Thesis advisor: James E. Anderson / The gravity model proved to to be one of the most successful framework for analyzing international trade flows, being referred to as the “workhorse” in the international trade literature (Head and Mayer (2014)). Microfoundations to this model has been provided in Anderson (1979) and it has often been employed to estimate the effects of a variety of trade policies (see Cipollina and Salvatici (2010) for a meta-analysis on reciprocal trade agreements, Rose (2000) for the effects of currency unions). The two chapters of this dissertation, which are independent empirical pieces, both make use of gravity equations for the estimation of trade flows, although with different purposes. The first chapter focuses on the specification of the gravity equation. In the second chapter, instead, gravity equations are employed for assessing the relationship between trade and growth: in fact, their estimation represents the first step for the creation of an instrumental variable for export flows. In the first chapter, a solo-authored work titled Scale Economies in European Trade, I show that European data support the existence of economies of scale in trade flows. The impact of trade costs on trade flows, in fact, is assumed to be constant by almost all empirical studies employing the gravity framework. Anderson et al. (2016) are the first to depart from this assumption, allowing trade costs to vary as a function of trade volumes. Their model nests the more traditional one and hence can be used to test for the existence of these scale economies, which are shown to be in place for trade between US and Canada. For my analysis I construct a comprehensive dataset for European trade in manufacturing over a long time span (from 1980 to 2013), on which I employ the same methodology. My results show that scale economies in trade costs are indeed a strong empirical fact outside of the American continent, and this holds for all the 26 manufacturing sectors considered, with an estimated average of 0.64% decrease in trade costs given by a 10% increase in trade volume. The focus on Europe allows me to test whether the EU expansion affected these economies of scale. While this is not true on average, it seems to be the case for some industries: trade with a EU partner entails scale elasticities 50% lower than trade with a non-EU member for 11 sectors out of the 26 considered. I also investigate whether scale elasticities can be rationalized by the existence of informational asymmetries. Using detailed product-level data, I do not find evidence that the degree of product homogeneity can account for the observed cross-sectoral variation. The scale coefficients are instead linked to country-specific institutional variables, such as the level of corruption: exporting to the country whose level of corruption is the lowest in the sample entails half the scale elasticity than exporting to the most corrupted one. In other words, corruption depresses trade to an higher extent on longer distances. In the second chapter, joint with Carlo Altomonte and Italo Colantone and titled Trade and Growth in the Age of Global Value Chains, we revisit the relationship between trade and income, taking into account the recent surge of global value chains (GVCs). First, we develop a new geography-based, time-varying instrument for export, exploiting the sharp increase (almost tripling) in the maximum size of container ships between 1995 and 2007. This global shock has an asymmetric impact on bilateral trade flows across countries, affecting disproportionately more countries endowed with a larger number of deep-water ports, which are needed to accommodate the new, much larger ships. We exploit this heterogeneity for identification, building up the instrument for export in a gravity framework. Our result show that export has a positive effect on GDP per capita, with a 0.6 elasticity. Evidence at the country-level shows that this effect works through capital accumulation. Exploiting the decomposition methodology by Wang et al. (2013), we show that differences in the value added composition of exports matter for trade-growth nexus. We find evidence in favor of an income premium for countries that upgrade their positioning in GVCs, whereas the degree of participation to GVCs does not seem to play a role. Consistent with this finding, we show that countries whose average level of upstreamness (a’ la Antras and Chor (2013)) increases the most over time exhibit a higher trade elasticity of income. Both papers indirectly deal with the effect of geographical distance on international trade flows. One of the strongest regularities in economics is certainly the negative role played on trade flows by the distance between origin and destination. Disdier and Head (2004), comparing 1,467 different studies, compute an average distance elasticity of trade of about -0.9. Hummels (2007) shows that the distance elasticity of trade does not seem to diminish over time, as it would do should distance be capturing only transportation costs, thanks to the technological developments witnessed in the transportation sector. Distance seems then to refer to trade costs in general, including institutional, policy and regulatory barriers that, also for historical reasons, often increase the further away countries are located. In the first paper, I show that the impact of distance on trade flows is not constant but varies with trade volumes. This corresponds with having a component of the composite friction described before, hidden in the distance term, being fixed and is consistent with micro-evidence on the export behavior obtained from firm-level data (Roberts and Tybout (1997)). It seems natural, then, to test whether some characteristics, either at the product-level or at the country-level, have a prominent role in explaining the non-linear effect that distance has on trade. My results find in level of corruption of the destination country an important determinant. In the second paper, we test whether the distance elasticity of trade varies as a function of the number of deep water ports on both the importer’s and the exporter’s shores, capturing the extent to which countries can trade via container vessels. The data support this claim for all the manufacturing sectors considered, showing that geographical distance, even though non-exclusively, captures the incidence of transportation costs on export flows. / Thesis (PhD) — Boston College, 2016. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

Commuting is becoming more and more important in every day life in the world of complex labour markets. If for hundred years ago most workers lived less than one hour walking distance from their workplace, then today they commute daily outside the cities and villages they live. Often is it a trade off between unemployment and employment. The subject of this thesis is to investigate if there are differences in female and male commuting behaviour. Especially it studies to what extent their spatial interaction is affected by variables such as labour in origin municipality, jobs in destination municipality and the distance between these two. Gravity models are used, and six different versions are estimated. The commuting data used in thesis comes from SCB (Statistics Sweden) and data about travel time collected by the Swedish Road Administration. Commuting data includes 1 003 771 people in Sweden who commute to a job located in another municipality than he or she was settled. The results imply that men are less sensitive to distance between two locations. Moreover, the amount of labour in origin and the amount of jobs in destination have higher effect on male commuting.

The considerable increase in international trade and migration flows can be treated as the consequence of globalization and economic integration process during the recent years. The issue of immigration impact on trade has been studied a lot since the middle of 1990s and a significant and positive effect was found in most of the cases. This paper contributes to previous studies by investigating the impact of immigrants from 155 countries on Sweden’s exports to and imports from these countries during the period from 1980 till 2010, using an augmented gravity model. The impact of immigrants on exports and imports is studied separately by looking at the whole period results and the dynamic of changes within the period. Besides this the influence of immigrants’ home countries peculiarities (by dividing them on regions and level of development) and immigrants’ type (immigrant stock, immigrant flow and asylum seekers) is tested. To the best of my knowledge it is the first study that implements this variety of classification tests for Swedish data. The empirical results suggest that a 10 % increase in immigrant stock facilitates a 1% increase in exports to and a 0.5% increase of Sweden’s imports from the immigrants’ home countries. There is a tendency of gradual decrease of immigrants’ impact on both exports and imports within the period under consideration. According to the different classification tests the immigrants from Africa have the largest impact on Sweden’s exports, though European immigrants have the largest impact on imports; Swedish foreign born population from developed countries more facilitate trade than those who are from developing; new comers and temporary immigrants have almost the same impact on exports as the total immigrant stock, but there is even slightly negative effect on trade by asylum seekers.

This paper examines whether European Monetary Union (EMU) countries share fairly the effect of their membership in Eurozone (EZ) or whether are winners and losers in this ''Euro-game''. By using panel data of 27 European Union (EU) Member States for the period 2001-2012 in the context of a gravity model, we focus on estimating the Euro’s effect on bilateral trade and we detect whether this effect differs across the Member States of EZ. Two estimation methods are applied: Pooled OLS estimator and Fixed Effects estimator. The empirical results come to the conclusion that the individual country effects differ and are statistically significant, indicating that EMU’s effect on trade differs across the Member States of EZ. The overall effect of the Euro is statistically insignificant, regardless the estimation method, demonstrating that the common European currency may have no effect on bilateral trade.

<p>Commuting is becoming more and more important in every day life in the world of complex labour markets. If for hundred years ago most workers lived less than one hour walking distance from their workplace, then today they commute daily outside the cities and villages they live. Often is it a trade off between unemployment and employment. The subject of this thesis is to investigate if there are differences in female and male commuting behaviour. Especially it studies to what extent their spatial interaction is affected by variables such as labour in origin municipality, jobs in destination municipality and the distance between these two. Gravity models are used, and six different versions are estimated. The commuting data used in thesis comes from SCB (Statistics Sweden) and data about travel time collected by the Swedish Road Administration. Commuting data includes 1 003 771 people in Sweden who commute to a job located in another municipality than he or she was settled. The results imply that men are less sensitive to distance between two locations. Moreover, the amount of labour in origin and the amount of jobs in destination have higher effect on male commuting.</p>

This thesis investigates the effect of US development aid on US exports to 134 recipient countries over the time period 1993 to 2015 with an application of the gravity model of international trade. Estimates of one-way panel dataset, specified by a dummy approach and estimated with OLS and PPML, suggest that for every aid dollar spent by the United States, US exports significantly increase by 1.59 US dollar. By lagging the aid variable for several years after disbursement, we find a declining effect of US aid on US exports, which indicates that tied aid is an important channel of the effect's magnitude. The effect does not vary systematically across income groups. Yet for geographical regions with a higher US export share, the impact of US aid on US exports is significantly larger suggesting that existing trading relations contribute to a larger effect of aid on donor's exports. The evidence shows that US aid increases US exports and reinforces economic relations with recipient countries and, thus, can be regarded as an important motive for the donor to provide development aid.

This paper empirically examines the effect of institutional quality on trade considering the gravity equation model. Taking data for 252 countries covering the period of 19 years from 1996 to 2014, the research has been done with two stage regression analysis. In the first stage, we estimate the effect of gravity factors that either benefit or hinder trade along with OECD membership and Linder’s effect by Poisson-Pseudo-Maximum-Likelihood (PPML) estimator with importer- time, and exporter- time fixed effects. Taking the estimated exporter- time fixed effects from the first stage, we regress it with institutional variables in the second stage by OLS method with country and time fixed effects. Results indicate that institutional quality has a significant and positive impact on bilateral export with mostly 1% and 5% significance level. Interestingly, our study also shows that Linder’s effect is negative only for trade among the OCED countries.

Relying upon two common trends in modern castle studies, this exploratory study works to combine the landscape context and the spatial interaction of the main building to create an assessment of the spatial and social interaction between the main residential structure of a noble’s estate and the landscape features attached to surrounding property features. To explore questions about this kind of interaction this project has taken the sheriffdom of Angus, Scotland, between the year 1450 and 1542, to examine non-royal residences in an area that offered a diverse topography. This project aims to gain a better understanding of the surroundings of late fifteenth and early sixteenth century noble residences in Angus while contributing to the growing discussion of castles and their landscapes, and testing methods for addressing the spatial and social interaction between the main structure and the landscape features. Section A discusses the three source types used for compiling the dataset for this project within the context of three key categories needed to create a GIS dataset: location, object, and attributes. From the landscape features the mills and fishings were the most commonly mentioned and further details regarding the contents of the lordly landscapes were rare. Section B explores three methods of examining the relationships between the main residence and the landscape features: a modified RA and RRA values assessment, which measured levels of segregation within the noble residence site as a whole; a version of the gravity model, which helped identify the draw for interaction within the arrangement of the noble’s landscape; and network analysis questions, which facilitated a clear assessment of any connections between the use of structural terms and landscape features mentioned over both temporal and social contexts. This exploration of spatial and social interaction opens up a discussion about Scottish noble landscape creation and new methods for studying the relationship between the main structure and the wider complex of a noble residence.

June 1 1997. After 155 years under British rule, Hong Kong was reunited with its ancient roots, China. The administrative power shifted. What happened then? In this paper we analyze how Hong Kong’s trade flows changed after the liberation. We conduct our analysis with main focus on the trade predicting factors of Gravity, Institutional quality and Hong Kong’s relationship to China. We have found that trade flows did not significantly change much, however, Hong Kong’s attitude towards its trade partners’ institutional quality seem to have. Further, Hong Kong seem to have embraced the reunification with China and is now more dependent of its new ruler, in terms of trade, than before.

The gravity model of trade is the workhorse model for international trade. In its most basic form, it stipulates that bilateral trade flow between two countries is proportional to the countries’ Gross Domestic Product (GDP) and the distance between them. According to the gravity model, the elasticity of trade flows to distance, or the “distance effect,” has increased since the early 1970s, a confounding empirical result known as the “distance puzzle.” This paper investigates the distance effect more closely by decomposing it. More specifically, it aims to isolate the effects from culture, constructing measures of cultural distance and examining their effects on bilateral trade levels and the distance effect. The results show that cultural differences do not account for the distance puzzle. However, it also finds that cultural distance has both a substantial and statistically significant effect on bilateral trade.