The polarized market hypothesis

Political polarization in the U.S. has increased noticeably in recent decades. Democrats and Republicans increasingly live in separate silos and are less likely to marry each other, live near each other, or talk to each other. What are the implications for the stock market?

There’s a strong argument that polarization will increase stock prices; that is, we’ll see a polarization premium which raises valuations today and lowers future expected returns. This premium could be present both in the aggregate U.S. stock market (we expect high valuation in highly polarized times) and in individual stocks (politically controversial stocks are overpriced). My logic is, as I’ve previously discussed, disagreement + short-sale constraints = overpricing.

Another implication of polarization is that the correlation of asset returns will change. If Republicans hold Republican stocks and Democrats hold Democratic stocks, we’ll have segmented stock markets where return covariation is driven partly by partisan sentiment. In addition to market, size, and industry factors in stock returns, we now must add a Republican/Democrat factor to explain covariation in returns.

The dystopian end game is a completely polarized financial system. Democrats invest in Democratic mutual funds with Democratic portfolio managers buying Democratic stocks in Democratic industries. Republicans work at Republican firms with Republican CEOs, issuing Republican bonds rated by Republican credit analysts. This dystopia is not entirely fanciful because, as I’ll explain, we already see these patterns emerging.

Evidence on polarized finance

A large body of evidence shows significant and sometimes increasing political polarization in U.S. capital markets. Let me discuss three separate areas: firms, retail investors, and finance professionals. I’ll quote extensively from Kempf and Tsoutsoura (2024) which I will call KT; see that paper for references and more details.

Firms

Corporate executives in the U.S. are predominantly Republican-leaning (using evidence from voter registration and political contributions). According to KT:

• Among corporate executives who have party affiliation, 66% are Republican.
• Republican affiliation is 75% in Florida, 50% in Massachusetts.
• Chief Financial Officers are most Republican (72%), Chief Legal Officers are least (50%).

Executives are increasingly self-sorting themselves into partisan firms, part of a larger pattern (also seen in marriages and dating) of “assortative matching” and “partisan homophily.” Here’s KT:

the likelihood that two executives work in the same firm increases by about 20% when they belong to the same political party. Moreover, the effect of shared party affiliation has increased substantially over time, especially after 2016.

Most of this increased polarization reflects increased polarization between regions: for example, Republican executives leave Democratic firms in California and move to Texas to work at Republican firms.

Another channel promoting polarized firms is mergers and acquisitions, according to KT:

Duchin, Farroukh, Harford, and Patel (2021) collect information on the political contributions of corporate employees and show the degree of political alignment between firms predicts a key cross-firm transaction, namely, mergers and acquisitions. They find companies which are less politically compatible are less likely to merge, and this relationship has strengthened over time. In addition, the effect is stronger during times of high affective polarization; i.e., when Democrats and Republicans dislike each other more. Political distance between the acquirer and the target is associated with worse post-merger performance and a lower value of synergies, consistent with political misalignment being an obstacle for post-merger integration. Hence, political misalignment between organizations can impose substantial economic costs.

This evidence suggests that while it might be rational for an individual firm to embrace its partisan identity (to avoid internal “political misalignment”), society as a whole is made worse by increasing partisanship.

Individual investors

A variety of evidence suggests that when the U.S. president is from your own party, you are more optimistic about the economy, more willing to make risky investments, and more willing to own stocks. Here’s Meeuwis, Parker, Schoar, and Simester (2022):

Using proprietary portfolio data on millions of households, we show that (likely) Republicans increase the equity share and market beta of their portfolios following the 2016 presidential election, while (likely) Democrats rebalance into safe assets …. These findings are driven by a small share of investors making big changes in allocation …

Looking at individual assets, KT discuss evidence that Democrats and Republicans have polarized views of issues such as ESG, climate change, and Covid, and this polarization impacts volume and prices. For example, during Covid, Republican-owned stocks had different responses to Covid news than Democratic-owned stocks, reflecting different attitudes towards Covid. Similarly, looking at house purchases and beliefs about climate change, optimistic Republicans are more likely to buy houses exposed to sea-level rise, with pessimistic Democrats selling those houses. Increased polarization leads to increased volume, as Republicans and Democrats trade with each other in response to differing interpretations of world events.

Finance professionals

Like individual investors, professionals are polarized both in their attitude towards the whole market and to individual stocks.

First, KT discuss evidence that political misalignment impacts behavior. When your preferred party does not control the presidency, you act more pessimistically:

• Corporate credit analysts issue lower credit ratings.
• Bankers demand more onerous terms on syndicated loans.
• Mutual fund managers hold more cash and less equity.

These facts have practical implications. If there’s a Democrat in the White House and you need a loan, seek out Democratic bankers.

Looking at individual stocks, we also see evidence of partisanship. Mutual fund managers overweight firms with CEOs who share their party affiliation. Mutual funds oriented around ESG and sustainability are more likely to have a Democratic portfolio manager. As discussed by KT in the case of credit analysts, different industries have different partisan identities: energy and technology firms have Republican analysts while finance and utilities have Democratic analysts.

Impacts on valuation

No effect?

One possibility is political polarization has zero effect on market prices. For example, suppose Alice is a Democrat and Bob is a Republican. Suppose that with no polarization, Alice and Bob would both hold the entire stock market. However, under extreme polarization, Alice only holds Democratic stocks while Bob only holds Republican stocks. Polarization will hurt Alice and Bob because they have less diversified portfolios, but they are willing to hold these riskier portfolios in order to affirm their political values.

The aggregate effect of presidential elections could also be neutral. When control of the White House switches from Democrat to Republican, Bob buys stocks from Alice, but if Alice and Bob both have roughly equal wealth, it could be that aggregate stock prices are unchanged. So elections bring high trading volume, but not much change in prices. That’s what Meeuwis et al. (2022) find in response to the 2016 elections: Republicans buy from Democrats, but the total impact is small.

Polarization discount?

An alternative view is that polarization lowers prices by lowering the potential pool of purchasers. Consider “sin stocks” involved in alcohol, tobacco, or gambling. If we view the world as polarized into sin-lovers and sin-haters, if there are more sin-haters, then sin stocks will be undervalued and will earn high future returns. That’s the finding of Hong and Kacperczyk (2009):

We hypothesize that there is a societal norm against funding operations that promote vice and that some investors, particularly institutions subject to norms, pay a financial cost in abstaining from these stocks. Consistent with this hypothesis, we find that sin stocks are less held by norm-constrained institutions such as pension plans as compared to mutual or hedge funds that are natural arbitrageurs, and they receive less coverage from analysts than stocks of otherwise comparable characteristics. Sin stocks also have higher expected returns than otherwise comparable stocks, consistent with them being neglected by norm-constrained investors …

Polarization premium?

According to Miller (1977) and Harrison and Kreps (1978), disagreement leads to higher prices. You could call this “overpricing,” or you could call this a “rational speculative premium,” but in any case it leads to high prices and low subsequent returns.

The basic idea in Miller (1977) is that the optimists always set the price. Suppose optimists think that the stock market is worth 7,000 while pessimists think it’s worth 5,000. If the total wealth of the optimists is small relative to the market, then you’d expect the equilibrium price to be somewhere between 7,000 and 5,000. But if the optimists are rich enough, they’ll bid up the price to 7,000, buying all the shares owned by pessimists. If short selling is not allowed, there’s nothing the pessimists can do but sell their shares and complain that the market is broken.

Moving now to the rational speculative premium of Harrison and Kreps, here’s a simple dynamic example. Consider the presidential election of 2024. Suppose the market consists of Democrats and Republicans. If Harris wins, Democrats will optimistically value the market at 7,000 while Republicans will pessimistically value the market at 5,000; if Trump wins, the reverse is true. Suppose that prior to the election, everyone agrees there’s a 50/50 chance that Harris will win. Thus before the election, both Democrats and Republicans agree that the market has fundamental value of 6,000.

In a world without short selling, what will happen after the election? The optimists always set the price. If Trump wins, joyful Republicans will bid up prices to 7,000. If Harris wins, joyful Democrats will bid up prices to 7,000. Thus no matter who wins, prices will be at 7,000 after the election. Given this structure, both Democrats and Republicans would be willing to pay 7,000 prior to the election, even though they all agree that the fundamental value is only 6,000.

This difference of 1,000 is the speculative premium or overpricing due to disagreement. Why are investors willing to pay more than fundamental value? Because each party believes that in the event that their candidate loses, they can unload their shares at inflated prices to those fools in the other party. This “greater-fool” dynamic drives overpricing. If you forced everyone to buy-and-hold forever, the pre-election price would fall to 6,000; it is only the option to resell that drives prices upward before the election.

Under this view, political polarization impacts both the aggregate stock market and individual stocks, as follows:

• Aggregate polarization premium. The whole U.S. stock market rises in value due to polarization, and future expected returns fall accordingly.
• Time variation in aggregate valuation. Prices are especially high at times of high polarization, and if there are predictable cycles in polarization due to election calendars, there can be predictable cycles in valuation.
• Stock-specific polarization premium. Politically controversial stocks are overpriced and will have low returns going forward.

A world divided

I’m not sure which effect is likely to dominate. We might see a polarization discount for some stocks and a polarization premium for other stocks. The outcome depends on the relative wealth of political groups, the extent to which political polarization aligns with other sources of disagreement, and how polarization might impact information flow.

The world is a complex place, political themes change over time, and markets adapt. Attitudes towards ESG, for example, vary over time as discussed in Baker, Egan, and Sarkar (2022). Firms might switch partisan identities, with Telsa in 2020 seen as a Democratic-leaning firm (Goldman, Gupta, and Israelsen (2024)), but maybe not so much today.[1] An enduring constant is that both asset managers and firms will seek to cater to investor demand, taking actions designed to attract more investors than they alienate.

One prediction that I’m confident making is the impact of market segmentation. If capital markets become more segmented, we’ll see prices in one segment become more disconnected to prices in other segments.

Most of the recent history of financial markets is about increased market integration and globalization. For example, when U.S. and Korean markets became more integrated, the correlation of U.S. and Korean stock returns went up. With political polarization, we might see a reverse pattern of market fragmentation and balkanization, with Democratic stocks decoupling from Republican stocks, and capital failing to flow across partisan lines.

Here, I’d expect a net loss to society, as market liquidity and beneficial risk-sharing falls. Just as it becomes more difficult to find a date when you eliminate half the population from the pool of candidates, it becomes difficult to find a financial counterparty in a polarized world.