Employment protection legislation and labor markets in transition: assessing the effects of the labor code in Armenia.

• Author: Hartwell, Christopher A.
• Position: Report

1. Introduction

   The question of the effect of employment protection legislation (EPL) on labor markets is a hotly debated one in both policy and economic circles. Broadly defined, EPL is designed, in the words of the World Bank, to "create conditions that are more conducive to job creation, protecting employment, and improving working conditions" (World Bank 2007a). In practice, this usually means a package of legislation that governs hiring and firing of workers, treatment of workers on the job (including holidays, sick leave, and maternity issues), and other related issues (such as workplace safety). (2)

   The creation of EPL often rests upon the assumption by economists and policymakers that legislation can be expertly designed in such a way as to improve the welfare of workers while not impacting the labor market in a negative manner (these assumptions mainly are based on a Coasean view of the world--that is, in a world of low transaction costs, negotiations can lead to redistribution without loss of efficiency). (3) In justifying the need for EPL, there usually is some reference to "market failure," with legislation needed to protect workers from the depredations of employers in a situation of informational asymmetry. However, there are strong economic arguments against the adoption of EPL. Theoretically, interference in the labor market could increase costs to hiring and thus create or maintain unemployment, advantaging those who already have a job at the expense of those who are still looking. Additionally, EPL can focus on only a symptom of a weak economy (turnover) rather than underlying structural flaws that lead to these outcomes, and thus create even more distortions.

   A growing literature, typified by the OECD's major initiatives over the past 4 years (see, for example, Bassanini and Duval 2006), has attempted to quantify the effect of EPL on labor market performance, but has thus far yielded mixed results (more of this in Section II below). Much of this work has been done at the cross-country level, but has been restricted to developed economies, ostensibly to research differences in labor market performance connected with continuing stagnation in Europe, but also because labor market regulation data is much more prevalent in developed countries. A much more interesting case can be made for attempting to observe the effects of EPL in emerging and transition economies, however, as their labor markets are still developing; thus, changes in EPL could be expected to have larger effects than in a developed economy and can help our understanding of the transmission channels of labor market policy. This, in turn, can offer more accurate policy prescriptions for both transition and developed economies.

   Sadly, little is known about the effects of EPL on the labor market in emerging market economies, and even less about how EPL can change the market in a specific country over time. The purpose of this paper is to help redress this lack of knowledge by examining the effect of EPL in one particular transition economy: Armenia. The Armenian Labor Code, introduced in late 2004, is a modern piece of employment protection legislation that sets restrictions on hiring and firing and represents a comprehensive package of EPL unlike that which has ever existed in Armenia. This paper represents the first attempt to quantify the scale of the Code's impact on the Armenian labor market; in particular, I will take a quantitative and econometric look at the effect of the Labor Code, based on performance in the job market in Armenia both before and after its inception. Through this analysis, I will seek to answer how recent employment protection legislation (EPL) affected the labor market in Armenia, and examine if previous results on the effects of EPL hold true in the Armenian case.

   The paper is organized as follows: Section 2 will provide a brief overview of literature dealing with EPL around the world, while Section 3 will examine the specific case of Armenia and the data available for analysis. Section 4 will describe the data and Section 5 the model, while Section 6 presents the results of the analysis. Section 7 offers some conclusions and avenues for further research.

2. Literature Review: Theoretical Predictions and Empirical Results

   The effects of EPL on the labour market have been well-understood in the context of economic theory, both in terms of its influence on the stock of current workers and the flows of workers into and out of the system. From a theoretical standpoint, any legislation that a government enacts to restrict the flexibility of employers should have a negative influence on the flow of workers, raising the cost to hiring and firing and thus protecting the stock of current workers at the expense of the unemployed. (4)

   In effect, this is a planned market distortion and exactly what EPL is designed to do; in pursuit of "stability" or "worker protection," EPL by its nature seeks to reduce turnover in the labor market that already exists and make it more difficult for turnover to occur in the future. However, the cost for this reduced turnover would not only be felt in the reduced flow of workers, but the cost to society of longer unemployment spells (Blanchard and Portugal 2001). With less turnover, there will be less vacancies and thus those out of work would have to wait longer for a job. EPL also can decrease wages for those seeking to re-enter the workforce: with a longer wait for a job, any applicant would be willing to settle for a lower wage for work (Blanchard 1999 also shows that the cost of unemployment for those already employed would also increase, thus making current workers more willing to accept a lower wage than to confront the threat of extended unemployment). (5)

   Given these negatives, it would appear that EPL is, on an economic level, a policy to be avoided. Yet theorists have conjectured that on various levels (apart from turnover) that EPL can be beneficial, as "policies that try to restore equality of opportunity are warranted even in the absence of efficiency reasons" (Bassnini et. al 2005, p. 10). In their study of workplace training in Europe, Bassanini et. al (2005) note that high workforce flexibility (i.e. high turnover) reduces the incentive to provide skills and knowledge to workers, and thus EPL, by keeping workers employed longer, would also increase investment in human capital on those workers that were employed. From a social/political standpoint as well, EPL reduces the risk of a particular employee being laid-off, reducing political tensions that can arise from job insecurity. Moreover, well-designed supporting or secondary institutions in the labor market (Abraham and Houseman 1994) can mitigate from the employers' side some of the rigidity imposed by EPL, thus finding alternatives to layoffs that also improve employer adjustment to changing market conditions. Finally, as Addison and Teixeira (2001, p. 3) note in their overview of the theory surrounding EPL, "employment protection legislation can enhance productivity performance by encouraging worker cooperation in the development of the production process, stimulate training investments, and reduce 'excessive' turnover" at the aggregate level. These effects are based mainly on the supposition of labor market imperfections, such as asymmetric information and/or misaligned incentives, which can be mitigated by appropriate legislation.

   The myriad types of EPL and how they are implemented are also presumed on a theoretical level to influence different segments of the labor market in different ways (see Table 1); for example, government-mandated maternity leave can be expected to increase the relative cost of hiring women, thus altering an employer's incentives towards hiring males instead. These effects can be mitigated in a Coasean world (see Lazear 1990) through flexible institutions that enable bargaining to distribute cost; for example, restrictions on firing workers could be alleviated for employers by having employees pay a sum equal to the net present value of the firing cost upon their hiring. However, in the real world, and especially in a world that imposes EPL, this kind of flexibility is usually lacking and the distortions that EPL impose can remain.

   While a large literature has coalesced around both the theoretical effects of EPL on labor markets, unfortunately there is little agreement on their effects empirically, with many studies finding no effect of EPL on broader employment trends, with others showing effects only in specific employment indicators. While the most sweeping study done by Botero et. al (2003) finds, consistent with the theoretical framework noted above, that "heavier regulation of labor is associated with a larger unofficial economy, lower labor force participation, and higher unemployment, especially of the young," (6) the record of other research is mixed. Indeed, most cross-country studies finding that EPL does not have a significant effect on labor markets in terms of aggregate employment or unemployment (see Micco and Pages 2007 for a brief review). For example, an early attempt by the OECD (1999) to estimate EPL effects over 1985-97, using two-period panel regressions, finds that in almost all cases that EPL has a negative but insignificant effect on employment (with the only significant effect on the sub-group of "prime [working]-age men"). Similarly, Heckman and Pages (2000) use a sample of OECD and Latin American countries but find little significance across their specifications for EPL and unemployment.

   The reason for the divergence between theory and evidence has also been a bone of contention for economists, with several ideas mooted, including that there are country-specific effects not captured in the panel data, or, more likely, omitted variable bias distorts the effect of EPL. Additionally, indicators of EPL can be misspecified, as they have been mostly...