If a socially conscious investor wants to have real impact on the way a company operates, what is the best way to achieve it? Given the increased interest in this question by policymakers, institutional investors, and retail investors alike, evaluating what is likely to work, and under what circumstances, deserves careful theoretical and empirical research. In our paper The Impact of Impact Investing which is forthcoming in the Journal of Financial Economics, we provide theoretical and empirical evidence that one often-advertised way of socially responsible investing is unlikely to materially change the way companies operate: divestment in secondary markets. Direct investments in primary markets, and increasing voting rights (engagement) by buying (rather than selling) stocks in secondary markets, both deserve careful attention.

Why is divestment so unsuccessful in achieving its advertised goals? Divestment typically involves selling the shares in harmful companies. However, this transaction in secondary markets merely shifts ownership rather than directly influencing the cash flows of the firm. The potential for divestment to affect corporate behavior therefore relies on two indirect mechanisms.  Either the new shareholders choose to exercise their rights of control thereby inducing socially responsible behavior, or the secondary market transaction changes the company’s cost of capital thereby affecting its growth prospects

The first mechanism can only be effective if the new investors are more inclined to push the company toward socially responsible practices than the old shareholders. However, this seems like an unlikely outcome of a divestment strategy since the motivation to sell is driven by a concern for social responsibility implying that the old shareholders are likely to care more than the new shareholders.

The second mechanism focuses on the cost of capital. When socially responsible investors sell their shares, the company’s stock price needs to fall to induce other investors to buy. This lower stock price increases the company’s cost of capital, meaning it needs to offer higher average returns going forward to its investors. As a result, it becomes harder for the company to fund new projects. With an increased cost of capital (discount rate), higher future benefits need to be achieved to overcome the current investment costs of any given project, i.e. a positive net present value (NPV). On the other hand, when socially conscious investors buy shares in companies with strong social and environmental practices, the stock price rises, lowering the company’s cost of capital and enabling greater growth through an increased number of positive NPV investment opportunities.

Consequently, for divestment to have a real effect, the change in the cost of capital must be substantial enough to alter the company’s decision-making. Our theoretical contribution is to provide a simple formula that calculates the change in the cost of capital resulting from divestment. This change is a function of three parameters: (1) the fraction of socially conscious capital, (2) the fraction of targeted (or `dirty’) firms in the economy, and (3) the return correlation between the dirty firms and the rest of the stock market (the `clean’ firms). Using our formula, we demonstrate that at current levels of socially conscious investing, the change is too small to materially affect company behavior.

Socially conscious investing currently represents less than 2% of stock market wealth in the United States, and the fraction of targeted firms is about a quarter of firms. Most importantly, we find that the correlation between the clean and dirty portfolios is very high. Large diversified portfolios of clean and dirty stocks are good substitutes to each other, implying that investors are willing to exchange them without much inducement. In our baseline calibration, we thus find that at current implementation levels, divestment causes a change in the cost of capital of only 0.44 basis points. This small change is unlikely to significantly influence the investment decisions of firms.  We further argue that for divestment to have a significant effect, socially responsible investors would need to account for over 80% of the market, an unrealistically high proportion.

Our empirical contribution is to corroborate this prediction in markets by studying index inclusions and exclusions in often-used stock market indices designed to capture social responsibility. Once again, we find effects that seem too small to matter for real capital allocation.

We focus on the index that the largest social index fund in the world, the Vanguard FTSE Social Index Fund, tracks, namely, the FTSE USA 4 Good Index. We find that the inclusion or exclusion of companies from this index, driven by changes in their ESG status, has little effect on their stock prices or cost of capital. This reinforces our conclusion that divestment in secondary markets has little to no effect on corporate behavior.

If divestment in secondary markets is unlikely to be effective, then what is? First, we argue that direct investment in the primary market, where an investor funds projects or companies with negative NPVs, is likely to have much greater impact. By funding these projects, an investor creates social value because without intervention the projects would not be undertaken. In contrast, refusing to fund positive NPV projects is unlikely to cause material change as there is an ample supply of other investors to step in and fund them. Second, socially responsible investors could engage directly with companies through their voting rights or by acquiring a majority stake and influencing company leadership. These last approaches, known as “impact investing,” could be more effective in promoting social goals since they require a smaller proportion of capital than divestment.