Labour market liberalisation is one of the most important structural reforms, and yet one of the least well understood. The conventional view is that such reforms increase welfare and improve economic performance. For example, changes in employment laws that lower the costs of hiring and firing workers encourage job separations, increase job creation and improve the ability of firms to respond to shocks (Saint-Paul 1997). Yet, these very same reforms can have negative effects on productivity if lower firing costs decrease the investments that firms make in human capital and research and development (Nickell and Layard 1999). The effects of employment protection on growth and productivity are therefore ambiguous (Holmlund 2014) and this can be explained by, for example, non-linearitites (Belot et al. 2007) and lengthy implementation lags (Dix-Carneiro 2014). Freeman (2010) has conjectured that, even if the effects of labour deregulation on growth and productivity are uncertain, employment protection might well reduce income inequality.
We argue that one factor that contributes to our limited understanding of labour market reforms is that much of the empirical evidence is confined to OECD countries and to the period after 1990. This column introduces a new index of labour market regulations rigidity covering over 140 countries since the 1950s to try to shed new light on the determinants and effects of changes in the rigidity of employment protection legislation (Campos and Nugent 2015.) We revisit the conventional explanation for the rigidity of labour laws (namely their ‘legal origins’) and find stronger support for an alternative account centred on trade liberalisation and per capita income levels. As to the effects of labour laws, our econometric results support the view that rigidity decreases income inequality without deleterious effects on growth.
The ‘legal origins’ explanation for the rigidity of a country’s labour market regulations stems from the seminal paper by Botero et al. (2004). They construct an index of the rigidity of employment protection legislation for a cross-section of 85 countries around 1997 and use it to test three alternative explanations: an efficiency theory (proxied by per capita income levels), political theories, and legal origins. Their results show that the ‘legal origins’ explanation dominates – countries with legal systems based on Common Law have less protective labour regulations than those with Civil Law systems. Accordingly, a (French) Civil Law origin tends to generate labour laws that are more rigid, protective, all-encompassing and imposed in a top-down manner than those from an (English) Common Law origin.
New index on the block
We introduce a new de jure index of labour market regulation rigidity covering more than 140 countries, both developed and developing, from 1950 to the eve of the Great Recession. In constructing this new index, our goal was to be as consistent as possible with the methodology used by Botero et al. (2004) to construct their index of employment protection law. Our measure extends their employment law protection to more countries and many more time periods. The longer time coverage is key for understanding the dynamics of a structural reform that is notoriously slow moving. This extension allows new research on determinants and effects of changes (rather than just levels) in employment protection legislation rigidity across countries as well as over time.
Moving across space and time
Although labour laws change slowly over long periods of time, understanding the determinants and the effects of such changes is important because they are deep, widespread, and long-lasting. Enlarging the available time coverage for such measures is thus essential for evaluating both the determinants and effects of changes in labour laws.
Figure 1 shows our index of the rigidity of employment protection legislation (regional averages and standard errors) for all available countries and time periods. It builds upon a database of labour laws from the International Labour Organization funded by the US Department of Labor.1 Its values range from 0 to 2.5 and, as in Botero et al. (2004), higher values indicate more rigid employment protection. Figure 1 supports the view that rigidity of employment protection legislation varies substantially across countries and over time.
Figure 1. Index of labour market regulation rigidity (‘LAMRIG’) across regions since 1950
Although employment laws in the US, Canada, New Zealand and Australia historically show low levels of protection, our index suggests these have increased in the last two decades or so. The rigidity of employment laws in Europe peaks around 1990 while Latin America has since experienced substantial labour market liberalisation. Maybe unsurprisingly, labour laws remain most protective in the former communist countries (‘Transition’) although there are important differences between Central European (new EU members) and former Soviet Union countries.
Using this new index, we fully confirm conclusion of Botero et al. (2004) that the rigidity of employment protection legislation varies, in the cross-section, significantly with ‘legal origins’. Indeed, support for the legal origins theory with our larger sample of countries for the late 1990s is, if anything, even stronger than in Botero et al. (2004). However, moving from levels to rates of change (that is, once one starts focusing on labour market reforms or liberalisation) weakens the relative power of the legal origins explanation. This has led us to consider a wider range of potential reasons. In addition to efficiency, legal and political theories, we examine the role of political economy factors, structural features, economic crises, and of other structural reforms.
Our results highlight the complex character of labour market reforms. We find the greatest support for the roles of trade liberalisation and the level of economic development. Although the patterns are by no means universal, we find labour law rigidity to be negatively associated with per capita income (richer countries tend to have more flexible labour laws) but positively related to trade reforms (countries that open up their economies in one period tend to increase the rigidity of their employment protection legislation in the subsequent period.) The ‘protection for sale’ theoretical framework (Grossman and Helpman 1994) is useful to explicate such a relationship between trade and labour market reforms.
We believe our results also provide further impetus for recent research on the labour market impacts of trade reform (see Caliendo et al. 2015, and references therein) because they highlight, from a very crowded ‘horse race’, such a prominent role for trade liberalisation. We find, among a large group of potential alternative explanations that trade reform is the most important driver of changes in labour laws rigidity. For example, we evaluate various structural factors (such as the share of natural resources in total exports, the share of agriculture in GDP, the Gini coefficient, and foreign aid to GDP), structural reforms (e.g. financial liberalisation, private credit to GDP, and black market exchange rate premium), aspects of the crises beget reform hypothesis (such as magnitude and extent of output contractions, debt crises, high inflation and unemployment), and political economy factors (such as democracy and political constraints capturing formal political institutions, strikes and assassinations reflecting political instability, and civil war and international war capturing violent political conflict). The effects of trade reform on changes in labour law rigidity are found to be more robust than any of these.
The new index also strongly supports the Freeman conjecture (2010) that labour market rigidity reduces income inequality without adverse effects on growth. While this conjecture may be intuitive and unsurprising to some, to the best of our knowledge, these are the first results to assess both its parts or halves using the same sample, time window, measure, and method.
We consider the results above a first step towards a fuller understanding of the determinants and effects of changes in employment protection legislation. Future analysis should be carried out in terms of levels and rates of change, across developed and developing countries and, crucially, over time windows that are sufficiently long to capture the slow dynamics of labour market reforms. The results offer encouraging new evidence on the determinants and effects of labour market reforms. They show that changes in labour market laws are mainly driven by prior trade reforms and income levels, and they also suggest that employment protection laws may decrease inequality without deleterious effects on economic growth. In our view, future research would benefit enormously from further measurement work. New indexes are needed that can differentiate between de jure and de facto liberalisation as well as new higher frequency measures (e.g. annual data would allow joint estimation of short- and long-run effects). As suggested above, it seems the expected payoffs from this type of research remain substantial.
Belot, M, J Boone and J van Ours (2007), “Welfare-Improving Employment Protection”, Economica 74: 381-396.
Botero, J, S Djankov, R La Porta, F Lopez-de-Silanes and A Shleifer (2004), “The Regulation of Labor”, Quarterly Journal of Economics 119: 1339-1382.
Caliendo, L, M Dvorkin and F Parro (2015), “Trade and Labor Market Dynamics,” NBER WP 21149.
Campos, N and J Nugent (2015), “The Dynamics of Labour Market Reform, Growth and Inequality: Evidence from a New Dataset” (Revised version of IZA DP 6881, October 2012).
Dix-Carneiro, R (2014), “Trade Liberalization and Labor Market Dynamics”, Econometrica, 82(3): 825–885.
Freeman, R (2010), “Labor Regulations, Unions, and Social Protection in Developing Countries: Market Distortion or Efficient Institutions,” in D Rodrik and M Rosenzweig (eds.) Handbook of Development Economics: Volume 5 (Amsterdam; North-Holland): 4657-4702.
Grossman, G and E Helpman (1994), “Protection for Sale”, American Economic Review 84(4): 833-850.
Holmlund, B (2014), “What Do Labor Market Institutions Do?”, Labour Economics 30: 62-69.
Nickell, S and R Layard (1999), “Labor Market Institutions and Economic Performance,” in O Ashenfelter and D Card (eds.), Handbook of Labor Economics: Volume 3C (Amsterdam; North-Holland): 3029-3084.
Saint-Paul, G (1997), “Is Labour Rigidity Harming Europe's Competitiveness? The Effect of Job Protection on the Pattern of Trade and Welfare”, European Economic Review 41: 499-506.
1 The NATLEX depository of labour laws is available on-line at http://www.ilo.org/dyn/natlex/