UCLA

This dissertation asks: how do global value chains influence currency conflict? I argue that global value chain integration shifts the traditional exchange rate preferences of exporting firms away from a desire for a competitive, underval- ued exchange rate—the cornerstone of a currency war—, towards a more stable, equilibrium-level exchange rate. As firms increasingly rely on the cross-border exchange of intermediate inputs, they will tend to prefer exchange rate stability over competitiveness, thus alleviating currency conflict.

In the late 1980s and into the 1990s, a revolution in information technology drastically lowered communication costs within and across borders. Concurrent innovations in shipping and containerization and a surge in regional trade agree ments considerably lowered trade costs between countries. Reduced trade and communication costs led many firms in advanced economies to outsource parts of their production process to lower-wage countries with whom they had a trade agreement. This fragmentation of the production process into global value chains reshaped international trade, and with it, currency politics.

In a globalized economy where firms import many of the inputs that comprise a final exported good, an undervalued exchange rate no longer gives a boost to exports due to the increased cost of the imported inputs. Currency undervaluation is a costly venture by policymakers, which requires a trove of foreign exchange reserves, a high savings rate, and, in most cases, controls on international capital mobility. Global value chain (GVC) integration decreases the benefits of an undervalued currency beyond its cost, thus constraining governments from manipulating their currencies for competitive gain. Additionally, as countries move up the value chain and specialize in complex intermediate inputs—e.g., airplane parts—, concern over exchange rate stability exceeds interest in exchange rate competitiveness due to exchange-rate pass-through—the effect of an exchange rate movement on the price of a good. All else equal, further GVC integration will tend to increase firm preferences for exchange rate stability and weaken preferences for an undervalued exchange rate.

These arguments are tested with cross-sectional time-series data that cover 62 countries—accounting for over 80% of global trade—between 1995 and 2011. I exploit the richness of the trade data by measuring annual GVC participation at the country, country-partner, and country-sector levels of observation. In the first empirical chapters, I test the argument that GVC integration puts upward pressure on undervalued exchange rates towards their equilibrium level. Utilizing two distinct measures of currency misalignment, I show that the more integrated a country becomes in GVCs—specifically, the more foreign inputs as a share of gross exports—the weaker its commitment to an undervalued currency. Moreover, as a country moves up the value chain, producing highly-specialized inputs, there is a similar revaluation of the exchange rate. All else equal, GVC integration constrains governments from utilizing competitive exchange rate policies.

The third empirical chapter evaluates the policy implications of the argument and recommends inclusive trade measures to promote value chain integration. Fol- lowing the information and communications technology (ICT) revolution of the late 1980s, firms could transmit data and communicate across borders quicker than in previous decades. The missing elements to take advantage of the low-cost labor were lowered trade barriers and protections for foreign firms. The regional trade agreements between North and South that ballooned in the 1990s provided this critical component for GVC integration. I test this argument using annual GVC data at the country-partner level regressed on the presence of a regional trade agreement between partners, as well as the standard gravity model covariates that predict trade between countries—distance, economy size, and shared history. My empirical strategy includes a marginal structural model with inverse probability weights, ensuring that there is balance in the time-varying covariates across different treatment histories, and allowing me to make causal claims (albeit with strong assumptions) about the treatment effect of regional trade agreements. Indeed, I find that regional trade agreements are a highly significant predictor of GVC integration, increasing GVCs between countries by 17% and 57% as a share of gross exports. The ICT revolution was a necessary condition for the emergence of GVCs, but not sufficient. Policymakers should consider using regional trade agreements, which increases general welfare in all countries involved, as an inclusive measure to combat currency conflict.

In sum, my dissertation shows that global value chain integration alleviates currency conflict. As firms rely increasingly on the cross-border exchange of in- termediate inputs, and, as these firms move up the value chain to produce highly- specialized goods or services, their exchange rate preference will tend to move away from an undervalued exchange rate. A main determinant of GVC integration is the presence of a regional trade agreement between partner countries. The deep provisions found in these agreements protect firms investing in foreign markets and the removal of trade barriers allows for freer movement of inputs across borders. Together, I have shown an inclusive approach to combating currency conflict.