How Far Do Stock Markets Influence the Real Economy?
There is a long-standing debate among economists about whether or not stock markets influence the real economy. One view is that stock markets are just a “sideshow,” since firms raise relatively little cash from public markets. Accordingly, stock markets play a minor role in channelling savings to investment, and stock prices passively reflect real economic conditions without affecting them. Opponents to this view argue instead that stock markets actively influence real outcomes through various channels (e.g., by relaxing financial constraints).
One distinctive feature of stock markets is their role as information aggregators. Thousands of dispersed investors have incentives to produce information about asset payoffs (e.g., cash flows on a growth opportunity for a firm). Indeed, by trading on their private information, investors can earn a return to cover costs of producing information. As investors buy undervalued assets or sell overvalued assets based on their private information, they collectively make stock prices more accurate predictors of firms’ future prospects. This ‘wisdom of crowds’ aspect of stock markets can be useful for decision-markers (e.g., firms’ managers, capital providers, regulators or central bankers) because they can use stock markets, in addition to their own information, as a large-scale predictive market.
In this view, the stock market is not a sideshow; it matters for the real economy. Stock prices actively influence firms’ capital allocation decisions because their movements affect managers’ beliefs about the value of their growth opportunities. This possibility has received increased attention from financial economists in recent years, and there is growing evidence supporting the view that decision-makers rely on stock prices as a source of information. In particular, cross-sectional variations in the correlation between investment and stock prices are consistent with implications of the hypothesis that managers use stock prices as a source of information. For instance, firms’ investment is more sensitive to their stock prices when these prices are more informative.
The Stock Market Also Affects Product Market Strategies
If decision-makers can improve their decisions by relying on information from stock prices, they have strong incentives to take actions that render prices more informative for them. In a paper we published earlier this year, we argue that these actions can also have real effects. We consider the particular case of product differentiation choices by firms. By differentiating their products from rivals, firms can gain market shares and increase their value. We show, however, that differentiation can bear an informational cost, as it makes the information that a firm’ manager can extract from its stock price less informative. The reason is that while stock prices aggregate investors’ private information, they also typically contain noise, which limits how informative they are. We show that it is easier for investors to filter out the noise from stock prices when they observe the prices of several firms whose cash flows load on the same fundamentals (e.g., they are exposed to the same demand shocks). As a result, the stock prices of firms following similar product market strategies are collectively more informative (i.e., contain less noise). One direct consequence is that a firm constrains its ability to learn from the stock market if it differentiates too much from its rivals. Overall, the equilibrium levels of product differentiation and stock price informativeness in the economy are jointly determined.
Going Public and Product Differentiation
One implication of our theory is that firms’ incentive to differentiate from rivals should increase when they publicly list their share on stock markets. In this case, managers switch from an environment in which they cannot learn from their own stock price to an environment in which they can. Thus, at the margin, an initial public offering mitigates the informational cost of differentiation, and therefore increases firms’ incentive to differentiate and opt for more unique product market strategies.
We provide evidence supporting this prediction using a sample of U.S. initial public offerings between 1996 and 2011. We measure the differentiation of a firm’s product offering relative to that of related (peer) firms using the innovative approach developed by Gerard Hoberg and Gordon M. Phillips. For every pair of public firms in the U.S., Hoberg and Phillips define an index of product similarity based on the relative number of words that firm-pairs share in the description of their products in their 10-K forms (these are forms that public firms must file every year with the U.S. Securities and Exchange Commission). We use (one minus) this index as our proxy for the level of product differentiation between each firm-pair.
Fig. 1 Firms that go public differentiate faster after their initial public offerings relative to rival firms.
Fig. 1 offers a visual representation of our main empirical finding. The blue line shows the evolution of differentiation between initial public offerings firms and their rivals (at the time of the initial public offerings) over the five years following the initial public offerings. The red line shows the evolution of differentiation between the rivals of each initial public offering firm and their own distinct rivals, which we use as controls for general trends in differentiation. Consistent with our theory, initial public offering firms increase their differentiation from their rivals in the years following their initial public offerings, and they do so at a faster rate (compare the blue and the red line) than the rivals do, relative to their own competitors. This result suggests that the initial public offering itself has an effect on differentiation choices. rather than other factors affecting both initial public offering firms and firms that have already been public for several years.
In additional tests, we show that the positive relationship between an initial public offering and changes in product differentiation is stronger when managers appear to possess better private information, or when rivals’ stock prices provide less information. Thus, in line with our theory, managers choose to differentiate more when the informational cost of differentiation is smaller (e.g., because managers are already very well-informed about their product markets).
Of course, there might be other factors explaining the pattern observed in Fig. 1. In particular, by going public, a firm might be able to more easily finance expenses necessary for differentiation, such as research and development or marketing. However, we establish that the pattern documented in Fig. 1 persists even after controlling for differences in access to financing and innovation policies (e.g., research and development expenses).
One important take-away of our analysis is that firms can increase the precision of signals that they get from the stock market by imitating each other. This effect has important implications for the structure of industries, innovation strategies, the diversity of product offering for consumers and scope for diversification for investors. In particular, one should account for the extent to which managers rely on stock market information in analyzing the determinants of industrial structure or co-movements in stock returns, which are higher when firms differentiate less.
The possibility for firms to obtain more accurate signals from stock prices when they imitate each other creates coordination problems among firms. Consider the situation in which firms choose between two strategies, A and B. If firms expect other firms to choose strategy A, the informational cost of choosing B will be relatively high (the stock market will produce less accurate signals about B, unless a large number of investors are willing to get informed about B). Thus, each firm has a natural incentive to choose strategy A. It might be, however, that collectively, all firms might be better with strategy B. Hence, firms’ incentive to learn from the stock market might lead them to cluster on inefficient strategies.
This problem is less likely to exist if investors have high incentives to produce information about new product market strategies. This might not be the case, however, because of economies of scope in information production. When firms choose the same strategy, information producers (e.g., financial analysts) can better amortize their cost of producing information because their information is useful to evaluate all firms following the same strategy. This suggests that considering the incentive structure of information producers in financial markets is important to understand whether the stock market provides too little incentive to differentiate.
This piece first appeared on VoxEU.