by Qontigo
- US short-horizon statistical-fundamental gap widens
- Banks’ woes caused Value’s return to plunge in Europe
- Trading volume has increased, but only in Financial stocks
US short-horizon statistical-fundamental gap widens
In the US, the risk forecast for the STOXX® USA 900 Index from Axioma’s US short-horizon statistical model pulled ahead of the prediction from the same-horizon fundamental model at the same time as the collapse of Silicon Valley Bank (SVB), and the gap has continued to widen ever since. Fundamental predicted risk has been higher than statistical for most of the past 12 months, and the differences between the two forecasts are far smaller in other regions.
Because of the nature of the statistical model, it is difficult to accurately attribute the reason for the differential, but we often note that if we see a higher forecast from the statistical model it may be picking up a risk that is not included in the market, style or industry factors. We note that the specific risk is very similar between the two models (and a small part of total benchmark risk), so suspect that there is a factor — maybe contagion risk beyond just the financial industries, or change of course by the Fed? — being picked up in the statistical model that is not included in the fundamental model.
For more thoughts on bank contagion see the recent blog post from Qontigo’s Christoph Schon here.
See chart 7 from the Unites States Equity Risk Monitor of 24 March 2023:
And for a closer view see chart below, which is not in the Risk Monitors, but it is available upon request:
Banks’ woes caused Value’s return to plunge in Europe
Value has had a tough go of it over the past month, and across the regions Qontigo Risk Monitors track closely the return is most negative in Europe. As problems for European banks surfaced over the past week or so, Value’s return in Europe was just about -1% in the week ended March 24, almost three standard deviations below its long-term average five-day return, based on the risk forecast at the beginning of the week from Axioma’s Europe Short-horizon fundamental mode. (Note that the volatility forecast for Value in Europe has been inching up, whereas in the US forecast risk for Value remains at the low end of its 12-month range.)
We investigated the source of this underperformance, assuming it is related to the drubbing taken recently by many financial stocks. The factor-mimicking portfolio that drives the factor return is, of course, sector-neutral, so long some financials and short others. We found that it was indeed the financials that drove the factor performance, and furthermore it was largely concentrated in the long positions, with Credit Suisse having the biggest impact on performance. Given the bailout, we would expect the performance of Value in Europe to stabilize assuming no more shoes drop…
See chart 7 from the Europe Equity Risk Monitor of 24 March 2023:
Trading volume has increased, but only in Financial stocks
For the past six months we have been discussing the low level of equity trading volume, noting that the overall level was much lower than might be expected from a decline in market values. Our assessment was that investors remained concerned about inflation and Fed moves among other issues, and therefore were not sure about what to trade, so were refraining from trading. Average daily volume (defined as the rolling average of 20-day trading volume in USD) bottomed out in January, although the current level remained far below the 12-month average.
We have seen volume pick up in the last couple weeks across the major markets we track. However, in most sectors volume remained close to its average level of the last 12 months. It was only Financials that have seen an extraordinary increase, with volume in the STOXX Global 1800 Index about 70% higher most recently as compared with the average level of the past year. Again, if markets stabilize and no more Financial shoes drop, we would expect to see volume once again retreat.
See charts from the Global Equity Risk Monitor of 24 March 2023:
