November 18, 2013
The article below is excerpted from a letter sent to clients on October 14, 2013. (Yes, we are just now getting around to posting it on the website). For some context, the letter was sent in the midst of widespread concern about the government shutdown and the debt ceiling standoff. We thought the importance of these issues was being exaggerated, and they were the “headline risks” to which we referred.
Values Trump Headlines
“Headline risks” have been practically constant through the ages, but for all the hype they generate, they’ve rarely had much lasting effect on market outcomes. Investors’ eventual results have been far more influenced by valuations: whether the investments they were buying were cheap or expensive compared to their economic fundamentals, based on a reliable and consistent measure of value.
This can be seen in the below chart of valuations and eventual returns for the postwar stock market. We’ve calculated the valuation of the S&P500 for each month over that period, and grouped together the months into quartiles: the cheapest 25% of months, the second cheapest 25%, an so on.
Sidebar: Measuring valuation — Because corporate earnings tend to fluctuate very widely over multi-year periods, the commonly used valuation measures based on one year’s worth of trailing or predicted earnings can be very misleading. Measures based on the longer-term trend in earnings have historically been much more reliably predictive of future returns and than the one-year measures. The charts below use the real S&P500 price divided by the postwar real earnings trend, which is the most reliable method we have found, but there are several other valid ways of “normalizing” earnings that provide a very similar result to what is discussed below.
The blue bars show the average inflation-adjusted returns for investors who bought during those months and held on for the next 7 years.
The difference in buying cheap vs. expensive is amazing. Investors who bought the S&P500 during the least expensive times averaged a fantastic 7-year real total return of 127%. As we get into the more expensive quartiles, average returns steadily decline, until we reach the most expensive times in the market, which averaged a 7-year return of just 8%! Putting that into annualized terms, that’s a real return of 12%/year for the cheapest times vs. 1%/year for the most expensive times.
Investors who bought into expensive markets were also more likely to suffer long-term losses. The next graph shows the same valuation categories along the bottom, but this time the red lines show how often investors took an inflation-adjusted loss after 7 years:
Here we see that investors in the cheapest times never once fell behind inflation after 7 years, whereas that happened to investors in the most expensive periods nearly half the time!
Sidebar: Why 7-year returns? Because that’s the kind of timeframe over which valuations have reliably predicted investment outcomes. There is no reliable way to know what will happen in markets over the next few months or even years. However, once you get into 5-or-more years timeframe, valuations start to become a lot more predictive (and generally, the longer you give it, the more reliable they are). There’s nothing special about 7 years in specific; it’s just a duration that has typically been long enough to let valuations do their thing.
Incidentally, this really helps explain why so few people actually follow a value investing approach — while it has a strong tendency to work very well over the long term, the wait can sometimes be excruciating.
Right now, the valuation measure used in these charts puts the S&P500 in that most expensive grouping. In fact, not only is it in the top 25% of most expensive times, it’s in the top 10% of most expensive times. The market’s valuation is now as high as it was during the bulk of the housing bubble, after which it went on to decline by 58% from the peak.
This — not the media-hyped crisis du jour — is the big risk we see for investors in the broad US market.
Good Investments are Out There
Fortunately, the US market is fairly unique in its extreme overvaluation. We see many areas of global markets (and even some pockets of the US) that are reasonably valued and priced at levels that have historically led to good returns for patient investors.
For example, the below chart shows comparative valuations for US and emerging market stocks. As the US (the black line) has gone from expensive to even more expensive, emerging market stocks (blue line) have gone from cheap to cheaper.
The bad news is that we’ve owned emerging market stocks for a while, and this was a big contributor to our lagging performance over the past couple of years. But the good news (for us, anyway) is that the eventual prospects for US and emerging stocks are markedly different. Markets have an extremely strong tendency to “mean revert” towards middle of the road valuations (the orange line). Such a reversion to fair value would be bad for US stocks, but quite good for emerging stocks.
A similar situation can be seen in this chart of US and European stock valuations:
It should be noted that both of these charts a few months out of date, so they don’t show the recent upturn in foreign stocks. Still, they are valuable in that they provide ballpark idea of relative value. They also show that it’s not the case that the US has a consistent premium to these other markets. Rather, different markets fall into and out of favor at different times, and right now, the US is trading at a large premium.
Below is a more recent snapshot of valuations (measured by the cyclically-adjusted price to earnings ratio, another great way to normalize earnings and the same one used in the above two charts). It shows that the US stock market is trading at a 60% premium to both developed international and emerging market stocks:
Many foreign countries have their problems, of course (though the US has its share too). The difference is that modestly priced international stock markets reflect an expectation that these economies will just do “okay.” If they simply do okay, returns could be good, and if they do better than that, returns could be great. US stock prices, in contrast, reflect the assumption that US companies will do spectacularly well. If they just do okay, that could be real trouble for the US stock market.
While we do have some exposure to pockets of the US market that we think are reasonable, the bulk of our stock exposure is in international markets. These markets have started to perform better over the past quarter, which has helped the accounts over that time. But more importantly, we think the above charts and data show that there is plenty of scope for further gains in these areas without taking undue risk.
A Cautious Approach
We are invested in a diversified manner in areas that are priced for good long-term returns, but there are a couple of reasons why we are maintaining a relatively conservative stance right now.
The first is that while there are some good valuations out there, there is very little that is really cheap. The second is that we continue to be concerned about some serious fundamental issues facing the economy: our large and growing debt burden, and the potential difficulties that could result from the Fed unwinding its ultra-easy monetary policies (or, from failing to unwind them fast enough).
While there is reason for caution, it’s also possible that our areas of concern don’t come to the fore for a long time —or, as unlikely as it seems, ever. In the meantime, there are good values out there, so we want to be invested to make money if global markets continue higher. But because of our concerns, the exposure we have is on the conservative side so that we can be positioned to take advantage if there is an opportunity to get stocks on the cheap side of the spectrum.
Based on all our research into what has worked for investors in the past, we believe this is the most reliable way to achieve long-term returns while mitigating the risk of the serious, long-lasting declines that we really want to avoid.
Unless otherwise noted, all stock valuation and return data in this letter have been calculated by PCA based on data from Standard & Poor’s, Robert Shiller/Yale University, the Bureau of Labor Statistics, and stockcharts.com.