Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Richard F. Kamierczak


This study demonstrates one way in which the combined nonparametric and parametric estimates of returns-to-research can be used to build a stronger argument for public investment in agricultural research. The data used in this study were constructed from time series information covering the period 1949--95. Tornqvist-Theil quantity indices were calculated to determine the implicit state price for each input category. The returns-to-research in Louisiana agriculture were estimated using both nonparametric and parametric estimators, with appropriate emphasis given to lag structures, data coherence, and functional forms. Model misspecification testing for the parametric model was examined. Internal rates of return were calculated from the empirical estimates. Results indicate that Louisiana agricultural research investments significantly contribute to productivity growth. Returns exhibited a pattern that has been observed in a previous national nonparametric study and numerous parametric studies of returns to agricultural research. The overall return pattern exhibited an investment return life cycle as long as 30 years, in which flows of research returns can be divided into four stages; incubation, growth, maturity, and senescence. Public and private researches appear to play alternative and complementary roles in Louisiana agricultural productivity. The public research plays a major role in the third and final periods of the cycle. Nearly 95 percent of the benefits from agricultural research flowed between 15 years and 30 years after the initial investment, with a peak at 24 years. Private investment largely affected the second and third stages of the overall return life cycle. Private return patterns had a stronger impact in the short term and the peak effect occurred 7--8 years earlier than those for public research investment. The extension service played an important role in the productivity growth period. The results suggest that internal rates of returns to agricultural research investment in Louisiana are at least 20 percent for public investment and 18 percent for private investment. Results also show that nonparametric and parametric approaches can generate consistent results and the nonparametric and parametric approaches play complementary roles. Comprehensive approaches to model misspecification. testing used in this study provided better insight into sources of possible misspecifications.