AGRICULTURAL ECONOMICS Agricultural Economics 34 (2006) 273–280 Efﬁciency effects of agricultural economics research in the United States David E. Schimmelpfennig ∗, a , Christopher J. O’Donnell b , George W. Norton a,c a Economic Research Service, U.S. Department of Agriculture, 1800 M Street, NW, Room 4179, Washington DC 20036, USA b Centre for Efﬁciency and Productivity Analysis (CEPA), School of Economics, University of Queensland, Australia c Department of Agricultural and Applied Economics, Virginia Tech, Blacksburg, VA 24061, USA Received 19 May 2004; received in revised form 14 January 2005; accepted 5 April 2005 Abstract Allocations of research funds across programs are often made for efﬁciency reasons. Social science research is shown to have small, lagged but signiﬁcant effects on U.S. agricultural efﬁciency when public agricultural R&D and extension are simultaneously taken into account. Farm management and marketing research variables are used to explain variations in estimates of allocative and technical efﬁciency using a Bayesian approach that incorporates stylized facts concerning lagged research impacts in a way that is less restrictive than popular polynomial distributed lags. Results are reported in terms of means and standard deviations of estimated probability distributions of parameters and long-run total multipliers. Extension is estimated to have a greater impact on both allocative and technical efﬁciency than either R&D or social science research. JEL classiﬁcation: Z00 Keywords: Social science research impacts; Bayesian estimation 1. Introduction One of the primary areas of emphasis of the agricultural economics profession has been on assessing the beneﬁts of production-oriented agricultural research and extension (ARE) (e.g., Bredahl and Peterson, 1976; Huffman and Evenson, 1992). Alston et al. (2000) and Evenson (2003) document the breadth and depth of this effort. Assessment studies have of- ten provided the information needed to evaluate ARE for ac- countability purposes and to make resource allocation decisions across programs. More recently, efforts have been made to con- ceptualize and measure the beneﬁts or impacts of social science research (SSR) in agriculture (Gardner, 2004; Lindner, 1987; Norton and Alwang, 2004). These studies suggest that the pri- mary output of SSR is information. Thus, the problem of quan- tifying SSR impacts becomes a matter of valuing information. Economic surplus, decision theory, and econometric methods have all been considered for this purpose. Economic surplus analysis (ESA) is particularly useful for valuing economic information from individual projects or well- deﬁned programs (Alston et al., 1998). Combining ESA with decision theory may help in establishing causality between project-level SSR and eventual decisions by an economic agent ∗ Corresponding author: Tel.: (202) 694-5507; fax: (202) 694-5775. E-mail address: des@ers.usda.gov (D. E. Schimmelpfennig). (Gardner, 2004; Norton and Schuh, 1981; Schimmelpfennig and Norton, 2003). However, for accountability purposes, one often prefers to evaluate aggregate research programs rather than individual projects. For nonsocial science research, aggre- gate beneﬁts of public and private ARE have been evaluated using econometric estimates of production, productivity, proﬁt, and cost functions (e.g., Huffman and Evenson, 1993). A ben- eﬁt of the econometric approach is that it provides a measure of the statistical reliability of the results. However, aggregate econometric assessment of SSR is difﬁcult because of the di- versity of the SSR programs affecting agriculture, the difﬁ- culty of separating out the effects of social science from those of other disciplines, and the fact that the users of social sci- ence research information are often one step removed from the beneﬁciaries. For economic information that eventually affects producers, there have been a few econometric attempts to value the agri- cultural SSR input. For example, Norton (1987) assesses the impacts of farm management and marketing research and ex- tension (MMRE) on improving technical (TE) and allocative (AE) efﬁciency in U.S. agriculture using a proﬁt function. The rationale for this work is that TE measures the ability of the ﬁrm to minimize the inputs required to produce given outputs, and this is an information problem that SSR should be able to help solve. AE measures the inequality between the marginal rate of technical substitution for a pair of inputs and the ratio of their c  2006 International Association of Agricultural Economists