As the first article in the series focussed on the broad contours of change in the entrepreneurial landscape at the state and gender level, now for the second, let us consider two questions- where we try to know more about entrepreneurs in India through PLFS surveys. First question is does higher education attainment result in greater proclivity towards self-employment. Second question is whether self-employment has higher earnings than regular salary or wages. 

For the first question, we will look at the highest formal educational attainment as well as technical education and how it correlates with usual status, that is, status of employment for the large part of the preceding year. We start with looking at what is the distribution of all types of employment statuses vis-a-vis general educational attainment, for 2017-18 and 2022-23. 

Table 1: General educational level and self-employment (as % of total)

At any level of educational attainment, employers account for a small fraction of the total while own account working constitutes a larger share. (Table 1) This remains true in 2017-18 as well as 2022-23. We see an increase in the percentage of employers between 2017-18 to 2022-23. We observe a similar increase in the percentage of own account workers as well. The increase in employers, though small in absolute percentage terms, is remarkable when we consider the low proportion of employers themselves. For all educational levels of primary education and above, we see nearly a 1% increase in the proportion of employers. The increase is larger for two groups with highest educational levels, graduate and postgraduate and above. But the group which shows the highest proportion of employers is those with diploma or certificate courses as the highest educational attainment. 

Diploma or certificate course is typically a technical education. In 2017-18, only 2.7% of individuals above age 15 have reported any technical education. Among the technical diplomas or certificates, engineering or technology diplomas or certificates accounted for nearly half of it. Among those 15 or more years old without any technical education, employers accounted for about 0.9% while among those with technical education, employers accounted for 2.7%, nearly 3 times more. Situation in 2022-23 shows little change. 3% of those 15 or more years old reported having technical education. Employers accounted for 1.4% among those without any technical education while 3.4% among those with.

Own account working falls as the education attainment rises. We observe it in both 2017-18 and 2022-23. One interesting observation in 2022-23 PLFS is that a large proportion, about 60%, of those with age 15 or more and highest general education attainment less than primary schooling have reported being in an educational institute. This has led to reduction in the proportion of own-account workers. There is a small increase in the percentage of own account workers from 2017-18 to 2022-23.   

As general educational attainment grows, self-employment in the form of being an employer accounts for a larger share with decline in own account working. This fits in with the expectation that higher general educational attainment should lead to more productive ideas as well as easier access to credit resulting in more productive enterprises which will employ people beyond the entrepreneur. An implication for those working in entrepreneurship development exists in the fact that those endowed with technical education are more likely to be employers than those without. The implication is perhaps there is low hanging fruit if we try to nurture entrepreneurship among those getting technical education. Equipping those in technical education with a foundational entrepreneurial skill set as well as providing them credit and market linkages if they chose to become an entrepreneur can be the pathways to realise this low hanging fruit. Another possibility is to consider innovative ways to improve technical educational attainments of existing entrepreneurs or workers so that more productive self-employment results. 

Table 2: Percentage of regular salaried or wage earners in total (%)

One possible argument to be considered is whether self-employment is the result of lack of the other livelihood opportunities, most importantly those which can lead to regular salary or wages. Table 2 shows that such does not seem to be the case prima facie. The proportion of regular salaried or wage-earning individuals grows as education attainment grows, showing that there is no obvious case of self-employment arising out of lack of choice. 

Table 3: Monthly earnings (in rupees) based on Current Weekly Status (CWS)

Table 3 provides further evidence that there is more to self-employment than lack of choice. In both 2017-18 and 2022-23, employers report higher monthly earnings than their own account workers and regular salaried/wage-earning individuals. Employers also have the highest growth rate of earning in the 2017-18 to 2022-23 than the remaining two groups. If there is any scope to consider ‘lack of choice or necessity’ argument, then it is for your own account working. It has lower earnings as well as higher incidence among those with lesser educational attainments. The further consideration among urban-rural and gender lines shows that these inferences are not driven by any subgroup. 

A word of caution is in order. There is a temptation to use samples of employers or own account workers to see more patterns associated with them. This will be an indirect sample arising out of a sample of households based on demographic distribution. The problem is this indirect subsample is unlikely to be representative of the underlying population of employers or own account workers in India since distribution of employers across Indian states will not be solely driven by demography but by economic factors as well. But PLFS sampling is based on demography only. Though the dearth of information on entrepreneurs forces us to be entrepreneurial in knowing more about them, we must acknowledge that our conclusions based on PLFS data are limited by lack of representativeness.