Applying Economics – Not Gut Feel – To ESG

Executives, investors, policymakers, the media, and business schools are taking environmental, social, and governance (“ESG”) issues more seriously than ever before. Given ESG’s potential to create long-term value for both shareholders and society, this attention is both much-needed and welcome. However, the enthusiasm for ESG – and, in particular, the pressure to do something or say something about ESG to demonstrate your commitment – can often lead to actions or statements that shoot from the hip and apply gut feel rather than being based on careful analysis.

One justification for shooting from the hip is that ESG is so new, and traditional finance research so focused on shareholder value, that there is no research to guide us. However, as explained in a recent paper, “The End of ESG”, ESG is both “extremely important and nothing special” – it’s no different to any other investment that creates long-term financial and social value – and decades of finance research have studied the risk and return to investment. My new paper, “Applying Economics – Not Gut Feel – To ESG” uses the insights from mainstream economics to help us understand ten key issues in ESG – and, in doing so, reaches different conclusions from conventional wisdom.

One common claim is that ESG risks – stranded assets, executive malfeasance, employee unrest, customer boycotts, regulatory fines, media shaming, and so on – increase the cost of capital. It seems a no-brainer that you should incorporate risks into valuations by increasing the discount rate – after all, Finance 101 tells you that the discount rate depends on risk. Indeed, the PRI’s (2016) comprehensive survey of valuation approaches used in practice concludes: “Some investors adjust the beta or discount rate used in company valuation models to reflect ESG factors: corporate governance, operational management, general quality of management, its strategic decision making etc.” It contains numerous case studies of investors that do so. One investor explains how they lower the discount rate for companies with above-average gender diversity, arguing that “Diversity also helps to reduce company-specific risk in the long term, leading to a lower cost of capital.” Similarly, question 12 of the specimen exam for the CFA UK Level 4 Certificate in ESG Investing is “What impact will a high ESG rating have on a company’s cost of capital?”. The answer key gives the correct response as “A: A lower cost of capital”.

But the main effect of a risk is to change the expected cash flows in the numerator of a valuation. If a cash flow should be $1 million, and there’s a 10% chance of a major ESG catastrophe that reduces it to $200,000, a 15% risk of a moderate disaster that lowers it to $300,000, and a 20% probability of a mild calamity that decreases it to $400,000, the expected cash flow becomes $695,000.

These risks need not increase the discount rate. Finance 101 tells us that the discount rate is affected only by market risk, and not by company-specific risk, in contrast to the statement by the investor in the PRI survey. The discount rate only increases if the risk of the disaster is correlated with market conditions – i.e. a disaster is more likely in bad times. In contrast, many ESG scandals, such as Volkswagen cheating emissions tests, Wells Fargo opening fake bank accounts, or Rio Tinto blowing up Juukan Gorge, are company-specific, not market-wide – they’re no more likely in a down market than an up market. In fact, one could argue that some ESG scandals are more likely in an up market – when times are good, companies get sloppy and don’t impose as tight controls.

It’s true that some ESG risks affect the broader market – for example, a carbon tax will impact many firms. However, what matters is not only whether a risk affects the broader market, but whether it’s more likely to manifest in up or down markets. As Giglio, Kelly, and Stroebel (2021) show, the government could be more likely to implement a carbon tax when the market is booming, since economic activity leads to high emissions and thus the need for a tax; then, brown assets actually bear lower market risk, since they do worse in a booming economy. Similarly, the model of Baker, Hollifield, and Osambela (2022) demonstrates that they are a hedge against the reduction in welfare that arises if the government fails to take action on climate change. Thus, while ESG risks definitely reduce expected cash flows, the impact on the discount rate is far from clear.

However, changing expected cash flows seems a hassle – you need to estimate the probability of different scenarios and what will happen to the firm in each scenario. Since such estimates will only be approximate, you’ll then need to do sensitivity analyses around your base case. Because this is cumbersome, some investors instead increase the discount rate because it’s easier – you just need to change one number – but the discount rate shouldn’t change if the risk is company-specific. Now it’s true that you can “fudge” the valuation by increasing the discount rate to reflect the effect of lower cash flows. But there’s no guidance as to how much this increase should be – there’s not even a logical midpoint about which to do a sensitivity analysis. It would be like incorporating the effect of higher energy prices by hiking the discount rate rather than raising the costs in the cash flow statement.

The classic textbook Principles of Corporate Finance by Brealey, Myers, Allen, and Edmans (2022) states the correct approach clearly: “Remember that a project’s cost of capital depends only on market risk. Diversifiable events can affect project cash flows but they do not increase the cost of capital… Don’t give in to the temptation to add fudge factors to the discount rate to offset things that could go wrong with the proposed investment… Adjust cash-flow forecasts instead. Fudge factors in discount rates are dangerous because they displace clear thinking about future cash flows.”

Note that there are reasons why brown companies might suffer a higher cost of capital, but they are more nuanced than “there’s a risk that cash flows fall”. The first is if the ESG risks are more likely to manifest in bad times – for example, if firms are more likely to engage in fraudulent behavior in recessions because they’re desperate, or if a climate disaster both reduces shareholder welfare and spurs government action. The second has nothing to do with risk. If investors have a taste for green stocks and dislike holding brown stocks, they’ll demand a higher cost of capital to own the latter (see Pastor, Stambaugh, and Taylor (2021) for a model of both types of risk). However, unless either condition is satisfied, sustainability only has an effect on cash flows, and not the cost of capital.

The complete paper is available for download here. The full list the ten issues it tackles is below:

1. Shareholder Value is Short-Termist (No, shareholder value is a long-term concept).
2. Shareholder Primacy Leads to an Exclusive Focus on Shareholder Value (No, shareholders have objectives other than shareholder value).
3. Sustainability Risks Increase the Cost of Capital (No, sustainability risks lower expected cash flows).
4. Sustainable Stocks Earn Higher Returns (No, sustainability may be priced in; tastes for sustainable stocks lead to lower returns).
5. Climate Risk is Investment Risk (No, climate risk is an unpriced externality).
6. A Company’s ESG Metrics Capture Its Impact on Society (No, partial equilibrium differs from general equilibrium).
7. More ESG Is Always Better (No, ESG exhibits diminishing returns and trade-offs exist).
8. More Investor Engagement Is Always Better (No, investors may be uninformed or undermine managerial initiative).
9. You Improve ESG Performance By Paying For ESG Performance (No, paying for some ESG dimensions will cause firms to underweight others).
10. Market Failures Justify Regulatory Intervention (No, regulatory intervention is only justified when market failure exceeds regulatory failure).