President-Elect Donald Trump recently announced that “in theory, I could run my business perfectly, and then run the country perfectly.” Since the President and Vice President enjoy the singular privilege of being exempt from conflict-of-interest statutes, he can perhaps make that position stick, though not everyone is convinced and his claim has touched off a lively debate.
In theory, the position of those elected to the U.S. Senate and House of Representatives is different. The current code of ethics of both chambers stresses the importance of not compromising their role as lawmakers, in which they are required to represent the interest of their voters rather than personal financial interests.
Mr. Trump’s ascendancy to the presidency has certainly given new urgency to the broader question of how the personal wealth interests of politicians could influence their voting and lawmaking. This is not only a question in the U.S., but should concern voters, policy makers, and academics across the world. It is this question that we address in our recent research, which investigates how the asset holdings of politicians in financial institutions affected their vote in Congress on the bailout of the financial sector during the crisis of September-October, 2008.
While scholars widely agree that a politician’s vote depends on their ideological position and their electoral prospects — which in turn are related to the politician’s ability to convince voters that he or she is the best advocate of their interests, and also by his or her ability to raise campaign funds — we show that the effect of the vote on a given politician’s personal wealth can also influence his or her voting on legislative proposals.
Using the politicians’ voting records on the bill that eventually became the Emergency Economic Stabilization Act — which bailed out American banks in the wake of the 2008 financial collapse — and detailed data on the equity stake of individual politicians in those financial institutions, we document that politicians whose portfolios were exposed to the stricken financial institutions are almost 60% more likely to vote in favor of the bailout plan than politicians whose wealth was relatively immune to the crisis.
Figure 1 illustrates these effects. Together, members of the House of Representatives held investments worth between $23.3m and $74.5m in the financial sector, with approximately 30 percent of the representatives owning shares in banks and other financial institutions. Our detailed data on the asset holdings of politicians (scraped from their financial disclosure forms), allow us to isolate the effect of personal wealth from competing determinants of voting, which include ideology and electoral prospects as well as a vast range of other potentially confounding effects.
In our empirical approach, we attempt to underpin our claim that it is the asset holdings of politicians per se that affects their voting, rather than these asset holdings reflecting the beliefs of politicians in the importance of financial institutions in the proper working of capital markets. Such “finance-friendly” politicians would then be more in favor of supporting a bailout regardless of the contents of their own portfolios. We use proxies for a politician’s beliefs in the importance of the financial sector derived from biographical details such as work experience in finance, educational background, and membership on finance-related congressional committees to tease out the effect of beliefs. We also consider, separately, the voting of politicians who had asset holdings in financial institutions before the 2008 crisis, but crucially not during this period. These “ex-investors” should share very similar beliefs about the sector’s importance as would current investors in the financial sector. Nevertheless, neither the biographical-based proxies for beliefs, nor the past asset holdings of politicians explain their vote on the bailout.
Perhaps even stronger evidence that our findings do not arise from differences in beliefs, but rather from the personal wealth effects of holding assets in the financial sector is provided by a test that uses information about the participation in defined contribution pension plans by the spouse of the politician. The return on these plans over 2008 should be a source of independent variation in personal wealth interest, in the sense that it is likely uncorrelated with other determinants of the politician’s voting behavior. Consistent with wealth interests per se and not beliefs affecting the vote on the bailout, we find a strong correlation between the returns on the spousal pension plans and the vote on the EESA. The data show that where the spouse was down in the market, the probability of voting in favor of the bailout rose.
While Congress has recently approved the Stop Trading on Congressional Knowledge Act, purportedly aimed at preventing politicians from taking advantage of non-public information gleaned from their activities as lawmakers, our evidence suggests that politicians don’t have to trade on information to become better off; they also have the ability to pass bills that benefit themselves financially. Researchers in both political science and economics need to look much more closely at this possibility than they have done, so far.
