UC San Diego

Past work on political careers assumes that who politicians are and how they got to be there will influence what they do. Unfortunately, several decades of empirical research have failed to conclusively link differences in previous political experience to the choices that politicians make. In this dissertation, I argue that previous experience matters. Empirically demonstrating this, however, requires both new data and new methods. New data are needed because existing resources do not allow researchers to recover the sequence of offices held by politicians before they reach institutions like the U.S. House of Representatives. New methods are needed because traditional measures of experience fail to capture differences among career sequences. To address deficiencies in data collection, I collected complete career sequences for 5,983 politicians who held the office of U.S. cabinet member, senator, representative, federal judge, state governor or big city mayor between 1809 and 1944. For each individual, all stints in public service were recorded, coded and assembled as sequences of office-holding events. To make sense of complex career sequences, I used an optimal matching algorithm used by molecular biologists to compare protein and DNA sequences. This algorithm was used to calculate a distance measure that summarizes differences in the number, type and order of offices occupied. I then used cluster analysis to group similar sequences together into meaningful career paths. Finally, these groups, or paths, were used as independent and dependent variables in statistical analyses. The application of these new methods to more comprehensive career data yielded several substantive findings. I find, for example, that political professionalization was pervasive in the U.S. during the late 19th and early 20th centuries. Features of professionalization included longer terms of service within the six offices studied here and increasing specialization in the career paths to these offices. Rather than reflect a random walk, I show that the pathways to power are shaped by political institutions. Finally, I find that previous political experience helps explain behavior in office, including the reelection experiences and retirement decisions of big city mayors and members of the U.S. House.