Value investing for crazy times

How should you invest if market prices are crazy? That’s the question posed by active manager David Einhorn in a recent interview with Barry Ritholtz.¹

As I’ve previously said,² I’m skeptical of Einhorn’s assertion that indexing has made the market “fundamentally broken.” But let’s put aside that question and consider the right response to broken markets. Here, I mostly agree with Einhorn and especially with his emphasis on cash distributions. I am totally on board with his solution: “We’re gonna have to get paid by the company.”

In my view, value investing is almost hardwired by math (specifically, the definition of returns) to work over the long term. Even if market prices are crazy, value investors should outperform due to cash distributions to shareholders such as dividends and repurchases; these don’t rely on the sanity of equity markets but rather on the nature of yields (lower prices mean higher yields).

Let me give you Einhorn’s diagnosis of what has gone wrong:

And the active long only managers are down a lot and they still have people paying attention to certain stocks. But there’s entire segments now mostly in the smaller part of the market where there’s literally nobody paying any attention. Like these companies could announce almost anything other than a sale of the company and… nobody would notice. And so we’ve had to adjust our thinking because our thinking before used to be if we buy this at this times earnings and they’re gonna do 20% better than everybody thinks, and the multiple re-rates as a result of that, we’re gonna do terrifically…

But what if those people aren’t in business anymore… There’s way fewer people listening. And the result is, is, is if we buy these things, we’re not going to get the same kind of return that we used to get.

… there’s just nobody who’s going to pay attention to notice that the earnings were 15% better. So if nobody notices, nobody’s there, nobody’s going to buy, nobody’s going to care.

To summarize: Back in the 1990s, markets responded to information about fundamentals. If you could identify stocks trading for below fundamental value, you could buy these stocks, confident that the market would soon agree with you, and the price would rise. But according to Einhorn, this pattern ceased, and therefore you cannot rely on prices going up for undervalued assets. If your previous strategy was “buy low, sell high,” you must abandon that strategy, because prices no longer get high in response to good fundamentals. Einhorn describes the conundrum as “a bargain that remains a bargain is no bargain.”

Now let me move on to his solution to this new market dynamic:

We’re buying things at four times earnings, five times earnings, and we’re buying them where they have huge buybacks and we can’t count on other long only investors to buy our things after us. We’re gonna have to get paid by the company. So we need 15, 20% cash flow type of numbers. And if that cash is then being returned to us, we’re gonna do pretty well over time.

…And if you pay four or five times earnings and the balance sheet is not levered and they’re able to return the cash and buy back 10, 15, 20% of the stock in four or five years, they’re going to run out of stock or the stock is going to go up. So you’re literally counting on the companies to, to make that happen for you.

I especially like his formulation that if a company does buybacks, “they’re going to run out of stock or the stock is going to go up.” Here’s how I would have phrased it: if I purchase stock in a company that buys back $X of stock every year, eventually either the price will go up or I will be the only owner of that stock. Thus, even if the market is crazy, there’s a benefit to cheap stocks that distribute cash to shareholders.

Seen through the lens of cash distributions, it’s not actually true that “a bargain that remains a bargain is no bargain.” No, a bargain that remains a bargain remains a bargain. Let me give you an example. Suppose the interest rate is 10% and I have a riskless asset that generates $1 per year. Plug in Gordon’s equation and you get a value of $10 for this asset. Suppose it is actually trading at a price of $5, so it is undervalued by 50%, and suppose that it remains undervalued forever. Is this a bargain? Of course! Wouldn’t you rather own a bond that yields 20% a year rather than 10% a year? A bargain is always a bargain.

If you were hoping to buy this asset for $5 and then sell it a year later for $10, a Warren Buffett-like strategy of just holding forever may disappoint. But it is still true that a long-only investor can prosper by purchasing cheap assets. Now, maybe that’s not a sustainable business model for a hedge fund, but it’s still true that value will work over time.

One issue that Einhorn raises is the duration of his holding period. In the good old days, you could buy an underpriced stock, and maybe in six months it would become less underpriced:

…we’re gonna figure out what somebody else is going to buy six months, a year, two years before they come to that conclusion.

But with less efficient markets, the holding period needs to become longer. Here’s Jeffrey Pontiff discussing why dividends are a value investor’s friend in crazy markets:³

Dividend payments effectively lower the duration of the arbitrage position…

The impact of dividends on mispricing deserves special consideration since dividend policy is determined by the firm. Thus, dividend policy provides a mechanism for management to influence the degree to which the firm is mispriced. … under-priced firms are more likely to initiate dividends.

Pontiff’s basic point is that while the stock can be underpriced, the dividend can’t be:⁴

Although the security is subject to mispricing, the dividend is not…Each time a dividend is paid, a partial liquidation of the mispriced security occurs…In equilibrium, securities that pay large dividends will be subject to less mispricing…

So to summarize, if you can’t rely on other investors to recognize and correct underpricing, you might be able to rely on the firms themselves, who will pay dividends and repurchase the underpriced stock. In other words, it used to be alert value investors who realized that firm XYZ was a bargain, and purchased the shares, driving up the price. Now, it is firm XYZ itself who does the buying (via repurchases). The actors have changed, but the plot is the same: buy cheap shares, get rewarded eventually.

So why has value struggled over the years? Let me go over three forces that impact value returns. I’ll discuss two concepts that Einhorn mentioned, distributions (“We’re gonna have to get paid by the company”) and migration (“the multiple re-rates”) and then one he did not directly address: changes in relative valuation.

Cash distributions

Let me focus on the simplest type of distribution: dividends. The Russell 1000 Growth Index has a dividend yield of 0.6%, while Russell 1000 Value has a dividend yield of 2.1%. If stock prices really were crazy and unrelated to fundamentals, including dividends, then you’d expect to earn 1.5% more per year by investing in value. For the Russell 2000 growth vs. value the spread is slightly smaller at 1.3%.

Looking at history from 1964 to 2006, Fama and French find similar dividend spreads.⁵ They say that the total return difference between small value and small growth was 8.5%, of which 1.7% is explained by differences in dividend yield. Thus, the distributions that Einhorn mentions have always been one benefit of small value strategies. Dividends are even more important for large-cap value; Fama and French find the difference in dividend yield between value and growth is 2.4% which is more than half of the 3.7% difference in annual returns for large cap.

Let’s look again at the statement “a bargain that remains a bargain is no bargain” and consider an extreme case: China Life Insurance.⁶ China Life has two classes of shares, the A shares trading in Shanghai and the H shares trading in Hong Kong. Amazingly, the A shares cost 3.5X more than the H shares despite having identical legal rights and identical dividends. This discrepancy, called the AH premium, is common to many Chinese stocks, and some have had high premiums for many years. In US terms, the H shares are value stocks that are stuck in a “value trap” and somehow remain underpriced (relative to the A shares) for years.

If you believe that the premium will continue in the future, does it make sense to ignore the price discrepancy? No. The H shares have a dividend yield of 5.4%. The A shares have a dividend yield of 5.4/3.5 = 1.5%. So if you buy the H shares instead of the A shares, you have a 3.9% (5.4-1.5) tailwind in your favor. A bargain is almost always a bargain when there are dividends involved.

However, consider a different Chinese firm: CanSino Biologics. Like China Life, CanSino A shares cost 3.5X as much as H shares. But this firm has no dividends, so there is no yield benefit of buying the cheaper H shares. When it comes to stocks with no dividends, value investors have lost a valuable ally.

Migration

In addition to cash distributions, another reason value stocks have high returns is migration: that is, value stocks tend to migrate out of the value category of cheapest stock and into the less cheap categories. In other words, bargains don’t remain bargains, they often eventually become normally priced.

Migration has historically been an important part of value returns. Fama and French say:⁷

We study how migration of firms across size and value portfolios contributes to the size and value premiums in average stock returns…The value premium has three sources: (i) value stocks that improve in type either because they are acquired by other firms or because they earn high returns and so migrate to a neutral or growth portfolio; (ii) growth stocks that earn low returns and as a result move to a neutral or value portfolio; and (iii) slightly higher returns on value stocks that remain in the same portfolio compared to growth stocks that do not migrate.

So, value firms have high return partly because they migrate up (point (i)) and partly because they have higher yields (point (iii)). Growth firms have lower returns because they migrate down (point (ii)).

So according to Einhorn, part of the problem is that value firms are not being recognized and rewarded but are staying cheap relative to growth firms (just as in China, H stay cheap relative to A). Is this true for U.S. stocks? Not according to Arnott et al. who argue that in the period 2007 to 2020:⁸

A related argument suggests that both the markets and the economy have evolved to a point where value stocks stay cheap and growth stocks stay richly priced, slowing the migration that drives the value advantage… Empirically, we found that migration is essentially unchanged from the past.

So why has value done poorly in recent years? This brings me to the widowmaker: widening value spreads.

Widening value spreads

Rather than asking whether a particular stock moves from the value category to the growth category (migration), a separate question is whether value stocks as a group get cheaper relative to growth stocks. You could rephrase this as “a bargain that becomes even more of a bargain leads to tears and drawdowns.”

Here, the systematic evidence is clear: value spreads have widened in recent years, thus hurting value investors. As shown by previous Acadian research, the value spread widened after 2017, mostly reflecting multiple expansion by growth stocks.⁹ In other words, value got cheaper relative to growth.

The dramatic widening of value spreads since 2017 has been documented over the years in a series of engaging pieces written (with erudition, humor, and panache) by Cliff Asness of AQR. These pieces trace out the Five Stages of Value Drawdowns: denial, anger, bargaining, depression, and acceptance. The latest, from January 2023, is titled “The Bubble Has Not Popped.”¹⁰ Sadly, the spread has likely widened since then.

The spread only started to widen in 2017 or so; prior to that, the poor performance of value did not reflect changes in valuation but rather probably reflected poor relative fundamentals for value stocks. In other words, the poor performance of value prior to 2017 was consistent with market efficiency, not craziness or a disconnect between prices and fundamentals.

What’s the bottom line? I believe value will, in the decades ahead, continue to be a useful framework for investors. Value investors are engaged in a socially useful activity: finding and correcting mispricing. Active value managers like Einhorn are serving on the front lines of price discovery.

While value is an important part of systematic investment, it is not the only tool in the toolbox. Systematic investment can diversify across many non-value signals including momentum, quality, peer-linkages, and sophisticated statistical prediction methods. When times get crazy and value spreads widen, these other systematic signals are a source of strength.