Semester

Spring

Date of Graduation

2014

Document Type

Dissertation

Degree Type

PhD

College

College of Business and Economics

Department

Economics

Committee Chair

Stratford Douglas

Committee Co-Chair

Ronald Balvers

Committee Member

Joshua Hall

Committee Member

Alexander Kurov

Committee Member

Harry Turtle

Abstract

My dissertation examines the information content of stock trades by members of Congress (politicians) in three separate essays. In my first essay, based on approximately 62,000 common stock transactions over the 2004--2010 period, I present evidence that past firm-level trading activity by politicians predicts future firm-level stock returns, and this prediction is stronger for the pre-recession period and for small firms. By constructing buy-minus-sell transactions-based portfolios, I document that the portfolios of Republicans outperform those of Democrats. Also, being a powerful member of Congress is associated with substantially higher abnormal returns for Republicans, but not for Democrats. Among powerful Republicans, the portfolios of unsophisticated traders perform significantly better than those of sophisticated ones with abnormal returns exceeding 34% on annual basis under one-week holding period. The abnormal returns in general disappear beyond three-month holding period suggesting that politicians trade on time-sensitive value-relevant information. This study contributes to the literature showing that power, party membership, and investor sophistication are important determinants of the abnormal returns in congressional stock portfolios.;In my second essay, I , for the first time in the literature, separately examine the performance of common stock transactions of politicians and their family members (spouses and dependent children). I document that the transactions by both politicians and spouses contain value-relevant information, but the content of information is less time-sensitive for spouses. On the other hand, the transactions in the accounts of the dependent children perform very poorly. I also present evidence that the transactions in tax-deferred accounts such as 401(k)s contain less information than those in taxable trading accounts for both politicians and their spouses. I finally show that the portfolios of the powerful politicians' spouses outperform those of the non-powerful politicians' spouses. These results suggest that members of Congress take advantage of their informational advantage trading through the accounts of their spouses.;My third essay examines the properties of aggregate congressional trading. I investigate the ability of aggregate trading activity by members of Congress to predict stock market returns using 77,663 stock transactions over the 2004-2010 period. I find that aggregate congressional trading predicts excess market returns 2 to 3 months in advance. These findings imply that politicians, similar to corporate insiders, trade on information containing macroeconomic components.

Share

COinS