Magic Trees

Imagine an island on which magic trees grow. These trees, as it turns out, are exactly like the trees everywhere else, except for three things. First, these trees never die. Second, the trees always grow to exactly 100 feet tall eventually. And third, to pass time on this boring island the villagers place bets on which specific trees in a given acre of land (on which all trees were planted at the same time) will grow the fastest over the next ten years.

Right after an acre is planted, there is much activity that can only be termed purely speculative. Without any obvious difference in the first shoots, the villagers place their bets based on which of the tree-market brokers tells the best story about a particular tree. These brokers do tend to change their minds frequently, however, so it turns out that there is rapid trading.

After a few years, some trees have clearly started to grow faster than other trees. Villagers tend to invest more on these trees that have “momentum.” And this trend continues, because the further in the lead a given tree is over its rivals, the more momentum it clearly has. There is, however, a class of investors who like to invest in the smaller trees, since the bet is on the rate of growth over time, and these investors think that the smaller trees are likely to revert to the mean (indeed, because all of these magic trees end up at the same height, they are correct on average).

The villagers who “own” the taller trees are generally happy, since their trees are “in the lead.” They don’t much care for the villagers who “own” the smaller trees, because they think these folk are just negative ninnies. The value-villagers are fairly confident, though, because they understand the math; and many of them are dismissive of the momentum-villagers and call them “lemmings.”

The odd thing is that both of these “investors” have their time in the sun. Early on in a tree’s growth pattern, the ones quickly out of the gate do tend to grow more rapidly. Consequently, momentum is a viable strategy. But the bigger the lead gets for these trees, the worse the bet becomes that the rate of growth will continue. Once the tree reaches 99 feet, for example, there are not many ways that it can beat a 20-foot tree going forward (remember – all of these magic trees always grow to be exactly 100 feet in time). And yet, the momentum-villagers remain true to their investment style, saying “the 20-foot tree must just be sick. And it can’t get as much sun because the 99-foot tree is shading it. The 99-foot tree may only grow 1 foot over the next ten years, but the 20-foot tree might not grow at all.” And, after all, it is fun to have your bet on the biggest tree in the forest. However, the value-villagers almost always win that bet.

This allegory isn’t really about growth versus value in equity investing. It is about asset class performance and, more specifically, the performance of equities versus commodities. There is a significant amount of history to suggest that over long periods of time, the average growth rate of equities and the average growth rate of commodity indices is approximately equal. This happens because the basic sources of both asset classes are fairly steady: aggregate economic growth, in the case of equities, and collateral return plus rebalancing effect, plus some other smaller sources of return, in commodities. So regardless of what you think about equities or commodities, in general when equities are dramatically outperforming commodities, you should be selling them to buy commodities, and vice-versa. But that isn’t how most investors bet. Most investors make up a story about why the tree is stunted, and will never grow, will never catch up, and then turn to bet on the tall tree.

It is a mistake.

It is a mistake, though, that the Street actively encourages because where the broker makes his money is on frenzy. Rallying markets tend to produce more volume and excitement, and that means more money for the broker. This is especially true when the market is equities, since there isn’t much underwriting of commodities to be done but there is quite a lot of underwriting of new equity issues.

Frankly, this over-exuberant cheerleading sometimes results in lies being disseminated to investors. Consider the following snapshot of a Bloomberg page describing the characteristics, including the P/E ratio, of the Russell 2000 index. You will see it says the Price/Earnings ratio is about 18.41. We all know that means that if you pay $18.41, you will get a set of stocks that will have $1 in earnings, collectively. Right?


Well, the following chart is also from Bloomberg, and you get it if you type RTY<Index>FA<GO>. What it says is that the Price/Earnings Ratio is actually 54.86, which means that your $18.41 actually only gets you $0.33 of earnings! What’s going on here? Well, the next line shows you that what Bloomberg considers the “real” P/E ratio – important enough to have on the front page – is really the Price/Earnings ratio if we only count the positive earnings.


So, that $18.41 does in fact get us $1 in earnings. Wall Street doesn’t want us to focus on the fact that it also gets us $0.67 in losses, so that the net is only $0.33. Because surely, those losses were one-time events, and obviously all 2000 companies will make money next year, right?

It may be that stocks are a great bargain here, and that multiples will expand further, the economy will surge in a way we haven’t seen in a decade or two, in an environment of tame inflation but ample liquidity. If that is the case, then equity investors will win over the next five years because they’re betting on the trees that have grown the most in the last two years. But in all other cases, commodity indices should grow faster than stocks as both revert to their long-term growth rates.

    May 7, 2013 at 5:16 pm

    Good one Mike

  2. Jim H.
    May 7, 2013 at 6:58 pm

    If one uses only four quarters of earnings, losses have to be excluded. Otherwise the wildly erratic P/E ratio blows up when earnings cross over to the negative side. #DIV/0!

    That’s why Robert Shiller invented the 10-year average P/E ratio called CAPE. Others use price-to-peak earnings, which also doesn’t blow up.

    Of course, Wall Street has its own permabullish twist: forecasted earnings, which of course never go negative (although they regularly prove to be wildly off the mark).

    Bloomberg’s ‘positive only’ earnings measure is simply innumerate gibberish. Time to pull the plug on that overpriced, lying terminal.

  3. HP Bunker
    May 7, 2013 at 7:56 pm

    To play devil’s (equity’s) advocate, Mike, would it not be logical to assume a lesser correlation between equities and commodities over time as developed economies increasingly come to be dominated by services rather than manufacturing and construction? “Products” like education and health care consume very little in the way of commodities, but consume an ever-growing share of developed nations’ GDP, so it seems logical that equity index earnings could increasingly diverge from commodity inputs over time.

    • May 7, 2013 at 8:33 pm

      There’s no real correlation between commodity indices and equities anyway. Since 1980, the correlation of monthly changes in the GSCI and S&P is 0.16. It’s only a recent phenomenon that they have been correlated. There’s no real reason they should be.

      • HP Bunker
        May 7, 2013 at 9:04 pm

        Ok, but then what I’m saying is I’m not sure why we would expect that the stock market and commodity indices would in the long term have the same relationship as has been the case historically. There are lots of reasons to think the US stock market is overvalued, but I’m not sure this (relatively low commodity prices, or the short trees in your example) is one of them. (Although the poor performance of commodities over the last few years might say something interesting about the true performance of China’s black box economy, which has a much larger share of GDP consumed by manufacturing and fixed investment and therefore should be more closely correlated with commodity prices than the US).

      • May 8, 2013 at 7:43 am

        Oh, I understand what you are saying now. No, I’m not suggesting any causality here. It isn’t BECAUSE commodity prices being low that stocks are high. Those are separate observations. One market has gone nowhere but down for two years; one has gone nowhere but up; but in the long run, they will reach about the same point. Which one is likely to be regressing to the mean and growing more slowly in the future, and which is likely to be regressing to the mean and growing faster? That’s my only point.

  4. May 9, 2013 at 5:33 am

    Michael, riddle me this, please. How can equities and commodities returns in the long run be about the same, if stocks have dividends re-invested , and commodities obviously don’t?

    • May 9, 2013 at 6:13 am

      Because commodities returns aren’t sourced from the same place as equity returns. Equities get returns from growth, which produces those dividends as well as higher earnings. Commodities get returns from rebalancing of a highly diverse portfolio, plus collateral return, plus normal backwardation and a few other sources. It happens that these returns over time add up to about the same as the equity returns (assuming that equity returns don’t decline because the economy’s natural growth rate declines).

  1. May 24, 2013 at 1:25 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: