Value is arguably the leading exhibit in the factor zoo. Quantifying the value factor is challenging, however, for multi-asset class portfolios for at least two reasons. Commodities aren’t easily valued because oil, gold, etc. don’t generate earnings or dividends. Even when there are cash flows to measure, putting a broad set of assets on a level playing field is difficult, perhaps impossible, if we use traditional valuation metrics, such as book value, price-earnings ratio, etc. Measuring valuation for real estate investment trusts (REITs), for instance, requires a different procedure vs. stocks, which is different from bonds. A solution (or at least a partial solution) that applies to everything is estimating valuation using return.

A 2013 paper by AQR Management’s Cliff Asness and two co-authors (“Value and Momentum Everywhere”) studies the value and momentum factors and outlines the logic for estimating the former with 5-year performance:

Momentum is straightforward since we can use the same measure for all asset classes, namely, the return over the past 12 months… For measures of value, attaining uniformity is more difficult because not all asset classes have a measure of book value. For these assets, we try to use simple and consistent measures of value… For commodities, we define value as the log of the spot price 5 years ago…. Fama and French (1996) show that the negative of the past 5-year return generates portfolios that are highly correlated with portfolios formed on BE/ME, and Gerakos and Linnainmaa (2012) document a direct link between past returns and BE/ME ratios [ratio of the book value of equity to market value of equity].

No one should confuse 5-year returns as a perfect solution, but it’s a useful first step for deciding how a broad mix of assets compare on a valuation basis. With that in mind, let’s crunch the numbers on the major asset classes based on a set of proxy ETFs for trailing 5-year performance. By that standard, valuation varies widely across markets. At the highest valuation extreme: US stocks, based on Vanguard Total Stock Market (VTI), which is currently posting a 13.9% annualized total return. On the flip side: broadly defined commodities via iPath Bloomberg Commodity (DJP), which is suffering with a 10.9% annualized loss through Apr. 25.

A simplistic interpretation of the chart above is to assume that the assets with the highest trailing return have the lowest expected return, and vice versa. That’s a reasonable first approximation, but it’s hardly a fail-safe way to select weights for a multi-asset class portfolio. A prudent asset allocation requires deeper analysis. But the numbers above suggest a number of possibilities.

One path is to consider how a 5-year-return-based set of weights compares with Mr. Market’s asset allocation (weighting assets based on market valuation). As an example, assume that the US stocks represent 30% of the global asset pool as measured by market values. Let’s also make a leap of faith and assume that US stocks are overvalued by 13.9%, as per the chart above. That implies that we should reduce Mr. Market’s US equity allocation by roughly 4 percentage points (30% * 13.9%). For assets with negative returns, the same formulation can be applied to raise the market-value-implied weights.

That’s just a toy example but it provides a basis for thinking about how to build a quantitative-based framework for a value-based dynamic asset allocation strategy. Such a strategy can be modified further by incorporating other factors into the mix. A value/momentum framework, for instance, is one possibility.

Keep in mind that while using a 5-year return (or something comparable) is an easy way to quantify valuation, it’s only a rough approximation. Accordingly, the methodology is a tool for routinely monitoring a motley mix of assets. In other words, this is where you should start your analysis on valuation for a multi-asset class portfolio.

Let’s also remember that the value premium, while stacking up an impressive run over the long term, can be a dog in the short run, and perhaps the medium run too. US equities appear to be richly valued these days, based on several yardsticks (including the cyclically adjusted price-earnings ratio). But that’s no assurance that stocks won’t remain overvalued for months or years to come and continue to deliver higher returns along the way.

“There is no clear message from all of this,” Professor Robert Shiller of Yale wrote earlier this month. “Long-term investors shouldn’t be alarmed and shouldn’t avoid stocks altogether. But my bottom line is that the high pricing of the market — and the public perception that the market is indeed highly priced — are the most important factors for the current market outlook. And those factors are negative.”

The one thing that’s clear: the 5-year valuation proxy agrees with Shiller’s analysis — US equity valuation appears to be high. Deciding if that’s timely information for making investment decisions is a separate question.