C P Chandrasekhar & Jayati Ghosh

The ‘missing middle’ in Indian industry

CP CHANDRASEKHAR JAYATI GHOSH | Updated on December 08, 2014

The middle order: It's alive and kicking.

bl9_Macroscan_chrt3_NET.jpg

bl9_Macroscan_chrt2_NET.jpg

bl9_Macroscan_chrt1_NET.jpg

Pundits and policymakers should stop making the strawman argument. Official data reveals Indian industry has a thriving pool of mid-sized companies with impressive output

Discussions on Indian industry — and in particular, on the implications of labour laws for employment especially in manufacturing industry — tend to take for granted the problem of the “missing middle”. Simply put, this is the idea that industries in developing countries like India tend to be dominated by a large number of tiny enterprises and a few large firms, but disproportionately few small and medium enterprises.

In terms of distribution, it is seen to be expressed in terms of a U-shaped curve whereby small and very large enterprises would dominate over medium-sized enterprises in terms of number of factories, number of workers, or share of output.

Such a feature has typically been attributed to various institutional features: for example, the difficulties of obtaining bank credit for expansion; and even more importantly, the plethora of regulations, especially labour protection regulations and laws that restrict the ability of firms to hire and fire at will, and that do not allow profitable firms to expand to medium size.

While this argument is advanced for many developing countries, it has a particular resonance in India, and is frequently used by those who argue for labour deregulation as a means of improving the conditions of operation for manufacturing industry.

The barrier argument

In the Indian case, this argument is scarcely new. It was, for example, the central argument of a World Bank study produced in the 1980s (“India: Industrial Regulatory Policy Study”, World Bank Industrial Strategy and Policy Division, Washington DC, December 1986).

In that study it was argued that barriers to both entry and exit are high for large firms, but significantly lower for small and tiny ones. Most of these barriers were seen to result from industrial regulation and its mode of implementation.

According to the Bank then, this resulted in notable losses in terms of potential efficiency and productivity gains for industry and for the economy as a whole. It was argued that “cross-country and Indian experience indicates that medium size firms often enjoy better labour relations and higher labour productivity than large firms, and respond more effectively to changing technological and market requirements. Their policy-induced absence has added an element of rigidity and contributed to the slowness of technological progress and structural change in Indian industry.”

Arguments like this are heard very commonly today, and the need to encourage small firms to grow into larger ones by reducing the regulatory barriers to their expansion (including labour laws) is one of the cornerstones of the current government’s approach to industrial policy.

The “Make in India” strategy may presently seem to be more about bluster and hyperbole than about action, but what policy action there is seems to be focussed on easing regulatory constraints and especially dismantling laws and rules that protect labour. The underlying perception is that such regulations need to be undone since they have contributed to the “missing middle” in Indian industry. Indeed, some recent studies (such as “Understanding the missing middle in Indian industry: The role of size dependent regulations and fiscal incentives by K Ramaswamy, 2013) argue that the missing middle is the outcome of the threshold effects of labour regulations defined by employment size and fiscal incentives determined by turnover of small scale enterprises.

Not missing

So widespread is this notion in official circles that it is taken for granted as axiomatic in all policy discussions and analyses propagated by mainstream economists and journalists. So it is almost embarrassing to ask: what missing middle? As it happens, the official data that we have suggest that there is no such thing!

Whether in terms of share of workers or share of output, the distribution of factories by number of workers suggests the opposite of a U-shaped distribution, with medium-sized factories accounting for the bulk of both.

Surely, it cannot be that those in charge of directing policy for industrial development in the country have failed to notice this rather obvious feature emerging from the official data? But surprisingly enough, that is what seems to be the case. Chart 1 shows the number of factories according to the Annual Survey of Industries, for the latest year 2011-12.

It is evident that, while micro-enterprises employing less than 15 workers are certainly the largest in number, accounting for more than one-third of the total number of factories, thereafter the distribution proceeds along ‘expected” lines, that is the number of factories decreases as the size of employment in the factory increases, in a fairly regular fashion.

The only sharp discontinuity is between the smallest two categories, that is, less than 15 workers and between 15 and 19 workers, where the number of factories drops sharply. But this merely suggests that really tiny enterprises dominate in number, which is no surprise.

There is no evidence of any sharp drop in number of enterprises related to the point at which labour laws are supposed to kick in, for example the sections of the Industrial Disputes Act that stipulate conditions (like government permission) for firing workers from companies employing more than 100 workers.

Indeed, as Charts 2 and 3 indicate, it is the medium-sized firms that actually dominate in both employment and output. Chart 2 that shows the distribution of workers across different factories by employment size, points to a very healthy inverted-V shape, rather than the U-shape posited by proponents of the existence of the Missing Middle.

As a result, firms employing from 100 up to 500 workers account for 45 per cent of total employment in registered manufacturing!

Similarly, these medium-sized firms — the ones, which according to the World Bank, should have better labour relations and higher labour productivity — also dominate in total output, as evident from Chart 3. This shows that more than a quarter of manufacturing output in 2011-12 was provided by factories employing between 100 and 500 workers. If these official data are to be believed, then not only is there no problem of a missing middle in Indian industry, but both employment and output are substantially delivered by the most desirable middle-sized firms!

Reality checks

Incidentally, such an analysis is confirmed by other recent cross-country studies that include India. A study of India, Indonesia and Mexico by Chang-Tai Hsieh and Benjamin A Olken (“The Missing “Missing Middle”,” Journal of Economic Perspectives, vol. 28) that used data on the distribution of both formal and informal sector manufacturing firms comes to very similar conclusions.

They find that, while there are a very large number of small firms, there is no “missing middle” in the sense of a bimodal distribution.

Instead, the fraction of firms of a given size declines smoothly with increasing firm size. They further examine the regulatory and tax notches in India, Indonesia and Mexico that are typically thought to discourage firm growth – and they find no economically meaningful bunching of firms near the notch points.

This is very important for our policy makers to understand. If they persist in designing policies on the basis of wrong hunches and with no respect for the reality revealed by even the official statistics, the likely outcomes may well also be the opposite of what they expect.

Published on December 08, 2014

Follow us on Telegram, Facebook, Twitter, Instagram, YouTube and Linkedin. You can also download our Android App or IOS App.

This article is closed for comments.
Please Email the Editor

Related

This article is closed for comments.
Please Email the Editor