Measuring the World Economy

Over the last decades there has been a sizeable increase in trade and financial openness, triggered by current and capital account liberalization as well as improvements in transport and communication technologies. As a consequence, the importance of spillover effects is likely to have increased at all levels of aggregation: among firms, industries, regions, and even countries. In other words, the relevance of geographic distance should have decreased over time, a view that has been popularized as ‘the death of distance’ by Cairncross (1997). Yet there is little overall quantitative evidence by how much globalisation has reduced economic distance between countries. A related albeit more specific strand of the literature considers either directly the evolution of trade costs over time or the sensitivity of trade flows to distance in gravity models. In their survey of the empirical trade literature, Leamer and Levinsohn (1995) conclude that the effect of distance on trade patterns has not diminished over time. More recently, Jacks et al. (2009) find that the role of distance has declined dramatically in the 19th century, but not over the period since 1950. This paper takes an alternative approach in addressing the question whether the world has become smaller in terms of economic distance. Instead of focussing on the transmission of single shocks (which requires high frequency data) or confining the analysis to a particular channel of interdependence (such as trade), it provides a bird-eye’s perspective, using a large cross-section of 135 countries and considering how cross-country interdependence of output volatility has evolved over the period 1955 to 2006. From a methodological perspective, we adopt a spatial econometric approach, relating a country’s output volatility to (distance weighted averages of) other countries’ output volatility, using descriptive measures, test statistics, and econometric estimates of cross-country interdependence. We also suggest a new method to approximate (unobserved) bilateral data from observed aggregate data in order to assess the importance of alternative channels of interdependence, in particular that of trade versus financial openness. We find that cross-country interdependence increased significantly till the mid 1980s and remained at high levels since then. In quantitative terms, estimation results for the most recent period suggest that a uniform shock to output volatility in the world economy roughly quadruplicates through spillover effects and the associated repercussions, which are equally transmitted through both trade and financial linkages. Keywords: financial openness, spillovers, trade, output volatility JEL No: C21, E32, F15, F40