ENTERPRISE SURVEYS 50337 ENTERPRISE NOTE SERIES INFORMALITY 2009 Entry Regulation, Labor Laws, and Informality Siddharth Sharma oes the regulation of entry cause informality in the manufacturing sector of developing D countries? This note presents evidence from India that suggests that a policy reform that ENTERPRISE NOTE NO. 1 lifted barriers to entry lessened informality. This reduction in informality was accom- panied by gains in average labor productivity in the informal sector. The most likely reasons for this productivity gain appear to have been a decline in inefficient single-household firms and an increase in more efficient multi-person firms. However, the impact of entry deregulation also ap- pears to have depended on state-level labor laws, being measurably larger in states with more pro-employer labor laws. Informality is endemic to firms in developing countries. For form led to growth in registered factories, the degree of this example, more than 80 percent of industrial employment in expansion differed across states because of differences in India is estimated to be in unregistered establishments.1 It is their labor laws. We will now describe the reform's effect on often asserted that despite obvious gains from registering, the informal sector, and compare it across states that differ such as better access to credit, firms choose to stay infor- in the flexibility of their labor laws.3 mal in order to avoid the burden of inflexible regulation. Consistent with this view, cross-country investigation re- Informality Declined after Entry Deregulation veals that informality is more prevalent in economies with Figure 1 below presents the aggregate numbers for India's greater entry regulation, and rigid labor laws (Djankov et informal manufacturing sector in 1988, 1994, and 2000. al. 2002; Djankov and Ramalho 2009). First, it shows that informality declined nationwide between This note presents the results of a new study that uses 1988 and 1994, the survey years falling right before and data from large, nationally representative surveys of India's after the wave of deregulation. In particular, the total num- WORLD BANK GROUP informal ("unregistered") enterprises2 to examine how in- ber of informal enterprises fell by 10 percent, while total formality is linked to entry barriers and labor regulation. employment fell by 3 percent. The lower decline in em- The point of departure for this inquiry is a major reform of ployment suggests that most of this contraction was driven entry regulation in the early 1990s. India's industrial li- by people leaving self-employment, rather than closures of censing system had made it notoriously difficult to set up or multi-worker informal firms. expand factories in the formal sector. In 1991, the Indian Second, Figure 1 shows that the informal sector recov- government "delicensed'' nearly half of all industries at a ered during 1994­2000. This could be a result of other re- stroke. This was the first step in a program of economic lib- forms that India undertook in the 1990s, reducing barriers eralization in response to a macroeconomic crisis. Previous to trade and FDI. The key difference between delicensing research (Aghion et al. 2008) has shown that while this re- and other reforms is that while delicensing was largely com- Figure 1 Informality before and after Delicensing tion states, delicensing expanded the formal sector in flex- Trends in Informality ible labor law states (Aghion et al. 2008) by reallocating 25 workers from informal to formal enterprises. Regression analysis on data disaggregated by state and industry confirms that this pattern is statistically significant: (in millions) 20 compared to neutral labor law states, states with pro- employer labor regulation saw a 25 percent larger contrac- 15 tion in the number of the informal enterprises (Sharma 2009). Pro-worker states, on the other hand, evidenced the 10 same lower level of contraction as neutral states. 1990 1995 2000 Year How Deregulation Improves Productivity Number of Establishments Employment Barriers to entry allow inefficient producers to survive. Re- National Aggregates cent research shows that delicensing played a major role in increasing productivity in India's formal manufacturing sec- Source: Survey of Unorganized Manufacturing. tor, with estimates of productivity gains in the order of 32 percent in the deregulated industries (Chari 2009). This plete by 1991, other reforms were phased in slowly, and in- productivity gain in the formal sector happened in two tensified during the late 1990s. Had the informal sector con- ways. First, relaxation of capacity expansion allowed the tinued to decline even after 1994, there would be reason to more efficient producers to expand, which in turn forced suspect that the 1988­94 contraction was related to reforms the least efficient firms to shut down. Second, relaxation of other than delicensing. Thus, the recovery of the informal entry increased the level of competition in the formal sector. sector in the late 1990s actually strengthens the case that In Figures 1 and 2, we saw that the same reform also en- the decline during 1988­1994 was an outcome of couraged exit from informality. Did this have an effect on delicensing. average productivity in the informal sector? It is held that the efficiency loss from informality is related to scale: since The Decline in Informality Depended on informal firms have to remain small in order to stay under Labor Laws the regulatory radar, they fail to take advantage of In India both the central and state gov- economies of scale. Moreover, unregis- ernments have jurisdiction over labor tered enterprises find it difficult to access legislation. This has caused the states to Delicensing reduced credit markets. Finally, given high trans- differ markedly in their labor regulation, informality in flexible port costs, small firms might be limited a fact that has been linked to growth di- to selling their products in local markets, vergence across Indian states. In a promi- labor law states which are easily saturated. nent study, Besley and Burgess (2004) But to what degree do these factors used data on state-level amendments to matter? Perhaps the best way to answer India's common labor law to classify each state as pro- this question is to see what happens to productivity in in- employer, neutral, or pro-worker, with pro-employer states formal enterprises when barriers to entering the formal sec- having the most flexible labor laws. They found that amend- tor are lowered. India's delicensing "experiment" thus ments in the direction of rigidity were associated with lower presents a unique opportunity to measure the productivity state-level growth. consequences of informality. The informality surveys reveal that the effect of deli- The results indicate that deregulation-induced exit from censing varied across states with different labor laws. Figure informality is associated with major productivity increases. 2, Panels A and B, show how changes in informality are dif- Panel C in Figure 2 shows that compared to inflexible states, ferent across pro-employer ("flexible") and neutral/pro- value added per worker in the informal sector increased in worker ("inflexible") labor law states. These graphs show flexible labor law states between 1988 and 1994. Thus, the that the number of enterprises and employment in the in- states that saw relative exits from informality also saw a rel- formal sector declined sharply during 1988-1994, in flexible ative increase in informal labor productivity. Regression es- labor law states. Inflexible labor law states, in contrast, saw timates also indicate that compared to neutral states, value no such decline. Interestingly, this difference in growth added per informal worker increased by 36 percent in pro- across the two groups of states vanished during 1994-2000, employer states. And in pro-worker states, which saw rela- which suggest that the pre-1994 divergence was the result tive entry into informality, value added per worker fell by of delicensing. Thus, compared to inflexible labor regula- 18 percent (relative to neutral states). 2 Figure 2 Comparing Changes in Informality across States with Inflexible and Flexible Labor Laws A. Number of Establishments B. Employment C. Value Added Per Worker 7 13 12 (in thousands) (in millions) (in millions) 10 9 4 7 6 1 4 3 1989 1994 1989 1994 1989 1994 1989 1994 1989 1994 1989 1994 Inflexible Flexible Source: Survey of Unorganized Manufacturing. The data also show that while flexi- question. One of the benefits of formal- ble labor law states saw a relative de- izing firms is that it offers workers some cline in total informal employment, Deregulation-induced degree of formal protection. In this there was no corresponding relative de- exit from informality is sense, more pro-worker labor laws bene- cline in total revenue generated in the fit those who are able to find formal jobs. informal sector. Thus, there might be associated with major However, labor laws that are too pro- such a thing as "overcrowding" in the productivity increases worker may inadvertently hurt overall informal sector: as more and more in- social protection by encouraging infor- formal firms crowd into the same local mality. In the informal sector there is no market, demand saturates, and prices worker protection. This tradeoff must be drop sharply. Conversely, as people leave the informal sec- considered when weighing the costs and benefits of labor tor, there is a bigger market share available to each re- regulation. maining informal worker. Studies suggest that there are substantial efficiency gains Notes in moving from the single person firm (self-employment) 1. This estimate is based on data from National Sample Surveys to multi-worker firms (Tybout 2000). In Figure 1, we saw and the Annual Survey of Industries. that the contraction in the informal sector was largely driven 2. The unregistered (or unorganized) sector, by definition, con- by people leaving self-employment. This suggests that sists of the self-employed, and those working in establishments deregulation also helped raise productivity by lowering the with fewer than 10 workers. This note uses survey data from the prevalence of inefficient single-person enterprises. 1988, 1994, and 2000 rounds of the Survey of Unorganized Man- ufacturing, conducted by the National Sample Survey Organiza- The Costs and Benefits of Labor Regulation tion. Most developing countries have a large population working 3. I will not present the technical details of these results in this in small, informal establishments or in self-employment. note. Much more detail, including regression results, can be found Given the extremely low productivity levels observed in the in the more extensive Sharma (2009). informal sector, there is a great deal of interest in designing policies that encourage a reallocation of employment to for- References mal enterprises. The findings described in this note suggest ., . Aghion, P R. Burgess, S. Redding, and F Zilibotti. 2008. The un- that removing regulatory barriers to industrial entry can be equal effects of liberalization: Evidence from dismantling the successful not only in stimulating growth in the formal sec- license raj in India. American Economic Review 98 (4). tor, but also in reducing informality. This happens mainly Besley, T., and R. Burgess. 2004. Can labour regulation hinder eco- because rising labor demand in the formal sector induces nomic performance: Evidence from India? Quarterly Journal of people to leave less profitable self-employment. Economics 119 (1). The effect of entry deregulation on labor demand, the re- Chari, A. V. 2009. The aggregate productivity effects of entry and sulting exit from informality, and the consequent produc- output restrictions: An analysis of license reform in India. tivity improvement all depend on labor laws. In fact, the Working Paper, Cornell University. Indian experience suggests that deregulation may succeed . Djankov, S., R. L. Porta, F L. de Silanes, and A. Shleifer. 2002. The only when labor laws are sufficiently flexible. regulation of entry. Quarterly Journal of Economics 117 (1). This complementarity between entry liberalization and Djankov, S., and R. Ramalho. Employment laws in developing the flexibility of labor regulation raises an important policy countries. 2009, Journal of Comparative Economics. 3 Sharma, S. 2009. Entry Regulation, Labor Laws and Informality. Working Paper, Enterprise Analysis Unit, The World Bank. Tybout, J. 2000. "Manufacturing Firms in Developing Countries: How Well Do They Do and Why?" Journal of Economic Litera- ture, Vol. 38, March. The Enterprise Notes Series presents short research reports to encourage the exchange of ideas on busi- ness environment issues. The notes present evidence on the relation between government policies and the ability of businesses to create wealth. The notes carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this note are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. 4