Temporal aggregation and the production smoothing model: Evidence from electronic parts and components manufacturing in Taiwan

To understand the inventory behaviors of manufacturers, production smoothing is one of the most discussed theoretical models from the perspective of macroeconomics or an individual firm. The basic motive of production smoothing is that companies can increase or decrease their finished goods inventories to allow production that is smoother than sales [1]. Hence, the production-smoothing model (PSM) of inventories depends on a convex short-run cost function and adjustment costs that induce firms to maintain inventories for dampening the effects of demand fluctuations [2]. In other words, production has to be less volatile than sales in PSM. The above hypothesis is reasonable because it's a common scene in manufacturing. Meanwhile, inventories will most usually serve as production smoother if adjusting production is costly in comparison with the costs of keeping inventories [3]. Based on the above framework, researchers have developed various formulations of PSM, which have been empirically implemented to different manufacturing sectors. However, the applications of PSM remain debatable despite the intuitive appeal of PSM. Previous empirical studies have shown mixed results regarding the validation of PSM, and Ghali [4] has shown that we should expect to see production smoothing for only a subset of manufacturing industries. He also claims that unless one confines the analysis solely to data on industries for which the PSM should a priori be applicable, the percentage of cases where smoothing is observed is irrelevant. Hence, it is important to verify different formations of PSM and to extend the empirical study of PSM to other key industries.