Executive compensation, especially cash bonus compensation, has come under fire by the Securities and Exchange Commission (SEC), the US Federal government, and the media for its role in the current economic crisis. Specifically, the SEC has argued that some compensation packages provide incentives for risk-taking that may undermine shareholder value over the long-term. Short-term incentive payments to executives in the form of cash bonuses are mostly contingent on reaching targets of accounting-related measures or financial performance measures (FPMs). However, the incentives from these payments may lead to accrual manipulation and earnings management (EM). Alternative measures are non-financial performance measures (NFPMs). We expect that firms that employ NFPMs in bonus contracts will have a lower prevalence of EM, since these measures tend to focus executives on the long-term. In this paper, we examine the type of performance measures used by firms in the S&P 500 index in their cash bonus compensation. We find that firms that use both FPMs and NFPMs have lower discretionary accruals compared to firms that use only FPMs, consistent with lower income-increasing EM. However, we do not find evidence of a reduction in EM behavior using the incidence of meeting or just beating analyst earnings benchmarks, another common EM proxy. In additional tests on a subset of firms with equity offerings, in which incentives for income-increasing manipulation are likely high, we find that firms with NFPMs have lower discretionary accruals. The implication is that NFPMs can be used in compensation contracts to reduce EM behavior and mitigate erroneous executive compensation. This is important to investors as well as regulators, especially in light of the recent debate on compensation reform.