Firms, Networks, and Trade

A conference on "Firms, Networks, and Trade" took place in Cambridge on March 15. Research Associates Laura Alfaro and Pol Antràs, both of Harvard University, and International Trade and Investment Program Director Stephen J. Redding of Princeton University organized the meeting. These researchers' papers were presented and discussed:

Alonso Alfaro-Urena, Central Bank of Costa Rica; Isabela Manelici, University of California at Berkeley; and Jose P. Vasquez, University of California at Berkeley

The Productivity Effects of Joining Multinational Supply Chains: Evidence from Firm-to-Firm Linkages

How integrated are developing countries in the supply chains of multinationals (MNCs) they host? What are the gains to domestic firms from supplying to MNCs? Alfaro-Urena, Manelici, and Vasquez answer these questions using administrative firm-to-firm transaction data from Costa Rica and two empirical strategies that identify the causal effects of starting to supply an MNC. Proxies of MNC linkages based on Input-Output tables conceal heterogeneities in buyer and seller characteristics relevant to the estimation of both the countrywide integration of MNCs and the effects of matching to an MNC. The researchers provide evidence on substantial linkage heterogeneity and its implications for the estimations above. They then identify the impact of becoming a supplier to an MNC by means of a quasi-experimental government program that matches MNCs with domestic firms. The researchers show that joining MNC supply chains led to strong and persistent gains in firm size and productivity. They add an economy-wide event study that, due to its larger sample size, allows them to explore levers of adjustment to joining MNC supply chains, heterogeneities in treatment effects based on supplier or MNC characteristics, and spillovers of productivity to domestic clients. Their results do not simply reflect firms' response to demand shocks, price effects, or improvements in tax compliance. Instead, they are suggestive of knowledge transfers by MNCs, that affect suppliers' production technology and organization, input choices, and relationships with domestic clients.

Johannes Boehm, Sciences Po, and Ezra Oberfield, Princeton University and NBER

Misallocation in the Market for Inputs: Enforcement and the Organization of Production

How costly is weak contract enforcement? Using microdata on Indian manufacturing plants, Boehm and Oberfield show that in states with weaker enforcement, as measured by judicial lags, production and sourcing decisions appear systematically distorted. They document that among plants in industries that tend to rely more heavily on inputs that require customization, those in states with more congested courts shift their expenditures away from intermediate inputs, whereas the researchers find the opposite in industries that tend to rely on standardized inputs. To quantify the impact of these distortions on aggregate productivity, they construct a model in which plants have several ways of producing, each with different bundles of inputs. Weak enforcement exacerbates a holdup problem that arises when using inputs that require customization, distorting both the intensive and extensive margins of input use.

Yimei Zou, Universitat Pompeu Fabra

Endogenous Production Networks and Gains from Trade

Zou develops a quantitative trade model with endogenous production networks, namely the collection of supplier-customer relationships among firms. In the model, firms form linkages with each other both within and across borders, balancing the tradeoff between extra revenue brought in by downstream connections and fixed costs required to establish these relationships. The structure of equilibrium production networks depends both on variable trade costs and linkage fixed costs. In particular, trade integration can lead to structural transformations of global production networks, which in turn bring about technological changes on both the firm and the aggregate level. The joint adjustments of domestic and international linkages constitute a new margin along which trade liberalization can affect welfare. Zou calibrates the model to trade data between the United States and the rest of the world (ROW) over 2000-2014. The model is able to replicate the actual time trend of the value added share in gross trade, as well as several cross-sectional patterns observed in the US-ROW input-output networks. Applying the model, Zou quantifies the welfare gains of moving from autarky to the 2014 equilibrium to be 15.5%, with a quarter of these gains arising solely from the rearrangement of linkages among firms.

Ernest Liu, Princeton University

Industrial Policies in Production Networks

Many developing countries adopt industrial policies that push resources towards selected economic sectors. How should countries choose which sectors to promote? Liu answers this question by characterizing optimal industrial policy in production networks embedded with market imperfections. My key finding is that effects of market imperfections accumulate through backward demand linkages, thereby generating aggregate sales distortions that are largest in the most upstream sectors. The distortion in sectoral sales is a sufficient statistic for the ratio between social and private marginal product of sectoral inputs, therefore, there is an incentive for a well-meaning government to subsidize upstream sectors. Liu's sufficient statistic predicts the sectors targeted by government interventions in South Korea in the 1970s and in modernday China.

Ayumu Ken Kikkawa, University of Chicago; Glenn Magerman, Université libre de Bruxelles; and Emmanuel Dhyne, National Bank of Belgium

Imperfect Competition and the Transmission of Shocks: The Network Matters

Kikkawa, Magerman, and Dhyne study the aggregate implications of the firm-to-firm production network structure. Using a dataset on all domestic transactions between Belgian firms, they establish two facts: firms charge higher markups if they have higher input shares within their customers, and firms experience larger churn of suppliers if they face a larger reduction in foreign goods' prices. Motivated by these two facts, the researchers build a model where firms compete as oligopolies to supply inputs to each customer and where firms optimally choose their suppliers. The network structure becomes irrelevant in a benchmark case where the researchers impose perfect competition and hold the network fixed. In this case, firm-level variables are sufficient to compute the welfare response to a large fall in import prices. Allowing for oligopolistic competition generates two counteracting forces within supplier-customer pairs. A supplier raises its markup to a customer when its costs decline, but it reduces the markup if other firms supplying the same customer receive the shock. Further, allowing for endogenous networks amplifies the impact of the shock as firms begin importing and begin sourcing from other firms exposed to the import shock. Due to the omission of these dynamics, the aggregate response in the benchmark case is less than one quarter of those in the full estimated model.

Jonathan Eaton, Pennsylvania State University and NBER; Samuel S. Kortum, Yale University and NBER; and Francis Kramarz, CREST-INSEE

Firm-to-Firm Trade: Imports, Exports, and the Labor Market

Firm-level customs and production data reveal both the heterogeneity and the granularity of individual buyers and sellers. Eaton, Kortum, and Kramarz seek to capture these firm-level features in a general equilibrium model that is also consistent with observations at the aggregate level. Their model is one of product trade through random meetings. Buyers, who may be households looking for final products or firms looking for inputs, connect with sellers randomly. At the firm level, the model generates predictions for buyer-seller connections and the share of labor in production broadly consistent with observations on French manufacturers and their customers in other countries of the European Union. At the aggregate level, firm-to-firm trade determines bilateral trade shares as well as labor's share of output in each country.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us