Is job creation hindered by rigidities in the labour market?

Theoretical arguments

According to the standard neoclassical theory, output growth is supposed to lead to employment growth, and interventions in the labour market, e.g., through labour laws, trade unions, or minimum wages, distort the labour market and prevent it from producing the optimal outcome in terms of employment. It is argued on the basis of that theory that employment growth, especially in developing countries, is constrained by restrictive labour laws and trade union interventions that create rigidities in the labour market. But the debate on this issue is far from settled because there are studies pointing out that available evidence does not lend support to the hypothesis of labour market rigidities constraining employment growth. Before looking at the empirical evidence on the two sides of this debate, it may be useful to recapitulate the theoretical arguments behind this hypothesis and break it down into its major components so that the evidence on each may be systematically examined.

There are at least three aspects of labour market imperfections that, in theory, can adversely affect employment by creating rigidity and making adjustment difficult. First, minimum wage legislation may prevent wages from falling at times and in situations when downward adjustment is needed in order to clear the market. For example, even in economies characterized by the existence of surplus labour, minimum wage legislation may prevent wages from being pushed downward beyond a certain level and employment from expanding. Also, when labour demand declines during an economic downturn, one way of bringing about adjustment would be to reduce wages, but the existence of minimum wages may prevent that from taking place.

Second, regulations relating to the labour market, e.g., employment protection law (EPL), severance pay, provision of unemployment insurance and the duration of the benefit, etc. may make employers reluctant to hire when needed. For example, provisions in a country’s EPL relating to firing of workers may create a perception among employers that it will be difficult to make downward adjustment in the workforce when they may have to. That, in turn, may discourage them from hiring, even when additional workers are needed during a period of economic expansion. Requirement to make severance payment for terminating workers may have the same effect mentioned above.

It is often argued that the existence of unemployment insurance and a generous duration of its benefits may discourage workers from actively looking for jobs and encourage them to remain on benefits for a longer period than is warranted by conditions in the labour market.

Third, imperfections may be created through the industrial relations route when trade union activities influence workforce adjustment or wage rates. That would imply that the level of unionization in an industry or firm may have an adverse effect on employment.

The empirical evidence

As for the impact of legislated minimum wages on employment, the available empirical evidence does not enable one to draw a definite conclusion in either direction. It would be important to note in this context that the literature on this subject is so vast that it would be almost impossible even to attempt a comprehensive review.⁵ Most of these studies focus on developed countries, although there are some on developing countries as well.⁶ Not surprisingly, studies differ considerably in terms of the methodologies employed and the type of data used. One study (Broecke et al., not dated) points out that the effect on employment depends on methodology and data used, and more recent (post-1995) studies are more likely to show positive effect on employment.

One review (Neumark, 2018) covering developed countries concludes that most country studies show negative effect on employment though there are some that show positive effect as well. For developing countries, the results are more mixed — studies showing both positive and negative effects and, in some cases, no impact at all. One review that covers studies on ten major emerging economies, viz., Brazil, Chile, China, Colombia, India, Indonesia, Mexico, Russian Federation, South Africa, and Turkey,⁷ concludes that increases in minimum wages have minimal or no impact on employment (Broecke et al., not dated). Menon and Rodgers (2017) report positive effect for Brazil and Mexico and negative effect for Indonesia and Bangladesh. Bel- man and Wolfson (2016) mention a study on Indonesia that shows negative employment effect for small domestically owned firms and positive effects for large domestically owned farms and farms with some foreign ownership. The same review reports result from a study on China that found positive employment effect for firms that survived.

Nataraj et al. (2014) conducts meta-regression analysis of the impact of minimum wages on employment and finds that minimum wages are associated with lower formal employment and higher informal employment, and so, the impact on overall employment is ambiguous.⁸ However, they admit that the study “cannot derive solid conclusions from the ... review because of the paucity of studies and the methodological difficulties encountered” (p. 556).

One study on India (Menon and Rodgers, 2017) that uses National Sample Survey Organisation’s employment survey data and data on minimum wages across states and industries present interesting results. For urban areas, the study shows no impact of minimum wages on employment, while in rural areas, there was positive income effect without any negative effect on employment.

In Bangladesh, the minimum wages for the ready-made garment industry were raised in 2006, 2010, and 2013 without any adverse effect on employment. As shown in Chapter 7, employment in the industry not only increased after the 2006 adjustment, but the rate of growth was higher compared to the years before the rise (Table 7.1). Employment in the industry stagnated in the industry after 2012, but that cannot be ascribed to the rise in minimum wages in 2013. After the major accident in 2013, the industry had to face severe scrutiny on a variety of fronts including safety standards and the situation regarding rights of workers. The changes that are taking place in the industry as a result include changes in the size structure and introduction of more modern machinery. So, it is difficult to identify a single factor (be it minimum wages or structural and technological change) that has caused employment to stagnate after 2012.

The upshot of the above review of the evidence on the impact of minimum wages on employment in developing countries is that there is very little basis to argue that minimum wages (or for that matter, high wages) cause rigidity in the labour market, which, in turn, is responsible for low or negative growth in employment. The experience of Bangladesh provides a good example of flexibility in wages and yet decline in the growth of employment. Data presented in Chapter 4 show that real wages in the country - overall as well as in manufacturing — declined between 2012 and 2106; and yet, employment growth declined.

There are studies pointing out that labour market institutions in the form of EPL or unemployment benefits do not necessarily hinder the growth of employment (e.g., Bean, 1994; Nickel, 1997; Forteza and Rama, 2002; Kap- sos, 2005). On Europe, a widely cited article by Bean (1994) argues that the available evidence does not show that the existence of generous unemployment benefits was the cause of persistent unemployment. Nickel (1997) also shows that all types of labour market rigidities do not have an adverse effect on unemployment rates.⁴ He concludes:

Labour market rigidities that do not appear to have serious implications for average levels of unemployment include the following: 1) strict employment protection legislation and general legislation on labour market standards; 2) generous levels of unemployment benefits, so long as these are accompanied by pressures on the unemployed to take jobs by, for example, fixing the duration of benefit and providing resources to raise the abilityAvillingness of the unemployed to take jobs; and 3) high levels of unionization and union coverage, so long as they are offset by high levels of coordination in wage bargaining, particularly among employers.

(Nickel, 1997, p. 72)

A study by Forteza and Rama (2002) covering 119 countries (i.e., both developed and developing) shows that minimum wages and mandated benefits do not hinder economic growth. They argue that curtailing social security benefits might not contribute much to economic performance.

An econometric exercise undertaken by Kapsos (2005) demonstrates that rigidities in the labour market do not have a negative effect on employment elasticity. In his cross-section analysis with data from 100 countries, he uses the World Bank’s “employment rigidity index” (which is the average of three indices, viz., difficulty of hiring, difficulty of firing, and rigidity of hours) and finds that there is no statistically significant relation between employment elasticity and employment rigidity index. Moreover, the sign of the coefficient is not in the expected direction.

One study (Chetty, 2013) addresses the question whether extending unemployment benefits increases unemployment rates in the USA. On the basis of the results of nearly a dozen economic studies on the country, it concludes:

...a 10-week extension in unemployment benefits raises the average amount of time people spend out of work by at most one week. This simple, unassailable finding implies that policy makers can extend unemployment benefits to provide assistance to those out of work without substantially increasing unemployment rates.

Labour market rigidity and employment: the debate on India

The debate that took place on labour market rigidity and employment in India during the 1990s and the first decade of the 2000s is perhaps the most notable among such debates covering developing countries. So, it may be useful to look at that debate, albeit briefly. The debate was basically spurred by the decline in employment growth during the 1990s compared to the earlier decades despite a notable rise in GDP growth, and some studies (e.g., Besley and Burgess, 2004) blaming it on labour market rigidity - especially on the difficulty of firing workers in large farms under the Industrial Disputes Act of 1947.¹⁰ On the practical side, after the introduction of wide-ranging economic reforms in 1991, employers of India started clamouring for reforms in the labour market and trade unions took the opposite position. In order to understand this debate, it would be useful to start by looking at some numbers that lay at its background. Table 8.1 presents some numbers compiled from various studies.

A few observations may be made on the basis of the figures presented in Table 8.1. First, for the economy as a whole, growth of employment declined after 1990. Although there was some recovery after 2000, the growth rate did not go back to the pre-1990 rates. Second, growth of employment in organized manufacturing rose substantially over the decades. Except for a period of negative growth during 1997-98 to 2001—02, employment growth in the sector was higher in the 2000s compared to the 1990s and the earlier decades.¹¹ It was growth of employment in the economy as a whole that declined in the 1990s compared to the earlier decades. And since that happened when overall economic growth was picking up, and people started wondering whether that growth could be dubbed as jobless, a debate on the causes of low employment ensued, and the spotlight fell on labour market institutions. But if one focusses on organized manufacturing, it cannot be said that there was a secular decline in employment growth. And as the labour market rigidity argument would apply more to the organized sector, it seems that there is no prima facie case for arguing that employment growth declined in India because of labour laws causing rigidity.

However, Besley and Burgess (2004) pointed out that states which amended the Industrial Disputes Act in a pro-worker direction experienced lower

Table 8.1 Annual Growth (%) of Employment in India

Sources: Goldar (2004), Chose (2015), ILO (2009), and Sharma (2007).

output, employment, and investment in registered formal manufacturing.¹² And on the basis of that evidence, they argued that the balance of power is tilted in favour of workers, and that is not good for long-term growth of output and employment. Fallon and Lucas (1991) argued that faster growth of wages relative to consumer prices abetted by job security regulations resulted in the reduction of employment during 1959—60 to 1981-82. But a number of subsequent studies produced empirical evidence to argue that (i) wage increases were justified by increases in productivity (Papola, 1994), and (ii) labour market institutions play a minor role, if any, in determining employment (Sharma, 2006, 2007).

Using data from a survey of 1300 manufacturing firms across nine industrial groups, Sharma (2007) showed the following: (i) increase or decrease in labour cost did not have a significant effect on employment; (ii) a higher proportion of the bigger firms were able to lower employment, which implies that there were factors other than labour laws that played a role in the ability to adjust the workforce; (iii) both unionized and non-unionized firms raised capital intensity; and (iv) there was flexibility in the labour market irrespective of provisions in the labour laws.

Like the above-mentioned study, Ghose (2015) concludes:

The evidence also shows that flexibility is very much there in practice. There is widespread use of regular informal and casual employees in the organized sector. ... Enterprise surveys show that most entrepreneurs view labour regulations as a minor problem.

(p. 11)

Based on a review of a set of studies, Shyam Sundar (2005) concludes: (i) minimum wage laws can be and are often by-passed, so much so that this is not a source of rigidity; (ii) revisions to minimum wages are so ineffective that labour costs are unlikely to impose a burden on employers; (iii) presence of trade unions does not act as a deterrent to raising employment; and (iv) employment in the organized manufacturing sector grew in the first half of the 1990s, “despite the prevalence of unaltered statutory regulations impacting on employment decisions of firms” (p. 2276).

Although the above summary of the labour market flexibility debate on India is not intended to be a synthesis of the entire literature in this field (which is vast), it nevertheless indicates that the empirical basis for blaming labour market rigidity for low employment growth is not strong. Admittedly, the studies cited above use different indicators of rigidity in labour markets and use different methods to examine the impact on employment growth. And the results often are influenced by methods and data sources. But they should provide adequate ground to at least cast doubt on the argument that making labour markets flexible would provide solutions to the problem of low employment growth.

Despite the lack of consensus on whether labour market rigidity is to be blamed for low employment growth in India, policymaking seems to be focussing primarily on labour laws and labour market reforms — if the chapter titled “Employment and Labour Reforms” in the latest strategy document of the government (NITI Aayog, 2018¹³) is any indicator. While the document mentions a few schemes for generating employment (including the well-known Mahatma Gandhi National Rural Employment Guarantee Scheme — MGNREGS), the seven items¹⁴ mentioned in the section on “Way Forward” does not include anything on economic policies needed to make growth more employment-intensive via a faster growth of sectors that are labour-intensive by their character. Labour law reforms seem to be the mechanism that is expected to facilitate the growth of such sectors.¹³