CHAPTER 1
International Macroeconomic Comovement: The Role of Globalization in Goods and Asset Markets
Paul R. Bergin
Department of Economics, University of California at Davis,
One Shields Ave., Davis, CA 95616, USA
1.Introduction
The US Great Recession of 2007â2009 has been dubbed as the first truly âglobal recessionâ of recent times, in that the severe economic downturn starting in the US coincided with downturns in many other countries. As the world struggles to recover from the global recession, this is an opportune time to reflect on the nature and mechanisms of International Macroeconomic Interdependence. This episode provides new evidence, as well as new urgency, regarding a set of questions long debated in the macroeconomics literature. To what degree do national economies move together in their cycles of recession and recovery? Does the trend of greater international integration in goods and asset markets provide greater opportunity for transmission of shocks from one country to another? In terms of policy implications, does greater comovement imply greater benefit from countries coordinating together with their domestic macroeconomic policies?
This chapter brings to bear updated empirical evidence and novel model simulations to provide a new perspective on these fundamental questions. Data indicate that the Great Recession was truly an unusual episode. Prior to it, business cycle comovement between the US and its major trading partners was moderately positive but far from perfect. During the Great Recession, the correlation of national gross domestic product (GDP) fluctuations was nearly complete, with recessions experienced by all the major trading partners. However, this does not appear to be a new norm, as the US recovery from the recession was not experienced universally, and the correlation of national GDP levels subsequently moved to low levels.
Can this episode of high international comovement be attributed to trends of globalization in goods and asset markets? I do document a longrun trend of rising integration, which appears to have accelerated in the decade prior to the crisis, both in terms of goods trade and asset trade. Both types of trade reversed trend and dropped significantly during the Great Recession, and even though both have since recovered their previous levels, the trends of rapid growth in both goods and asset trade of the precrisis period have not returned. To focus on how goods and asset trade affect business cycle comovement, rather than the reversed direction of causality, we look at the cross sectional experience of the US main trading partners in the year prior to the crisis, and see how this related to the varying degrees of GDP comovement with the US among these countries. One finds suggestive evidence that greater integration in both goods and asset markets promoted greater business cycle comovement during the particular episode of the Great Recession.
To understand better the mechanisms by which goods and asset trade can transmit shocks across countries, I construct a simple dynamic stochastic general equilibrium model with two countries, two goods, and two assets. Two countries are needed as a minimum, of course, to study transmission across a border. Two goods are needed if one wants to consider integration in goods markets separately from asset markets, so that a country trades one good for another good under the condition of balanced trade. Finally, two internationally tradable assets are needed to consider the hypothetical case of international integration in asset markets, where there is no trade in goods, but only trade of one asset exchanged for another asset. I believe this is the first paper of to study separate goods and asset market integration in this manner.
Stochastic simulations of the model indicate that the greatest transmission occurs when both goods and asset integration are operative, so that a country may trade assets to finance a level of expenditure on goods different from domestic production. Financing of a trade imbalance thus allows for a large effect on the level of domestic demand for foreign goods, and hence induces the greatest spillover to the level of foreign production. This transmission typically takes place through changes in the world interest rate, which has direct effects on foreign consumption, investment, and production decisions. Transmission via the goods market alone, in isolation from asset market trade, can also be fairly large, but the transmission there takes place largely via the terms of trade, the relative price of the two countryâs export goods. Changes in the relative value of foreign exports also affects the consumption and production decision in the foreign country. Finally, we find that integration in asset markets alone, in isolation from goods trade, does not provide a transmission channel for the usual supply or demand shocks, but can provide a potent transmission channel for shocks hitting the financial market of one of the countries. This last result may offer insight into the mechanisms by which the financial crisis in the Great Recession in the US could be transmitted so strongly to other countries.
The model also makes clear that, even while greater international market integration can strengthen spillovers of shocks, these spillovers can make foreign output either move the same direction as home output or in the opposite direction. The sign of the spillovers depends on the shocks involved and the structure of the economies. In general, comovement will be positive in the case of demand shocks like government spending, or in a case where home and foreign goods are viewed by demanders as more complementary.
This chapter contributes to a large body of research on international business cycle comovement. Backus, Kehoe, and Kydland (1992) first showed that comovement can be negative in theory, using a simple one good business cycle model. This chapter differs in modeling the asset trade explicitly, rather than assuming complete asset markets operating in the background, so as to be able to track the asset trades that underlie shock transmission. I also extend the analysis to study the possibility of asset trade in isolation of goods trade. Early theoretical work actually had difficulty demonstrating a strong linkage between trade and positive business cycle comovement (see Kose and Yi, 2006). Burstein, Kurz, and Tesar (2008) showed that international trade linkages can promote positive output comovement, depending on the complementarity of home and foreign goods. But they interpret their comovement in terms of production rather than consumption, and again, they do not consider the separate role of asset market integration.1
There is also mixed evidence regarding the relationship between financial integration and business cycle comovement. Among limited empirical evidence, some work has found financial market integration can either promote positive comovement (see Kose, Prasad, and Terrones, 2004), and other work finds negative comovement (see Kalemli-Ozcan, Papaioannou, and PeydrĂł, 2013). In theoretical work, Heathcote and Perri (2002) find that a business cycle model with financial autarky does a better job matching certain features of the data than a model with complete asset markets. Recent business cycle research has focused on shocks to the financial market that can generate recession, which one might think is particularly relevant to the Great Recession. Perri and Quadrini (2016) find that in this context financial integration can potentially contribute to positive output comovement. One paper that focuses on the Great Recession episode is Imbs (2010), which finds empirical evidence that asset market integration promotes output comovement during this particular event. We differ, among other things, in that our more updated data covers both the recession period and the recovery afterward, which gives a very different perspective, as will be explained below.2
The next section of this paper presents empirical methodology and stylized facts regarding international output comovement. Section 3 constructs a novel theoretical framework, and Section 4 presents simulation results. Section 5 concludes, and summarizes the contribution of the other chapters in this volume to the constellation of questions raised in this chapter.
2.Empirical Motivation
This section will document several empirical regularities regarding international comovement in real GDP. Data come from the International Financial Statistics collected by the International Monetary Fund. As standard in business cycle research, data are quarterly and HP-filtered to remove long run trends. Countries were chosen because...