International credit markets and global business cycles

• Author: Yi Wen,Xiaochuan Xing,Patrick A. Pintus
• DOI: http://doi.org/10.1111/ijet.12206
• Date: 01 March 2019
• Published date: 01 March 2019

doi: 10.1111/ijet.12206

International credit markets and global business cycles

Patrick A. Pintus,∗Yi Wen†and Xiaochuan Xing‡

This paper stresses a new channel through which global financial linkages contribute to the

co-movement in economic activity across countries. We show in a two-country setting with

borrowing constraints that international credit markets are subject to self-fulfilling variations in

the world real interestrate. Those expectation-driven changes in the borrowing cost in turn act as

global shocks that induce strong cross-country co-movementsin both financial and real variables

(suchas asset pr ices,g rossdomestic product, consumption, investment, and employment). When

firms around the world benefit from unexpectedly low debt repayments,the y borrowand invest

more, which leads toexcessive supply of collateral and of loanable funds at a low interest rate, thus

fueling a boom both at home and abroad. As a consequence, business cycles are synchronized

internationally. Such a stylized model thus offers one way to rationalize both the existence of

a world business-cycle component, documented by recent empirical studies through dynamic

factor analysis, and the factor’s intimate link tog lobal financial markets.

Key wor ds world interest rate, international co-movement, self-fulfilling equilibria

JEL classification E21, E22, E32, E44, E63

Accepted 11 October 2018

1 Introduction

Empirical evidence suggests that there exists a set of common factors behind business cycles across

nations. This set of global factors has affected national outputs of many countries simultaneously

over the last 50 years. For example, Kose, Otrok, and Whiteman (2003) document the existence of

a distinct world business cycle in a 60-country sample that spans seven regions of the world over

the period 1960–90. More specifically, they show that on top of country-specific shocks, a world

factor is also important and accounts for a significant fraction of country-specific business-cycle

variations. In particular, this global factor accounts for about 35% of the variances of output and

of consumption in the USA, which turns out to be more than the regional or the country-specific

factors can explain.

Kose, Prasad, and Terrones (2003) report similar results and further show that financial linkages

between countries are more important for international business-cycle transmissions in economies

∗CNRS-InSHS and Aix-Marseille University, CNRS-InSHS D´

el´

egation Paris Michel-Ange, Paris, France. Email:

patrick.pintus@univ.amu.fr

†Federal Reserve Bank of St Louis, St Louis, Missouri, USA and TsinghuaUniversity, Beijing, China.

‡YaleUniversity, New Haven,Connecticut, USA.

This article has been written in honor of Roger Farmer,whose intellectual cur iosity,academic achievements and generous

friendship have both enriched and inspired us. Wethank very much Jess Benhabib, Kazuo Nishimura, and Makoto Yano

for inviting us to contribute to this special issue, as well as a referee for comments and suggestions and St´

ephane Auray

for discussions. The usual disclaimer applies.

International Journal of Economic Theory 15 (2019) 53–75 © IAET 53

International Journal of Economic Theory

International credit markets Patrick A. Pintus et al.

that are more open to international capital flows (see also Imbs 2004). More recently, Crucini et al.

(2011) also confirm that the world-wide business cycle is as important as the nation-specific business

cycles and they also find that “real interest rates are dominated by a common G-7 component,w ith

two-thirds of the variance attributable to the common factor and almost none attributed to the

nation-specific component” (p. 167). In other words, in G7 countries real interest rates exhibit

variations that appear to be driven by a common global factor.Therefore, not only do national levels

of output and consumption follow a world component, but so also do financial variables.

In sum, the existing body of empirical evidence strongly points to the existence of a few world

common factors that explain international co-movements of both output and financial variables. In

addition, financial linkages between countries through international capital and debt markets are

suspected to play a key role in explaining these common factors. Of striking importance is the fact

that a common component drives much of the co-movementin different countries’ interest rates, at

least in developed economies.

However,little is known theoretically about the nature of such common factors behind world real

output and financial variables, uncovered through dynamic factor analysis in the studies reviewed

above. One obvious candidate is global technology shocks. But the hypothesized existence of global

technology shocks suffers from the lack of micro-level empirical evidence that innovations at the

firm level are synchronized globally despite different institutional arrangements, such as patent laws

and government efforts to protect technological secrets. It is not obvious that different countries’

technology frontiers progress and regress at the same time.

International trade is another possible source of global business cycle propagation (see, for

example, Miyamoto and Nguyen 2017). But it takes time for goods and services to move across

borders: shipping time often takes weeks if not months from one country to another. So the role of

shocks to the speed of global shipping or trade costs in generating the world business cycle is also

likely to be limited.

This paper proposes that the existence of a global financial market sets the conditionfor the r ise of

financial and real global factors that cause global business cycles. Weshow how the existenceof a g lobal

financial market makes it possible for investors’expectations to be coordinated around the world. In

particular, global sentiments originating from the world financial market can simultaneously drive

both global interest rates and national outputs across the world. Intuitively, financial markets are

connected much more tightly and in a more timely manner across nations than goods and labor

markets, yet within each country the goods and labor sectors are closely connected to the financial

sector. Asa result, animal spirits in the world financial market can simultaneously affect the financial

and real variables in each country (especially across developed economies, which are morefinancially

globalized).

Thus, we develop a very simple setting with two countries that are open to international financial

capital flows, within which cross-national credit borrowing and lending operate under collateral

constraints to alleviate defaults. We deliberately assume that production factors such as labor, land,

and capital are not mobile across national borders to highlight the role of global financial linkages

in generating global real business cycles.

More specifically,we assume that in each country households are international lenders and firms

are international borrowers, through a global bond market, under the standard assumption that the

former are more patient than the latter (`alaKiyotaki and Moore 1997). Although borrowing and

lending are channeled through a world financial market, production factors in each country cannot

be traded across countries. In other words, in each country firms (households) can borrow (lend)

internationally (subject to a collateral constraint) and the world interestr ateon financial debt adjusts

to reflect the tightness of the global supply and demand for loans.

54 International Journal of Economic Theory 15 (2019) 53–75 © IAET