International trade and trends

Abstract

One of the most important events in the economic world is undoubtedly Globalization. This term is frequently employed by professionals and the general public, but rarely with a precise meaning. Broadly, we speak of globalization to mean that the intensity of trade relations between domestic economies has reached a certain threshold. In the recent decades, international trade has grown dramatically and this dramatic growth (averaging 2-3 percent per year for the past fifty years) has been reflected in a remarkable increase of the ratio between world trade and income; in other words, the increase of the so-called openness ratio. This increase cannot be explained without considering the advance of vertical specialization, a kind of trade reflecting the phenomenon of fragmentation of global production across countries. In this scenario, each country specialises in the stage of the production chain in which it has comparative advantage rather than producing the whole good, as used to happen. Defining vertical specialization as the part of foreign added value embodied in the domestic exports, this proportion accounts for more than 21%, in average, of the total exports of the ten OEDC countries studied in depth in this Thesis. Briefly, lees than 80% of the merchandise exported by these countries are genuinely made at home The aim of this Thesis is to show how the international fragmentation of production and the corresponding change in the nature of international trade, moving from horizontal to vertical specialization, affect the ratio of openness. It takes official trade statistics as a base and, consequently, affects the comparisons of openness processes between the first wave (1870-1913) and second wave (1948-2007) of globalization. We also study how this change in the nature of international trade is reflected in the stochastic properties of the openness ratio. We start, in Chapter 1, by carrying out a preliminary analysis of the stochastic properties of merchandise exports over GDP ratios in 57 countries. We cover the period 1948-2007, offering an overview of the integration order of the series in a twofold perspective: time series and panel data. From these analyses we can assert that a different representation exists for the data generating process depending on the degree of development of countries. Following the World Bank criteria to characterise countries as developed or less developed, we find a stationary pattern for less developed countries and a unit root pattern for developed countries. We could say, by focusing on the share of exports over GDP, that the new conditions of international trade have not affected the intensity of trade integration equally in developed countries (DCs) and less developed countries (LDCs). This finding is fully consistent with the theory of the disintegration of production, which assigns a major role to the decline of trade costs in the process and predicts difficulties in joining the international network for LDCs. Poorer infrastructure and institutional frameworks hinder the entrance of LDCs into the network as final good exporters, so their series are not necessarily affected by the multi-accounting bias that vertical specialization introduces into the openness ratios. The disintegration of production leads inputs to cross borders several times during the productive process, which introduces an upward bias into the openness ratio, since the numerator is valued (in the official export statistics) at final prices, while the denominator is value-added (GDP). Thus, our hypothesis is that the progress of vertical specialization, together with the abovementioned multiple-accounting bias, is the reason for the presence of unit roots in the international trade series (exports to GDP) for DCs. This is the possibility that we explore in the following chapter. In Chapter 2, our goal is to assess to what extent the upward bias of exports that is now partly seen as responsible for the boom of the world openness ratio and might have been distorting the measurement of globalization in some specific countries and periods (1948-2007). Here we try to find a link to relate the presence of unit roots and vertical specialization in ten OECD countries for which we have available data. We have scattered vertical specialization data (as the percentage of imports embodied in domestic exports). Using this information, provided by Hummels et al. (2001) and Chen et al. (2005), we propose a modified openness measure that reveals how much domestic added value is contained in the exports of a particular country. Domestic added value is the value of exports when the content of vertical specialization is subtracted. More interestingly, when analysing the statistical properties of the unbiased series of openness, we find that these become trend stationary for nearly all the countries, corroborating the responsibility of vertical specialization for the presence of unit roots. In Chapter 3, we provide additional support to the previous findings by studying the stochastic properties of the openness ratios in the first wave of globalization (1870-1913). The first wave of globalization shares several similarities but also exhibits some differences in relation to the second globalization. In the first globalization, the overwhelming trade in finished products and the consequent absence of multi-accounting biases make us anticipate a stationary trend in the openness ratios. This hypothesis is confirmed when analysing the stochastic properties of the merchandise X/GDP ratios for the same ten OECD countries that we studied in the second chapter. We find a stationary pattern for all series but one, thus supporting the idea that the change in the official statistics due to vertical specialization must be behind the presence of unit roots in the second wave of globalization. Finally, Chapter 4 tries to shed new light on the debate on the intensity of trade integration in the two waves of globalization, the pre-WWI (1870-1913) and post-WWII (1948-2007) globalizations. Although a way of comparing periods has been to confront the evolution of trade shares of GDP, the procedure is not free from biases. Firstly, because it does not consider the change in the nature of trade in the second period, when the advance of vertical integration and the corresponding multiple-accounting effect on the statistics biases the ratio upwards. Secondly, the procedure does not consider the change in the GDP composition in the second period, when the advance of the service sector biases de ratio downwards. To overcome the problems of comparison that arise from the use of this doubly biased ratio, we also propose a new synthetic measure of openness. By using the corrected measure, the outcomes show that the intensity of the integration process was considerably superior in the second wave.