I examine whether dividend policy affects investment decisions through two-stage regressions with diagnostic analyses sufficient to highlight endogeneity. Contrary to Miller and Modigliani [1961, Dividend policy, growth, and the valuation of shares, Journal of Business 34, 411-433], neither the magnitude of dividends nor their changes are merely determined as the residual cash flows after investment spending. Results also indicate that discretionary accruals mitigate the competition for capital resources between investment decisions and dividend policy. However, the marginal propensity to pay dividends is similar between discretionary accruals and the other components of earnings.