Exporters add and drop destination markets in response to a variety of global, national and industry-specific shocks. This paper develops empirical measures of these market changes and documents a set of key stylized facts using the customs databases of China (2000-2006) and the United Kingdom (2010-2016). First, I find within-firm changes in destination markets involve large trade values and 30-40% of all market changes involve simultaneously adding and dropping markets. Second, around 20% of within-firm market changes are driven by fluctuations in bilateral exchange rates and local CPI measures. Taken together, these facts suggest that firms face large destination-specific fluctuations in the demand for their products. Third, while adding and dropping markets, firms simultaneously adjust prices and quantities across all other destinations they serve. I build a multi-country general equilibrium model to investigate the channels that can generate the observed data patterns and study the aggregate implications of mutable markets (within-firm market changes) on the distribution of markups, trade volumes, and welfare. Applying the multi-country model to analysis of a bilateral trade war, I find that aggregate productivity for countries directly involved in the trade war drops more (1-2%) and that of countries not involved rises more (8-10%) when firms endogenously vary their markets in response to the new conditions of competition in local markets induced by the bilateral trade war.
Markets and Markups: A New Empirical Framework and Evidence on Exporters from China
We build a new empirical framework for analyzing destination-specific markup and quantity adjustments by exporters. Our first contribution is an unbiased estimator of the destination-specific markup elasticity to the exchange rate that isolates marginal costs in large unbalanced panels where the set of markets served by a firm varies endogenously with currency movements. Relatedly, we estimate firms’ cross-market supply elasticity—defined as the adjustment in relative quantities across markets associated to exchange rate-induced adjustment in markups. Our second contribution is a new classification of Harmonized System products into high and low differentiation goods—which we used as a proxy for exporters’ market power. Exploiting information about Chinese “measure words” reported in customs declarations, we add value to existing classification systems including Rauch (1999) and the UN’s Broad Economic Categories. Applying this framework to exporters from China, we find that the average markup elasticity is higher for high differentiation goods (20%) than for low differentiation goods (6%). The cross-market supply elasticities are correspondingly lower for high than low differentiation goods, 0.83 and 2.47, respectively. Finally, we discuss how our estimated elasticities can serve as a diagnostic tool to guide the development of open macro models.
Invoicing and Pricing-to-Market: A Study of Price and Markup Elasticities of UK Exporters
In this paper, we provide novel micro evidence that the currency in which exports and imports are invoiced is a good proxy for the currency in which firms set prices. Using detailed data on UK customs transactions, we document that destination-specific markup adjustment is substantial only for export shipments which are invoiced in the destination market's currency, consistent with the view that firms invoicing in local currency price to market. Conversely, we find no destination-specific markup adjustments by firms that invoice a shipment in either their own currency or a vehicle currency, consistent with a firm setting one price either in their own or in a vehicle currency. However, we also document that, while the aggregate shares of invoicing currencies for the UK's exports and imports are stable over time, there is substantial heterogeneity at the firm-product-destination level. A firm's shipments of the same product to the same destination are often invoiced in multiple currencies, with a non-trivial degree of switching from one invoicing currency to another within a twelve-month period. This is more pronounced for firms that are multi-product and serve several destinations, pointing to a potentially important margin of adjustment so far understudied in the literature.
Renegotiation of Trade Agreements and Firm Exporting Decisions: Evidence from the Impact of Brexit on UK Exports
The renegotiation of a trade agreement introduces uncertainty into the economic environment. In June 2016 the British electorate unexpectedly voted to leave the European Union, introducing a new era in which the UK and EU began to renegotiate the terms of the UK-EU trading relationship. We exploit this natural experiment to estimate the impact of uncertainty associated with trade agreement renegotiation on the export participation decision of firms in the UK. Starting from the Handley and Limao (2017) model of exporting under trade policy uncertainty, we derive testable predictions of firm entry into (exit from) a foreign market under an uncertain 'renegotiation regime'. Empirically, we develop measures of the trade policy uncertainty facing firms exporting from the UK to the EU after June 2016. Using the universe of UK export transactions at the firm and product level, and cross-sectional variation in 'threat point' tariffs, we estimate that in 2016 over 5300 exporters did not enter into exporting new products to the EU, whilst over 5400 exporters exited from exporting products to the EU. Entry (exit) in 2016 would have been 5.0% higher (6.1% lower) if firms exporting from the UK to the EU had not faced increased trade policy uncertainty after June 2016.
Media Coverage: VoxEU; The Economist; The Financial Times; The Telegraph
Understanding how exporters react to exchange rate shocks is important for evaluating international shock transmissions and setting the optimal international monetary policy. Empirical studies have documented huge heterogeneity in the degree to which different firms and products respond to exchange rate shocks. In addition, estimates of exchange rate pass through (ERPT) are time varying and depend on observed and unobserved variables in a nonlinear way. This paper proposes a machine learning algorithm that systematically detects determinants of ERPT and estimates ERPT at the firm-level in a large-scale custom dataset. The accuracy of the algorithm is tested on simulated data from an extended multi-country version of Atkeson and Burstein (2008). Applying the algorithm to China’s custom data from 2000-2006, this paper estimates the ERPT of China’s exporters and documents new evidences on the nonlinear relationships among market structures, unit value volatility and ERPT.
The Impact of Brexit Uncertainty on UK Exports
A Granular Analysis of the Exposure of UK Exports to EU Tariffs, Quotas and Antidumping under ‘No Deal’
Plying to Paradise: The Role of Airlift in the Caribbean Tourism Industry