 0.2878 | 0.3013 |
| Within | 0.1080 | 0.1100 | 0.1430 | 0.1730 | 0.1720 | 0.1610 | 0.1410 | 0.1240 |
| Between | 0.0404 | 0.0440 | 0.0493 | 0.0579 | 0.0510 | 0.0545 | 0.0510 | 0.0481 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Urban/Rural | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Urban | 0.1422 | 0.1496 | 0.1804 | 0.2292 | 0.2208 | 0.2108 | 0.1898 | 0.1705 |
| Rural | 0.1597 | 0.1633 | 0.2217 | 0.2340 | 0.2284 | 0.2260 | 0.1958 | 0.1765 |
| Within | 0.1470 | 0.1530 | 0.1920 | 0.2300 | 0.2230 | 0.2150 | 0.1910 | 0.1720 |
| Between | 0.0018 | 0.0010 | 0.0005 | 0.0008 | 0.0003 | 0.0005 | 0.0004 | 0.0004 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| County of residence | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Tallinn | 0.1360 | 0.1340 | 0.1748 | 0.2142 | 0.2316 | 0.2161 | 0.1822 | 0.1622 |
| Harjumaa | 0.1326 | 0.1319 | 0.2295 | 0.2693 | 0.1998 | 0.2211 | 0.1823 | 0.1640 |
| Hiiumaa | 0.1347 | 0.1736 | 0.1622 | 0.1842 | 0.1857 | 0.1802 | 0.2016 | 0.1485 |
| Ida-Virumaa | 0.1284 | 0.1461 | 0.1405 | 0.1847 | 0.1756 | 0.1949 | 0.1657 | 0.1511 |
| Jogevamaa | 0.1587 | 0.1382 | 0.1561 | 0.1483 | 0.1853 | 0.1733 | 0.1584 | 0.1441 |
| Jarvamaa | 0.1485 | 0.1493 | 0.1379 | 0.1862 | 0.1901 | 0.1689 | 0.1505 | 0.1550 |
| Laanemaa | 0.1208 | 0.1247 | 0.1643 | 0.1978 | 0.2169 | 0.1904 | 0.1461 | 0.1689 |
| Laane-Virumaa | 0.1208 | 0.1516 | 0.1786 | 0.1985 | 0.1766 | 0.1559 | 0.1778 | 0.1681 |
| Polvamaa | 0.1145 | 0.1362 | 0.1383 | 0.2067 | 0.2092 | 0.1847 | 0.1236 | 0.1466 |
| Parnumaa | 0.1373 | 0.1350 | 0.1529 | 0.1611 | 0.1662 | 0.1811 | 0.1673 | 0.1457 |
| Raplamaa | 0.1562 | 0.1715 | 0.2032 | 0.1983 | 0.1858 | 0.1893 | 0.1649 | 0.1435 |
| Saaremaa | 0.1081 | 0.1262 | 0.1403 | 0.1608 | 0.1441 | 0.1644 | 0.1648 | 0.1452 |
| Tartumaa | 0.1414 | 0.1562 | 0.1952 | 0.2531 | 0.2210 | 0.1988 | 0.1929 | 0.1781 |
| Valgamaa | 0.1398 | 0.1420 | 0.1216 | 0.1529 | 0.1858 | 0.1674 | 0.1556 | 0.1442 |
| Viljandimaa | 0.1067 | 0.1329 | 0.1508 | 0.1926 | 0.1800 | 0.1718 | 0.1549 | 0.1525 |
| Vorumaa | 0.1379 | 0.1359 | 0.1091 | 0.1598 | 0.2163 | 0.1789 | 0.2101 | 0.1690 |
| Within | 0.1340 | 0.1400 | 0.1740 | 0.2120 | 0.2080 | 0.2010 | 0.1770 | 0.1610 |
| Between | 0.0144 | 0.0144 | 0.0187 | 0.0198 | 0.0157 | 0.0148 | 0.0150 | 0.0121 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Gender | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Male | 0.138181 | 0.14787 | 0.181055 | 0.23483 | 0.210411 | 0.204671 | 0.184357 | 0.160909 |
| Female | 0.140235 | 0.147322 | 0.183589 | 0.200648 | 0.214667 | 0.212765 | 0.183991 | 0.171243 |
| Within | 0.1390 | 0.1480 | 0.1820 | 0.2210 | 0.2120 | 0.2080 | 0.1840 | 0.1650 |
| Between | 0.0095 | 0.0067 | 0.0101 | 0.0107 | 0.0110 | 0.0073 | 0.0077 | 0.0072 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Age groups | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| 15–24 | 0.1773 | 0.2251 | 0.2333 | 0.2436 | 0.2223 | 0.1834 | 0.2188 | 0.2217 |
| 25–54 | 0.1385 | 0.1517 | 0.1810 | 0.2144 | 0.1973 | 0.1898 | 0.1684 | 0.1401 |
| 55–64 | 0.1609 | 0.1514 | 0.2082 | 0.2216 | 0.2523 | 0.2309 | 0.1709 | 0.1629 |
| >65 | 0.0838 | 0.0728 | 0.0756 | 0.1097 | 0.1131 | 0.1304 | 0.1198 | 0.1333 |
| Within | 0.1330 | 0.1390 | 0.1690 | 0.1980 | 0.1920 | 0.1860 | 0.1610 | 0.1450 |
| Between | 0.0154 | 0.0155 | 0.0235 | 0.0329 | 0.0314 | 0.0295 | 0.0306 | 0.0277 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Economic Sector | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Agriculture, Forestry and Fishing (A) | 0.1344 | 0.1352 | 0.1246 | 0.2148 | 0.2116 | 0.1660 | 0.1475 | 0.1348 |
| Industry including energy | 0.0994 | 0.1040 | 0.1332 | 0.1633 | 0.1370 | 0.1360 | 0.1063 | 0.0891 |
| Construction (F) | 0.1007 | 0.1058 | 0.1906 | 0.2108 | 0.1848 | 0.1518 | 0.1753 | 0.1300 |
| Wholesale and retail trade (G,H, J) | 0.1212 | 0.1245 | 0.1697 | 0.2160 | 0.1904 | 0.1909 | 0.1554 | 0.1346 |
| Financial, real-estate (K,L) | 0.0966 | 0.0973 | 0.1584 | 0.1938 | 0.2102 | 0.1902 | 0.1729 | 0.1407 |
| Public sector and other services(O,P,Q,T) | 0.1021 | 0.1070 | 0.1358 | 0.1530 | 0.1758 | 0.1472 | 0.1261 | 0.1098 |
| Within | 0.1080 | 0.1090 | 0.1390 | 0.1720 | 0.1700 | 0.1600 | 0.1390 | 0.1250 |
| Between | 0.0405 | 0.0448 | 0.0530 | 0.0596 | 0.0537 | 0.0561 | 0.0526 | 0.0480 |
| Total | 0.1480 | 0.1530 | 0.1920 | 0.2320 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
Agriculture, hunting and forestry; fishing and operation of fish hatcheries and fish farms;
Industry, including energy;
| Education | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| No formal education | 0.0874 | 0.0567 | 0.0445 | 0.0374 | 0.0265 | 0.0178 | 0.0172 | 0.0146 |
| Primary | 0.0494 | 0.0451 | 0.0540 | 0.0535 | 0.0658 | 0.0541 | 0.0596 | 0.0386 |
| Lower secondary | 0.1186 | 0.1220 | 0.1339 | 0.1495 | 0.1919 | 0.1499 | 0.1385 | 0.1291 |
| Upper secondary | 0.1366 | 0.1397 | 0.1843 | 0.1967 | 0.1873 | 0.1756 | 0.1761 | 0.1514 |
| Post-secondary | 0.1161 | 0.1182 | 0.1275 | 0.1731 | 0.2124 | 0.1665 | 0.1667 | 0.1459 |
| Tertiary education (1) | 0.1203 | 0.1292 | 0.1668 | 0.2171 | 0.2049 | 0.1949 | 0.1645 | 0.1510 |
| Tertiary education (2) | 0.0665 | 0.0302 | 0.0884 | 0.1459 | 0.1785 | 0.1618 | 0.1236 | 0.1086 |
| Within | 0.1250 | 0.1300 | 0.1680 | 0.1990 | 0.1950 | 0.1810 | 0.1650 | 0.1480 |
| Between | 0.0234 | 0.0240 | 0.0247 | 0.0325 | 0.0284 | 0.0349 | 0.0264 | 0.0249 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Economic status | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Employee | 0.1095 | 0.1137 | 0.1556 | 0.1843 | 0.1773 | 0.1646 | 0.1425 | 0.1188 |
| Self employed | 0.2282 | 0.3866 | 0.2927 | 0.3258 | 0.3595 | 0.3725 | 0.3309 | 0.3504 |
| Retired | 0.0460 | 0.0418 | 0.0463 | 0.0700 | 0.0900 | 0.0821 | 0.0756 | 0.0896 |
| Other nonactive | 0.2189 | 0.2125 | 0.2302 | 0.3248 | 0.3098 | 0.3191 | 0.2878 | 0.3013 |
| Within | 0.1080 | 0.1100 | 0.1430 | 0.1730 | 0.1720 | 0.1610 | 0.1410 | 0.1240 |
| Between | 0.0404 | 0.0440 | 0.0493 | 0.0579 | 0.0510 | 0.0545 | 0.0510 | 0.0481 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Urban/Rural | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Urban | 0.1422 | 0.1496 | 0.1804 | 0.2292 | 0.2208 | 0.2108 | 0.1898 | 0.1705 |
| Rural | 0.1597 | 0.1633 | 0.2217 | 0.2340 | 0.2284 | 0.2260 | 0.1958 | 0.1765 |
| Within | 0.1470 | 0.1530 | 0.1920 | 0.2300 | 0.2230 | 0.2150 | 0.1910 | 0.1720 |
| Between | 0.0018 | 0.0010 | 0.0005 | 0.0008 | 0.0003 | 0.0005 | 0.0004 | 0.0004 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| County of residence | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Tallinn | 0.1360 | 0.1340 | 0.1748 | 0.2142 | 0.2316 | 0.2161 | 0.1822 | 0.1622 |
| Harjumaa | 0.1326 | 0.1319 | 0.2295 | 0.2693 | 0.1998 | 0.2211 | 0.1823 | 0.1640 |
| Hiiumaa | 0.1347 | 0.1736 | 0.1622 | 0.1842 | 0.1857 | 0.1802 | 0.2016 | 0.1485 |
| Ida-Virumaa | 0.1284 | 0.1461 | 0.1405 | 0.1847 | 0.1756 | 0.1949 | 0.1657 | 0.1511 |
| Jogevamaa | 0.1587 | 0.1382 | 0.1561 | 0.1483 | 0.1853 | 0.1733 | 0.1584 | 0.1441 |
| Jarvamaa | 0.1485 | 0.1493 | 0.1379 | 0.1862 | 0.1901 | 0.1689 | 0.1505 | 0.1550 |
| Laanemaa | 0.1208 | 0.1247 | 0.1643 | 0.1978 | 0.2169 | 0.1904 | 0.1461 | 0.1689 |
| Laane-Virumaa | 0.1208 | 0.1516 | 0.1786 | 0.1985 | 0.1766 | 0.1559 | 0.1778 | 0.1681 |
| Polvamaa | 0.1145 | 0.1362 | 0.1383 | 0.2067 | 0.2092 | 0.1847 | 0.1236 | 0.1466 |
| Parnumaa | 0.1373 | 0.1350 | 0.1529 | 0.1611 | 0.1662 | 0.1811 | 0.1673 | 0.1457 |
| Raplamaa | 0.1562 | 0.1715 | 0.2032 | 0.1983 | 0.1858 | 0.1893 | 0.1649 | 0.1435 |
| Saaremaa | 0.1081 | 0.1262 | 0.1403 | 0.1608 | 0.1441 | 0.1644 | 0.1648 | 0.1452 |
| Tartumaa | 0.1414 | 0.1562 | 0.1952 | 0.2531 | 0.2210 | 0.1988 | 0.1929 | 0.1781 |
| Valgamaa | 0.1398 | 0.1420 | 0.1216 | 0.1529 | 0.1858 | 0.1674 | 0.1556 | 0.1442 |
| Viljandimaa | 0.1067 | 0.1329 | 0.1508 | 0.1926 | 0.1800 | 0.1718 | 0.1549 | 0.1525 |
| Vorumaa | 0.1379 | 0.1359 | 0.1091 | 0.1598 | 0.2163 | 0.1789 | 0.2101 | 0.1690 |
| Within | 0.1340 | 0.1400 | 0.1740 | 0.2120 | 0.2080 | 0.2010 | 0.1770 | 0.1610 |
| Between | 0.0144 | 0.0144 | 0.0187 | 0.0198 | 0.0157 | 0.0148 | 0.0150 | 0.0121 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Gender | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Male | 0.138181 | 0.14787 | 0.181055 | 0.23483 | 0.210411 | 0.204671 | 0.184357 | 0.160909 |
| Female | 0.140235 | 0.147322 | 0.183589 | 0.200648 | 0.214667 | 0.212765 | 0.183991 | 0.171243 |
| Within | 0.1390 | 0.1480 | 0.1820 | 0.2210 | 0.2120 | 0.2080 | 0.1840 | 0.1650 |
| Between | 0.0095 | 0.0067 | 0.0101 | 0.0107 | 0.0110 | 0.0073 | 0.0077 | 0.0072 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Age groups | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| 15–24 | 0.1773 | 0.2251 | 0.2333 | 0.2436 | 0.2223 | 0.1834 | 0.2188 | 0.2217 |
| 25–54 | 0.1385 | 0.1517 | 0.1810 | 0.2144 | 0.1973 | 0.1898 | 0.1684 | 0.1401 |
| 55–64 | 0.1609 | 0.1514 | 0.2082 | 0.2216 | 0.2523 | 0.2309 | 0.1709 | 0.1629 |
| >65 | 0.0838 | 0.0728 | 0.0756 | 0.1097 | 0.1131 | 0.1304 | 0.1198 | 0.1333 |
| Within | 0.1330 | 0.1390 | 0.1690 | 0.1980 | 0.1920 | 0.1860 | 0.1610 | 0.1450 |
| Between | 0.0154 | 0.0155 | 0.0235 | 0.0329 | 0.0314 | 0.0295 | 0.0306 | 0.0277 |
| Total | 0.1490 | 0.1540 | 0.1920 | 0.2310 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
| Economic Sector | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 |
|---|---|---|---|---|---|---|---|---|
| Group Specific Theil’s Index | ||||||||
| Agriculture, Forestry and Fishing (A) | 0.1344 | 0.1352 | 0.1246 | 0.2148 | 0.2116 | 0.1660 | 0.1475 | 0.1348 |
| Industry including energy | 0.0994 | 0.1040 | 0.1332 | 0.1633 | 0.1370 | 0.1360 | 0.1063 | 0.0891 |
| Construction (F) | 0.1007 | 0.1058 | 0.1906 | 0.2108 | 0.1848 | 0.1518 | 0.1753 | 0.1300 |
| Wholesale and retail trade (G,H, J) | 0.1212 | 0.1245 | 0.1697 | 0.2160 | 0.1904 | 0.1909 | 0.1554 | 0.1346 |
| Financial, real-estate (K,L) | 0.0966 | 0.0973 | 0.1584 | 0.1938 | 0.2102 | 0.1902 | 0.1729 | 0.1407 |
| Public sector and other services(O,P,Q,T) | 0.1021 | 0.1070 | 0.1358 | 0.1530 | 0.1758 | 0.1472 | 0.1261 | 0.1098 |
| Within | 0.1080 | 0.1090 | 0.1390 | 0.1720 | 0.1700 | 0.1600 | 0.1390 | 0.1250 |
| Between | 0.0405 | 0.0448 | 0.0530 | 0.0596 | 0.0537 | 0.0561 | 0.0526 | 0.0480 |
| Total | 0.1480 | 0.1530 | 0.1920 | 0.2320 | 0.2230 | 0.2160 | 0.1920 | 0.1730 |
Agriculture, hunting and forestry; fishing and operation of fish hatcheries and fish farms;
Industry, including energy;
Appendix II. Estimation Results
| (1) | (2) | |
|---|---|---|
| Property Tax Revenues/GDP (t-1) | -0.7709*** (0.2683) | -0.5653** (0.2480) |
| Social Protection Spending/GDP (t-1) | -0.3829*** (0.0789) | -0.4574*** (0.0773) |
| Share of Employment in the Service Sector | -0.2980** (0.1235) | 0.7099 (1.0683) |
| Share of Employment in Industry | -0.3834** (0.1520) | -1.4193*** (0.4847) |
| Share of Urban Population | -0.2312** (0.1060) | -1.3987*** (0.5219) |
| Share of the Population with Tertiary Education | -0.1202*** (0.0380) | -0.2172 (0.1441) |
| Log Real GDP per Capita | -6.6083*** (2.3794) | -7.6205*** (2.1407) |
| Unemployment Rate | 0.1242** (0.0533) | 0.0432 (0.0603) |
| Output gap | 0.0526 (0.0576) | -0.0148 (0.0609) |
| Share of Employment in the Service Sector^2 | -0.0119 (0.0165) | |
| Share of Employment in Industry^2 | 0.0274** (0.0121) | |
| Share of Urban Population ^2 | 0.0074** (0.0033) | |
| Share of the Population with Tertiary Education^2 | 0.0016 (0.0026) | |
| Observations | 285 | 285 |
| Adjusted R-squared | 0.904 | 0.913 |
| Country Fixed Effects | Yes | Yes |
| Year Fixed Effects | Yes | Yes |
| (1) | (2) | |
|---|---|---|
| Property Tax Revenues/GDP (t-1) | -0.7709*** (0.2683) | -0.5653** (0.2480) |
| Social Protection Spending/GDP (t-1) | -0.3829*** (0.0789) | -0.4574*** (0.0773) |
| Share of Employment in the Service Sector | -0.2980** (0.1235) | 0.7099 (1.0683) |
| Share of Employment in Industry | -0.3834** (0.1520) | -1.4193*** (0.4847) |
| Share of Urban Population | -0.2312** (0.1060) | -1.3987*** (0.5219) |
| Share of the Population with Tertiary Education | -0.1202*** (0.0380) | -0.2172 (0.1441) |
| Log Real GDP per Capita | -6.6083*** (2.3794) | -7.6205*** (2.1407) |
| Unemployment Rate | 0.1242** (0.0533) | 0.0432 (0.0603) |
| Output gap | 0.0526 (0.0576) | -0.0148 (0.0609) |
| Share of Employment in the Service Sector^2 | -0.0119 (0.0165) | |
| Share of Employment in Industry^2 | 0.0274** (0.0121) | |
| Share of Urban Population ^2 | 0.0074** (0.0033) | |
| Share of the Population with Tertiary Education^2 | 0.0016 (0.0026) | |
| Observations | 285 | 285 |
| Adjusted R-squared | 0.904 | 0.913 |
| Country Fixed Effects | Yes | Yes |
| Year Fixed Effects | Yes | Yes |
Prepared by Kodjovi Eklou.
The acceleration of inequality around the year 2013 and 2014 could be due to a slow-down in GDP and wage growth respectively. Real GDP growth was 1.3 percent in 2013 and wage growth has slowed down to 5.9 in 2014 (from 7 percent a year earlier).
The Gini index ranges from 0 (perfect equality meaning everyone has the same income) to 100 (maximal inequality, a single person or household has all the income).
The Gini index decreased further by 1 unit in 2018. The median equivalized disposable income is the total income of a household that is available for spending or saving, divided by the number of household members weighted by age.
Only accommodation and food service activities, and, professional, scientific and technical activities are exceptions as they have a relatively high gender pay gap.
A few exceptions exist in sectors such as mining and quarrying, financial and insurance activities with also the highest gender pay gap and the lowest female employment.
The persistent at-risk-of poverty rate is the share of the population living in households in which the equivalized disposable income was below the at-risk-of-poverty threshold (60 percent of the national median equivalized disposable income) for the current year and for at least two out of the three preceding years.
This situation could be explained by the relatively high life expectancy of women relative to men (82.4 compared to 72.9 years for men) as the at-risk of poverty is particularly high for persons living alone in this age group.
The headcount ratio from the World Bank is defined as the percentage of the population living in households with consumption per capita below $1.90 a day (2011 PPP).
ISCED is the International Standard Classification of Education.
We use household level data from Statistics Estonia that yields a GPG slightly different from the headline figures published in Eurostat that aims at cross-country comparison.
Meriküll et al (2019) find similar result in Estonia with the GPG in wealth at the mean being driven by the top of the wealth distribution.
See for instance Ferraro et al. (2018) who show that minimum wages have contributed to lower the GPG in Estonia. See also Majchrowska and Strawiński (2018).
While the GPG is important for all sectors it stands out particularly in sectors such as wholesale and retail trade, in financial and real estate and Public sector and other services (See also, Estonia 2018 Staff Report).
We do not control for the number of children and the marital status as they are potentially endogenous to female labor force participation decision (see Blau and Kahn, 2017).
See for instance Anspal et al (2011) and Anspal (2015).
A non-parametric approach following Ñopo (2008) also yield similar results, suggesting that the large unexplained part of the GPG is not related to the method of estimation.
In this regard, we welcome the project of the Ministry of Finance aiming at collecting new data and better understanding GPG, more specifically the unexplained component.
The share of women low wage earners was 29.4 percent in 2014 compared to 13.9 for men. The EU28 average were respectively 21.2 and 13.5.
See Meriküll and Mõtsmees (2017) who show that longer breaks between jobs can explain an additional part of the GPG when considering wages asked by the employee, albeit small.
We include the lagged policy variables to mitigate potential endogeneity issues. There might be evidently role for other policy variables such as the progressivity of the tax system, we focus here on variables for which we have a large data coverage for most countries. Also, while we have few control variables, our specification is also constrained by data availability and follows the related literature (Jain-Chandra et al (2018) and Chen et al (2018)). In addition, our model allows to replicate the recent trend in inequality in the data while also having a relatively high explanatory power.
The sample size is only determined by data availability.
We estimate also dynamic panel models and models including lagged control variables, but the models shown here performed better.
Property tax revenues and social protection spending as share of GDP were respectively 0.3 and 16 percent in Estonia while the same figures were 1.7 and 23.9 percent at the EU level. Property tax revenues are taken from the Global Revenue Statistics of the OECD and include components such as recurrent taxes on immovable property, on net wealth, estate inheritance and gift taxes. Social protection spending data are from Eurostat.
