Protecting Capital: Economic Elites, Asset Portfolio Diversification, and the Politics of Distribution

Abstract

This dissertation explores why, how, and to what extent economic elites influence distributive outcomes. In answering these questions, it challenges standard political economy approaches built on the assumption that economic elites' interests can be traced to a single sector or asset. Building on the literatures on business and financial economics, the history of elites, and the latest contributions to the political economy literature, it proposes that the structure of economic elites' asset portfolio is crucial to explain their role in shaping distribution. This work thus develops a theory that explains how asset portfolio diversification shapes economic elites' preferred policies to protect their capital, how such preferences shape the political strategies they use to advance their interests and, ultimately their capacity to influence distributive outcomes. In the first part, it shows that diversified members of the economic elite tend to pursue policies that have a multiplier effect on the overall economy, whereas specialized elites prioritize policies that narrowly target their sector alone. Furthermore, to achieve their preferred policy outcome, the former type is likely to embed in the state structure by directly participating in the state administration and electoral politics, and the latter is more prone to lobby in favor of their interests from the outside, relying on sectoral business associations. In the second part, it demonstrates that diversification was a risk hedging strategy pursued by landed elites that faced a high risk of expropriation during the transition from a traditional to a modern economy. Where this process of early diversification took place, rates of development where higher both in the short and the long run. Furthermore, areas where landed elites remained specialized became economic laggards. The evidence to sustain these claims comes from a combination of historical narratives from Argentina and Chile and the use of multiple statistical techniques drawing on original individual- and subnational-level data from previously untapped historical archives that altogether spans between the mid-nineteenth and the mid-twentieth centuries.