Indexology^® Blog

The Federal Open Market Committee voted to raise the Federal Funds rate by 25 bps on March 16, 2022. This move was well telegraphed and not at all surprising—but that doesn’t mean that we won’t hear concerns about how rising rates will impact equity returns. Finance theory teaches us that, other things equal, rising interest rates are not good for the performance of stocks, as rising borrowing costs and higher discount rates tends to translate to lower future performance. For the much of history, empirical evidence has aligned with the theory. But in more recent data, we have noticed that “other things” may not have always been equal.

At a cursory glance, rising rates have not necessarily boded ill for equity performance, at least in the period from 1991 through 2021. There were eight episodes when the 10-Year U.S. Treasury yield rose. The S&P 500^® declined in none of these; in two cases equities were flat, and the S&P 500 rose in six, in some instances quite substantially.

Breaking down the period in Exhibit 1, there were 156 months when the 10-Year U.S. Treasury yield rose (and 216 months when it declined). Of the months when the 10-Year U.S. Treasury yield rose, the S&P 500 gained in 115 (74%) and declined in 41; the S&P 500 rose nearly three times as often as it fell when interest rates rose. On average, the S&P 500 gained 1.57% each month that rates rose, versus just 0.55% in months when rates declined.

We can also look at these months graphically in the scatter plot in Exhibit 3. Here we plot the change in the 10-Year U.S. Treasury yield against the performance of the S&P 500 for the same period from 1990 through 2021. Each point represents a monthly observation, and we see no discernible relationship. The blob speaks for itself—or rather, it doesn’t. History does not provide evidence of a clear link between changes in interest rates and changes in the equity market.

The Russia-Ukraine conflict is now in its third week and markets remain volatile. The major U.S. equity benchmarks dropped about 10% from their peaks, with the exception of the Energy sector. The CBOE Volatility Index (VIX^®), the so-called “fear gauge,” has been hovering above 30, which is the 90^th percentile of its historical value. Its level on March 10, 2022, was more than two standard deviations above its one-year average. Although it remains unclear how long these geopolitical tensions will last and how much it will affect the global economy, the U.S. equity market has managed to avoid the VIX levels seen two years ago, which were triggered by pandemic-driven sell-offs.

More importantly, historical data show that the equity markets tend to bounce back quickly after elevated volatility. We look at all the trading days on which VIX hit above 30 and calculate the S&P 500® performance in the subsequent 6 months and 12 months. The scatter charts in Exhibit 2 show that the vast majority of these 557 days were followed by positive performance in the next 6 months (82%) and 12 months (88%).

We further compare the 6- and 12-month performance after these highly volatile days with rolling returns on all historical days. On average, the 12-month performance after VIX hit above 30 was two times higher than the 12-month rolling returns on any business day since Dec. 31, 1999. Similar results are reflected in the small-cap space (see Exhibit 3).

After the exceptional performance of large-cap stocks in recent years, concentration concerns naturally come to mind.

There are many ways to measure concentration. A simple method is to add up the weight of the top names, but the drawback with this approach is it doesn’t incorporate all the constituents in an index. The Herfindahl-Hirschman Index (HHI), defined as the sum of the squared index constituents’ percentage weights, is more favorable from this aspect and is widely used.

But the HHI faces an issue, which is that even for completely unconcentrated equal weight portfolios, the HHI value is inversely related to the number of names. If we want to use the HHI to examine the history of concentration within an index or to make cross-sector comparisons, we need to adjust for the number of names. In our paper Concentration within Sectors and Its Implications for Equal Weighting, we define the adjusted HHI as the index’s HHI divided by the HHI of an equally weighted portfolio with the same number of stocks.

A higher adjusted HHI means that an index is becoming more concentrated, independent of the number of stocks it contains. Exhibit 1 shows that the adjusted HHI for the Energy sector decreased from 2014 to 2019, despite an increase in its raw HHI. This is because the number of constituents in the sector decreased from 43 in 2014 to 28 in 2019.

Concentration tends to mean-revert in most sectors. This is particularly noticeable in Energy, Industrials, Information Technology, and Materials, as we see in Exhibit 2. These data imply that when concentration is relatively high, as we see for Information Technology presently, it subsequently tends to decline. Meanwhile, when concentration is relatively low, as we see for Industrials, Energy, and Materials, it subsequently tends to increase.

Rising sector concentration implies that larger-cap names are outperforming smaller caps, and that a cap-weighted index should outperform its equal-weighted counterpart. Falling concentration implies the opposite. Since concentration tends to mean revert, using relative concentration to alternate between cap-weighted and equal-weighted sector exposures is a potential source of value added.

The S&P Indices Versus Active (SPIVA^®) Scorecard¹ provides a rich tapestry of data and insights on the active versus passive debate by comparing the performance of actively managed funds to their corresponding passive benchmarks. The SPIVA Latin America Scorecard is one of several regionally focused reports and covers the performance of active funds in Brazil, Mexico, and Chile. Below are some takeaways from our latest year-end 2021 report.

1. It Is Difficult to Outperform the Benchmark over the Long Run.

In the short run, most Brazilian Government Bond Funds outperformed the benchmark, but this swiftly turned into underperformance for holding periods of three years or more. Brazilian Large-Cap Equity Funds fared better, as slightly more than half of actively managed funds outperformed the S&P Brazil LargeCap for up to five years. In the short run, it is not unusual to see the average active manager flipping between underperformance and outperformance. After a decade, however, at least 70% of active funds underperformed in each category; underperformance among Chile’s active equity funds was the highest, with 98% of funds lagging their benchmark over the past 10 years.

2. Funds Struggle to Survive across Categories.

The price of being a poor-performing active fund is hefty, with the worst-performing funds liquidating or merging with other funds. Exhibit 2 shows that Brazil Corporate Funds had the lowest survival rate, with only 32% of funds still alive after 10 years. Interestingly, despite low rates of outperformance among Mexican equity funds, 78% managed to keep their doors open after a decade.

3. Even Among Survivors, Outperformance Rates Are Poor.

Using Exhibits 1 and 2, we can divide the SPIVA database into three groups: funds that did not survive, funds that survived but underperformed, and funds that both survived and outperformed. We can see that even among the funds that survived, more than two-thirds underperformed in nearly every fund category.

4. The Picture Doesn’t Improve Much on an Asset-Weighted Basis

Exhibit 1 shows the outperformance rate of funds on an equal-weighted basis. Some may make the argument that funds with larger assets have more assets because they have historically outperformed, and that they maintain relatively large assets because they produce “alpha.” Report 4 of our recent SPIVA Latin America Scorecard shows that in only one out of seven active fund categories were funds able to provide an average asset-weighted return above its appropriate benchmark over a 10-year period. Brazil Large-Cap Funds provided the one “success” story, but only narrowly.

5. And Finally, There Is a Wide Range of Performance among Active Funds.

While “alpha” may exist, identifying outperforming managers is difficult and the performance distribution of actively managed funds is skewed. The interquartile range, which is the spread between the first and third quartile breakpoints of active funds in their respective category, is quite wide, showing there is a large difference between top and bottom fund performances.

Get the latest numbers and more from the SPIVA Latin America Scorecard here.

¹ SPIVA Scorecards: An Overview. https://www.spglobal.com/spdji/en/education/article/spiva-scorecards-an-overview

U.S. equities have had a tough start to the year amid rising inflation concerns, anticipated rate hikes by the Federal Reserve, earnings misses from several mega-cap companies, and ongoing geopolitical tensions emanating from the Russia-Ukraine conflict. Exhibit 1 shows that the S&P 500^®, S&P MidCap 400^®, and S&P SmallCap 600^® all declined in the first two months of 2022, as did the majority of their respective sector, style, and pure style indices. Energy was a notable outlier, boosted by surging commodity prices.

While recent news flow had a similar directional impact on U.S. equities, the magnitude of impacts varied across sector and style segments. For example, Exhibit 2 shows that a whopping 41% separated the best-performing S&P 500 sector (Energy, up 28%) and the worst-performing sector (Real Estate, down 13%) in the first two months of 2022, while the S&P 500 Value (-3%) beat the S&P 500 Growth (-12%) by 9%. Such differences suggest that sector and style perspectives could have been helpful to navigate the recent market environment; avoiding the worst-performing segments or identifying the outperformers could have offered sizeable performance improvement.

Comparing dispersion figures for S&P 500 stocks, sectors, and styles offers a more formal way to assess the potential opportunity for outperformance (or embarrassment) from picking S&P 500 segments. Exhibit 3 shows that, on average, annualized monthly stock level dispersion (24%) was typically higher than sector dispersion (11%), which in turn was greater than style dispersion (3%).

In other words, about half the potential value of stock picking came from picking sectors, with about 14% of the potential value of selecting stocks coming from style selection. This is unsurprising, given that the S&P 500 comprises 505 stocks and 11 GICS sectors, but only two styles. Notably, though, style and sector dispersion figures have been similar in the first two months of 2022, averaging 8% and 13%, respectively: the potential value of style insights was similar to sector insights in recent months.

The potential value of style perspectives is even more pronounced after adjusting for capacity, since both the potential opportunity for outperformance (measured by dispersion) and the potential size of active positions (measured by the average constituent size) are relevant when comparing the value of insights into different segments. This is especially the case given the proclivity for smaller market segments to exhibit higher dispersion.

Exhibit 4 shows the average capacity-adjusted dispersion figures for S&P 500 style and S&P 500 sectors relative to S&P 500 stocks. Over the entire period, driven by the relative size of their respective components, insights into style were eight times more valuable than those for stocks, with sector insights being six times more valuable than for stocks. More recently, elevated style dispersion at the start of 2022 meant style insights could have been more than twice as valuable as sector insights at the start of the year, and over 10 times as valuable as stock insights.  

Hence, while it can sometimes be useful to do nothing, the recent divergence in style performance means some may find it useful to express views through a style perspective.