These essays explore the debilitating role played by financial crises in economic performance, analyzed at the microeconomic level of the household; the impact of institutions on human capital accumulation, both at the microeconomic and macroeconomic levels; and the role played by human capital in economic performance. The broad policy implication of this work is that deep institutional reform may improve economic performance through a variety of channels.

In Chapter 1 (co-authored with Jamus Lim), effects of a financial crisis on households in Bulgaria are estimated using longitudinal survey data. Changes in consumption expenditures between the years 1995 and 1997, during which time the economy experienced a twin banking and currency crisis, are found to vary widely across different groups in society, represented in terms of their production dimensions, with the impact of the crisis associated with heterogeneous exchange rate pass-through related to differences in tradability, and also in terms of asset ownership and other socioeconomic, demographic and geographic dimensions.

In Chapter 2, the Bulgarian household survey data is further used to explore linkages between ethnicity and educational attainment. Econometric analysis at the individual level is used to estimate associations of Romany and Turkish ethnicity with differences in educational attainment, controlling for other individual and household variables; differences in the effects of the control variables by ethnicity; interactive effects of these variables with ethnicity; and association of ethnicity with selection into vocational secondary school, accounting for sample selection bias. The results suggest that cultural and social institutions play a role in human capital accumulation at a microeconomic level.

Chapter 3 (co-authored with Jamus Lim) zooms out from the microeconomic level to consider interactions between governance, educational outcomes, and economic performance, using a two-stage strategy that estimates national-level educational production functions which include institutional governance as a covariate, and uses these estimates as instruments for human capital in cross-country income regressions.