Continue active refreshing of this index's data?

Continue active refreshing of this index's data?

Equity Risk Monitors — July 5, 2023

Equity Risk Monitor Highlights | Week Ended June 30, 2023: Major Themes of the First Half

  • Risk Down Everywhere (Except Japan)
  • Size Factor Returns and Diversification
  • Statistical-Fundamental Risk Spreads

Risk Down Everywhere (Except Japan)

The decline in equity index volatility for the 1st half of 2023 has been a global phenomenon and quite smooth in most markets we track.  Factor volatilities are lower across the board.  When we examine this phenomenon using a “dense” asset/asset covariance matrix we see that the drop in pairwise correlations has kept pace with the drop in stock volatilities.  The CBOE VIX index is down more than 40% since the start of the year as well, demonstrating that factor models are not the only type of forecast volatility metric that’s plunging.  The following chart shows the change in risk forecast averaged over all 4 models in each market from the end of 2022 to the 30th of June 2023.

This chart is not published in the Equity risk monitors, but is available upon request:

With the exception of Japan which is up slightly, all of these markets show markedly lower risk than six months ago, even as macroeconomic conditions continue to tighten.  Even the regional bank /Credit Suisse banking crisis in March barely registered, except for two markets not directly affected- the UK and Japan.

Size Factor Returns and Diversification

As we have noted here for the last few months, the returns to the Size factor in developed markets are large in magnitude and the opposite sign relative to expectations from the history of the risk models.  This is not the case with Emerging markets, as shown by negative returns to Size in the EM, Asia Pacific Ex-Japan and China models.

This chart is not published in the Equity risk monitors, but is available upon request:

The index diversification levels in each of these markets appears to be inversely correlated with the performance of the Size factor as well. 

The following charts are available in the corresponding Axioma Equity Risk Monitors for June 30 2023:

Even though diversification appears to be improving in markets such as the US and Europe (although not Japan!), the levels are much lower (diversification ratios around 3) where, in China and EM they are around 6 or even 7.  The diversification ratio is the difference between the weighted average sum of the individual stock variances in the index versus the index-level variance forecast.  The higher this ratio, the more the index benefits from diversification across its constituents.  It would appear that diversification in developed markets is qualitatively different from that of emerging markets at present.  In developed markets, any improving diversification is from decreasing correlation between the mega-caps driving index performance and the rest of the market; in emerging markets it would appear to be a more “traditional” diversification where the asset correlations are falling more broadly, so the overall median correlations appear lower.

This chart is not published in the Equity risk monitors, but is available upon request:

Statistical-Fundamental Risk Spreads

We noted a few weeks ago that both the short-horizon and ultra-short trading horizon models in the US were showing growing spreads between the statistical model forecasts and the fundamental model forecasts, implying a “hidden” factor that the stat models picked up, and we hypothesized that this was related to mega-cap concentration.  This risk spread is now apparent in the Medium Horizon model as well.   This stands in sharp contrast with the Europe model, which shows almost no spread between the models, and the China model, which shows a significant spread in the opposite direction, at least over the short horizon.

The following charts are available in the corresponding Axioma Equity Risk Monitors for June 30 2023:

In the China model, the fundamental model’s forecast is 220 bps higher for the China A 900 index than the statistical model.  There is a directionally similar, but less significant gap between the medium horizon models.  Looking at the difference in asset-level risk contributions, it would appear that the differences are concentrated in stocks from a few key industries:

Other than Chemicals and non-bank financials, the theme would appear to be technology and EV related.  The lack of structure in the statistical model does not see this clearly systematic risk.

In the US, the forecast difference is 509 basis points in the Short-Horizon model.  189 basis points of that comes from the top seven names in the index:

When sorted by industry:

Meaning that 60% of the difference in risk grouped by industries is coming from just one or two names in that industry- very different from the situation in China.