Searching for an effective measure of R&D performance. (includes appendix)

Notes that measurement of R&D performance has always posed great problems due to the nature of R&D activity and the difficulties in identifying a tangible output. Also that performance measurement systems have often been built on input variables or on qualitative measures of R&D performance and to single out those related to activities under complete and partial control of R&D managers. Takes as a starting point the concept of economic value creation as a firm's normative objective and the analysis of the contribution of R&D to it. Builds on this a performance measurement system that leads to identify proxy measures of both R&D effectiveness and efficiency.

This paper is based on the results of the research work "La misura delle prestazioni per i progetti di R&D". The authors gratefully acknowledge the CNR who has financially supported the authors. However, Cristina Masella has written the introduction, sections 2 and the example; Vittorio Chiesa section 1 and section 3; while the authors have jointly written sections 4, 5 and 6

Introduction

The evaluation of the contribution of R&D to corporate profitability has always posed complex and multidimensional problems. Macroeconomic studies have shown that there is a strong relationship between long-term profitability and investments in R&D[1-3]. Nevertheless, in the strategic management field recent approaches have demonstrated that long-term growth can be brought back to the ability to identify and cultivate core capabilities and competences of the corporation and that this does not mean outspending rivals in R&D but being effective[4].

In other words, the present competitive environment, in which competition increases and the pace of technological change accelerates, stresses even more the need for deploying R&D investments more efficiently and more effectively The issues facing individual companies are how much to spend on R&D and how to monitor R&D effectiveness[5,6].

This paper attempts to put up a performance measurement system for the R&D function. It starts from the concept of economic value creation as a firm's normative objective and analyses the contribution of R&D. It draws on this a performance measurement system which leads to identify proxy measures of both R&D efficiency and effectiveness.

The paper is articulated as follows. First, a brief state of the art of R&D measures is presented. Then, the firm's long-term model of profitability is presented and discussed and the contribution of R&D identified. Finally the performance measurement system framework for the R&D function is presented.

State of the art

Measuring R&D performance has always posed many problems. This can be related to the nature of the R&D activity: first, the degree of uncertainty of an R&D activity is very high; second, once completed the R&D output is itself often highly fuzzy and not definable and, thus, not measurable; finally, the ultimate result of R&D activity can be viewed after years, once an innovation has been brought to the market, but, at this time, the outcome is the result of the efforts of both the R&D unit and the other company functions. For these reasons, R&D has always been treated as an expense centre and R&D planning has been the result of a negotiation between R&D and corporate on the amount of resources to allocate in R&D.

Contributions to our understanding of how to measure R&D performance have been given by works which have taken different angles of view. A taxonomy has been put forward taking into account two dimensions: when the evaluation is done and whether it is internal or external[7-10]. Here, we propose a modified taxonomy which also considers exante measurement. Figure 1 summarizes the main approaches to R&D performance measurement and classifies them according to the mentioned dimensions: the time when performance is measured (whether it occurs before, during or after the completion of an R&D project) and whether it involves external comparisons or not. These approaches can also be classified according to the unit of analysis: some contributions have focused on the individual R&D project and others on the R&D process (or, in other words, the whole function). Finally there is a problem related to the managerial aspect of a measurement system. To be effective, this should concern the domain within which a certain manager can take decisions. Because of the nature of R&D activities, R&D performance measurement systems often adopt variables which are not completely under control of R&D managers, as measures of the R&D success. This can be another dimension by which the R&D evaluation approaches could be classified.

[Figure 1 ILLUSTRATION OMITTED]

Traditionally measures have more frequently been related to R&D input than output. This lies on the belief that there should be a positive relationship between amount of resources allocated to R&D and R&D output and, therefore, the higher R&D expenses, the more effective the output. There is strong evidence that this is not true in a number of cases. Companies well known for their innovative capabilities often spend far less than do their less effective competitors. More complex approaches of the same kind look at the breadth and the depth of the firm's R&D skills and/or to the R&D organization as input variables which could provide explanations for superior performance. These approaches are grouped in quadrant 1 of Figure 1. In fact, they are based on input variables measures and thus undertake an ex-ante approach. Moreover, they involve external measures because competitors within the same industry represent the reference basis. The unit of analysis is the function as a whole.

Another approach is that of the R&D project evaluation procedure. This can be used exante for R&D project selection, but also in-process and after project termination. Actually, project evaluation is a continuous process. What changes over time is the quality of the information on which the …