Investors know that stock returns depend on both a firm’s current assets and growth potential. Current assets are easy to assess, but determining a firm’s growth opportunities is less clear cut. Dimitris Papanikolaou, an assistant professor of Finance at the Kellogg School of Management and Leonid Kogan, a professor at MIT, examined how the heterogeneity of firms’ growth opportunities leads to heterogeneity in their investment behavior and stock returns.
A firm’s share price has two fundamental components—the value of the firm’s assets in place and the value of the firm’s future growth opportunities. “We try to identify the composition of a firm’s assets, what part of a firm’s value consists of projects already implemented—assets already in place—versus assets that have yet to be implemented—future growth opportunities,” Papanikolaou says.
While a firm’s current assets may be stable, its growth opportunities may not. The difference in risk between these two could help explain differences in investment behavior and stock returns. But quantifying the riskiness of future opportunities is complicated. Though a firm’s total value is observable, it is difficult to estimate how much is derived from its existing assets versus its growth opportunities. Papanikolaou and Kogan had to develop a clever proxy that would reveal the otherwise unobservable growth opportunities.
Though a firm’s total value is observable, it is difficult to estimate how much is derived from its existing assets versus its growth opportunities.
One frequently used proxy for growth opportunities is Tobin’s Q, named after the late Nobel Prize-winning economist James Tobin. Tobin’s Q is a ratio that relates the market value of a firm to the replacement cost of its assets. The extent to which the former exceeds the latter indicates the firm’s future growth opportunities. In empirical applications, the book value of a firm’s assets often serves as a proxy for the replacement cost of capital.
While Tobin’s Q may be a useful measure for growth opportunities, it has limitations. “Tobin’s Q is a proxy for many things,” Kogan observes. “It incorporates information about both current assets as well as future growth. Because it contains information about many things, it is not a perfect proxy for anything.” Moreover, market value may include such subjective elements as analyst views and speculation. Book value may depend on subjective estimation of assets, which differs from the replacement cost of capital. Papanikolaou and Kogan needed a better proxy for growth opportunities.
So they constructed their own using an IMC portfolio, a portfolio that takes long positions on the firms that produce investment goods and short positions on the firms that produce consumption goods, or final goods. This portfolio mimics the factor of investment shocks, which as Kogan points out is based in theory rather than just derived from a statistical construct.
The completed proxy compares a firm’s stock return with the IMC portfolio. “We correlate a firm’s stock returns with the returns of the firms that produce investment goods,” Papanikolaou says, adding, “If a firm has a lot of growth opportunities, its stock returns are going to be more sensitive to the returns of firms that produce capital, and this enables us to predict the quantity of investment—as well as stock returns—at the firm level.”
The IMC portfolio allowed the authors to identify technological improvements that are specific to investment goods producers. Such advances tend to lower the cost and increase the quantity of new investment. “These technology shocks,” Papanikolaou notes, “may increase the supply of investment goods—there are more machines available—thereby tending to lower the price of investment goods.” Furthermore, “These technology shocks may increase the quality of the investment goods, so we have better computers and other equipment. Holding productivity fixed, it is basically easier for firms to invest.” This situation often benefits investment goods producers by shifting resources from the production of consumption goods to the production of investment goods. In the face of a technology shock, for example, high-growth-opportunity firms tend to invest at a higher rate than low-growth-opportunity firms.
Papanikolaou and Kogan’s approach not only helps clarify the effect of shocks, it also resolves those effects at the firm-level rather than the industry-level. “The investment behavior of the average firm in response to economic factors may be misleading, since not all firms are the same,” Papanikolaou says.
The researchers used a standard technique called “factor-mimicking portfolio,” but changed the way it was constructed. Factor-mimicking portfolios correlate dependent variables—here stock returns—with a portfolio that mimics an unobservable independent variable—like growth opportunities. Such portfolios are often created using statistical techniques, but Papanikolaou and Kogan constructed theirs using a theoretical approach. “This approach, using the IMC portfolio, lets us identify dispersion in growth opportunities among firms better than the current state of the art,” Papanikolaou says.
Studying how individual firms respond to shocks rather than relying on industry averages may yield additional valuable information compared to old methods. Papanikolaou points out that their approach using factor-mimicking portfolios could prove useful in identifying “what type of firm is more likely to be affected by a credit shock” like that seen in the recent financial crisis.