There are some conditions that doctors cannot conclusively diagnose through examinations and tests. To determine whether a patient has one of these conditions, doctors must perform what is known as a “diagnosis by exclusion,” in which they eliminate other possible causes. A cynic might say the doctors are diagnosing the patient as a hypochondriac. 1 1 Close Or the doctor missed something. There may be a parallel in investing. Value stocks have been underperforming growth stocks for several years. Few, if any, macro factors seem to explain this mysterious underperformance. In the search for macro causes, it may end up simply becoming a diagnosis by exclusion. Perhaps we are just hypochondriacs looking for a macro disease that does not exist.
One of the problems in the search for causes is that value stocks do not have uniform macro drivers. Think of two very different stocks that are both cheap on value. A stock with steady earnings might be cheap because it has very poor growth prospects. Another stock might have high potential but might be cheap because of poor management and high debt levels. These two stocks could have very different reactions to macro data. Higher inflation, for example, could erode the value of the first company’s steady earnings, but could help the second company by devaluing the debt it holds.
A lot of discussion of value underperformance has centered on industry performance. And it is true that some growth industries, such as tech, have outperformed some of the more value-oriented ones, 2 2 Close The S&P 500 Information Technology Sector index has outperformed the broader S&P 500 index by around 77% over the past 5 years. Source: Bloomberg. Data from 5/9/13 to 5/9/18. but this doesn’t fully explain value’s underperformance. According to our own findings, value stocks have, on average, underperformed growth stocks within their sectors. 3 3 Close Source: AQR. So it is more than just tech or a couple of FANG stocks driving the results. 4 4 Close Take that strawman.
If you had asked for a macro explanation for value underperformance a year ago, you probably would have gotten an answer that included at least one of the macroeconomic conditions of the time: moderate economic growth, low volatility, low interest rates and low inflation. It wouldn’t take much of a stretch to come up with a story for how each of those factors is hurting value or helping growth. The problem is that this year we’ve seen at least a partial reversal of three of those conditions. Volatility spiked early in the year, interest rates have risen a bit, and inflation has moved up to the Fed’s target. 5 5 Close The VIX index moved from a low of 8.92 on 1/4/18 to a high of 50.30 on 2/6/18. The U.S. 10-yr Treasury yield has moved from 2.41% at the close on 12/29/17 to 3.00% at the close on 5/9/18. The Core PCE index, the Fed’s preferred measure of inflation, last printed at 1.9% YoY for March 2018, versus 1.5% YoY as of December 2017. Source: Bloomberg, Bureau of Economic Analysis. While we aren’t seeing high rates or hyperinflation, you would still expect that value would be performing a little better, but instead it has continued to lag. Economic growth is still moderate, but value has performed well in past periods of moderate growth, such as the mid-2000s. 6 6 Close These strawmen are getting burned worse than Zozobra.
Some have blamed indexing. The rationale is that growth stocks are more heavily weighted in some indexes, and people are buying those indexes. It’s difficult to disprove this one, but it seems like it could overestimate the differences within the indexes and the effects of passive investing. Still, it is difficult to totally dismiss the possibility.
Others have cited quantitative easing as a cause. Citing QE seems like it has become a diagnosis by laziness. If an analyst can’t think of any other cause, he can safely say something like in a time of large central bank balance sheets, what else can you expect? Since the Fed stopped new purchases a few years ago and has started to let the bonds roll off of its balance sheet, it is less credible as a cause of value underperformance. Perhaps now folks will say it is the QE unwind that is hurting value stocks.
It is helpful to look at the rare past times when value has underperformed growth. Two really stick out: the tech bubble in the 1990s and the Nifty Fifty period of the late 1960s. 7 7 Close One other period of growth outperformance was post-stock market crash of 1987, which has little in common with the current period. There are plenty of differences between those times and the current, almost too many to list, but they do have one thing in common: sentiment. In each period there was a feeling that change was coming and there would be big winners and losers in the future. In the late 1960s, folks thought that a few large well-managed companies had big advantages over others and buying them would be the “one decision” investors needed to make. 8 8 Close Another thing the three periods have in common is annoying catch-phrases to describe them. In the tech bubble it was the “new paradigm” where valuations didn’t matter because any company with even a slight connection to the internet would outperform others in its sector. Now we have disrupters that are upending their industries and allegedly smashing their outdated competitors. Admittedly, sentiment is not a clear macro cause, but this psychology may at least partially explain the investor behavior. 9 9 Close It also is not a very quanty explanation.
We know that after those last two periods, value bounced back quickly, but investors should be aware of some differences. The recent underperformance has been more drawn out, though it also hasn’t been as sharp as during the 1960s and 1990s. In those past periods dispersion among value and growth stocks was quite wide. In other words, value stocks were particularly cheap relative to their histories. Despite value’s recent underperformance, that is not the case now.
Perhaps someday we’ll have a clearer macro explanation for why value has underperformed, but it shouldn’t be worrying if there isn’t. There are plenty of ways for investors to get macro exposures, and part of the appeal of value investing is that it is different from just getting beta to the market or the economy. Value has performed well over the long run. We should get over our macrochondria and just accept that sometimes there are symptoms without a syndrome.
What We Are Watching
U.S. Retail Sales (Tuesday) U.S. retail sales data published by the Census Bureau was unexpectedly soft in January and February before rebounding in March. As a result, consumer spending data for the first quarter as a whole was fairly sluggish (up only 1.1% QoQ annualized according to the BEA), but economist forecasts for the remainder of the year continue to look upbeat. If this view is correct, April sales numbers should show a continuation of the rebound in spending seen in the March data. If sales fall short of consensus forecasts, it could cause economists, policymakers and market participants to question their expectations for stronger consumer activity moving forward.
U.K. Labor Market Data (Tuesday) U.K. inflation has been above the Bank of England’s 2% target over the last year, in large part due to the impact of a weaker pound on import prices. In the May Quarterly Inflation Report (QIR), The BoE concluded that “the inflation rates of the most import-intensive components of the CPI appear to have peaked.” However, the central bank’s forecasts see continued above-target inflation in the near term because “wage growth and domestic cost pressures are firming gradually.” This suggests that the BoE will likely be watching monthly labor market data closely to assess the trajectory of wage growth. Gains in average weekly earnings have been trending higher over the last year, and a continuation of this pickup would likely add to the BoE’s confidence in its forecasts. Conversely, disappointing readings on earnings growth could cast doubt on the need for further rate hikes from the central bank.
Brazil Central Bank Meeting (Wednesday) The Banco Central do Brazil (BCB) has been in an easing cycle since late 2016, during which time policymakers have reduced interest rates from 14.25% to a historic low of 6.5%. The dramatic reduction in rates has been possible due to stability in the Brazil currency (the real) and a significant decline in inflation to the lowest level in two decades. Markets are pricing high odds of one last 0.25% cut at this month’s meeting, but do not see much prospect for further easing beyond that. The real has depreciated in recent weeks, perhaps reflecting diminished support from high interest rates, which could argue for caution on the part of the BCB. Elections scheduled for later this year further complicate the policy decision, as the central bank has historically been hesitant to adjust rates too close to a vote. As a result, any decisions made in the near term may be difficult to reverse a few months from now if conditions have shifted. If the BCB chooses to leave rates unchanged, or signals a more hawkish outlook for future meetings, it could provide some support to the real.