New Researcher Prize: “Why sell oneself into (temporary) slavery?”

Free labour dominates our thinking about markets today, yet unfree labour has been and continues to be an important part of the economy.  For example, large numbers of workers from the Indian subcontinent and southeast Asia work in the Middle East under the kafala system.  West Germany recruited Gastarbeiter for decades in the mid-20th century.  Although not widely publicized, the US maintains guest worker programs that tie individuals to a particular employer for a set period and wage.

The existence of these and other contractual forms prompts the question, why do workers voluntarily forego their options and become temporary slaves?  In order to answer this, I turn to an historical example of bonded-labour contracts and international migration:  Indian indentureship in the nineteenth and early twentieth centuries.  I turn to new, unique, individual-level data collected and collated by me on roughly 250,000 Indian indentured servants sent around the world.

A key tenet of the indentureship contract is a guaranteed, fixed wage.  This in turn points to a key driver of unfree labour contracts:  uncertainty about the future and fear of bad economic outcomes may make temporary slavery a viable strategy.  The certainty from the contract outweighs the losses from lack of mobility.

This research traces Indians back to their home districts and calculate measures of volatility (to proxy for uncertainty) at the time of departure. It then models out-migration as a result of push factors (low wages, high prices, and volatility) vs. the pull factors from the guaranteed overseas wage.

The research finds that migration is consistent escaping volatility.  Districts in India with higher volatility were more likely to send immigrants and sent higher numbers than low-volatility districts.  The results were not uniform, though.  Castes heavily involved in owning and investing in land responded more than subsistence-only castes.  This reflects the different time horizons across castes.  Finally, Indians who left from more volatile areas were less likely to return to India.  Thus, not only did volatility affect Indians at the time of departure but also years later at the potential time of return.

Overall, volatility matters.  It caused Indians to sell themselves into temporary slavery and may have important other effects for people near subsistence who are unable to protect themselves against major economic shocks or even minor fluctuations.  More generally, this research highlights how uncertainty about economic conditions, not merely average wage differentials between markets, affects migration and labour choice.