Indexology^® Blog

The equity markets roared to the finish line, easily passing commodities and bonds with a 26.3% return for the S&P 500^® in 2023. Much of that rally could be attributed to declining consumer price indices and favorable Federal Reserve guidance, with increased confidence in a soft landing. Declining inflation can best be observed at the kitchen table, where the S&P GSCI Grains saw 15% declines on the year and the S&P GSCI Livestock remained flat.

Despite another strong year of equity performance, the Dow Jones Commodity Index (DJCI) Gold maintained its top perch for this century after reaching an all-time high in December. Gold achieved a 12.8% return, outpacing both bonds and broad commodities, with the S&P GSCI recording a  -4.3% annual return. Going back to the start of the century, the DJCI Gold produced a 7.8% annual return, compared to a 7% return for the S&P 500 during that time. Adjusting for volatility, gold has demonstrated better risk-adjusted returns than stocks, with a Sharpe ratio of 0.48 versus 0.45 for equities.

Central banks appear to have taken notice in the outperformance of gold. After trading flat for much of the year, the S&P GSCI Gold and S&P GSCI Silver staged a fourth quarter rally, rising 11.4% and 7.2%, respectively. Global central banks supported gold demand, adding over 800 metric tons to their portfolios with data through October. Gold has historically provided investors an alternative to fiat currency. Digging into the central bank purchase data, countries including Russia and China have led the increase in central bank holdings. Foreign central banks appear to increasingly value gold’s hedge against inflation, debt default and dollarization.

Commodity Market Recap

Energy weighed down commodity performance in 2023, with the S&P GSCI Natural Gas falling 62.6%. The decline reflects a return to the longer-term price levels observed during the past decade, as well as robust production along with reduced export demand of liquefied natural gas. The broader petroleum complex was mixed, as lower oil didn’t translate to falling gas prices. The S&P GSCI Energy dropped 5.2% for the year, while the S&P GSCI Unleaded Gasoline rose 5.0%.

As the leading dividend index provider, S&P DJI is constantly looking for new approaches and methodologies to bring novel ideas to the market. Our recently launched S&P 500^® High Dividend Growth Index is a prime example of this innovative thinking, as it incorporates a forward-looking assessment into its methodology. This index tracks companies in the S&P 500 that have not only offered consistent or growing dividends in the past but also have the highest forecast dividend yield growth. In doing so, constituents are selected based on what dividend they are expected to pay instead of being assessed solely on what they have paid in the past.

In this blog, we will present the index methodology, introduce the S&P Global Dividend Forecasting Dataset and review the index’s risk/performance profile.

Methodology Overview

To be eligible for selection, constituents must have maintained or grown their historic dividends for 5 consecutive years and must also be projected to do so over the next 12 months.

From this pool of eligible stocks, the top 100 constituents with the highest forecast dividend growth scores are selected. The score is computed as the 12-month forecast yield minus the 12-month historical yield.

Finally, the index constituents are weighted by forecast dividend yield with constraints placed on individual stocks and GICS^® sectors.¹ To reduce turnover, the index uses a 20% buffer.

Dividend Forecasting Data

The Dividend Forecast Dataset is sourced from S&P Global Market Intelligence, which is another division within S&P Global. This team has more than 40 professionals performing fundamental analysis with the goal of delivering precise forecasts of the size and timing of dividend payments. They serve over 150 customers across the world, including most of the top-tier global banks.²

Performance Review

The performance statistics that follow are calculated starting in 2010, when the data for the dividend forecasting dataset began. Hence, the back-tested data incorporates forecasts that were stored as and when they were made, with no look-ahead bias.

While this period has been a strong performance period for the S&P 500, the S&P 500 High Dividend Growth Index has more than kept pace. Since 2010, it has had an annualized return of 11.94% while delivering a significantly higher yield.

Downside Protection and Upside Participation

The historical capture ratios versus the S&P 500 show that the S&P 500 High Dividend Growth Index has exhibited moderately defensive characteristics (94.5% downside capture). Moreover, on average, the index has historically participated in 96% of the market return in up-market periods.³ This is higher than most dividend strategies and is likely a result of its relatively lower value tilt and higher growth tilt.

Historical Yield and Dividend Growth Analysis

This innovative methodology offers a unique blend of dividend growth and high yield. Since 2010, the index has had an average yield of over 3%, comfortably outperforming its benchmark and other strategies within the dividend growth category.

Impressively, for a dividend strategy offering a high yield, the index also has a high annualized dividend growth rate. From 2010 to 2022, the S&P 500 High Dividend Growth Index grew its dividend at an annual rate of 13.8% (see Exhibit 3). This outpaces the long-term U.S. inflation rate, even with the recent spike in inflation over the last few years.

Factor Exposure

Exhibit 4 shows the factor exposure difference between the S&P 500 High Dividend Growth Index and S&P 500 High Dividend Index in terms of Axioma Risk Model Factor Z-scores. The S&P 500 High Dividend Growth Index demonstrated less value tilt and had lower dividend yield than the S&P 500 High Dividend Index.

Sector Exposure

From a sector perspective, the S&P 500 High Dividend Growth Index had lower sector weights in Utilities (-11.5%) and Real Estate (-5.8%), while having higher sector weights in Industrials (9.6%) and Information Technology (6.3%) than the S&P 500 High Dividend Index (see Exhibit 5).

In a market full of passive dividend solutions, the S&P 500 High Dividend Growth Index stands out by utilizing a forward-looking approach while historically offering high dividend growth and high yield. This index’s historical lower value tilt and higher growth tilt may help to avoid sacrificing potential upside when seeking high yield.

¹ For further information about the index design, please see the S&P 500 High Dividend Growth Index Methodology.

² For more information, please see this link.

³ The market is defined as the monthly performance of S&P 500 benchmarks from April 16, 2010, to Oct. 31, 2023.

Clean energy has certainly been a prominent topic in 2023, especially among those at the front and center of discussions at the United Nations Climate Change Conference (COP28) in Dubai, which just concluded. As we are approaching the end of 2023, we wanted to review the results of the S&P Global Clean Energy Index Series rebalance from October and share some of the key developments from the second half of the year in the clean energy space.

October Rebalance

Launched in 2007, the S&P Global Clean Energy Index has been a benchmark to measure clean energy-related companies’ performance over the past 16 years. In April 2021, we also launched the S&P Global Clean Energy Select Index, which is designed to measure the 30 largest companies in global clean energy businesses that are listed on developed market exchanges.

Both the S&P Global Clean Energy Index and the S&P Global Clean Energy Select Index went through a semiannual rebalance on Oct. 20, 2023. In the index methodology, we assign companies to four buckets of exposure scores from 0 to 1 with an increment of 0.25 to measure their purity of exposure to the clean energy business. Exhibit 1 shows the change in exposure before and after the October rebalance. We can see that for the S&P Global Clean Energy Index, post rebalancing, we have 10 more companies with an exposure score of 1 being added to the index. The weighted average exposure score of the index improved from 0.92 to 0.95. This shows the effect of rebalancing to improve the purity of index exposure to clean energy companies. The S&P Global Clean Energy Select Index, on the other hand, selects 30 companies with an exposure score of 1 listed in the developed market exchanges.

On the market allocation breakdown, the major change of the S&P Global Clean Energy Index post rebalancing is the 3.11% weight increase in India and 2.72% weight increase in China, together with a 5.01% weight decrease in Spain. For the S&P Global Clean Energy Select Index, the weight of the U.S. increased by 11.86%, with a drop of 6.19% in New Zealand and a drop of 5.41% in Brazil.

S&P Global Clean Energy Index Performance YTD in 2023

After outperforming the S&P Global BMI in 2022, both the S&P Global Clean Energy Index and the S&P Global Clean Energy Select Index underperformed YTD through the end of November 2023. The S&P Global Clean Energy Select Index was down 21.49% and the S&P Global Clean Energy Index was down 27.88% in USD total return terms. There was significant dispersion seen among constituents; some of the performance laggers included Sunpower Corp (-76.98%), SolarEdge Technologies (-71.98%) and Plug Power (-67.34%), while Chubu Electric Power (up 38.24%), VERBUND AG (up 16.54%) and First Solar (up 5.33%) made up for some of the loss with positive performance contributions.

Despite the performance headwind, we continue to see encouraging discussions around the world on the energy transition, including the following selected highlights.

Key Developments

The International Energy Agency (IEA) Released the World Energy Outlook 2023

In October, the IEA released the World Energy Outlook 2023, in which it says that the use of fossil fuels is not declining quickly enough, but the move to renewable energy is “unstoppable”.¹ According to the report, “Tripling renewable energy capacity, doubling the pace of energy efficiency improvements to 4% per year, ramping up electrification and slashing methane emissions from fossil fuel operations together provide more than 80% of the emissions reductions needed by 2030 to put the energy sector on a pathway to limit warming to 1.5 °C.”²

Global Pledge on Renewables and Energy Efficiency

At COP28, the Global Pledge on Renewables and Energy Efficiency was signed by 121 countries.³ Among other objectives, those who sign the pledge commit to “work together to triple the world’s installed renewable energy generation capacity to at least 11,000 GW by 2030, taking into consideration different starting points and national circumstances.”⁴

COP 28 Concluded with a Deal to Transition away from All Fossil Fuels

After many nights of discussion, almost 200 countries reached a deal to transition away from all fossil fuels. This first-ever agreement once again enforces the global commitment to net zero emissions by 2050. Although the deal is not legally binding, the message is loud and clear. It is now on each country to set up its own agenda in order to phase out fossil fuels “in a just, orderly and equitable manner.”⁵

¹ https://www.bbc.com/news/science-environment-67198206

² IEA (2023), World Energy Outlook 2023, IEA, Paris. https://www.iea.org/reports/world-energy-outlook-2023

³ https://energy.ec.europa.eu/news/cop28-eu-energy-days-focus-implementing-clean-energy-transition-after-launch-global-pledge-2023-12-04_en

⁴ https://energy.ec.europa.eu/system/files/2023-12/Global_Renewables_and_Energy_Efficiency_Pledge.pdf

⁵ https://www.wsj.com/business/energy-oil/cop28-leaders-call-for-transitioning-away-from-fossil-fuels-in-final-push-at-climate-talks-48f4b1c3

“…sometimes I’ve believed as many as six impossible things before breakfast.”
– The White Queen, Through the Looking Glass

Should an asset owner rely on historical performance data to select managers? The efficacy of doing so depends on the answers to three questions:

• What fraction of the manager universe is truly gifted?
• How gifted are they?
• How lucky might the “ordinary” managers be?

For example: Suppose we assume that 60% of all managers are “gifted” and 40% are “ordinary,” that a gifted manager has a 75% probability of achieving above-median results, and that an ordinary manager has a 12.5% chance of doing the same. Exhibit 1 shows some implications of these assumptions for a 1000-manager universe.

Exhibit 1 contains both good and bad news for our hypothetical asset owner. The good news is that after one period, 90% (450/500) of the above-median managers are genuinely gifted; if our assumptions are correct, hiring only from the above-median pool will lift the odds of success. The bad news is that our assumptions are almost certainly incorrect, not to say wildly unrealistic. Why? Because these assumptions imply that 69% (344/500) of period 1’s above-median managers will also be above median in period 2—a persistence rate far greater than those we actually observe. Exhibit 1 is, sadly, an artifact of wishful thinking.

If Exhibit 1’s assumptions are clearly wrong, what alternatives might be more realistic? To be more modest, we can reduce the population of gifted managers from 60% to one-third, reduce their probability of ranking above median from 75% to 60%, and narrow the gap between the gifted and the ordinary by setting the ordinary managers’ chance of being above median at 45%. As before, Exhibit 2 contains both good and bad news for our hypothetical asset owner.

The good news is that using Exhibit 2’s assumptions, 51% (255/500) of period 1’s above-median managers should repeat that performance in period 2. Although we don’t often see results that good, 51% persistence is not unheard of, and so Exhibit 2 is at least a somewhat plausible model of reality.

The bad news in Exhibit 2 is that only 40% (200/500) of period 1’s above-median managers are genuinely gifted; 60% of them got there through luck rather than skill. And perhaps worse news: only 47% (120/255) of the managers who are above median in two consecutive periods are genuinely gifted. In other words, an asset owner who hires from the above-median pool is more likely to get an ordinary manager than a gifted one. Even if we assume that genuinely gifted managers exist—and that they stay gifted over time—hiring an above-median performer provides a less-than-even chance of finding the gifted manager we are seeking.

Active management is difficult, as readers of our SPIVA Scorecards know well; identifying outstanding managers is perhaps equally challenging. Relying on historical performance rankings is unlikely to be helpful.

Make more informed sustainability decisions with deeper data – our indices are powered by analytics from the world-renowned S&P Global Corporate Sustainability Assessment.