Selection, Reallocation, and Spillover: Identifying the Sources of Gains from Multinational Production
Executive Summary — Nations with greater openness to multinational production exhibit, on average, higher productivity and faster economic growth. This positive relationship, likely conditional on many factors, is often attributed to knowledge spillover, such as direct knowledge transfer through partnership, the possibility to learn from the innovation and experiences of foreign firms, and the interaction and movement in labor markets. However, another less stressed explanation centers on firm selection whereby competition from multinationals leads to market reallocation and survival of only the most productive domestic firms. Moreover, as only firms with greater productivity are able to overcome the fixed cost of foreign investment, countries with greater openness to multinational production are thus attracting foreign firms that are, by selection, more productive. The above mechanisms all imply a positive relationship between multinational production and host-country productivity but represent sharply different economic causalities and policy implications. The self-selection of multinational firms suggests that higher host-country productivity can reflect the productivity of self-selected multinational firms, instead of the causal effect of multinational production. In contrast, domestic firm selection and knowledge spillover imply multinational production causes higher aggregate, domestic productivity. How the latter two affect domestic production is countervailing: tougher domestic firm selection results in a contraction of domestic production while knowledge spillover creates positive externalities. In this paper the authors disentangle the roles of knowledge spillover and firm selection in determining the aggregate productivity and welfare impact of multinational production and quantify the relative importance of these distinct sources of gains. Results show that firm selection and market reallocation constitute an important source of productivity gains while its relative importance varies across nations. There are crucial implications for policy aimed at influencing foreign direct investment (FDI) flows. Key concepts include:
- If foreign firms have knowledge spillover effects for domestic firms, special policy treatment may be justified.
- If, however, as these results suggest, increases in productivity can also arise from tougher selection on domestic firms as a result of competition for scarce labor and capital, it is important to improve domestic conditions, including conditions of labor (in particular, skilled labor) supply and credit access, and to eliminate barriers in order to facilitate gains from competition and reallocation of resources.
- The entry of multinational firms raises the cutoff productivity of domestic firms, pushing the least productive domestic firms to exit the markets.
- New multinational production also leads to an increase in the minimum revenue of continuing domestic firms, indicating an increase in fixed production cost and capital price.
- The authors analyzed a 60-country dataset of financial, operation, and ownership information for over one million public and private manufacturing companies in 2002-2007.
Quantifying the gains from multinational production has been a vital topic of economic research. Positive productivity gains are often attributed to knowledge spillover from multinational to domestic firms. An alternative, less stressed explanation is firm selection whereby competition from multinationals leads to market reallocation and survival of only the most productive domestic firms. We develop a model that incorporates both aspects and identify their relative importance in the gains from multinational production by exploring their distinct predictions on domestic productivity and revenue distributions. We show that knowledge spillover shifts both distributions rightward while selection and reallocation raise the left truncation of the distributions and shift revenue leftward. Using a rich firm-level panel dataset that spans 60 countries, our structural estimates suggest firm selection and market reallocation constitute an important source of productivity gains while its relative importance varies across nations. Ignoring the role of this source can lead to significant bias in understanding the nature of gains. We also perform counterfactual analysis and quantify both the aggregate and the decomposed welfare effects of multinational production.