Systematic Relative Strength

Ibbotson Associates has put out, for many years, the familiar multi-color mountain chart of stock returns beating bond returns. They contended that equity returns historically were far superior to fixed income returns, and this was used as fuel by many to hold large equity allocations in every market environment, regardless of client circumstances.

The last decade has not been kind to stocks, so now bond returns have exceeded stock returns over the last 40 years. Various commentators have begun to attack Ibbotson’s position by suggesting that retail investors should just own bonds, since they have done better than stocks. Now Ibbotson is firing back with new justifications for why stocks should do better than bonds going forward.

I’m imagining that when economists disagree on important topics like unknowable future returns that they would pull hair or hit each other with pillows if they weren’t able to write journal articles to fight with one another.

Ultimately, the whole argument is silly. No one knows the future, so it is impossible to know who will be right 10 years down the road. Is there some rule that you have to guess which asset will be better? Why not just hold stocks when they are strong, and switch to bonds (or some other asset) when stocks weaken?

“Since the beginning of the financial crisis, there have been two principal explanations for why so many banks made such disastrous decisions. The first is structural. Regulators did not regulate. Institutions failed to function as they should. Rules and guidelines were either inadequate or ignored. The second explanation is that Wall Street was incompetent, that the traders and investors didn’t know enough, that they made extravagant bets without understanding the consequences. But the first wave of postmortems on the crash suggests a third possibility: that the roots of Wall Street’s crisis were not structural or cognitive so much as they were psychological.

In “Military Misfortunes,” the historians Eliot Cohen and John Gooch offer, as a textbook example of this kind of failure, the British-led invasion of Gallipoli, in 1915. Gallipoli is a peninsula in southern Turkey, jutting out into the Aegean. The British hoped that by landing an army there they could make an end run around the stalemate on the Western Front, and give themselves a clear shot at the soft underbelly of Germany. It was a brilliant and daring strategy. “In my judgment, it would have produced a far greater effect upon the whole conduct of the war than anything [else],” the British Prime Minister H. H. Asquith later concluded. But the invasion ended in disaster, and Cohen and Gooch find the roots of that disaster in the curious complacency displayed by the British.

The invasion required a large-scale amphibious landing, something the British had little experience with. It then required combat against a foe dug into ravines and rocky outcroppings and hills and thickly vegetated landscapes that Cohen and Gooch call “one of the finest natural fortresses in the world.” Yet the British never bothered to draw up a formal plan of operations. The British military leadership had originally estimated that the Allies would need a hundred and fifty thousand troops to take Gallipoli. Only seventy thousand were sent. The British troops should have had artillery—more than three hundred guns. They took a hundred and eighteen, and, for the most part, neglected to bring howitzers, trench mortars, or grenades. Command of the landing at Sulva Bay—the most critical element of the attack—was given to Frederick Stopford, a retired officer whose experience was largely administrative. Stopford had two days during which he had a ten-to-one advantage over the Turks and could easily have seized the highlands overlooking the bay. Instead, his troops lingered on the beach, while Stopford lounged offshore, aboard a command ship. Winston Churchill later described the scene as “the placid, prudent, elderly English gentleman with his 20,000 men spread around the beaches, the front lines sitting on the tops of shallow trenches, smoking and cooking, with here and there an occasional rifle shot, others bathing by hundreds in the bright blue bay where, disturbed hardly by a single shell, floated the great ships of war.” When word of Stopford’s ineptitude reached the British commander, Sir Ian Hamilton, he rushed to Sulva Bay to intercede—although “rushed” may not be quite the right word here, since Hamilton had chosen to set up his command post on an island an hour away and it took him a good while to find a boat to take him to the scene.

Cohen and Gooch ascribe the disaster at Gallipoli to a failure to adapt—a failure to take into account how reality did not conform to their expectations. (my emphasis) And behind that failure to adapt was a deeply psychological problem: the British simply couldn’t wrap their heads around the fact that they might have to adapt. “Let me bring my lads face to face with Turks in the open field,” Hamilton wrote in his diary before the attack. “We must beat them every time because British volunteer soldiers are superior individuals to Anatolians, Syrians or Arabs and are animated with a superior ideal and an equal joy in battle.”

This long quotation about the failure to adapt is taken from Malcolm Gladwell’s article in The New Yorker on the psychology of overconfidence. (You can read the complete article here.) Just as the problems in the banking system were born of overconfidence, so too were the problems in asset allocation. The failure to adapt can have serious consequences whether in Gallipoli or your client’s portfolio.

Proponents of strategic asset allocation believed so strongly that their approach was superior that they ignored existing problems with the theory, such as instances of rising correlations in previous declines. Even now, after widespread failure in 2008, I continue to read articles about how strategic asset allocation performed like it was expected to. James Montier of Societe Generale recently referred to the efficient markets hypothesis as “the financial equivalent of Monty Python’s Dead Parrot. No matter how much you point out that it is dead, the believers just respond that it is simply resting!” The argument that 2008 was a one-time aberration and that things will soon be back to normal might be true—and it also might not be true.

If you assume for a moment, as we did, that strategic asset allocation might not be ideal, you would be motivated to do research and explore other options. When we used relative strength as the engine to do tactical asset allocation across a broad range of asset classes, we discovered that periodic trends generated returns that were not dependent on the assumption of lack of correlation between asset classes for their success. It would be entirely possible for assets that should be uncorrelated to be trending the same way at the same time, since we made no a priori assumptions. Rather than having to forecast forward returns to re-jigger allocations or to re-optimize correlation matrixes on a frequent basis, we designed the portfolio to flow with the trends. The result is a clean, simple, and understandable process that, we think, turns out to be as good a description of reality as strategic asset allocation.