an, 1995), high objectivity (e.g. Bourguignon and Chiapello, 2005), high outcome orientation (e.g. Eisenhardt, 1989), low distortion (e.g. Baker, 2002), few in number and high understandability (e.g. Meyer, 2002). These attributes are now presented in turn. 1985) and objectively measured16 (Bourguignon and Chiapello, 2005). Traditional agency research has suggested that outcome and objective measures of performance should be linked to incentive pay, whereas behavioural and subjective measures of performance should be linked to base pay (Colon and Parks, 1990; Eisenhardt, 1985; 1988).Furthermore, it is known that users tend to distort the data used for measuring performance (Baker, 2002; Indjejikian, 1999). Thus, another characteristic that designers aiming to create high quality measures should take into consideration is the degree to which a measure can be manipulated by managerial actions. For instance, a study conducted by Eccles and Mavrinac (1995) has found that investors and financial analysts had the perception that some of the information reported to external shareholders had been prone to managerial manipulation. For instance, non-financial performance measures do not need to be audited before being reported to shareholders and this may motivate executives to distort some of their computations or chose for external reporting only the non-financial performance measures that show good firm results (Eccles and Mavrinac,According to agency researchers, together with the informativeness principle –that supports diverse measurement–, the controllability and congruency15 principle may also have an effect on the performance impact of incentive systems (Indjejikian, 1999). On the one hand, the controllability principle suggests that individuals should be evaluated and rewarded byperformance measures that they can have control over or significantly influence (Antle and Demski, 1988). On the other hand, the congruency principle specifies that performance measures should assess the achievement of goals that are congruent with principal’s business objectives (Anthony and Govindarajan, 1995).In spite of the informativeness, controllability and congruency agency principles, performance measurement research has also found other characteristics that may affect the likelihood of a measure to positively affect individual behaviour and, ultimately, firm performance. For instance, there is a vast body of research that shows that performance information used for compensation purposes should be outcome oriented (Eisenhardt,
đang được dịch, vui lòng đợi..
