“We thought it was the eighth inning, and it was the ninth. I did not think it would go down 33 percent in 15 days.”
In April 2000, Stanley Druckenmiller, who managed the phenomenally successful Quantum fund for George Soros, called it quits – saying “I overplayed my hand” in technology stocks. The Nasdaq composite had suffered the first blow of what was to become a much deeper loss, with the index losing an additional 70% by 2002 low. Still, getting out was a good idea in hindsight.
With one of the best records in the industry, Druckenmiller has expressed increasing concerns recently that “all the lobsters are in the pot.” A few weeks ago, he said of equities that he holds “the smallest positions I’ve had,” and warned “a necessary condition for a financial crisis, in my opinion, is too loose monetary policy that encourages people to take undue risk.”
Floyd Norris of the New York Times reported in 2000 that Druckenmiller was actually the second high-profile hedge manager to call it quits in that cycle. The first was Tiger Fund’s Julian Robertson, who had lagged the advance because he (correctly, in hindsight) viewed technology stocks as vastly overvalued. The article quoted an analyst saying “The moral of the story is that irrational markets can kill you. Julian said ‘This is irrational and I won’t play,’ and they carried him out feet first. Druckenmiller said ‘This is irrational and I will play,’ and they carried him out feet first.”
As I noted a few weeks ago “The problem with bubbles is that they force one to decide whether to look like an idiot before the peak, or an idiot after the peak. There’s no calling the top, and most of the signals that have been most historically useful for that purpose have been blazing red since late-2011.” My impression remains that the downside risks for the market have been deferred, not eliminated, and that they will be worse for the wait.
For those inclined to understand the unfinished half-cycle since 2009 in its full context, rather than dismissing objective evidence and discipline that have proved valid in complete market cycles throughout history, several prior commentaries may be helpful (see The Diva is Already Singing, The Elephant in the Room, and Aligning Market Exposure with the Expected Return/Risk Profile). This would be a terrible moment to lose that discipline.
Estimating the Risk of a Market Crash
I want to preface this section by emphasizing that our central investment outlook is driven by none of it. That is, a defensive outlook here does not presume, require, or rely on a market crash. Our ongoing discipline is to align our investment outlook with the market return/risk profile that we estimate on the basis of a broad ensemble of evidence that we can test historically and validate in out-of-sample data. That outlook will shift as that evidence shifts, period. In contrast, what follows is more of an intellectual curiosity – something that we don’t rely on (and neither should you), but that also highlights the fascinating correspondence of market dynamics since about 2010 with the existing literature surrounding market bubbles and crashes.
Imagine a game that offers you one of two payoffs. If you choose to play, you’ll get a return of $W with probability p(win) and you’ll lose $L with probability p(lose). What’s your expected return? Simple:
Expected return = $W x p(win) + $L x p(lose)
For example, if $W is $20, $L is -$10 and p(win) = 60% (so p(lose) is 40%), your expected return is $8.
If you’re neutral to risk, you’ll play the game only if the expected return is greater than zero. If you’re risk-averse, you need the expected return to exceed zero by even more. If you’re risk-seeking (which often occurs with small amounts of money like lottery bets), you might even play the game even if the expected return is less than zero, but probably only if the absolute amount of possible loss $L isn’t terribly large.
Continue reading here.