This study tries to measure and to analyze the evolution of the total productivity of the factors (TPF) in the Cauca s manufacturing sector between the year 1993 and 2006. The measurement developed carries out for an unbalanced panel of firms across three alternative econometric methodologies: ordinary least square (OLS), fixed effects (FE), and a semiparametric model proposed by Levinsohn-Petrin (LP). The results demonstrate that biases exist in the parameters under OLS and FE, finding constant returns to scale but under LP we found, under consistent parameters, that TPF follows closely the cauca s business cycle, in presence of increasing returns to scale.