Revisiting the ticking time bomb

Since the question “when is the crash going to happen?” is always asked, we thought it particularly timely to update the research we have done on the topic. Timing a crash can be a fool's errand, and fortunately such efforts are largely irrelevant if you are tail hedging (though they are quite relevant if you aren't). When tail hedging efficiently, the extreme asymmetries in payoffs, by definition, removes any need to time the top. But this doesn't mean that exercises in timing are without merit.

As we showed in previous research, without a doubt (or at least with over 99% confidence), bad things happen with increasing expectation when conditioning on higher Q ratios ex ante. That is, when Q is high, large stock market losses are no longer a tail event but become an expected event. Factoring time into the equation, and again based on history, the confidence interval around the median time would point to an expectation that the crash should commence right about now.

Monetary policy has proven to be very effective over the past seven years in elevating asset markets. However, its effect has been limited to the price of assets (the “title” to existing capital), but not the price of new capital. This differential is depicted in the Q ratio, where one can think of the numerator as representing the aggregate price of the stock market and the denominator as the aggregate book value. The higher the ratio, the further the stock market is priced relative to the reality of the underlying capital, and the greater the implied return on that aggregate capital above the average aggregate cost of capital. This ratio has always had its breaking point, much to the frustration of interventionist monetary policy, as the numerator ultimately crashes back to the denominator, rather than the denominator catching up to the numerator (a fact that Keynesians from Paul Krugman to James Tobin himself have considered a central puzzle of economics). Indeed, the continued deviation of this ratio from its long run historical average is something that both economic history and, best of all, economic logic dictate as unsustainable.

The question becomes how deviations and extremes in the Q ratio are ultimately corrected. The short answer is: they are corrected via the numerator, i.e., through corrections in the aggregate stock market value. The further the Q ratio has deviated from its long run historical average, simply put, the further the stock market has to fall to correct that deviation (this is what the market's homeostatic process does so predictably well).

There are regularities in the “stopping time” to the market's homeostatic correcting of extreme Q deviations, and as we saw recently in China, even massive interventions can't ultimately stop such corrections. An equity holder should be very aware of the current valuation environment, the magnitude of the drop that is to be expected, and the inherent cyclicality behind the amount of time between crashes. We are currently beyond the median amount of time, historically, before we would expect to see at least a 20% correction of the stock market (the numerator). Most importantly perhaps, the majority of the losses tend to happen in a concentrated plunge at the tail end of the path down to minus 20%. For instance, in just the last two months before the market passes through our 20% drawdown trigger, it typically (on average) has experienced a loss of nearly the entire 20%.

The very high probability of a crash currently implied by history flies in the face of a very low probability of a crash currently implied by the options market. The same beliefs that have pushed the market to extreme valuations have also returned option prices back to near record lows. If there is elevated risk in the equity market to the degree we have seen, counter-intuitively, it is not at all priced into options markets.

To use my favorite investing metaphor, the pot odds—the payoff, or the size of the pot relative to the price of calling—are very favorable compared to the hand odds—the likelihood of making the best hand; that is, we are getting the best of it.

In the recent August volatility (or in any other crashes we have seen), the tide turned both too surprisingly and too quickly for most to fully re-position until it was much too late. The future need not look like the past, but for an equity holder (or an opportunistic trader), the price of equity tail risk is not currently representative of that which has proven itself throughout history under similar (if not far less risky!) circumstances. How much further the rally stretches, whether another 10% or 100%, does not matter to an efficient tail hedger; it only adds to the expected magnitude and timing of a pending crash—which grows larger and sooner with each uptick in the stock market and tick of the clock—thus adding to the expected profitability and strategic advantage of the hedge.

Mark Spitznagel is the CIO of Universa Investments LP.