It’s been five years since we published the first article on the “weirdest chart ever,” which shows the cyclical relationship between implied volatility on equity index options (represented by the VIX index) and the yield curve. Now, Anne Lundgaard Hansen of the Federal Reserve Bank of Richmond has published the first ever academic study on the VIX-yield curve cycle concluding the following:
“The proposed indicator significantly outperforms the yield-curve spread in predicting U.S. recessions from 1990–2021 both in- and out- of-sample using both static and dynamic probit models. VIX-yield-curve cycles also contain predictive power above and beyond other leading economic indicators.” Here’s the link to her paper: Predicting Recessions Using VIX-Yield-Curve Cycles
If it is a good economic indicator, what is it telling us now?
At first glance the relationship between the VIX index and the yield curve is not obvious. A two-year moving average of the two indicators charted since 1990 looks like this (Figure 1):
Figure 1: Is this weird chart the most powerful indicator of economic downturns?
However, when broken down into three bite-size pieces and color-coded by year to get a sense of movement and direction, a distinct repeating pattern immediately jumps out (Figures 2, 3 and 4).
Figure 2: The first cycle lasted from the 1990-91 recession to the expansion peak in 1999-2000
Figure 3: The second cycle lasted from the 2001 tech wreck until the global financial crisis
Figure 4: The 3rd cycle began in 2009 but has taken an unprecedented turn in recent months
The cycle works in four phases. As in any cycle, where it begins and ends is a bit arbitrary but we present it here from early recovery to recession:
Early Recovery: Economic recoveries begin with easy monetary policy but still feature high volatility lingering from a recent economic downturn. Easy monetary policy usually means low short-term interest rates and therefore a steep yield curve. Over time, easy monetary policy tends to calm market volatility. As the central bank injects liquidity, the economy typically recovers, reducing investor uncertainty. Also, additional liquidity in markets makes it easier for buyers to find sellers and sellers to find buyers, so large market orders can be executed without prices having to move as far into order books to get filled. As such, as the recovery matures volatility begins to subside.
Mid-expansion: the middle part of economic expansions is usually characterized by falling implied volatility in markets and yield curves that remain steep. At some point, with calm markets and falling unemployment, the central bank concludes that it is an opportune time to begin policy tightening. Policy tightening begins the process of flattening the yield curve as short-term rates rise towards long-term rates.
Late-expansion: Policy tightening does not immediately produce a recession. In fact, recessions don’t usually occur until six months to two years after a tightening cycle is complete. Thus, the late expansion stage usually begins with a flat yield curve and still low volatility. However, as the central bank’s tight monetary policy eventually tightens credit conditions and reduces market depth. In this context, filling large market orders sometimes involves risking larger prices moves. Volatility begins to rise in line with decreasing market depth and increasing economic risks.
Eventually high volatility and a tight monetary policy, characterized by a flat yield curve, produce an economic downturn. High short-term rates make lending and borrowing uneconomic while high volatility makes it more difficult for private sector entities to raise capital from equity and bond markets. The economy falls into a recession and the central bank eases monetary policy, lowering short-term rates and steepening the yield curve.
Wash. Rinse. Repeat. At Least Up Until Now.
The cycle played out three times from 1990 to 2021. However, the pandemic and its aftermath have left it in a strange place. Most of the observed cycle, from 1992 to early 2021, took place in the context of low, stable rates of inflation. Inflation is no longer low or stable. Moreover, the 2020 recession, though not a surprise in the context of the cycle, happened largely as a result of exogenous factors (pandemic and lockdowns) and not for endogenous reasons such as credit problems in the private sector, the usual proximate cause of downturns.
In this case, the Fed cut rates to near zero and engaged in quantitative easing, as it had done from 2009 to 2014. However, this most recent cycle was also accompanied by an unprecedented fiscal expansion which took Federal spending from 21% to 35% of GDP while the Federal budget deficit grew from 5% to nearly 20% of GDP. In this context, the labor market recovered very quickly with unemployment falling from 14% to below 4% in about two years as inflation surged from its pre-pandemic 2% to around 9.1%.
Now the Fed is rapidly tightening policy. It’s already raised rates by 150bps and, judging from the public statements of FOMC members, it intends to raise rates another 50-75bps at each of its next two meetings. Unless there is a sharp rise in long-term bond yields, the yield curve appears set to flatten much earlier than it has in the past few cycles. This could take us down towards the bottom of the chart.
The next question would be: what happens to volatility? Volatility subsided after the tremendous volatility spike in March 2020, which recently rolled off the two-year rolling window, but volatility is not particularly low (Figure 5). Indeed, there are multiple reasons to be concerned:
Figure 5: Is volatility about to rise sharply again?
- Tighter Fed monetary policy often pushes volatility higher.
- Volatility has been trending higher for a year now, since hitting bottom on June 30, 2021.
- Equities could still be overvalued compared to investor expectations for future dividends and the current level of long-term rates (our article here).
In some ways the current situation is vaguely reminiscent of the situation in 1999-2001 at the peak of the 1990s economic expansion. Then the VIX-yield curve cycle did a series of loop-tee-loops (scientific expression) in the far lower right-hand corner of the chart. Then, as now, the Fed had cut rates to deal with an emergency (Russian debt default and now defunct hedge fund Long-Term Capital Management) and then had to conduct a second tightening cycle in 1999 and 2000. Then, as now, technology stocks peaked and began falling. Eventually this translated into a recession in 2001 and 2002.
What was missing then was high inflation. By 2000 inflation had perked up to around 3.3%, high by the standards of the period from 1990 to 2020 but not high by historical standards nor by today’s standards. The slight bump in inflation in 2000 was enough to prompt the Fed to raise rates to 6.5%, which was enough to invert the yield curve and send the economy into a recession.
Currently, inflation is at 9.1% headline and around 6% core. Fed Funds are still at 1.625% as of this writing. It could be that the VIX-yield curve cycle spends considerable time in the lower right-hand portion of the chart until the Fed is able to bring down inflation and begin easing monetary policy, which could be a year or more in the future. It remains to be seen whether, as Hansen’s analysis suggests, the VIX-yield curve cycle will be successful at predicting the next economic downturn. That said, if it does return to the lower right-hand corner, it would be blinking red and warning of a possible economic downturn.
Interest Rates Data
Get comprehensive views of the U.S. dollar-denominated interest rate markets across the entire yield curve with our robust datasets.
All examples in this report are hypothetical interpretations of situations and are used for explanation purposes only. The views in this report reflect solely those of the author and not necessarily those of CME Group or its affiliated institutions. This report and the information herein should not be considered investment advice or the results of actual market experience.