This thesis proposes a general test of endogenous growth models based on the observation that per capita growth rates in the U.S. appear to be stationary over the last century. The test is used to provide time series evidence against two major alternatives to the neoclassical growth model: the "AK"-style models and the R&D-based models. A modified version of the Romer [1990] R&D model is introduced that is not rejected by the stationarity test, but the extended model alters a key implication usually found in endogenous growth theory. Although growth in the extended model is generated endogenously through R&D undertaken by profit-maximizing agents, the long-run growth rate depends only on parameters that are usually taken to be exogenous. Empirical evidence on TFP growth and R&D in the U.S., France, Germany, and Japan provides support for the extended model. The author thanks colleagues and friends for their support and acknowledges the contributions of various economists. The thesis examines time series tests of endogenous growth models, focusing on the stationarity of growth rates and the implications for accumulation-based models. The results show that growth rates are stationary, contradicting the predictions of accumulation-based models. The analysis also considers R&D-based models and finds that they are potentially consistent with the time series evidence. The paper concludes that the stationarity of growth rates suggests that accumulation-based models are inconsistent with the data, while R&D-based models may be more consistent. The author also discusses the implications of these findings for economic growth and the relative price of capital.This thesis proposes a general test of endogenous growth models based on the observation that per capita growth rates in the U.S. appear to be stationary over the last century. The test is used to provide time series evidence against two major alternatives to the neoclassical growth model: the "AK"-style models and the R&D-based models. A modified version of the Romer [1990] R&D model is introduced that is not rejected by the stationarity test, but the extended model alters a key implication usually found in endogenous growth theory. Although growth in the extended model is generated endogenously through R&D undertaken by profit-maximizing agents, the long-run growth rate depends only on parameters that are usually taken to be exogenous. Empirical evidence on TFP growth and R&D in the U.S., France, Germany, and Japan provides support for the extended model. The author thanks colleagues and friends for their support and acknowledges the contributions of various economists. The thesis examines time series tests of endogenous growth models, focusing on the stationarity of growth rates and the implications for accumulation-based models. The results show that growth rates are stationary, contradicting the predictions of accumulation-based models. The analysis also considers R&D-based models and finds that they are potentially consistent with the time series evidence. The paper concludes that the stationarity of growth rates suggests that accumulation-based models are inconsistent with the data, while R&D-based models may be more consistent. The author also discusses the implications of these findings for economic growth and the relative price of capital.