Economists like Keynes, Minsky, and Kindleberger have highlighted the importance of shifting perceptions of risk in driving markets and the economy. When risk perceptions are low, risky markets tend to perform well. But when risk perceptions increase, safer assets come into demand. Over the years, numerous measures have been used to measure risk perceptions – one widely followed by investors is equity volatility or VIX. In this paper, Financial Market Risk Perceptions and the Macroeconomy, Carolin Pflueger, Emil Siriwardane, and Adi Sunderam introduce a new measure that they call ‘the price of volatile stocks’ or PVS. But does it work?
This article is only available to Macro Hive subscribers. Sign-up to receive world-class macro analysis with a daily curated newsletter, podcast, original content from award-winning researchers, cross market strategy, equity insights, trade ideas, crypto flow frameworks, academic paper summaries, explanation and analysis of market-moving events, community investor chat room, and more.
Economists like Keynes, Minsky, and Kindleberger have highlighted the importance of shifting perceptions of risk in driving markets and the economy. When risk perceptions are low, risky markets tend to perform well. But when risk perceptions increase, safer assets come into demand. Over the years, numerous measures have been used to measure risk perceptions – one widely followed by investors is equity volatility or VIX. In this paper, Financial Market Risk Perceptions and the Macroeconomy, Carolin Pflueger, Emil Siriwardane, and Adi Sunderam introduce a new measure that they call ‘the price of volatile stocks’ or PVS. But does it work?
How Is PVS Measured?
The authors calculate the book-to-market ratios for all publicly traded US stocks. They use the previous quarter’s book value to accurately capture the available data at the time, and they use a trailing six month average of market capitalisation. A high ratio means that a stock is cheap, while a low ratio means it is expensive.
Then they sort all the stocks by their two month realised volatility of stock prices. They group the companies into five buckets of volatility from highest to lowest, equally weighting the book-to-market ratios in each bucket. Finally, to arrive at the PVS value, they take the difference between lowest volatility bucket’s ratio and the highest volatility bucket’s ratio.
A low PVS value would imply perceptions of risk are elevated as the valuations of low volatility stock are expensive compared to that of the high volatility stocks. A high PVS value would imply perceptions of risk are subdued as the valuations of low volatility stocks have cheapened relative to high volatility stocks.
Does PVS Correlate Well with Other Measures of Risk?
Yes. The PVS measure is (correctly) inversely correlated with earnings forecast risk and options volatility – so a low PVS (high perception of risk) is associated with higher earnings risk. The PVS is also positively correlated with real rates and bank lending such that a low PVS (high risk) is associated with low interest rates (high demand for bonds) and falling lending (Figure 1).
How Does the PVS Indicator Relate to the Economy?
First, there is a link between PVS and real rates. Since 1970, a one-standard deviation increase in PVS has been associated with a 1.3 percentage point increase in the real rate. This holds in expansions and recessions (shown in grey), as well as in both high- and low-inflation periods (Figure 2).
Figure 2: One-Year Real Rate and PVS
Source: Financial Market Risk Perceptions and the Macroeconomy
Importantly, the authors find no relationship between the book-to-market ratio of the aggregate stock market and the real rate. So there is clear value in splitting companies into buckets of volatility as the PVS measure does.
On a forward-looking basis, they also find a one-standard deviation increase in PVS is associated with an investment-capital ratio that is 0.22 percentage points higher at a one-quarter horizon. The magnitude is 0.35 percentage points at a four-quarter horizon. They find similarly strong results for the output gap and the unemployment rate.
Can the PVS Indicator Predict Markets?
They find that the PVS measure does predict stock market performance. A one-standard deviation increase in PVS forecasts a 15.1 percentage point higher annual return on the volatility-sorted portfolio. The Sharpe ratio is around 0.5.
They also find significant predictive relationships for volatility-sorted portfolios in US corporate bonds, options, and CDS. The results are poor for commodities and currencies.
What Drives the PVS?
There appears to be a statistically significant relationship between surprises in real GDP and changes in PVS. A one-standard deviation higher real GDP growth surprise is associated with a 0.6 standard deviation increase in PVS. There are similar results for the surprise in corporate profit growth and changes in credit growth. Overall, the results here show that perceived risk falls after good news about the economy.
Final Takeaways
It seems the is a new and useful additional to the risk indicator family, and it uses equity data in a richer and more sophiscated manner than its siblings. It appears to be driven by and correlates with factors that investors would reasonably expect. The main challenge is whether the indicator can work well on a high-frequency basis – much of the testing was done on quarterly and monthly data. At Macro Hive, we’ll be investigating this indicator further to see whether we can include it in our Macro Hive Risk Barometer.