Famine, institutions, and indentured migration in colonial India

By Ashish Aggarwal (University of Warwick)

This blog is part of a series of New Researcher blogs.


Women fetching water in India in the late 19th century. Available at Wikimedia Commons.

A large share of the working population in developing countries is still engaged in agricultural activities. In India, for instance, over 40% of the employed population works in the agricultural sector and nearly three-quarters of the households depend on rural incomes (World Bank[1]). In addition, the agricultural sector in developing countries is plagued with low investments, forcing workers to rely on natural sources for irrigation as opposed to perennial man-made sources. Gadgil and Gadgil (2006) study the agricultural sector in India during 1951-2003 and find that despite a decline in share of agriculture in GDP in India, severe droughts still adversely impact GDP by 2-5%. In such a context, any unanticipated deviation from normal in rainfall is bound to have adverse effects on productivity and consequently, on incomes of these workers. In this paper, I study whether workers adopt migration as a coping strategy in response to income risks arising out of negative shocks to agriculture. And, if local institutions facilitate or hinder the use of this strategy. In a nutshell, the answers are yes and yes.

I study these questions in the context of indentured migration from colonial India to several British colonies. The abolition of slavery in the 1830s led to a demand for new sources of labour to work on plantations in the colonies. Starting with the “great experiment” in Mauritius (Carter, 1993), over a million Indians became indentured migrants with Mauritius, British Guyana, Natal, and Trinidad being the major destinations. The indentured migration from India was a system of voluntary migration, wherein passages were paid-for and migrants earned fixed wages and rations. The exact terms varied across different colonies, but generally the contracts were specified for a period of five years and after ten years of residency in the colony, a paid-for return passage was also available.

Using a unique dataset on annual district-level outflows of indentured migrants from colonial lndia to several British colonies in the period 1860-1912, I find that famines increased indentures. However, this effect varied according to the land-revenue collection system established by the British. Using the year the district was annexed by Britain to construct an instrument for the land revenue system (Banerjee and Iyer, 2005), I find that emigration responded less to famines in British districts where landlords collected revenue (as opposed to places where individual was responsible for revenue payments). I also find this to be the case in Princely States. However, the reasons for these results are markedly different. Qualitative evidence suggests that landlords were unlikely to grant remissions to their tenants; this increased tenant debt, preventing them from migrating. Interlinked transactions and a general fear of the landlords prevented the tenants from defaulting on their debts. Such coercion was not witnessed in areas where landlords were not the revenue collectors making it easier for people to migrate in times of distress. On the other hand, in Princely states, local rulers adopted liberal measures during famine years in order to help the population. These findings are robust to various placebo and robustness checks. The results are in line with Persaud (2019) who shows that people engaged in indentured migration to escape local price volatility.


[1] https://www.worldbank.org/en/news/feature/2012/05/17/india-agriculture-issues-priorities



Banerjee, Abhijit, and Lakshmi Iyer (2005): “History, Institutions, and Economic Performance: The Legacy of Colonial Land Tenure Systems in India”, American Economic Review, Vol. 95, No. 4, pp. 1190-1213.

Carter, Marina (1993): “The Transition from Slave to Indentured Labour in Mauritius”, Slavery and Abolition, 14:1, pp. 114-130.

Gadgil, Sulochana, and Siddhartha Gadgil (2006): “The Indian Monsoon, GDP and Agriculture”, Economic and Political Weekly, Vol. 41, No. 47, 4887-4895.

Persaud, Alexander (2019): “Escaping Local Risk by Entering Indentureship: Evidence from Nineteenth-Century Indian Migration”, Journal of Economic History, Vol. 79, No. 2, pp. 447-476.



Strangling Speculation: The Effects of the 1903 Viennese Futures Trading Ban

By Laura Wurm (Queen’s University Belfast)

This blog is part of a series of New Researcher blogs.


Farmland in Dalat, Vietnam. Available at Wikimedia Commons.


Ever since the emergence of futures markets and speculation, the effects of futures trading on spot price volatility have been subject to intense debate. While the populist discourse affirms the adverse and price-disturbing consequences of futures trading, the work of scholars stresses the risk allocation and information transmission function of futures towards spot markets, essential for pricing cash transactions. My research tests whether these volatility-lowering effects of futures trading towards the cash market hold true by assuming the opposite: what happens if futures trading no longer exists?

To do so, I go back to the early 20th century, when futures trading in the Viennese grain market was, unlike at other trade locations at the time, such as Germany, England, or Texas, banned permanently. The 1903 parliament-enforced prohibition of futures trading was the consequence of an aversion against speculators, who were blamed for “never having held actual grain in their hands”. Putting an end to the vibrant futures market of the Agricultural Products Exchange, the city’s gathering place for farmers, millers, large-scale customers, and speculators, was thought to be the last resort to curb undue speculation. Up to the present day, futures trading has not been resumed. The uniqueness of this ban makes it an ideally suited natural experiment to test the effects of futures trading and its abolishment on spot price volatility. Prices from the Budapest Stock and Commodity Exchange, which was not affected by the ban, are used as a synthetic control. The Budapest market, as part of the Austro-Hungarian Empire, operated under similar legal-economic and geographic conditions, and was, in addition to Vienna, the only Austro-Hungarian market offering a trade in futures. This makes Budapest an ideally suited control.

My project examines the information transmission function of futures to spot markets and finds a heightened spot price volatility in Vienna and a lower accuracy in pricing cash transactions after futures trading was banned. The intra-day variation of spot prices increased after the ban. Without futures trading, the Viennese market lacked pricing accuracy and efficiency. The effect on volatility holds true when using a commodity traded exclusively on the Viennese spot market as a control. In addition, assessing Granger causality, information flows between the futures and spot markets of the two cities are found to have existed prior to the ban, which links to the information transmission function of futures towards cash markets and the close ties between the two markets. After futures trading was prohibited in Vienna, Budapest futures prices with 3-6 months maturity continued to significantly Granger-cause Viennese spot prices.



Laura Wurm