Why do U.S. Firms Hold So Much More Cash than They Used To?

Bates, Thomas W., Kathleen M. Kahle, and René M. Stulz, 2009, “Why Do U.S. Firms Hold So Much More Cash than They Used To?” Journal of Finance, Vol 64, No. 5 (2009), 1985-2021.

Purpose:  To document the increase in U.S. firms’ cash holdings from 1980-2006 and investigate its causes.

Motivation:  U.S. firms in 2006 hold twice as much cash (as a percentage of assets) than they did in 1980.  The average firm in 2006 has enough cash on hand to pay off all its debts.  This cash buildup has excited considerable popular attention.

Findings:

  • Firms of all sizes have a much higher cash-to-asset ratio in 2006 than in 1980.
  • This also holds for firms with no foreign income, so repatriation taxes do not explain it.
  • Cash buildup is observed in firms that do not pay dividends, but not in dividend-payers.
  • Firms whose idiosyncratic risk increased saw more growth in cash holdings.
    • Firms in industries that grew riskier
    • Younger firms (with more recent IPOs)
  • As the increase in idiosyncratic risk has reversed in recent years, cash holdings have declined slightly.

Data/Methods:

  • Data is a sample of 13,599 U.S.-based, nonfinancial, non-utility firms with positive sales and positive assets, over the period 1980-2006 (from WRDS merged CRSP/Compustat database).
  • Look at descriptive statistics of the average cash ratios, with the data segmented by size, industry risk, year of IPO, dividends paid, and net income.
  • Ask whether firms’ cash demand curve shifted or whether firms moved along the curve.
    • Regress cash-to-assets ratio on several firm characteristics (R&D expenditures, size, etc).
  • Connect the rising cash levels to specific firm characteristics.
    • Use Fama-Macbeth regressions with 1980s data to predict cash holdings, based on several firm characteristics, in the 1990s and 2000s.
    • Compare prediction to actual holdings.
  • Examine whether increased cash holdings is due to agency problems.
    • Compare yearly average cash for quintiles based on the GIM index of management entrenchment.

Conclusions:

  • Firms whose cash holdings increased the most are newer, do not pay dividends, and are in industries that have become riskier since 1980.
  • Cash holdings rose because inventories fell, R&D expenses increased, capital expenditures fell, and cash flow risk increased.
  • Cash holdings did not increase more quickly for firms with entrenched management.