[This blog was originally published by Summitas at www.Summitas.com]

One of the most remarkable discoveries of the modern portfolio theorists was the existence of a huge excess equity risk premium or ERP.  This premium is the incremental return investors demand to own stocks rather than bonds.

We all know that, looking back over nearly 200 years, investigators learned that, on average, stocks had beaten Treasury bonds by roughly 7% per year. And we’re all familiar with those Ibbotson charts showing that if we had invested $1 in stocks in 1926 we would now have a gazillion dollars, whereas if we had invested that same dollar in bonds we would now have $3.26. Or something like that.

Ironically, however, during most of this period of exceptional equity outperformance, people mainly didn’t own stocks – they either owned no securities at all or they owned bonds. True, there was a flurry of interest in stocks in the late 1920s, but this soon fizzled out (possibly having something to do with the fact that stock prices plummeting 80% between 1929 and 1932.)

For a very long time after that stocks continued to outperform bonds and investors continued not to own them. It wasn’t until the late 1960s that equities began to work themselves into investment portfolios in any major way – progress that was stopped in its tracks by the slow economic growth, high inflation, and general malaise of the 1970s.

Finally, in the mid-1980s, with an historic Bull Market in full steam, portfolios began to load up on equities. But it wasn’t just the Bull Market that got investors excited about equities – it was the studies showing a huge excess ERP.

Financial advisors –the main medium for transmitting Modern Portfolio Theory ideas from the universities to the real world – glommed onto this data and convinced their clients to own heavily equity-oriented  portfolios. Chief investment officers at large endowed institutions took it for granted that all portfolios with time horizons longer than few years should be equity-oriented, and they extended this thinking by concluding that institutions with infinite time horizons should have (almost) infinite equity exposures. By the mid-1990s (and again by 2006-2007), it was not at all unusual to find family and institutional investors with 80 percent – or even 90 percent – equity exposures, once private equity was included and hedge fund exposure was beta-adjusted.

At last, investors had seen the light and had begun to build portfolios designed to take advantage of one of the great Free Lunches in the investment world: stocks, which had always beaten bonds by far more than portfolio theory predicted. (According to the portfolio theorists, stock buyers should have been satisfied with about a 1% premium over bond returns.) But there was one problem: no sooner had investors begun to buy stocks in droves than bonds began to outperform stocks! Over the 30-year period ending with calendar year 2008, not only was there no excess ERP, and not only was there no ERP at all, there was actually a BRP (bond risk premium).

What was going on? Several things, I suggest. First, the portfolio theorists’ data was bad. It was bad because the early data suffered from very serious survivorship bias and because all the data suffered from a breathtaking sampling error: the US was hardly a “typical” capital market. Over the 200-year period of the studies the US went from nowhere to the world’s most dynamic emerging market to the most dominant economy in the history of the world. This didn’t happen in any other society and can never again happen in the US. Meanwhile, outside the US, evidence for an ERP is, to put it bluntly,  “nasty, brutish and short.”

Second, investors and their advisors misunderstood what an ERP meant. It didn’t mean that you got paid more to own stocks than bonds, it was simply the incremental return at which point an investor would be indifferent as between stocks and bonds. In an equilibrium model, investors would own portfolios roughly equally weighted between stocks and bonds and they would only over-weight stocks if they expected to receive an excess ERP.

Since an excess ERP doesn’t exist – and probably never existed in equilibrium – investors should re-think the equity-heavy portfolios they have owned for decades. The way to make money in the investment world is not to buy equities and hope for the best. The way to make money is to buy assets when they are selling below fair value and to sell them as they approach fair value: buy low, sell high.

A note about ERP Lite

When most advisors talk about the “equity risk premium” they are addressing a slightly different, more tactical idea. They are saying that, given the level of stock prices currently, they expect the forward-looking “equity risk premium” to be X percent. The “X” is often expressed as the reciprocal of the price/earnings ratio. For example, at the bottom of the 1973-74 Bear Market, with price/equity ratios below 7, an advisor might have suggested that the forward-looking ERP was 13 or 14 percent. At the top of the Bull Market in early 2000, with P/E ratios north of 40, the forward-looking ERP might have been 2 or 3 percent. ERP Lite is simply a slightly more quantitative way of saying that equity prices mean-revert, hardly an earthshaking proposition. The ERP I am talking about is a more fundamental idea and doesn’t mean-revert at all, but is a permanent condition of stock and bond prices.


Please note that this article is intended to provide interested persons with an insight on the capital markets and is not intended to promote any manager or firm, nor does it intend to advertise their performance. All opinions expressed are those of Gregory Curtis and do not necessarily represent the views of Greycourt & Co., Inc., the wealth management firm with which he is associated. The information in this report is not intended to address the needs of any particular investor.