This paper reviews the literature on R&D to provide guidelines for recent efforts to include R&D in the national income accounts. The main conclusions are:
1. Measures of R&D as an asset held by a particular owner must be complemented by estimates of the spillover effect of R&D in order to obtain a reliable measure of the overall effect of R&D on productivity growth.
2. If research financed by the government and research financed by business are both counted as investment, some double counting occurs and growth accounting analysis overstates the role of research relative to other factors.
3. The overall rate of return to R&D is very large, perhaps 25 percent as a private return and a total of 65 percent for social returns. However, these returns apply only to privately financed R&D in industry. Returns to many forms of publicly financed R&D are near zero.
4. Firm R&D should be allocated to the different industries in which a firm produces, rather than all credited to the firm's main industry. An allocation procedure is proposed.
5. Much further work needs to be carried out to understand how R&D conducted in the richest countries is transmitted to developing countries. Detailed microeconomic data on firms or establishments in developing nations will be necessary to understand the channels of technology transfer more fully.