UC Berkeley

To which extent may history help us understand current financial outcomes? Financial

History typically tests for financial theories using historical data, or studies past financial

outcomes to understand the present through analogy. This dissertation focuses on an alternative approach to the use of history in finance, which I label "History & Finance." This approach exploits facts and institutions of the past, and the persistence of the economic and social phenomena they determine, to directly explain current financial outcomes. In the first chapter of the dissertation, I define the scope of History & Finance based on its differences with Economic and Financial History. History & Finance has its roots in the Political Economy approach that exploits the long-run effects of historical shocks and institutions to understand current economic outcomes. I survey the research in History & Finance, organizing it across the subfields of Finance which have used this approach so far. I discuss the challenges that this method poses to researchers, and in particular the difficulty of making plausible causal statements in long-run settings, and of determining the channels of transmission of the effects of historical phenomena on current financial outcomes. I then propose five directions that scholars may follow to enlarge the scope of this approach to research in finance.

In the second chapter of the dissertation, I use the History & Finance approach to understand the current spatial distribution of innovation. I focus on the innovation of traditional industries, a margin of innovation which is largely neglected in the literature, although it includes the vast majority of the innovations produced every year in Europe and the United States. I exploit newly assembled data on innovation and education at the level of European regions, as well as a unique rm-level data set on the innovation of traditional industries, to show that the amount of formal education of blue-collar workers is an important determinant of the innovation produced in traditional industries. The relevance of the formal education of blue-collar workers for innovation is alternative to the learning-by-doing hypothesis and the directed technical change literature, none of which recognizes a role for formal education in the productivity of jobs on the line. Moreover, I document for the first time that the variation in the amount of basic education across European regions is highly persistent over the decades, despite the major institutional and economic shocks that have differentially affected European regions over time. Hence, I use the historical variation of basic education across regions to address the issue of the reverse causality between current innovation and current basic education. I also propose an historical natural experiment, the quasi-exogenous diffusion of the printing press across Europe after 1450 AD, as a source of variation for cross-regional basic education in the past. Higher basic education increases the amount of innovation produced by firms in traditional industries, whose workers are in large

part basically educated individuals. The effect is indeed stronger for firms that employ more blue-collar workers. I also document that higher basic education determines higher capital expenditures and lower financial constraints for those firms that innovate more.

In the third chapter of the dissertation, which is coauthored with Marcel Prokopczuk and Michael Weber, I use the History & Finance approach to understand the limited stock

market participation of households. Because the Jewish population has been associated with the provision of financial services in Europe for more than eight centuries, we test whether those German counties with a higher persecution of local Jewish communities over time tend to invest less in stocks. Indeed, in those counties where historical persecution was higher as far back as in the Middle Ages, present-day households invest less in stocks. The size of the effect is similar across cohorts, and it does not fade away over time. The evidence is consistent with a cultural norm of distrust in finance that has started in those areas where Jews were persecuted more in the past, and has transmitted across generations over the centuries. To obtain quasi-exogenous variation in the extent of Jewish persecution across German counties in the distant past, we exploit the forced migrations of the first German Jewish communities from the Rhine Valley to current Poland to instrument for the existence of Jewish communities in counties in the Middle Ages, and hence the likelihood that persecution against the communities was documented. This dissertation defines History & Finance as a rising approach to the study of financial phenomena, and it proposes several promising routes that researchers across subfields of finance may take to bring this approach from its infancy to a mature stage.