The use and misuse of Economics C. P. Chandrasekhar

When the final session, prior to the impending election, of the current Parliament ended in February, high on the list of the unfinished business of the Modi-led NDA government was its aggressive effort to rewrite the laws regulating wages and employment conditions of workers in India. While opposition by workers’ movements, trade unions (including those aligned to the ruling BJP), the parliamentary opposition and democratic opinion has managed to stall the effort, it is more than likely to be revived by future governments. The conservative legislative push of the NDA was presented not just as an effort at rationalisation that would replace 44 central laws with four comprehensive codes relating to wages, industrial relations, social security, and health and occupational safety, but also as a win-win strategy for all. Employers would gain, it is argued, because they would experience an improvement in the ease of doing business, with greater freedom to set the terms and duration of employment and to fix wages and benefits. So would workers, who would also ostensibly benefit from the so-called “reform” of labour laws that would inexplicably promote processes of formalisation and improve wages, benefits and conditions of work. Finally, by triggering investment and growth, the economy as a whole would benefit, not least through an acceleration of the hitherto tardy growth of employment that falls short of labour force growth in the country.

This claim that policy changes that alter the terms of the contested relationship between employer and employee or the relative emphasis on private profit as opposed to social benefit would improve the welfare of all, should under normal circumstances be dismissed as that much propaganda. But across the world this kind of case for more flexible labour markets with lighter regulation of capital-labour relations has been supported with research claiming to be objective and even scientific, buttressed often by the use of mathematical models and statistical analysis. The burden of that analysis has been that laws that perceive inequality in the relationship between employers and employees, and protect workers with regulations relating to wages, working conditions and the process of collective bargaining, hurt rather than help workers, by limiting growth and destroying existing or potential jobs. Conceptually, it is argued that “restrictive” labour laws raise wage costs and encourage labour displacing technologies, hurt profits and limit investment, and discourage expansion, especially in labour intensive industries, resulting in inadequate exploitation of economies of scale. But, more than such questionable, and often simplistic, theorising, the work more widely cited by advocates of deregulation is empirical in nature, making the case for policy change “evidence-based”.

With respect to India too, there is much by way of a literature that argues that stricter labour laws hurt employment and growth, and, therefore, the workers whom they seek to protect. Among the works frequently cited, not just to make this case with respect to India, but to use India as an example that has implications for other developing countries is an article by two economists from the prestigious London School of Economics, Timothy Besley and Robin Burgess, entitled “Can labor regulation hinder economic performance? Evidence from India”, published in 2004 in the highly rated Quarterly Journal of Economics. By using national and state level evidence on organised industrial sector performance and examining its relationship with the level of labour regulation, while ostensibly controlling for other influences, B&B claimed to have established “that much of reasoning behind labor regulation was wrong-headed and led to outcomes that were antithetical to their original objectives” and that “attempts to redress the balance of power between capital and labor can end up hurting the poor.” Trying to attract attention to their results, the two economists went on to declare that “the “battle cry of labor market regulation is often that pro-worker labor market policies redress the unfavourable balance of power between capital and labor, leading to a progressive effect on income distribution. We find no evidence of this here—indeed the distributional effects appear to have worked against the poor.”

There have been many scholarly critiques of such arguments for dismantling worker protection. But in a recently released paper, Servaas Storm, an economist at the Delft University of Technology in the Netherlands, has provided a damaging critique ^[1] of this contribution from Besley and Burgess defending what Storm (following Albert Hirschman) calls the “perversity trope” that argues that labour laws meant to protect workers actually hurts them. Some conclusions in the Besley and Burgess (B&B) paper are startling, explaining in part its influence. Thus, according to their “estimates”, Andhra Pradesh generated 199,000 more jobs by 1990 than would otherwise have been the case, because of pro-employer labour legislation. On the other hand, West Bengal is seen to have missed out on creating 180,000 more jobs by 1990 because of its adoption of pro-worker measures. Not surprisingly, the paper has even been referred to in official Indian assessments making a case for labour market “reform”. The official Economic Survey 2006 referred to the paper to defend labour market deregulation on the grounds that it shows “that States, which have enacted more pro-worker regulations, have lost out on industrial production in general.”

Given that background, Storm’s critique of B&B is a revelation. It reports a failure to replicate their claimed results, which in any case are shown to be ‘flimsy’ and non-robust. It also points to the internally contradictory and empirically implausible nature of their results. Storm concludes that the B&B paper in no way ‘establishes’ the perversity thesis, but rather illustrates “how a combination of scientific pretension and a deep desire for respectability can lead to a gratuitous empiricism in which priors trump evidence—to the detriment of socio-economic progress.” The next government would do well to look at this critique seriously before being carried away by the reactionary rhetoric of the likes of B&B.

Central to B&B’s empirical investigation is a metric of labour regulation they construct, identifying individual amendments of labour legislation as being pro-worker, neutral or pro-employer and giving them a value (of +1, 0 or -1). As more than one scholar has pointed out this arbitrary allocation and ‘valuation’ of amendments is often problematic or downright wrong. Moreover, the final measure of the nature of regulatory change in a state based on adding up these numbers is arrived at based purely on amendments to the Industrial Disputes Act, ignoring the influence of a large number of other state and central laws that are often more important in determining the nature of labour market regulation in individual states. Finally, the B&B exercise completely sidesteps the issue of poor enforcement, or the failure to actually implement labour laws, which Indian evidence suggests is rampant. Storm demonstrates that when attempts are made to correct for these errors of omission or commission, the results reported by B&B either turn insignificant or are reversed.

In sum, the measure of “labour market regulation” B&B use to explain industrial performance is not just weak but a distortion. It is not surprising, therefore, that in the 1980s, when according to B&B a number of pro-worker amendments were reportedly introduced, other indicators point to a weakening of workers’ collective bargaining power in India. In particular, workers had failed to protect their shares in valued added in industry and got none of the benefit resulting from increases in productivity, with the wage share falling and that of profit rising. What is more, it is in those states where regulation was seen as “pro-worker” going by the B&B metric, that the wage share declined the most.

Storm’s concern is not just with the simplistic arguments and ‘flimsy’ statistical analysis of a paper that has provided grist to the mill of those who want to undermine even the limited protection afforded by law to workers in India and elsewhere. Rather, he sees it as part of a larger set of papers arguing that labour market regulation negatively impacts economic performance, which have been rendered influential through publication in high-profile, “peer-reviewed” journals that are seen as purveying high-quality research.  Almost all of them have been found to be “robustly non-robust” when subjected to scholarly scrutiny by other economists.

The fact that these papers contained arguments that are implausible and statistical exercises that are not robust shows that both the research agenda and method have been crafted to deliver results that suit a specific slant in policy making. That such partisan and non-robust work gets academic sanctity through publication in “prestigious” journals is indicative of the fact that recognition of excellence is often driven by shared biases rather than objective scrutiny. Economics was never a “value-free” science. But it is precisely those who have scientific pretensions, who appear in such instances to unabashedly serve the cause of advancing reactionary ideas and policies.

[1] Servaas Storm, Labor Laws and Manufacturing Performance in India: How Priors Trump Evidence and Progress Gets Stalled”, Working Paper No. 90, Institute of New Economic Thinking, January 2019, available at https://www.ineteconomics.org/uploads/papers/WP_90-Storm-Labor-Laws.pdf

(This article was originally published in the Frontline Print edition: March 15, 2019)