Diversification is a fundamental principle of investing – examples of the concept date back as far as Talmudic texts estimated to have been written over 3,000 years ago, stating “Let every man divide his money into three parts, and invest a third in land, a third in business and a third let him keep by him in reserve.” The diversification principle received a further boost in the recent era when Harry Markowitz’s Modern Portfolio Theory supported the notion that the volatility of a portfolio may be less than the volatility of its parts, such that the introduction of even a high-return high-risk asset to the portfolio may improve the portfolio’s risk-adjusted return (or even outright reduce its volatility). Yet at the same time, both the rabbis of the Talmud and Markowitz would probably agree that the first step of investing your assets is even more basic: make sure you own stuff that has a reasonable expectation of providing a useful return in the first place. Unfortunately, though, we seem to have lost sight of this rule in recent years!
The inspiration for today’s blog post comes from some recent conversations I’ve been having with a few planners about portfolio construction and how to select investments for a client. In our discussion, the planners were making the case that this past decade has not been a lost one for investors – despite the fact that the S&P 500 is back at price levels it first reached in 1998 and can barely manage to show a positive return, including dividends, since it peaked over 1,500 in March of 2000 – because a well diversified portfolio shows a higher return than the near-0% of the index. This point was further made recently in a New York Times Bucks Blog article by planner Carl Richards, who showed “Why It Shouldn’t Have Been A Lost Decade For Investors.”
While I’ll grant that it’s true there are other equity indices that have done better than U.S. large-cap stocks – as the Richards article notes – that’s actually the point. There are indices that have done better than the S&P 500. And it shouldn’t be surprising. Because the S&P was priced, in advance, to deliver incredibly mediocre returns!
Below is a graph of the Price/Earnings (P/E) ratio of US large-cap stocks, using the Shiller P/E10 data. The long-term average is approximately 16. Notably, though, any time the graph gets materially over a P/E ratio of 20, “bad stuff” happens – from the peak in 1904 that preceded the crash of 1907, to the peak in 1929 that preceded the Great Depression, to the peak in 1966 that preceded 16 years of price stagnation, to the astonishing peak in 2000 – so high, that the scale on the graph has to be increased by 50% just to FIT the peak. Sure enough, once again a tremendous peak in the overpricing of stocks led to a decade (or more?) of stagnation.
On the other hand, this wasn’t true of all types of equities. The graph below, from Hussman Funds, shows in blue the relative valuation of large cap P/E ratios to small cap P/E ratios over the past few decades. When the blue line is high, it means large cap P/Es are much higher than small cap P/Es, indicating that large cap is relatively “more overpriced”; conversely, when the blue line is low, large caps are favorably valued relative to small caps. The purple line, on the other hand, is the difference in returns between small cap and large cap over the subsequent 5 years. The results are pretty clear – the lines basically move together. When the blue line is high (large is expensive relative to small), the purple line is high (small outperforms large over the next 5 years); similarly, when the blue line is low (large is cheap relative to small), the purple line is low (small loses to large in performance over the next 5 years).
The point here is pretty simple. Advocates of diversification alone have made the case that the past decade was a perfect example of the value of diversification – by owning lots of different equity types, and not just the S&P 500, investors earned better than the near-0% return that the S&P has generated since the peak. The problem is, by looking at any basic measure of valuation, it was evident, in advance, that the S&P was going to have a meager long-term return, and that other asset classes – which had far more favorable valuation – were going to outperform. In other words, the success of diversification over the past decade was not the principle of owning lots of different stuff in your portfolio – it was the success of NOT owning a grossly overvalued investment.
In turn, this means that in reality, diversification only goes so far. Yes, it may have been difficult to know by just how much small caps would outperform large, or at what exact day, week, or month the outperformance would begin. But if you and your clients are truly long-term investors, the data was pretty clear – 2000-2010 was going to be a really bad decade for large-cap investors who bought and held over the time period, regardless of whether it took a few weeks, months, or even a year or few for the differences to manifest.
So the next time you’re considering investments for a diversified portfolio, remember: being diversified in an array of good investments that all have the opportunity for favorable return is one thing; owning every investment under the sun, regardless of its investment merits, still fills your portfolio with a bunch of mediocre investments that will likely deliver mediocre returns. Sometimes, the best path is not to “just” own a diversified portfolio; it’s to avoid owning bad investments. Don’t sacrifice your future at the altar of blind diversification. Every investment should stand on its own merits, too.
So what do you think? Was the past decade an example of the success of diversification, or simply a demonstration that valuation matters? Is it appropriate to follow the long-term impact of valuation, even if there’s a possibility of short-term underperformance? Should diversification be the only lens we use to consider an investment, or are factors like valuation relevant too?