Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group


First Advisor

Guillermo L. Ordoñez


This dissertation consists of two chapters on macroeconomic dynamics. The first chapter examines the consequences of technological diversification on long-run growth. I claim that the typical undiversified emerging economy would grow 0.23% faster if it were as technologically diversified as advanced economies. I obtain such estimate by building a growth model in which technological diversification facilitates inventor mobility across sectors, and through it, R\&D. Invention-relevant-knowledge is costly to accumulate, and more importantly, it is inalienable; inventors cannot disinvest cumulative knowledge capital when hit by a bad shock, and cannot sell it. Anticipation to this asymmetric effect leads inventors to refrain to accumulate too much capital. Technological diversification provides a way out. Inventors in a diversified economy can transport themselves to a better sector. A second result of the model is the quantification of the effect of volatility on growth. The model predicts that reducing standard deviation by 1% implies 0.08% faster growth. In the second chapter I extend the sovereign default model in the tradition of Eaton-Gersovitz (1981), to consider the consequences of strategic bailout from a lender country. Default is strategic as debt is not enforceable, and bailouts are strategic, as there is no obligation to extend them. The introduction of this implicit guarantee on sovereign debt has two opposing effects on its pricing. First, spreads are lower because the expected recovery rate is higher. However, if bailouts are a possibility after declaring default, and part of outstanding debt will dilute, then the value of exerting the default option is higher, and these events more frequent. This would raise spreads. I show that bond price schedule is decreasing in the haircut fraction of debt after re-structuring, is bounded from below, and is less sensitive to income fluctuations with the inclusion of a bailout probability. I also show that bailouts are more likely to happen when there are good realizations of income for the creditor economy. A final application can generate spreads close to zero for Italy even when income fluctuations and debt accumulation would predict otherwise. Shutting off the bailout implicit guarantee would have raised spreads from 0.03% to 1.8%.

Available for download on Friday, August 02, 2019

Included in

Economics Commons