This study identifies a J-shaped relation between dividends and firm value. On average, top-dividend-payers are valued higher than all other firms, while non-dividend-payers are valued higher than low-dividend-payers. This J-shaped relation is highly stable over time as it is observed in nearly every year over the period 1962–2010, and it remains significant after controlling for firm characteristics such as profitability, growth and firm size. We also find similar J-shaped relations in stock markets outside the U.S. Our evidence suggests that dividend theories, such as the dividend catering, free-cash-flow and dividend clientele hypotheses, are inadequate to explain the J-shaped relation. It does not appear that the J-shaped pattern reflects mispricing. There is mixed evidence as to whether the J-shaped relation is driven by dividends or unobservable firm characteristics.

This study uses the quantile regression method to determine the relative importance of internal and external sources of funds in financing corporate investments across different levels of investment activities. Our findings lend support to the first (internal–external) rung of the pecking order, as investments are more responsive to internal funds than to external funds at almost all investment levels except at top investment levels. However, empirical support for the second (debt–equity) rung of the pecking order is tenuous, as investments are more responsive to equity issuance than to debt issuance at low and medium investment levels. Intriguingly, firms rely primarily on internal funds for organic growth (via capital expenditures or R&D spending), whereas they use debt capital heavily in financing acquisition-led growth. Finally, firms at high investment levels hold large cash holdings and save a significantly large fraction of equity issue proceeds as cash.