The Recent Increase in the Volatility of Economic Indicators
In recent months, we have seen a staggering increase in stock market volatility. One popular measure of market volatility, the Chicago Board Options Exchange’s Volatility Index (VIX), jumped to 62.6 in November 2008, higher than it’s ever been. The VIX is computed from the S&P500 stock index option prices, and higher numbers imply that investors expect more volatile movements in the S&P index in the near term. (Numbers correspond to the annualized percentage point change expected over the next 30 days). The index has since dropped to 44.7 for January 2009, but this is still very high.
Is there a corresponding increase in volatility of macroeconomic variables? To answer this question, we focused on four main economic indicators, GDP growth, employment growth, productivity growth, and inflation. We obtained their volatilities by calculating their deviations from long-run trends. When an indicator is far from its long-run trend, regardless of whether it is above or below, its volatility is higher. More precisely, we computed the current volatility of an indicator as a weighted average of its past volatility and its deviation from its historic mean (in the case of GDP, employment, and productivity) or its two-year moving average (in the case of inflation).
Plotting the growth rate of GDP and its volatility from 1950 to the present, we can easily see what is commonly referred to as the “great moderation,” the long-run decrease in the volatility of GDP growth (and other variables) that started in the 1980s. Averaging 2.0 percent during the 1950–1984 period, the volatility of GDP growth fell to 0.9 percent after 1984. A variety of explanations for the great moderation have been proposed, including a change in the structure of the economy due to advances in information technology, increased resilience of the economy to oil shocks, increased access to financial markets, changes in financial market regulation, improvements in the conduct of monetary policy, a reduction in the size of domestic and international shocks (see “Why has output become less volatile” and “The Great Moderation: Good Luck, Good Policy, or Less Oil Dependence?”).
Our graph shows that the volatility of GDP growth has increased during the current recession. However, volatility also increased during the 1991 and 2001 recessions. As we will see, the volatility of other economic indicators also exhibits this cyclical pattern, increasing during recessions and then falling back. Indeed, the asymmetry of the business cycle may account for part of the increase in measured volatility during recessions. Because expansions last longer than contractions, the historic mean of a variable lies closer to the values it reaches during expansions. As a result, deviations of the variable from its mean are larger during contractions than during expansions. Taking this cyclical pattern into account, there is no evidence of a long-run increase in the volatility of GDP growth.
The chart below shows the similar pattern followed by the volatility of nonfarm payroll employment growth. Note that the severity of the current recession is reflected in that the latest spike in volatility is already higher than those of the previous two recessions.
While labor productivity growth also shows clear evidence of participating in the great moderation, it does not exhibit the same cyclical pattern as GDP growth and employment growth. Consequently, it should not be surprising that its volatility shows absolutely no sign of increasing in this downturn.
Inflation volatility spiked recently to a level surpassing the peaks reached during the 1970s. The recent high level of volatility is due to very low readings of inflation (even deflation) rather than high levels of inflation, as was the case during the 1970s. It is hard to judge how much of the recent increase in inflation volatility can be attributed to cyclical factors. First, the pattern of cyclical volatility is less pronounced in the case of inflation. Also, after its sizeable decrease during the 1980s, inflation volatility has drifted up since the late 1990s.
In summary, the volatility of most macroeconomic variables has recently increased. Except for the case of inflation, the high levels of volatility seem to be mainly due to cyclical factors, and there seems to be no evidence of any long-run increase in volatility.