Last year, scientists in central Italy witnessed a strange phenomenon. On April 1, toads near the town of L'Aquila fled from their breeding sites en masse. Five days later, a 6.3-magnitude earthquake struck L'Aquila, killing hundreds.
The toads, it seems, sensed the coming quake.
If only there was an amphibian that could predict tremors in the market. Days before financial crises strike, we could flee, like those Italian toads, to the shelter of bonds, and then return to stocks before prices settle again.
MIT Sloan School senior finance lecturer Mark Kritzman, who is also president and chief executive of Windham Capital Management, a Cambridge, Massachusetts financial advisory firm, thinks he has found such a warning signal – not in a toad, but in a new statistical model called the absorption ratio. The absorption ratio predicts systemic risk by measuring how tightly markets are coupled, and thus how vulnerable they are to the spread of negative shocks. It is calculated by dividing the volatility of a fixed portfolio of assets by volatility across the entire market. When systemic changes in the market correlate more closely with changes in value of the fixed portfolio, the absorption ratio increases, and so does predicted market fragility.
Investors may be able to use the absorption ratio as a market timing mechanism. While tightly connected markets do not always lead to declines in asset values, when market drops do occur, a spike in the absorption ratio nearly always precedes them. A backward-looking analysis conducted by Kritzman found that an individual trader could have escaped both the 2001 dot-com meltdown and the 2008 financial crisis unscathed if he had shifted assets from stocks to bonds in proportion with the absorption ratio as it approached its maximum value, then moved back to stocks as the ratio fell.
If the Fed had known about the absorption ratio several years ago, Kritzman says, it would have noticed the national housing bubble. Between January 1997 and September 1998, the housing market absorption ratio jumped from a relatively average 64.12 percent, to 77.68 percent, just as the national housing bubble began to inflate. It went on to peak several times over the next decade, reaching an all-time high of 93.19 percent in March 2008. Once home prices began to stabilize in 2009, the ratio started falling again.
The absorption ratio seems to predict financial turbulence – periods when prices behave in an unusual fashion relative to historical patterns. In a study of the MSCI USA stock index covering the period between July 1, 1998 and June 30, 2009, Kritzman found that the ratio tended to shift upwards 40 days prior to major turbulent financial events such as the global financial crisis and the tech bubble. It usually then continued to rise, peaking a few days after the conclusion of the event.
The absorption ratio also tracks one measure of international contagion (financial market turmoil that spreads across countries). In 2008, Anna Pavlova of London Business School and Roberto Rigobon of MIT Sloan School released a statistical model showing historical asset prices relative to levels of global financial constraint in a discussion paper for the Center for Economic Policy Research. When measured against a common currency, changes in the absorption ratio and the contagion model track each other nearly perfectly, underscoring the close connection between global financial contagion and systemic risk.
