|A sharp decline in the Chinese stock market, along with worries about the domestic sub-prime lending market, rattled investors a bit in the first quarter. Stocks rebounded in March to post modest gains in the first quarter. Since last August, stock market returns have been quite good.
In assessing the possible impact of an economic slowdown on earnings and P/E multiples, we looked at various scenarios in which earnings were weak. to gauge return prospects. This work suggests that the market is discounting some, but not all, of the risk of a sharp slowdown in earnings.
We continue to own tactical positions in short-term emerging-markets bonds and commodity futures, both of which are funded from our domestic bond allocation, and remain overweighted to large-caps relative to small-caps.
Quarterly Investment Commentary
Equities rebounded in March, bringing returns back into positive territory for the year. Both the large-cap S&P 500 and the small-cap Russell 2000 iShares gained 1.1% for the month. Year to date through March, the S&P was up 0.6%, while the Russell 2000 gained a more robust 1.9%. Foreign equities had a strong month, gaining almost 3%, and posted a first-quarter return of almost 4%. On the fixed-income side, Vanguard Total Bond Market Index Fund was flat in March, but was up 1.4% for the quarter. Emerging markets short-term bonds and commodity futures both performed well in the first quarter.
What’s Rattling the Market?
After enjoying a seven-month run of mostly climbing stock prices, investors were spooked by the big drop in the Chinese market in late February, and the subsequent dip in U.S. stock prices. It appeared that investors were concerned that the stock market decline in China might signal slowing growth there, which would have an effect on the rest of the globe. Given that both the U.S. and Chinese economies have been expanding for a long time, it is realistic to expect that slowdown will occur at some point, and any evidence of a slowdown will impact the markets. Still, we were a little surprised to see other markets react as they did to the decline in China’s market. The Chinese A-share market is mostly limited to domestic Chinese investors and is their only equity investment option. As a result, short-term swings in that market can be driven by lots of factors beyond fundamentals.
While the one-day decline in the Chinese stock market spooked investors around the globe, U.S. investors have grown increasingly concerned about the potential impact of a weak housing market. In particular, problems in the sub-prime lending market have garnered lots of attention.
Most of us—especially those living in areas with very high housing costs—are aware that lenders have pushed the envelope in recent years to come up with loans to enable people to buy homes they would otherwise be unable to afford. In so doing, many of these buyers have stretched themselves financially to the point where they have no margin for error. With rates having climbed, and low starter rates and temporary interest-only terms winding down or expiring, defaults among sub-prime mortgages have climbed sharply. Consider that loans in this segment accounted for 24% of originations in 2006, defaults in this sector are in the 13% to 14% range, and some sub-prime lenders have either experienced big financial losses or gone out of business. Firms we respect like PIMCO have posted that this is a meaningful source of risk to the housing market, since they are the first rung on the housing-market food chain. So on its face, rising delinquencies in a high-growth part of the market could be a serious concern.
Looking more closely, however, the picture is not as clear. The growth in sub-prime originations has moved in lockstep with a decrease in Federal Housing Authority loans (FHA borrowers are typically first-time home buyers who are unable to make a meaningful down payment). As of year-end 2006, more than 76% of all loans outstanding were still prime loans. This means that even if the combined sub-prime/FHA sector had a default rate of 20%—well above the prior peak in 2002—fewer than 5% of outstanding loans would be impacted. While this would be enough to cause pain, it would probably not be enough to result in a disaster for the economy.
Earnings Will Be the Key
Developments like a slowdown in China or a crunch in sub-prime lending would definitely impact the economy, perhaps significantly. The real question, though, is not whether they will have an impact, but rather how much of that potential risk is already reflected in stock prices. This is a question we regularly ask ourselves, and we can take a careful look at market history to come up with some assumptions to use in the context of the current environment. If the market has already discounted these events, then maybe we needn’t worry about them as much.
The way we typically tackle this topic is to imagine what a bearish scenario might look like in terms of earnings and P/E multiples for the S&P 500. Since the price of the S&P 500 is a function of the underlying earnings and the P/E multiple applied to those earnings, we can simply take a bearish earnings number and put a best-guess P/E multiple on top of it to arrive at a prospective target price for the S&P 500. We can then compare this target price to today’s price to see what kind of return this implies over a multi-year period. If the return is decent, it tells us the market is already discounting a negative scenario.
We’ll start with earnings. How bad is bearish when it comes to earnings? There’s no magic answer to that, but historically, it has been very rare for nominal earnings to end any five-year stretch at a lower level than they began. That gives us some context, but we’re not willing to rely solely on history—we also want to do the math. Earnings are a function of how much money companies take in (revenues) and the profit margin on those revenues. If we assume corporate revenues will increase at roughly the rate of inflation over the next five years (which is pretty bearish), and at the end of that time period we apply a historically average profit margin (which implies a material decline from the current level of profit margins), we come up with an earnings number for the S&P 500 that is somewhere in the low $80s. This is around the level of current earnings, and thus would comprise a five-year earnings scenario that is in line with prior periods of extended and noteworthy earnings weakness (in other words, a pretty bearish earnings scenario).
What would this mean for stock prices? To get at that, we must put a multiple on top of those earnings (see table on next page if you want to cut to the chase). One might assume that a poor earnings environment would result in an overall glum mood for investors—one where they weren’t willing to pay as much for stocks (meaning the market would trade at a relatively low P/E ratio). Historically, however, P/E multiples have been higher at the earnings low point than the historical average P/E and higher than the P/E during the prior earnings peak. The reason is that the market is forward looking, and even when things are really bad at a point in time, investors are already looking ahead to what comes next.
Using data from past earnings troughs, we can make a table that shows us the expected return from equities in some scenarios, assuming a starting point of 1420 for the S&P 500 (it’s level on March 30), and including the impact of dividends.
For us to say that the market is already discounting a bearish scenario, we’d want to see decent stock market returns even if that scenario comes to pass. There are a lot of ways we could define “decent,” but one good one is that stocks do better than bonds. The shaded regions in the table represent pre-tax returns that are equal to or better than the potential return range for bonds over the same time horizon.
When looking at the probable returns in a bearish environment, it does not appear that the market is fully discounting that scenario. A five-year period of flat earnings is not common, but is far from unprecedented. Given the risks posed by the housing market, the extremely high level of profit margins (which will be difficult to sustain), the fact that current earnings are far above their trend growth line, and the fact that the economic cycle is already longer than average, a bearish outcome is not a stretch.
However, these are all cyclical risks, which, while more notable than average, are not enough to cause us to run for the exits. Cycles are normal, and the challenge is in evaluating how big those risks are, and in differentiating cyclical risks from events or developments that are likely to cause lasting damage. One thing we have learned is that the world is a scary place. In the past 20 years, we can point to many events—including the stock market crash in 1987, fears of a major banking collapse, wars in the Persian Gulf, the Asian currency crisis and near-collapse of Long-Term Capital Management, panic over Y2K, and the deflation scare in 2002, to name just a few—that were serious causes for concern and in some cases led investors to panic in the short term, but that didn’t end up having lasting negative consequences. The table shows us some pretty unpleasant scenarios for stocks, but these are just possible outcomes on the negative side and we don’t think they are likely enough or damaging enough to justify avoiding stocks. On the other side, even modest earnings growth and an average P/E multiple results in returns that are better than bonds, so remaining invested in stocks still offers a likely reward. For example, if earnings simply grew in line with inflation (say, 3%) and we put the 25-year average P/E multiple of 20 on top of those earnings, we’d get a 9% return from equities. We also believe that our managers can add value above the market—or more specifically their benchmarks—over longer time frames, and that would be an additional source of return for our clients.
Our analysis might seem at odds with what we’ve said about valuations being attractive. Right now, valuations look good when computed based on the current price of the S&P 500 and the current earnings of the S&P 500. However, valuation work is different from scenario forecasting, although both are important and related to one another. As we have mentioned in many of our prior communications, we view risks—including a meaningful slowdown in earnings—as being above average, and the data in the preceding table represents one way of evaluating how much of those risks are already priced into the market. The answer seems to be that valuations are probably in the right ballpark, but they don’t fully price in the possibility of a major earnings slowdown over the next five years.
Sometimes months and quarters go by with only incremental changes to our overall assessment of asset classes and the broader economic environment. It can be helpful to look over longer periods to see the gradual evolution in our thinking. What do we see if we look back from this point? At the end of the second quarter of last year we were coming off a difficult stock market environment that resulted in attractive valuations. In the period since, stocks have generated strong returns. Meanwhile, earnings growth has begun to slow, and developments that we think are probable—such as declining profit margins, further declines in earnings growth, and an overall economic slowdown—have moved closer. The result is that while current valuation metrics show stocks as undervalued, we think a forward-looking assessment suggests a more tempered view.
As always, we will continue to challenge the assumptions that underlie our view, consider new information as it becomes available, and stay intellectually honest in making well-reasoned investment decisions for our clients. We appreciate your confidence and trust.
Jon Houk, CFP®