Indexology^® Blog

As financial conditions have tightened materially since the start of the year and pressure is mounting on the U.S. equity market, it is worth circling back to see how the “wisdom of the crowd” has held up. Specifically, does a sentiment indicator derived from Tweets on Twitter hold as an informative investable factor? It’s been over eight months since we launched the S&P 500^® Twitter Sentiment Index Series in November 2021. Since market conditions changed significantly following the launch, let’s take a look at how the indices have been performing.

Previously, using back-tested results, data showed the sentiment factor helped provide better annualized performance when markets turned down compared to the indices’ underlying benchmarks: the S&P 500 and S&P 500 Equal Weight Index.

When reviewing the performance data for the first two quarters of 2022, we see that U.S. equity markets have experienced a significant downturn, with the S&P 500 dropping 19.96% and the S&P 500 Equal Weight Index dropping 16.68%. During the same period, we have seen that Sentiment acted as a cushion, with the S&P 500 Twitter Sentiment Index dropping only 18.78% and the S&P 500 Twitter Sentiment Select Equal Weight Index dropping only 14.94%—ultimately offering some downside protection relative to their benchmarks.

Even though both indices showcased this outperformance, we saw it more readily in the S&P 500 Twitter Sentiment Select Equal Weight Index, which outperformed its benchmark by 174 bps over the first half of 2022. This index is more sentiment concentrated—it reflects the 50 most positively talked about companies on Twitter from the S&P 500 and is equal weighted, versus a market cap weighting of 200 stocks in the S&P 500 Twitter Sentiment Index.

Essentially what the indices have demonstrated is that in a downturn, the wisdom of the crowd seems to have held.

Since January 2021, the conversation about stocks and indices on Twitter has grown by 52% and the number of users Tweeting about stocks and indices on Twitter has increased by 30%.¹

“The finance community on Twitter continues to grow, featuring robust real-time discussion among a wide range of professional and individual investors. The S&P 500 Twitter Sentiment Index helps quantify the value of those conversations and gives people on Twitter an exciting way to measure the impact of their Tweets about publicly traded companies,” said Jared Podnos, Head of Strategic Market Development for Twitter’s Developer Platform.

An increased Tweet and user volume as a base for our indices, combined with a challenging two quarters for the U.S. equity market, provided a good test period for the S&P 500 Twitter Sentiment Indices, and they have managed to come out on top.

¹ Source: Twitter Internal Data. Stocks and Indices Conversation Dec. 31, 2021-June 23, 2022. Global. PT, ES, DE, JA, TR, EN, HI, AR language only.

DISCLAIMER

S&P® and S&P 500® are registered trademarks of Standard & Poor’s Financial Services LLC. Twitter® is a registered trademark of Twitter, Inc. These marks have been licensed for use by S&P Dow Jones Indices (S&P DJI) for use with the S&P Twitter Sentiment Index Series. The Indices are meant for informational purposes only and are not recommendations to buy or sell any securities. Any investment entails a risk of loss. Please consult your financial advisor before investing. S&P DJI and Twitter receive fees in connection with licensing and use of the S&P Twitter Indices.

The widespread global adoption of ESG and climate indices has created an accompanying need for better understanding regarding the performance impact of the various choices made in their design. As introduced in S&P DJI’s new Climate & ESG Index Dashboard, a Brinson-like “carbon attribution” teases out one particularly important source of returns.

Untangling the exact impact of individual goals on ESG and climate index returns can be a challenge, as sophisticated indices can incorporate a range of objectives. For example, the S&P PACT^TM Indices (S&P Paris-Aligned & Climate Indices) includes explicit carbon-reduction targets, temperature alignment goals and climate objectives such as mitigation and adaption, while an optimization process is used to target these multiple objectives simultaneously.¹ This can make the individual contributions hard to unpack.²

Conveniently, the impact of weighting to higher- or lower-carbon-emitting stocks may be assessed analogously to the way sector or country effects are measured by a Brinson attribution.³ To do so, first, we rank each benchmark security according to their weighted average carbon intensity (WACI),⁴ and then slice up the underlying benchmark universe into five quintile portfolios, from highest to lowest, each with an equal number of benchmark constituents. Then, for any index based on the same benchmark universe, we can simply apply the standard Brinson methodology (using the performance of our hypothetical carbon portfolios instead of the traditional sector or country portfolios) to measure the allocation effects of over- or underweighting each carbon quintile.

The S&P 500^®-based S&P 500 Net Zero 2050 Paris-Aligned ESG Index offers a case study. Exhibit 1 shows the weights of the index and its underlying benchmark, as well as the active weight of the index, in each S&P 500 carbon quintile as of June 30, 2022.

The results of the subsequent return attribution by carbon quintile portfolio—as measured over the prior three months—are shown in Exhibit 2.

Together, Exhibits 1 and 2 measure the extent and impact of carbon-based weights over the period: the High Carbon Quintile 1 outperformed the S&P 500 by 6.1% (see Exhibit 2), and the S&P 500 Net Zero 2050 Paris Aligned ESG Index underweighted this carbon quintile by -6.9% (see Exhibit 1), which detracted from the relative index performance. Conversely, the overweight in the Low Carbon Quintile 5 contributed positively to the relative index performance.

Repeating and cumulating such analysis over multiple periods offers a longer-term perspective, as shown in Exhibit 3 for the five-year period ending in June 2022. Over this longer-term period, an underweight in the High Carbon Quintile 1, and an overweight in the second-least carbon intensive quintile, contributed the most to the index’s overall outperformance.

By quantifying the links between index design choices and index performance, a carbon-based attribution analysis such as that in Exhibit 3 can offer insight and perspective that may prove useful in assessing the merits of one index over another. Investors seeking similar attributions for a range of our flagship indices are now able to find them in S&P DJI’s newly launched Climate & ESG Index Dashboard.

Register here to receive quarterly insights and performance attributions for our range of flagship ESG and climate indices.

¹ For more details, see the full index methodology, available at www.spglobal.com/spdji/en/documents/methodologies/methodology-sp-paris-aligned-climate-transition-pact-indices.pdf

² www.ssga.com/content/dam/ssmp/library-content/pdfs/insights/reducing-carbon-in-equity-portfolios.pdf

³ https://jpm.pm-research.com/content/11/3/73; https://www.jstor.org/stable/4478947

⁴ https://www.spglobal.com/spdji/en/documents/additional-material/faq-sp-paris-aligned-climate-transition-indices.pdf

With the first half of 2022 in the books, commentators have noted that this year’s -20.0% total return for the S&P 500® is the worst January-June result in more than 50 years. Painful as the first six months were, what might they tell us about the rest of the year?
History gives us both bad news and good news. Bad news first: the correlation between the past six months’ return and the next six months’ return is vanishingly small. Predictions are problematic. Exhibit 1 illustrates this for the S&P 500; results for the S&P MidCap 400® and S&P SmallCap 600® are comparable. (The data here encompass not just the first six months of the year, but every six-month period from 1995 onward.)

In general, knowing how well or poorly an index did in the past six months tells you nothing about how well or poorly it will do over the next six months.

But…with a little legerdemain, we can tease out some good news as well. The fact that returns have been above average, or below average, does not help us forecast what returns will be going forward. This means that regardless of what has already happened, the next six months’ return is best regarded as a random draw from the same distribution that generated the last six months’ return.

We can use this insight by sorting the data points in Exhibit 1 into deciles based on the last six months’ performance. Within each decile, we can measure the average monthly performance in the last six months and the next six months. The difference between the next six months and the last six months represents the improvement (or worsening) of performance by decile and is graphed in Exhibit 2.

On one level, Exhibit 2 is just a demonstration of mean reversion in action. But it also has a practical implication: if historical returns have been especially good, future returns are likely to be worse, and if historical returns have been especially bad, future returns are likely to be better. At the end of June 2022, historical results across the capitalization range were indeed especially bad, as Exhibit 3 illustrates.

There are no guarantees, but history tells us that when returns are as bad as the first half of 2022’s have been, improvement has been much more frequent than continued decline. When returns have been especially bad, patience tends to be especially valuable.

The United Nations estimates a 70% increase in food supply is needed to meet a projected 2050 global population of 9.6 billion. With 70% of its agricultural production exported today, Australia could play an integral role in meeting increased demand from the global supply chain. Join Nadine Blayney of Ausbiz, Daphne van der Oord of S&P DJI, and Ken Chapman from ASX for a closer look at how the broad S&P/ASX Agribusiness Index could help market participants track and access growth in Australia’s agribusiness sector.

Explore the Australian Market with Confidence www.spglobal.com/spdji/australia

The S&P 500^® posted its worst first-half performance since 1970,¹ as inflation concerns, Fed rate hikes and slowing economic growth have weighed on markets. Mega caps were hit particularly hard, with the S&P 500 Top 50 posting a loss of 22%, underperforming the S&P 500 by 2%. Exhibit 1 shows that Information Technology was the biggest sectoral detractor from the S&P 500’s performance, followed by Consumer Discretionary and Communication Services. Energy was the sole positive contributor and the only sector to post a gain YTD, up 32%.

This market environment, characterized by higher volatility and wide disparity among sectoral performance, has positive implications for skillful sector allocators. To understand why, it is important to remember that volatility is linked to dispersion, which measures the spread among returns of an index’s components. When volatility goes up, dispersion also tends to rise, as the gap between the winners and losers widens. The greater the spread, the greater the opportunity to add value. Dispersion can be measured at various levels of granularity, such as among stocks or sectors.

Total market dispersion can be decomposed into the average dispersion within each sector and the dispersion across sectors. The ratio of cross-sector effects to total S&P 500 dispersion has remained above average, which implies that the rewards for skillful sector picks have been increasing (see Exhibit 2).

A natural consequence of the weakness among mega caps was a tailwind for Equal Weight because of its small-cap bias, with the index outperforming the S&P 500 by 3% so far this year. Exhibit 3’s YTD performance attribution of the S&P 500 Equal Weight Index illustrates that the underweight to IT and Communication Services were key contributors to Equal Weight’s outperformance. 

Another consequence of the outperformance of smaller caps is that most factor indices, which generally have a small-cap tilt, outperformed the S&P 500 (see Exhibit 4). Low Volatility and Dividend strategies took the lead, as factor performance was importantly influenced by the extraordinary outperformance of Value versus Growth.

The comeback of smaller caps and Equal Weight might have positive implications for active managers, as their portfolios are often closer to equal than cap weighted.  Exhibit 5 plots the underperformance of large-cap funds compared to the relative performance of the S&P 500 Equal Weight Index versus the S&P 500, as a proxy measure for smaller-cap outperformance. We notice that two out of the three years when most active large-cap managers outperformed (2005, 2007 and 2009) coincided with Equal Weight’s outperformance.

The current challenging environment, characterized by high inflation, concerns about potential Fed rate hikes and weak economic fundamentals, is reminiscent of 1970. That was the last time we experienced this level of underperformance for the first half of the year, which was subsequently followed by a strong turnaround in the second half of the year—an ultimate indicator that past performance is not indicative of future results. 

¹ Based on S&P 500 Price Return.