Mobility and tenure choice in urban India

The story of migration is the story of aspiration. Molloy et. al. (2011) [12] open their well-cited paper on US migration with examples: the Okies, suffering from the dust bowl, moving to California; African-Americans in the rural south migrating for manufacturing jobs in the cities of the United States north; etc.

But migration (and for this paper, we are discussing internal migration) is not just an American Story: all over the world, migrants have moved from country to city, and from economically troubled cities to economically vibrant ones. Beginning with the 1980s, the economic powerhouse of London has drawn people especially young people from all over the UK. Honshu, with the economic centers of Tokyo, Osaka, and Nagoya, have drawn people from the rest of Japan. China has urbanised rapidly, as hundreds of millions have quickly transformed it from an overwhelmingly rural country to a majority urban one, presumably in large part because the economic opportunities in Chinese cities are so much greater than in the countryside.

The migration story has not always worked out so well, however. Latin America and Africa are also urbanizing rapidly: Brazil, Peru, Libya, Gabon are now more than 75 percent urban, and Lagos, Kinshasa, etc. are among the fastest-growing cities in the world. And yet it is not clear that in these instances migration has produced large benefits, as living conditions in peri-urban areas have many of the deficiencies of rural areas: absences of electricity, sanitation, and treated water. As Henderson (2003 [7], 2010 [8]) has shown, while advances in economic development are almost always accompanied by urbanization, the converse is not necessarily true. The implications of migration particularly rural to urban vary by country, and are therefore worth studying on a country by country basis.

This paper examines the things we know and don’t know about migration for the world’s second-largest country: India. In studying India, we will begin by using the template that Molloy et. al. (2011) [12] used to investigate migration in the US: we will discuss the data choices involved in studying migration, and the strengths and weaknesses underlying these choices, present some basic facts about migration in India, discuss the robustness of these facts, and look at differences in migration characteristics across different types of people.

We will then go one step further, and investigate the interaction of migration and tenure choice (i.e., the choice between owning and renting) in India. In the United States, there has long been a strong association between the propensity to migrate and tenure choice those who have identified themselves as sticky with respect to location have a tendency toward home-owning, while those who have been footloose have had a greater tendency to rent.

One manifestation of this is the relationship in the US between marriage and tenure: using a simple linear probability model of homeownership in the US, one finds that, after a long list of controls, married couples are 22 percent more likely to be owners than single people. One fact about India is that the rental sector of the housing market has been rapidly shrinking over the past 50 years (Tandel et. al., 2016 [15]). As we think about mobility in India, we may consider the impact that the shrinking rental sector has had on the opportunities for Indians to migrate. We model the impact of migration on tenure choice in India and test whether the relationship between the two characteristics has remained stable across time.

A number of findings surprise us. The first is that Indians do not migrate very much, both locally and across states. The second is that the variables that we would expect to predict migration do not seem to do so, until we run separate regressions by sex. Finally, the relationship between migration and tenure choice in India is much weaker than we expected, based on literature exploring the relationship in other countries.

This paper is organized as follows: Section 2 summarizes the relevant literature, section 3 provides a discussion of the available data, section 4 presents some stylized facts on migration in India, section 5 has a discussion of tenure choice (theory and the Indian context), section 6 presents some regression results, and section 7 ends the paper with concluding remarks.

DOWNLOAD THE WORKING PAPER