Much of our knowledge of change in rural areas depends on longitudinal village studies. Drawing upon a number of village studies carried out over the years in India, this column provides a broad picture of how the economic and social structures of villages are changing, and the consequences for production, employment, migration, inequality and other key issues.
2016 is the centenary of Gilbert Slater’s famous village studies, which he launched from Madras University soon after his appointment as Professor of Economics there. While there were earlier village studies, his were pioneering in their breadth and content, and in their contribution to survey methods. They provided a baseline for several later revisits to his villages, and have inspired many successors. In fact, there are now quite a few long-term studies of villages in different parts of India, and elsewhere too, though longitudinal village research (that follows the same village or set of villages over time) seems to be something of a South Asian specialisation. Much of our knowledge of rural change depends on these studies.
Village studies come in and out of fashion, but there is a community of researchers both in India and elsewhere who have invested much time and effort in this type of research. A number of those who had been working in this area met in Jawaharlal Nehru University in 2012, and have put together a book, which illustrates the longitudinal method and its applications: Himanshu, Praveen Jha and Gerry Rodgers (eds.) (2016), The changing village in India. Insights from longitudinal research, Oxford University Press, New Delhi.
The 24 contributors to the book report on research in different parts of the country. They provide a broad picture of change in a number of villages and present findings on a variety of specific issues, ranging from public services to migration, from land ownership to female work. Some of the contributions concern a single village, others several villages. One looks at how a small town fits into the picture. Another looks at what can be learned from the repeated studies of the ‘Slater villages’. One reviews village studies in Bangladesh. Several articles give results from one of the best known longitudinal studies in India, concerning the village of Palanpur in Uttar Pradesh (UP), which has now been studied by successive teams of researchers for almost 60 years. Others draw on a long-term project that has been tracking change in villages in Bihar since the 1970s. And there are studies in the book from other villages in UP and Bihar, from Punjab and Haryana, from Tamil Nadu and from Maharashtra.
One important conclusion is that there is no common story across India. Villages are diverse, and are changing in diverse ways. And these studies only cover a few villages among 500,000-600,000 villages in India. So one should be careful not to generalise. But some issues come up time and again in different contexts. When you find echoes of the same sets of relationships in different places, perceptions and patterns that are repeated, it is possible to build a qualitative understanding of change which takes us beyond the study of one or a handful of villages.
The process of transformation
Longitudinal village studies are almost essential to understand changing agrarian structures. For example, the next two figures illustrate changes in class structures and relations over time in Bihar and UP. Based on surveys in 12 villages in Bihar over a period of almost 30 years, Figure 1 shows the decline in the numbers and influence of the class of landlords and a corresponding decline in attached (often bonded) labour.
Figure 1. Changes in class structures and relations in Bihar, 1981-2009
In a UP village, a different class classification shows a similar decline over two decades in the numbers of both dominant and dependent classes, with a corresponding rise of intermediate groups.
Figure 2. Changes in class structures and relations in UP, 1981-2009
These changes in class structure reflect how the village economy is itself changing. The village studies show a declining importance of agriculture in household income, almost everywhere. But there is diversity across villages (agriculture has held up quite well in some Tamil Nadu villages and in parts of Bihar). However, it is rarely the former landlords who invest in agriculture.
Non-agricultural income is now more important than agricultural income in most villages but it largely comes from outside the village, apart from some construction, trade and local services. Commuting has become important if towns are not too far; so has long-distance migration from more remote villages. This radically changes the village labour market and class relations.
In order to understand these processes, we need to study flow of labour, goods and capital into and out of villages. Local towns may be providing health and education services, and sucking in rural labour. But the connections between urban and rural areas are complex. A study of a small town in Tamil Nadu illustrates this; many of the flows from local rural areas bypass the town and go straight to Chennai or even Mumbai. The dynamics of India’s economy shift many of the relationships to the metropolis, so the links may not be local.
Alongside these changes in production relations, there are changes in the role of caste and political organisation.
There are huge variations in caste structures across Indian villages. But some patterns are quite widely repeated. For instance, several studies show how the same castes maintain their economic dominance over time through the capturing of new opportunities. These opportunities may be in nearby urban areas, seen for example, in the development of a new class of political entrepreneurs among upper castes in Haryana towns, or investment in the garment industry by the Gounders community from villages near Coimbatore.
There is also some sign that the greatest access to new opportunity is found at the bottom as well as at the top. For instance, scheduled castes are found to be moving out of agriculture in UP, Bihar and Tamil Nadu for they have access to casual labour markets outside the village.
This may imply less exploitative relations within the village, but not necessarily – there are contradictory findings in different places. And new forms of exploitation (for example, of migrants) may emerge. Still, the social gap may be reduced even while the economic gap remains.
There is a substantial political dimension to this change, which involves struggle and often overt conflict within villages. But an increasing external orientation in many places can reduce the importance of political struggle within the village, along with a weakening of the sense of collective identity. Collective action is very weak in Palanpur in UP and in Bihar villages, and this may be because the most important concerns for villagers are increasingly outside the village.
Migration is a key factor, which tends to empower the migrants (though there are counter examples). Migration, which in the past was often just a question of an income supplement or a response to distress may now be part of a longer-term economic strategy of multi-locational households. But there are large differences in the role of migration between villages even within the same state.
The nature and distribution of the benefits of change
What are the distributional consequences of these transformations? Again, the longitudinal method is extremely helpful. For example, with panel data1on individual households, you can look at their fortunes and what factors led to success or failure.
In Bihar, for instance, changes in income (in money terms) between 1998 and 2011 were largest for upper caste and upper Muslim households, and lowest for middle castes (Other Backward Castes), with Scheduled Castes in between. This pattern of largest gains at the top but also significant gains at the bottom, while the middle does less well, seems to be fairly widespread. The gains at the bottom often reflect increasing opportunities and wages for casual non-agricultural wage work.
Also in Bihar villages, Figure 3 below compares household income in 2011 with the income of the same household in 1999. Comparing local (within-village) household income in 2011 with income in 1999 by educational level of the household head (in blue), the proportional gain is largest for the least educated in 1999 (250% increase in income in money terms for those with primary education or less, against 150% for higher education; the price index for agricultural labourers rose by 88% in this period so these are substantial real increases). But when you add in the incomes of migrants to household income in 2011 and compare the result with household expenditure in 1999 (in red), the picture is reversed. Now there is a 280% increase at the bottom, but 330% at the top. In other words, the gains to education were realised by migrating. Or the gains from migration were dependent on education.
Figure 3. Education, migration and income (Bihar)
Longitudinal studies also help track changes in land ownership patterns, which vary greatly from one village to another. Landlessness grew in villages in Tamil Nadu and Maharashtra, but there was much less change in Bihar. In Palanpur, the Gini coefficient2of land distribution, both owned and cultivated, was extraordinarily stable over 50 years, in the range 0.45 to 0.5. But some village studies show inter-caste dynamics in land, especially where upper castes develop urban interests and middle castes acquire land. This was found in Tamil Nadu, Bihar and Maharashtra, for example.
There is clear and widespread evidence of widening gender inequality in the labour market and a feminisation of agriculture. In most villages, there has been diversification of employment for men, but much less or not at all for women. This is significant in the context of the debate at national level on declining female labour force participation. The village evidence suggests that this reflects lack of opportunities rather than a response to rising income.
There is also interesting evidence on how different castes and communities have different labour market strategies and opportunities outside agriculture. In Palanpur, for instance, Muslims are found more often in self-employment, upper castes in regular employment, and Scheduled Castes in casual work.
How does the overall pattern of income distribution change? Overall inequality in rural areas reflects both distribution within and distribution between villages. Villages are very different in their character and development path. Within villages, a very long-term study such as that in Palanpur finds fluctuations in inequality over time, with no clear long-term trend but a recent increase. Many factors are involved but it is frequent for income from outside the village to be less equally distributed than income within the village.
These are just some illustrations of findings and issues that can be addressed. And one has to resist the temptation to generalise. These relationships are observed in just a few of India’s more than 500,000 villages, and the different studies also use different concepts and methods. But these studies do make it possible to better understand the dynamics of change.
These studies also discuss the longitudinal method itself. Alongside their advantages such studies also raise a number of challenges.
First, long-term village studies pose the question - what is the village. Villages are increasingly part of a wider economic and social framework. That has always been the case, of course, and the village republic is something of a myth; but the question can increasingly be asked, how far is the village an appropriate unit of analysis? Villages are swallowed up by towns, or serve as residences for commuters. Both economy and society becomes less distinct. Many of the village studies document this process.
So not only is it more difficult to distinguish the object of the research, but the nature of that object is itself changing. The village remains a crucial reference point for the bulk of the population, and an identity persists across time. But it is important to capture not only change within the village but also change in connections between the village and the wider society.
Second, longitudinal research faces a particular problem of change not only in the village, but also in the researcher. There may be different researchers in successive studies, or the same researcher but whose perspective changes as he or she ages. This may have an impact on both theoretical framework and method. And in repeated studies over the long term, there is likely to be a changing relationship between the researcher and the village, for personal connections develop and it is difficult for the researcher to remain a neutral, external observer.
Third, there is a trade-off between depth and breadth. The greater the investment in a single village, the more accurate the information and the deeper the understanding. But such efforts involve considerable resources and efforts over long periods of time, which are not easy to sustain, and are geographically concentrated. Also there are significant benefits from studying several villages in order to take a comparative approach or to achieve a degree of representation. Each approach has its advantages and disadvantages, and in fact there is value added in applying diverse methods.
- Panel data is a dataset in which the behaviour of entities is observed over time.
- Gini coefficient is the most commonly used measure of inequality. The coefficient varies between 0 (reflecting complete equality) and 1 (reflecting complete inequality).
- Himanshu, P Jha and G Rodgers (eds.) (2016), The changing village in India. Insights from longitudinal research, Oxford University Press, New Delhi.