Tag: Transition economies

Regional Economic Development Along the Polish-German Border: 1992-2012

Image of Europe at night from sky via NASA representing regional economic development

In this brief, we summarize the results of a recent analysis focused on the regional economic development in Poland and Germany along the Oder-Neisse border (Freier, Myck and Najsztub 2021a). Economic activity is approximated by satellite night-time light intensity, a comparable proxy available for regions on both sides of the frontier consistently between 1992 and 2012. This period covers the time of economic transformation and the first eight years of Poland’s membership in the European Union. We find that convergence in overall activity across the border has been complete: Polish municipalities that used to be economically much weaker have caught up with those on the German side of the Oder and the Neisse rivers.


The question of the harmonious development of economic activity is at the heart of the European integration project (Art. 2, Treaty of Rome, 1957), and the Maastricht Treaty (1992) made economic convergence between member states an explicit objective. In a forthcoming paper (Freier et al. 2021), we take a new approach to the question of regional European integration.

This brief derives from a recent publication in Applied Economics (Freier et al. 2021a), in which we examine the degree of regional economic convergence along the German-Polish border by taking advantage of satellite night-time illumination data covering the period between 1992 and 2012. The data allows us to study detailed regional patterns of economic development along the river-delimited part of the frontier and further inland.

The seminal work by Henderson et al. (2012) was the first to use night-time light intensity data which covers the entire globe to measure economic activity. Unlike traditional regional economic indicators, light intensity data is independent of administrative border reforms and has been collected in a consistent format over the studied two decades.

Our analysis suggests that, over the analysed period from 1992-2012, there has been essentially full convergence in economic activity between municipalities on both sides of the Polish-German border. While the average value of night-time illumination in our selected group of municipalities in 1992 was 3.7 (on a scale between 0 and 63) in Poland and 7.7 in Germany, the respective values were 9.0 and 9.7 by 2012, and the latter difference is not statistically significant. This convergence suggests a much stronger rate of growth in economic activity on the Polish side of the border. Additionally, we show that within Germany, the distance to the border has much less relevance for economic activity compared to Poland, where it reflects interesting trends. In 1992, Polish towns farther from the border showed significantly higher economic performance. Within Poland, this gap has been greatly reduced over the 20 years we analyse, with regions closer to the border growing much faster compared to those farther away.

Night Lights Along the Polish-German Border

In our dataset, we include municipalities that are located within 100 km from the river delimited part of the PL-DE border. To avoid the sensitivity of the analysis to top censoring of the night-time light intensity data, we removed regional capital cities: Berlin (with surrounding municipalities), Dresden, Gorzów Wielkopolski, and Zielona Góra. This leaves us with 488 municipalities on the German side of the border and 193 municipalities on the Polish side.

The night lights data series, provided by the National Oceanic and Atmospheric Association (NOAA), starts as early as 1992 and continues in a consistent, comparable format to 2012. The data is independent of the administrative structures of local governments, which over time have changed on both sides of the border. This allows us to aggregate the night-time lights information for municipalities using the most recent available administrative borders. This data is essentially the only source of information on economic activity that is consistently available and comparable on both sides of the border over such a long period of time.

The night-time lights data has been applied widely as a proxy of economic development on the country and regional level (Henderson et al., 2012; Bickenbach et al., 2016). Clearly, the intensity of night-time lights does not capture the entire spectrum of economic activity. It has been pointed out that the relationship between night-time light intensity and conventional measures of economic development, such as GDP, is likely to differ depending on a region’s stage of economic development (Hu and Yao, 2019). However, we focus on mostly rural and sparsely populated areas (where there is little risk of top censoring of the data), and compare dynamics between regions that are similar in terms of their stage of economic development, geography, and weather. All these factors support the use of night lights as a proxy for regional development in our application (a number of technical steps are necessary to validate and calibrate the data for use in our analysis, see: Freier et al. 2021).

Economic Convergence Along the PL-DE Border

To understand the overall development of economic activity over the period of interest, we map the changes in the night-time light intensity in Figure 1. The colour scale on the map represents differences in light emissions between 1992 and 2012, with the range going from -40 to 40. A negative value indicates a reduction, and a positive value highlights an increase in light intensity. The negative values have been coloured in a blue-green scale (-40 to 0), while positive values in a red scale (0 to +40).

Figure 1. Night lights: changes in light intensity between 1992 – 2012 along the Polish-German border

Notes: municipalities along the PL-DE river border up to 100 km to the border; municipalities marked in grey treated as outliers and excluded from analysis due to high proportion of top-coded lights pixels in 1992; municipality borders as of 2013 (DE) and 2012 (PL). Source: GeoBasis-DE / BKG 2013, PRG 2012, DMSP OLS v4, OpenStreetMap, own calculations. For details see Freier et al. (2021).

As notable in Figure 1, the red areas are predominant. This exemplifies that between 1992 and 2012, nearly all municipalities in this area witnessed positive economic development as manifested in the intensity of night-time lights. We have a few areas that reflect negative dynamics on the German side of the border. This is mainly due to the regional implications of shutting down activity in agriculture and traditional industries as they were unable to compete with West-German technology and productivity. In Poland, green-blue areas are essentially non-existent, illustrating a universally positive economic development over the studied period. This difference in the pace of changes in light intensity between the German and the Polish side reflects a process of rapid convergence of economic development between municipalities on both sides of the border. These developments are represented in Figure 2 which shows the difference between the night-time light intensity in Germany and Poland by year and provides a test for its statistical significance. The estimation is done on mean log pixel values per municipality and clearly highlights the steep path of convergence. In the early nineties, the difference in mean light intensity was around 100 percent – i.e., the mean difference was as high as the mean level of lights on the Polish side of the border.  Already ten years later it reduced to around 50 percent and disappeared by the end of the analysed period. It is notable that, after an initial steep convergence, the difference in light intensity had a period of stagnation between 2002 and 2008. Interestingly, the full convergence which followed coincides with Poland’s entry into the Schengen agreement in December 2007. As seen in Figure 2, the difference in the average night-time light intensity between Poland and Germany was statistically insignificant and essentially zero since 2009.

Figure 2. Difference in mean night-time lights between Germany and Poland over time

Notes: Difference in log of average pixel values per municipality; year fixed effects included, weighted by municipality area; 95% CI. Source: see Figure 1.

Regional Development and Distance from the Border

Thanks to its high degree of geographical precision, the night-time lights data allows us to study the detailed spatial patterns within each country and, in particular, the relationship between distance to the border and economic activity. This is done by looking across the years 1992 to 2012 and examining three-year windows at each end of the analysed period. Our results, which are reported in Table 1, confirm a strong positive relationship between economic activity and distance to the border on the Polish side of the Oder-Neisse rivers. Overall, Polish regions farther from the border show a greater degree of economic activity, but this relationship has substantially diminished over time. While in Germany, economic activity was higher in regions farther from the border and increasing at the average rate of about 0.3% per km, this rate was about three times higher in Poland, falling from about 1.2% per km in 1992-94 to 0.6% in 2010-2012.

 Table 1. Total night-time lights along the Polish-German border, 1992-2012

Notes: Notes: municipalities along the PL-DE river border up to 100 km to the border; municipality borders as of 2013 (DE) and 2012 (PL); mean municipal total lights calculated using average pixel values per municipality and weighted by municipality area. Standard errors in parentheses, statistical significance: + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001. Source: see Figure 1.

Table 2 reports changes in light intensity between the beginning and the end of a specific period. Here, we find some interesting and perhaps disconcerting results on the relationship between the distance to the border and changes in light intensity. While the distance-to-border coefficient in the Polish case for the full period is negative, suggesting that regions closer to the border were catching up to the more developed regions farther away, the corresponding coefficient for the final three years is positive. This means that, in the years 2010-2012, economic development was faster in municipalities farther away from the border. Although the relationship is not very strong (the change in light intensity grows by about 0.1% per kilometre of distance to the border), it still suggests a reversal in the fortunes of municipalities close to the border on the Polish side. This result points towards the fact that homogeneity of development cannot be taken for granted and that physical distance might continue to play a role in determining the regional rate of growth in the future.

Table 2. Changes in night-time lights along the Polish-German border: 1992-2012

Notes: Notes: municipalities along the PL-DE river border up to 100 km to the border; municipality borders as of 2013 (DE) and 2012 (PL); mean municipal total lights calculated using average pixel values per municipality and weighted by municipality area. Standard errors in parentheses, statistical significance: + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001. Source: see Figure 1.


In this brief, we report results from a forthcoming paper (Freier et al. 2021) in which we evaluate regional development in municipalities on the German and Polish side of the Oder-Neisse border between 1992 and 2012, using night lights data as a proxy for economic activity. We find that driven by rapid growth in Polish municipalities and somewhat sluggish growth in German ones, the light intensity levels across the Oder-Neisse border show no significant differences by the end of our observation period. This is despite significant initial differences just 20 years earlier and the fact that municipalities on the German side also experienced increases in economic activity. In as far as economic development can be proxied by the intensity of night-time illumination, it seems that economic convergence between regions on both sides of the border was complete by 2012.

We also show interesting patterns regarding the relationship between economic activity and distance from the border. For Germany, this relationship is weakly positive and remains stable throughout the analysed period. In Poland, distance is strongly and positively correlated with light emissions at the beginning of the period, hence indicating that municipalities farther from the border show higher average economic activity. By 2012, however, the border regions have closed most of the gap and the distance to the border is a substantially weaker predictor of economic activity, suggesting a much more homogenous pattern of activity.


This brief draws on results reported in Freier et al. (2021a). The authors gratefully acknowledge the support of the Polish National Science Centre (NCN), project number: 2016/21/B/HS4/01574. For the full list of acknowledgements and references see Freier et al. (2021a).


  • Bickenbach F, Bode E, Nunnenkamp P and Söder M (2016) Night Lights and Regional GDP. Review of World Economics 152(2): 425–47.
  • Freier, R., Myck, M., Najsztub, M (2021a) Lights along the frontier: convergence of economic activity in the proximity of the Polish-German border, 1992-2012. Applied Economics, available online: doi: 10.1080/00036846.2021.1898534.
  • Freier, R., Myck, M., Najsztub, M (2021b) Night lights along the PL-DE border 1992-2012. Dataset used in Freier et al. (2021a), Zenodo, DOI: 10.5281/zenodo.4600685.
  • Henderson JV, Storeygard A and Weil DN (2012) Measuring Economic Growth from Outer Space. American Economic Review 102(2): 994–1028.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Addressing the Covid-19 Pandemic: Policy Responses Across Eastern Europe

20200601 Addressing the Covid-19 Pandemic FREE. Network Image 01

The world holds its breath as Covid-19 continues to spread and challenge local health care systems as well as local economies. The focus of international media has mostly been on China and then Western Europe and the US. However, countries around the Baltic Sea, Eastern Europe and the Caucasus differ from the West with respect to their socio-economic development, trade integration, and political systems. The webinar “Addressing the Covid-19 Pandemic in Eastern Europe: Policy Responses Across Eastern Europe” hosted by the the Forum for Research on Eastern Europe and Emerging Economies (FREE) Network on May 28, 2020 aimed to fill this gap in the current discourse and give voice to experts from Latvia, Russia, Georgia, Belarus, Poland, Ukraine as well as Sweden, in order to contextualize their countries’ policy choices and experiences in the crisis. Policy recommendations can only be of preliminary nature at this point of time. Yet, it becomes clear that even though transition countries have fared relatively well during the health crisis, they will not be spared from the ensuing economic crisis and will require policy tools which are adapted to the local context.


Less than six months after the outbreak of the Covid-19 crisis in China, the pandemic has spread across the globe. The epicenter has moved from Asia to Europe and the US, and in late May 2020 some voices are warning that it is now shifting towards Latin America. While the world´s eyes have been on Milan and Paris, little has been said about how the new EU member states and countries to the East of the European Union cope with the pandemic. Some observers have claimed the emergence of a new “iron curtain” in the corona crisis; Eastern Europe, the Baltic States and the Caucasus having been relatively unscathed compared to the West. Persisting differences in trade and travel patterns, demographic and socio-economic differences, as well as differences in trust levels could account for such an observation.

Yet, the most recent statistics suggest that this may be a premature interpretation and the overall picture is much more heterogeneous. Infections in Russia seem to be rising quickly, Georgia by contrast has turned out to be one of the top students.

Figure 1: Total confirmed Covid-19 cases vs. deaths per million.

Source: Our World in Data, 2020. • CC BYa.
Note: Data includes the most recent numbers as of May 25, 2020. Both measures are expressed per million people of the country’s population. The confirmed counts are lower than the totals. The main reason for this is limited testing.

On May 28, 2020, the Forum for Research on Eastern Europe and Emerging Economies (FREE) Network hosted a webinar with its member institutes: BEROC in Belarus, BICEPS in Latvia, CEFIR@NES in Russia, CenEA in Poland, ISET-PI in Georgia, KSE in Ukraine, and SITE in Sweden to discuss how their countries have fared in the corona crisis so far. The webinar provided an opportunity to share experiences and to add some interpretations and insights to the crude statistics, which often become unintelligible in the current overflow of information.

Figure 2: FREE Network Countries.

Source: SITE 2020.

The webinar started with Torbjörn Becker, director of SITE, introducing recent developments in terms of health statistics in the region and the research being done within the framework of the FREE Network.

SITE on Sweden

Jesper Roine, Professor at the Stockholm School of Economics and SITE, then presented the case of Sweden, the country which – with regards to death rates – has surpassed all other FREE Network countries by far. The Swedish case has been very controversially discussed in international media throughout the pandemic. Yet, the common claim that in Sweden everything was “business as usual” is not true, according to Roine. Compared to its direct neighboring countries Finland, Denmark and Norway, Sweden has chosen a relatively lenient approach to Covid-19, but high schools and universities have moved to distance learning since March and working from home is highly encouraged. Mobility reports show that Swedes have reduced their movement a lot, but less so than their Scandinavian neighbors. Roine confirmed that the Swedish health policy has been dominated by the public health agency, Folkhälsomyndigheten. Even though this is the default option in Swedish law, Roine stressed that this does not mean that the government’s hands are tied.

He presented two preliminary conclusions regarding the impact of the Swedish strategy: first, Sweden’s mitigation strategy has worked relatively well; the public health system is seriously strained but not overwhelmed. Yet, Roine said that the “lack of testing [remained] a mystery”, even for advocates of the current mitigation strategy. Second, in Roine’s opinion the attempt to protect the elderly has failed. The virus has spread to numerous nursing homes and excess death rates indicate that mortality has increased sharply for citizens above 65 years of age, much less for other age groups. Geographically, Stockholm has been the center of the epidemic. Other parts of the country have been affected to a much lesser degree.

BICEPS on Latvia

Sergejs Gubins, Research Fellow at the Baltic International Centre for Economic Policy Studies (BICEPS) presented the Latvian experience of the corona crisis. A small country of about 2 million inhabitants, Latvia currently presents the second lowest Covid-19 mortality rate within the EU. Gubins related this to the Latvian government’s quick and determined policy reaction. After the first cases were reported in early March, schools and universities were closed, public gatherings forbidden, international travel halted, and a two-meter social distance rule imposed. Given the success of this strategy, Latvia has started to loosen its restrictions. A “Baltic Schengen area” was announced very recently and travel among the Baltic states is now possible again. The economic support package announced by the government amounts to 45 percent of GDP and includes a large equity investment in the airline airBaltic as well as important investments in infrastructure. According to Gubins, the current policy discussion focuses on the accessibility and size of help funds, widely deemed insufficient. Furthermore, the economic outlook of the country in terms of unemployment rates and GDP growth is bleak despite its success in containing the virus.

CEFIR on Russia

According to Natalia Volchkova, Director of the Centre for Economic and Financial Research (CEFIR) at the New Economic School in Moscow, Russia has pursued a “standard European strategy” in its fight against Covid-19. Two new hospitals exclusively for Covid-19 patients were created in Moscow, the current epicenter of the pandemic, and nearby. Most money spent on health care went to these new facilities, less was transferred as bonuses to medical workers. Russia has emphasized testing: around 10 million tests were performed; close to 400,000 cases of Covid-19 were confirmed. On May 27, free antibody testing was started in Moscow and is to be extended to other parts of the country. State-financed testing will serve to measure the potential degree of immunization of the population. While cases have started to decline in Moscow, other regions of Russia lag behind and are still expected to peak.

Volchkova stressed the role of the Russian shadow economy, which has been severely hit by the crisis. The size of the informal sector makes it difficult for the Kremlin to pass efficient support packages for the economy. Another policy problem lies in the weakness of the social security net, particularly unemployment benefits are hard to obtain. Therefore, most policy measures have focused on companies. Family allowances are the government’s second heavily used tool, which to Volchkova’s mind is an efficient policy choice. She concluded that the current help measures may already amount to 3 percent of GDP.

ISET-PI on Georgia

As of May 28, 2020, Georgia had only reported 12 corona deaths. According to Yaroslava V. Babych, Lead Economist at ISET Policy Institute in Tbilisi, the key explanation for Georgia’s relative success in the corona crisis is that, as in Latvia, testing started very early. She explained that even before Georgia’s neighbor Iran confirmed an outbreak of Covid-19, passengers’ temperatures were taken at the border crossing. The government in Tbilisi then soon imposed harsh quarantine measures, local quarantines in regional hotspots, a shutdown of public transport, an evening curfew and very high fines. Compliance with the measures was very high. Orthodox Easter celebrations were allowed to take place under strict hygiene measures and did not result in a spike in infection rates.

The country, largely reliant on tourism and agriculture, is now focusing on the economic consequences of the crisis. According to Babych, Georgia holds the ambition to become the first European country to open up to international tourism again from July 1, 2020. The government is also determined to avoid another meltdown of the important construction sector, as happened in 2008 – 2009. However, similar to the Russian case, Babych identified two factors which crucially weaken the Georgian economy: the lack of automatic stabilizers in the form of unemployment benefits and the large informal sector. Policymakers have therefore resorted to monthly cash payments to those who stopped paying income tax around March and fixing prices for specific food products. While the effectiveness of these measures still has to be evaluated, the policy discourse in Georgia has moved on to the socio-economic consequences of the crisis.

BEROC on Belarus

Lev Lvovskiy, Senior Research Fellow at the Belarusian Economic Research and Outreach Center (BEROC), provided an overview of the Belarusian policy measures. According to Lvovskiy, Belarus has a high number of nurses and doctors and a relatively efficient “Soviet style of fighting pandemics”. There have been hardly any restrictions to public gatherings and events, both the Orthodox and the Catholic Easter festivities were maintained, as were soccer games and the national Victory parade. Initially, the official policy was to trace and isolate cases, but this did not prove to be very efficient, supposedly due to poor enforcement. Lvovskiy said that testing is rare which is why statistics on the spread of the virus and its effects remain of questionable quality.

While Belarus disposed of a solid health care system, it was not well prepared economically, which explains why the government has not been very proactive in Lvovskiy’s opinion. The Belarusian industrial production decreased by 7 percent in April 2020 compared to the same month the year before; unemployment has started to increase, yet, there are no significant unemployment benefits. Increasing the height of unemployment pay is the key policy issue under discussion in Minsk but in the absence of international loans, the government´s hands are tied. The issue is urgent: the most recent BEROC survey suggests that 46% of individuals living in urban areas have already seen their income decrease. Lvovskiy’s preliminary conclusion is that the Belarusian policy response to the Covid-19 crisis was not as bad as expected by many international observers: the health crisis has mostly been contained. But like in the Georgian case, the socio-economic implications of the crisis are becoming more pressing now.

CenEA on Poland

Michal Myck, Director of the Centre for Economic Analysis (CenEA) in Szczecin, explained that Poland also successfully avoided a spike in infection rates thanks to a quick policy response. Poland was one of the first countries to impose international travel restrictions and very harsh social distancing measures, yet, infection rates remain higher than in other FREE Network countries. Since the second half of April, most measures have been lifted and the spread of the virus seems under control and concentrated in the region of Silesia.

All limitations were implemented without invoking a state of emergency. Myck suggested that the government may have made this choice because the presidential elections would have been automatically postponed otherwise, an outcome the government wanted to avoid. The elections were eventually postponed, but doubts persist with regards to the constitutional validity of the way this decision was taken. Myck stressed the persisting political uncertainty. Economic policy in Poland has focused on protecting jobs and providing liquidity to enterprises. State loans have been primarily directed to SMEs and will be partly written off, conditional on continued activity and employment. In Myck’s opinion, the economic outcome for Poland will depend on whether investments from and exports to Western Europe quickly resume or not.

KSE on Ukraine

Tymofiy Mylovanov, President of the Kyiv School of Economics and former Minister of Economic Development, Trade and Agriculture, stressed that in the first few weeks of the pandemic, Ukraine enforced harsher policy measures than its neighbors. The lock down was almost complete, with only grocery stores and pharmacies allowed to open. Compliance was high during the first few weeks but then started to decline.

The government allocated 3 percent of GDP to a Covid-19 support fund, there has been a lot of deregulation on the labor market, but the central bank’s key interest rate remains at 8 percent. Pressure for a looser monetary policy increases according to Mylovanov, as GDP has fallen by 1.2 percent and unemployment is expected to reach up to 10 percent by the end of the year.

Mylovanov’s thoughts about Ukraine’s economic prospects are mixed: average salaries continue to grow during the crisis which may be explained by the fact that low-skilled employees get laid off first, suggesting a potentially long-lasting change of the composition of the workforce. At the same time, the political situation is volatile with local elections coming up in October 2020 and public pressure mounting. As Poland, Ukraine did not declare a state of emergency. While Mylovanov thinks that the policy response could have been better, he is optimistic that Ukraine was better prepared to Covid-19 than to previous crises and will not have to resort to international loans.

Preliminary Conclusions

It is too early to draw any definite conclusions, but undoubtedly, a lot can be learned from the very diverse experiences of the corona crisis in the region. The former Soviet countries have a different historical and political legacy than Western European countries and accordingly, have found different ways of handling the crisis. Some have been more successful than their Western neighbors. But even those countries which have not faced a large health crisis have been severely hit economically and are likely to suffer economic hardship in the future.

The lack of a strong tradition of unemployment benefits and automatic stabilizers renders countries like Georgia, Belarus and Russia particularly vulnerable to the economic crisis which will inevitably follow the Covid-19 outbreak. In some countries, the corona shock may also accelerate or trigger political changes. In the view of this, the FREE Network will organize a series of follow-up webinars and briefs on more specific corona-related topics, with the aim of contextualizing statistics and providing a wider picture of the socio-economic consequences and policy implications of the crisis.

Please find a full recording of the webinar below. Updates on further events will be posted on the FREE website and on social media channels (Facebook, Twitter).

List of Speakers

  • Jesper Roine, Professor at the Stockholm Institute of Transition Economics (SITE / Sweden)
  • Sergejs Gubins, Research Fellow at the Baltic International Centre for Economic Policy Studies (BICEPS / Latvia)
  • Natalia Volchkova, Director of the Centre for Economic and Financial Research at New Economic School (CEFIR@NES / Russia)
  • Yaroslava V. Babych, Lead Economist at ISET Policy Institute (ISET / Georgia)
  • Tymofiy Mylovanov, President at the Kyiv School of Economics (KSE / Ukraine)
  • Lev Lvovskiy, Senior Research Fellow at the Belarusian Economic Research and Outreach Center (BEROC / Belarus)
  • Michal Myck, Director of the Centre for Economic Analysis (CenEA / Poland)
  • Torbjörn Becker, Director of the Stockholm Institute of Transition Economics (SITE)

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.

Stylized Facts from 25 Years of Growth in Transition

20180226 Stylized facts from 25 years of growth Image 01

This brief summarizes the growth experience of transition countries 25 years after the dissolution of the Soviet Union. We divide our sample into two main groups: the 10 transition countries in Eastern Europe and the Baltics that became EU members in 2004 and 2007 (EU10); and the 12 countries (ex Baltics) that emerge from the Soviet Union (FSU12). The growth experiences of these two groups have been distinctly different. The magnitude of the initial transition decline in output was much more severe in the FSU12 group. Despite growing almost 2 percentage points faster than the average EU10 for the following fifteen years, the FSU12 group is still further behind the EU10 group than they were at the beginning of transition. This illustrates how hard it is for countries to recover from large negative income shocks and thus the importance for countries to avoid such negative events. However, there are no signs of transition countries being stuck in a low or middle-income trap or that natural resource wealth leads to lower growth during this period.

2017 marked the 25-years anniversary after the dissolution of the Soviet Union and the beginning of the transition for the economies in the region. In a recent paper, we explore the growth experience of transition countries over these 25 years (Becker and Olofsgård, 2017). The paper has four main parts: an overview of the transition literature focusing on growth; a part that provides a detailed description of growth in transition; an analytical section that investigate if we can explain growth in transition countries with a standard growth model; and finally an exploration of whether institutional and other variables that have been highlighted in the transition literature (but are excluded from the basic growth model) are correlated with growth in transition countries. This brief summarizes the descriptive part of the paper, while the more analytical sections will be the topic of future briefs.

For most of the paper, we divide our sample into two main groups; the 10 transition countries in Eastern Europe and the Baltics that became EU members in 2004 and 2007 (EU10); and the 12 countries that emerged from the Soviet Union (FSU12). In addition, we include three transition countries that are not part of either group (Croatia, Albania and Macedonia – Other3) and we also divide the FSU12 group into the four countries that export significant amounts of fuel (FSUF) and the eight countries that do not (FSUNF). There are of course remaining differences within these groups, but this aggregate analysis allows us to see certain patterns in the transition process more clearly.

Initial output collapses

The focus in economics is often on how to generate higher growth and not about protecting against significant drops in output. There are some exceptions, including Becker and Mauro (2006) and Cerra and Saxena (2007), where the focus is on output losses and how countries recover after crises. For transition countries, a very important feature of the economic development process is exactly the initial drop in income and the time it has taken countries to recover from the initial phase of transition. Table 1 shows how much income fell in the different country groups and the time it took to get back to the pre-transition income level.

Table 1. Output drops and recoveries

Source: Becker and Olofsgård (2017)

The initial collapse in the FSU12 group was enormous, with income cut in half. The EU10 countries also had massive output losses, but “only” lost a quarter of their income on average. This took over a decade to recover from, while the path back to pre-transition income levels in the average FSU12 country was almost twice as long. There have been many papers written on the economic chaos that was part of the initial transition process, and explanations for this decline has been attributed to, e.g., misleading data, lack of functioning markets, shock therapy and poor economic and legal institutions in general. All of these factors have likely played important roles in the process, but regardless of the explanation, this was a very unfavorable time in terms of economic outcomes for hundreds of millions of people in these countries. Avoiding such costly drops in output should be a top priority for economic policy makers in any country at all times, not just in transition.

From collapse to growth

In most transition countries, the initial phase of decline in transition lasted several years, but eventually the negative growth rates turned positive (Figure 1). Again, we can see that the EU10 group had fewer years of declining incomes with growth resuming in 1993, while for the FSU12 group, growth in transition only started in 1996/7.

Figure 1. Bust-Boom countries

Source: Becker and Olofsgård (2017)

What is less visible in Figure 1 due to the wide scale needed to capture the initial output drops is that the FSU12 groups has shown significantly higher growth than the EU10 group in the last 15 years. Over the more recent period, the average FSU12 country has grown by close to 6 percent, while growth for the EU10 has been around 4 percent per annum (Table 2).

Table 2. Real GDP/cap growth

Source: Becker and Olofsgård (2017)

The faster growth in FSU12 countries is particularly pronounced among the fuel exporters, which were growing by one and a half percentage point faster than the non-fuel exporters between 2000 and 2015. But the table also shows that the very negative growth experience during the first ten years of transition is hard to erase and the EU10 countries have grown faster over the full 25-year period compared to the FSU12 countries. In terms of understanding the growth experience of the different country groups and time periods, it is clear that the sharp increase in international oil prices during the last 15 years of the period generated high growth in the fuel exporting countries in the FSU12 group. Interestingly though, also the non-fuel exporters grew faster than the EU10 in this time period. This is likely linked to spillovers from Russia to the other countries in the region, but could also be related to some recovering after the massive initial declines in output. Such macro and external factors are not always stressed in discussions of growth in transition countries, which more often focus on the pace of reforms or strength of institutions, but seem to be relevant at this aggregate level when comparing the initial and later phases of transition.

Relative incomes in transition countries

Growth or the lack thereof is of importance in determining income levels, which is what we generally think is what influences welfare. The question is then what the growth processes we have analyzed imply for income levels in transition countries, and in particular, how the income levels in these countries compare with other countries.

Figure 2. Income relative to 15 old EU countries

Source: Becker and Olofsgård (2017)

The short story here is that the relative ranking of the different groups is largely unchanged from the start of transition until the end of 2015. The group of countries that eventually joined the EU has the highest income level while the non-fuel exporting FSU countries have the lowest. However, the leading group still only has around 60 percent of the income of the average “old” EU country while the average FSU12 country has half of that or around 30 percent of the income of the old EU countries. This puts the relatively high growth rates of the FSU12 group over the last 15 years in perspective; the road to reach old EU level incomes is long indeed. Also, within the FSU group, it is clear that there is a sharp dividing line between the fuel exporters and the rest. This is in stark contrast to the notion of a “natural resource curse” that is often blamed for poor growth in oil and mineral rich countries.

Growth traps in transition?

One issue that comes up with regards to both low and middle-income countries is if they are stuck at a certain level in the relative income rankings of the world. This is referred to as the low or middle-income trap and the question is if there are signs of transition countries being stuck in such traps.

Figure 3. Moving up the income ladder

Source: Becker and Olofsgård (2017)

Figure 3 shows how transition countries are classified into the World Banks income groups low income (1 in the Figures scale), lower middle income (2), higher middle income (3) and high income (4) groups.

It is clear that the FUS 12 group of countries was sliding down the scale initially, but since the beginning of the 2000’s, all of the transition countries have been climbing up the World Bank income ranking scale without any apparent signs of a low or middle-income trap.

Policy conclusions

There are of course country differences along all the dimensions discussed in this brief but grouping the transition countries together provides some interesting general observations of growth in transition. First of all, it is clear that it is very hard to fully recover from large drops in income. Even with the help of some extra growth following a crisis, it seems to take a long time for most countries to make up for lost ground. This suggests that policy makers in transition as well as other countries need to take measures to hedge the really bad outcomes and not only focus on how to generate an extra one percent of growth.

The other observation is that at the aggregate level, external factors and more mechanical macro boom-bust-boom type of growth factors may dominate what we generally think of as the long-run determinants of growth (such as institutions, education, and micro level reforms to make markets work better) over very long time spans. This does not mean that the focus on the more fundamental growth drivers should diminish, but it is important that reforms in these areas are complemented with a macroeconomic framework that reduces the risks of costly output collapses.

Finally, it is clear that the incomes generated by natural resources can produce growth at the macro level and that there is little evidence that transition countries should be stuck at any particular level in the global income rankings. Go transition countries!


  • Becker, T, and A. Olofsgård (2017), “From abnormal to normal—Two tales of growth from 25 years of transition”, SITE Working paper 43, September.
  • Becker, T., and P. Mauro, (2006). “Output Drops and the Shocks That Matter”. IMF Working Papers 06/172.
  • Cerra, V., and S.C. Saxena (2008). ”Growth Dynamics: The Myth of Economic Recovery”. American Economic Review, 98(1), 439–457.

Economic Gender Equality Issues in Transition Economies

20170129 Economic Gender Equality Issues in Transition Economies Featured Image 12 | 8 size 01

Until a couple of decades ago, gender was almost a non-topic within development economics.[1] But in the 1990s research gradually showed that gender inequality could have substantial impact on macroeconomic outcomes. At the same time it became clear that women and men were hit differently by economic shocks.[2] These insights triggered an unprecedented focus on gender both in research and at the policy level – see Duflo (2012) for a brilliant overview with a developing country focus. The largest collective action process in history targeted at reducing world poverty, the Millennium development goals, focused on gender inequalities in several dimensions when enacted in year 2000.[3]

In the so-called transition economies, economic gender issues came on the agenda in the late 1990s as it became evident that the transition process had affected men and women differently – see e.g. Dijkstra (1997) – and that these growing gender inequalities had important humanitarian and economic costs. For instance, in many transition economies men’s mortality skyrocketed in the 1990s while the gender wage gap rapidly increased.[4] In particular, Pastore and Verashchagina (2011) show that the gender wage gap in Belarus doubled during the decade from 1996 to 2006, partly as a result of women’s increased segregation into low-wage industries.

From a gender perspective, the Soviet model had focused on full employment for both men and women, but without aspiring to dismantle traditional gender roles. Women therefore tended to work full time alongside with men, while remaining primary caretakers of children and household. The differences in gender equality were, however, significant across the Eastern and Central European countries already before the transition process started. It is thus essential to carry out country-specific analysis of gender equality so as to fully account for context-specific institutional, economic and cultural aspects.

This paper aims to provide a short overview of research on economic gender inequality that might be of particular relevance to transition economies. Given the extensive literature on gender inequality on the one hand and transition economies on the other, this report hopes to serve as an introduction and therefore provides extensive references to the literature to ease further reading.

The structure of the paper is as follows. Section 2 presents the efficiency gains associated with gender equality; while the subsequent section examines education from a gender perspective. Section 4 reports on the research on gender differences in the labour market, while the following section exposes how gender stereotypes lead to less competent politicians, missing women, etc., while stereotypes at the same times can be changed quickly. The report ends with an overview of current research and policy relevant questions for transition economies.

Research based on economic gender equality

Had gender equality been a universally accepted goal, no further arguments would have been needed to promote it. In this report, the presumption is that men and women are equally worthy of human rights and civil liberties. Given conflicting policy goals, scarce resources and a lack of women decision-makers, more knowledge about the economic gains associated with gender equality is needed. Furthermore, research on the economic impact of gender inequality might not only provide arguments for promoting gender equality, but can also ease the formulation of actual policies by suggesting mechanisms through which gender equality and economic development are linked.

Economists’ argument for gender equality

From an economic point of view, the main argument to strive for gender equality is that men and women on average have the same cognitive and non-cognitive abilities. Few scientists would today question the statement that the differences within genders with respect to abilities are larger than the differences across genders. In other words, men and women are in terms of innate productive capacities more similar than men among men and women among women are. As long as we define our productive capacity only in terms of brains, most would also agree on the productive equality of men and women. But brawn is often raised as a divisive trait that makes men on average more productive than women. Galor & Weil (1996) even posits that there is no reason for women to enter the formal labour market as long as brawn is more important than brains in production as an explanation as to why women were not on the formal labour market in big numbers until the event of industrialization. Albeit seductive, this line of argument has several fundamental flaws.

First of all, no formal labour market existed before the industrial revolution. In agrarian economies everyone works – men, women and children – but are seldom paid with a monetary salary and have no formal contract regulating pay and work hours. With industrialization men came to constitute the majority of the workforce early on as a consequence of women being the main caretakers, and hence not being able to work far from home once they became mothers (until the children themselves were old enough to work). Moreover, social norms prescribing women to stay at home further impeded mothers to work during certain historical phases. Ultimately, there are few occupations – historically and especially now – that were too brawn-intensive for women.  Rather social norms assigned occupations according to one of the genders and occupation-specific technologies developed accordingly. As a first step in the overview on the mechanisms of economic gender inequality, follows in the next section an exposition on its relation to economic development.

Engendering economic development

Two flagship reports from the World Bank (2001, 2012A) were exclusively dedicated to the role of women in economic development.[5] The point of departure for both reports was the strong correlation between any measure of gender equality and economic development (measured for instance as GDP per capita). While it is clear that gender equality in education and formal labour force participation enhance economic growth – see e.g. Klasen (1999) and Klasen and Lamanna (2009) – it is also clear that sustained economic growth generates a new demand for women’s human capital and indirectly promotes gender equality. From a policy perspective the direction of causality is not unimportant in the short and medium run. In the very long run it is unlikely that a high-income economy can flourish without utilizing the female half of the country’s productive capacity.

Recent research – as Bandiera and Natraj (2013) and Cuberes and Teignier (2014) – indicate that the methodological problems are such that it is challenging to draw policy conclusions on the link between gender equality and economic development based on cross-country studies, and that country-specific analyses are needed to be able to formulate precise policy conclusions.

In the transition economies, gender equality varies greatly along with economic standard. There are clearly efficiency gains to be made by increasing gender equality, but each country needs to perform an analysis of which factors are most crucial to improve. For instance, Hsieh, Hurst, Jones och Klenow (2016) calculates that 15-20 per cent of GDP per capita growth during the period 1960 to 2008 can be attributed to the increased efficiency in the allocation of talent in the American economy. This increase in efficiency is mainly explained by the improved allocation of women’s talents according to Hsieh, Hurst, Jones och Klenow (2016). In a closely related study, Cuberes och Teignier (2016), it is estimated that the OECD’s GDP per capita is 15 per cent lower at present compared to a situation without gender segregation on the labour market and where equally many women and men become entrepreneurs.

In the following, the main gender differences that are central for gender equality and economic efficiency (and thereby growth) are discussed. Out of these, it has been viewed as a first priority to assure that girls and boys both get primary and possibly secondary and tertiary education. Secondly, from an economic standpoint, women’s activity on the formal labour market is essential for sustained economic development. Thirdly, gender norms and their relevance for a wide spectrum of economic (and political) issues are discussed.

Men and women’s education

At the beginning of the 1990s, there were few gender differences in terms of level of education and the labour force was highly educated in most transition economies, although there are considerable regional differences. Gender segregation in terms of field of study was relatively low and gender differences in math performance small. While in most transition countries there has been a feminization of higher education  – in line with the trend in most countries in the world – in other transition economies the increase in economic gender inequalities post 1991 has led to a widening of the gender gaps in both primary and secondary schooling.[6]

While it is debated – see for instance Breierova and Duflo (2004) – that girls’ education is more important than boys’ education for economic growth, it is uncontested that a gender gap in basic education harms future possibilities of a gender equal labour market and economic gender equality in a broad sense.

On a more positive note, the general math-intensity of education in transition countries is still associated with a relatively small gender gap in math performance. In some countries, girls even have a relative advantage in math relative to boys according to Unicef (2013). This becomes of special interest, since recent research has pointed to the importance of math-intensive higher secondary studies for future labour market outcomes – see Buser, Niederle and Oosterbeek (2014). This research also suggests that young women in the Netherlands (and in other European countries) are disadvantaged by their lack of math and science interests. More generally, there is an extensive literature on the existence of stereotype threat of women in mathematics, implying that especially the most talented women shy away from mathematics due to the fear of being found lacking in terms of mathematical performance – see e.g. Spencer, Steele and Quinn (1999).[7]

In most developed countries, math-intensive sciences, engineering and computer science are heavily male-dominated fields of higher education, maybe partly as a consequence of the predominant norm of math being a “male” subject. Thus, there is ample scope to promote women in IT and technology (by more research and explicit policy) in transition economies, where the preconditions for women entering these fields are generally more advantageous. At present Mexico and Greece have the largest share of women graduates in computing (around 40 per cent) according to OECD (2014). Transition countries have the potential to reach similar levels.

Women and men in the labour market

In this section, the overall findings regarding women’s labour force participation (and how it relates to economic development) and the gender wage gap are reviewed. Gender segregation on the labour market is only briefly discussed, but the following section reviews some evidence on vertical segregation. (Gender segregation varies across cultural and technological context and thus requires a more in-depth analysis.)

Development and women’s labour force participation

Women’s labour force participation has been shown to be sensitive to production technology. Research indicates that married women’s labour force participation is U-shaped of over the industrialization process – as first documented in Goldin (1994) and in Mammen and Paxson (2000) in a developing country-context. The line of arguments goes as follows. Before industrialization, most economies had a limited formal labour market. This does not imply that men and women do not work, but rather that they work in self-subsidence farming, or in the informal labour market. As economies develop, the labour force participation of married women tends to decrease for two main reasons. As production moves out of the homes, it becomes more difficult for women to combine work and the care for children. While in agricultural economies, children simply follow the mother when she works, this becomes unfeasible as production occurs in factories and under regulated conditions both because it is practically difficult to find someone to mind the children but also socially unacceptable often for a woman to leave home and children. Moreover, as economies develop there is a strong income effect, which makes it economically possible for married women not to work. Therefore, there is a decline in married women’s labour force participation as an industrialization process occurs. As the economy continues to develop the substitution effect comes into play. By this time, both men and women are more educated and eventually the family’s loss of well-educated married women’s salary becomes notable. Therefore, as the return on education increases with industrialization, the labour force participation of married women increases.

Women’s labour force participation in general has been shown to be sensitive to the introduction of new technology and new medicines. Greenwood, Seshadri and Yorukoglu (2005) indicate that the washing machine and the vacuum cleaner made home production less time-consuming, thereby freeing up time for women to dedicate more time to formal labour market work. Moreover, Goldin and Katz (2002) and Bailey (2006) show how the introduction of the Pill made it possible for women to control and plan their fertility and thereby made labour market work more feasible. Furthermore, Albanesi and Olivetti (2016) suggest that medical progress that led to improved maternal health in the US during the period 1930-1960 positively affected women’s labour force participation. Even though technological breakthroughs might come at a specific point in time, Fogli and Veldkamp (2011) has shown that it takes time for a change in social norms to occur. More precisely, their research shows how women’s labour market entry is closely related to the spread of information from working to non-working women at the local level.

Summing up, while it is clear that there is an overall tendency of women’s labour force participation increasing as a country develops into an industrialized economy with a well-developed service sector, this development is far from automatic or linear. Therefor it is important to identify country-specific conditions, technologies and norms that might enhance or hinder women to enter the labour force.

Gender wage gap

A persistent overall gender wage gap is often mistakenly interpreted as a prime indicator of women being discriminated against in the labour market. While a gender wage gap within a specific occupation in a sector might suggest the existence of discrimination, the overall wage gap is often more of an indication of gender segregation on the labour market or of low female labour force participation.

Even though a large gender wage gap is not synonymous with gender discrimination, it is associated with economic inefficiency. By simulating a theoretical growth model of the American economy, Cavalcanti and Tavares (2016) calculate that GDP per capita in the US would be 17 per cent higher if the US would have the same (relatively low) gender wage that Sweden has.

At an international level the trends in the gender wage gap appears to be related to several differences between men and women on the labour market. One correlation in international cross-country comparisons – that for long puzzled researchers – is that countries with high female employment rates tend to have higher gender wage gaps than countries with a lower female employment rate. The expectation would, if anything, be the reversed: in countries with a high share of women in formal employment, women are more emancipated and thus do not accept a considerable gender wage gap. But Olivetti and Petrongolo (2008) convincingly show that more than half of this cross-country correlation is due to selection. In countries with a high gender employment gap, such as southern Europe and Ireland, there is a selection of high-skilled women into the labour market resulting in a relatively high average wage for women, and thus in a comparatively low gender wage gap. Another potential mechanism explaining why the gender wage gap is smaller in for instance Scandinavia than in the UK and the US would be that the overall wage distribution is more compressed and thereby the gender wage gap is mechanically smaller – see Blau and Kahn (2003).

Even in countries with small gender employment gaps, women on aggregate tend to work fewer hours on the formal labour market. Recent research in Olivetti and Petrongolo (2016) suggests that for industrialized countries it is the growth in the service sector that drives the number of hours women are working. It is further shown that half of the variation in female working hours across industrialized countries is explained by the share of the service sector.

But even as men and women work to the same extent and the same hours, in most countries occupational gender segregation on the labour market is widespread. Horizontal segregation signifies that men and women tend to work within different occupations and even sectors, while the vertical segregation implies that women to a less extent than men tend to be managers. In the next section we will examine some of the costs related to vertical gender segregation.

Gender stereotypes, political quotas and missing women

For a long time, women were underrepresented in politics around the world. This constituted a democratic problem since it implied that half of the constituency in a country was not represented politically. Therefore, quotas for women at different levels in politics have been introduced around the world with considerable success. Pande and Ford (2011) review the evidence on the Indian case, where quotas have been shown not only to increase the representation of women but also to dismantle the negative stereotypes towards female politicians – see Beaman et al (2009). As suggested in Besley et al (2017), the introduction of gender quotas in politics can considerable also improves the quality of politicians. With an exceptionally rich dataset, Besley et al (2017) show that the voluntary quota, implying that every second candidate to the local elections in Sweden in the mid 1990s was a female politician, increased the average competence of politicians. This was achieved by the quota allowing for competent women to be elected and by less competent male politicians not being re-elected.

Even though quotas to increase the share of women on corporate boards are more controversial – despite several European countries having implemented them (see European Commission, 2015)– there is ample evidence that the social norm envisioning the leader/executive to be a man further cements vertical gender segregation – see e.g. Babcock and Laschever (2003) and Reuben et al (2012). Changing leadership norms is indeed a most important measure for increasing economic efficiency at the firm and societal level. Sekkat, Szafarz and Tojerow (2015) investigate which governance characteristics at the firm level are most likely to yield a female CEO in a vast sample of developing countries and find that a female dominant shareholder as well as the firm being foreign-owned are most conducive to women at the corporate top.

Generally, gender norms are known to be persistent and difficult to change. But there are examples where stereotypes change quickly, such as when the introduction of cable television to remote rural villages in South India almost instantly wiped out the traditional son preference with the introduction of more modern gender norms – see Jensen and Oster (2009). Unfortunately, son preferences can also be intensified due to worsening economic conditions, as for instance happened in South Caucuses after the breakup of the USSR. Georgia, Azerbaijan and Armenia all experienced a significant decline in fertility after 1990 and a sharp increase in the de facto son preference, measured as of the average share of boys to girls at birth. Research – see Das Gupta (2015), Dudwick (2015), and Ebenstein (2014) – suggest that this is the outcome of a combination of factors that all concurred to emphasize sons’ larger economic capability in helping their parents economically. In times of economic crises, increased availability of ultrasound technology and abortion together with having fewer children per family, the traditional preference for sons, at least temporarily, peaked to Chinese levels (after the One-Child policy).

Economic gender analysis in transition economics

In the following, the need for sex-disaggregated data and country-specific research are discussed, as well as recent policy work on gender equality.


The prerequisite for well-informed research and policy is data availability. At the international level an impressive effort has been made during the last decades to create sex-disaggregated data, and there are now many gender databases as, for instance, the World Bank’s Gender data portal (http://datatopics.worldbank.org/gender/). While there are surveys such as the Life in Transition Survey (LiTS, http://www.ebrd.com/what-we-do/economic-research-and-data/data/lits.html), Demographic and Health Services (DHS, http://dhsprogram.com) and others being made, there is still a lack of gender-disaggregated data in most transition economies.

The national Statistics Bureau should have the mission of collecting and reporting sex-disaggregated data. Moreover, it is excellent if all interesting gender statistics regularly are published in an overview report to increase accessibility both for the general public but also for policy-makers. In Sweden, Statistics Sweden biannually since 1984 publishes “Women and Men in Sweden – Facts and Figures” (http://www.scb.se/en_/Finding-statistics/Publishing-calendar/Show-detailed-information/?publobjid=27675), a much appreciated publication. Since 1989, the Swedish government publishes, in an Appendix to its annual Autumn Budget, an overview of the “Economic Allocation of Resources between Men and Women”, where both past policy and current statistics are presented. Initially, the intention was to in this way guarantee the production of sex-disaggregated statistics that was necessary for the formulation of gender-sensitive economic policies.

An even more ambitious step would be to create longitudinal micro-datasets where individuals are followed in terms of family, education, work, health and other characteristics so as to be able to fully evaluate the effect of economic policy.

Country-specific research

Gender-specific analysis of labour market conditions and economic outcomes exist for several countries, see e.g. Khitarishvili (2016). However, there is a vast array of dimensions and mechanisms within the field of research about economic gender equality in need of further investigation, particularly incorporating deep knowledge about country-specific economic circumstances.

As discussed in Section 2, the correlation between gender equality and economic development is generally strong but the direction of causality is unclear. There is therefore scope to analyse the precise nature of the gender inequality within each transition economy with respect to the driving forces of economic growth. Are there, for example, any differences in accumulation of human capital at young age between men and women? Are women able to capitalize on their human capital in the labour market? Are there regulations in place impeding women to work in certain sectors and how is the availability of childcare? Is male mortality higher than female mortality – as has been the case in some transition countries in recent years?

In Section 3 about gender inequality in human capital, there are several dimensions that need country-specific contextualization. Higher education has generally undergone a feminization during recent decades in many transition economies, but not in all. To map such trends, it is essential both to analyse whether the economy capitalizes on women’s newly gained human capital and to study why men are becoming less present in higher education. Moreover, by field of study, transition economies have been exceptionally gender equal in math from an international perspective. One could try to exploit such an advantage by channelling women into programming and IT. This could provide transition economies with a considerable comparative advantage by them using their talent pool better than most countries.

Regarding gender inequality in the labour market, there are a number of interesting research projects that must be pursued at the country level as exemplified in Section 5. For instance, in Moldova there is only a tiny gender gap in labour force participation. While this can pass as an indication of a gender equal labour market, in reality it masks a highly (horizontally and vertically) gender segregated labour market, which might also be one explanation of Moldova’s elevated rates of human trafficking – see further World Bank (2014).


Gender inequality has been perceived as one of the most important dimension to both investigate and address by part of the international organizations working with development assistance. Three major policy areas can be identified, beyond the policy initiatives addressing basic health, violence against women and trafficking: a) the labour market; b) norms; and c) political representation. Regarding gender inequalities in the labour market, the trend is now for a deeper analysis attempting to identify the mechanisms at work in the labour market – see for instance Morton et al (2014).

The policy work on social norms is innovative and often uses surveys and interviews to map gender-specific stereotypes and expectations in order to provide a background and explanation for the wide gender differences in economic outcomes. World Bank (2012B) constitutes such an example, where gender norms are contextualized and at the same time put into a cross-country perspective. Here the attempts of involving men by at least mapping their attitudes are well on their way.

Lastly, there is a considerable amount of policy work – hand in hand with the extensive research on the topic – on women’s low degree of political representation. Introducing quotas for women in parliament is not enough to assure women’s political representation as overly evident in the report by the European Commission on the topic (European Commission, 2015). Further policy work is of the essence to support and ease the implementation of quotas and other measures to assure women’s political representation actually improves.

Concluding remarks

This report touches upon main gender issues in transition economies with a focus on economic dimensions, but essential human rights issues as equal access to health care and legislation, and policies against trafficking are, of course, presupposed. Ultimately gender equality is not a women’s issue. But women are the most engaged so far and efforts must continue to involve men and make them active stakeholders.

Even with the best intentions, it remains crucial to formulate actions on the basis of research. Given that economic resources for policy interventions are limited and that we strive for having policy-impact, continuous effort has to be made to let research inform policy on how to best use available resources.


Albanesi, S., and C. Olivetti (2016). “Gender Roles and Medical Progress”. Journal of Political Economy 124(3): 650-695.

Alesina, Alberto, Giuliano Paola and Nathan Nunn (2013). “On the Origins of Gender Roles: Women and the Plough”, Quarterly Journal of Economics 128(2): 469-530.

Babcock, Linda and Sara Laschever (2003). Women Don’t Ask: Negotiation and the Gender Divide. Princeton University Press, Princeton.

Baden, Sally (1993). “The Impact of Recession and Structural Adjustment on Women’s Work in Selected Developing Countries”. Institute of Development Studies, University of Sussex, BRIDGE Report 15.

Bailey, Martha J. (2006). “More Power to the Pill: The Impact of Contraceptive Freedom on Women’s Lifecycle Labor Supply”. Quarterly Journal of Economics 121(1): 289-320. Uppdatering “Erratum and Addendum,”, September 2009.

Beaman, Lori, Chattopadhyay, Raghabendra, Duflo, Esther, Pande, Rohini and Petia Topalova (2009). “Powerful Women: Does Exposure Reduce Prejudice?”. Quarterly Journal of Economics 124: 1497-1540.

Bandiera, Oriana and Ashwini Natraj (2013). “Does Gender Inequality Hinder Development and Economic Growth? Evidence and Policy Implications”. World Bank Research Observer 28(2): 2-21.

Becker, Gary S., Hubbard, William H. J. and Kevin M. Murphy (2010). “The Market for College Graduates and the Worldwide Boom in Higher Education of Women”. American Economic Review 100(2): 229-33.

Besley, Tim, Folke, Olle, Persson, Torsten and Johanna Rickne (2017). “Gender Quotas and the Crisis of the Mediocre Man: Theory and Evidence from Sweden”. American Economic Review 107(8): 2204-42.

Bhattacharya, Jay, Gathmann, Christina and Grant Miller (2013). “The Gorbachev Anti-Alcohol Campaign and Russia’s Mortality Crisis”. American Economic Journal: Applied Economics 5(2): 232-60.

Blau, Francine and Lawrence M. Kahn, 2003. “Understanding International Differences in the Gender Pay Gap”. Journal of Labor Economics 21: 106-144.

Boserup, Ester (1970). Woman’s Role in Economic Development. London: George Allen & Unwin.

Breierova, Lucia and Esther Duflo (2004). “The Impact of Education on Fertility and Child Mortality: Do Fathers Really Matter Less Than Mothers?”. NBER Working paper 10513.

Buser, Thomas, Niederle, Muriel and Hessel Oosterbeek (2014). “Gender, Competitiveness, and Career Choices”. Quarterly Journal of Economics 129(3): 1409-1447.

Cavalcanti, T. och J. Tavares, 2016. “The Output Cost of Gender Discrimination: A Model-Based Macroeconomic Estimate”. Economic Journal 126: 109–134.

Cuberes, David and Marc Teignier (2014). “Gender Inequality and Economic Growth: a critical review”. Journal of International Development 26: 260–276.

Cuberes, D. och M. Teignier, 2016. “Aggregate Effects of Gender Gaps in the Labor Market: A Quantitative Estimate”. Journal of Human Capital 10(1): 1-32.

Das Gupta, Monica (2015). “’Missing Girls’ in the South Caucasus Countries: Trends, Possible Causes, and Policy Options”. Policy Research Working Paper 7236. Washington, D.C.: World Bank Group.

Djikstra, Geske, A. (1997). “Women in Central and Eastern Europe: A Labour Market Transition” in Djikstra, Geske and Janneke Plantega (eds.). Gender and Economics. London: Routledge.

Dudwick, Nora (2015). “Missing Women in the South Caucasus: Local Perceptions and Proposed Solutions”. Washington, DC: World Bank Group.

Duflo, Ester (2012). “Women Empowerment and Economic Development”. Journal of Economic Perspectives 50(4): 1051-1079.

Ebenstein, Avraham (2014). “Patrilocality and Missing Women”. Mimeo, Hebrew University of Jerusalem.

Elborgh-Woytek, Katrin et al. (2013). “Women, Work, and the Economy: Macroeconomic Gains from Gender Equity”. IMF Staff Discussion Notes No. 13/10.

European Commission (2015). “Gender Balance on Corporate Boards Europe is Cracking the Glass Ceiling”. http://ec.europa.eu/justice/gender-equality/files/womenonboards/factsheet_women_on_boards_web_2015-10_en.pdf

European Commission (2015). “Women in Power and Decision-Making in the Eastern Partnership Countries”. http://eige.europa.eu/sites/default/files/documents/gender_equality_and_decision_making_in_eap_countries_2015_-_report_and_annex_one_file.pdf

Fogli, Alessandra and Laura Veldkamp (2011). “Nature or Nurture? Learning and the Geography of Female Labor Force Participation”. Econometrica 79: 1103–1138.

Galor, Oded and David N. Weil (1996). “The Gender Gap, Fertility, and Growth“. American Economic Review 86(3): 374-38.

Goldin, C., 1994. “The U-shaped Female Labor Force Function in Economic Devlopment and Economic History”. NBER Working Paper 4707.

Goldin, Claudia and Larry F. Katz (2002). ”The Power of the Pill: Oral Contraceptives and Women’s Career and Marriage Decisions”. Journal of Political Economy 110(4): 730-770.

Greenwood, J. Seshadri, A. and M. Yorukoglu (2005). ”Engines of Liberation”. Review od Economic Studies 72(1): 109-133.

Hsieh, C.-T., Jones, C. I., Hurst, E. och P. J. Klenow (2016). ”The Allocation of Talent and U.S. Economic Growth”. Mimeo. Older version in NBER Working paper 18693.

Jensen, Robert and Emily Oster (2009). “The Power of TV: Cable Television and Women’s Status in India”. Quarterly Journal of Economics 124: 1057-94.

Kabeer, Naila (2003). Gender Mainstreaming in Poverty Eradication and the Millennium Development Goals – A handbook for policy-makers and other stakeholders. International Development Research Centre, Ottawa.

Kazandjian, Romina, Kolovich, Lisa, Kochhar, Kalpana and Monique Newiak (2016). “Gender Equality and Economic Diversification”. IMF Working Paper 16/140.

Khitarishvili, Tamar (2016). “Gender Dimensions of Inequality in the Countries of Central Asia, South Caucasus and Western CIS”. Levy Economics Institute Working Paper 858.

Klasen, Stephan (1999). “Does Gender Inequality Reduce Growth and Development? Evidence from Cross-Country Regressions”. Background paper for Engendering Development, World Bank, Washington DC.

Klasen, Stephan and Francesca Lamanna (2009). “The Impact of Gender Inequality in Education and Employment on Growth: New Evidence for a Panel of Countries”. Feminist Economics 15(3): 91-132.

Mammen, K. and C. Paxson (2000). “Women’s Work and Economic Development”. Journal of Economic Perspectives 14: 141-164.

Morton, Matthew, Klugman, Jeni, Hanmer, Lucia and Dorothe Singer (2014). Gender at Work : A Companion to the World Development Report on Jobs. Washington, DC: World Bank Group.

OECD (2015). Education at a Glance 2015: OECD Indicators. OECD Publishing, Paris.

Olivetti, Claudia and Barbara Petrongolo (2008). “Unequal Pay or Unequal Employment? A Cross-Country Analysis of Gender Gaps”. Journal of Labor Economics 26: 621-654.

Olivetti, C. and B. Petrongolo (2016). “The Evolution of Gender Gaps in Industrialized Countries”. forthcoming Annual Review of Economics.

Pande, Rohini and Deanna Ford (2011). “Gender Quotas and Female Leadership: A Review”. Background Paper for the World Development Report on Gender

Pastore, Francesco and Alina Verashchagina (2011). “When Does Transition Increase the Gender Wage Gap?”. Economics of Transition 19(2): 333-369.

Reuben, Ernesto, Rey-Biel, Pedro, Sapienza, Paola and Luigi Zingales (2012). “The Emergence of Male Leadership in Competitive Environments”. Journal of Economic Behavior and Organization 83(1): 111–117.

Sekkat, Khalid, Szafarz, Ariane and Ilan Tojerow (2015). “Women at the Top in Developing Countries: Evidence from Firm-Level Data”. IZA Discussion paper 9537.

Spencer, Steven J., Steele, Claude M. and Diane M. Quinn (1999). “Stereotype Threat and Women’s Math Performance“. Journal of Experimental Social Psychology 35: 4–28.

UNICEF (2013). Equity in Learning? A Comparative Analysis of the PISA 2009 Results in Central and Eastern Europe and The Commonwealth of Independent States. Geneva: United Nations Children’s Fund.

World Bank (2001). Engendering Development – Through Gender Equality in Rights, Resources and Voice. Washington, DC.

World Bank (2012A). World Development Report 2012: Gender Equality and Development. Washington, DC.

World Bank (2012B). On Norms and Agency Conversations about Gender Equality with Women and Men in 20 Countries. Washington, DC.

World Bank (2014). “Moldova: Gender Disparities in Endowments and Access to Economic Opportunities”. Report 76077-MD, Washington, DC.

[1] The exception was the seminal Boserup (1970).

[2] See for instance Baden (1993).

[3] See Kabeer (2003) for an overview of research in development economics and policy experience relevant to the achievement of the Millennium Development Goals from the perspective of gender equality.

[4] Research – see Bhattacharya, Gathmann and Miller (2013) – however suggests that it might have been changing alcohol policy rather than transition per se that caused the sudden increase in mortality.

[5] The IMF has published a number of reports recently, such as Elborgh-Woytek et al (2013) and Kazandjian, Kolovich, Kochhar and Newiak (2016).

[6] See for instance, Becker et al (2010) and OECD (2015).

[7] Stereotype threat is defined as when an individual perceives to be ”at risk of confirming, as a self-characteristic, a negative stereotype about one’s social group” in the seminal paper by Steele and Aronson (1995).

The Economic Complexity of Transition Economies

FREE Network Policy Brief Featured Image 01

‘Diversification’ is a constant concern of policy-makers in resource rich economies, but measurement of diversification can be hard. The recently formulated Economic Complexity Index (ECI) is a promising predictor of economic development characterizing the overall complexity and diversity of the economy as a system. The ECI is based on the diversity and ubiquity of a country’s exports. This brief uses ECI to discuss the economic diversity of transition economies in the post-Soviet decades, and the relationship between economic diversification and per capita income.

The search for and construction of appropriate predictors of economic development are among the main goals of economists and policy-makers. Education, infrastructure, rule of law, and quality of governance are all among the commonly used indicators based on inputs. The recently formulated Economic Complexity Index (Hidalgo and Hausmann, 2009) is a new promising predictor of economic development characterizing the overall complexity and diversity of the economy as a system.

Indeed, the importance of production and trade diversification for economic development has been highlighted by the economic literature. Numerous studies have found a positive relationship between diversified and complex export structure, income per capita and growth (Cadot et al., 2011; Hesse, 2006; Hausmann et al., 2007). In line with this, Hausmann et al. (2014) demonstrate the predictive properties of the ECI for economic development and GDP per capita, which implies that the ECI can serve as a useful complement to the input-based measures for policy analysis by reasoning from current outputs to future outputs.

This brief uses the ECI to discuss the evolution of economic diversification, its relationship to per capita income in transition economies in the post-Soviet decades, and its policy implications.

How is economic complexity measured?

The economic complexity index (ECI) is a novel measure that reflects the diversity and ubiquity of a country’s exports. The index considers the number of products a country exports with revealed comparative advantage and how many other countries in the world export such goods. If a country exports a high number of goods and few other countries export these products, then its economy is diversified (a wide range of exports products) and sophisticated (only a few other countries are able to export these goods). Thus, the measure tries to capture not a specific aspect of the economy, but rather its overall sophistication.

For example, Japan, Switzerland, Germany and Sweden have been in a varying order at the top of the ranking of the Economic Complexity Index from 2008 until 2013. This means that these countries export a large number of highly sophisticated products.

In contrast, Tajikistan is among the countries at the bottom of the world ranking by the ECI with raw aluminum, raw cotton and ores making up 85% of all Tajikistan’s exports in 2013. However, not only are Tajikistan’s exports concentrated among very few narrow products, these products are also ubiquitous and the ability to export them does not require knowledge and skills that can be used in the production and exports of many other products.

As the index for each country is constructed relative to other countries’ exports, it is comparable over time.

What can we learn from the economic complexity of transition economies?

The economic complexity index can serve as a useful indicator for understanding transition economies in the post-Soviet period. A strong relationship between GDP per capita and economic complexity is found in the sample of transition economies in Figure 1. This figure presents the relationship for the last year for which data is available for the sample of 13 post-Soviet states and Poland. As can be seen in Figure 1, the economic complexity is positively related to income per capita. This is especially true for Poland, Estonia, Lithuania, Latvia and Russia, who all have higher than average economic complexity and high levels of per capita income. While Belarus and Ukraine also have diverse and complex economies, they have somewhat lower income per capita than the first group.

Figure 1. Economic Complexity and GDP per capita

Figure1Source: Data on GDP per capita is from the World Bank, and the data on the Economic Complexity Index is from the Observatory of Economic Complexity.

Natural resource-rich, or rather, oil-rich countries are the exception from the abovementioned correlation. Most transition countries with below than average economic complexity are characterized by low income per capita levels, except for Kazakhstan and Azerbaijan, which are oil-rich countries. Still, the overall picture is straightforward: countries with a complex export structure have a higher level of income.

One of the advantages of a systemic measure like export complexity is its straightforward policy application. The overall diversity and sophistication of the economy can thus be a complementary measure for the assessment of economic progress and development to GDP and GDP per capita, which are more susceptible to the volatile factors such as commodity prices.

Figure 2 shows the development of economic complexity for 14 post-Soviet countries and Poland between 1994 and 2013 (due to data availability issues, only one year is available for Armenia).

First, we see that the economic complexity has diverged over time, although there is some similarity in the rankings among countries over time. The initial closeness is likely related to the planned nature of the Soviet economy that aimed to distribute production among Soviet Republics. In the post-Soviet context, however, the more complex economies (Estonia, Belarus, Lithuania, Ukraine, Latvia, Russia) kept or increased their sophistication and diversity of exports. Poland is the leading economy in terms of complexity, both in the beginning and towards the end of the sample period. Belarus, the second most complex economy in 2013 and the most complex economy in several years prior, shows an increasing trend in its sophistication of exports. Although its GDP per capita is noticeably lower than what would be expected from such a sophisticated economy, the complex production structure may explain its ability to withstand a permanent high inflation and external macroeconomic shocks. Some others, e.g., Tajikistan and Azerbaijan, saw a decreasing trend in economic complexity; Georgia and Kazakhstan, notably, lost in economic complexity but also in their ranking among their peers.

Figure 2. Economic Complexity of Transition Economies

Figure2Source: Data on GDP per capita is from the World Bank, and the data on the Economic Complexity Index is from the Observatory of Economic Complexity.


This brief revisited the economic complexity of transition economies and its evolution since the 1990s. The post-Soviet and other transition countries have had diverging economic development paths: Some have managed to build complex production economies, while others’ comparative advantage remains in raw materials. These differences are also reflected in their income levels.

Across the world, economic diversification is associated with higher per-capita income. As the brief showed, this relationship also holds for the post-Soviet countries; policy-makers should take economic diversification seriously. Increasing economic complexity may well pave the path to higher income levels.


  • Cadot, O., Carrère, C., & Strauss-Kahn, V. (2011). Export diversification: What’s behind the hump?. Review of Economics and Statistics, 93(2), 590-605.
  • Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., Simoes, A., & Yildirim, M. A. (2014). The atlas of economic complexity: Mapping paths to prosperity. Mit Press.
  • Hausmann, R., Hwang, J., & Rodrik, D. (2007). What you export matters. Journal of economic growth, 12(1), 1-25.
  • Hesse, H. (2006). Export diversification and economic growth. World Bank, Washington, DC.
  • Hidalgo, C. A., & Hausmann, R. (2009). The building blocks of economic complexity. proceedings of the national academy of sciences, 106(26), 10570-10575.

Macroeconomic Performance and Preferences for Democracy

20191231 Default Image 01

This policy brief summarizes the results of our research on factors influencing preferences for democracy in transition countries. The aim of this work was to detect which macroeconomic and individual factors impact the choice of supporting democracy. The results showed that the performance of the country, described by level of GDP, unemployment, level of corruption and economic growth, has a serious impact on an individual’s perception of democracy. At the same time, individual factors like education and age also influence people’s choice of support of democratic authorities.

Individual perception of democracy is a question that attracts the attention of policymakers.  The macroeconomic instability that has been observed worldwide lately is likely to impact individual attitude toward democratic values and political institutions. The recent economic crisis brought a deterioration of the economic situation around the world and provided new challenges to cope with. It is likely that macroeconomic indicators have an impact on how a person perceives democracy. Literature studying similar questions has shown that GDP growth, unemployment and inflation all affect personal attitude to democratic institutions (Clarke et. al., 1994; Barro, 1999; Papaioannou and Siourounis, 2008). As for individual characteristics, the level of education is revealed by the literature as a very important factor in the context of the individual’s propensity of democracy approval.

The literature on the determinants of political support and attitudes to democracy was mostly focusing on exploring stable world economies with long-formed and steady-functioning democracies. We tried to look at a similar question in the context of transition economies, where democratic institutions are still under development.

We intend to estimate individuals’ propensity to favor democratic values. The specification of our econometric model was based on the literature addressing the same topic. The estimation procedure used probit econometric techniques, which allows for the calculation of the propensities of interest while taking into account the influence of both macroeconomic factors and individual characteristics. The paper used two sources of data: macroeconomic information was collected from the World Development Indicators of the World Bank, and individual-level cross-sectional data was obtained from Life in Transition Survey (LITS) 2010, which initially covered 38864 individuals from 35 countries. However, as the paper focuses on countries in transition, the final set only included individuals from 30 countries, most from Eastern Europe, Baltics and CIS, and excluded representatives of Western Europe. This data allowed for substantial data variation in the context of economic development vs. perception of democratic values (Graph 1).

Figure 1. Support of Democracy and GDP Per Capita
Source: WDI and LITS 2010

Inclusion of different macroeconomic variables together with individual factors allowed for an evaluation of their importance and level of impact on the perception of democratic values (Table 1). The results show that GDP per capita has a positive and significant effect on individuals’ perception of democratic values, which is in line with the literature claiming that standard of living in countries with not so high level of GDP is positively correlated with satisfaction with their life and the political system (Easterlin, 1995; Clark et al., 2008; Stevenson and Wolfers, 2008). Inflation rates are not significant and do not influence individuals’ attitude to democracy. On the other hand, economic growth is strictly positive and significant, and an increase of the economic growth rate raises propensity of democratic support by around 1.6 percentage points. The possible explanation here is that the growth rate of GDP works as a proxy of expectations for improvements of the standard of living in the future.

Table 1. Influence of Macroeconomic and Individual Factors on Perception of Democracy

Unemployment works as an indicator of a country‘s economic performance and has an expected negative sign in terms of individuals’ satisfaction with life and political institutions, which is also in line with the results in the literature (Di Tella et al., 2001; Wagner and Schneider, 2006). Impact of unemployment was tested using a cross product of unemployment and the Freedom House Index (this latter indicator shows the level of political and civil rights from 1 (most free) to 7 (least free)). The sign on this cross product is positive, which captures their mutual positive impact on the support for democracy. Thus, the higher the unemployment in a country with a low level of democratization is, the larger the probability of democratic support by individuals in these countries is.  The indicator for the level of corruption in a country was also taken into account, via the Corruption Perception Index. This index ranks countries on a scale from 0 (highly corrupt) to 10 (effectively, corruption-free). The results show that the less corrupt a country is, the higher the propensity that an individual in that country will support democracy is. In fact, one additional point in the index increases the propensity of support by almost 4 percentage points. Military expenditures negatively affect the support of democratic values, and so does the existence of oil in the country. Here, military expenditures may be seen as a proxy for a less democratic regime, so that the leaders there have higher incentives to rule using suppressive measures with a support of military force in the country (Mulligan, Gil and Sala-i-Martin, 2004).

As for the individual factors, both secondary and higher education appear to be very important factors with a positive impact on the satisfaction with democracy. This finding follows the literature (Barro, 1999; Przeworski et al., 2000; Glaeser et al., 2004). In our results, people with secondary or higher education degree showed 10 and 18 percentage points higher propensity of support, respectively. Age also seem to matter: positive perception of democracy is specific to those aged 18-54, compared to the older generation, which goes in line with the explanation that senior citizens are more conservative than younger citizens. We also observe a negative significant coefficient on female gender, which may, perhaps, be related to women being more conservative than men.

Subjective relative income measure (answer to the question “to which income quintile do you think you belong to?”) has a positive impact on the support for democracy. Surprisingly, individuals from middle-income group have a more positive attitude than those who regard themselves as rich. Employment status is positively correlated with the support for democracy. Moreover, self-employment and employment in the public sector have a larger effect on the propensity of positive attitude to democratic values than employment in the private sector.

Divorced and widowed people expressed less support for democracy than single individuals, which might signal some dissatisfaction that impacts on personal attitude. Urban residency is positively correlated with the support of democracy. The same relationship is present for the risk tolerance of an individual. Finally, inclusion of a subjective measure of life satisfaction brought some changes to the general picture. It appeared that those who are satisfied with life strongly support the democratic values and such mentality raises the propensity of support by 7 percentage points. Moreover, inclusion of this variable makes the effect of being rich insignificant.

To sum up, the results showed that economic performance of the country described by various macroeconomic indicators has a serious impact on individual’s perception of democracy and, most probably, of other forms of government. At the same time individual factors also influence people’s satisfaction with the authorities. Thus, individual support of a political system is based on the results of performance of both the individual and the country.


  • Barro R. 1999. “Determinants of Democracy.”Journal of Political Economy 107, #S6.
  • Clark A. and Oswald A.J. 1994.“Unhappiness and Unemployment.”EconomicJournal104.
  • Clark A., FrijtersP. and Shields M. 2008. “Relative Income,Happiness and Utility: An Explanation for the Easterlin Paradox and Other Puzzles.” Journal of Economic Literature46,# 1.
  • DiTellaR., MacCulloch R.J., Oswald A.J. 2001. “Preferences over inflation and unemployment: Evidence from surveys of happiness.”American Economic Review91.
  • Easterlin R. 1995. “Will Raising the Incomes of All Increase the Happiness of All?”Journal of Economic Behavior and Organization27, # 1.
  • Glaeser E., La PortaR., Lopez-de-SilanesF. and ShleiferA. 2004.“Do Institutions Cause Growth?” Journal of Economic Growth.9, #3.
  • Mulligan C.B., Gil R. and Sala-i-Martin X. 2004. “Do Democracies HaveDifferent Public Policies than Nondemocracies?” Journal of Economic Perspectives18, #1.
  • Papaioannou, E. and Siourounis G. 2008.“Economic and Social Factors Drivingthe Third Wave of Democratization.” Journal of Comparative Economics36, #3.
  • Prezworski A., Alvarez M., Cheibub J. and LimongiF. 1996. “WhatMakes Democracy Endure?” Journal of Democracy 7, #1.
  • Stevenson, B. and Wolfers, J. 2008. “Economic Growth and SubjectiveWell-Being: Reassessing the Easterlin Paradox.” Brookings Papers on Economic Activity  1.
  • Wagner A.F. andSchneider F. 2006. “Satisfaction with Democracy and the Environment in Western Europe: A Panel Analysis.” IZA Discussion Papers 1929, Institute for the Study of Labor (IZA).

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.