55 Bulletin UASVM Horticulture, 69(2)/2012 Print ISSN 1843-5254; Electronic ISSN 1843-5394 Generation and Distribution of Productivity Gains in French Agriculture. Who are the Winners and the Losers over the Last Fifty Years? Jean-Philippe BOUSSEMART 1) , Jean-Pierre BUTAULT 2) , Oluwaseun OJO 3) 1) University of Lille and LEM/IESEG-School of Management, 3, rue de la Digue, 59000 Lille, France; jp.boussemart@ieseg.fr. 2) INRA Paris and INRA Nancy, AgroParisTech–Paris, UMR Économie publique 16, rue Claude Bernard 75231 Paris Cedex, France; butault@nancy-engref.inra.fr. 3) CNRS/LEM and IÉSEG School of Management, 3, rue de la Digue, 59000 Lille, France; o.ojo@ieseg.fr. Abstract. This paper offers an approach based on the economic theory of index numbers that revisits the classical surplus accounting technique. We measure the productivity gains and the combined effects of output and input price variation on French farmers’ income between 1959 and 2011, for the whole agricultural sector. During this period, total factor productivity grows at an average annual rate of 1.4% mainly due to a decrease of input quantity over the last thirty years while output volume has stagnated since the end of the nineties. Over the whole period, with a share of nearly 70% of the global surplus, the customers appear as the main beneficiaries of these productivity gains through a decrease in agricultural and food prices. Farmers only retained 23% of the surplus corresponding to a low increase in farm income. Finally, the suppliers and taxpayers are the losers in the surplus distribution via respectively a significant decrease of relative intermediate input prices and a substantial growth of public subsidies in favour of the agricultural sector. Keywords: index numbers, total factor productivity, factor income distribution, agricultural and food policy INTRODUCTION As a major source of growth and the main determinant of real prices, productivity is a key variable in economics. As Zvi Griliches once stated it, if there was only one thing that should be measured in economics, one should focus on Total Factor Productivity (TFP). In fact, productivity is a variable of interest because its time changes determine welfare. However measuring productivity gains is only one side of the problem. Attention should also be paid to the distribution of productivity gains among the different inputs and outputs retained by the technology in order to assess which of them recover price advantages from technical innovations and better management. For a long period, this last question has been considered as a key issue in productivity analysis. Kendrick (1961), Kendrick and Sato (1963), devoted a large part of their works to measure TFP growth from quantity changes and Productivity Surplus (PS) shares from price variations simultaneously. During the seventies, such analyses became a standard practice in France or other European countries. Particularly, the agricultural sector was one of the industries where numerous studies were conducted in order to conclude if the productivity gains generated by farmers were captured or not by the upstream and downstream sectors. More recently, thanks to the index number theory and the use of flexible parametric functional forms or non-parametric data envelopment techniques which allowed new technology modeling developments, TFP estimations have been an extraordinarily innovative