Have Senators beaten the market during the recent downturn?
Our index indicates that since January 2014, senators have not, on average, outperformed the S&P 500. This is not necessarily surprising. Older senators may have chosen to hold lower volatility mutual funds that under-performed US equities in the bull market. Some senators may have picked individual stocks with little insider knowledge, and prove little better at doing so than most retail investors. These effects are enough to overwhelm excess returns that senators may have enjoyed by trading on privileged information.
However, it might be easier to tease out senators’ trading on insider knowledge during the market turmoil surrounding the Covid-19 pandemic. First, numerous senators received confidential information on the scope and likely US response to the pandemic. Second, all senators with substantial equity exposure had an interest in acting on that knowledge. Third, due to substantial market volatility, the potential return from acting was very high. Finally, the appropriate strategy – sell equities to avoid losses – was easy to implement.
To investigate, we pulled holdings and transactions from our database from January 1 – May 6 2020 and tracked senator portfolio returns relative to the S&P 500 index. One note of caution: these data are not entirely complete, since (1) senators have 45 days to file periodic transaction reports, so some recent trades may not be reported; and (2) we may miss some senators who file paper, as opposed to electronic, transaction reports.*
Since the beginning of 2020 through May 6, 2020, our “all senator” index indicates that senators beat the S&P 500 by 3.4 percentage points, losing 8.9% instead of 12.0% (Figure 1). To put that in perspective, the median senator, with $1.2M in stocks at the start of the year, would have avoided $40,000 in losses by holding our index instead of holding the S&P.
Digging a bit deeper, Figure 2 shows that while many senators suffered losses in the recent sell-off, we estimate that 35 senators beat the S&P 500. Two senators actually made gains since the start of the year – Pat Roberts (R-KS) and Tina Smith (DFL-MN). Figure 3 shows the full list of senators who have beaten the market since the start of the year. You can explore their full holdings on our main dashboard.
Of course, the S&P isn’t guaranteed to outperform a carefully-curated, well-hedged, and possibly professionally managed portfolio. In that case, we’d expect that senators would behave passively when faced with market turmoil, with perhaps minor portfolio adjustments. To test this, in Figure 4, we calculated net weekly transactions for each senator, normalized by portfolio value on Jan 1, 2020, and plotted the transaction distribution since the start of the year.
Some senators make minor portfolio adjustments in January, along with a few large sales equal to around 10% of portfolio value. There is much more action in February – some purchases in the second week, and some sales in the third week. Most of these transactions are small as a fraction of portfolio value, and look more like minor rebalancing transactions than massive sell-offs. Interestingly, there’s a big spike in purchasing in the third week of February, right after the market peaked (on February 19). This could indicate senators trying to buy into the initial dip – not necessarily the behavior of someone with insider knowledge about the true scope of plausible economic damage. More net purchases show up in March, which roughly track with the market’s partial recovery from March – April.
Were these active trades responsible for senators’ above-market returns? As a partial answer, Figure 5 plots gross transactions (left panel) and net transactions (right panel) from January - May against portfolio returns over that period. The horizontal line indicates S&P 500 returns over that period. In the left panel, it’s generally clear that senators who traded a smaller (gross) fraction of their portfolio earned higher returns. In the right panel, senators who made net sales over the period suffered relative losses; those with net purchases enjoyed relative gains. In most recent reporting on congressional stock trading scandals during the Covid-19 pandemic, we hear that senators (like Richard Burr**) made excess returns by dumping their stocks in February. These data suggest a different story – the senators who performed best were those who bought when the market was down.
Ultimately, however, active trading didn’t do much to benefit individual senators. We re-calculated the portfolio returns of individual senators with positive gross transactions from January – May, assuming that instead of actively trading during that period, they instead passively held their January 1 portfolio. Figure 6 plots the counterfactual returns (horizontal axis) against their estimated actual returns (y-axis), along with the 45-degree line indicating equal returns under both strategies. Senators with points above the 45-degree line would have earned a higher return had they avoided actively trading and instead just passively held their portfolio. Senators with points below the 45-degree line earned higher returns by trading actively. Pretty much none of the senators making active trades were better off for doing so.
Based on these preliminary estimates, it looks like senators behaved much like many individual investors during the recent market crash – most passively rode the market, and those who tried to actively trade ended up losing out when they did. It’s too early to make any definitive conclusions, however – we’re still working on gathering more data, refining our statistical methodology for estimating financial holdings, and conducting exploratory research on individual members. On the data front, we’re especially excited to get our hands on data that is harder to access, subject to less scrutiny, and therefore more likely to conceal suspicious trading behavior – specifically, paper-filed Senate transaction reports and data from members of the House and the Executive branch.
——————————
* We’re working on digitizing these reports – sign up to volunteer or back our Kickstarter to help!
** Richard Burr isn’t included in these analyses because he has filed paper transaction reports since 2015, which are much more difficult to access and analyze -- suspicious!
