Systematic Relative Strength

“ALICE laughed: ‘There’s no use trying,’ she said; ‘one can’t believe impossible things.’ ‘I daresay you haven’t had much practice,’ said the Queen. ‘When I was your age, I always did it for half an hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast.’”

—-Lewis Carroll, Alice in Wonderland

This is the beginning of an article from The Economist entitled “Something Doesn’t Fit.” The piece goes on to discuss something that is happening now in the capital markets, but according to handbooks on strategic asset allocation, isn’t supposed to happen. Both gold and Treasury bonds are doing well. Market lore says gold is supposed to do well during times of inflation and Treasury bonds are supposed to do well during deflationary periods. As The Economist points out:

There is something remarkable about this combination. You would expect the performance of gold and Treasury bonds to be inversely correlated. When gold was at its real all-time high in 1980, the ten-year Treasury-bond yield was 10.8%. Fixed-income investors had suffered years of negative real returns in the 1970s.

Adding other assets to the mix does not solve the dilemma. David Ranson of Wainwright Economics, a consultancy, has examined how gold and corporate-bond spreads have worked as a guide to asset allocation since 1978. The gold price is an indicator of inflationary pressures. Bond spreads are an indicator of growth: investors are happy to take the risk of owning corporate bonds, and spreads come down, when the economy, and thus business revenues, are strengthening. Over the year to end-May gold was up by 25% while corporate-bond spreads narrowed sharply. History suggests that such a combination should be bad news for government bonds.

One of the more striking things about financial markets is their repeated ability to do impossible things; impossible at least in the finance textbooks. It turns out that a lot of the things history suggests just don’t pan out.

This is one of the reasons that a trend-following approach based on relative strength can be so helpful. Trend following does not make any a priori assumptions about how the world works or about how things are correlated. It just goes with the flow. A trend follower has no difficulty reconciling gold and Treasury prices going up at the same time. What is, is.

Models that are based on historical correlations are prone to break when (not if) the correlations go haywire. Relative strength models are designed to be adaptive. This is one of the primary reasons we choose to have all of our investment processes driven by relative strength.

As relative strength flows from one asset class to another, an adaptive simply model changes the holdings. The changes in relative strength between asset classes are generally somewhat durable. The flow in relative strength, after all, usually has a rational cause. In the case of the recent conundrum, The Economist suspects a couple of possible causes:

So what explains the current situation? Both gold and Treasury bonds could be classed as “safe haven” assets that investors buy when they are risk-averse.

So the safe haven trade is one possibility. The next possibility is very different:

Martin Barnes of Bank Credit Analyst, a research firm, points out that the direction of official policy (low rates, quantitative easing, big deficits) looks inflationary but the economic fundamentals (a big output gap, sluggish credit growth) look deflationary. Faced with this dichotomy, investors who buy both Treasury bonds and gold are not displaying cognitive dissonance. They are just hedging their bets.

In the absence of being able to figure out what will happen next, perhaps investors are just hedging their bets-otherwise known as “having no clue.”

From a fundamental point of view, these choices are quite different. From the point of view of a relative strength investor, the explanation is less important than what is actually happening. Impossible things will contine to happen, but relative strength will just continue to adapt.