The Penalty of Being Young: India’s Workers During the Pandemic

Abstract

With the evolution of Covid-19 since its emergence in 2020, the pandemic has had multiple economic effects—effects which manifest as immediate shocks—but also as scarring effects having long-term repercussions. Certain demographics may be more exposed or vulnerable to these long- and short-term impacts. This chapter focuses on young workers who entered the Indian labor market for the first time during the pandemic. Using all-India CMIE-CPHS data, we track a panel of both the young workers, and the young entrants to examine this. Our findings reveal that even though there is only a marginal difference in the likelihoods of finding employment when comparing between the pandemic and the pre-pandemic entrants, the pandemic entrants face a greater disadvantage in the intensive margin in terms of the type of employment. There was a rise (drop) in the more precarious forms of employment like daily wage (permanent salaried) for the pandemic entrants as compared to their pre-pandemic counterparts. Further, they suffer disproportionately in terms of the associated earnings from this employment. The pandemic entrants made 60% lower monthly income than the pre-pandemic entrants in 2019. Even by 2022, the temporary salaried workers among the pandemic entrants continued to make 4% lower income as compared to the starting income of their pre-pandemic counterparts.

Both authors have contributed equally to this work.

6.1 Introduction

The Covid-19 pandemic had far reaching effects extending beyond its immediate health repercussions. The disruption in economic activities, restrictions in mobility, and the contraction of the global economy in the months thereafter meant that this was not a short-term disruption and likely had long-run implications for workers across the globe. In particular, as many have pointed out, the pandemic exacerbated existing inequalities in the labour market with marginalized communities and groups suffering disproportionately. In this chapter, we focus on one of these groups, i.e. young workers.

In a global survey of youth (18–29 year olds) conducted by the International Labour Organisation (ILO) across 112 countries, 17% reported having lost their jobs. Nearly a quarter reported a reduction in working hours while about 42% reported a reduction in their income (ILO, 2022a). Country-specific studies also indicated young workers being disproportionately affected. In the United States, for instance, while 16–29 year olds accounted for only a quarter of the workforce, nearly a third of the rise in unemployment rate between February and April of 2020 were attributed to them. Additionally, the rise was much higher among Black and Hispanic youth (Alba & Aaronson, 2020). In countries in Latin America too, the youth were worst impacted compared to older workers. However, they were able to return to employment faster than older workers. However, most of this recovery was into informal employment and occupations that were more favorable towards young workers (gig and platform work (ILO, 2022b)). Similarly, in Asia and Pacific, unemployment rates increased in Australia, Indonesia, Japan, Malaysia, and Vietnam, as well as in Hong Kong and China with youth unemployment rates predicted to double by 2021(ADB & ILO, 2020).

In India, the issue of youth unemployment had already been a major policy challenge. Between 2011–12 and 2018–19, youth (18–30 years) unemployment rate had increased from about 6% to nearly 15%. If youth were categorized further by their level of education, unemployment rate was even higher among educated youth. Graduates reported an increase in unemployment rate from about  21% to 32% during the same period (Basole et al., 2021). It was in this context that a crisis like Covid-19 hit. Preliminary evidence found that nearly 60% of older workers did not face any job loss during the economic lockdown while the corresponding share among younger workers was only 30%. Moreover, younger workers were also less likely to return to work after a job loss (Basole et al., 2021).

In this context, the second section of this chapter uses the most recent all-India household survey data to see how young workers have fared vis-a-vis older workers. We track individuals who were employed pre-pandemic and examine what happened to them during the lockdown period (in 2020) and then two years subsequently (in 2022). This allows us to understand both the immediate impact as well as the relatively long term persistence of job loss during the lockdown. We compare the impact and persistence between young and old workers to understand if and how young workers are differently affected in terms of job loss, transitions from jobs, as well as persistence of job loss compared to older workers.

Besides the immediate and more obvious impact of the lockdown on young workers, there is also a ‘scarring’ effect that youth are particularly vulnerable to as new entrants into the labour market. Kahn (2010), for instance, found that youngsters graduating during a recession are in lower-level occupations with lower wages and persistent negative impacts. Similarly, Schwandt and von Wachter (2020) found that youngsters entering the labour market for the first time in 2020 potentially stood to forfeit earnings to the extent of $400 billion over the next ten years of their working lives, a penalty of graduating into a bad economy. It is likely that in India too, in a labour market that was already plagued by high youth unemployment, young entrants to the labour market during these bleak years may also be particularly disadvantaged in comparison to workers who had entered during normal years. In the third section, we specifically track the young, erstwhile students, who would have entered the labour market in 2020. In doing so, we examine if there are any ‘scarring’ effects owing to entering the labour market during an economic downturn. The fourth section concludes.

6.2 The Costs of Being a Young Worker

The Center for Monitoring Indian Economy’s (CMIE) Consumer Pyramids Household Survey (CPHS) provides a unique high frequency panel dataset that allows us to track individuals over multiple times in the year, across several years. The CMIE interviews households three times a year collecting information about the demographics of all household members including their employment status, type of employment, industry, and occupation. More importantly, during the economic lockdown, CMIE (temporarily) transitioned from a field-based survey to a phone survey, effectively being one of the only large scale surveys of individuals during the economic lockdowns of 2020.

Since we are particularly interested in how workers were impacted and recovered after the economic lockdown, we leverage CMIE-CPHS panel dataset of workers. The first economic lockdown, one of the most stringent in the world (Mathieu et al., 2020), was imposed in India starting March 24th, 2020. It was extended multiple times, and for most of the next two months, the country was in a near-full lockdown. Economic activities contracted severely as mobility was restricted and the economy effectively shut down. Although the national lockdown was only announced by March-end, from early March onwards, mobility had begun to be severely restricted as captured by the stringency index (Fig. 6.1).

Fig. 6.1

Stringency index for India

The severe restrictions clearly had economic impacts as is evident from Fig. 6.2.

Fig. 6.2

Pandemic phases, infection, and mobility rates. Source Jha and Lahoti (2022)

Figure 6.2 gives the overall picture of the manifestation of the pandemic in the country vis-a-vis the different phases of the pandemic by overlaying reported number of Covid-19 cases (left axis) against various measures of mobility restrictions as measured by Google mobility index (right axis).

To understand the economic impact of these months, we identify the months of March and April 2020 as ‘impact’ months. Using these as anchors, we construct a panel identifying the same individual pre-pandemic, exactly one year ago in March–April 2019. This is the baseline on the basis of which we benchmark impact. Similarly, to uncover the long run recovery after the impact, we track these same individuals two years after the impact, i.e. in the months of March and April 2022. Finally, since we are interested in the labour market impact, we restrict the analysis to pre-Covid workers. Therefore, we essentially track pre-lockdown workers during the lockdown months, and two years thereafter, to see what was the impact and their long run recovery in labour market outcomes.

By restricting the analysis to only workers, for every individual, we are able to identify four possible trajectories. Table 6.1 describes these trajectories.

Table 6.1 Employment trajectories

An individual may have been completely unaffected in terms of labour market outcomes during and after the lockdown. This is the ‘No effect’ trajectory. On the other hand, a worker may have lost employment during the lockdown period and not have been able to return to work which we refer to as the ‘No recovery’ trajectory. They may also have been able to hold on to their jobs during the lockdown but subsequently lost employment two years later—‘Delayed job loss’. Finally, an individual having lost work during the lockdown may have subsequently returned to work by the beginning of 2022, constituting the ‘Recovery’ trajectory.

6.2.1 Employment Costs

In general, nearly 80% of the pre-pandemic workforce were unaffected and followed a ‘no effect’ trajectory. They did not lose work during the first four months of 2020, and remained employed two years later. A marginal share, approximately 7%, faced a ‘no recovery’ trajectory. However, this overall number hides large variations across different groups. Women and young workers were particularly affected as were less educated workers, as has been explored elsewhere (although using a slightly different trajectory panel) (Abraham et al., 2022; Basole et al., 2021; Deshpande, 2020). Since we are particularly interested in the impact on young workers, we explore the trajectories of young workers vis-a-vis older workers.

We broadly categorize the workforce into four age categories with the youngest at 18–23 year-olds, and the oldest being individuals 45 years and above. Figure 6.3 provides the distribution of trajectories by each age group.

Fig. 6.3

Employment trajectories by age group. Source Authors’ calculations using CMIE-CPHS

Examining the share of individuals who were unaffected by the lockdown, it is evident that middle-aged workers were the least affected with nearly 80–90% of these workers not losing employment during or the years after the lockdown. In contrast, the worst off were the youngest of workers, and to a lesser extent, the oldest workers. Only 55% of young workers were unaffected. Instead, about 17% of young workers had no recovery from a job lost during the lockdown period. Combined with those who suffered a delayed job loss (16.7%), workers who were unemployed by the beginning of 2022, accounted for about 34% of young workers. For the middle aged group individuals, the corresponding share was only between 12% and 8%. Clearly, younger workers were far more likely to lose their jobs and not return to work compared to older workers. This is not surprising since young workers have less experience and firms find it less costly to fire these workers compared to more experienced (older) workers. Further, young workers often have fewer networks and social capital in the labour and consequently find it harder to return to the labour market after having lost jobs (ILO, 2020; ILO, 2022a).

To what extent does the trajectory differ within different kinds of young workers? Similar to other studies, we find a disproportionately larger impact on young women compared to young men. Nearly 41% of young women workers followed a no recovery trajectory compared to only 16% of young men. Similarly, about 57% of young men were unaffected by job loss, while for women, this was only 23%.

When young workers are categorized by their level of education, interestingly, we can see that it is the more educated young workers who are more likely to suffer from job loss and not recover (Fig. 6.4). As education levels increase, there is a clear trend with a decline in the share of workers following a no-effect trajectory and an increase in the share having a no recovery trajectory. About a quarter of young workers who had education above graduate level had a no recovery trajectory, compared to only 10% of less educated workers. This curious pattern may be explained by aspects from the demand and supply side. On the demand side, it is likely that the kind of jobs that were more likely to recover or could be recovered into, were also the ones that were more likely pursued by less educated individuals. These include self-employment and casual wage work which are characterized by an ease of entry (in terms of required education levels, capital investment) as well as ease of exit. Salaried work, on the other hand, is less easy to return to having lost employment, and hence does not have the ease of entry associated with casual and self-employment. Indeed, we find that among permanent salaried workers, nearly a quarter experienced a no-recovery trajectory, compared to only 12% of daily wage workers. Since salaried work is likely to be more pursued by higher educated individuals, this could explain why higher educated workers witnessed a more muted recovery.

Fig. 6.4

Employment trajectories of young (18–23) workers, by education level. Note Authors’ calculations using CMIE-CPHS

On the supply side, the less recovery among higher educated workers could be explained by the fact that these kinds of workers were more likely to come from richer households and hence could ‘afford’ to remain unemployed or return to education. Indeed, this conjecture is confirmed in that the households that experienced a recovery were in fact the poorest households, while individuals who faced no recovery were more likely to come from the richest households (based on pre-pandemic average household income level).

It is also the case that for many older workers who are likely to be the main earner in their family, it is imperative to return to work. Therefore, they are more likely to resort to fallback employment or less paid precarious work. Young workers, on the other hand, have the ‘luxury’, in a sense, of not having to work. In fact, in times of economic downturn, many workers often return to education/training in an effort to acquire more skills while jobs are in shortage. We compare the transitions to different employment arrangements as well as out of the workforce entirely using the transition matrix below (Fig. 6.5).

Fig. 6.5

Employment transitions across employment types and out of the labour force. Note OOLF stands for out of labour force. Darker shades represent larger shares. Authors’ calculations using CMIE-CPHS

The transition matrix indicates what employment arrangements people have moved into over the two years. So, the 48% in the first cell under daily wage workers indicates that 48% of 2019 daily wage workers remained as daily wage workers. Another 15% moved out of the labour force while a marginal 3% and 5% moved into temporary and permanent salaried employment. Darker shades represent larger shares. For older workers, across all employment arrangements, we see a large movement into self-employment. So about 30–25% of salaried workers had moved into self-employment. In contrast, for younger workers, we see very little transition to other forms of employment, but rather between 30%–40% of young workers, irrespective of employment arrangement had left their jobs entirely, as can be seen in the high share of OOLF categories for all employment types. Since the large exit out of the labour force obscures intra-workforce transitions, we restrict the analysis to individuals who remained employed between the two periods (Fig. 6.6).

Fig. 6.6

Intra-workforce transitions. Note Authors’ calculations using CMIE-CPHS

Comparing intra-workforce movements, we find that there is more stickiness, as represented by the diagonal elements, for younger workers in daily wage work and temporary salaried work compared to older workers. The permanent salaried workers have similar levels of stickiness, whether young or old. Older workers in self-employment are more likely to continue as self-employed compared to younger workers. Further, older workers see far more transitions into self-employment, unlike younger workers. Therefore, analogous to the case of men compared to women, as found by Abraham et al. (2022), here too, we find that younger workers seem to have a disadvantage in finding fallback employment and are more likely to withdraw from work entirely.

Not surprisingly, the majority of the young workers who left the workforce are now reporting themselves as students. However, the data does not allow us to identify whether they are indeed enrolled in education or not. About 85% of the displaced young workers were now students, compared to only 13% of displaced older workers. Even two years after the most stringent economic lockdown and the inevitable contraction of the economy, many of these erstwhile workers have not been able to return to work. This also has important policy implications. India already has a problem of educated unemployed. As more and more individuals have withdrawn to pursue education/skilling, this problem will only be exacerbated unless targeted policy interventions are in place to bring them back to appropriate jobs in the workforce.

6.2.2 Earnings Costs

For those workers who managed to retain their jobs, what has been the income implications of the lockdown in its immediate aftermath, and two years thereafter? Here too, have young workers suffered a larger loss of earnings compared to older workers? CMIE-CPHS allows us to track monthly earnings of workers.¹ We compare the average monthly earnings of workers across different employment arrangements, pre and post the lockdown. Note that average earnings here is limited to those who remained employed in both periods. Given the intra-workforce flux that we saw earlier, it is likely that a salaried worker in 2019 may no longer be salaried, but rather may be self-employed or a casual wage worker. Nevertheless, a comparison of the average earnings between the two periods can provide an understanding of relative change in earnings between younger and older workers (Fig. 6.7).

Fig. 6.7

Average monthly earnings by type of employment arrangement for older and younger workers, pre and post lockdown. Note The blue bars represent older workers, red bars represent younger workers. Earnings are restricted to those who are employed and report a non-zero income. Earnings are in real terms, in 2000 prices. Authors’ calculations using CMIE-CPHS

We can see that, in keeping with the traditional Mincerian wage and experience predictions, average earnings for older workers are always higher than that of younger workers (Mincer, 1958). And, not surprisingly, permanent salaried workers earn the highest at approximately Rs. 30,000 per month, followed by temporary workers, self-employed and daily wage workers. Between 2019 and 2022, there has been a secular increase in earnings for all employment types, except self-employed, for both young and older workers. Therefore, although younger workers have suffered disproportionately at the extensive margin of employment loss, in the intensive margin of earnings, they have fared similar to their older counterparts.

6.3 The Costs of Being a Young Entrant

In order to understand the implications of entering the labour market during this economic turmoil, we track a subgroup of the youth who are identified as the young entrants who would have entered the labour market in 2020, i.e. the year of the pandemic.² We identify this cohort as individuals between the age of 18 and 23 who report themselves as students and were out of the labour force in 2019.³ These are individuals likely to enter the labour market in the coming year. We follow them till 2022⁴ and document their employment trajectories along with their corresponding incomes. Our working sample thus consists of a balanced panel of individuals who were students, out of the labour force, and between 18 and 23 years of age in 2019. Employment and income outcomes of these pandemic entrants are then compared with those who entered the labour market in ‘normal’ times or the baseline. To do this we again create a balanced panel of individuals who were students, out of the labour force, and in the 18–23 age bracket in 2018. We track the employment and income outcomes of this cohort, who are likely to enter the market in 2019—a normal, pre-pandemic year.

In the analysis we are thus tracking two cohorts—(i) the pandemic cohort are tracked in 2020 and in 2022, (ii) the pre-pandemic cohort are tracked in 2019. The two cohorts (pandemic cohort; and normal year, pre-pandemic cohort) are not likely to be different in characteristics—the only significant difference between them being the year they entered the market. Comparing the outcomes for these two cohorts thus gives us the difference on account of their entering the labour market in the year of the pandemic. In doing so, we examine if there are any penalties or ‘scarring’ effects owing to entering the labour market during an economic crisis. Since we have data for only two years for the pandemic cohort, we cannot comment on the long-term impact yet, rather we use the employment and income information for this cohort in two years (2020 and 2022) to gauge the impact and recovery vis-a-via the baseline.⁵

6.3.1 Employment Outcomes of the Pandemic Cohort

There are 27,636 individuals who form a part of the balanced panel of the pandemic cohort, i.e., individuals who were interviewed in all three years (2019, 2020, and 2022), belong to the age bracket of 18–23, and report themselves as students and out of the labour force in 2019. Of these labour market entrants, 6% were able to find employment in 2020, 13% remained unemployed, and 82% continued to remain out of the labour force (Fig. 6.8).

Fig. 6.8

Distribution of the baseline/pre-pandemic cohort and the pandemic cohort. Source Authors’ calculations using CMIE-CPHS

We compare the above individuals against a similar panel of ‘young’ workers who entered the labour market during a ‘normal’ year. The balanced panel for this baseline group consists of 33,230 individuals. They reported to be students, out of the labour force, and in the 18–23 age bracket in 2018, forming the non-pandemic cohort of 2018. Of them, 9% were able to secure employment, 15% were unemployed, and 76% remained out of the labour force in the year 2019. The labour force thus shrank by 5% in the pandemic year on account of the young entrants, as compared to the normal pre-pandemic year. Even after two years of having entered the labour market, only 10% of the young entrants of the pandemic cohort were able to find employment.

In addition to the question of how many were able to find employment, it is also important to look at the kinds of employment that they were able to secure, and how it compares vis-a-vis their predecessors, or the labour market entrants during a normal year.

Figure 6.9 gives us the distribution of the nature of employment that the labour market entrants were able to secure. The baseline (2019) gives the nature of employment for workers who were students in 2018 and entered the workforce in 2019. This is compared with the labour market entrants of 2020 who in turn are tracked over two years—2020 and 2022. We compare the first year of entrance in the market for both cohorts, i.e., 2019 for the baseline cohort, and 2020 for the pandemic cohort, to understand the differential impact of entering the market in the year of the pandemic. Further, the employment distribution for the pandemic cohort in 2022 is used to understand how much of the difference persists.

Fig. 6.9

Type of employment secured by labour market entrants in 2019, and those entering in 2020 tracked in 2022. Source Authors’ calculations using CMIE-CPHS

There was a drop in the percentage of individuals who were able to get into the most secure form of employment—permanent salaried. While 10% of the baseline cohort was able to get a permanent salaried job on entering the labour market, the corresponding number for the pandemic cohort was only 7%. A greater proportion of the pandemic cohort got absorbed in daily wage work, and self-employment—the more precarious kinds of employment—in comparison to the baseline cohort. The pandemic cohort however makes some recovery in about two years time. By 2022 the proportion of workers in permanent salaried jobs went up to 12% and the self-employed fell to 33%, though the proportion of daily wage workers and those in temporary salaried jobs continued to remain higher than the respective proportions in the pre-pandemic cohort. So, even though there is only a marginal difference in the extensive margin of finding employment between the pandemic and the pre-pandemic cohort, the pandemic cohort faces a greater disadvantage in the intensive margin of the type of employment.

6.3.2 Earnings of the Pandemic Cohort

While there is a difference in the nature of employment secured by the pandemic cohort and that secured by the baseline pre-pandemic year cohort—with the pandemic cohort performing poorly both in terms of the size of the total workforce as well as the share of those getting the more secure salaried jobs—there is also an associated income story which too is likely to differ between these two cohorts.

We go deeper to examine the intensive margin of income loss for the labour market entrants of the pandemic cohort who are able to secure employment. The average real monthly income of the employed labour market entrants in 2020 was Rs. 3,946.⁶ In contrast, the employed labour market entrants in 2019 were earning Rs. 9,588 on an average. The pandemic cohort was thus making around 60% lower monthly income than the baseline year, pre-pandemic cohort. The income difference varied depending on the nature of employment (Fig. 6.10). The difference in incomes for the daily wage workers and the self-employed of the pandemic cohort vis-a-vis the respective pre-pandemic cohort was the highest at 62%, i.e., the daily wage workers and the self-employed of the pandemic cohort were earning 62% lower monthly income on average as compared to their pre-pandemic counterparts. The difference was the least for the permanent salaried workers of the pandemic cohort, who were earning 38% lower income than their pre-pandemic cohort.

Fig. 6.10

Average monthly earnings of pandemic and baseline cohort, by employment type. Source Authors’ calculations using CMIE-CPHS

Even after gaining a two-year experience in the market, the pandemic cohort of employed workers was able to make only around 9% higher income as compared to the starting income of the pre-pandemic cohort. By 2022 the permanent salaried workers of the pandemic cohort were making 22% higher income than the starting income of the permanent salaried workers of the pre-pandemic cohort, while the temporary salaried workers of the pandemic cohort continued to make 4% lower income even after two years as compared to the starting income of their pre-pandemic counterpart. Increment in earnings of the daily wage workers and the self-employed after two years of working was only around 5–7% higher than the starting salary of the respective workers in the baseline pre-pandemic year cohort.

Comparing the young entrants of the labour market in the pandemic year with those in the pre-pandemic year, we find that the penalty in terms of finding a job is marginal in the extensive margin of not getting employment. However, there is a significant difference in the intensive margin of the kind of employment one is able to secure, and the intensive margin of the earnings. So the pandemic entrants faced a penalty at the extensive margin of securing employment, which was relatively less than the penalty they suffered at the intensive margin in terms of the kind of employment, and the associated earnings from this employment.

6.4 Conclusion

The Indian economy is going through its potentially most productive period where the youth bulge can be turned into a demographic dividend. However, on the downside, if the economy is not able to provide satisfactory employment and income earning opportunities to this mass, it can turn into a ‘demographic bomb’ (Lin, 2012).

The analysis in this chapter explores the economic fate of two kinds of youth—the young who were already working on the eve of the pandemic (young workers); and the young, erstwhile students, who entered the labour market in the year of the pandemic (young entrants).

On comparing the effect of the pandemic on the young workers vis-a-vis the older workers, we find that the younger workers have suffered disproportionately more in terms of losing employment during the pandemic. In terms of their earnings however, for the young who were able to retain their employment, they fared similar to their older counterparts. The penalty for the young workers thus was primarily in terms of their ability to retain their employment.

In contrast, when we focus on the young entrants during the pandemic, we find that vis-a-vis their predecessors they were only marginally worse-off in securing employment.⁷ However in terms of both the kind of employment they were able to secure, and their earnings from those employment, the young entrants of the pandemic were at a greater disadvantage than the pre-pandemic cohort. A greater proportion of the young entrants of the pandemic cohort got absorbed in the more precarious kinds of employment, and a smaller segment was able to secure permanent salaried employment. Further, the starting earnings of these entrants took a large hit vis-a-vis the pre-pandemic cohort.

It is also pertinent to note that we continue to live in deeply uncertain and volatile times with the infection rates on the rise even as we write this chapter. Our analysis suggests that the young are at a greater disadvantage—whether they are already a part of the labour market or a fresh entrant. We will be able to get a complete picture of the actual penalty only after some more years when the economy has moved past the continuing damages of the pandemic, and the employment and income trajectory of the workers has stabilized. But till then it is imperative that we take measures to provide productive employment and decent pay to our youth to exploit the golden period of the economy’s youth bulge. Poor employment opportunities and a lack of decent income avenues at the beginning of one’s career is likely to have scarring effects on the workers for their entire labour market life-cycle. In the face of this unprecedented crisis, safeguarding the economic outcomes of our youth is thus critical to exploit the economy’s demographic dividend, aside from its intrinsic social-ethical merits.