An interview with Professor Ken Merchant
Interview by: André A. de Waal
Kenneth A. Merchant is the Deloitte & Touche LLP Chair of Accountancy at the University of Southern California. He also serves on the boards of two US firms, one that is listed on the New York Stock Exchange and one that is privately owned.
His book Modern Management Control Systems from 1998 and its successors have had a lasting impact in both the university and the business worlds. Merchant's model of action controls, results controls and personnel and cultural controls has turned out to be one with real practical value.
Professor Merchant held the keynote speech during the lustrum congress of the Dutch Controllers Institute in November 2004.
Professor Merchant, what are currently your fields of research?
I'm at the moment busy with four topics, some of which are related. At present we are studying which incentives are applied within performance management systems (PMS) of car dealerships in The Netherlands, China and the USA. How are these systems constructed in the different countries and do cultural issues cause the differences between them? A question in respect to this is why these systems can differ so much within the same industry. The second research topic is the study of incentives that are applied within loss-making companies. The central question here is: why aren't incentives in these companies based on accounting measures of performance? Why don't managers get a bonus if they stop the loss-creating activities? Instead the incentives seem to be based largely on non-financial measures of performance or merely subjective performance evaluations. I'm also busy with a research project into the evolution of management accounting systems in China, especially regarding what happens to these systems in the transition of state-owned companies to privatized enterprises. You notice that the systems are 30 years behind because they are still organized to support the traditional goal of providing as many people with a job as possible. Finally, we are testing whether non-financial measures like customer satisfaction and employee satisfaction are reliable leading indicators of future profitability. Ah yes, I'm also thinking about studying the relation between corporate governance and PMS, but it is difficult to get access to boards of directors to get the needed data.
What developments do you see in the areas of management control and performance management?
An important development is the connection between leading non-financial indicators and financial indicators. The question is: do non-financial indicators provide good leading indications of value creation? The limitation here is obtaining good quality data with which one can calculate these indicators. An increasingly important skill of managers is to be able to identify, make and maintain relations with the people in the organization that can provide the required data and that are willing to openly and honestly share this data. Another development is cross-cultural performance management, as we have just discussed: how do effective PMS in different countries look? These are the two most important trends, I think activity based management, value based management and EVA as research topics have passed their peak. All companies who were interested in these techniques have by now looked at them and either implemented them or not.
I just finished my research into trends and developments that will keep Dutch organizations occupied this year. As the six most important trends I discovered the following: make information flows more reliable more forecasting, add more non-financial information, better implementations, more ICT and focus on integrity. Are these also the trends in your home country the United States?
In the States at this moment only one thing seems to be important: do we comply with Sarbanes-Oxley? All other things have been put on hold, probably until deep in 2005. People are not worried in the least about rolling forecasts and such, only information and communication technology receives some attention. Integrity is important but is seen as part of Sarbox, namely the setting up of a code of ethics and whistleblower programs. In relation to the further development of non-financial indicators American academics are interested but American companies have not shown much interest. Incidentally, I'm of the opinion that Sarbox will not prevent other accounting scandals from happening. Enron for example occurred because of top management's doing; it had nothing to do with control systems.
What is your opinion on the balanced scorecard of Kaplan and Norton, and the levers of control of Simons?
The balanced scorecard (BSC) has been oversold. Normative statements are made, such as that the BSC is suitable for every company in every situation and will have positive effects, and there is very little scientific evidence to back up these statements. The advocates basically sell everybody a hammer and then state that everything is a nail, hitting it as much as possible will solve all problems. Take a good look at the BSC. I think there are too many performance indicators in it, easily twenty or more, which is far too much for an average manager to handle. The question how then to link all these indicators to the reward system has still not been answered satisfactorily. Next to this, the average BSC is not balanced at all. There is still a strong emphasis on the financial indicators. Finally we do not know if and how the BSC works. We do know that when it does not work, meaning the organization is not obtaining better results, Kaplan and Norton always counter that the scorecard has not been implemented properly. At the moment strategy maps are hot but I seriously wonder: do we really get a better strategy by using this framework? Again the evidence is lacking. In relation to Simons I don't think the contribution has been that great. He did show us convincingly that the design of a control system is not all that is important. Also important is the way in which the system is used. But I don't think that Simons framework has caught on well. I have never heard any managers claim to need more or less diagnostic- or interactive-type controls.
You state that performance management systems world wide will converge. Which methods and performance indicators do you see "surviving"?
I do state that performance measurement techniques will converge. There will be greater emphasis on what I call "results controls" which involve rewards for good performance. However, differences will persist in firms' reward allocation techniques, levels of merit pay, and types of performance standards. I expect that several specific summary financial measures of performance, like ROA, ROE, and residual income, will to dominate the PMS of the future, but they will be supplemented with a limited number of non-financial indicators. The latter will depend on the industry in which the organization operates. There will certainly be indicators which give information on the quality of personnel because it becomes more and more crucial to the success of a company. EVA® in its full, complex form will probably not be one of the survivors because, despite all the marketing claims, objective, scientific studies have not found a strong relation between this indicator and value creation.
What do you recommend us to use then?
In my opinion there are two best practice measurement methods. With the first method an organization uses summary financial measures, that is to say indicators that adequately reflect the financial results of the company or organizational unit. These measures allow managers maximum operating freedom. They can follow whatever path they think will best yield the desired result. But in many cases the measurement horizon has to be extended for these indicators, from a quarterly outlook to a period of multiple years. The advantages of this are that a better picture is obtained of the long term performance of the organization and the operational freedom managers get to try out new things, like paying more attention to innovation, without directly being punished by lagging quarterly figures. The second method uses a combination of lagging indicators of accomplishments (e.g. earnings, ROA) and leading indicators of future accomplishments which are often called "value drivers". Multiple indicators give managers some guidance as to how to create value. It is important to make these value drivers specific and tangible on every organizational level and to keep their number limited. Probably three to eight indicators will do the trick on each level, but we need more research to establish the definitive number, which will vary across settings. A nice task for both researchers and practitioners!