Research

Rising competition for prestigious white-collar jobs and higher education can incentivize elites to behave as rivals rather than sources of support. This paper asks whether elite peers behave as rivals and whom this rivalry targets. Using novel data on high school and university graduates across five provinces in colonial India between 1894 and 1919, I study how exposure to elite upper-caste peers affected subsequent educational and occupational mobility. Colonial India provides a useful setting since caste groups occupied distinct social ranks, while expanding access to higher education and government employment increased competition for opportunities historically monopolized by elites. Exploiting plausibly random variation in the share of elite peers across cohorts within the same school, I find that greater elite exposure reduced the probability that the intermediate-ranked merchant castes progressed through university or entered the legal profession. They, however, had little impact on castes ranked above or below the merchant castes. The negative effects are driven by educated elites -- Brahmins and Kayasthas -- who historically dominated higher education and government employment, rather than by landed elites. Additionally, similarly represented but higher-ranked commercial castes were unaffected. These findings are consistent with elite rivalry being domain-specific -- driven by educated elites whose status was tied to higher education -- and concentrated among the lower-status group posing the greatest competitive threat.

This paper studies the effect of trade protection and women labour on the Japanese cotton spinning industry in the early twentieth century. A decline in British exports of cotton textiles to Asian markets provided a temporary protection, creating an opportunity to export for the Japanese cotton weaving industry and resulting in increased demand for domestically produced cotton yarn. In the first part of the paper, I use WWI as a quasi-natural experiment, and using a differences-in-difference design, I study the effect of the trade shock, using distances to the nearest port as a measure for exposure to international trade, on factory size and labour productivity. At the extensive margin, the number of factories increased in prefectures closer to ports. However, at the intensive margin, factories located further away from ports increased in size and made more efficient use of their spinning capacity. In the second part of the paper, I provide suggestive evidence as to whether factories with a higher proportion of women workers increased labour productivity during and after the war. I find that factories that had a higher proportion of women workers in 1913, increased their labour productivity during the war. These effects were a result of factories being able to increase the number of spindles that each worker operated, rather than due to an increase in the intensity of work. On the other hand, there was a negative effect on the total number of workers employed in these factories during and after the war. These results suggest that factories were able to undertake more productivity-enhancing measures by increasing the number of machines they operated, without increasing the number of workers to increase production.

We analyze the long-run political economy effects of temporary trade protection from British imports during WWI in Colonial India. In the first part of our paper, we study the effects of the WWI trade shock on firm-level profits and the formation of political connections with the pro-independence party, the Indian National Congress (INC), which increasingly advocated for the boycott of British goods under the leadership of Gandhi. We measure political connections as the share of directors of the firm that corresponded with Gandhi. For this purpose, we make use of a novel data source: the complete collection of letters sent by Gandhi. We construct the firm-level WWI trade shock by leveraging both the district-level variation in trade costs and the industry-wide variation in the drop of British imports. Using firm-level panel data constructed from the Investors India Yearbook, we find that more protected firms (i.e., experienced greater trade shock) during WWI generated more profits during and just after the war, and became more politically connected with the INC. In the second part of the paper, we find that more politically connected firms generated more profits, but only after independence. Using a two-stage least squares approach, we show that firms that became more politically connected after WWI also generated more profits after independence. Furthermore, they saw no significant increase in profits during the WWII. Overall, our results suggest that firms that benefited from temporary protections  turned into lobbyists, demanding more protection in the long run.

Abstract: Mayshar et al. [2022] argue that dependence on appropriable cereals, rather than increased land productivity driven by the agricultural revolution, eased the emergence of tax-levying elites and eventually, states. This conjecture is consistent with correlations between measures of the emergence of hierarchy and settlement and proxies for land productivity and the productivity advantage of cereals over tubers, both based on present-day agro-climatic factors, i.e., radiation and temperature. Since, however, global warming has dramatically altered climate conditions, a test based on coeval data is needed to credibly evaluate the appropriability theory. To devise such a test, we construct historical potential caloric yield indexes combining simulated paleoclimatic data with coefficients from present-day crop yield response functions. Our analysis yields four main findings. First, Bronze Age barley yields reconstructed from the available administrative cuneiform texts are strongly positively related to our coeval index for potential caloric yield of barley and strongly negatively correlated with a measure of the present-day potential caloric yield of barley. These correlations are consistent with a climate change-driven reversal of fortune of the Mesopotamian polities that we consider. Second, the degree of appropriability is only weakly related to hierarchy in the cross-sectional analyses. Third, state-building is strongly linked to lower land suitability and adverse climate conditions, particularly when multicollinearity among independent variables based on agro-climatic factors is properly addressed. These findings are consonant with a time-inconsistency theory of state-building. Finally, the regressions with the diffusion of settlement point as the dependent variable are in the same direction of those studying state-building, although the estimates are significantly noisier.