According to one prominent investment firm’s analysis of the VIX, the stock market has not yet experienced the capitulation that typically signals the end of the bear market.
By capitulation, I’m referring to the deep despair that leads investors to throw in the towel and swear off equities. Though not every bear market ends with capitulation, most have. So Wall Street analysts are scouring the historical record for reliable indicators of capitulation.
The VIX — the CBOE Volatility Index VIX, -6.56% —reflects option traders’ expectation of the S&P 500’s SPX, +1.49% volatility over the subsequent month, with higher levels indicating greater expected volatility. Since 1990, the earliest year for which the CBOE has historical data for the VIX, its highest-ever close was 82.69 (in March 2020). Its lowest-ever close came in November 2017 at 9.14. It currently stands in the low 20s.
A recent analysis by equity strategists at BNP Paribas concludes that the VIX is a reliable indicator of market capitulation and, therefore, useful for determining whether the bear market has come to an end. They found that the median VIX level at past bear-market bottoms was 40.5, well-above the VIX’s highest level hit (at least so far) in the current bear market (which is 36.45). Furthermore, since the firm found that spikes “in volatility have on average come at the same time as the trough in the market,” they conclude that the bear market has not yet hit bottom.
The firm’s argument seems plausible, since the VIX has stubbornly refused to spike upwards during this bear market — no matter how much turmoil the market has suffered. Take what happened on Dec. 15, when the stock market suffered its biggest decline in three months — with the Dow Jones Industrial Average DJIA, +1.60% slumping more than 750 points. The VIX that day closed up just 1.69 points, at 22.83. This closing level stands at the 74th percentile of the VIX’s historical distribution since 1990, which means that 26% of daily closes over the past 32 years were higher. This certainly suggests that we’re yet to experience capitulation.
Investors nevertheless shouldn’t bet too heavily on this message from the VIX. The median VIX level identified by BNP Paribas at past bear market bottoms — 40.5 — conveys a false precision, since in fact it sits is the middle of a wide range.
Consider where the VIX stood at the bottom of the eight bear markets since 1990 in the Ned Davis Research calendar of bull and bear markets. It ranged broadly from 28.14 to 61.59. In two of those eight, in fact, the VIX was lower than the levels it hit in both the spring and in October of 2022. It therefore seems like a stretch to confidently conclude, from the VIX itself, that the bear market didn’t hit bottom at the market’s spring or October lows.
This wide range is also illustrated in the chart above, which reports the S&P 500’s subsequent 12-month return as a function of the VIX. Though average return is correlated with VIX level, notice from the green columns the spread between the stock market’s best- and worst returns. Any bet based on this chart’s data would need to be a low-confidence bet.
Consider what happened during the Global Financial Crisis. Prior to the GFC, the VIX had never risen above the high 40s. So when the VIX rose to that level in October 2008, many of the market timers my firm monitors confidently bet that the bear market was at or near its end. They were wrong. Stocks kept sliding. The VIX in November 2008 would spike to near 90, and the bear market wouldn’t end until the following March, when the S&P 500 was nearly a third lower.
Consider also the notion that a spike in the VIX indicates that the bear-market bottom is near. For each bear market since 1990 in the Ned Davis Research calendar, I calculated the number of days from the date on which the VIX hit its peak to the date on which the bear market ended. The average was 57 calendar days, or almost two months. While in the case of one bear market the VIX’s peak occurred on the exact day of the bear market low, in another case 171 calendar days (almost six months) transpired between the peak and the end. Again, that’s quite a wide range.
So even if the VIX in recent days had spiked up by enough to suggest capitulation, we still couldn’t conclude that the bear market was at or near its end.
These observations are not intended as a criticism of the PNB Paribas research. Since there is no agreed-upon definition of what capitulation is, imprecision is inherent to any attempt to measure it. This is why some analyses have suggested that capitulation has already occurred, while others — such as the research from PNB Paribas — suggest it has not.
The bottom line: The picture is mixed, but that is hardly a surprise. It will never be the case that the indicators all point in the same direction. On the one hand, it is true that, had the VIX spiked much higher in recent sessions, the weight of the evidence would be tilted more towards believing the bear market is near its end. But, on the other hand, such a tilt would be extremely slight.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at firstname.lastname@example.org