ISET Economist Blog

Discrimination in Georgia
Monday, 26 May, 2014

On May 2, 2014, the Georgian parliament unanimously passed the law on the elimination of any form of discrimination. The stated objective of the law is to ensure that any physical or legal entity equally benefits from all rights defined by Georgian legislation, irrespective of race, skin color, language, sex, citizenship, place of origin, birth or residence, wealth or class status, religion or belief, national, ethnic or social belonging, profession, marital or health status, disabilities, sexual orientation, gender identity, political or other considerations, etc. Acute discussions between parliament and church representatives, especially about the phrase on sexual orientation and gender identity, preceded the adoption of the law. Leaving aside moral and religious considerations, it may be interesting to discuss the economic aspect of discrimination, in particular labor market discrimination.

The Nobel Prize-winning economist Kenneth Arrow refers to discrimination as a situation in which personal characteristics of the worker that are unrelated to productivity are also valued on the market (Arrow. K. (1973), “The Theory of Discrimination”, Princeton N.J.: Princeton University Press). As a result, people from disadvantaged groups are either refused jobs or are accepted but have lower salaries compared to the majority.


Economic theory distinguishes between two kinds of discrimination: a taste for discrimination and statistical discrimination.

Taste for discrimination results from genuine likes and dislikes for certain groups of people. Racial, ethnic, and gender discrimination belong to this category. Statistical discrimination, on the other hand, is rooted in informational asymmetries in the labor market. An employer typically has very limited information about a prospective worker, and if there is a perceived or real correlation between group membership and performance in the job, an employer may (partly) derive their expectation about an applicant from their group membership. For example, an employer may have made the experience that on average members of group X are less competent for doing a certain job. Lacking information about the competency of a specific applicant who belongs to group X, the employer may expect that applicant to perform as well as the average of group X and thus reject employment. This can happen even if the employer is solely deciding based on legitimate criteria (performance, competency, etc.) without having a general dislike for group X.

Taste for discrimination was explored in-depth by Gary Becker, another Nobel Prize winner (he passed away recently). Becker showed that competition between companies can help decrease discriminating behavior due to taste for discrimination. This is because discriminating firms have to pay the price for their behavior. Namely, if they do not hire productive workers just because of “taste”, other firms can hire them and become more productive and more profitable than discriminating firms. In the long run, only non-discriminatory firms are able to make enough profits to survive (Becker, G. (1957), “The Economics of Discrimination”, Chicago: University of Chicago Press).

However, as Arrow (1973) (“The Theory of Discrimination”, Princeton N.J.: Princeton University Press) pointed out later, in the absence of perfectly competitive environment discrimination can persist and take different forms, depending on the type of market failure in place. For example, if the discriminator is not an employer but another group of workers, segregation may occur. Firms will have an incentive to employ either discriminating or discriminated workers only to avoid the two groups working together. If a firm would employ members of both groups, it would have to compensate the discriminators who despise the members of the other group for the disutility they get from working together. Likewise, those who are discriminated against might demand compensation for working together with people who regard them with hostility and suspicion.

Statistical discrimination may reinforce itself and become the source of persistent inequality. The vicious cycle works the following way: disadvantaged group members internalize the fact that employers are not going to compensate for the human capital they accumulate (as only their group’s average will be taken into account), reducing their efforts to acquire proper education, thereby confirming the prejudice and becoming even more marginalized. So, by endogenizing the prejudice the members of the discriminated group end up increasing the likelihood of being trapped in a negative equilibrium.  In some extreme cases, the discriminated start feeling and believing themselves to be inferior to the advantaged groups.


There are not many studies devoted to discrimination in Georgia, so one cannot say a lot about the nature and the extent of the problem. One of the first (and maybe even the very first) scientific approaches to labor market discrimination in Georgia was taken by Sophiko Skhirtladze, an ISET graduate. In her Master Thesis (2008) she explored the wage gap between Georgians and important ethnic minorities residing in Georgia. She found that there was no “taste for discrimination” against ethnic minorities, as the existing wage gap could be explained by differences in human capital, location, occupations, and industries.

In 2010 (in a study sponsored by the Economics Education and Research Consortium) Sophiko explored gender and ethnic discrimination at the selection stage for relatively high-paid jobs. While no gender discrimination was revealed, ethnic minorities were found to be disadvantaged in this case. While per 20 applications on average 7 Georgian males were invited for an interview, ethnic minority males received just 1 invitation.

These results are quite telling and need to be tracked over time in order to observe the recent trends and developments. Without a proper understanding of roots and patterns, it is impossible to tailor appropriate policies dealing with discrimination. One reason for the discrimination found by Sophiko could be language issues. Often, ethnic minorities residing in Samtskhe-Javakheti (Armenians) and Kvemo Kartli (Azerbaijanis) are not good at speaking the Georgian language. This is true even for many of those who study at Georgian-language universities. If language is the cause of the identified pattern one could improve the situation by teaching the Georgian language more actively at an early age to children from the ethnic minority groups.

Statistical discrimination, without second-round (no education, no skills…) effects, could be addressed by fostering inter-group contact and communication. Again here, language can be key.

Georgians do not seem to have a very strong “taste for discrimination”. According to the Caucasus Barometer, while 99% of Georgians approve of doing business with their fellow Georgians, about 80% approve of doing business with local Armenians or Azerbaijanis. The difference is non-negligible, but not huge. As recently discussed in an article by Maka Chitanava (to be found on the ISET Economist Blog), Georgians are generally quite open to business relations with almost any ethnic group. This indicates that discrimination in the labor market is more of a “statistical” nature, possibly related to language issues.

Ethnic minorities are not the only type of minorities that could be discriminated against in the labor market. Unlike ethnic minorities, the issue of sexual minorities may be more of the kind entailed by a taste for discrimination. The events that happened in May 2013 provide obvious evidence of prevalent social prejudices towards them. In this case, easy solutions probably do not exist. Conservative Georgians will need a lot of time until (if ever) they accept these people as they are, without negative emotions. So, it would be a difficult task for the government to ensure that such prejudices do not lead to any form of discrimination in the labor market.

To an ordinary person, the Georgian law might look more like a declaration of principles rather than a tool designated to fight discrimination. Even before the adoption of this law, the Georgian Constitution was declaring that all citizens have equal rights and all those who felt oppressed could defend their rights at the court. Should one be disappointed then about the new law? Not necessarily. Reiterating non-discriminatory principles and spelling them out is always positive. Moreover, the fact that the law does not impose heavy regulations might actually be good. Affirmative action (i.e. positive discrimination) is sometimes used in developed countries to fight discrimination. Yet while such policies certainly constrain the freedom of employers, their success in terms of promoting equality is not guaranteed. Actually, in some cases, they risk aggravating the “taste for discrimination” rooted in some parts of society.

The success of the Georgian law will crucially depend on the determination of the Georgian government to enforce it and to ensure that no discriminatory behavior is tolerated. Time will tell whether that is going to happen.

The views and analysis in this article belong solely to the author(s) and do not necessarily reflect the views of the international School of Economics at TSU (ISET) or ISET Policty Institute.