We hypothesize that, unlike for-profit firms, nonprofit hospitals have incentives to manage earnings to a range just above zero. We consider two ways managers can achieve this. They can adjust discretionary spending [Hoerger, T.J., 1991. 'Profit' variability in for-profit and not-for-profit hospitals. Journal of Health Economics 10, 259-289.] and/or they can adjust accounting accruals using the flexibility inherent in Generally Accepted Accounting Principles (GAAP). To test our hypothesis we use regressions as well as tests of the distribution of earnings by Burgstahler and Dichev [Burgstahler, D., Dichev, I., 1997. Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics 24, 99-126.] on a sample of 1,204 hospitals and 8,179 hospital-year observations. Our tests support the use of discretionary spending and accounting accrual management. Like Hoerger (1991), we find evidence that nonprofit hospitals adjust discretionary spending to manage earnings. However, we also find significant use of discretionary accruals (e.g., adjustments to the third-party-allowance, and allowance for doubtful accounts) to meet earnings objectives. These findings have two important implications. First, the previous evidence by Hoerger that nonprofit hospitals show less variation in income may at least partly be explained by an accounting phenomenon. Second, our findings provide guidance to users of these financial statements in predicting the direction of likely bias in reported earnings.