Stocks in the U.S. recently hit new all-time highs. Sounds great right? It’s good when stocks go higher! However, it’s not the record level of stocks that have Wall Street and investors postulating, it’s the value. After all, the price (i.e. the level) of the S&P 500 index tells us nothing about value. Ben Graham, the father of value investing, reminds us “that price is what you pay and value is what you get.”
The answer to the question of whether or not investors are currently paying too much for stocks (particularly in the U.S.) is ultimately based upon the amount of value that companies can produce in the future. This is an important concept because there is some connection between the price you pay today and the longer term returns you can potentially earn in the future. Let’s try to answer some important questions.
Is the stock market expensive?
The answer to this, as with most finance and economics questions, is it depends. You can get different answers depending upon the particular market and the metrics used to evaluate it. The widely cited Shiller P/E (named after Nobel Laureate Robert Shiller who created the measure) says U.S. stocks are somewhere between fair value and expensive depending on what time period is used for a historical average. The Shiller P/E is currently at 27x, which is high, but nowhere near the highs of 44x reached during the dot com bubble. As for markets abroad, the Bank Credit Analyst notes that valuations in Japan and the U.K. (which comprise approximately 40% of APCM’s int’l developed exposure) are more attractive on both an absolute and relative basis. Valuations in emerging markets are generally low as well.
Don’t get me wrong, by no means are equities cheap. As a result investors should not expect the 20% annualized returns that the S&P 500 has generated since the market bottom in 2009; back then the forward P/E was about 10x. Post financial crisis stock market returns were supported by earnings growth which averaged over 11% per year. Support from strong earnings helped get the market to where it is today and is the reason why valuation levels are not near extremes.
Having said that, as of late there has been six quarters of negative earnings growth primarily due to the combination of a precipitous drop in oil prices and a strengthening U.S. dollar. Going forward earnings are expected to rise next year which will be an important part of sustaining current S&P 500 price trends. Mid-single digit return expectations are reasonable when factoring in modest earnings growth, 2% dividends, and little to no P/E multiple expansion.
Should you sell all your stocks now and reinvest later when valuations are more attractive?
Today’s stock valuations along with low (and even negative) global bond yields have reignited market timing discussions. Completely selling out of the equity market is a drastic move for a long-term investor and is something that APCM does not recommend.
However, historical evidence indicates that P/E ratios and other valuation measures are somewhat inversely correlated with future long term returns. When the market’s P/E is high, stocks prices typically have muted returns during the next ten years. So why not implement a simple strategy of buying when the P/E is low and selling when it is high? Research shows that this contrarian strategy does not actually translate into superior performance for multiple reasons.
First off, the “average” that is used to determine if market valuations are expensive or cheap changes, and is really only known for sure in hindsight. The average Shiller P/E for the decade immediately following WWII was 12x and the average since 2000 has been over 25x. If it is the year 1945 and you are trying to determine if stocks are expensive you can only compare the current value with 12x and you would not have anticipated this richening trend. Solely relying on this contrarian strategy to time the market would not have worked over the past few decades. Similarly in the bond market, it would have been disastrous to wait for yields to rise back towards historical average over the past 20 or 30 years.
Another reason against timing the market based on valuation levels is that P/E ratios often look extremely attractive during a bear market or a financial crisis, but during these periods the outlook for earnings becomes very cloudy. If earnings don’t recover then stocks weren’t cheap after all. Forecasting earnings is difficult in normal times and even harder in a crisis. In the end, the idea of simply selling when P/E ratios are high and buying when they are low turns out to not be so easy!
It is best to avoid the major market timing sin of drastically moving in and out of stocks. At APCM we recommend moderation. If you are going to sin, sin a little. The price you pay for deviating too far from your plan can be detrimental and permanent. To reduce exposure to higher P/E S&P 500 stocks we purchased a variation of the typical market cap weighted index which quantitatively screens for stocks with higher quality earnings. Nothing drastic, just a tweak that lowered exposure to higher valuation stocks and should add some incremental value over time.
There are always many reasons to worry and with so many headlines calling for much lower prices it can be difficult to stick with your investment plan and process. Instead of asking if you should sell everything the more important question to answer is exactly how much equities should you own given your circumstances and today’s valuations?
Can starting P/E ratios predict future stock market returns at all?
According to research conducted by Vanguard:
Valuation metrics such as P/E ratios, have had an inverse or mean-reverting relationship with future stock market returns, although it has only been meaningful at long horizons and even then, P/E ratios have “explained” only about 40% of the time variation in net-of-inflation returns.
Because valuation is only one of the many factors that influence market performance, it is important to account for all variables when making investment decisions. During the portfolio construction process it is necessary to develop realistic long term return expectations based on structural trends like debt, demographics, and valuation levels. Starting valuations do matter, but you also need to incorporate today’s modest global economic growth, low interest rates, modest inflation expectations, demographic headwinds, technological advancements and productivity potential.
Callan, the investment consulting firm for the Alaska Permanent Fund has long term equity return forecasts of around 7%. This is in line with the assumptions APCM utilizes for financial modeling in client accounts. Vanguard notes “that today’s low dividend and Treasury yields are in part, associated with lower expected inflation.” Thus, the likelihood of a balanced portfolio achieving real returns in excess of those since 2000 are higher despite lower nominal return expectations.
I know some readers are saying “I have a short time horizon” so thoroughly examining and quantifying risk exposure is particularly important. Your portfolio should be in line with your time horizon and maybe you will own little to no stocks at all. In the end, valuations do matter but a successful investment experience hinges upon the combination of multiple factors that determine how much stocks and bonds you should own. Volatility in the markets is inevitable, but don’t let emotions and headlines drive your decisions!
Brandy Niclai, CFA®
CIO Multi Asset Strategies