MacroScan (Printable Version) - Migrating for Work

One of the important features of labour markets and working conditions of workers in India that has always been inadequately captured by our statistical system is economic migration. The Census data collection exercise is concerned only with current residence and permanent migration. It does not even attempt to capture short-term of seasonal flows of people, and - because of its rather strict definition of permanent migration - it even tends to leave out fairly prolonged periods of migration.

The National Sample Survey Organisation, until recently, had also tended to ignore short-term migration. In the 55th Round survey conducted in 1999-2000, some questions were asked about those who worked away from their normal residence for more than three months, and this did provide some hints of evidence about short-term migration for work. But micro studies indicated that much of such movement is even more short-term and often seasonal in character, and this was simply not captured. The next large survey of 2004-05, the 61st Round, did not address the issue of migration at all.

That is why the results of the 64th Round survey conducted in 2007-08, with a special focus on migration, were so eagerly awaited. Although it was a “small” survey (the most recent large NSS survey was conducted over 2009-10), it was nevertheless large enough to give us some useful indicators about some national and state-level trends with respect to this feature. One again, migration for work is only a subset of the questions that were addressed, but the greater depth of the questions and the revision of the time period of movement with respect to which questions were asked do provide some more data.

The basic trends in permanent or longer term migration that are indicated from that survey are shown in Chart 1. There is a significant increase in migration rates for females, but for urban males the rates are stagnant and those for rural males have even declined somewhat.

It is well known that the dominant proportion of female migration in India is for purposes of marriage, since most marriage patterns in the country are based on virilocal residence. More than 91 per cent of rural female migrants and 61 per cent of urban female migrants had moved because of marriage. This also explains why most of the migration (more than 90 per cent) in India is permanent in nature. Another 30 per cent of urban female migration was accounted for by the need to move because the head of household or main earning member had moved.

Since female migration rates remain so strongly determined by marriage, it is likely the male movement will better capture changing trends. Therefore Charts 2 and 3 provide the evidence on changing migration rates of rural and urban males, respectively, by broad social category.

This is where the story becomes interesting. The latest NSS survey actually shows a significant decline in rural male migration rates (Chart 2), on average by as much as 28 per cent compared to the previous survey. The drop is evident across all social categories, though it is largest for the SC and OBC categories. This is possibly the first time that such declines have been shown in the aggregate survey data, and if this is a correct representation of reality it is also likely to have implications for many other trends, such as urbanisation and changing demographic structures across different regions. Rural to urban male migration increased by about 5 percentage points compared to the previous survey, but it still accounted for less than 40 per cent of total male migration.

It is also worth noting that such declines in migration rates are not evident for urban males (Chart 3). In fact, while migration rates for urban males of the ST and OBC categories decreased to some extent, they increased for SC and general categories, such that the average rate increased slightly. One quarter of all male migration was from one urban centre to another.

Similar surprises are evident in terms of the causes of migration, as shown in Chart 4. Work-related reasons (dominated by the search of new or better employment) actually accounted for falling proportions of the moves made by rural residents, both male and female, even as they continued to increase for urban males.

This is without question a noteworthy development. What explains this trend? This issue needs to be explored in much more detail, and with more investigation into regional characteristics. But it is possible that the introduction of the NREGA has played some positive role in preventing the extreme distress migration that was observed to characterise many rural parts of the country.

Of course, 2007-08 was still an early year for the programme, which had not yet been extended to all the rural districts. We may have to wait for the latest NSS survey to provide some more information on this. The impact on rural employment would be felt not only in terms of the direct employment effects but also through the multiplier effects of the incomes earned on what has been a very depressed rural economy.

And of course, it is also possible that in addition to positive factors, adverse features - such as the fact that our urban areas have become less welcoming or even tolerant of rural migrants - may also have been significant. But certainly this survey shows that even short-term migration, of between one to four months, was relatively low and had probably decreased since the previous survey.

On thing is clear though: once they move, migrants are more likely to look for jobs and to find them (Chart 5). This is true of all categories of migrants, and holds true whether they move for work-related reasons or for any other reason. However, the survey also shows that a growing proportion of the work that migrants find is in the form of self-employment, indicating the continuing difficulty in our growing economy, of finding paid employment.

Construction remains by far the dominant employer for rural migrants (Chart 6) though agriculture accounts for the activity of nearly a quarter. Manufacturing accounts for only about 17 per cent of rural migrants’ work, a declining proportion compared to the previous survey.

For migrants from urban areas, construction and manufacturing account for around equal shares of the activity (Chart 7) - amounting to just over half the work found. Trade, hotels and restaurants are the other large employers of migrants from urban areas.

There is obviously a great deal more that is revealed by this latest migration survey of the NSS, including the type of work, the remittances sent, the details of the short-term employment, and so on. But the early results already suggest that there are some important changes in the pattern of movement for work, especially for rural residents. We need to consider more systematically how much this is related to patterns of public intervention, and also how public policy should address these changes.



Themes > Features	28.12.2010


Migrating for Work

C.P. Chandrasekhar and Jayati Ghosh

One of the important features of labour markets and working conditions of workers in India that has always been inadequately captured by our statistical system is economic migration. The Census data collection exercise is concerned only with current residence and permanent migration. It does not even attempt to capture short-term of seasonal flows of people, and - because of its rather strict definition of permanent migration - it even tends to leave out fairly prolonged periods of migration. The National Sample Survey Organisation, until recently, had also tended to ignore short-term migration. In the 55th Round survey conducted in 1999-2000, some questions were asked about those who worked away from their normal residence for more than three months, and this did provide some hints of evidence about short-term migration for work. But micro studies indicated that much of such movement is even more short-term and often seasonal in character, and this was simply not captured. The next large survey of 2004-05, the 61st Round, did not address the issue of migration at all. That is why the results of the 64th Round survey conducted in 2007-08, with a special focus on migration, were so eagerly awaited. Although it was a “small” survey (the most recent large NSS survey was conducted over 2009-10), it was nevertheless large enough to give us some useful indicators about some national and state-level trends with respect to this feature. One again, migration for work is only a subset of the questions that were addressed, but the greater depth of the questions and the revision of the time period of movement with respect to which questions were asked do provide some more data. The basic trends in permanent or longer term migration that are indicated from that survey are shown in Chart 1. There is a significant increase in migration rates for females, but for urban males the rates are stagnant and those for rural males have even declined somewhat. It is well known that the dominant proportion of female migration in India is for purposes of marriage, since most marriage patterns in the country are based on virilocal residence. More than 91 per cent of rural female migrants and 61 per cent of urban female migrants had moved because of marriage. This also explains why most of the migration (more than 90 per cent) in India is permanent in nature. Another 30 per cent of urban female migration was accounted for by the need to move because the head of household or main earning member had moved. Since female migration rates remain so strongly determined by marriage, it is likely the male movement will better capture changing trends. Therefore Charts 2 and 3 provide the evidence on changing migration rates of rural and urban males, respectively, by broad social category. This is where the story becomes interesting. The latest NSS survey actually shows a significant decline in rural male migration rates (Chart 2), on average by as much as 28 per cent compared to the previous survey. The drop is evident across all social categories, though it is largest for the SC and OBC categories. This is possibly the first time that such declines have been shown in the aggregate survey data, and if this is a correct representation of reality it is also likely to have implications for many other trends, such as urbanisation and changing demographic structures across different regions. Rural to urban male migration increased by about 5 percentage points compared to the previous survey, but it still accounted for less than 40 per cent of total male migration. It is also worth noting that such declines in migration rates are not evident for urban males (Chart 3). In fact, while migration rates for urban males of the ST and OBC categories decreased to some extent, they increased for SC and general categories, such that the average rate increased slightly. One quarter of all male migration was from one urban centre to another. Similar surprises are evident in terms of the causes of migration, as shown in Chart 4. Work-related reasons (dominated by the search of new or better employment) actually accounted for falling proportions of the moves made by rural residents, both male and female, even as they continued to increase for urban males. This is without question a noteworthy development. What explains this trend? This issue needs to be explored in much more detail, and with more investigation into regional characteristics. But it is possible that the introduction of the NREGA has played some positive role in preventing the extreme distress migration that was observed to characterise many rural parts of the country. Of course, 2007-08 was still an early year for the programme, which had not yet been extended to all the rural districts. We may have to wait for the latest NSS survey to provide some more information on this. The impact on rural employment would be felt not only in terms of the direct employment effects but also through the multiplier effects of the incomes earned on what has been a very depressed rural economy. And of course, it is also possible that in addition to positive factors, adverse features - such as the fact that our urban areas have become less welcoming or even tolerant of rural migrants - may also have been significant. But certainly this survey shows that even short-term migration, of between one to four months, was relatively low and had probably decreased since the previous survey. On thing is clear though: once they move, migrants are more likely to look for jobs and to find them (Chart 5). This is true of all categories of migrants, and holds true whether they move for work-related reasons or for any other reason. However, the survey also shows that a growing proportion of the work that migrants find is in the form of self-employment, indicating the continuing difficulty in our growing economy, of finding paid employment. Construction remains by far the dominant employer for rural migrants (Chart 6) though agriculture accounts for the activity of nearly a quarter. Manufacturing accounts for only about 17 per cent of rural migrants’ work, a declining proportion compared to the previous survey. For migrants from urban areas, construction and manufacturing account for around equal shares of the activity (Chart 7) - amounting to just over half the work found. Trade, hotels and restaurants are the other large employers of migrants from urban areas. There is obviously a great deal more that is revealed by this latest migration survey of the NSS, including the type of work, the remittances sent, the details of the short-term employment, and so on. But the early results already suggest that there are some important changes in the pattern of movement for work, especially for rural residents. We need to consider more systematically how much this is related to patterns of public intervention, and also how public policy should address these changes.



© MACROSCAN 2010