MacroScan (Printable Version) - Employment trends in West Bengal

Since the early 1990s, West Bengal like other Indian states has seen its employment growth lagging behind output growth, so much so that even a relatively dynamic sector like agriculture has been successful in creating very few additional jobs.

Chart 1 indicates the change in rates of male work participation according to the 1991 and 2001 Census of India. The Census defines 'main' workers as those who participate in economically productive work for 183 days or more in a year. Marginal workers, on the other hand, are those who participate in economically productive work for less than 183 days. While this is a measure of underemployment, obviously it would be an underestimate, since even those with more than 183 days of work could be underemployed.



The work participation rate of male workers in terms of main activity fell quite significantly over the decade. Some of this was due to a rise in participation in education, especially for those in the age group of 15–19 years, which is to be welcomed. But this reason is not enough to explain the decline, and it is clear that the absence of productive employment opportunities was dominantly responsible for this. Thus, the decline in the rate of male main workers was countered by the increase in the share of male marginal workers, which points to fewer jobs being available over the year. It also suggests that more secure forms of employment may have been replaced by less secure forms-a possibility also reiterated by the NSS data on the type of contracts.

Chart 2 shows the rates of work participation of women. West Bengal has one of the lowest rates of female work participation in the country. However, unlike the rest of the country, these rates have been rising, both for main and marginal workers. It is not clear how much of this is due to better enumeration of women's work and how much is the real trend. Even here, by far the more significant increase is in terms of marginal work. (It should be noted that the Census definitions and coverage of 'economically productive work' are still rather restricted and tend to leave out a range of household related economic activity. This is one of the reasons why women's work participation as indicated by the Census is so low in West Bengal. The coverage by the NSS is slightly better in this regard.)



(It should be noted that the Census definitions and coverage of "economically productive work" are still rather restricted and tend to leave out a range of household related economic activity. This is one of the reasons why women's work participation as indicated by the Census is so low in West Bengal. The coverage by the NSS is slightly better in this regard.)

Charts 3 and 4 indicate the changes in terms of main occupation, between 1991 and 2001 for men and women separately. The most significant feature in both cases is the relative decline in agricultural work. For male workers, the shift has been in terms of decline in the share of cultivators (who have also declined in terms of absolute numbers). This may reflect the increase in the share of landless households, which is reported by the NSS, whereby few rural men report themselves as mainly working on their own household plots. By contrast, the proportion of male agricultural labourers has remained broadly stable at around 23 per cent. The proportion of 'other workers' too have increased commensurately.





For women workers in West Bengal, the share of both cultivators and agricultural labourers has declined. In this case, the importance of 'household industry' has gone up considerably, probably as part of a survival strategy of working households, given the generally weak employment situation.

While the Census showed a substantial increase in marginal work for both men and women in West Bengal, the results of the NSS Survey point to a rather different conclusion. The NSS defines 'usual status' activity as one where the person is engaged in an activity for a long period of time in the preceding year. Within this, 'usual principal status' refers to the main activity while 'usual subsidiary status' refers to those who would be characterized as 'unemployed' according to the usual status definition, but pursued some activity in a 'subsidiary' capacity. It is evident that this corresponds approximately (but of course not exactly) to the Census definitions of main and marginal work.

According to the NSS results, which are presented in Charts 5 and 6, the ratio of subsidiary status workers to usual principal status workers, has been going down for quite a long time in West Bengal, and especially during the period 1993–94 to 1999–2000. Such declines are evident for men and women, and for rural and urban areas. For male workers the decline was continuous. For women workers, while the ratio increased in the decade 1983–1993–94, it went down substantially thereafter.





In addition to 'usual status' activity, the NSS has other concepts of employment-the 'weekly status' employment that counts a person as employed if s/he has been engaged in any economic activity for at least one hour on any day of the week preceding the survey and 'daily status' employment that defines a person as employed if s/he has worked for 4 hours or more during the day preceding the survey. The 'usual status' definition may be said to provide a stock estimate of the work available, since it deals with the existence of work over a fairly long period of time. By contrast, the 'weekly' and 'daily' status measures can be seen as flow estimates that also give a sense of the current state of labour demand.

Charts 7 and 8 indicate the growth of rural employment according to the NSS, both by the 'usual status' and 'daily status' indicators. In terms of usual activity, while male employment shows a sharp decline, that for women has actually turned negative during the later period. In terms of daily status, which possibly captures the real state of the labour market from the point of view of workers more accurately, both men and women experienced a decline in employment, after reasonable growth in the earlier period.





Charts 9 and 10 show the NSS results for the same variables in the urban sector. Here the condition of male employment appears to be slightly better, in that, while there was deceleration in employment by both usual and daily status indicators, it was relatively less and male employment growth continued to increase above the rate of increase in labour force. However, female employment shows a collapse, with absolute declines according to both indicators.





The change in type of employment must be seen in relation to this aggregate employment stagnation, since it defines the broader conditions under which labour contracts change. In what follows, we consider patterns specifically in the rural areas of West Bengal.

Chart 11 shows the changing terms of employment for rural men. It is clear that the most significant changes are decline in regular employment and increase in casual work. While these are part of long-term trends, the recent period appears to have seen an acceleration in these changes. The change in self-employment is more complex, since the share of self-employment increased in 1993–94, only to come down to almost the earlier levels by 1999–2000.



For rural women, as shown in Chart 12, the pattern is rather different. One has to keep in mind that the work participation rate of women in West Bengal is already very low, and that the latest period witnessed an absolute decline in employment. Therefore, the increase in share of a particular type of employment does not mean an absolute increase in such work; rather it indicates that such work declined less than other types of work contracts.





This may be why an increase in the share of regular work and a decline in the share of casual work can be observed for rural women. It is likely that most of the job losses were in the sector of casual work, which is why the share has gone down. The increase in the share of self-employment is likely to reflect disguised unemployment in such a context, since this can be a residual or 'refuge' activity when other paid work is not available.

It remains to consider the extent of diversification of employment, given the overall stagnation of employment and the changes in types of work contract. Chart 13 describes the share of non-agricultural employment in total rural employment (usual status, principal and subsidiary activities taken together). The trend is clearly for an increase in the share of non-agriculture in total employment, although this is more prominent for women workers.

By 1999–2000, nearly half of the women engaged in some usual status activity in rural West Bengal were not employed in agriculture. Once again, given the overall decline in employment, this suggests that the job losses for women have been greater in agriculture, rather than a significant expansion in non-agricultural jobs. For men, the increase in jobs in the non-agriculture sector was quite sharp till 1993–94, but seems to have reduced subsequently.

Table 1 provides a picture of the overall extent of employment diversification from agriculture by districts, using data from the 2001 Census. It needs to be borne in mind that these shares of non-agricultural employment (which refer to both main and marginal workers taken together) include other primary activities (such as forestry and fishing) and also refer to both rural and urban areas.



This is why the shares of non-agriculture in total employment are very high, especially for some heavily urbanized districts. While there is almost complete absence of agricultural work in Kolkata, the share of non-agriculture is also very high is Howrah, which is dominantly urban, part of greater Kolkata, and in North 24 Parganas, which is also part of the greater Kolkata hinterland, as well as in Darjeeling, which is dominated by the city of Siliguri.

What is interesting, however, is that very few districts show a domination of agriculture in total employment. Indeed, for male workers, only four districts show a share of non-agriculture in total employment of less than 40 per cent. They are among the districts that are generally considered 'backward': Uttar and Dakshin Dinajpur (with around 32 per cent each), Bankura (36 per cent) and Birbhum (39 per cent).

For the state as a whole, well above half of those employed are engaged in non-agricultural pursuits. This is also well above the national average of more than 60 per cent of workers engaged in agriculture, indicating that in West Bengal, employment diversification is relatively more advanced.

What are the areas in which such diversification has occurred and which of these has been more dynamic? It is worth considering this in detail by looking at the relationship between growth and employment.

It is notable that in West Bengal, the period 1993–94 to 1999–2000, which experienced such stagnation and decline in aggregate employment generation, actually showed a higher rate of growth of economic activity. State Domestic Product in constant price terms actually increased at an annual rate of 6.7 per cent, compared to 5.8 per cent in the period 1983–1993–94. However, aggregate employment growth in the later period in West Bengal according to the NSS was only 0.76 per cent, compared to 2.44 per cent in the earlier period. These data reflect employment elasticities in the recent period which are substantially lower for West Bengal than for the rest of India.

This suggests that the problem was not growth per se so much as the pattern of growth, which became progressively less employment-intensive. This was especially marked in manufacturing and services (other than transport and trade), in both of which employment actually declined, and was despite the fact that in both sectors, output growth was impressive, even double than the rates achieved earlier.

The pattern of services employment is surprising, given that the services sector, particularly self-employment, is typically a residual sector that can become a refuge for those unable to find productive employment elsewhere. However, in West Bengal the freeze on new state government employment after 1991 may have played a role in this. This is also reflected in the decline in regular employment in urban areas that was indicated by the NSS data.

While agriculture has been an important source of employment generation in the past, the rate of growth in jobs in this sector has been declining in the recent period. Of course, the situation in West Bengal is better than in the rest of India, where agricultural employment has actually fallen in the period 1993–94 to 1999–2000. This is probably because of boro rice cultivation, which has meant increased labour demand because of more intensive cropping in many parts of the state.

The few sectors that still show some dynamism in terms of employment generation are wholesale and retail trade, transport and storage, and construction. These are the sectors for which employment elasticity has not declined during the period of 1999–2000. However, the dynamism in these sectors is more prominent for urban areas; in rural areas the impact on employment appears to be less.

Obviously, the lack of adequate employment generation is one of the most pressing economic and social problems in West Bengal at the moment, much as in the rest of India. But the problem seems to be even more acute in this state than elsewhere, given the sharper declines in employment expansion that have been noted. Clearly, this has got to be a critical focus of future policy.

This can be achieved by moving away from those sectors that exhibit low employment elasticity, or by changing the pattern of growth in these sectors. Within agriculture, diversification of cropping pattern as well as the development of improved storage and distribution facilities enhancing cultivators' ability to grow new crops, may lead to more employment generation. In addition, the collapse in formal finance access for farmers, as evidenced for example in the sharp decline in rural credit–deposit ratios of banks, must be reversed.

However, simply encouraging the shift of activities away from employment towards manufacturing and services may not be adequate to cause employment expansion, given the current pattern of growth in these sectors. What is needed is a change in the nature of such growth, which would make these activities also more labour-using than they are at present.

Within manufacturing, the basic problem seems to be that much of the growth has come from large units, which are less employment-intensive, even as many more labour-using small units close down. The pressure of import competition and worsening credit access are, at one level, national problems that are common across the states, and cannot be dealt with through state government policy. But it is still possible to think of methods of encouraging and safeguarding the interests of small and more employment-intensive producers through various state-level incentives.

In services, much of the employment decline has come about because of the ban on fresh hiring by the state government that has affected public employment very drastically. The point is that such declines have not been compensated for by increases in private service employment. While services cannot be treated as one sector for policy purposes as it is such a heterogeneous, catch-all category, it is true that there is scope for expansion of a range of new service activities, especially IT-enabled services which can be provided by educated workers.

The expansion of IT-enabled service activities is a process that has entailed a drop in the proportion of educated unemployed in urban India in general, but there is plenty of scope to develop such activities (which range from call centres and back-office work to accountancy and other activities) in urban areas of West Bengal. For this, of course, better infrastructural facilities (especially power, communication and connectivity) as well as more systematic and directed training of educated youth are required.

On the whole, it appears that there is scope for redirecting economic processes within the state of West Bengal to encourage more employment generation, notwithstanding the overall macroeconomic processes that are currently operating to inhibit the growth of jobs.



Themes > Features	30.06.2003


Employment trends in West Bengal

Since the early 1990s, West Bengal like other Indian states has seen its employment growth lagging behind output growth, so much so that even a relatively dynamic sector like agriculture has been successful in creating very few additional jobs. Chart 1 indicates the change in rates of male work participation according to the 1991 and 2001 Census of India. The Census defines 'main' workers as those who participate in economically productive work for 183 days or more in a year. Marginal workers, on the other hand, are those who participate in economically productive work for less than 183 days. While this is a measure of underemployment, obviously it would be an underestimate, since even those with more than 183 days of work could be underemployed. The work participation rate of male workers in terms of main activity fell quite significantly over the decade. Some of this was due to a rise in participation in education, especially for those in the age group of 15–19 years, which is to be welcomed. But this reason is not enough to explain the decline, and it is clear that the absence of productive employment opportunities was dominantly responsible for this. Thus, the decline in the rate of male main workers was countered by the increase in the share of male marginal workers, which points to fewer jobs being available over the year. It also suggests that more secure forms of employment may have been replaced by less secure forms-a possibility also reiterated by the NSS data on the type of contracts. Chart 2 shows the rates of work participation of women. West Bengal has one of the lowest rates of female work participation in the country. However, unlike the rest of the country, these rates have been rising, both for main and marginal workers. It is not clear how much of this is due to better enumeration of women's work and how much is the real trend. Even here, by far the more significant increase is in terms of marginal work. (It should be noted that the Census definitions and coverage of 'economically productive work' are still rather restricted and tend to leave out a range of household related economic activity. This is one of the reasons why women's work participation as indicated by the Census is so low in West Bengal. The coverage by the NSS is slightly better in this regard.) (It should be noted that the Census definitions and coverage of "economically productive work" are still rather restricted and tend to leave out a range of household related economic activity. This is one of the reasons why women's work participation as indicated by the Census is so low in West Bengal. The coverage by the NSS is slightly better in this regard.) Charts 3 and 4 indicate the changes in terms of main occupation, between 1991 and 2001 for men and women separately. The most significant feature in both cases is the relative decline in agricultural work. For male workers, the shift has been in terms of decline in the share of cultivators (who have also declined in terms of absolute numbers). This may reflect the increase in the share of landless households, which is reported by the NSS, whereby few rural men report themselves as mainly working on their own household plots. By contrast, the proportion of male agricultural labourers has remained broadly stable at around 23 per cent. The proportion of 'other workers' too have increased commensurately. For women workers in West Bengal, the share of both cultivators and agricultural labourers has declined. In this case, the importance of 'household industry' has gone up considerably, probably as part of a survival strategy of working households, given the generally weak employment situation. While the Census showed a substantial increase in marginal work for both men and women in West Bengal, the results of the NSS Survey point to a rather different conclusion. The NSS defines 'usual status' activity as one where the person is engaged in an activity for a long period of time in the preceding year. Within this, 'usual principal status' refers to the main activity while 'usual subsidiary status' refers to those who would be characterized as 'unemployed' according to the usual status definition, but pursued some activity in a 'subsidiary' capacity. It is evident that this corresponds approximately (but of course not exactly) to the Census definitions of main and marginal work. According to the NSS results, which are presented in Charts 5 and 6, the ratio of subsidiary status workers to usual principal status workers, has been going down for quite a long time in West Bengal, and especially during the period 1993–94 to 1999–2000. Such declines are evident for men and women, and for rural and urban areas. For male workers the decline was continuous. For women workers, while the ratio increased in the decade 1983–1993–94, it went down substantially thereafter. In addition to 'usual status' activity, the NSS has other concepts of employment-the 'weekly status' employment that counts a person as employed if s/he has been engaged in any economic activity for at least one hour on any day of the week preceding the survey and 'daily status' employment that defines a person as employed if s/he has worked for 4 hours or more during the day preceding the survey. The 'usual status' definition may be said to provide a stock estimate of the work available, since it deals with the existence of work over a fairly long period of time. By contrast, the 'weekly' and 'daily' status measures can be seen as flow estimates that also give a sense of the current state of labour demand. Charts 7 and 8 indicate the growth of rural employment according to the NSS, both by the 'usual status' and 'daily status' indicators. In terms of usual activity, while male employment shows a sharp decline, that for women has actually turned negative during the later period. In terms of daily status, which possibly captures the real state of the labour market from the point of view of workers more accurately, both men and women experienced a decline in employment, after reasonable growth in the earlier period. Charts 9 and 10 show the NSS results for the same variables in the urban sector. Here the condition of male employment appears to be slightly better, in that, while there was deceleration in employment by both usual and daily status indicators, it was relatively less and male employment growth continued to increase above the rate of increase in labour force. However, female employment shows a collapse, with absolute declines according to both indicators. The change in type of employment must be seen in relation to this aggregate employment stagnation, since it defines the broader conditions under which labour contracts change. In what follows, we consider patterns specifically in the rural areas of West Bengal. Chart 11 shows the changing terms of employment for rural men. It is clear that the most significant changes are decline in regular employment and increase in casual work. While these are part of long-term trends, the recent period appears to have seen an acceleration in these changes. The change in self-employment is more complex, since the share of self-employment increased in 1993–94, only to come down to almost the earlier levels by 1999–2000. For rural women, as shown in Chart 12, the pattern is rather different. One has to keep in mind that the work participation rate of women in West Bengal is already very low, and that the latest period witnessed an absolute decline in employment. Therefore, the increase in share of a particular type of employment does not mean an absolute increase in such work; rather it indicates that such work declined less than other types of work contracts. This may be why an increase in the share of regular work and a decline in the share of casual work can be observed for rural women. It is likely that most of the job losses were in the sector of casual work, which is why the share has gone down. The increase in the share of self-employment is likely to reflect disguised unemployment in such a context, since this can be a residual or 'refuge' activity when other paid work is not available. It remains to consider the extent of diversification of employment, given the overall stagnation of employment and the changes in types of work contract. Chart 13 describes the share of non-agricultural employment in total rural employment (usual status, principal and subsidiary activities taken together). The trend is clearly for an increase in the share of non-agriculture in total employment, although this is more prominent for women workers. By 1999–2000, nearly half of the women engaged in some usual status activity in rural West Bengal were not employed in agriculture. Once again, given the overall decline in employment, this suggests that the job losses for women have been greater in agriculture, rather than a significant expansion in non-agricultural jobs. For men, the increase in jobs in the non-agriculture sector was quite sharp till 1993–94, but seems to have reduced subsequently. Table 1 provides a picture of the overall extent of employment diversification from agriculture by districts, using data from the 2001 Census. It needs to be borne in mind that these shares of non-agricultural employment (which refer to both main and marginal workers taken together) include other primary activities (such as forestry and fishing) and also refer to both rural and urban areas. This is why the shares of non-agriculture in total employment are very high, especially for some heavily urbanized districts. While there is almost complete absence of agricultural work in Kolkata, the share of non-agriculture is also very high is Howrah, which is dominantly urban, part of greater Kolkata, and in North 24 Parganas, which is also part of the greater Kolkata hinterland, as well as in Darjeeling, which is dominated by the city of Siliguri. What is interesting, however, is that very few districts show a domination of agriculture in total employment. Indeed, for male workers, only four districts show a share of non-agriculture in total employment of less than 40 per cent. They are among the districts that are generally considered 'backward': Uttar and Dakshin Dinajpur (with around 32 per cent each), Bankura (36 per cent) and Birbhum (39 per cent). For the state as a whole, well above half of those employed are engaged in non-agricultural pursuits. This is also well above the national average of more than 60 per cent of workers engaged in agriculture, indicating that in West Bengal, employment diversification is relatively more advanced. What are the areas in which such diversification has occurred and which of these has been more dynamic? It is worth considering this in detail by looking at the relationship between growth and employment. It is notable that in West Bengal, the period 1993–94 to 1999–2000, which experienced such stagnation and decline in aggregate employment generation, actually showed a higher rate of growth of economic activity. State Domestic Product in constant price terms actually increased at an annual rate of 6.7 per cent, compared to 5.8 per cent in the period 1983–1993–94. However, aggregate employment growth in the later period in West Bengal according to the NSS was only 0.76 per cent, compared to 2.44 per cent in the earlier period. These data reflect employment elasticities in the recent period which are substantially lower for West Bengal than for the rest of India. This suggests that the problem was not growth per se so much as the pattern of growth, which became progressively less employment-intensive. This was especially marked in manufacturing and services (other than transport and trade), in both of which employment actually declined, and was despite the fact that in both sectors, output growth was impressive, even double than the rates achieved earlier. The pattern of services employment is surprising, given that the services sector, particularly self-employment, is typically a residual sector that can become a refuge for those unable to find productive employment elsewhere. However, in West Bengal the freeze on new state government employment after 1991 may have played a role in this. This is also reflected in the decline in regular employment in urban areas that was indicated by the NSS data. While agriculture has been an important source of employment generation in the past, the rate of growth in jobs in this sector has been declining in the recent period. Of course, the situation in West Bengal is better than in the rest of India, where agricultural employment has actually fallen in the period 1993–94 to 1999–2000. This is probably because of boro rice cultivation, which has meant increased labour demand because of more intensive cropping in many parts of the state. The few sectors that still show some dynamism in terms of employment generation are wholesale and retail trade, transport and storage, and construction. These are the sectors for which employment elasticity has not declined during the period of 1999–2000. However, the dynamism in these sectors is more prominent for urban areas; in rural areas the impact on employment appears to be less. Obviously, the lack of adequate employment generation is one of the most pressing economic and social problems in West Bengal at the moment, much as in the rest of India. But the problem seems to be even more acute in this state than elsewhere, given the sharper declines in employment expansion that have been noted. Clearly, this has got to be a critical focus of future policy. This can be achieved by moving away from those sectors that exhibit low employment elasticity, or by changing the pattern of growth in these sectors. Within agriculture, diversification of cropping pattern as well as the development of improved storage and distribution facilities enhancing cultivators' ability to grow new crops, may lead to more employment generation. In addition, the collapse in formal finance access for farmers, as evidenced for example in the sharp decline in rural credit–deposit ratios of banks, must be reversed. However, simply encouraging the shift of activities away from employment towards manufacturing and services may not be adequate to cause employment expansion, given the current pattern of growth in these sectors. What is needed is a change in the nature of such growth, which would make these activities also more labour-using than they are at present. Within manufacturing, the basic problem seems to be that much of the growth has come from large units, which are less employment-intensive, even as many more labour-using small units close down. The pressure of import competition and worsening credit access are, at one level, national problems that are common across the states, and cannot be dealt with through state government policy. But it is still possible to think of methods of encouraging and safeguarding the interests of small and more employment-intensive producers through various state-level incentives. In services, much of the employment decline has come about because of the ban on fresh hiring by the state government that has affected public employment very drastically. The point is that such declines have not been compensated for by increases in private service employment. While services cannot be treated as one sector for policy purposes as it is such a heterogeneous, catch-all category, it is true that there is scope for expansion of a range of new service activities, especially IT-enabled services which can be provided by educated workers. The expansion of IT-enabled service activities is a process that has entailed a drop in the proportion of educated unemployed in urban India in general, but there is plenty of scope to develop such activities (which range from call centres and back-office work to accountancy and other activities) in urban areas of West Bengal. For this, of course, better infrastructural facilities (especially power, communication and connectivity) as well as more systematic and directed training of educated youth are required. On the whole, it appears that there is scope for redirecting economic processes within the state of West Bengal to encourage more employment generation, notwithstanding the overall macroeconomic processes that are currently operating to inhibit the growth of jobs.



© MACROSCAN 2003