Fiscal policy, government size and EMU business cycle synchronization
| Published date | 01 May 2020 |
| DOI | http://doi.org/10.1111/sjpe.12233 |
| Author | David Duffy,Ishmael Tingbani,Sabrina Bunyan,George Filis |
| Date | 01 May 2020 |
Scott J Polit Econ . 2020;67:201–222.
|
201
wileyonlinelibrary.com/journal/sjpe
|
The Global Fin ancial Crisis of 2007–2009 and European D ebt crisis since 2010 have revived the di scussion of the
suitabilit y of the EMU as a common currency area . Business cycle synchroniz ation is considered a pre‐requisite
for a well‐functio ning common currency are a, according to the Optimum C urrency Area theor y (Alesina & Barro,
Accepted: 9 September 2019
DOI: 10 .1111/sjpe.1 2233
ORIGINAL ARTICLE
cycle synchronization
1|1| George Filis2|2
This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which
permits us e and distributio n in any medium, provid ed the original wor k is properly cited , the use is non‐comme rcial and no
modifications or adaptations are made.
© 2019 The Authors . Scottish Journal of Political Economy published by John W iley & Sons Ltd on behalf o f Scottish Economic
Soci ety.
1Departme nt of Accounting, Fin ance and
Economics, U lster Business Sc hool, Ulster
University, Newtownabbey, UK
2Departm ent of Accounting, Fin ance
and Economics, Bournemouth Business
School, The E xecutive Busine ss
Centre, Bournemouth University,
Bournemouth, UK
George Filis , Department of Ac counting,
Finance and Economics, Bournemouth
Business Sch ool, The Executi ve Business
Centre, Bournemouth University,
Bournemouth, UK.
Email: gfilis@bournemouth.ac.uk
We provide new evidence on the ef fects of fiscal policy a nd
government size on pairw ise business cycle synchronizat ion
in EMU. A novel time‐varyin g framework is employed to es‐
timate business cycle s ynchronization and s ubsequently a
panel approach is us ed to establish the ro le of fiscal vari‐
ables in determining the pairwise synchronization observa‐
tions across time. Th e findings suggest similar ities in the size
of the public sector, yet diverge nce in fiscal policy s tance,
matter for the deter mination of business cycle synchron iza‐
tion. Hence, increa sed fiscal federalism in EMU will cont rib‐
ute to increased busine ss cycle synchronization. Our r esults
remain robust to dif ferent specifications and sub‐pe riods.
business cycle s ynchronization, EU bus iness cycles, fiscal p olicy,
time‐varying correlation
C32; C33; E32; E62; O52; F44
202
|
BUNYAN et Al.
2002). Kappl er and Sachs (2013) mainta in that without a cer tain level of synchro nicity “a common monet ary policy
may not satisfy t he needs of all member count ries and may even contribute to c yclical divergence” (p. 1).
Hence, the level of s ynchronization is a matter of im portance to policymake rs, particularly in a comm on cur‐
rency zone. Mor eover, business cycle sy nchronization enables a more ef fective coordination of fisc al and mon‐
etary poli cies (Mundell, 1961). Business c ycle synchroniz ation may also impac t upon the long ru n viability of
monetary u nion, partic ularly in the pres ence of ‘decoupling’ of b usiness cycles , such as in the EMU, where d e‐
coupling bet ween the periphery co untries relative to the cor e EU countries is observed i n the post‐financial crisis
period (Ahm ed et al., 2018; Degiannakis, D uffy, & Filis, 2014).
A vast amount of res earch has focused on b usiness cycle syn chronization and it s determinants . Belke, Domnick,
and Gros (2017) provi de an extensive review of the li terature, along with the e arlier research by Degia nnakis et al.
(2014), Papageorgiou , Michaelides, and Milio s (2010), de Haan, Inklaar, and Jong‐A‐Pin ( 2008) and Altavill a (2004).
The aim of the pres ent study is not to present a thorou gh account of the existing findi ngs, but rather to identif y
relevant gaps in t he literature so to highlight t he contribution made. I n short, the literatu re related to the determi‐
nants of busines s cycle synchronization foc uses mainly on bilateral tra de, industrial specializ ation, monetary and
financial integration, distance between countries, political ideology and global economic shocks.1
Neverthele ss, according to Ka ppler and Sachs (2013 , p. 1), business cycle s ynchronizatio n is determined by
“the degree of symmetry between macroeconomic shocks, transmission channels and institutional features (in‐
cluding fisc al policy), as well as, the level of e conomic integration” bet ween countries. This cla im is rather import‐
ant as the fact t hat the level of synchron ization might be impac ted by fiscal policy d ecisions and other ins titutional
features, has b een rather neglected by the l iterature. There are only a han dful of studies focusing on th e poten‐
tial impact of f iscal policy on b usiness cycle sy nchronization (se e, for instance , Gächter, Gruber, & Riedl, 2 017;
Inklaar, Jong‐A‐Pin, & Haa n, 2008). Interest ingly, there is no consensus amo ng this limited number of s tudies as to
whether fiscal policy can increase business cycle synchronization.
Overall, the cu rrent strand in this line of res earch has neglected several i mportant aspects w hen considering the
impact of fisc al policy on business cycle sy nchronization. First, un like in the present research, prev ious studies have
not considered th e size of the government sec tor (by means of government ex penditure) along wit h discretionary f iscal
policy (proxie d by the cyclically adjust ed net lending) in order to ex plain business cycle syn chronization, with the on ly
exception being t he study by Camacho, Perez‐Q uiros, and Saiz (2006). This is ra ther important as under standing the
role of fiscal po licy and government size wil l help shape policy desi gn and implementatio n to support moneta ry union.
Second, we do not as sume an EU‐wide business cycle to es timate the level of synchronizat ion between an EU
aggregate business cycle and the individual countries’ business cycles. Rather, we consider bilateral synchronization
levels across count ry‐pairs, in a similar fa shion to Gächter et al. (2017) and Dar vas et al. (2005). This app roach over‐
comes the need to a ssume that a specific c ountry acts as a n “attractor” or t hat there is a force which dr ives a common
business cycle . It also means that we do not assu me the existence of any commo n European or world busin ess cycle.
Third, unlike Gäc hter et al. (2017) and Dar vas et al. (2005), we e mploy a robust tim e‐varying fra mework to
estimate the pairwise business cycle synchronization, which overcomes issues related to the use of rolling‐win‐
dow correlation s. For instance, re sults based on rolling‐window ap proaches are infl uenced by the choice of t he
window lengt h, whereas no such d ecision is require d using the time‐var ying framewo rk that we apply in thi s
study. Even more, rolling‐window correlation exhibits slow dynamics due its overlapping calculation (i.e. when one
observatio n is dropped at the start of th e window length, it is subsequ ently replaced by another obse rvation at
the end of this wind ow period.). This latter p oint is also responsible f or the observed autoc orrelation betwee n the
rolling‐window correlation figures at successive time points.
1 See, inter ali a, Montinari an d Stracca, (2016) ; Kappler and Sa chs (2013); Cerqu eira and Mart ins, (2009, 2011, (20 09, 2011); Kose et al., (20 08);
Inklaar et al ., (2008); de Ha an et al., (200 8); Calderon et a l., (2007); Kos e and Yi, (2006) ; Imbs, (2006); C amacho et al., ( 2006); Böwer an d
Guillemin eau, (2006); Ba xter and Koupa ritsas (200 5); Imbs, (200 4); Morgan et al ., (2004); Kose e t al., (2003a, ( 2003b); Kale mli‐Ozcan et al ., (2001);
Frankel and Ros e, (1998); Krugman, (1993 ); Canova and De llas, (1993).
To continue reading
Request your trial