Recent revelations show Republican Sens. Richard Burr (R-N.C.) and Kelly Loeffler (R-Ga.) unloaded up to a million or more dollars’ worth of stock last month. Reports suggest they publicly expressed confidence — while privately voicing alarm — over the nation’s ability to handle the covid-19 crisis. Reporters have raised similar questions about stock sales by Sens. Dianne Feinstein (D-Calif.) and James M. Inhofe (R-Okla.), although it is unclear whether they personally directed those trades. House members and senior staffers have also come under scrutiny.

Notably, Burr voted in 2012 against the Stop Trading on Congressional Knowledge Stock Act, which ultimately passed. That’s the law enacted to penalize members of Congress for insider trading, or buying or selling securities based on nonpublic information. We don’t know yet whether any of these newly reported transaction run afoul of the Stock Act. But we do know that legislators’ financial self-interest can shape their congressional votes.

Here’s what you need to know.

How I did my research

For this study, I started by creating a measure of lawmakers’ financial self-interest. This metric captures the idea that the greater the number of firms that would gain from enactment of a new policy and the more money a particular legislator has invested in those firms, the greater that lawmaker’s financial self-interest in that policy area.

The measure varies by policy area. For example, while many firms benefit from trade, virtually all benefit from immigration. Because trade policy matters a lot less for firms that do not produce goods that are easily traded, I don’t count such firms in measuring a legislator’s financial self-interest in trade.

As a measure of overall financial self-interest, I sum up the value of stock in firms a politician owns at the time of a particular roll call vote. This is akin to net worth, with the important difference that certain firms receive more weight when a policy will probably affect them more.

For instance, trade legislation will affect Apple, with a sprawling global supply chain, more than Dominos, which has mostly domestic sourcing and production. A politician owning $1 million shares in Apple stands to gain more from more open trade policy than another who owns $1 million in Dominos.

To construct this measure, I take advantage of the Ethics in Government Act of 1978 that requires senators and representatives to disclose annually most assets they own worth more than $1,000 dollars. The Center for Responsive Politics makes these data available in an easy-to-use form for the years 2004 to 2014.

I analyze all votes on immigration — as categorized by the Comparative Agendas Project — and all votes on foreign trade agreements between 2004 and 2014. Political scientists have shown that policy change in these two areas can directly affect the bottom line of firms.

1. Investments can shape votes

I find a clear association between financial self-interest and lawmakers’ votes on trade and immigration measures in both the House and Senate — controlling for other variables, like how much of the lawmaker’s constituency is foreign born and the lawmaker’s pre-congressional career — that might also influence votes. As their financial self-interest increases in a policy, they are more likely to support that policy.

I double-checked the result by comparing how senators from the same state and party behaved on the same roll call vote; did stock ownership appear to influence differences between the two? I found 5,696 votes with such pairings over 11 years.

And yes, stock ownership does appear to make a difference in how these pairs vote. When these pairs of senators split their votes on a measure, the senator who scored higher on financial self-interest on the particular policy in question almost always voted in favor of the measure.

Consider the 2011 roll call vote by GOP Sens. James M. Inhofe and Tom Coburn approving the free trade agreement with South Korea. Large firms, which tend to be more capable of operating in foreign markets and establishing global supply chains, should benefit as free trade agreements remove barriers to trade and integration. Inhofe was more heavily invested in those firms than Coburn. Inhofe voted in favor, while Roberts voted nay.

2. Facing voters can moderate the influence of financial self-interest

To some degree, elections limit this sort of behavior. The Senate has staggered elections for members’ six-year terms, meaning only a third of those seeking reelection face voters in any given election year. Again, I used pairs of same party, same state senators to see whether an approaching election influenced senators’ votes.

And again, it does. Financial self-interest strongly predicts support of policies when elections remain distant. That’s less true as elections approach. What’s more, after legislators have announced their retirement, financial self-interest has an outsized influence on their votes. Notably, Burr announced in 2016 that he would retire at the end of his current term. Knowing he wouldn’t face voters again might have let him feel free to sell his stock.

3. Sunshine can limit influence of financial self-interest

Given the sheer volume of congressional votes in any particular term, we might expect many votes would fly under the radar. That suggests legislators might be more likely to vote their financial self-interest on less noteworthy issues.

According to recent studies, trade is unlikely to be as salient as the more emotionally charged and racialized issue of immigration. Indeed, the impact of financial self-interest on both House and Senate members’ votes is nearly twice as strong for trade as for immigration.

Scholars have struggled to find strong evidence of systematic insider trading among members of Congress. But it is difficult to prosecute insider-trading, and few elected officials leave paper trails or sell stock immediately after closed-door meetings. Occasionally, of course, there are exceptions. Just this past fall, former congressman Chris Collins (R-N.Y.) was convicted of insider trading.

Of course, politicians might invest in a firm not to make money but to signal commitment to the company in exchange for electoral support, or perhaps, eventually, a job. Even in such cases, business interests can sway legislative behavior. Because financial self-interest can shape legislators’ choices in more subtle ways, it will inevitably be tough to fully detect the extent to which lawmakers’ private interests mold their public behavior.

Caleb Ziolkowski is a PhD candidate in the political science department at the University of California, Los Angeles.