Inputs and sources

This project is implemented within 12th Global Development Network (GDN) Regional Research Competition for Central and Eastern Europe. The proposal is a revision of our submission to the 11th annual GDN RRC for CEE. We have revised the project in response to the criticism raised by CERGE-EI referees. We have have acquired micro-datasets for 12 European, 14 non-European transition countries for numerous years, the data has been organised and partially also standardised in terms of variables definition to enable reliable cross-country analyses. In addition, we have advanced the development of software needed to implement the non-parametric techniques intended in this proposal.

There are three main contributions distinguishing this study from previous cross-country analyses:

  1. We will explicitly address the institutional context using coherently estimated measures of gender discrimination, including the distributional effects in the gender wage gap. Bringing to the fore the institutional effects will serve the purpose of providing reliable policy guidance, especially necessary in the context of the transition, where the extent of gender discrimination tends to be larger than in highly developed economies used as benchmark.
  2. We will do so using non-parametric techniques, which are particularly well suited for analysing the countries with a large informal sector. This approach will allow to go beyond the size of the gap at “average” earnings or median of income distribution, as would be necessary in the case of parametric approach. Accounting for the distributional effects is also extremely relevant from a policy perspective.
  3.  We will disentangle the observed long term effects associated with transition from the cyclical variation in the extent of gender discrimination, exploring also the cross country variation in how the contributions from these two sources differ. We will do this by accounting for the cyclical effects separately from long term trends. Cyclical effects will be proxied by output growth and output gap (when possible).

Hence, a focus on institutions will enable us to both integrate a gender perspective into the labour market analysis and to delineate contextual difference between time periods as well as between countries’ different institutional paths of transition.


Research on gender inequalities in labour markets which explicitly takes an institutional perspective, beyond merely adding a dummy variable for a policy measure or legal change, is scarce and has not been extended yet to study of transition economies, to the best of our knowledge. What is largely lacking as well is that the institutional literature on labour markets and on gender in labour markets needs to be connected to the literature on labour markets in transition economies, in order to contextualize the role of institutions in labour market analysis. The literature is strong in the empirical side but is relatively limited when it comes to explanations and contribution to labour market theory from a transition perspective and a gender perspective.

Read more


Upon completion of the project, we will produce a deliverable report which will contain all our results. This includes: (i) theoretical framework providing grounds for deriving policy implications from the empirical analysis; and (ii) empirical analysis of the determinants of gender discrimination in CEECs. Even more importantly, however, we intend to write two publishable articles.

Read more


A careful examination of the empirical studies uncovers two major limitations of the literature findings: (i) excessively simplistic econometrics in most of the literature, as most of the older studies employ Oaxaca-Blinder or Juhn-Murphy-Pierce decomposition, not always tackling of the self-selection effects and (ii) focus on developed, industrialised countries (predominantly the US and the EU). Recently, a new wave of literature has sprung, addressing the first of the two caveats. It has been already forcefully argued that the sample selection into the workforce is crucial for assessing gender wage gaps (Olivetti & Petrongolo, 2008). Moreover, cross-country comparisons are considerably flawed if the self-selection issue is not properly tacked (Albrecht, Vuuren and Vroman, 2009). Finally, also the comparisons across time are not valid without adequate selection correction (Mulligan and Rubinstein, 2008; Machado 2010). Picchio and Mussida (2010) suggest even that the parametric estimators are bound to underestimate the extent of gender discrimination, especially the wage gap. It is the nonparametric approach that we intend to follow throughout the project. However, the major limitation of the Picchio and Mussida (2010) methodology follows from the fact that it rests upon random assignment, thus making the results potentially susceptible to the choice of the seed. Read more