Understanding the migrants

Availability of credible data is critical to integrating these workers in development programmes

Much is being written on the plight of migrant labour. However, little is known about their magnitude. The Indian Constitution guarantees freedom of movement for all citizens. The foundational principles of free migration are enshrined in clauses (d) and (e) of Article 19(1) of the Constitution, which guarantee all citizens the right to move freely — and reside and settle in any part — of the territory of India.

As movement is a dynamic process, it is extremely difficult for any statistical system to capture population mobility. The best possible way to capture such data is to approach the household to which the individual belongs. Unfortunately, India does not have a single definition of migrants. The two major agencies — the Census and the National Sample Survey Office (NSSO) — that collect information on migration use different criteria, though both are based on the change in place of residence, to define migrants.

According to the Census, an individual is classified as a migrant if he has changed his place of residence in the past from one village/town to another. It also has a place of birth classification. Neither a change in place of residence within a village/town nor a temporary change in the place of residence is considered migration. 

On the other hand, the NSSO defines migration on the basis of last usual place of residence, which, unlike the Census, is defined as a place where one has stayed continuously for a period of six months or more. If the present place of residence of an individual (i.e. the village or town where the person is being enumerated) is different from his last usual place of residence, then he is classified as a migrant.

According to the 2011 Census, 37.6 per cent of the population were migrants, compared to 30.6 per cent in 2001 by place of last residence, whereas these percentages were 36.5 and 30, respectively, by place of birth. Of the 450 million last-residence migrants, only 11.9 per cent (or about 54.3 million) were inter-state migrants. Based on the NSSO definition, 28 per cent of India’s population in 2007-08 were classified as migrants. 

The availability of credible data is critical to developing a robust understanding of migration, including factors such as the reasons why people migrate, their origins and destinations, sectors of employment, and living conditions. In India, it is also crucial to understand the nature of short-term and circular migration (where the migrant does not move permanently from the place of origin to the destination), since such migration forms a significant share of population movements. The statistical agencies’ focus on this issue has been limited.

There are a number of reasons for seeking better data on internal migration, in addition to their effect on the cities. These include the linkages between the Mahatma Gandhi National Rural Employment Guarantee Scheme and migration, enhancing financial inclusion, and the nature of domestic remittances. The specific nature of migration in India, with migrants seeking to retain relationships at both their place of origin and destination, needs to be better understood in order to anticipate the nature and growth in urbanisation.

There are four broad sources of data on migration: Numerous specific surveys conducted by researchers with varying methodologies and on different scales; the Census of India, conducted by the Registrar General of India (RGI); periodic surveys undertaken by the NSSO; and administrative data at various levels, from local governments to the Union government, generated as part of administrative record-keeping. 

The specific surveys collect disaggregated information on caste/tribe, gender, age, education, migration duration, occupation, seasonality and wages. They use various definitions of migration based on different time definitions, and use different methods of data collection, e.g., interviewing the individual vis-à-vis the household, using different sampling methods. These micro studies contain detailed information on the vulnerability of different groups to social exclusion and discrimination in the labour market, specific to certain groupings which are not captured in larger surveys, but have significant local relevance for labour market outcomes.

Migrants gather outside Dharavi slums to take buses to reach a railway station and board a special train to their native places. PTI

The RGI recently released most of the data on migration based on the 2011 Census. The unit of data collection is currently the district. However, within districts, there are specific sub-districts that may be more migration-intensive than the average for the district, since there are strong local externality effects in migration. Currently, the data available from the RGI is only for in-migration.

The latest migration survey by the NSSO — the 64th round — was conducted in 2007-08, as part of the employment-unemployment schedule. There is some migration information in other NSSO surveys also. There have been six rounds since 1999-2000 in which some questions pertaining to migration have been asked. Migration has been usually investigated as a part of the employment-unemployment survey. In the 2007-08 survey, the focus was mainly on the migrant workforce.

However, broad issues regarding reasons for migration, spatiality (where migrants come from and where they go), nature of movement (permanent, semi-permanent or temporary) and some general particulars about last usual place of residence were also collected. The 2007-08 survey was also the first when questions about out-migration were asked, such as places and reasons for out-migration and remittances received from out-migrants. It also enlisted their economic activities.

Governments at various levels also generate data pertaining to migration. This can be at the panchayat level (recording of information about migrating children), or in providing services to migrants (such as the district-level facilitation cells in Odisha), or it can be at the central level, where some schemes of the Union government now generate data that is useful for understanding migration flows.

In conclusion, the two data sources — NSSO surveys and the Census — define migration differently. The NSSO treats a person as a migrant if he/she has been residing at the place of enumeration for less than six months. In the Census, a person is treated as a migrant if her/his place of enumeration is different from the place of birth (POB) and place of last residence (POLR). Detailed information from the Census is based on POLR and only one table is generated based on POB, to provide lifetime migration data. The Census is conducted once in 10 years and is due in 2021, but may be delayed by at least a year owing to the Covid-19 pandemic.

The writer is former director general, NSSO