This paper investigates the relative importance of cyclical fluctuations in labor and capital utilization, increasing returns to scale, and technology shocks as explanations for procyclical productivity. It exploits the intuition that materials inputs do not have variable utilization rates, and materials are likely to be used in fixed proportions with value added. Therefore, materials growth is a good measure of unobserved changes in capital and labor utilization. Using this measure shows that cyclical factor utilization is very important, returns to scale are about constant, and technology shocks are small and have low correlation with either output or hours growth.