The Validity of ROI as a Measure of Business Performance
检验了会计投资回报率（ROI）作为经济回报率衡量指标的有效性，通过分析公司层面ROI与股票回报的关联性，发现ROI的有效性存疑。
Business performance at the corporate level is widely assessed by the return, through changes in the price the stock and dividends, to shareholders. However, corporate level performance analysis, whether for strategic or public policy purposes, is inappropriate in many instances because the heterogeneity the corporation's operations. The strategic business units (SBUs) comprising a corporation are often very diverse. Profitability analysis grouping these entities together yields few the insights needed for strategic or public policy decision making. Because this heterogeneity, profitability analysis is often advocated and undertaken at the SBU level. The absence an equity market at the SBU level means that a measuring instrument other than that based on stock price must be used to assess profitability. Due to the lack alternatives, the performance an SBU is almost always based on accounting data. In particular, the accounting measure, return on investment (ROI), is widely regarded as the most useful measure and ultimate bottom line test business performance (James Reese and William Cool, 1978). It is used both as an objective management and as a dependent/ criterion variable to evaluate the effect various factors on performance. Despite its widespread use, ROI has been extensively criticized as being a inadequate indicator economic rate return (G. C. Harcourt, 1965; Ezra Solomon, 1971; Franklin Fisher and John McGowan, 1983). ROI, most typically defined as Net Income,/ Total Assets,-1, does not properly relate the stream profits to the investment that produced it. The earnings numerator is a consequence investment decisions made in the past, but the assets denominator can be expected to have had influence on not only past and current earnings but also to have influence on future earnings. Because this failure to produce an accurate mapping, ROI has been criticized as being so seriously flawed that it bears little, if any, resemblance to the crucial concept internal or economic rate return. The shortcomings ROI are said to be so severe that its cross-sectional variations can be completely explained by the inappropriateness the measure. Due to a lack proven validity, empirical investigations using ROI have been labeled by Fisher and McGowan as totally misleading enterprises and by George Benston (1985) as of doubtful value. Those making use accounting ROI argue that the noise created by the accounting distortions need not be expected to drown out the underlying signal economic return contained in ROI and, therefore, ROI is still appropriate for use in analysis, (F. M. Scherer, 1979, and William Long and David Ravenscraft, 1984). This paper examines the validity ROI as a measure economic rate return by ascertaining the existence and extent the association corporate level accounting ROI with stock return, a widely accepted indicator business performance. The null hypothesis for the analysis is that ROI has no validity as a measure business performance and, as such, will not exhibit an association with stock return. While the tests provide direct evidence on the validity corporate level ROI, the same underlying theorized inadequacies accounting ROI are also present at the SBU level.1 Conclu-