Underemployment is the key problem India needs to target. Doing so will help cut outright unemployment as well. Underemployment manifests itself in low-productivity employment. Some of this low productivity can be observed in daily life. For instance, it is commonplace in India to find two or more workers performing tasks that one worker can do. If a home appliance such as air conditioning or television requires servicing or the plumbing needs fixing, often two or more workers show up. But only one of them will actually perform the task, with the others either carrying the tools or simply providing company.

In government offices and even private companies, it is common to find workers sitting outside office doors nearly all day with little to do. Even workers in small and informal private enterprises can be seen taking long breaks or simply shirking their duties.

Concrete evidence of dominance of low-productivity jobs in India is to be found in a surprisingly large concentration of workforce in small, often tiny, and informal enterprises in both industry and services. The Indian industry and service sectors are perhaps unique in the world for being populated by such a large number of micro and small enterprises and such a small number of medium and large enterprises. The combination of low per worker productivity in micro and small enterprises and heavy concentration of employment in them translates into overall low per worker productivity in industry and services.

Before presenting the evidence, it bears noting that enterprise surveys, which are the only sources of detailed estimates on the distribution of workforce, output and wages by enterprise size, remain highly partial in coverage in India. Indeed, even the periodic Economic Censuses, which aim to collect information on a handful of variables from all enterprises in all non-agricultural enterprises, fall well short of covering the entire non-agricultural workforce. For instance, the number of non- agricultural workers reported in the latest Economic Census, conducted in 2013/14, was only 131 million, compared with 243 million based on the Employment–Unemployment Survey of 2011/12.

There are two sets of enterprise surveys in India: the Annual Survey of Industries (ASI) conducted annually by the Central Statistical Office (CSO) and the survey of unincorporated or unorganized sector non-agricultural enterprises periodically conducted by the NSO. The former is confined to manufacturing enterprises registered under certain sections of the Factories Act, 1948. The latter covers unincorporated enterprises in both industry and services with some sectors such as the financial services or construction occasionally excluded. It explicitly excludes enterprises covered by the ASI. The two surveys are, thus, mutually exclusive in their coverage.

The latest NSO survey on unincorporated enterprises relates to the year 2015/16 and covers all non-agricultural sectors except construction. The enterprises account for a total of 111.3 million workers. Keeping in mind that the total number of workers identified by the 2013/14 Economic Census, which also covered incorporated enterprises, was 131 million, it can be concluded with a reasonable degree of confidence that the survey covers 80 per cent or more of non- agricultural workforce identifiable in all non-agricultural enterprises. Therefore, an extremely large part of the workforce in India remains employed in the informal sector.

To gain further insight into the characteristics of the workforce, consider the nature of the enterprises covered by the survey. These enterprises can be classified according to two criteria: employment of hired workers on a regular basis—or lack thereof—and ownership pattern. Under the nomenclature used by the NSO, enterprises that run without employing any hired worker on a regular basis are called own account enterprises (OAEs), and those employing one or more hired workers are classified as establishments. Using the ownership criterion, the enterprises are classified into proprietorship, partnership, self-help groups (SHGs) and trusts.

At an all-India level, 84.2 per cent of the enterprises in 2015/16 were OAEs. In rural areas, the share of these enterprises was even higher at 91.4 per cent. Therefore, the vast majority of the enterprises hired no workers on a regular basis. In terms of ownership, 96 per cent of enterprises were proprietorships on India-wide basis. Of the remaining, 1.4 per cent enterprises were partnerships and 1.8 per cent SHGs. This, once again, indicates the generally informal nature of employment.

The full impact of the concentration of non-agricultural employment in unincorporated enterprises, especially OAEs, can only be appreciated when we see it in conjunction with the gross value added (GVA) per worker.

First, the GVA per worker is consistently significantly lower in manufacturing than in services. In rural areas, the gap is particularly large. Second, predictably, OAEs consistently exhibit lower GVA than establishment enterprises. And, finally, there is a large gap in per worker GVA between rural and urban enterprises. At the aggregate level, urban enterprises exhibit 1.9 times per worker GVA than rural enterprises. In manufacturing, urban GVA per worker is more than twice that in rural areas.

In the manufacturing sector, rural OAEs exhibit less than one-sixth of the GVA per worker at nationwide level. This low level of productivity relative to the national average in unincorporated enterprises is almost as bad as in agriculture. Even in urban areas, the GVA per worker in manufacturing is less than half of the nationwide GVA per worker. The inevitable conclusion is that too many workers, even in industry and service, are employed in low-productivity enterprises.

Just as there is a preponderance of tiny farms in agriculture, there is a dominance of tiny firms in industry and services in India. This feature has resulted in low output per worker and hence low incomes for the bulk of the Indian workforce. India needs to create more productive jobs in industry and services so that workers may exit agriculture.

A key conclusion following from the analysis in this chapter is that the creation of high-productivity jobs requires the entry of significantly larger firms in industry and services than is the case. It is not merely that large-scale firms themselves are more productive; their presence also helps small and medium firms achieve high productivity. Large firms effectively become agents of change in technology, skills and management. They also demand high-quality components that small and medium firms may supply.

Therefore, the key policy question we must address is how best to facilitate the emergence and entry of large firms in India.