Fossil fuel divestment has been a dominant topic for socially responsible investors for the last few years, and a hotly debated topic for universities, cities and major endowments.
The divestment campaign started by writer and environmental advocate Bill McKibben continues to have strong momentum, supported by efforts from the organization he started, 350.org. Prominent commitments to divest fossil fuels have been announced by the Rockefeller Brothers Fund; cities including Portland, San Francisco and Seattle; and colleges such as Stanford, Hampshire and Pitzer. Fossil fuel divestment is now a mainstream issue that we hear about from clients who previously hadn't considered taking social issues into consideration for their investment portfolios.
Our latest research extends prior analysis of the investment implications of divestment, refining it to address the emerging consensus about what "fossil fuel divestment" means. We also explore a key consideration emerging among investors considering divestment, which is what to do with the proceeds from divesting fossil fuels. Recent experience in a declining oil price environment has provided some surprising results for some investors.
Divestment Analysis
We created a simulated portfolio that excludes the Energy sector within the S&P 500 Index to represent the "divestment" portfolio. The divestment portfolio excludes energy stocks, while increasing the weights of the remaining index constituents on a pro rata basis. We compared the divestment portfolio to the "no divestment" policy alternative, represented in our study by the S&P 500 index.
The study simulated historical performance by creating a full divestment portfolio as of the end of 1989. Simulated performance over the 25-year period provides support to both advocates and skeptics of fossil fuel divestment.
Simulated performance of the full divestment portfolio was virtually indistinguishable from that of the S&P 500 index, as shown in Chart 1.
Returns are virtually identical for the 25-year period ending December 31, 2014, implying no tradeoff between values and performance for divestment proponents. Investors with long time horizons may be comforted by these results.
However, the study went beyond a point-to-point analysis to examine the pattern of returns over three-year rolling periods. Over long periods of time, performance differences canceled out.
Over shorter time horizons, there are significant periods in which divestment falls out of favor. The full divestment portfolio has a significant performance advantage in the early years of the simulation and in recent years, while the S&P 500 index had a significant performance advantage during several years in the second decade of analysis, as seen in Chart 2.
The wide variations in performance may be easier for an institution to weather than an individual investor.
Given the minimal exposure of the full divestment portfolio to energy stocks, the study compared performance of the full divestment portfolio to the trend in oil prices during the simulation. Chart 3 illustrates that the doubling of oil prices from 2002 to 2004 coincides with the performance deterioration of the full divestment portfolio relative to the S&P 500 index. The subsequent doubling of oil prices in 2007 had a similar impact to performance.
Evaluating the risk of divestment
In addition to simulating performance retrospectively, the study examined prospective risk of the divestment portfolio relative to the S&P 500 Index. The study used standard deviation to measure projected risk in absolute terms, determining the projected variability of each portfolio. As shown in Table 1, the divestment portfolio has a slightly higher simulated (backward-looking) and projected standard deviation, but the differences are small.
Examining risk in relative terms, the study also examined predicted tracking error, which is a statistic that measures deviation from a target benchmark. The divestment portfolio has a predicted tracking error of 1.07%. In statistical terms, a portfolio with predicted tracking error of 1.07% is expected to have annual returns within plus or minus 1.07% of S&P 500 returns, two-thirds of the time.
Risk models and optimizers have inherent limitations; past performance and simulated results are no guarantee of future results. The study doesn't account for potential changes in company behavior resulting from engagement efforts, nor does it consider potential regulatory changes that could restrict the ability of companies to extract carbon.
Reinvestment Implications
Energy stocks were poor performers in 2014, suffering through an unexpected collapse of oil prices during the second half of the year. Many observers expected declining oil prices to be a positive development for portfolios that divested their energy stocks, but some investors experienced unexpected consequences.
A key consideration influencing investment results is the decision about how to reinvest the proceeds from divesting fossil fuels. There are a few distinct approaches that carry different advantages and disadvantages. The approaches we see most frequently include:
Pro-rata reallocation: The approach we used for the study used the proceeds from divestment to increase the weights of the remaining index constituents on a pro rata basis. This is the easiest approach to implement, though it may not be the optimal approach from a risk perspective.
Risk-based reallocation: The approach we often recommend for clients, in which divestment proceeds are redeployed into companies that may be statistically correlated but are not directly involved with energy activities. For example, industry groups such as construction and engineering, aerospace and defense, and machinery have high historic correlations with oil companies. Another option for investors with a global perspective is investing in the currencies or non-energy equities of countries that are major energy producers, such as Canada, Norway and Russia. The broader economy in these countries tends to do well when oil prices rise, so investing in them could be a way to benefit from rising oil prices without providing direct funding to fossil fuel companies.
Reinvest in clean energy-related stocks: The approach that is likely to be the most satisfying from a social value-driven perspective, but which carries risks that aren't always apparent to clients. Last year provides a good illustration, as oil industry leaders Exxon and Chevron experienced stock market declines of about 6% for the year. As a comparison, the Powershares Clean Energy ETF declined about 16%, First Solar about 18% and Yingli Green Energy more than 50%. The other side of the spectrum is represented by Vestas Wind Systems, which gained more than 20% in 2014 after being up more than 400% in 2013. The booming stock price of Vestas, however came after a three-year period in which it lost about 48% in 2010, 66% in 2011 and 46% in 2012!
Consequences of the rapid fall in oil prices included reduced demand for alternative sources of energy and a period of heightened risk aversion for investors, creating the dramatic performance results discussed anecdotally above. As a result, "old" energy companies such as Exxon and Chevron delivered better stock market performance than many "new" energy companies. Clean energy companies often have more volatile business models than the established companies targeted in divestment campaigns, featuring steeper adoption curves, a higher degree of technological and regulatory uncertainty, and less financial cushion. Clean energy companies may offer attractive long-term prospects, but it's important to recognize the higher near-term volatility they exhibit.
Closing Thoughts
Our investment analysis suggests that removing energy stocks from a well-diversified portfolio may have a small to moderate impact on investment risk; however, the magnitude of the impact is very much a function of the investor's time horizon. Endowments with a multi-decade time horizon are likely to be much better equipped to weather near-term volatility than an individual investor nearing retirement.
The question of reinvestment is equally important, and risk and reward associated with divestment are intricately linked to the decision about how to reinvest. As with the original divestment question, time horizon plays a critical role influencing the success or failure of a given strategy.
