Time series testing and endogenous growth

Using the theory of cointegration, we show that the theory of endogenous growth is compatible with the time series evidence found in the post World War Il (WWII) American data but does not accommodate the socio-economic events prior to 1946. The anomaly of the disequilibria caused by the Big Depression of 1929 and the expansion triggered by WWII is measured in three dimensions: I. The cointegrating vector between output and physical capital is statistically different from (I, -1), 2. The magnitude of these disequilibria is three times larger than that of any other disequilibrium caused by any expansion or recession in the post-war era, and 3. The persistence of these disequilibria is about four times larger than the persistence of any post-war disequilibrium. For the post-war data, we find that output growth can forecast capital growth but not viceversa.