"Lift-Off" Lies And The Fed's Reputational Risk

Submitted by Ben Hunt via Salient Partners' Epsilon Theory blog,

Every Fed watcher’s favorite word these days is “lift-off”. As if the Fed’s first rate increase, whenever that comes to pass, is the ignition of some giant Saturn V rocket that will inexorably carry interest rates up, up, and away. Please. This is Narrative creation … really, Narrative abuse … of the first order. The next time you read or hear someone use the word “lift-off”, I’m begging you to remember Jim Mora’s classic press conference when he was asked about the Colts’ chances of making the play-offs, because it’s a dead ringer for what Janet Yellen is saying in her heart of hearts.

You think Yellen is thinking ahead to a rates lift-off? Really? The reputational risk and future payday risk associated with this first rate increase is astronomical for Yellen, much less a series of rate increases. Yes, wage inflation is slowly staggering up off the floor after being knocked unconscious for the past five years. Yes, it would be nice if the Fed were not scared to death of spooking the bond market (thank you, Captain Obvious … err, I mean former Fed governor and current BlueMountain millionaire Jeremy Stein). But the notion that we’re either off to the races in the real economy or that Yellen woke up on Monday with a political and personal deathwish to tell off the bond market is just ludicrous.

The Fed wants to raise short rates to put some bullets back in the exhausted gun of ordinary monetary policy, and that gives a totally different meaning to the notion of a rate increase today than in, say, 1994. Since when was the Fed concerned about getting in front of – gasp! – 2.3% wage inflation? No, the Fed wants to reload with conventional ammo before the next external shock or the next inventory-led slowdown, which is all smart and good and thoughtful, but they’re being forced to reload while the battle is still raging. That creates a very different decision-making and implementation path for rate increases than any historical corollary of the past 60 years, which means that any investment conclusion based on those historical corollaries is almost certainly a category error, the worst possible methodological mistake you can make. Sorry, but my crystal ball is still broken, and I think yours is, too.

Second, I want to call attention to the crucial distinction in logic (and gambling) between probabilities and odds. Will the Fed raise interest rates one day? Sure. There is a 99.999% probability that this event will occur over a long enough time period, in exactly the same way there is a 99.999% probability that a Triple Crown winner will materialize over a long enough time period. Is it profitable to attempt to predict when that day will materialize? No. The payoff odds associated with any specific meeting being THE meeting of the Fed rate increase will inevitably be poor, in exactly the same way the Belmont Stakes betting odds on American Pharoah (it kills me to misspell this word) and every other Triple Crown candidate over the past 35 years were poor. Why? Because when human beings pay great attention to any probabilistic event, they ALWAYS over-estimate the likelihood of that event occurring. This is one of the strongest findings in all of behavioral economics, and it runs rampant at both the track and the stock market. Over-estimated probabilities mean bad odds. To wit: just because American Pharoah won the Belmont stakes, and if you bet on him to win you were absolutely right, that doesn’t mean it was smart to make a bet at 3:5 odds.

I’ve got a lot more to say on the issue of investor attention, as it’s one of the most interesting areas of modern academic research on markets, but for now I’ll leave you with this. We will never know the approximately “true” probability of American Pharoah winning the Belmont stakes, because we can’t repeat the experiment a dozen or so times. What we know with near certainty, however, is that the expressed odds of 3:5 for the single experiment were worse than whatever the true probability might have been. What we also know with near certainty is that pundits LOVE infrequent probabilistic exercises like Fed meetings or Triple Crown races, and their attention magnifies investor attention in a profoundly unhealthy way. Why do pundits swarm like flies around these events? Because they are tautological exercises – without a repeated experiment, you can NEVER be proven wrong in your pre-race or pre-meeting assessment of probabilities – and that’s a great business model for them. Of course, that means it’s a terribly weak signaling model for you. My view: once you hear more than a handful of Missionaries arguing about any sort of well-publicized probabilistic event, whether it’s a horse race or an election or a Fed meeting or a jobs report or whatever, it’s un-investable. Run away. Far better to adapt to whatever the outcome turns out to be as part of a prepared strategy than to fling yourself from pillar to post in an attempt to anticipate an overly-examined, poorly-estimated, totally-gamed event. That’s not the sexy way to invest. It’s not the heroic way to invest. But if there’s one Epsilon Theory lesson that I never get tired of repeating, the Golden Age of the Central Banker is a time for investment survivors, not investment heroes.