A large body of research documents that long-term relationships are an important feature of firm-to-firm transactions. Survey evidence suggests that on average around 80% of firms' sales go to regular customers, and these customers are usually other firms. Nevertheless, the macroeconomic implications of long-term relationships on firms' growth and on firms' price setting are poorly understood. For example, long-term relationships with suppliers could be an important determinant of firms? performance, as well as of firms' decisions to exit and to expand. On the other hand, a firm?s desire to maintain a relationship may constrain it from changing price, thus generating or amplifying price rigidities. This might be an important driver behind the persistent effects of monetary policy shocks on GDP observed in the data. This dissertation research uses a new dataset, the Longitudinal Foreign Trade Transactions Database (LFTTD) from the U.S. Census Bureau. This dataset records for every import transaction undertaken by U.S. firms the identities of the U.S. importer and of the foreign exporter, along with the values and quantities shipped. This feature of the data makes it possible to classify transactions as taking place between long-term trading partners, and to examine the implications of relationships on macro dynamics. The dissertation comprises of two sub-projects. (1) The first part of the research consists of examining the relationship dynamics in U.S. trade. The purpose of this project is to study why relationships are formed, how long they last, why they are broken up, and how they evolve over the business cycle. This research seeks to shed light on the effect supplier relationships have on the performance of U.S. importers over the business cycle and on their decisions to exit and to expand. It will also explain whether the collapse in trade during recessions is due to the destruction of relationships or due to reduced trade within ongoing relationships. (2) The second part of this dissertation aims at resolving an important puzzle in monetary economics. On the one hand, VAR evidence suggests that monetary policy shocks have persistent effects on GDP. On the other hand, micro evidence indicates the average firm frequently changes price, thus in principle adjusting quickly to shocks and eliminating any real effects. Most existing research investigating this puzzle has so far ignored producer markets (transactions between firms) as a source of price rigidities. This project examines whether firms? desire to maintain a relationship with their transaction partners is a plausible source of rigidity that can reconcile the conflicting micro and macro evidence. Since most firm-to-firm transactions take place in long-term relationships, it is a first-order question to understand how relationships affect aggregate outcomes. One important question is what drives the large, persistent differences across firms in performance and productivity: long-run relationships with reliable suppliers could be an important explanation. On the other hand, if producer relationships matter for aggregate price dynamics, this could have important implications for monetary policy. Understanding how producer prices are set and how they affect aggregate variables will enable central bankers to better respond to economic fluctuations.
