Board of Governors of Federal Reserve System : Global Trade and GDP Co-Movement

May 27, 2020 at 02:53 pm

K.7

Global Trade and GDP Co-Movement

de Soyres, François and Alexandre Gaillard

Please cite paper as:

de Soyres, François and Alexandre Gaillard (2020). Global Trade and GDP Co-Movement. International Finance Discussion Papers 1282.

https://doi.org/10.17016/IFDP.2020.1282

International Finance Discussion Papers

Board of Governors of the Federal Reserve System

Number 1282 May 2020

Board of Governors of the Federal Reserve System

International Finance Discussion Papers

Number 1282

May 2020

Global Trade and GDP Co-Movement

François de Soyres and Alexandre Gaillard

NOTE: International Finance Discussion Papers (IFDPs) are preliminary materials circulated to stimulate discussion and critical comment. The analysis and conclusions set forth are those of the authors and do not indicate concurrence by other members of the research staff or the Board of Governors. References in publications to the International Finance Discussion Papers Series (other than acknowledgement) should be cleared with the author(s) to protect the tentative character of these papers. Recent IFDPs are available on the Web at www.federalreserve.gov/pubs/ifdp/. This paper can be downloaded without charge from the Social Science Research Network electronic library at www.ssrn.com.

Global Trade and GDP Co-movement*

François de Soyres†	Alexandre Gaillard‡
Federal Reserve Board	Toulouse School of Economics

This version: January 2020

Abstract

We revisit the association between trade and GDP comovement for 135 countries from 1970 to 2009. Guided by a simple theory, we introduce two notions of trade linkages: (i) the usual direct bilateral trade index and (ii) new indexes of common exposure to third countries capturing the role of similarity in trade networks. Both measures are economically and statistically associated with GDP correlation, suggesting an additional channel through which GDP ﬂuctuations propagate through trade linkages. Moreover, high income countries become more synchronized when the content of their trade is tilted toward inputs while trade in ﬁnal goods is key for low income countries. Finally, we present evidence that the density of the international trade network is associated with an ampliﬁcation of the association between global trade ﬂows and bilateral GDP comovement, leading to a signiﬁcant evolution of the trade comovement slope over the last two decades.

Keywords:International trade, international business cycle comovement, networks, input- output linkages

JEL Classiﬁcation: F15, F44, F62

*We thank the 2020 World Development Report team as well as seminar and conference participants for helpful comments. The views in this paper are solely the responsibility of the authors and should not necessarily be interpreted as reﬂecting the views of the Board of Governors of the Federal Reserve System or of any other person associated with the Federal Reserve System.

Email:francois.m.desoyres@frb.gov; Corresponding author. Address: Board of Governors of the Federal Reserve System, 2051 Constitution Avenue NW, Washington, DC
‡Email:alexandre.gaillard@tse-fr.eu.

Introduction

Over the past decades, both import and export ﬂows have increased much faster than GDP for almost all countries in the world. This march toward more open economies has been accompanied by a reorganisation of he world's production across different locations, with both trade in intermediate inputs and in ﬁnal goods trade representing an increasing share of world GDP, now reaching around three times the share observed in the 1970s. In valued-added terms, trade increased at an average annual growth rate of more than 5 percent during the 1990-2009 period, with the share of trade in intermediate inputs roughly constant at around 70% of total trade. During the same period, the average GDP co-movement across all pairs of countries rose from 6% to 38%.

The general surge in trade-over-GDP also implies more complex patterns for international propagation: when two countries are increasingly connected to the same direct or indirect trade partners, the associated surge in "third country" exposure can create systemic interdependence that operates over and above direct trade linkages. The consequences of these changes in trade patterns for the synchronization of economic activity are an important issue because they can have implications for macroeconomic policies.1In light of these global trends, several questions arise: did the rise of Global Value Chains (GVCs) have a speciﬁc effect on the correlation of GDP and its association with both direct and indirect trade ﬂows? Did the rise in production fragmentation have the same effect across income groups? Are direct trade linkages more important than common exposure to third markets? Did the sensitivity of GDP co-movement to an increase in bilateral trade ﬂows evolve over time?

Since the seminal paper by Frankel and Rose(1998), hereafter FR, a large empirical literature has studied the determinants of cross-country business cycle co-movement, showing that bilateral trade is an important and robust element associated with changes in GDP correlation while measures of ﬁnancial linkages or countries' sectoral similarity are not statistically associated with higher bilateral synchronization.2In this paper we re-assess the association between global trade and cross-country business cycle correlation using a large sample of 135 countries from 1970 to 2009, including high and low income countries. Using constructed panel data and controlling for both observed and unobserved heterogeneity between countries and

For example, the extent to which the Euro Zone can be considered as an optimal currency area (and, therefore a common monetary policy could be optimal) largely depends on the synchrony of business cycles among the
member countries.
2Among many others, seeFrankel and Rose (1998),Clark and van Wincoop (2001),Imbs (2004),Baxter and Kouparitsas (2005),Calderon et al. (2007),Inklaar et al. (2008),Di Giovanni and Levchenko (2010),Ng (2010),Liao and Santacreu (2015),di Giovanni et al. (2016) andDuval et al. (2015). The literature mostly focused on high income countries, with the notable exception ofCalderon et al. (2007), and set up estimation equations that unveil a singletime-invariantvalue for the association between bilateral trade ﬂows and business cycle correlation.

over time, we estimate the trade co-movement slope (TC-slope) across different income groups and unveil a series of new determinants of GDP co-movement, including the different role of the content of trade ﬂows for each income group as well as the presence of network effects and how they interact with bilateral proximity. Moreover, we also uncover important time variations in the TC-slope, which suggests that the sensitivity of GDP correlation to changes in trade proximity is not akin to a time-invariant deep parameter but is a function of other elements that evolve over time.

Building on earlier literature, this paper makes several contributions. First, starting with the role of bilateral trade ﬂows, we update previous analysis by separating trade ﬂows into trade in intermediate inputs and trade in ﬁnal goods and investigate separately their speciﬁc role for GDP synchronization for high and low income countries. As shown in de Soyres and Gail- lard(2019) and conﬁrmed in this paper, trade in intermediate inputs plays a particular role in the TC-slope for OECD countries. However, this ﬁnding is complemented and nuanced here by a novel insight regarding low income countries. Using only within country-pair variations and controlling for several factors including changes in the similarity of industrial structure across country pairs, we show that economies at the lower end of the income distribution experience an increase in the correlation of their GDP with their trade partners when the content of their trade ﬂows is more tilted toward ﬁnal goods trade. To understand this difference, we use disaggregated trade data and show that country-pairs with a large TC-slope in intermediate inputs are also characterized by high proximity in the sectoral composition of their trade ﬂows. All told, our analysis suggests that trade in inputs is associated with higher GDP correlation when countries have a similar industrial structure.3

Second, guided by recent debates on the role of Global Value Chains and the systemic interdependence that can arise from worldwide input-output linkages, we move beyond bilateral trade linkages and construct new indices of network proximity for all country pairs. We argue that changes in GDP synchronization between two countries can be the result of an increased common exposure to third markets, which can happen either at the ﬁrst order when two countries have similar trade partners or at the second order when countries' direct partners have similar partners. On the whole, our results reveal that ﬁrst order common exposure is particularly strong for high-income countries, while second-order proximity, a measure of more indirect propagation, is more prevalent for low income economies. Moreover, we show theoretically and empirically that the marginal increase in GDP comovement associated with the increase in any trade link is itself increasing in the overall density of the network. As such, this ampliﬁcation aspect linked with overall network density helps rationalize the wide array

To the extend that such similarity is in turn associated with a higher degree of input speciﬁcity, then this ﬁnding is fully consistent with results inBarrot and Sauvagnat(2016).

of TC-slopes found in the literature since any estimate depends on both the time and country coverage. Interestingly, this result challenges the usual assumption of a single time-invariant relationship between trade and GDP comovement. While the complementarity between network and bilateral trade could rationalize our ﬁnding that the TC-slope signiﬁcantly increased in the last two decades, we cannot rule out the possibility that other factors weighed into this evolution. In particular, the growth of price distortion could have also have played a role.

Finally, we provide various robustness checks, using different controls, measures and sample selection. For instance, controlling for bilateral ﬁnancial interconnection of the banking sector or foreign direct investment does not affect our main ﬁndings (although it reduces our sample due to data coverage). Overall, our results are robust to a wide range of speciﬁcations and trade indexes and highlight important disparities among country groups and over time.

Relationship to the literature.Starting with Frankel and Rose(1998), a large number of papers have studied and conﬁrmed the positive association between trade and GDP comove- ment in the cross-section.4This paper is mostly related to a few recent contributions. First, di Giovanni et al.(2016) uses a cross-section of French ﬁrms and presents evidence that international input-output linkages at the micro level are an important driver of the value added comovement observed at the macro level. Their evidence is in line with the ﬁndings of this paper and supports the role of Global Value Chains in the synchronization of GDP ﬂuctuations across countries.5Second, Liao and Santacreu(2015) is the ﬁrst to study the importance of the extensive margin for GDP and TFP synchronization and shows that changes in the number of products traded across countries (rather than the average shipment per product) plays an important role in the synchronization of GDP. Huo et al.(2019) uses a more structural approach and proposes a perfectly competitive production framework to measure technology and non- technology shocks and subsequently analyzes their cross-sectional properties. In this setup, international transmission through trade accounts for a third of total comovement. Third, Calderon et al.(2007) investigate the relationship between trade and business cycle comove- ment for both developed and developing countries. Based on cross sectional estimates, they ﬁnd that the impact of trade integration on business cycles is higher for industrial countries than for developing countries. Fourth, our paper is related to a recent series of papers developing accounting and theoretical frameworks to measure GVC participation, including Bems

See papers cited for instance in footnote 2.

Relatedly,Burstein et al.(2008) uses a cross section of trade ﬂows between US multinationals and their afﬁliates as well as trade between the United States and Mexican maquiladoras to measure production-sharing trade and its link with the business cycle. Moreover, Ng(2010) uses cross-country data from 30 countries and shows that bilateral production fragmentation has a positive effect on business cycle comovement. The concept of bilateral production fragmentation used is different from this paper as it takes into account only a subset of trade in intermediates, namely imported inputs that are then further embodied in exports. Moreover, the cross-sectional nature of the analysis allows for neither dyadic nor time windows ﬁxed effects.

et al.(2011) and others.

If the empirical association between bilateral trade and GDP comovement has long been known, the underlying economic mechanism leading to this relationship is still unclear. Using the workhorse IRBC with three countries, Kose and Yi(2006) have shown that the model can explain at most 10% of the slope between trade and business cycle synchronization, leading to what they called the Trade Comovement Puzzle (TCP). Since then, many papers including John- son(2014) or Duval et al.(2015), have reﬁned the puzzle, highlighting different ingredients that could bridge the gap between the data and the predictions of standard models.6

The rest of the paper is organized as follows. We ﬁrst provide a simple trade network model 2highlighting the role of trade in the global GDP-comovement. We then turn to our empirical contribution. Section 3presents the data and the different constructed variables used throughout the paper. Section 4investigates the global TC-slope not only across countries in different income groups, but also over time. We discuss the main implications of our results in section 5and, in section 6, test several possible explanations for some of the key differences between the results relative to high and low income countries. Finally, section 7concludes.

A simple trade network model

To motivate our empirical work and formalize our intuition, we begin by writing a parsimonious static model of international trade with multiple countries and sectors. Our main goal is to illustrate through a series of example several mechanisms through which GDP in two countries can be correlated. In particular, we show that GDP comovement is the result of a combination of many factors, including the correlation structure of shocks hitting every country in the world, bilateral trade linkages between countries as well as their indirect exposure to the rest of the trade network, and the association between gross output and GDP which can be time varying.7For simplicity, our framework abstracts from other relevant considerations such as the presence of ﬁnancial linkages or the possibility of common (or coordinated) monetary policy. Note, however, that we will control for these and other elements in our empirical

For a quantitative solution to the Trade Comovement Puzzle, seede Soyres and Gaillard(2019), where it is shown that production linkages alone are not sufﬁcient for a macro model to deliver a trade co-movement slope in
line with the data.
7As discussed in Johnson(2014), comovement in intermediate input, and the resulting comovement in gross output, does not necessarily translate into real value-added comovement. Building on this insight, de Soyres and Gaillard(2019) shows that the introduction of markups and extensive margin adjustments can create a mechanical link between input correlation and GDP correlation. We simplify the discussion here by introducing a simple ad hoc proportional transformation between output and real value-added that illustrates the fact that the sensitivity of GDP comovement to trade proximity is a function of other elements - which could include the prevalence of markups for example.

investigations in subsequent sections.

2.1 Basic setup

Production and pricing.Consider a world with many countries (i, j 2 f1, ..., Ng) and many sectors (s, s02 f1, ..., Sg). In country i and sector s, gross output is the result of a Cobb Douglas combination of three main elements: (1) an exogenous technology shock (Zi,s), (2) intermediate inputs from all other sector-countries in the world (Xij,,ss0), and (3) inelastic domestic factors of production (Li,s).

	Yi,s= Zi,sÕ(Xij,,ss0)ai,s	!	Lig,si,s,	(1)
	j,s0
	j,s0
with åj,s0aij,,ss0	+ gi,s= 1. The production cost of a representative ﬁrm in each country i and

sector s is a function of the price charged by its input suppliers and the suppliers of its sup- pliers. For simplicity we assume that there are no trade costs. Moreover, we also assume that ﬁrms' markups (mi,s) are completely exogenous and independent of the destination market which further implies that prices are equal across all destination markets. Denoting pi,sthe price of output produced by country-sector (i, s) and withe price of domestic factor in country i, standard cost minimization conditions imply that the price in (i, s) is given by:

				ci,s	j,s0
	pi,s =mi,s MCi,s =mi,s	wigi,sÕ(pj,s0)ai,s	(2)
	Z
				i,s	j,s0
With MC	the marginal cost in (i, s) and c	i,s	a constant depending only on parameters.8	As
i,s

is usual in all models with input-output linkages, the price in a given sector-country is a direct function of all other prices in the economy. To simplify notation, we stack prices in all countries and sectors into an (N S, 1) vector P, where the ﬁrst S rows contain the prices of all sectors in country 1, subsequent S rows contain all prices in country 2, etc... Taking the log and denoting by W the cross-countryinput-output matrix of the economy, prices are the

j,s0

8The variable ci,sis deﬁned as: ci,s= ggi,sÕj,s0aj,s0ai,s

i,s i,s

j i,s

j,s i,s

solution of a simple linear system:9

P= (

W) 1

k1,1

log(Z1,1)..

+ g1,1log(w1)

(3)

INS

log(

ZN,S

+ g

N,S

log(

)

BkN,S

Clearing conditions.Gross output is used both as an intermediate input in production

and to produce a composite ﬁnal good used by consumers. With Cobb Douglas production

function, the representative ﬁrm in country j and sector s0spends a fraction aon goods coming from (i, s), so that:

p	i,s	Xj,s0	=aj,s0	p	j,s0	Y	,	for all i, j, s, s0	(4)
	i,s	i,s		j,s0

Aggregate demand in each country j is denoted by Dj.10Country j addresses a fraction bof its total demand to country-sector (i, s), so that market clearing in the ﬁnal goods market can be written as:

pi,syij,s =bij,s Dj,

for all i, j, s

(5)

where y is the amount of good produced in (i, s) that are absorbed as ﬁnal demand in country j. We store all shares binto a (NS, N) matrix Bwhere each row corresponds to a sector-country (i, s) and the columns correspond to all countries.11Finally, the resource constraint condition is given by

Yi,s= åyij,s+ åXij,,ss, for all i, s(6)

j,s0

9We denote ki,s= log(mi,sc input-output matrices Wi,jas:

W =

i,s) and the Input-Output matrix W can be deﬁned by block using country-pair

	W1,1	W1,2. . .	1	, with (Wi,j)s,s0= aij,s	0
0 ..	..	..
B	WN.,1	. ...	WN.,N	C	,s
@				A

10A natural general equilibrium closing of the model would be to assume that total demand Diequals total income of domestic production factor wiLias well as domestic proﬁts. We keep things more general here and solve for gross output for any level of ﬁnal demand, which makes it possible, in principle, to study both supply shocks (through shocks to technology Zi,s) as well as demand shocks if were to introduce shocks to Di.

11Matrix Bis deﬁned as:

0	b11,2	b21,2	. . .	b1,2N	1
	b11,1	b21,1	. . .	b1,1N	C
B =B...	...	...	...
B					C
B b1	b2	. . .	bN	C
B	N,S	N,S		N,S	C
@					A

Combining (4), (5) and (6), we can solve for nominal output in each country and sector:

p1,1..Y1,1

D..

(7)

BpN,SYN,S

}

In this stylized framework, solving for gross output in each sector-country amounts to jointly solving for prices using (3) and nominal output using (7).

Deﬁning Real Value Added.Measuring real value added in this framework is not straight- forward. Statistical agencies measure real value added in each sector as the difference between gross output and intermediate input, measured using base period prices. As discussed in Ke- hoe and Ruhl(2008) or in Johnson(2014), in a perfect competition setting, this procedure amounts to measuring changes in domestic factor supply (i.e. changes in labor Li,s). Hence, without markups, our assumption that domestic factors are completely inelastic would lead to constant measured real value added. However, Basu and Fernald(2002), de Soyres and Gaillard(2019) and others note that things differ markedly when one introduces markups. By introducing a wedge between marginal cost and marginal revenue product of inputs, the presence of markups creates a proportional relationship between gross output and proﬁts ﬂuctuations. In such a case, even with inelastic domestic factor supply, real value added can still ﬂuctuate owing to movements in proﬁts.

We parsimoniously account for such a channel by positing a reduced form relationship between gross output Yi= åsYi,sand measured real GDP, so that RGDPi= åsLi,s+ kiYi. With ﬁxed domestic factor supply, changes in real GDP come only from gross output ﬂuctuations. In the rest of this section, we show how correlation of gross output ﬂuctuations can emerge from a variety of different channels, which we then formally test in the rest of the paper.

2.2 Propagation channels

Considering technology as the only source of shocks, the proportional change in gross output in any country-sector is a function of the vector of shocks and the Leontieff inverse:

Yb = [IW] 1Zb

(8)

In the rest of this section, we present stylized examples with speciﬁc Input-Output matrices to illustrate several determinants of bilateral comovement. In particular, we show that, absent of any bilateral trade between two countries, and, indeed, even in situations where two countries

do not export at all the global trade network could give rise to endogenous output correlation.

Consider a world with six countries and only one sector per country. We choose a speciﬁc

structure of input-output linkages in order to show how (i) bilateral trade, (ii) direct common

trade exposure, and (iii) indirect common trade exposure all play a role in bilateral output

(and ultimately GDP) comovement. The structure is described in ﬁgure 1and the associated W matrix is given by:

0	0	a2	a3	a4	0	0	1
B 02	1	1	1	02	0	C
0	02	0
W =B	a1	0	a3	0	a5	0	C
					6
B	0	0	0	0	0	C
B	a4	C
B							C
B	0	0	0	0	0	6	C
B	a5	C
B							C
B	0	0	0	0	0	0	C
B							C
@							A

Figure 1.Network representation of input-output linkages

Using equation (8), we can write the proportional change in gross output in country 1 as

a function of all shocks and trade linkages:12

Y1=	1		Z1+ a12Z2+ (a13+ a12a23)Z3+ a14Z4+ a12a25Z5+ (a14a46+ a12a25a56)Z6	(9)
j W	j
b	1		c1	c	3	1 3 c	1 c4	5 c	5 6	1 4 6 c	(10)
Y2=j W j	a2Z1	+ Z2	+ (a2	+ a2a1)Z3	+ a2a1Z4	+ a2Z5	+ (a2a5	+ a2a1a4)Z6
b			c	c		c	c	c		c

where we recall that aijis the spending share in country i on goods coming from country j.

The multifaceted effect of global trade.Let us consider the case where technology shocks

12The variable j W j is the determinant of matrix W

are uncorrelated, so that Cov(Zi, Zj) = 0 for all i and j. In such a case, correlation between Yb1and Yb2is solely due to global trade linkages. Using equations (9) and (10), we can write a simple expression for corr(Yb1, Yb2):13

corr(Y1	, Y2) = la12	+ a21+ (a13+ a12a23)(a23	+ a21a13) + a21(a14)2+ a12(a25)2	(11)
b	b

(a41a64 +a21a52a65)(a52a65 +a12a41a64)

Equation (11) reveals that several types of trade linkages can give rise to endogenous output co-movement. We will now examine three cases that provide economic intuition for the empirical exercise we perform in the next sections.

1. Only bilateral trade. Consider a situation where countries 1 and 2 both import inputs from one-another but do not trade with the rest of the world. In other words, a21and a12are strictly positive but all other input-output shares are zero. Following (11), the correlation between Yb1and Yb2is simply a function of bilateral trade shares:

corr(Y1	, Y2) = la12	+ a21		(12)
b	b

2. No bilateral trade and only ﬁrst order "third country" exposure. This situation happens when countries 1 and 2 import intermediates from country 3 (meaning that a31and a32are strictly positive) but all other input-output shares are zero. This situation is interesting because neither country 1 nor country 2 exports any value added to any other country. With uncorrelated technology shocks, the only reason countries 1 and 2 co-move is that they are commonly exposed to country 3. Using (11), we then get a simple expression for bilateral correlation of output:

corr(Y1	, Y2) = l a13a23	(13)
b	b

3. No bilateral trade and only second order "third country" exposure. In this case, the only trade ﬂows are as follows: country 6 exports inputs to countries 4 and 5, which themselves export to countries 1 and 2 respectively. In such a conﬁguration, countries 1 and 2 have neither bilateral ties nor any ﬁrst order network proximity since there is no overlap between their trade partners. However, they are both indirectly exposed to country 6.

13where lis deﬁned by

l= j W j2	q		1
Var(Y1)Var(Y2)
		c c

Equation (11) then yields the following expression of bilateral output correlation:

corr(Y1	, Y2) = la14a46a25a56		(14)
b	b

For simplicity, we chose here a sparse and symmetric second order network exposure, but other types of indirect exposure will naturally arise in the data given the high density of the actual network of trade linkages. For instance, indirect exposure could arise if country 6 is both directly linked to country 1 and indirectly linked to country 2.14

An inspection of equations (12) to (14) reveals that GDP co-movement is the result of several type of linkages, including direct bilateral trade links (case 1) as well as common exposure to third countries using ﬁrst- or second-order partners (cases 2 and 3, respectively).

The role of network density.So far, we have considered situations where different types of linkages were analyzed in isolation from one another. In practice, these mechanisms do not operate independently, and the density of the global trade network can act as a powerful ampliﬁcation factor. We illustrate this point by considering a situation where countries 1 and 2 trade with each-other (a21and a12are non-zero) and are also commonly exposed to country 3 (a31and a32are non-zero). Using equation (11), we can write bilateral output correlation as:

corr(Y1	, Y2) = la12	+ a21+ (a13+ a12a23)(a23+ a21a13)	>la12+ a21+ a13a23		(15)
b	b

The inequality in equation (15) reveals that the correlation stemming from the combination of both bilateral trade and common exposure is larger than the sum of each channel individually. As such, it shows the complementarity that arises from these channels that amplify one an- other. More broadly, the strength of each channel increases with the presence of other linkages in the trade network, which means one should not expect that the marginal effect of increasing any given link in the sparse network of the 1970s is the same as the effect of increasing a link in today's network. In the empirical exercise below, we test and provide support for such ampliﬁcation through network density.

The role of sectoral composition.We slightly modify our setup and consider a world with only two countries and two sectors. To streamline the discussion, we also assume that there are no trade ﬂows at all, implying that countries do not have any link with one another and technology shocks do not propagate across countries. Furthermore, technology shocks are sector speciﬁc and do not embed any country-speciﬁc component, in the sense that sectors s are hit by the same shock Zsin both countries. This assumption creates a link between

14Formally, this case happens when a61, a65and a52are all strictly positive.

sectoral specialization and bilateral comovement even in the absence of any trade ﬂows. We assume these shocks are uncorrelated across both sectors and follow a distribution with a common variance s2. Proportional changes in output in each country i and sector s are given by Yci,s= cZs.

Introducing pi,sas the share of sector s in country i's gross output, we can write the

aggregate change in country i as a function of sectoral changes: Yi

=pi,1Yi,1 +pi,2Yi,2 =

i,1Z1

+ p

i,2Z2

. Using our assumptions on shocks orthogonality and

the fact that p

= 1

i,1

ci,2

for both countries, we can write the following:15

corr(Y1, Y2) =

p1,1p2,1 + (1

p1,1)(1

p2,1)

(16)

qp1,12+ (1 p1,1)2

qp2,12

b b

+ (1

p2,1)2

From equation (16), it is apparent that bilateral output correlation equals zero whenever countries are fully specialized in different sectors, while it equals one if and only if p1,1= p2,1. In other words: bilateral co-movement increases with countries' similarity in their sectoral composition.

2.3 Key Takeaways from the Model

To summarize, the model developed in this section gives rise to four testable predictions:

1. Bilateral GDP correlation increases with direct bilateral trade.

2. Bilateral GDP correlation increases with common exposure to third countries through direct as well as indirect linkages (this prediction is what we call the network channel).

3. The previously mentioned channels complement one another in the sense that the marginal effect of increasing bilateral trade linkages or increasing common exposure to other countries depends on the density of the overall network of trade linkages. Owing to increasing trade linkages over the past four decades, this result also implies that both trade- and network-comovement slopes are expected to increase over time.

4. Bilateral GDP correlation increases with the similarity of sectoral composition between two countries..

In the rest of the paper, we will test for the presence of all these channels in the data as well as other related aspects of the relationship between global trade and GDP correlation.

15Note that the (common) variance s2disappears from this equation since it appears in both the numerator and the denominator.

It is worth noting that, on top of the forces discussed in the framework developed in this section, an obvious additional source of bilateral comovement is simply the correlation of country-speciﬁc shocks. Since this source is unlikely to be fully captured by our index of sectoral similarity discussed below, we will add country-pair ﬁxed effects in our speciﬁcation that effectively control for any time invariant factor affecting bilateral correlation.

Data sources and construction of our main variables

One of the objectives of this paper is to investigate the heterogeneity of the TC-slope across different levels of development as well as across different time periods. To be able to do so, we build on and expand previous studies by broadening both time and geographical coverage and we build a sample containing 40 years of data and a total of 135 countries, which accounts for almost the totality of world trade ﬂows and world GDP. To investigate the role of income level in the determinants of bilateral GDP correlation, we create four types of country-pairs:

pairs where both countries belong to the OECD, (ii) pairs where both countries are high income (deﬁned as HH pairs) according to the World Bank deﬁnition of income group, (iii) pairs where one country is high income and the other is not (deﬁned as HL pairs), and (iv) pairs where no country is categorized as high income (deﬁned as LL).16Note that for clarity of exposition we do not separate middle and low income countries, and only investigate the differences between high income and other countries. Moreover, the ﬁrst sub-sample (constructed based on OECD membership) is not informed by income level but is designed to capture possible speciﬁcities related to being part of what is usually considered as an "rich countries' club." Our analysis will reveal that results in the OECD and HH sub-samples turn out to be qualitatively similar but quantitatively different.

As will be clear below, all of our speciﬁcations are designed to control for unobserved country-pair heterogeneity by using only within country-pair time series variations. Hence, we divide our 40 years of time coverage, stretching from 1970 to 2009, into four non-overlapping time windows. In table 1, we report the share of total trade ﬂows of each income group in our sample, relative to total world trade ﬂows.

The extent to which countries have correlated GDP can be inﬂuenced by many factors beyond international trade, including correlated shocks, ﬁnancial linkages, common monetary policies, and so on. Because those other factors can themselves be correlated with the index of trade proximity in the cross section, using cross-sectional identiﬁcation could yield biased

16The classiﬁcation of high, middle or low income countries is taken from the World Bank classiﬁcation: http:

//databank.worldbank.org/data/download/site-content/OGHIST.xls.

Table. 1.Trade ﬂows in the different income groups b

	Total Flows a	Share of total trade (%)
Period	OECD	HH	HL	LL
1970:1979	303	60.7	65.2 32.4	2.1
1980:1989	881	64.5	70.6	29.4	1.9
1990:1999	1864	61.9	64.3	34.9	2.6
2000:2009	3972	48.1	47.8	46.5	6.2

in billions of US dollars.

selected income groups are not exclusive. Some countries among the LL group also appear in OECD. For instance, this is the case for Mexico.

results. Indeed, in their seminal paper, FR use cross-sectional variations to evaluate whether bilateral trade intensity correlates with business cycle synchronization, but their speciﬁcation does not rule out omitted variable bias such as, for example, the fact that neighboring countries have at the same time more correlated shocks and larger trade ﬂows. By constructing a panel dataset and controlling for both country-pair and time windows ﬁxed effects, this paper relates to recent studies that try to control for unobserved characteristics.17Therefore, in order to separate the effect of trade linkages from other unobservable elements, we construct a panel dataset by creating four periods of ten years each.18Within each time window, we compute GDP correlation as well as the average trade intensities deﬁned below.

3.1 Trade Proximity and GDP-comovement

GDP.We use annual GDP data from the Word Development Indicators (WDI) of the World Bank, measured using constant 2010 prices in US dollars.19For our analysis, GDP series need to be ﬁltered in order to extract the business cycle component from the trend. Our main and benchmark ﬁlter is the standard Hodrick-Prescott (HP) ﬁlter with a smoothing parameter of 100 which is consistent with the yearly frequency of our data. Such a transformation allows us to capture the standard business cycle ﬂuctuations. With this setting, we mostly keep ﬂuctuations that have a frequency between 8 and 32 quarters. In section 6, we provide robustness checks using a Baxter and King (BK) ﬁlter and a simple log-ﬁrst difference.20With

Di Giovanni and Levchenko(2010) includes country pair ﬁxed effects in a large cross-section of industry-level data with 55 countries from 1970 to 1999 in order to test for the relationship between sectoral trade and output (not value-added) comovement at the industry level. Duval et al.(2015) includes country pair ﬁxed and year effects in a panel of 63 countries from 1995 to 2013 and test the importance of value added trade in GDP comovement.
Adding time windows ﬁxed effects controls for the recent rise of world GDP correlation since the 1990s, which could be unrelated to trade intensity.
We used the data series called "NY.GDP.MKTP.KD".
We use a Baxter and King (BK) ﬁlter to isolatemedium-term ﬂuctuations in the spirit of Comin and Gertler(2006). We keep ﬂuctuations between 32 and 200 quarters, following Comin and Gertler(2006). A simple log-ﬁrst

the ﬁltered GDP, we compute the GDP correlation for each country-pair (i, j) within each time-window t of 10 years, denoted Corr GDPijt.

Trade Proximity.We collect data on bilateral trade ﬂows from the Observatory of Economic Complexity (MIT). This database covers 215 countries over the period 1962-2014. The data are classiﬁed according to the 4-digit Standard International Trade Classiﬁcation (SITC), Revision 2. Only products and commodities are considered. To classify trade ﬂows into ﬁnal goods and intermediate inputs, we use a concordance table from SITC Rev. 2 to Broad Economic Categories (BEC).21,22Finally, we exclude country-pairs with less than two time-windows for which trade proximity is available.

We then aggregate trade ﬂows in each category at the country-pair level. For each type of ﬂow d 2 ftotal, inter, f inalg (for total trade ﬂows, trade in intermediate inputs and trade in ﬁnal goods respectively) we construct an index for bilateral trade proximity of a country-pair (i, j) in a given time-window t, as follows:

			d		Td
				i$j,t	8d 2 ftotal, I, Fg	(17)
		Tradeijt	=
		GDPit+ GDPjt
where Td	= Td	+ Td	is total trade ﬂows between countries i and j, deﬁned as the sum
i$j,t	i!j,t	j!i,t

of exports from i to country j and exports from j to country i.23In the result tables below, we refer to Total Ttotal, Inter Tinterand Final T f inalfor simplicity.

3.2 Network Effects

A key contribution of this paper is to provide evidence that the association between GDP comovement and trade linkage operates not only through bilateral trade intensity, but also through common exposure to third countries, which we refer to as network effects.

First order network index.In a world with many countries, the bilateral index of trade proximity is not a sufﬁcient measure of trade linkages.24We ﬁrst complement the above co-

difference is a more "agnostic" transformation that accounts for both the cyclical and the trend components embodied in any year-to-year ﬂuctuation, but it is sometimes considered as less sensitive to researcher's assumptions and preferences regarding the parameters of the ﬁltering method.

21The concordance table from SITC Rev2 to BEC can be found on the UN Trade Statistics webpage: https: //unstats.un.org/unsd/trade/classifications/correspondence-tables.asp.

22We merge capital goods and intermediate inputs as a single bundle of intermediate inputs. Trade in capital goods is roughly 14% to 15% of total trade ﬂows. For robustness, we also consider trade in capital goods separately. The main results remain unchanged.

23This speciﬁcation is widely used in the literature. As a robustness check, we also adopt an alternative used

		Td	+Td	i,t		Td	+Td	i,t
		i	j,t	j		i	j,t	j
index: Tradeijtd	= max n		!GDPit!		,		!GDPjt!	o.

24The importance of third country effect is also mentioned in Kose and Yi(2006) and Duval et al.(2015) analyzes

2nd ijt

Ti,t

variate with an index of ﬁrst order network proximity that is constructed to reﬂect the fact that two countries might experience a surge in their GDP correlation if their exposure to a common third country increases. In other words, over and above changes in bilateral trade ﬂows, two countries that are increasingly linked to similar partners are likely to become more synchronized. To account for such a common exposure mechanism, we construct a third country index aiming to capture the ﬁrst order component of a trade network, such that:

networkijt= 1	2	åk	Ti,t	Tj,t	(18)
1st	1		Ti$k,t	Tj$k,t

where Ti$k,trepresents the total trade ﬂows between country i and country k and Ti,tdenotes the total trade ﬂows of country i vis-a-vis all of its partners. This index effectively measures the geographical overlap in two countries' trade partners. Note that country-pairs with very similar trade partners have an index close to one while two countries trading with completely different partners have an index of zero.

Second order Network effect. As a measure of 2nd order network proximity for any pair (i, j), we build an index measuring to what extend country i's partners are linked with country j's partners, weighted by the importance of the partners in terms of total trade ﬂows of the two countries i and j:

network2ijtnd=	1	å å(wt(i, z) + wt(i, y) + wt(j, z) + wt(j, y)) networkzyt!	(19)
4
		z2P(i) y2P(j)

where wt(i, z) = Ti$z,t. Under this speciﬁcation, the more the partners of my partner are similar to my partners in terms of 1st order network, the higher the index network .

Cross-Network effect. In the robustness tests (section6), we go a step further and construct another index capturingnon-symmetricsituations where a country's direct partners are linked with another country second order partners. We refer to this situation as across-networkeffect of trade proximity, denoted (cross networkijt) and deﬁned as follows:

cross networkijt= 1

= 12

å(j)wt(j, z) åk

Ti,t

Tz,t

+ z

å(i)wt(i, z) åk

Tj,t

Ti$k,t

Tz$k,t

Tj$k,t

wt(j, z)networkizt+ åwt(i, z)networkjzt

z2P(j)

z2P(i)

Tz$k,t

Tz,t

(20)

the role of indirect trade linkages between two countries using a value-added approach. Our approach differs from Duval et al.(2015) because common exposure to third countries can happen even when two countries do not exchange any value added with one-another.

Figure 2.Illustration of ﬁrst order, second order and cross network proximity indexes.

1stOrder Network	2ndOrder Network		Cross-Network
C	D	A	B	A	B
	E
		C	D	C	D

Note: dashed areas represent 1st order network. The second order effect and the cross-network effect can be represented as a combination of 1st order network effects.

where wt(i, z) = Ti$z,t. The index measures the extent to which a country i in the country-pair

Ti,t

(i, j) is similar in terms of trade partners (i.e. in terms of direct network index) to all countries

z 2 P(j) trading with its partner j, weighted by the importance of z in the total trade of j.

As an illustration, we combine the three network representations in ﬁgure 2. Finally, we

summarize in table 2the evolution of our three network indexes in our sample. Interestingly,

the ﬁrst order network effect is much larger in OECD countries relative to the other group

considered.

Table. 2.Network index in the different income groups a

						Network index *100

		First order			Second order			Cross-network
Period	OECD	HH	HL	LL		OECD	HH	HL	LL		OECD	HH	HL	LL


70:79	53.6	47.4 47.1	50.1		46.4	46.5 47.5	47.8		42.1	42.2 40.9	40.0

80:89	55.7	48.0	47.3	50.1		48.0	47.9	48.8	49.0		42.4	42.1	40.9	40.3
90:99	56.5	49.6	46.6	48.6		48.5	49.3	49.4	49.2		42.7	42.7	41.7	41.0
00:09	55.0	47.2	44.5	46.7		48.4	49.3	49.3	49.2		43.3	43.3	42.6	42.3

Numbers reported are the average over allcountry-pairs.

3.3 Proximity in sectoral composition

As discussed in section 2, if shocks have a sectoral component then two countries with increas-

ing similarity in sectoral specialization could experience a corresponding surge in business

cycle co-movements even in the absence of any trade linkages. In order to account for such a mechanism, we build two bilateral indexes of proximity in sectoral composition. The ﬁrst index is based on countries' proximity in terms of sector share in GDP while the second focuses on the proximity in traded goods, at the 4-digit SITC level or ISIC level, as proxy for domestic specialization in exported goods. Data for sector shares in GDP come from the World Bank's WDI. We use the share in value added of nine main sectors composed of service, agriculture and seven manufacturing sectors (textile, industry, machinery, chemical, high-tech, food and tabacco, and other).25Such an index is a direct measure of two countries' specialization, but its usefulness is somewhat limited by the high level of sectoral aggregation which allows us to capture only specialization in broad sectors. Moreover, data are available only for a subset of all countries.

We deﬁne the sectoral proximity index in terms of traded goods denoted exportijtproxfor a given country-pair (i, j) in time-window t as:

exportijt	= 1	2 s	åEX	EXi,ti,t		EXj,t		(21)
prox	1		2S		EX (s)		EXj,t(s)

where EXi,t(s) refers to total export of country i in sector s 2 SEX, with SEXbeing the set of sectors (each 4-digit SITC code or ISIC code, depending on the deﬁnition adopted). We deﬁne the sectoral proximity index in terms of sector shares in GDP, denoted sectorijtprox, for a given country-pair (i, j) in time-window t as:

sectorijt	= 1	2 så	iY,ti,t		Yj,t	(22)
prox	1	2S		Y (s)	Yj,t(s)

where Yi,t(s) refers to total value-added of country i in sector s 2 S, with S being the set of sectors. For both indexes, country pairs with very similar sectoral/trade composition have an index close to 1, while countries that completely specialize in different sectors have an index of 0. We provide in table 3the evolution of sectoral proximity and export proximity over time for the income groups considered. In section 4, we use the export proximity constructed at the 4-digit SITC level and leave the ISIC speciﬁcation as a robustness exercise in section 6.

Looking at indices based on exports as well as GDP, we note that country-pairs in the OECD are signiﬁcantly more similar than those in other groups. Moreover, the time evolution of these indices also reveals a higher convergence, in terms of economic structure, among OECD countries compared with other sub-samples.

25Data are available here:https://databank.worldbank.org/data/source/.

Table. 3.Sectoral and export proximity index in the different income groups a

			Export proximity*100					Sectoral proximity*100

		4-digit SITC				ISIC b					WDI c
Period	OECD	HH	HL	LL		OECD	HH	HL	LL			OECD	HH	HL	LL


70:79	29.9	21.9 11.3	14.2		46.1	35.4 28.7	36.8			85.2	83.3 75.3	78.4

80:89	32.6	21.2	12.0	15.0		48.4	34.6 26.8	33.7			88.8	84.5	78.6	81.5
90:99	37.3	24.5	14.0	15.6		52.6	38.6	26.8	30.0			89.5	88.1	78.8	81.6
00:09	38.1	26.0	16.1	17.1		53.3	39.4	28.7	30.7			89.6	83.5	75.5	78.2

Number reported is the average over allcountry-pairs.

classify

goods

and

products

the

ISIC

level

following

the

corre-

spondencetablehttps://unstats.un.org/unsd/tradekb/Knowledgebase/50054/Correlation-between-ISIC-and-SITC-codes-or-Commodity-and-Industry.

WDI refers to sectoral proximity in terms of share of WDI sectors GDP in total GDP. Data is available here:

https://databank.worldbank.org/data/source/.

The GlobalTrade-Comovement Slope

In this section, we revisit the seminal FR analysis and use all variables deﬁned in the previous section as well as additional controls to investigate the determinants of business cycle correlation for different income groups and time periods. We proceed step-by-step and gradually introduce our variables.

4.1 The initial Frankel and Rose(1998) speciﬁcation

We ﬁrst review the FR results by extending the analysis to a large sample of countries separated into different income groups. Following the more recent literature, we use a panel ﬁxed effect in order to control for unobserved heterogeneity between country-pairs, as well as changes in economic conditions over time that are not related to trade.26As a ﬁrst step, we estimate a panel with country-pair (CP) and time-window (TW) ﬁxed effects with the following speciﬁcation:

Corr GDPijt= b1ln(Tradeijttotal) + Xijt+ CPij+ TWt+ eijt,

(23)

where Xijtis a vector of additional control variables that includes dummies for URSS countries, the euro area, and the different waves of the European Union. On the one hand, the introduction of CP ﬁxed effects means that we are using only within country-pair time variations for the identiﬁcation. These dummies effectively control for time invariant factors

26In order to discriminate between ﬁxed or random effects, we run a Hausman test which display a signiﬁcant difference (p < 0.001), and we therefore reject the random effect model.

that can inﬂuence GDP comovement between two countries, such as distance, common border, common language, etc. On the other hand, TW ﬁxed effects capture aggregate changes in GDP comovement for all country-pairs in the world that could be due to aggregate shocks. In this speciﬁcation as well as all subsequent analysis, standard errors are robust to clustering at the country-pair level, which accounts for serial correlation across time. That is, we allow for the error term to have a ﬁxed country-pair component common to all (i, j) observations.

In a second step, we introduce our network indexes (ﬁrst and second order), which aim to capture the network effect of trade on GDP comovement stemming from both direct and indirect exposure to third countries. For this exercise, we use the following speciﬁcation:

Corr GDPijt= b1ln(Tradeijttotal) + gnetworkijt+ Xijt+ CPij+ TWt+ eijt

(24)

In equation (24), networkijtdeﬁnes a vector composed of the ﬁrst and second order network measures discussed above. The results are gathered in table 4.

Two main results emerge. First, as previously highlighted in the literature, trade proximity using total trade ﬂows is signiﬁcantly associated with more GDP correlation, for all considered groups. However, the strength of this association is very heterogeneous. Using the point estimate obtained with all country pairs, we ﬁnd that moving from the 25th to the 75th percentiles of log total trade is associated with an increase in GDP correlation of 5.0 percentage points. The same number increases up to 16.7 percentage points for OECD country pairs, 7.1 percentage points for pairs in the HH group, 5.9 percentage points for the HL group and 9.3 percentage for the LL sub-sample.

Second, the effect of trade through the network effect is high and signiﬁcant. According to our point estimate, moving from the 25th to the 75th quantiles of the direct network index implies an increase in GDP correlation of about 7.3 percentage points for all country-pairs, again with stark differences across sub-samples. For pairs in the OECD group, moving from the 25th to the 75th percentiles is associated with an impressive 30.9 percentage point increase in bilateral GDP correlation, while it is 14.6 percentage point for pairs in the HH group and only 6.2 percentage points for pairs in HL. Interestingly, the strength of a marginal increase in the direct network indexes is decreasing as the sample includes countries at the lower end of the income distribution, with the latter effect becoming statistically insigniﬁcant for the LL group.

Interestingly, the second order network index also plays a signiﬁcant role for GDP co- movement when using the whole sample. Concerning the classiﬁcation in terms of income group, the point estimates increase as we move to the low income group. For example, for

the LL group, when moving from the 25th to the 75th quantiles of this index, GDP-correlation increases 8.8%, while the ﬁrst order network effect is insigniﬁcantly correlated with more GDP-comovement. This result highlights the possible strong dependence of those country- pairs to the global network as opposed to their more direct bilateral network. When focusing on the particular OECD group, we ﬁnd the second order network effect is particularly high, with an increase in GDP correlation of 21.3% when moving from the ﬁrst to the last quartiles.

All told, this ﬁrst exercise reveals a subtle association between trade and GDP co-movement. While previous investigations highlighted the role of either direct or indirect bilateral trade, the economic and statistical signiﬁcance of our network indices sheds light on an additional channel stemming from increasing exposure to other countries. As we will further show below, the strength of this new channel is increasing over time, which makes it all the more relevant for understanding recent and future changes in cross-country business cycle synchro- nization.

Table. 4.Trade Comovement slope with total trade index

					corr GDP
	All	All	OECD	OECD	HH	HH	HL	HL	LL	LL
ln(Trade)	0.023	0.021	0.083 0.090	0.038	0.028	0.024 0.023 0.032 0.038
	(0.005) (0.006) (0.030) (0.033)	(0.013)	(0.015)	(0.006)	(0.006)	(0.012)	(0.006)
network1st		0.293		1.082		0.540		0.254		0.044
		(0.067)		(0.262)		(0.163)		(0.073)		(0.073)
network2nd		0.457		2.914		0.201		0.567		1.143
		(0.221)		(0.913)		(0.555)		(0.243)		(0.243)
CP+TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	13,079	1,224	1,224	2,541	2,541	10,538	10,538	2,745	2,745
R2	0.006	0.009	0.068	0.094	0.033	0.039	0.002	0.005	0.005	0.009
Notes:							p<0.1; p<0.05; p<0.01.

4.2 Accounting for trade in intermediate inputs and ﬁnal goods

Expanding on our results using total trade ﬂows, we now reﬁne the analysis and decompose total trade ﬂows into two sub-categories: trade in intermediate inputs and trade in ﬁnal goods. As discussed in de Soyres and Gaillard(2019) for the case of high income countries, trade in intermediate inputs is signiﬁcantly correlated with GDP comovement, while trade in ﬁnal goods is not.27In this section, we test the differential relationship between business cycle

27In de Soyres and Gaillard(2019), we also show theoretically how international input-output linkages, coupled with market power and extensive margin adjustments, can quantitatively generate a strong link between trade in

synchronization and both trade in intermediate inputs and trade in ﬁnal goods in a larger sample covering different income groups.

We run the following speciﬁcation (with and without network effects), where we disag- gregate total trade ﬂows (Tradeijttotal) into trade in intermediate inputs (Tradeinterijt) and trade in ﬁnal goods (Tradeijtf inal):

Corr GDPijt= b1ln(Tradeijtinter) + b2ln(Tradeijtf inal) + Xijt+ CPij+ TWt+ eijt

(25)

Corr GDPijt= b1ln(Tradeinterijt) + b2ln(Tradeijtf inal) + gnetworkijt+ Xijt+ CPij+ TWt+ eijt

(26)

The results are shown in table 5. When focusing on country-pairs in OECD and HH, we see that the TC slope is signiﬁcantly driven by trade in intermediate inputs as opposed to trade in ﬁnal goods28Turning to country-pairs in the HL and LL groups, we ﬁnd an opposite result: the TC slope is signiﬁcantly related to more trade in ﬁnal goods while trade in intermediate inputs is not signiﬁcantly associated with higher GDP comovement. These ﬁndings are also strongly economically signiﬁcant: according to the point estimate obtained when controlling for network effects, moving from the 25th to the 75th quantiles of log trade in intermediate inputs is associated with a 21.7 percentage points increase in GDP correlation for pairs in the OECD group and a 12.9 percentage point increase for pairs in the HH group. For pairs in the HL and LL groups, moving from the 25th to the 75th quantile of log trade in ﬁnal goods increases respectively GDP comovement by 4.9 and 5.1 percentage points respectively. In section 6, we show these results are robust to a number of alternative speciﬁcations, including ﬁnancial controls (FDI and constructed ﬁnancial interconnection BIS indexes), different GDP ﬁlters and different measures of trade intensities.

There are several possible explanations for the difference between high and low income countries. For example, one could argue that intermediate inputs traded by low income countries are more standardized than the heavily customized products traded between high income countries.29We further investigate this issue below and argue that at least two channels might be at play: (i) similar sectoral specialization between two countries seems to amplify the effect of intermediate input trade on GDP correlation ; and (ii) there is a different time evolution of market power in low versus high income countries.

intermediate inputs and GDP-comovement, resolving the Trade-Comovement Puzzle

Notice that we combine trade in capital goods with trade in intermediate inputs. Separating those ﬂows to the regression provides similar results as shown in the sensitive analysis.
Note that intermediate inputs include many commodities sold on the global market and little bilateral stick- iness in thebuyer-supplier relationship. For such goods, it is not surprising that direct bilateral trade is not associated with bilateral GDP comovement.

Table. 5.Trade Comovement slope with disaggregated trade index

					corr GDP
	All	All	OECD	OECD	HH	HH	HL	HL	LL	LL
ln(inter)	0.011	0.009	0.106 0.103	0.043 0.034	0.008	0.007	0.014	0.018
	(0.005)	(0.005) (0.030) (0.030)	(0.013)	(0.014)	(0.006)	(0.006)	(0.011) (0.006)
ln(ﬁnal)	0.012	0.012	0.023	0.009	0.011	0.009	0.017 0.016	0.015	0.017
	(0.005)	(0.005) (0.024) (0.025)	(0.012)	(0.012)	(0.005)	(0.005)	(0.010) (0.005)
network1st		0.305		1.046		0.521		0.264		0.040
		(0.067)		(0.259)		(0.163)		(0.073)		(0.073)
network2nd		0.416		2.906		0.170		0.518		1.076
		(0.221)		(0.938)		(0.550)		(0.243)		(0.243)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	13,079	1,224	1,224	2,541	2,541	10,538	10,538	2,745	2,745
R2	0.006	0.009	0.074	0.099	0.035	0.040	0.003	0.005	0.004	0.008
Notes:							p<0.1; p<0.05; p<0.01.

4.3 The role of sectoral proximity and export proximity

We complement the analysis with additional controls that aim to capture the proximity in terms of sectoral composition. As shown in section 2, if two countries have similar sectors, they are more likely to experience GDP comovement. We test this prediction using two measures of proximity: (i) proximity in terms of exported goods using our index exportijtproxand (ii) proximity in terms of GDP sector shares captured by the index sectorijtprox. While exportijtproxis available for all considered country-pairs, sectorijtproxis constrained by data availability. We run the following two speciﬁcations:

Corr GDPijt= b1ln(Tradeijtinter) + b2ln(Tradeijtf inal) + gnetworkijt+ a1exportijtprox
+ Xijt+ CPij+ TWt+ eijt	(27)
Corr GDPijt= b1ln(Tradeijtinter) + b2ln(Tradeijtf inal) + gnetworkijt+ a1exportijtprox
+ a2sectorijtprox+ Xijt+ CPij+ TWt+ eijt	(28)

The results are gathered in table 6. On the one hand, focusing on the ﬁrst ﬁve columns, we ﬁnd that similarity in the type of traded goods has a surprisingly negative effect on GDP- comovement, which is not captured by our simple model in section 2. This result could be explained by the possible substitutability between traded goods, which implies that a positive technology shock in one country decreases the market share of producers in other countries, leading to ﬂuctuations in the opposite direction when they trade similar goods. On the other

hand, other estimated coefﬁcients are not sensitive to the addition of exportijtprox. Turning to the effect of sectoral proximity (last ﬁve columns) on the sub-sample for which data are available, we do not ﬁnd a statistically signiﬁcant effect of sectorijtproxon GDP-comovement.

Table. 6.The role of sectoral proximity and export proximity

					corr GDP
	All	OECD	HH	HL	LL	All	OECD	HH	HL	LL
ln(inter)	0.010	0.102	0.037	0.008	0.018	0.010	0.142	0.089	0.009	0.044
	(0.005)	(0.030)	(0.014)	(0.006)	(0.011) (0.014)	(0.082)	(0.040)	(0.014) (0.022)
ln(ﬁnal)	0.013	0.008	0.007	0.017	0.017	0.002	0.046	0.189	0.013	0.010
	(0.005)	(0.025)	(0.013)	(0.005)	(0.010) (0.012)	(0.050)	(0.042)	(0.013) (0.021)
network1st	0.307	0.967	0.445	0.271	0.027	0.321	0.703	0.287	0.324	0.345
	(0.066)	(0.269)	(0.162)	(0.073)	(0.156) (0.201)	(0.660)	(0.504)	(0.221) (0.368)
network2nd	0.446	2.745	0.302	0.554	1.081 2.103	3.216	1.107	2.471 2.396
	(0.221)	(0.955)	(0.550)	(0.243)	(0.469) (0.607)	(1.885)	(1.544)	(0.656) (0.976)
exportprox	0.394	0.381	0.754	0.318	0.075	0.091	0.916	0.928	0.024	0.333
	(0.083)	(0.237)	(0.182)	(0.092)	(0.146) (0.188)	(0.473)	(0.439)	(0.206) (0.295)
sectorprox						0.296	0.249	0.430	0.172	0.235
						(0.189)	(0.705) (0.430)	(0.213) (0.409)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	1,224	2,541	10,538	2,745	4,499	655	893	3,606	1,091
R2	0.012	0.101	0.049	0.007	0.008	0.025	0.165	0.166	0.017	0.022
Notes:							p<0.1; p<0.05; p<0.01.

4.4 The evolution of the TC Slope from 1970 to 2009

Having established the link between global trade ﬂows and GDP comovement for different income groups, we now turn to the issue of the potential time evolution of the association between a marginal increase in trade and business cycle synchronization. More precisely, we provide evidence regarding a noticeable evolution of the TC slope between 1970 and 2009. This evidence is of particular importance since GDP-comovement surged between 1990 and 2009, for many countries, including those in the low income groups. Moreover, establishing that the association between trade and business cycle synchronization not only varies by income group but is also time varying yields two additional beneﬁts. First it helps reconcile different values of the TC-slope that are found in the literature and which rely on different geographic and time coverages. Second, it would lend support to the hypothesis that the marginal effect of increasing (either direct or indirect) trade ﬂows between two countries changes with economic

conditions. Such a hypothesis can then be investigated further in order to uncover what are the factors that enable and amplify the relationship between trade and GDP comovement.

We introduce a dummy variable LTWtwhich equals to 1 for the last two time-windows in our sample - that is for the periods 1990:1999 and 2000:2009 - and 0 otherwise. This "Late Time Window" dummy is then interacted with the determinants of GDP comovement, allowing us to formally test for the difference between the TC-slope in earlier time windows and the slope observed toward the end of the time coverage.30Formally, we now test the change in the slope using the following speciﬁcations:

Corr GDPijt= b1ln(Tradeijtinter) + b2	LTWtln(Tradeijtinter) + b3ln(Tradeijtf inal)
+ b4LTWtln(Tradeijtf inal) + gnetworkijt+ Xijt+ CPij+ TWt+ eijt(29)
Corr GDPijt= b1ln(Tradeijtinter) + b2	LTWtln(Tradeijtinter) + b3ln(Tradeijtf inal)

b4LTWtln(Tradeijtf inal) + g1networkijt+ g2LTWtnetworkijt

+ Xijt+ CPij+ TWt+ eijt

(30)

where Xijtrefers again to controls. By adding these interaction terms, we specify that co- efﬁcients b2, b4and g2indicate whether the TC slope estimated with respect to trade in intermediate inputs and ﬁnal goods and the coefﬁcients associated with network effects in the period 1990-2009 are different from the coefﬁcients estimated using the period 1970-1989.

We present our results in table 7. Looking at non-interacted point estimates, we see that the ﬁndings are consistent with previous speciﬁcations, with trade in intermediate inputs being signiﬁcantly correlated with more GDP comovement for country pairs in the OECD and HH sub-sample, while pairs in the other groupings feature a more prominent role for trade in ﬁnal goods. Moreover, the results show that the TC-slope changed signiﬁcantly over time for some country groups. Among developed countries (that is, in the OECD and HH groups), the TC slope in intermediate inputs signiﬁcantly increased over time. Between 1970 and 1989, the estimates indicate a signiﬁcant TC-slope of about 0.066 for country-pairs among the OECD group, while it rises to 0.156 for the period 1990 to 2009. In terms of magnitude, moving from the 25th to the 75th percentile of log trade in intermediate inputs (for all time-windows) would have implied an increase in GDP comovement of about 13.0 percentage point from 1970 to 1989, which is much lower than the 31.2 percentage point increase corresponding to the slope estimated using the 1990 to 2009 period. For HH, the TC-slope is statistically

30Note that with CP ﬁxed effects we are only using within country-pair time variations in trade proximity and GDP correlation. Hence, it is important for our Late Time Window dummy to cover (at least) two time-windows so that there are time variations within the late sub-sample.

different from zero only for the 1990 to 2009 period. In turn, it is also worth noting that the TC-slope in ﬁnal goods also increased over time for the HL group, implying that an increase in ﬁnal goods trade between high and low income countries is associated with a stronger increase in GDP comovement in recent time windows.

Finally, as regards the time evolution of the association between network proximity and GDP comovement, we ﬁnd a signiﬁcant positive increase in the network-comovement slope in the period 1990 to 2009 relative to the period 1970 to 1989 for almost all sub-samples. The high point estimates of these interacted terms reveal that the surge in the association between trade network and business cycle synchronization is very large. Altogether, these ﬁndings show that the association between international trade linkages and GDP correlation experienced a strong increase over time, either directly (through bilateral trade) or indirectly (via the network effect).

Table. 7.Time evolution of the Trade Comovement slope

					corr GDP
	All	All	OECD	OECD	HH	HH	HL	HL	LL	LL
ln(inter)	0.002	0.004	0.074	0.066	0.013	0.008	0.003	0.005	0.012	0.015
	(0.006)	(0.006) (0.033) (0.033)	(0.016)	(0.016)	(0.007)	(0.007)	(0.013) (0.007)
LTW*ln(inter)	0.010	0.006	0.074 0.090	0.028	0.025	0.006	0.002	0.012	0.007
	(0.005)	(0.005) (0.029) (0.031)	(0.011)	(0.011)	(0.006)	(0.006)	(0.013) (0.006)
ln(ﬁnal)	0.007	0.008	0.019	0.019	0.021	0.022	0.011	0.011	0.019 0.020
	(0.006)	(0.006) (0.028) (0.028)	(0.014)	(0.014)	(0.006)	(0.006)	(0.011) (0.006)
LTW*ln(ﬁnal)	0.008	0.007	0.034	0.045	0.009	0.004	0.010	0.010	0.006	0.009
	(0.005)	(0.005) (0.032) (0.033)	(0.011)	(0.011)	(0.006)	(0.006)	(0.012) (0.006)
network1st	0.317	0.211	0.800 0.699	0.608 0.482	0.273 0.189	0.033	0.133
	(0.066)	(0.073) (0.269) (0.273)	(0.165)	(0.176)	(0.073)	(0.082)	(0.156) (0.082)
network2nd	0.375	0.030	2.830	1.944	0.637	1.154	0.509	0.102	1.101 1.302
	(0.221)	(0.227) (0.984) (1.012)	(0.552)	(0.553)	(0.244)	(0.252)	(0.469) (0.252)
LTW*network1st		0.236		0.211		0.369		0.180		0.154
		(0.058)		(0.158)		(0.117)		(0.068)		(0.068)
LTW*network2nd		0.661		1.056		0.846		0.643		0.157
		(0.129)		(0.404)		(0.246)		(0.150)		(0.150)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	13,079	1,224	1,224	2,541	2,541	10,538	10,538	2,745	2,745
R2	0.015	0.019	0.110	0.121	0.062	0.071	0.010	0.013	0.008	0.009
Notes:							p<0.1; p<0.05; p<0.01.

Ti,t

4.5 Network density as an ampliﬁcation channel

In section 2, we discussed a mechanism that could explain the recent increase in the strength of the association between global trade and business cycle synchronization - namely, the role of overall network density. Our model highlights how network density interacts with the channels discussed above and strengthens the connection between individual links and bilateral GDP comovement.

To show the change in network density over time, we ﬁrst display in table 8the average share of total worldwide trade ﬂows over total worldwide GDP in each of our four time windows, where we normalize with respect to the value in the ﬁrst time window. Since the 1970s, the average trade ﬂow over GDP has more than tripled. As a result of such an increase in network density, we expect a corresponding surge in the correlation between trade and GDP comovement.

Table. 8.Total trade ﬂows over worldwide GDP

Period	70:79 80:89 90:99 00:09
Trade ﬂows / GDP	1.0 1.81 2.13 3.10

We then move on to formally testing our intuition and implement a bilateral measure of overall network connectiveness that reﬂects how much countries in a given country-pair are connected to the rest of the trade network. We compute the average bilateral network density for a given country-pair, as follows:

Network Densityijt=	w(i, j) åz2P(i)jTradeitot$z,t+ w(j, i) åz2P(j)iTradeitot$z,t	(31)
w(i, j) + w(j, i)

where P(i) jdeﬁnes the set of i-partners except the country j and wt(i, j) = Ti$j,t. This index measures the average trade volume over GDP of the two countries within the country-pair (i, j) when bilateral trade ﬂows are not taken into account, and it aims to measure the connectedness of two countries to the rest of the network. In this sense, it should be interpreted not as a measure of overall network density, but rather as a measure of trade proximity between the pair at hand and the rest of the world. Table 9presents the time evolution of this measure for each sub-sample.

Table. 9.Evolution of our bilateral measure of network densitya

			Income Group
Period	All	OECD	HH	HL	LL

70:79	0.3	0.5	0.5	0.3	0.2
80:89	0.7	1.0	1.0	0.7	0.4
90:99	1.0	1.6	1.5	0.9	0.6
00:09	1.6	2.4	2.2	1.4	1.0

Reported numbers are average over allcountry-pairs.

We test the following speciﬁcation:

Corr GDPijt= b1ln(Tradeinterijt) + b2ln(Tradeijtf inal) + b3densityijtln(Tradeinterijt)

b4densityijtln(Tradeijtf inal) + g1networkijt+ g2densityijtnetworkijt

+ Xijt+ CPij+ TWt+ eijt

(32)

Notice that our measure of bilateral density is directly linked to the ﬁrst order network index as the later measures the intensive margin of the ﬁrst order trade network, while the former can be interpreted as measuring similarity in the ﬁrst order trade network.31Table 10summarizes the ﬁndings.

The results present interesting differences across income groups. Regarding the ﬁrst column of table 10which presents the results using the whole sample, the interaction between our bilateral measure of density and trade network effects is positive, suggesting that country- pairs that are more connected to the rest of the world feature a higher marginal effect of network proximity. We also ﬁnd that the interaction between bilateral trade and bilateral density is signiﬁcant with different patterns in different sub-samples: bilateral density acts as an am- pliﬁer of intermediate trade proximity for the OECD and HH groups, while it strengthens the role of ﬁnal goods trade in the HL group. Overall, our ﬁndings imply that the TC-slope usually measured in the literature is a function of overall connectivity between a country-pair and the rest of the world. In other words, bilateral trade ﬂows have a higher marginal effect on GDP-comovement when two countries trade more with each other.

31As a robustness, we also used total trade ﬂows over worldwide GDP as a measure of network density and interacted it our ﬁrst order network effect. Results are similar in this case, although the logic of the estimation differs markedly: using world trade over world GDP as a measure of density means means the index is not bilateral and mostly measure an increasing trend for the whole sample. We see this exercise as conﬁrming the ﬁndings in section 4.4in the sense that there is a worldwide increase in the association between global trade and bilateral GDP co-movement.

Table. 10.Network density and Trade Comovement Slope

			corr GDP
	All	OECD	HH	HL	LL
ln(inter)	0.004	0.033	0.005	0.009	0.015
	(0.006)	(0.035)	(0.017)	(0.007)	(0.014)
ln(ﬁnal)	0.003	0.042	0.007	0.003	0.024
	(0.006)	(0.031)	(0.015)	(0.006)	(0.012)
density*ln(inter)	0.004	0.095	0.030	0.006	0.005
	(0.004)	(0.021)	(0.008)	(0.005)	(0.013)
density*ln(ﬁnal) 0.011	0.030	0.000	0.016	0.020
	(0.004)	(0.017)	(0.007)	(0.004)	(0.012)
network1st	0.230	0.931	0.506	0.165	0.212
	(0.075)	(0.298)	(0.173)	(0.085)	(0.175)
network2nd	0.213	2.588	0.671	0.302	1.480
	(0.228)	(0.999)	(0.573)	(0.250)	(0.499)
density*1st net.	0.085	0.076	0.003	0.106	0.244
	(0.035)	(0.093)	(0.061)	(0.044)	(0.104)
density*2nd net.	0.162	0.607	0.272	0.158	0.582
	(0.063)	(0.260) (0.099) (0.083)	(0.233)
CP + TW FE	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes
Observations	13,079	1,224	2,541	10,538	2,745
R2	0.016	0.122	0.066	0.012	0.016
Notes:		p<0.1; p<0.05; p<0.01.

4.6 Summary of empirical evidence

The preceding sections provided empirical support for the theoretical results discussed in section 2and offered novel insights on the complex association between global trade ﬂows and bilateral GDP comovement. In particular, our ﬁndings can be summed up as follows:

1. The correlation between trade in intermediate inputs and GDP comovement is signiﬁcant and positive for countries in the OECD and HH groups, suggesting a speciﬁc role for Global Value Chains. Interestingly, trade in ﬁnal goods is signiﬁcantly correlated with higher business cycle synchronization for the low income groups.

2. Common exposure to third countries, measured using our network indices is signiﬁ- cantly positively correlated with more GDP comovement. First order network effects decay as we move to low income group while second order network effects increase for those in the lowest income group.

3. Similarity in exported goods composition is signiﬁcantly negatively correlated with GDP comovement while there is no statistically signiﬁcant effect of sectoral composition on GDP-comovement.

4. The correlation between GDP comovement and both bilateral trade and network effects tends to increase over-time. As suggested by our simple model, this increase could be rationalized by a surge of trade network density which can amplify the association between global trade and bilateral comovement. In the next section, we also discuss a complementary channel that could rationalize the time evolution of point estimates - namely, an increase in the sensitivity of GDP to foreign shocks due to higher markups.

Discussion

We now dig further into two particular ﬁndings: (1) the differential role of sector versus trade proximity, and (2) the time evolution of the marginal association between trade and GDP comovement. First, we show that the association between trade and business cycle correlation strongly depends on the sectoral composition of the economy, which creates another link between our explanatory variables. Second, we show that market powers have substantially risen in the 1090s, with the possible implication that international shock transmission could have strengthened even in the absence of increased trade ﬂows.

5.1 Global Value Chain and Sectoral Composition

A number of studies document the strong inﬂuence of global value chain (GVC) in transmitting shocks across countries. Indeed, to the extent that inputs tend to be more customized and less substitutable than ﬁnal goods, bilateral production fragmentation can be associated with higher complementarity between production factors, which would then lead to higher GDP comovement. Conversely, when two countries trade ﬁnal goods while producing very similar goods, they can be seen as competing in the same market, which implies that a positive supply side shock (such as a technology innovation) in one country leads to an increase in this country's market share at the expanse of its trade partner, leading to opposite GDP movements.

In this section, we test the hypothesis that two economies with similar sectors and similar exported goods are more likely to comove when trade in intermediate inputs is high, and that they are less likely to comove when trade in ﬁnal goods is high. To do so, we run the

following two speciﬁcations:

Corr GDPijt= b1ln(Tradeinterijt) + b2ln(Tradeijtf inal) + b3exportijtprox

b4exportijtproxln(Tradeinterijt) + b5exportijtproxln(Tradeijtf inal)

+ Xijt+ CPij+ TWt+ eijt

(33)

Corr GDPijt= b1ln(Tradeinterijt) + b2ln(Tradeijtf inal) + b3sectorijtprox

b4sectorijtproxln(Tradeinterijt) + b5sectorijtproxln(Tradeijtf inal)

+ Xijt+ CPij+ TWt+ eijt

(34)

where Xijtrefers to controls, including ﬁrst and second order network indices. The results are shown in table 11and can be stated as follows. Country-pairs with similar export composition co-move more when they trade more in intermediate inputs and co-move less when they trade more ﬁnal goods. Given the mean export proximity index in table 3and the estimates in table 11, moving from the 25th to the 75th quantiles of the log trade index in intermediate inputs would imply an increase in GDP comovement of 0.20 for the HH group against 0.012 for the HL group. On the opposite side, moving from the 25th to the 75th quantiles of the log trade index in ﬁnal goods would imply an increase in GDP comovement of negative 0.01 for the HH group against 0.05 for the HL group.

Regarding the similarity in GDP sector shares, the results are signiﬁcant and consistent for OECD group only. For this sub-sample, given the mean sectoral proximity index, moving from the 25th to the 75th percentiles of the log intermediate inputs (respectively ﬁnal goods) distribution would lead to an increase in GDP comovement of 0.29 (respectively negative 0.11). Notice, however, that moving from the 25th to the 75th quartile of the sectoral composition index would imply an average decrease in GDP comovement of 0.23.

5.2 Market Power

In de Soyres and Gaillard(2019), we argued that the rise in market power (and hence a greater share of proﬁts in overall GDP) can help solve for the Trade Comovement Puzzle by generating a disconnect between movements in production factors (capital and labor) and GDP. The mechanism can be stated as follows: ﬁrms that charge a markup have a disconnect between the marginal cost and the marginal revenue product of their inputs (which are partly imported). The difference between these two is accounted as value added in the form of

Table. 11.The role of sectoral composition as an ampliﬁer of intermediate input's role

						corr GDP
	All	OECD	HH	HL		LL	All	OECD	HH	HL	LL
ln(inter)	0.012	0.044	0.007	0.012		0.007	0.059	1.025	0.247	0.082	0.025
	(0.008) (0.062)	(0.019)	(0.008)		(0.017) (0.081)	(0.545)	(0.255)	(0.088)	(0.160)
ln(inter)*exportprox0.155	0.227	0.266	0.131 0.147
	(0.044)	(0.176)	(0.070)	(0.052)	(0.073)
ln(inter)*sectorprox							0.085	1.324	0.210	0.109	0.082
							(0.097)	(0.614)	(0.294)	(0.106) (0.188)
ln(ﬁnal)	0.023	0.026	0.029	0.020	0.035	0.006	1.705	0.143	0.053	0.035
	(0.007)	(0.051)	(0.017)	(0.008)		(0.014) (0.082)	(0.578)	(0.314)	(0.088)	(0.174)
ln(ﬁnal)*exportprox	0.084	0.094	0.229	0.037		0.112
	(0.042)	(0.161)	(0.066)	(0.048)	(0.073)
ln(ﬁnal)*sectorprox							0.010	1.998	0.355	0.082	0.027
							(0.102)	(0.650)	(0.347)	(0.110)	(0.209)
exportprox	0.135	0.462	0.775	0.489		0.074
	(0.249)	(0.765)	(0.487)	(0.287)	(0.566)
sectorprox							0.979	8.253	4.987	2.124	0.856
							(0.953)	(2.859)	(2.936)	(1.021) (1.723)
CP + TW FE	Yes	Yes	Yes	Yes		Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes		Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	1,224	2,541	10,538		2,745	4,499	655	893	3,606	1,091
R2	0.015	0.105	0.054	0.011		0.012	0.025	0.194	0.180	0.020	0.021
Notes:								p<0.1; p<0.05; p<0.01.

proﬁts. Therefore, any change in input usage leading to a change in proﬁts triggers a change in value added. Since higher markups are associated with higher proﬁt shares in GDP (ceteris paribus), the higher the market power, the larger the association between foreign shocks and domestic GDP movements.

Since 1980, market powers have been increased substantially. For instance, Diez et al.(2018) and De Loecker and Eeckhout(2018) have shown that market powers have considerably increased over time in almost all developed countries. Figure 3shows that markups have increased over time in Europe, North America, Asia and Oceania, while they seem to have been more subdued in Africa and South America. This evolution suggests that countries that experienced higher markup increases over time should also have experienced surges in the sensitivity of their GDP to foreign shocks. Admittedly, this phenomenon could also contribute to the time evolution of the TC-slope uncovered in the previous section, though studying this mechanism is beyond the scope of this paper.

Figure 3.Global Market Powers. Source: ﬁgure 3 in De Loecker and Eeckhout(2018).

Robustness and additional exercises

This section provides additional results corroborating the ﬁndings that there is a signiﬁcant correlation between bilateral trade and the trade network with GDP-comovement. We ﬁrst investigate if the addition of ﬁnancial integration (FI), such as FDI and ﬂows of assets sig- niﬁcantly affects our results. We then separate intermediate inputs and capital trade ﬂows and then include cross network effects. Finally, we provide sensitive results with respect to additional controls, alternative measures, sets of countries, and time periods.

6.1 Financial Integration: role of FDI and ﬂows of assets

Previous studies found that ﬁnancial interconnection is signiﬁcantly (and negatively) associated with GDP comovements. Kalemli-Ozcanet al.(2013) identiﬁes a strong negative effect of banking integration on output synchronization, conditional on global shocks and country-pair heterogeneity. Such a result is consistent with a resource shifting hypothesis where an integration of capital market between two countries means that global savings are invested in the countries with the highest marginal productivity of capital - at the expense of investment in

the rest of the world.32

Bilateral data on ﬁnancial integration (FI) is scarce for pairs with two low income countries, but it is relatively widespread for other pairs. Hence, we focus our attention on the OECD and the HL groups for this exercise and account for the role of ﬁnancial ﬂows by using the consolidated banking statistics from the Bank for International Settlement and construct an index of ﬁnancial proximity (FP).33We use the total bilateral cross-border claims (including bank and non-bank sectors for all maturities) between countries i and j to construct an index

of ﬁnancial proximity (FP), such that FP	=	Ci!j,t +Cj!i,t	, where here C	i!j,t	refers to total
ijt		GDPit+GDPjt

cross-border claims from country i to country j in period t. Additionally, we control for FDI which might affect GDP co-movement independently of trade proximity.34We use up- to-date and systematic FDI data for 206 economies around the world from the UNCTAD's Bilateral FDI Statistics, covering inﬂows, outﬂows, inward stock and outward stock by region and economy.35We use the inﬂows and outﬂows in order to construct a bilateral ﬁnancial

integration (FI) controls, such that: FI	=	FDIi!j,t +FDIj!i,t	, where here FDI	i!j,t	refers to total
ijt		GDPit+GDPjt

FDI from country i to country j in period t. For this measure, we report only the effect of including this control for the whole sample.

The results of the benchmark speciﬁcation with additional controls are shown in table 12. The bilateral trade comovement slope and the trade network comovement slope are not affected by the inclusion of ﬁnancial variables, suggesting that the link between trade and GDP comovement remains unaffected by the inclusion of these controls.

6.2 Separating capital and intermediate inputs

We combined in our benchmark speciﬁcation trade in intermediate inputs and trade in capital goods. We now relax this assumption and test separately the effect of bilateral trade in capital goods and in intermediate inputs. Trade in capital goods account for only around 15% of total trade, and some country-pairs do not trade capital goods. We construct the index relative to

In other words, if savings can be allocated across borders, a positive technology shock in one country relative to its partners creates an inﬂow of capital into this country at the expense of other economies.
The dataset is available here:https://stats.bis.org/.
According toFontagné(1999), trade and FDI are positively correlated, which implies that failing to control for FDI is likely to bias our estimates of the relationship between trade and GDP correlation.
Data are in principle collected from national sources. In order to cover the entire world, where data are not available from national sources, data from partner countries (also called mirror data) as well as from other international organizations have also been used. Data can be downloaded on the UNCTAD website.

Table. 12.Effect of ﬁnancial integration

				corr GDP
	All	All	All	All	OECD	OECD	HL	HL
ln(inter)	0.169	0.164	0.006	0.006 0.405	0.402	0.013	0.015
	(0.078)	(0.078)	(0.018)	(0.018)	(0.093)	(0.094)	(0.023)	(0.023)
ln(ﬁnal)	0.091	0.088	0.008	0.007	0.059	0.054	0.023	0.021
	(0.054)	(0.054)	(0.015)	(0.015)	(0.046)	(0.047)	(0.018)	(0.018)
network1st	0.802	0.832	0.855	0.855	0.558	0.573 0.941	0.942
	(0.634)	(0.636)	(0.189)	(0.189)	(0.807)	(0.804)	(0.218)	(0.218)
network2nd	1.620	1.650	0.112	0.111	0.616	0.681	0.404	0.378
	(1.735)	(1.731)	(0.568)	(0.569)	(1.936)	(1.876)	(0.650)	(0.649)
BIS index		23.453
		(27.939)
FDI index				0.531		6.283		15.823
				(3.573)		(5.380)		(6.547)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	846	846	2,947	2,947	492	492	2,030	2,030
R2	0.098	0.099	0.026	0.026	0.229	0.233	0.021	0.024

Notes: p<0.1; p<0.05; p<0.01. Note that column 1 shows the results using the sub-sample containing data on the BIS index while column 3 shows the result using only country-pairs with data on the FDI index.

capital goods Tradeijtcapitalas we did in section 3. We then test the following speciﬁcation:

Corr GDPijt= b1ln(Tradeinterijt) + b2ln(Tradeijtcapital) + b3ln(Tradeijtf inal)

+ gnetworkijt+ Xijt+ CPij+ TWt+ eijt

(35)

The results are gathered in table 13, where we also show the benchmark speciﬁcation without separating trade ﬂows. Overall, our previous results are still valid except that now intermediate inputs seem to also play a role for HL pairs. Trade in capital goods is, however, not signiﬁcant for the OECD and LL groups. Interestingly, trade in capital goods seems to be positively correlated with more GDP comovement for HH, while we ﬁnd the opposite for HL, suggesting again possible heterogeneity in the effect of bilateral trade ﬂows along the development stage for capital goods.

Table. 13.Trade Comovement slope with disaggregated trade index and capital ﬂows

					corr GDP
	All	All	OECD	OECD	HH	HH	HL	HL	LL	LL
ln(inter)	0.009		0.103		0.034		0.007		0.018
	(0.005)		(0.030)		(0.014)		(0.006)		(0.006)
ln(inter)		0.014		0.117		0.026		0.016		0.017
		(0.006)		(0.031)		(0.013)		(0.006)		(0.012)
ln(capital)		0.010		0.006		0.028		0.015		0.012
		(0.004)		(0.019)		(0.009)		(0.004)		(0.008)
ln(ﬁnal)	0.012	0.016	0.009	0.013	0.009	0.010	0.016	0.021	0.017 0.024
	(0.005)	(0.005)	(0.025)	(0.024)	(0.012)	(0.013)	(0.005)	(0.005)	(0.005) (0.011)
network1st	0.305	0.349	1.046 1.026 0.521 0.603	0.264	0.301	0.040	0.034
	(0.067)	(0.069)	(0.259)	(0.258)	(0.163)	(0.163)	(0.073)	(0.076)	(0.073) (0.159)
network2nd	0.416	0.490	2.906 2.971	0.170	0.119	0.518	0.613	1.076 1.199
	(0.221)	(0.226)	(0.938) (0.938)	(0.550) (0.556)	(0.243)	(0.249)	(0.243) (0.492)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	12,702	1,224	1,222	2,541	2,508	10,538	10,194	2,745	2,614
R2	0.009	0.011	0.099	0.103	0.040	0.047	0.005	0.009	0.008	0.009
Notes:							p<0.1; p<0.05; p<0.01.

6.3 Cross network effects

In the baseline speciﬁcation, we have estimated two different network effects: a ﬁrst order network effect and a second order network effect. We now test the implication of a third network index which aims to capture how a country i is similar in terms of trade partners to the partners of a country j. This measure is described in section 3, and table 14provides the results and show that the inclusion of such a measure does not change the estimated coefﬁ- cients regarding bilateral trade ﬂows and ﬁrst order network effects. However, it inﬂuences point estimates associated with the second order network measure, which is likely to be the case, as those two indexes measure trade network effects taking place further down in the trade network.

6.4 Other robustness exercises

Our results are robust to a number of other alternative speciﬁcations that we gather in table 15. We ﬁrst conﬁrm that among non-OECD pairs, trade in ﬁnal goods is signiﬁcantly associated with more GDP-comovement. When considering only the last two time windows, we ﬁnd

Table. 14.Trade Comovement Slope with cross network effects

					corr GDP
	All	All	OECD	OECD	HH	HH	HL	HL	LL	LL
ln(inter)	0.009	0.008	0.103 0.107	0.034	0.035	0.007	0.006	0.018	0.017
	(0.005)	(0.005) (0.030) (0.031)	(0.014)	(0.014)	(0.006)	(0.006)	(0.006) (0.011)
ln(ﬁnal)	0.012	0.012	0.009	0.009	0.009	0.009	0.016 0.016	0.017 0.017
	(0.005)	(0.005) (0.025) (0.025)	(0.012)	(0.012)	(0.005)	(0.005)	(0.005) (0.010)
network1st	0.305	0.294	1.046 1.084	0.521 0.523	0.264 0.246	0.040	0.035
	(0.067)	(0.067) (0.259) (0.264)	(0.163)	(0.165)	(0.073)	(0.074)	(0.073) (0.153)
network2nd	0.416	0.364	2.906 3.019	0.170	0.160	0.518	0.432	1.076 0.866
	(0.221)	(0.221) (0.938) (0.961)	(0.550)	(0.554)	(0.243)	(0.242)	(0.243) (0.504)
networkcross		0.166		0.330		0.056		0.245		0.343
		(0.115)		(0.409)		(0.258)		(0.126)		(0.279)
CP + TW FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	13,079	13,079	1,224	1,224	2,541	2,541	10,538	10,538	2,745	2,745
R2	0.009	0.009	0.099	0.099	0.040	0.040	0.005	0.006	0.008	0.009
Notes:							p<0.1; p<0.05; p<0.01.

consistent results except for sub-samples with the lowest number of observations (the LL and

groups), for which bilateral trade ﬂows turn out to be statistically insigniﬁcant. Network effects, however, are in line with previous results.

We then ﬁnd that under alternative sectoral proximity indexes, the results remain very similar when using ISIC export proximity or SITC 2-digit export proximity.

Finally, we conﬁrm the general pattern using three additional analysis: (i) an alternative measure of trade proximity built using the mean of log trade intensities instead of the log

of the mean trade intensities within time-windows, (ii) an alternative measure using the max

operator when computing the trade intensities: ln(Trade) = max Ti$j/GDPi, Ti$j/GDPj,

BK ﬁltered GDP instead of HP ﬁltered GDP,36(iv) First difference instead of HP-ﬁltered GDP. For the alternative bilateral trade measures using the max operator or the mean log, the results are very close to the benchmark speciﬁcation. With the alternative ﬁlters, all results are consistent except for the LL group for which the correlation between trade in intermediate inputs and GDP-comovement turns out to be positive and signiﬁcant in the case of BK-ﬁltered GDP.

36In this exercise, we keep ﬂuctuations between 32 and 200 quarters.

Table. 15.Sensitive analysis: Trade and GDP-comovement

		Estimated coefﬁcient

	ln(input)ln(ﬁnal)network1st	network2nd	Sample	Pairs\|Obs.
(i) Sample selection
Alternative Group	0.007	0.012**	0.25***	0.23	Non-OECD	4094\|11830
Period (1990:2009)	0.206***	-0.107*	1.38*	1.93	OECD	316\|616
Period (1990:2009)	0.025	0.018	0.98**	0.56	HH	754\|1434
Period (1990:2009)	0.044***	0.034***	0.49***	3.80***	HL	3598\|6582
Period (1990:2009)	0.027	0.004	0.64***	4.09***	LL	1140\|1972
(ii) Alternative controls
SITC 2-digit exportprox	0.097***	-0.007	0.96***	2.75***	OECD	320\|1224
SITC 2-digit exportprox	0.036***	-0.008	0.53***	-0.30	HH	764\|2533
SITC 2-digit exportprox	0.009	0.017***	0.27***	0.49**	HL	3650\|10521
SITC 2-digit exportprox	0.018*	0.017*	-0.01	1.08**	LL	1165\|2744
ISIC exportprox	0.103***	-0.010	1.02***	2.77***	OECD	320\|1224
ISIC exportprox	0.038***	-0.011	0.56***	-0.29	HH	764\|2533
ISIC exportprox	0.008	0.016***	0.28***	0.49**	HL	3650\|10521
ISIC exportprox	0.018	0.017*	-0.02	1.09**	LL	1165\|2744

(iii) Alternative Measures

mean ln(index)mean ln(index)mean ln(index)mean ln(index)

Max trade index Max trade index Max trade index Max trade index

BK ﬁlter

First difference First difference First difference First difference

	0.111***	-0.034	1.05***	2.68***	OECD	320\|1224
	0.037***	-0.016	0.58***	-0.28	HH	764\|2533
	0.007	0.016***	0.26***	0.45*	HL	3650\|10521
	0.017	0.016*	-0.04	1.03***	LL	1165\|2744
c	0.103***	-0.013	1.06***	2.92***	OECD	320\|1224
c	0.038***	-0.011	0.57***	-0.18	HH	764\|2533
c	0.009	0.018***	0.26***	0.47*	HL	3650\|10521
c	0.018	0.017*	-0.05	1.12**	LL	1165\|2744
	0.115***	-0.010	1.05***	3.24***	OECD	324\|1260
	0.047***	-0.017	0.57***	-0.10	HH	320\|1224
	0.010*	0.014***	0.26***	0.52**	HL	3650\|10521
	0.026**	0.015	-0.05	1.11***	LL	1165\|2744
	0.078***	0.004	0.67**	2.74***	OECD	320\|1224
	0.017	-0.025*	0.59***	-0.38	HH	764\|2533
	0.009	0.023**	0.25***	0.62**	HL	3650\|10521
	0.021**	0.020**	0.14	1.03**	LL	1165\|2744

Notes: p<0.1; p<0.05; p<0.01. In parenthesis: std. deviation.

We provide the results using EU and USSR dummies since adding those controls substantially
reduce the signiﬁcance of trade in ﬁnal goods.
bFDI and BIS are mostly available for OECD and HH groups. We report only the robustness with

those two groups.

cWe deﬁne max trade index as the measure using max Ti$j/GDPi, Ti$j/GDPj.

Conclusion

This paper takes a fresh look into an old question: what is the association between trade ﬂows and GDP comovement at business cycle frequencies? Guided by a simple theory, we provide novel evidence on the role of both bilateral and global trade ﬂows and emphasize the strong interaction arising between bilateral linkages and the global trade network, which implies that the previously studied Trade-Comovement slope should not be - and indeed is not - constant over time. Taking a closer look at different income groups, we also present new facts on the role of sectoral composition and on the type of trade that seems to be associated with GDP correlation.

Looking ahead, we believe the paper provides interesting scope for future research. Most notably, it becomes crucial to understand why the TC-slope seems to be mostly driven by trade in intermediate inputs in developed countries, while it is mostly driven by trade in ﬁnal goods in developing countries. Moreover, the TC-slope has signiﬁcantly increased over time. While the literature has documented the possible role of the global rise of markups, it seems important to investigate further the channels that could explain this pattern.

References

Barrot, J.-N. and Sauvagnat, J. (2016). Input Speciﬁcity and the Propagation of Idiosyncratic Shocks in Production Networks *. The Quarterly Journal of Economics, 131(3):1543-1592.

Basu, S. and Fernald, J. G. (2002). Aggregate productivity and aggregate technology. European Economic Review, 46(6):963-991.

Baxter, M. and Kouparitsas, M. A. (2005). Determinants of business cycle comovement: a robust analysis. Journal of Monetary Economics, 52(1):113-157.

Bems, R., Johnson, R. C., and Yi, K.-M. (2011). Vertical linkages and the collapse of global trade. American Economic Review, 101(3):308-12.

Burstein, A., Kurz, C., and Tesar, L. (2008). Trade, production sharing, and the international transmission of business cycles. Journal of Monetary Economics, 55(4):775-795.

Calderon, C., Chong, A., and Stein, E. (2007). Trade intensity and business cycle synchroniza- tion: Are developing countries any different? Journal of international Economics, 71(1):2-21.

Clark, T. and van Wincoop, E. (2001). Borders and business cycles. Journal of International Economics, 55(1):59-85.

Comin, D. and Gertler, M. (2006). Medium-term business cycles. American Economic Review, 96(3):523-551.

De Loecker, J. and Eeckhout, J. (2018). Global market power. Working Paper 24768, National Bureau of Economic Research.

de Soyres, F. and Gaillard, A. (2019). Value added and productivity linkages across countries. FRB International Finance Discussion Paper No. 1266.

Di Giovanni, J. and Levchenko, A. A. (2010). Putting the parts together: Trade, vertical link- ages, and business cycle comovement. American Economic Journal: Macroeconomics, 2(2):95- 124.

di Giovanni, J., Levchenko, A. A., and Mejean, I. (2016). The Micro Origins of International Business Cycle Comovement. NBER Working Papers 21885, National Bureau of Economic Research, Inc.

Diez, M. F., Leigh, M. D., and Tambunlertchai, S. (2018). Global market power and its macroeconomic implications. International Monetary Fund.

Duval, R., Li, N., Saraf, R., and Seneviratne, D. (2015). Value-added trade and business cycle synchronization. Journal of International Economics, pages 251-262.

Fontagné, L. (1999). Foreign Direct Investment and International Trade: Complements or Substitutes? OECD Science, Technology and Industry Working Papers 1999/3, OECD Publishing.

Frankel, J. A. and Rose, A. K. (1998). The Endogeneity of the Optimum Currency Area Criteria. Economic Journal, 108(449):1009-25.

Huo, Z., Levchenko, A. A., and Pandalai-Nayar, N. (2019). The Global Business Cycle: Measurement and Transmission. Working Papers 669, Research Seminar in International Eco- nomics, University of Michigan.

Imbs, J. (2004). Trade, ﬁnance, specialization, and synchronization. Review of Economics and Statistics, 86(3):723-734.

Inklaar, R., Jong-A-Pin, R., and De Haan, J. (2008). Trade and business cycle synchronization in oecd countries: A re-examination. European Economic Review, 52(4):646-666.

Johnson, R. C. (2014). Trade in intermediate inputs and business cycle comovement. American Economic Journal: Macroeconomics, 6(4):39-83.

Kalemli-Ozcan, S., Papaioannou, E., and Peydro, J.-L. (2013). Financial regulation, ﬁnan- cial globalization, and the synchronization of economic activity. The Journal of Finance, 68(3):1179-1228.

Kehoe, T. J. and Ruhl, K. J. (2008). Are shocks to the terms of trade shocks to productivity? Review of Economic Dynamics, 11(4):804 - 819.

Kose, M. A. and Yi, K.-M. (2006). Can the standard international business cycle model explain the relation between trade and comovement? Journal of International Economics, 68(2):267- 295.

Liao, W. and Santacreu, A. M. (2015). The trade comovement puzzle and the margins of international trade. Journal of International Economics, 96(2):266 - 288.

Ng, E. C. (2010). Production fragmentation and business-cycle comovement. Journal of International Economics, 82(1):1-14.

Attachments

Original document
Permalink

Disclaimer

Board of Governors of the Federal Reserve System published this content on 27 May 2020 and is solely responsible for the information contained therein. Distributed by Public, unedited and unaltered, on 27 May 2020 13:52:03 UTC

Stock Market News in real time

Board of Governors of Federal Reserve System : Global Trade and GDP Co-Movement

Headlines

Most Read News