Stein, Jeremy, “Efficient Capital Markets, Inefficient Firms: A Model of Myopic Corporate Behavior,” The Quarterly Journal of Economics (1989), 655-669.

Claim: Managers will invest myopically, even when they care about stock price and even when markets behave rationally.

Model:

• Stock markets determine company valuation based on current earnings.
• Managers have an incentive to forgo profitable investments now in order to boost earnings (cash flow) now.
• Market are rational, so they will expect managers to do this.
• Managers who care about stock prices will be trapped into behaving myopically, since they will be penalized if they do not boost earnings now.
• In a steady-state “signal-jamming” model, managers will inflate earnings by “borrowing” from the future (i.e. not investing optimally), and markets will correctly estimate this borrowing.
• Capital market pressure determines the strength of this phenomenon, and may take one of the following forms:
  • Threats of takeover (when a firm’s stock price is low)
  • Lack of financial slack (the ability to undertake investments without issuing new stock)
  • Distance between the firm and its creditors (creditors who are intimate with a firm will rely less on the stock price to appraise firm value)
  • The degree to which current earnings are a good signal of future earnings.

Conclusions:

• Myopic behavior, where managers underinvest to boost short-term earnings and market rationally expect this and adjust valuations accordingly, is a Nash equilibrium.
• This positive market reaction to announced investments is not evidence that managers do not underinvest, as argued by Jensen (1986).
  • Underinvesting managers only undertake the best projects, so markets are pleased when myopic managers invest.
• Startups with high stock prices and high investments are not evidence that managers are not myopic.
  • For startups, current earnings have little correlation with long-run success (most startups have negative earnings). The link between stock prices and underinvestment does not yet exist.
• Corporate divestitures and breakups are evidence supporting the signal-jamming model.
  • Markets can better interpret the investments and future earnings of stand-alone companies than of conglomerates, so breakups reduce the signal-jamming inefficiencies