Jalles, João To a
Wo king Pape
Es ima ing s uc u al budge balances in de eloping Asia
ADB Economics Wo king Pape Se ies, No. 719
P o ided in Coope a ion wi h:
Asian De elopmen Bank (ADB), Manila
Sugges ed Ci a ion: Jalles, João To a (2024) : Es ima ing s uc u al budge balances in de eloping
Asia, ADB Economics Wo king Pape Se ies, No. 719, Asian De elopmen Bank (ADB), Manila,
h ps://doi.o g/10.22617/WPS240087-2
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/298165
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
ESTIMATING STRUCTURAL
BUDGET BALANCES
IN DEVELOPING ASIA
João To a Jalles
ADB ECONOMICS
WORKING PAPER SERIES
NO. 719
Ap il 2024
Es ima ing S uc u al Budge Balances in De eloping Asia
This pape e iews he cu en discussions, me hods and p ac ices su ounding he es ima ion o easonable
p oxies o he unde lying iscal posi ion, a use ul ancho o iscal policy. An empi ical applica ion o
de eloping Asian economies is ca ied ou . The e is no one-size i s all ype o app oach and he sensi i i y
and disce nmen ega ding speci ic economies a e impo an in a ious s ages.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 68 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.
ASIAN DEVELOPMENT BANK
The ADB Economics Wo king Pape Se ies
p esen s esea ch in p og ess o elici commen s
and encou age deba e on de elopmen issues
in Asia and he Paci ic. The iews exp essed
a e hose o he au ho s and do no necessa ily
e lec he iews and policies o ADB o
i s Boa d o Go e no s o he go e nmen s
hey ep esen .
ADB Economics Wo king Pape Se ies
Es ima ing S uc u al Budge Balances in De eloping Asia
João To a Jalles
No. 719 | Ap il 2024
João To a Jalles ([email p o ec ed]) is a senio
associa e p o esso o economics a he Lisbon School
o Economics and Managemen (ISEG), Uni e si y o
Lisbon.
C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
© 2024 Asian De elopmen Bank
6 ADB A enue, Mandaluyong Ci y, 1550 Me o Manila, Philippines
Tel +63 2 8632 4444; Fax +63 2 8636 2444
www.adb.o g
Some igh s ese ed. Published in 2024.
ISSN 2313-6537 (p in ), 2313-6545 (elec onic)
Publica ion S ock No. WPS240087-2
DOI: h p://dx.doi.o g/10.22617/WPS240087-2
The iews exp essed in his publica ion a e hose o he au ho s and do no necessa ily e lec he iews and policies
o he Asian De elopmen Bank (ADB) o i s Boa d o Go e no s o he go e nmen s hey ep esen .
ADB does no gua an ee he accu acy o he da a included in his publica ion and accep s no esponsibili y o any
consequence o hei use. The men ion o speci ic companies o p oduc s o manu ac u e s does no imply ha hey
a e endo sed o ecommended by ADB in p e e ence o o he s o a simila na u e ha a e no men ioned.
By making any designa ion o o e e ence o a pa icula e i o y o geog aphic a ea, o by using he e m “coun y”
in his publica ion, ADB does no in end o make any judgmen s as o he legal o o he s a us o any e i o y o a ea.
This publica ion is a ailable unde he C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/. By using he con en o his publica ion, you ag ee o be bound
by he e ms o his license. Fo a ibu ion, ansla ions, adap a ions, and pe missions, please ead he p o isions
and e ms o use a h ps://www.adb.o g/ e ms-use#openaccess.
This CC license does no apply o non-ADB copy igh ma e ials in his publica ion. I he ma e ial is a ibu ed
oano he sou ce, please con ac he copy igh owne o publishe o ha sou ce o pe mission o ep oduce i .
ADB canno be held liable o any claims ha a ise as a esul o you use o he ma e ial.
Please con ac pubsma ke [email protected] g i you ha e ques ions o commen s wi h espec o con en , o i you wish
oob ain copy igh pe mission o you in ended use ha does no all wi hin hese e ms, o o pe mission o use
heADB logo.
Co igenda o ADB publica ions may be ound a h p://www.adb.o g/publica ions/co igenda.
ABSTRACT
This pape e iews he cu en discussions, me hods, and p ac ices su ounding he
es ima ion o easonable p oxies o he unde lying iscal posi ion, a use ul ancho o
iscal policy. An empi ical applica ion o de eloping Asian economies is ca ied ou . The e
is no one-size i s all ype o app oach and he sensi i i y and disce nmen ega ding
speci ic economies a e impo an in a ious s ages. The choice o he il e o decompose
ends om cycles ma e s. The way o adjus e enues and expendi u es en e ing he
cyclically adjus ed balance also ma e s—choices ega ding economy-speci ic s. panel
es ima ions o he use o s a ic s. ime- a ying app oaches need o be made. To deal
wi h one-o ope a ions, a na a i e-based app oach can complemen he sugges ed
iden i ica ion based on la ge changes in cyclically adjus ed go e nmen capi al ans e s.
A discussion o o he impo an ac o s ha can a ec he es ima es o s uc u al balances
such as asse and commodi y p ices is also p o ided.
Keywo ds: budge elas ici y, ime- a ying es ima ion, end-cycle decomposi ion,
cyclically adjus ed balances, one-o iscal ope a ions
JEL codes: C33, E62, H30, H60, O53
1. In oduc ion
Fo iscal su eillance and moni o ing, i is pa amoun o assess he s a e o public
inances and iscal policies, ha is, o unde s and he na u e o budge a y de elopmen s.
Ac ual budge balances a e inadequa e measu es o hese pu poses as hey a e a ec ed
by mul iple ac o s ha a e bo h empo a y and ou side he di ec con ol o he
go e nmen .
P ime amongs hese ac o s a e luc ua ions in economic ac i i y. The
unde s anding ha economic luc ua ions, which a e ansi o y in na u e, ha e on he
in e p e a ion and e alua ion o iscal de elopmen s suppo ed he no ion ha nominal
budge s could no be aken a ace alue since hey combine empo a y and pe manen
dis u bances (La ch and Tu ini 2009). Disen angling he wo elemen s led o he c ea ion
o he concep o he cyclically adjus ed budge balance (CAB), which came o li e mo e
han 30 yea s ago in a seminal con ibu ion by Blancha d e al. (1990).1 The CAB seeks
o co ec iscal ou comes o he in luence o cyclical mo emen s. In o he wo ds, hey
aim a de e mining wha iscal posi ion would p e ail i he economy ope a ed a i s ull
capaci y (ECB 2014).
In addi ion o he impac o he economic cycle, o he aspec s ha ha e become
p og essi ely impo an a e he one-o measu es. These a e empo a y go e nmen
measu es such as one-o e enues ( o ins ance, s emming om he sale o licenses o
elecommunica ion) o one-o capi al ans e s ( o ins ance, inancial aid o he banking
sec o ).
1 In ac , B own (1956) pionee ed he ex ac ion o he unde lying budge balance by compu ing he ull employmen
su plus, he p edecesso o he CAB. The concep ual ad an age o po en ial GDP o po en ial ou pu o e ull
employmen did no acili a e hings in p ac ical e ms. The compu a ion o po en ial ou pu s ill emains a specula i e
issue despi e many me hods and app oaches ha ing been de eloped (Sec ion 3.1).
2
S ipping he impac o sho - e m luc ua ions and one-o s yield wha some
in e na ional o ganiza ions e e o as “s uc u al budge balance” (SB) (In e na ional
Mone a y Fund [IMF]) o he “unde lying iscal posi ion” (Eu opean Commission [EC]).2
The a ac ion o he SB si s wi h i s goal o measu e, a low cos , he “ ue” iscal posi ion
co ec ed o ansi o y cyclical ac o s.
Howe e , es ima es o he s uc u al balance can be subjec ed o signi ican
measu emen s e o s as hey a e su ounded by conside able unce ain y. These e o s
a e ela ed o accu a ely es ima ing he ou pu gap—which is an unobse able a iable—
and he nonlinea eac ions o ax e enues o sha p changes in eal g oss domes ic
p oduc (GDP) g ow h. The de ini ion o iden i ica ion o empo a y ac o s can also cause
di icul ies. These p oblems complica e he applicabili y o his concep in economies ha
a e egula ly a ec ed by se e e shocks (e.g., wea he e en s); hose ha ha e a
p oduc ion s uc u e ha is no su icien ly di e si ied and, hence, ende he economy
hos age o commodi y cycles; o hose whose g ow h ou look is cha ac e ized by a lack
o some ancho ed p edic abili y. This epo aims o e iew he cu en discussions
su ounding he es ima ion o easonable p oxies o disc e iona y iscal s ance by
e iewing me hods and p ac ices. An empi ical applica ion o a selec ion o de eloping
membe economies o he Asian De elopmen Bank (ADB) is ca ied ou bu ull esul s
a e a ailable in he Appendixes.3 We can conclude om he empi ical exe cises ha he e
is no one-size i s all and desk sensibili y and judgmen is impo an in a ious deg ees.
The choice o he il e o decompose ends and cycles ma e s wi h a p e e ence o a oid
2 Fo he emainde o he documen , I will be e e ing o s uc u al balance as he measu e s ipped om bo h cyclical
a ia ions and one-o measu es.
3 The appendixes a e a ailable a h ps://dx.doi.o g/10.22617/WPS240087-2.
3
he ou da ed and p oblema ic HP il e in a o o he Hamil on (2018) app oach.
Mo eo e , he ake on how o adjus e enues and expendi u e en e ing he CB o SB is
pa amoun . The e is a ple ho a o choices o be made, om p io empi ical es ima ion a
he economy-speci ic le el, o whe he such analysis should be done in a s a ic o ime-
a ying ashion, o which subcomponen s o e enues and expendi u es a e likely o
luc ua e wi h he cycle o be o hogonal o i . Fu he mo e, i one hinks abou one-o
ope a ions, hen a mo e na a i e-based app oach can complemen he ad hoc
iden i ica ion o la ge changes in cyclically adjus ed go e nmen capi al ans e s. Finally,
he e is he discussion o o he ac o s in addi ion o he cycle, such as asse and
commodi y p ices, which includes changes in housing p ices and e ms o ade
adjus men s ha can p o e o be di icul o be e lec ed in a measu e o he unde lying
iscal posi ion. Ul ima ely, SBs a e a use ul ancho o iscal policy, bu should no be
aken oo se iously o eal- ime policy making.
The emainde o he epo is o ganized as ollows. Sec ion 2 elabo a es on he
a ionale behind needing a measu e o cyclically adjus ed budge balances; Sec ion 3
p esen s he exis ing me hodologies o do so con as ing p os and cons o each one, he
so-called agg ega ed s. disagg ega ed app oaches; Sec ion 4 deals wi h one-o iscal
ope a ions and he issues conce ning he passage om cyclically adjus ed balances o a
measu e o unde lying balance o s uc u al balance; Sec ion 5 pe o ms se e al empi ical
applica ions on ADB membe economies; and he las sec ion concludes.
4
2. Why Do We Need a Measu e o he Cyclically Adjus ed Budge Balance o
S uc u al Budge Balance?
Despi e he laws and p oblems, he CAB (and/o he SB) con inues o be one o he main
indica o s necessa y o iscal policy analysis. Mo e speci ically, i is an impo an concep
because i allows:
(i) spli ing he in luence o disc e iona y iscal policy o a gi en change in he de ici
om he e ec o he economic en i onmen while simul aneously allowing
au oma ic s abilize s o ope a e eely;
(ii) e alua ing iscal impulses; and
(iii) assessing i iscal policy in he long- un is sus ainable.
On (i), he discussion in he li e a u e su ounds he ole o iscal policy in
mac oeconomic s abiliza ion. Calcula ing he s uc u al balance means spli ing he iscal
s ance in o he cyclical and he s uc u al componen s. Mo e gene ally, iscal policy can
help he s abiliza ion o he business cycle by means o h ee di e en channels (Silgone ,
Rei schule , and C espo-Cua esma 2003). The i s e e s o he ole o he cyclical
componen , ha is, he au oma ic s abilize s. The e is, none heless, a second indi ec
channel ela ed o public spending i ems ha con inue ixed, independen o he phase o
he business cycle and by no eac ing o i also ha e a s abilizing unc ion. The hi d, is
he use o disc e iona y iscal policy (despi e mos li e a u e poin ing o a p o-cyclical bias
o hese policies and, hence, a des abilizing e ec ( an den Noo d 2000). Changes in he
s uc u al balance need policy ac ions and ansla e disc e iona y iscal policy changes.4
While he Global Financial C isis (and, mo e ecen ly, he co ona i us disease [COVID-
4 To Dos Reis e al. (2007) i is p e e able o e e o au oma ic s abilize s as “a passi e policy esponse o he cycle.”
11
Table 1 shows a sample o OECD economies’ agg ega e budge a y elas ici ies
pa ame e s e ie ed om La ch and Tu ini (2009).
Table 1: Budge a y Sensi i i y Pa ame e s
Economy Re enue Expendi u e Budge
Balance
Economy Re enue Expendi u e Budge
Balance
Belgium 0.47 -0.07 0.54 Hunga y 0.45 -0.01 0.46
Bulga ia 0.35 -0.01 0.36 Mal a 0.35 -0.01 0.36
Czech Republic 0.36 -0.01 0.37 The Ne he lands 0.39 -0.17 0.55
Denma k 0.50 -0.15 0.65 Aus ia 0.43 -0.04 0.47
Ge many 0.40 -0.11 0.51 Poland 0.33 -0.06 0.40
Es onia 0.29 -0.01 0.30 Po ugal 0.41 -0.04 0.45
G eece 0.42 -0.01 0.43 Romania 0.28 -0.02 0.30
Spain 0.38 -0.05 0.43 Slo enia 0.42 -0.05 0.47
F ance 0.44 -0.06 0.49 Slo ak Republic 0.27 -0.02 0.29
I eland 0.36 -0.05 0.40 Finland 0.41 -0.09 0.50
I aly 0.49 -0.02 0.50 Sweden 0.48 -0.10 0.58
Cyp us 0.39 -0.01 0.39 Uni ed Kingdom 0.40 -0.02 0.42
La ia 0.26 -0.02 0.28
Li huania 0.26 -0.01 0.27 Eu o a ea 0.42 -0.06 0.48
Luxembou g 0.48 -0.01 0.49 EU27 0.39 -0.04 0.43
Sou ce: La ch and Tu ini (2009).
The disagg ega ed app oach ine unes he o e all p e ious sequence o s eps by
u he decomposing he wo main budge a y i ems o e enues and expendi u es and
es ima ing indi idual e enue elas ici ies. When doing so, one needs o ake a s ance on
e enue o expendi u e i ems ha may no equi e cyclical adjus men . This means ha
only a subse o e enue componen s is adjus ed, hose ypically being pe sonal income
axes, co po a e income axes, indi ec axes o alue-added axes and social secu i y
con ibu ions. Fo expendi u e, some ely on cyclical adjus men o only cu en p ima y
expendi u es, he e o e, excluding in e es paymen s and capi al expendi u e).9 O he s
simply ocus on one i em o cu en p ima y expendi u es, he unemploymen bene i s.
9 In e es expenses may display cyclicali y in andem wi h iscal de ici s, esul ing in inc eased (dec eased) bo owing
needs when he e is a posi i e (nega i e) ou pu gap. This, in u n, can lead o luc ua ions in he in e es cos s incu ed.
Coun e cyclical shi s in in e es a es ha e he po en ial o o se he cyclical na u e o bo owing equi emen s,
po en ially esul ing in a minimal ne impac , i any, as sugges ed by Fa ing on e al. (2008).
12
Howe e , hey no mally accoun o a small p opo ion o o al o p ima y spending in mos
economies (i a all).
Ma hema ically, his e sion o he CAB can be exp essed as:
𝐶𝐴𝐵 = ∑𝑅
−𝐺
+𝑅 −𝐺 (8)
whe e 𝑅
is he cyclically adjus ed componen i o e enues, 𝐺
is he cyclically
adjus ed unemploymen - ela ed expendi u es, 𝑅 and 𝐺 deno e he non-cyclically
adjus ed e enue and expendi u e componen s espec i ely.
Fo cyclically adjus ed e enues, he new 𝜌, can be ob ained as:
𝜌, =∑𝜌,
(9)
whe e 𝜌, is he elas ici y o each e enue componen wi h espec o he ou pu gap
and
is he sha e o each e enue componen i in o al e enues.
Fo cyclically adjus ed expendi u es, he new 𝜌, can be ob ained as:
𝜌, = 𝜌,
(10)
whe e 𝜌, is he elas ici y o unemploymen bene i s wi h espec o he ou pu gap and
is he sha e o unemploymen bene i s in cu en p ima y expendi u es.
3.2 Po en ial Ou pu
In addi ion o being a measu e o agg ega e supply in he economy, po en ial
ou pu is equally an es ima e o end GDP. GDP’s long- e m end is no mally upwa d as
mo e esou ces—p ima ily labo and capi al—become a ailable and as echnological
change allows mo e-e icien use o hese esou ces (A nold 2009). Al e na i e measu es
o po en ial GDP we e o iginally concei ed o in luence decisions abou mone a y and
13
iscal policy. I an economy was ound o be below po en ial hen mone a y o iscal policy
could be used o accele a e he g ow h o ou pu wi hou isking oo much in la iona y
p essu e. The concep o po en ial ou pu is pe cei ed as a ool o assis policymake s
manage agg ega e demand and hence keep a s eady economic g ow h pa h. Di e en
de ini ions a e o en used by al e na i e in e na ional o ganiza ions (Box 1).
Box 1: Al e na i e De ini ions o Po en ial Ou pu
Eu opean Commission. “Po en ial ou pu cons i u es he bes composi e indica o o he
agg ega e supply side capaci y o an economy and o i s scope o sus ainable, non-in la iona y,
g ow h.... Po en ial g ow h cons i u es a summa y indica o o he economy’s capaci y o gene a e
sus ainable, non-in la iona y, g ow h.”a
OECD. “Po en ial g oss domes ic p oduc is de ined in he OECD’s Economic Ou look publica ion
as he le el o ou pu ha an economy can p oduce a a cons an in la ion a e. Al hough an
economy can empo a ily p oduce mo e han i s po en ial le el o ou pu , ha comes a he cos
o ising in la ion. Po en ial ou pu depends on he capi al s ock, he po en ial labou o ce (which
depends on demog aphic ac o s and on pa icipa ion a es), he non-accele a ing in la ion a e o
unemploymen (NAIRU), and he le el o labou e iciency.”b
ECB. “Po en ial ou pu is a key economic concep as i s e olu ion de e mines how as an
economy can g ow in a sus ainable way. I is ypically hough o as he highes le el o economic
ac i i y ha can be sus ained by means o he a ailable echnology and ac o s o p oduc ion, in
pa icula labou and capi al, wi hou c ea ing in la iona y p essu e.”c
ECB. “Po en ial ou pu is gene ally unde s ood o p o ide an indica ion o he medium- o long-
e m le el o sus ainable eal ou pu in he economy and i s a e o g ow h. I is also e e ed o as
he le el o ou pu which can be achie ed using a ailable p oduc ion ac o s wi hou c ea ing
in la iona y p essu es.”d
IMF. “Po en ial ou pu is he maximum amoun o goods and se ices an economy can u n ou
when i is mos e icien — ha is, a ull capaci y. O en, po en ial ou pu is e e ed o as he
p oduc ion capaci y o he economy.”e
ECB = Eu opean Cen al Bank, IMF = In e na ional Mone a y Fund, OECD = O ganisa ion o Economic Co-ope a ion
and De elopmen .
a Eu opean Commission. 2014. “The P oduc ion Func ion Me hodology o Calcula ing Po en ial G ow h Ra es & Ou pu
Gaps.” Economic Pape s 535. No embe .
b OECD S a . h ps://s a s.oecd.o g/ (accessed 2022).
c Ande sson, Malin, Bela Szö i, Má é Tó h, and Nico Zo ell. 2018. “Po en ial Ou pu in he Pos -C isis Pe iod.” ECB
Economic Bulle in Issue 7/2018.
e Jahan, Sa wa , and Ahmed Sabe Mahmud. 2013. “Wha is he Ou pu Gap?” Finance & De elopmen 50 (3). IMF.
Sou ce: Eu opean Pa liamen . 2020. “Po en ial Ou pu Es ima es and Thei Role in he EU Fiscal Policy Su eillance.”
14
The e a e mul iple app oaches o ob ain es ima es o end GDP. In spi e o
subs an ial p og ess in he econome ics o e he yea s, he e is s ill no a gene ally
accep ed me hod in he economics p o ession o calcula e po en ial ou pu (Bo io e al.
2013, 2014).10
To summa ize, a ious app oaches o es ima ing he ou pu gap exis , wi h some
based on s a is ical echniques and o he s on economic heo y-in o med models. Each
me hod has i s own se o s eng hs and weaknesses, as ou lined in CBO (2004).
Howe e , he e is a consensus ha es ima ing he ou pu gap in eal- ime poses
signi ican challenges, as no ed by O phanides and Van No den (2002). In he ollowing
discussion, we will unde ake a his o ical analysis ha looks backwa d in ime, p ima ily
ocusing on uni a ia e s a is ical me hods. I ’s impo an o no e ha s a is ical me hods
a e no en i ely de oid o heo e ical conside a ions, as highligh ed by Bassane i e al.
(2010) and Oksanen (2018). Uni a ia e s a is ical me hods, in pa icula , ely on he
undamen al heo e ical assump ion ha ac ual ou pu ends o luc ua e a ound i s
po en ial le el. In o he wo ds, hese me hods a e designed o ensu e ha ac ual ou pu ,
by cons uc ion, only empo a ily de ia es om po en ial ou pu .
Taking in o accoun he c i icisms di ec ed a he widely used Hod ick-P esco
(HP) il e , such as i s endency o iden i y alse cycles (Ha ey and Jaege 1993, Cogley
and Nason 1995), a mo e ecen il e ing echnique in oduced by Hamil on (2018) is
employed. Hamil on’s c i ique o he Hod ick and P esco il e e ol es a ound h ee
issues, namely spu ious cycles, end-o -sample bias, and a bi a y assump ions
conce ning he smoo hing pa ame e . As an al e na i e, Hamil on p oposed a eg ession
10 Fo a e iew o di e en me hods see Gibbs (1995), Gio no e al. (1995), Ladi ay e al. (2003), Ho n e al. (2007),
Bassane i e al. (2010), Ande on e al. (2014), and Alichi e al. (2017).
15
il e . Howe e , i is wo h no ing ha Schüle (2018) demons a es ha Hamil on’s
eg ession il e sha es some o he sho comings o he HP il e , indica ing ha i is no
a uni e sal solu ion. Schüle a gues ha he choice o he “co ec ” il e depends on he
esea che ’s speci ic objec i es, ha is, wha aspec s o he da a hey aim o emphasize.
To da e, only a limi ed numbe o s udies in addi ion o Schüle (2018) ha e empi ically
e alua ed Hamil on’s me hodology.11 Hod ick (2020) subsequen ly examined i
Hamil on’s (2018) al e na i e app oach was ac ually be e han he HP il e a ex ac ing
he cyclical componen . Hod ick (2020) ound ha o ime se ies in which he e a e
dis inc g ow h and cyclical componen s, he HP il e is be e a isola ing he cyclical
componen .
3.3 Budge a y Elas ici ies
To make i simple, many ely on Gi oua d and And e’s (2005) assump ions o ze o
expendi u e elas ici y and one e enue elas ici y. In ac , o his g oup o economies
elas ici ies compu ed o speci ic ax ca ego ies yield an agg ega e e enue elas ici y
close o 1 (Eu opean Commission 2005). Rega ding he ze o-elas ici y assump ion on
expendi u e (meaning ha he cyclically adjus ed expendi u e is equal o ac ual
expendi u e, 𝐺 = 𝑔), his is jus i ied economically because go e nmen expendi u e is
o en iewed as disc e iona y in i s en i e y, and hus independen om he business
cycle. Fedelino e al. (2009), o ins ance, assumed no adjus men on he expendi u e
side. While his may be a easonably good app oxima ion, in p ac ice, some expendi u e
11 In he con ex o New Zealand, Callaghan e al. (2018) employed bo h he HP il e and Hamil on's (2018) me hod o
calcula e labo o ce pa icipa ion gaps. D ehman and Ye man (2018) e alua ed he e ec i eness o a HP end e sus
Hamil on's linea p ojec ion o he c edi - o-GDP gap as an ea ly wa ning indica o o inancial c ises. Phillips and Shi
(2019) in oduced an enhanced HP il e and conduc ed an assessmen o i s pe o mance in compa ison o Hamil on's
app oach.
16
i ems, such as unemploymen bene i s, exhibi a cyclical pa e n. This me hod does no
sepa a e be ween he a ious e enue and expendi u e i ems. Howe e , he loss o
accu acy may be accep able in some cases acco ding o hese au ho s. As Sancak e al.
(2010) demons a e, in some economies, ax buoyancy ends o inc ease du ing
expansions and go down du ing ecessions. Consequen ly, in hese economies he
con ibu ion o m au oma ic s abilize s may be o e s a ed i one/ze o elas ici ies a e used
o es ima e au oma ic s abilize s.
Agains his backg ound, economy-speci ic elas ici ies o o e all e enue and
expendi u e should be used whene e possible, ei he by elying on p e-exis ing s udies
o by es ima ing hem in a eg ession se ing.
Wha Does This En ail?12
Conce ning e enue, he elas ici y o indi idual e enue ca ego ies can be b oken down
in o wo componen s. The i s is he ou pu elas ici y o ax e enue (𝜌,), which esul s
om mul iplying he elas ici y o ax e enues conce ning he co esponding ax base
(𝜌,),, by he elas ici y o he ax base in ela ion o he ou pu gap (𝜌,):
𝜌, = 𝜌,×𝜌, (11)
The i s s ep is o assume o de i e he alue o he ax elas ici y wi h espec o
i s base. De i a ion equi es, in addi ion o s a u o y ax a es, knowledge o he income
dis ibu ion. Bu in p ac ice, one migh ely on esul s o exis ing s udies.
The second s ep in ol es an econome ic p ocess whe e we es ima e how
sensi i e he ele an ax bases a e o changes in he ou pu gap. This es ima ion
12 This sec ion d aws hea ily om Bo nho s , Dob escu, Fedelino, Go schalk, and Naka a (2011).
17
necessi a es he de ini ion o mac oeconomic indica o s ha ac as p oxies o hese ax
bases. Fo ins ance, wage bills a e o en used as a p oxy o income axes and social
secu i y con ibu ions, co po a e p o i s a e employed as a measu e o he ax base o
co po a e income axes, and p i a e consump ion is he basis o indi ec axes. Once
hese wo elas ici ies a e de e mined, we can hen compu e he ax e enue elas ici ies in
esponse o changes in he ou pu gap.
Gi oua d and And e (2005) p esen a able (see Table 2 below) wi h ypical ax
elas ici ies. The elas ici ies o e enue componen s conce ning changes in he ou pu gap
a e usually g ea e han one o income axes, which is indica i e o he p og essi e na u e
o many income ax sys ems. They end o ho e a ound one o indi ec axes, e lec ing
he ypically uni o m indi ec (VAT) ax a es. In he case o social secu i y con ibu ions,
hese elas ici ies a e somewha lowe han one.
Table 2: Common Tax Elas ici ies
Tax
Ca ego y
Elas ici y o Tax
Re enue Rela i e
o i s
Base
Elas ici y o Base
Rela i e o Ou pu
Gap
Elas ici y o Tax
Re enue Rela i e
o
Ou pu Gap
Pe sonal income
ax
1.5–2.0 0.6–0.9 1.0–1.7
Co po a e income
ax
1.0 1.2–1.8 1.2–1.8
Social secu i y
con ibu ions
0.8–1.1 0.6–0.9 0.5–0.9
Indi ec axes
1
1
1
Sou ce: Gi oua d and And e (2005).
Rega ding expendi u es, i ’s possible o b eak down he elas ici ies o cu en
expendi u e ca ego ies in o wo componen s. Cu en spending, especially o i ems like
unemploymen bene i s, is mo e likely o exhibi cyclical beha io due o he na u e o he
bene i sys em. In con as , nominal spending on o he i ems such as public sec o wages
18
and consump ion o goods and se ices is likely o be la gely una ec ed by he business
cycle, equi ing no adjus men (Bo nho s e al. 2011). The ou pu elas ici y o
expendi u es (𝜌,) is hen de e mined by mul iplying he elas ici y o cu en
expendi u es (𝜌, ) conce ning i s base, which could be unemploymen , by he
elas ici y o ha base conce ning changes in he ou pu gap (𝜌,):
𝜌, = 𝜌, ×𝜌, (11)
Simila o e enues, he elas ici ies o expendi u e conce ning hei espec i e bases
can ei he be assumed o de i ed. In equa ion (11), i we a e conside ing a speci ic
expendi u e ca ego y, such as unemploymen bene i s alone, i may be easonable o
assume a p opo ional ela ionship wi h i s base, which in his case is unemploymen ,
esul ing in an elas ici y o 1. Consequen ly, he ou pu elas ici y o ha pa icula
expendi u e ca ego y is de e mined by he elas ici y o unemploymen conce ning
changes in he ou pu gap. In p e ious esea ch, due o a lack o addi ional da a, he
Okun’s me hod has been employed o his pu pose. The elas ici y o he unemploymen
a e in ela ion o changes in ou pu is essen ially he ecip ocal o he Okun coe icien .
Gio no e al. (1995) and Bou he illain e al. (2001) adjus ed expendi u e o accoun o
cyclical e ec s using an a e age elas ici y o -0.2.
19
Box 2: Es ima ing Budge a y Elas ici ies and he Okun’s Coe icien
All hese budge a y-ou pu elas ici ies can be es ima ed in a eg ession se ing.
Elas ici ies ep esen he pe cen age shi in one a iable, deno ed as X, in esponse o a
one-pe cen age-poin al e a ion in ano he a iable, Y. A commonly employed me hod o
calcula ing he elas ici y o a ime se ies, X, in ela ion o a p oxy o economic ac i i y,
deno ed as Y, in ol es es ima ing he ollowing equa ion:
∆log (𝑋) = 𝑎 +𝜌, ×∆log(𝑌)+𝜀
whe e ∆ is a i s di e ence ope a o . Mo e speci ically, he le -hand-side dependen
a iable should be exp essed in eal e ms ( ha is, e enues o expendi u es de la ed by
he GDP de la o ) and he igh -hand-side key eg esso should be eal GDP g ow h ( ha
is, he i s di e ence o he log o eal GDP).
The Okun’s Law assumes a nega i e ela ionship be ween cyclical luc ua ions in GDP
and he unemploymen a e. Shocks o he economy a e assumed o lead GDP o
luc ua e a ound i s po en ial; his, in u n, causes i ms o hi e and i e wo ke s, changing
he unemploymen a e in he opposi e di ec ion. The es ima ion o he Okun coe icien
is needed in he disagg ega ed app oach o he expendi u e side elas ici y. Ball e al.
(2019) sugges es ima ing a “changes” e sion o he o iginal “cyclical” Okun’s law, as a
ela ionship be ween changes in he unemploymen a e (u) and he g ow h a e o ou pu :
∆𝑢= 𝛼 +𝛽∆𝑌+𝜀
Ou coe icien o in e es is 𝛽 which gi es how much a one pe cen age poin inc ease in
ou pu g ow h ansla es in o a educ ion in he unemploymen a e. Fu ce i e al. (2020)
es ima e an a e age Okun coe icien in ad anced economies o -0.4 while o de eloping
economies i is equal o -0.2.
Fo he lis o Asian De elopmen Bank membe s, Appendix 2 shows he es ima ed
economy-speci ic elas ici ies o hose pai s o g ow h and unemploymen a e wi h a
leas 15 con inuous obse a ions. Coe icien s a e es ima ed by o dina y leas squa es
(OLS). Fo hose ha came ou s a is ically signi ican , hey a y be ween -0.02 in
Aze baijan and -0.19 in Fiji wi h an a e age o -0.12.
GDP = g oss domes ic p oduc .
Sou ce: Jalles (2019) and Fu ce i e al. (2020).
20
4. The Role o Asse and Commodi y Cycles
S anda d cyclical adjus men s can be accompanied wi h an adjus men —when jus i ied—
o signi ican changes in asse o commodi y p ices (Bo nho s e al. 2011):
Commodi y p ices o e ms o ade adjus men . In cases whe e go e nmen
budge s depend on income gene a ed by commodi y expo s o a e signi ican ly
in luenced by luc ua ions in he e ms o ade, i may be bene icial o make
adjus men s o ab up shi s in hese p ices in o de o e eal he ue iscal
si ua ion benea h.
Asse p ice adjus men . The iscal si ua ion a i s co e can also be in luenced by
asse p ices, pa icula ly hose o eal es a e and s ocks. As demons a ed by
Fa ing on e al. (2008), a sus ained 10% ise in asse and housing p ices esul ed
in an annual inc ease in cyclically adjus ed ax e enues anging om 0.1% o
0.4% o GDP.
How o add ess empi ically he abo e issues? As wi h he cyclical adjus men , bo h
agg ega e and disagg ega e app oaches can be used. In wha ollows, he o me will be
discussed and employed. Basically, he app oach consis s in adding a sepa a e e m o
he de ia ion o asse p ices o e ms o ade om hei benchma k le el, deno ed as
asse p ice gap (A∗/𝐴) o e ms o ade gap (T∗/𝑇):
𝑅, = 𝑟∗
, ∗
, (12a)
𝑅,, = 𝑟∗
, ∗
, ∗
, (12b)
whe e 𝑅, and 𝑅,, s and o e enues adjus ed o he ou pu and asse p ice gaps
and he ou pu , asse p ices, and e ms o ade gaps, espec i ely. I he elas ici ies o
e enues wi h espec o asse p ices o e ms o ade is ze o𝜌, = 0 o 𝜌, = 0, he
27
play a c ucial ole in iden i ying one-o s. Mo e echnically, o each economy, he e is a
le el o a end o ne capi al ans e s conside ed as “no mal” such ha la ge depa u es
o ne capi al ans e s om his benchma k p o ides a good p oxy o one-o s.
Summing up, om an ope a ion poin o iew, calcula ing s uc u al balances
in ol es a se o in e ela ed s eps (Figu e 1).
S ep 1: Iden i ying and emo ing one-o s.
S ep 2: E alua ing he e ec o he business cycle by elying on he agg ega ed o
disagg ega ed app oach.
S ep 3: Es ima ing he e ec s o o he cycles o ac o s, including hose associa ed
wi h asse p ices o commodi y p ices.
One can s ill ob ain he CAB by omi ing s ep 1.
28
Figu e 1: S eps o Ob aining S uc u al Budge Balances
Sou ce: Bo nho s e al. (2011).
6. Empi ical Applica ion: ADB Membe ship
6.1 Po en ial Ou pu
To ex ac he cyclical and end componen s o a nonspeci ic a iable 𝑥 (deno ed
𝑥 and 𝑥, espec i ely) whe e in ou case 𝑥= {𝐺𝐷𝑃}, we begin by employing he
equen ly used Hod ick-P esco il e . This il e minimizes he ollowing unc ion:
𝑚𝑖𝑛
{∑ (𝑥−𝑥)+𝜆∑[(𝑥−𝑥)−(𝑥
−𝑥)]
} (13)
whe e 𝜆 is he smoo hing pa ame e . The la ge he alue o 𝜆, he bigge is he penal y
on a ia ions o he end’s g ow h a e. Hod ick and P esco sugges 100 o 1,600 as a
C oss-economy
compa isons?
29
alue o 𝜆 o annual o qua e ly da a, espec i ely. Ra n and Uhlig (2002) a gue ha 𝜆
should equal 6.25 (,
) o annual da a and 129,600 (1,600 × 3) o mon hly da a.
Fo Hamil on (2018), we es ima e:
𝑥 = 𝛾+∑𝛾+𝑥 +𝑢
(14)
whe e 𝑥= 𝑥+𝑥. The non-s a iona y pa o he eg ession p o ides he cyclical
componen :
𝑥= 𝑢 (15)
while he end is gi en by
𝑥= 𝛾+∑𝛾+𝑥
(16)
Hamil on (2018) claimed ha h and k should be chosen such ha he esiduals
ob ained om equa ion (15) a e s a iona y and sugges ed ha , o b oade p ocesses,
he ou h di e ences o a se ies a e s a iona y. We chose h = 2 and k = 3. Hamil on’s
(2018) be e al e na i e is, in essence, o use he eg ession o a a iable a da e on he
ou mos ecen alues as o da e -h since his would achie e “… all he objec i es
sough by use s o he HP il e wi h none o i s d awbacks.” (p. 831). In pa icula , his
al e na i e “… can isola e a s a iona y componen om any I(4) se ies, p ese es he
unde lying dynamic ela ions, and consis en ly es ima es well-de ined popula ion
cha ac e is ics o a b oad class o possible da a-gene a ing p ocesses.” (pp. 839–840).
Equa ions (14), (15), and (16) a e es ima ed using o dina y leas squa es (OLS).
Figu e 2 shows o a selec ed se o ADB membe economies he cyclical
decomposi ion o eal GDP, ha is, he ou pu gap exp essed—as common in he
30
li e a u e—in pe cen o po en ial o end GDP.15 E idence seems o sugges ha he HP
is smoo he han he Hamil on- esul ing gaps, ha is, booms and bus s a e less
p onounced. In he case o he end-poin s, his is mo e ele an in he case o HP due o
i s known limi a ions and esul s would imply a mo e a o able ou pu gap han ha
ob ained using ins ead he Hamil on app oach.
Figu e 2: Compa ing Ou pu Gap be ween HP and Hamil on
in Selec ed Membe Economies
(% po en ial GDP)
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , HP = Hod ick-P esco .
No e: HP il e wi h lambda=100 and Hamil on il e -based decomposi ions. Ou pu gaps exp essed in
pe cen o espec i e po en ial GDP.
Sou ce: Au ho .
Figu e 3 plo s he po en ial GDP in na ional cu encies o he same se o selec ed
economies o a sho e pe iod ( he las 10 yea s) o isually be e inspec di e ences.
15 The eason behind he selec ion o hese ou economies ela es wi h ep esen a i eness: wo a e om Sou heas
Asia and wo a e om Sou h Asia. Also, size-wise hey do no ep esen small island s a es as da a quali y is ypically
sca ce and poo e and clima e- ela ed ulne abili ies makes hem unique.
-15
-10
-5
0
5
10
1990 1994 1998 2002 2006 2010 2014 2018
Yea
HP100 Hamil on
Philippines
-20
-10
0
10
20
1990 1994 1998 2002 2006 2010 2014 2018
Yea
HP100 Hamil on
Indonesia
-5
0
5
10
1990 1994 1998 2002 2006 2010 2014 2018
Yea
HP100 Hamil on
Pakis an
-2
-1
0
1
2
3
1990 1994 1998 2002 2006 2010 2014 2018
Yea
HP100 Hamil on
Bangladesh
31
A he end o he pe iod, he Hamil on-based po en ial GDP is always below ha coming
om he HP decomposi ion.
Figu e 3: Compa ing Po en ial be ween HP and Hamil on
in Selec ed Membe Economies
(local cu ency)
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , HP = Hod ick-P esco .
No e: HP il e wi h lambda=100 and Hamil on il e -based decomposi ions.
Sou ce: Au ho .
6.2 Budge Elas ici ies
6.2.1 Ou pu
The p ima y challenge in calcula ing measu es like he Cu en Accoun Balance (CAB),
Cyclically Adjus ed P ima y Balance (CAPB), o he S uc u al Balance (SB) lies in
de e mining he app op ia e iscal elas ici ies. A s aigh o wa d and simplis ic app oach
would be o adop Gi oua d and And é’s (2005) con en ion, which assigns an elas ici y o
one o e enues and an elas ici y o ze o o expendi u es, as p e iously discussed o he
easons p o ided.
12000
14000
16000
18000
20000
2011 2013 2015 2017 2019
Yea
eal GDP HP100
Hamil on
Philippines
7000000
8000000
9000000
10000000
11000000
2011 2013 2015 2017 2019
Yea
eal GDP HP100
Hamil on
Indonesia
9000
10000
11000
12000
13000
2011 2013 2015 2017 2019
Yea
eal GDP HP100
Hamil on
Pakis an
7000
8000
9000
10000
11000
12000
2011 2013 2015 2017 2019
Yea
eal GDP HP100
Hamil on
Bangladesh
32
Fo he ADB sample o membe s—Appendix 1 has he comple e lis —i is likely
ha he s eng h o he expendi u e side o he budge o eac o he economic cycle is
small. Speci ically, by examining IMF Go e nmen Financial S a is ics da a pe aining o
unemploymen - ela ed expenses as a pe cen age o GDP, we can disce n ha ac oss 17
economies wi h da a spanning he mos ecen 5-yea pe iod (2015-2019, in en ionally
excluding he yea s impac ed by COVID-19), his igu e anges om 0% (in se e al
economies) o 1.9% o GDP (in Thailand). On a e age, i s ands a 0.36%, in s a k
con as o he a e age cu en go e nmen expendi u e, which amoun s o 14.85% o
GDP. Figu e 4 shows he economy-speci ic alues. Addi ionally, one can es ima e he
agg ega e expendi u e elas ici y wi h espec o a p oxy o economic ac i i y as de ailed
in Box 2. Doing so o a sample o 38 ADB membe economies wi h a ailable da a yields
he coe icien s displayed in Figu e 5. Elas ici ies a y om 0.809 (in Myanma 16) o 2.422
(in Timo -Les e) wi h an a e age alue o 1.23. The majo i y o he membe s ha e an
elas ici y o ze o (27 ou o 38 economies). Agains his e idence, i seems e y
easonable o assume hen o his sample o economies, a leas as a s a ing poin o
ou exe cises, a budge a y spending elas ici y o ze o.
16 E ec i e 1 Feb ua y 2021, ADB placed a empo a y hold on so e eign p ojec disbu semen s and new con ac s in
Myanma .
33
Figu e 4: Unemploymen - ela ed Bene i s, Membe Economies wi h A ailable Da a:
A e age, 2015–2019
(% GDP)
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , IMF = In e na ional Mone a y Fund.
No e: ADB placed on hold i s egula assis ance in A ghanis an e ec i e 15 Augus 2021. E ec i e 1
Feb ua y 2021, ADB placed a empo a y hold on so e eign p ojec disbu semen s and new con ac s in
Myanma .
Sou ce: IMF Go e nmen Financial S a is ics.
Figu e 5. To al Expendi u e o Ou pu Elas ici y Coe icien s o Membe Economies
ADB = Asian De elopmen Bank
No es: OLS es ima ion o he change in eal o al expendi u e e enues agains eal GDP g ow h. Only
economies wi h a leas 15 obse a ions o eal o al go e nmen e enues we e conside ed. Economies
wi h no ba s deno e insigni ican coe icien es ima es ( eplaced by ze o). ADB placed on hold i s egula
assis ance in A ghanis an e ec i e 15 Augus 2021. E ec i e 1 Feb ua y 2021, ADB placed a empo a y
hold on so e eign p ojec disbu semen s and new con ac s in Myanma .
Sou ce: Au ho .
0
0.5
1
1.5
2
2.5
3
Bhu an
B unei Da ussalam
Cambodia
India
Indonesia
Republic o Ko ea
Malaysia
Maldi es
Philippines
Thailand
Vie Nam
Solomon Islands
Fiji
Ki iba i
Vanua u
Papua New Guinea
Samoa
Tonga
Tu alu
A menia
Kazakhs an
Ky gyz Republic
Tajikis an
People’s Republic o China
Tu kmenis an
Uzbekis an
Myanma
S i Lanka
Geo gia
Aze baijan
A ghanis an
Mongolia
Bangladesh
Pakis an
Ma shall Islands
Nepal
Timo -Les e
coun y-speci ic elas ici y a e age
economy
-
speci ic elas ici y
34
Now, ega ding he e enue side, one can es ima e he agg ega e e enue
elas ici y wi h espec o a p oxy o economic ac i i y as de ailed in Box 2. Doing so o a
sample o 38 ADB membe economies wi h a ailable da a yields he coe icien s
displayed in Figu e 6. Elas ici ies a y om 0.55 (in Papua New Guinea) o 2.5 (in
Kazakhs an) wi h an a e age alue o 1.6. The majo i y o he economies ha e an
elas ici y ela i ely close o 1 om abo e.
Figu e 6: To al Re enues o Ou pu Elas ici y Coe icien s o Membe Economies
ADB = Asian De elopmen Bank.
No es: OLS es ima ion o he change in eal o al go e nmen e enues agains eal GDP g ow h. Only
economies wi h a leas 15 obse a ions o eal o al go e nmen e enues we e conside ed. Economies
wi h no ba s deno e insigni ican coe icien es ima es ( eplaced by ze o). ADB placed on hold i s egula
assis ance in A ghanis an e ec i e 15 Augus 2021. E ec i e 1 Feb ua y 2021, ADB placed a empo a y
hold on so e eign p ojec disbu semen s and new con ac s in Myanma .
Sou ce: Au ho .
0.00
0.50
1.00
1.50
2.00
2.50
3.00
Papua New Guinea
Bangladesh
Bhu an
Maldi es
A menia
Malaysia
Uzbekis an
Philippines
Tu kmenis an
Timo -Les e
1. Republic o Ko ea
Aze baijan
Nepal
Myanma
Pakis an
Indonesia
Ky gyz Republic
Ma shall Islands
Singapo e
S i Lanka
India
Ki iba i
Vanua u
Fiji
Samoa
Tonga
A ghanis an
Tu alu
Thailand
Solomon Islands
Geo gia
Mongolia
Cambodia
Tajikis an
People’s Republic o China
Kazakhs an
coun y-speci ic elas ici y a e age
economy
-
speci ic elas ici y
35
6.2.2 Asse P ices and Te ms o T ade
Le us begin by plo ing he ime se ies o eal housing p ices ( om Bank o In e na ional
Se lemen s, BIS) and he eal e ms-o - ade-adjus men ( om Wo ld Bank, Wo ld
De elopmen Indica o s, WDI) o ou sample o ADB membe s o which hese se ies
ha e a leas 15 con inuous obse a ions. Figu es 7 and 8.1 plo he log o each se ies
oge he wi h he log o eal GDP in he seconda y igh hand side (RHS) axis o
compa ison.17 Looking a Figu e 7 o housing p ice dynamics, we obse e ha in some
economies, bo h pa e ns seem o go ela i ely hand-in-hand (e.g., Malaysia and
Singapo e) while in o he s, hey seem disjoin (e.g. Indonesia and Thailand).
Figu e 7: Real Housing P ices and Real GDP (logs) O e Time o Membe Economies
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , ln = na u al loga i hm.
Sou ce: Au ho .
17 Acco ding o he Wo ld Bank, he lis o commodi y expo e s ha a e pa o he ADB membe ship a e: B unei
Da ussalam, Myanma , Indonesia, Timo -Les e, Solomon Islands, Fiji, Papua New Guinea, A menia, Aze baijan,
Kazakhs an, Ky gyz Republic, Tajikis an, Tu kmenis an, Uzbekis an, Mongolia. Many o hese a e islands and o he s
a e economies o which he e is no a ailable da a o conduc a se ious analysis.
15.215.415.615.81616.2
5.5 6 6.5 7 7.5
10 10.5 11 11.5
1212.51313.51414.5
5.25.45.65.8 6 6.2
8 8.5 9 9.5
4.6 4.7 4.8 4.9
4 4.5 5
4.5 4.55 4.6 4.65 4.7
4.4 4.6 4.8 5
4.2 4.4 4.6 4.8
4.6 4.7 4.8 4.9
19801990200020102020 19801990200020102020 19801990200020102020
Indonesia Malaysia People’s Republic o China
Republic o Ko ea Singapo e Thailand
ln ealhousep ice ln gdp
yea
ln eal house p ice
ln eal
GDP
36
Figu e 8.1: Real Te ms-o -T ade Adjus men P ices and Real GDP (logs) O e
Time o Membe Economies
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , ln=na u al loga i hm,
TOT = e ms o ade.
Sou ce: Au ho .
Now le us es ima e join ly he elas ici y o e enue accoun ing addi ionally o
housing and e ms-o - ade. Resul s in Table 3 seem o sugges ha , while he coe icien
on eal GDP g ow h is posi i e and signi ican as one would expec om a e enue
elas ici y (e en i di e en om hose displayed in Figu e 6 as he unde lying speci ica ion
is no he same), he g ow h o eal housing p ices does no seem o ha e an e ec on
e enues (all coe icien s o he 6 economies o which we can es ima e wi h some
deg ee o s eng h a e s a is ically insigni ican ). In Table 4, we expand he speci ica ion
used in Table 3 o also include he eal e ms-o - ade-adjus men g ow h. He e, he
common se dic a es an e en smalle sample o only ou ADB membe economies. S ill,
in gene al, he ole o housing p ices emains o hogonal o e enues in hese economies
and only in Indonesia i seems ha e ms-o - ade ha e a ce ain ole o play. In pa icula ,
he coe icien is posi i e and signi ican , bu in e ms, he one om eal GDP g ow h
6.5 7 7.5 8
7.588.599.5
2.533.544.5
99.51010.511
1010.51111.512
1414.51515.516
8 8.5 9 9.5
4.24.44.64.855.2
7.588.599.5
88.599.510
1212.51313.51414.5
5 5.5 6 6.5
1.41.61.822.22.4
7.588.599.5
7.588.599.5
23.52424.52525.5
10152025
5 10 15
24262830
22242628
30313233
24262830
2122232425
2425262728
20222426
29303132
18202224
18.51919.520
2022242628
2022242628
19801990200020102020
19801990200020102020 19801990200020102020 19801990200020102020
A menia Bangladesh Bhu an Cambodia
India Indonesia Kazakhs an Ky gyz Republic
Pakis an Philippines Republic o Ko ea Singapo e
Solomon Islands S i Lanka Thailand
ln eal o adj ln gdp
yea
eal TOT
a
djus men
p
ices
ln eal
GDP
43
Fo comple eness, he cha s equi alen o Figu e 9.1 o all indi idual ADB
membe economies— o which i is possible o ob ain he CAPB—a e a ailable in
Appendix 3 Figu e A3.1. Unde lying da a is a ailable upon eques .
2
Tu ning o exe cise (2), we now use he es ima ed e enue elas ici ies om Sec ion 6.2.1
o ob ain he CAPB. The esul is shown in Figu e 10. As only Indonesia and he
Philippines ha e elas ici ies ha a e di e en ( o Pakis an and Bangladesh, he
assump ion o uni a y e enue elas ici y is applied), only hese wo economies a e shown.
Compa ed o Figu e 9.1 o he Philippines, he lines seem o be close o one ano he ,
while in he case o Indonesia, his is ue only o he pos -GFC pe iod.
Figu e 10: CAPB (% GDP) wi h Es ima ed Re enue Elas ici ies
in Selec ed ADB Membe Economies (Al e na i e Fil e s)
ADB = Asian De elopmen Bank, CAPB = cyclically adjus ed p ima y balance, GDP = g oss
domes ic p oduc , HP = Hod ick-P esco , OB = o e all balance.
Sou ce: Au ho .
3
Rega ding (3), ins ead o assuming a ze o-expendi u e elas ici y, we es ima ed hese
economy-by-economy simila ly o wha we did o e enues, again ecall Sec ion 6.2.1.
Resul s o hese elas ici ies we e shown in Figu e 5 whe e missing ba s deno e
-8
-6
-4
-2
0
2
1990 1994 1998 2002 2006 2010 2014 2018
Yea
OB HP
Hamil on
Philippines
-6
-4
-2
0
2
4
1990 1994 1998 2002 2006 2010 2014 2018
Yea
OB HP
Hamil on
Indonesia
44
s a is ically insigni ican coe icien s. As p e iously no ed, mos economies a e
cha ac e ized by he ze o-expendi u e elas ici y. Tha said, he new esul ing CAPB o
Bangladesh and Pakis an is shown in Figu e 11 (no e ha now he cases o Indonesia
and he Philippines a e he e e se o (2); hey ha e insigni ican expendi u e elas ici ies,
which in e ec would yield he esul shown in Figu e 10).
Figu e 11: CAPB (% GDP) wi h Bo h Re enue and Expendi u e Es ima ed Elas ici ies
in Selec ed ADB Membe Economies (Al e na i e Fil e s)
ADB = Asian De elopmen Bank, CAPB = cyclically adjus ed p ima y balance, GDP = g oss
domes ic p oduc , HP = Hod ick-P esco , PB = p ima y balance.
Sou ce: Au ho .
4
Finally, o (4), we now e u n o a ze o-expendi u e elas ici y bu now e enue elas ici ies
a e es ima ed in a ime- a ying ashion o each economy. Resul s o hese elas ici ies
a e shown in Figu e 12 whe e we add he 90% con idence bands. Fo he Philippines,
e enue elas ici y has been mos ly posi i e and signi ican o e he 1990–2021 ime
-6
-4
-2
0
2
4
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Philippines
-6
-4
-2
0
2
4
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Indonesia
-10
-5
0
5
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Pakis an
-10
-5
0
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Bangladesh
45
pe iod. This pa e n con as s wi h he o he h ee economies o which e enue
elas ici ies ha e been posi i e bu su ounded by la ge unce ain y (wi h mos o he
yea s wi h con idence bands abo e and below he ze o line, excep Indonesia in he las
decade). Hence, Indonesia, using a ime- a ying e sion o he elas ici y coe icien would
po en ially unde es ima e he ue his o ical cyclically adjus ed e enue, which was
posi i e and a ound 1 acco ding o Figu e 6. Fo Bangladesh and Pakis an s a ic and
ime- a ying e idence is consis en as Figu e 6 also shows an insigni ican coe icien
es ima e.
Fo comple eness, he cha s equi alen o Figu e 12 o all indi idual ADB membe
economies a e a ailable in Appendix 3 Figu e A3.2. Unde lying da a is a ailable upon
eques .
Figu e 12: Time-Va ying Re enue Elas ici ies
in Selec ed Membe Economies (Al e na i e Fil e s)
Philippines
Indonesia
Pakis an
Bangladesh
ADB = Asian De elopmen Bank.
Sou ce: Au ho .
0 5 10
1990 2000 2010 2020
Philippines
Coe icien o g ow h ub10
lb10
yea
-10 0 10 20
1990 2000 2010 2020
Coe icien o g ow h ub10
lb10
yea
-5 0 5 10
1990 2000 2010 2020
Coe icien o g ow h ub10
lb10
yea
-10 0 10 20 30
1990 2000 2010 2020
Coe icien o g ow h ub10
lb10
yea
G aphs by Coun y
46
Figu e 13: CAPB (% GDP) wi h Time- a ying Re enue Es ima ed Elas ici ies
in Selec ed Membe Economies (Al e na i e Fil e s)
ADB = Asian De elopmen Bank, CAPB = cyclically adjus ed p ima y balance, GDP = g oss
domes ic p oduc , HP = Hod ick-P esco , PB = p ima y balance.
Sou ce: Au ho .
To conclude his sec ion, we plo in Figu e 14 he compa ison o he Hamil on-
based CAPB agains he PB o each o he ou exe cises and he ou economies unde
sc u iny. We see ha he CAPB e sions ha ollow closes he p ima y balance a e hose
om exe cises (1) and (3). Exe cise (4) wi h he ime- a ying e enue elas ici ies is he
one gi ing he mos di e en ime p o ile and gene ally sugges ing a wo se iscal posi ion
( he blue line ends o be below he o he ones). S ill wi hin each economy he e is
dispe sion among he ou lines ha is la ge a he beginning (end) o he ime pe iod o
Bangladesh (Pakis an and he Philippines). Hence, no one size i s all can be applied.
-10
-5
0
5
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Philippines
-10
-5
0
5
10
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Indonesia
-30
-20
-10
0
10
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Pakis an
-15
-10
-5
0
5
1990 1994 1998 2002 2006 2010 2014 2018
Yea
PB HP
Hamil on
Bangladesh
47
Figu e 14: CAPB (% GDP) Compa ison ac oss he Fou Exe cises
in Selec ed Membe Economies (Hamil on Fil e )
Philippines
Indonesia
Pakis an
Bangladesh
ADB = Asian De elopmen Bank, CAPB = cyclically adjus ed p ima y balance, GDP = g oss domes ic
p oduc , PB = p ima y balance.
No e: “PB” deno es he p ima y balance (% GDP). “CAPB_ham_i”, o i=1,…,4, deno es he esul ing
Hamil on-based CAPB (% GDP) o exe cises 1–4 desc ibed ea lie in his sec ion.
Sou ce: Au ho .
6.4 Es ima ing One-O s
Decomposing ne capi al ans e s in o end and cycle poses i s p ac ical p oblems. Fi s ,
a ne a iable can ake bo h posi i e and nega i e alues ha ende s impossible he
applica ion o loga i hm, which is a p io condi ion, as commonly used in he li e a u e, o
apply a s a is ical il e . Tha said, since mos one-o s a e ope a ions ha inc ease ne
ans e s o imp o e he iscal s ance, we will ocus on c edi s alone. This a iable is
ans o med in o eal quan i ies using he GDP de la o , hen he loga i hm is compu ed
48
and applied o a gi en il e . Ou o he p e ious selec ion o economies, we need o pick
a new se as he ans e s gap in pe cen o GDP o he p e ious g oup is e y small
(below 0.1%, which is he alue used o conside one-o s). In wha ollows, we look a
A menia, Bhu an, he Ky gyz Republic, and Timo -Les e o mo e he e ogenei y.19 The
capi al ans e s (c edi ) gap (exp essed in % o GDP) is shown in Figu e 14 o hese ou
economies. Fo comple eness, he cha s equi alen o Figu e 15 o all indi idual ADB
membe economies o which he e is da a, a e a ailable in Appendix 3, Figu e A3.3.
Unde lying da a is a ailable upon eques .
Figu e 15: Go e nmen Capi al T ans e s-Gap (% GDP)
in Selec ed Membe Economies, Al e na i e Fil e s, 2000–2021
19 O he eligible economies ha ha e ans e s gap in pe cen o GDP la ge han 0.1 include mos ly islands in he
Paci ic: Ki iba i, Samoa, Solomon Islands, Tonga, Tu alu, and Vanua u.
.2
.4
.6
.8
T ans e s Gap (c edi ) (% GDP)
2000 2004 2008 2012 2016 2020
Yea
HP Hamil on
Timo -Les e
0
.2
.4
.6
.8
1
T ans e s Gap (c edi ) (% GDP)
2000 2004 2008 2012 2016 2020
Yea
HP Hamil on
A menia
Con inued on he nex page
49
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , HP = Hod ick-P esco .
Sou ce: Au ho .
Table 6 iden i ies he yea s o which he change in he ans e s (c edi ) gap is in
he economy’s op 90 h pe cen ile and a he same ime, ha gap is la ge han 0.1% o
GDP. Whe he o no hese yea s co espond o “ ue” one-o s, his is some hing o be
con i med by desk economis s and local au ho i ies using a mo e na a i e-based
app oach. Appendix 4 p o ides he ull lis o ADB membe economies wi h iden i ied one-
o yea s compu ed using ei he he HP o Hamil on app oaches.
Table 6: Yea s o De ici Reducing One-O s in Selec ed Membe Economies
Compu ed Using Cyclical Capi al T ans e s (C edi )
App oach
HP
Hamil on
Economy
Po en ial
O
ne
-
o
Yea s
Po en ial
O
ne
-
o
Yea s
A menia
2002, 2018
-
Bhu an
2010, 2013
-
Ky gyz Republic
2002
2002, 2003
Timo
-
Les e
2017, 2018
2009, 2011, 2018
ADB = Asian De elopmen Bank, GDP = g oss domes ic p oduc , HP = Hod ick-P esco .
No e: Yea s a he op 90 h pe cen ile o he dis ibu ion o he change in capi al ans e s gap and wi h
capi al ans e s gap la ge han 0.1% o GDP.
Sou ce: Au ho .
0
.05
.1
.15
T ans e s Gap (c edi ) (% GDP)
2000 2004 2008 2012 2016 2020
Yea
HP Hamil on
Bhu an
0
.5
1
T ans e s Gap (c edi ) (% GDP)
2000 2004 2008 2012 2016 2020
Yea
HP Hamil on
Ky gyz Republic
50
7. Conclusion
This epo aims o e iew he cu en discussions su ounding he es ima ion o
easonable p oxies o disc e iona y iscal s ance by e iewing me hods and p ac ices. An
empi ical applica ion o ADB’s membe s is ca ied ou bu ull esul s a e a ailable in he
Appendixes.
The main akeaway is ha he e is no one way o adjus iscal balances. We can
conclude om he empi ical exe cises ha he e is no one-size i s all and desk sensibili y
and judgmen is impo an in a ious deg ees. The choice o he il e o decompose ends
and cycles ma e s wi h a p e e ence o a oid he ou da ed and p oblema ic HP il e in
a o o he Hamil on (2018) app oach. Mo eo e , he ake on how o adjus e enues and
expendi u es en e ing he CAB o SB is pa amoun . The e is a ple ho a o choices o be
made, om p io empi ical es ima ion a he economy-speci ic le el, o whe he such
analysis should be done in a s a ic o ime- a ying ashion, o which subcomponen s o
e enues and expendi u es a e likely o luc ua e wi h he cycle o be o hogonal o i .
Fu he mo e, i one hinks abou one-o ope a ions, hen a mo e na a i e-based
app oach can complemen he ad hoc iden i ica ion o la ge changes in cyclically adjus ed
go e nmen capi al ans e s. Finally, he e is he discussion o o he ac o s in addi ion
o he cycle, such as asse and commodi y p ices, which includes changes in housing
p ices and e ms o ade adjus men s ha can p o e o be di icul o be e lec ed in a
measu e o he unde lying iscal posi ion.
Ul ima ely, I would say ha o he selec ed se o membe s looked a mo e closely,
asse p ices, commodi y p ices, and one-o s do no seem o be impo an . Mo eo e , as
he go e nmen size is ela i ely small and s able in mos economies, assuming a ze o
51
elas ici y o his i em seems app op ia e. As o e enue elas ici ies, whe e some ine
uning wi h economy-speci ic elas ici ies can be made, he imposi ion o a uni a y alue
o simplici y also seems app op ia e in mos con ex s. Bu i p io economy-speci ic
es ima ions o e enue elas ici ies a e o be conduc ed, I would a he go wi h a s a ic
es ima ion in he agg ega ed o disagg ega ed app oaches.
52
REFERENCES
Abiad, Abdul, Da ide Fu ce i, and Pe ia Topalo a. 2016. “The Mac o-economic E ec s
o Public In es men : E idence om Ad anced Economies.” Jou nal o
Mac oeconomics 50: 224–40.
Ada o , A., B. Clemen s, and J. Jalles. Fo hcoming. “G ow h E ec s o Public
In es men and Public Consump ion: The Role o he Business Cycle, Fiscal
Space and E iciency.” Wo ld Bank Wo king Pape .
A onso, An onio, and João To a Jalles. 2014. “Assessing Fiscal Episodes.” Economic
Modelling, 37: 255–70.
Alichi, Ali, Oli ie Bizimana, Sil ia Domi , Emilio Fe nandez Co ugedo, Douglas Lax on,
Kadi Tanye i, Hou Wang, and Fan Zhang. 2015. “Mul i a ia e Fil e Es ima ion o
Po en ial Ou pu o he Eu o A ea and he Uni ed S a es.” IMF Wo king Pape
No. 15/253. Washing on, DC: In e na ional Mone a y Fund (IMF).
Alichi, Ali, Oli ie Bizimana, Douglas Lax on, Kadi Tanye i, Hou Wang, Jiaxiong Yao,
and Fan Zhang. 2017. “Mul i a ia e Fil e Es ima ion o Po en ial Ou pu o he
Uni ed S a es.” IMF Wo king Pape No. 2017/106. Washing on, DC: In e na ional
Mone a y Fund (IMF).
Ande sson, Malin, Bela Szö i, Má é Tó h, and Nico Zo ell. 2018. “Po en ial Ou pu in
he Pos -C isis Pe iod.” ECB Economic Bulle in Issue 7/2018.
Ande on, Robe , Ted A anki, Alis ai Dieppe, Ca he ine Elding, S ephan Ha ou unian,
Pascal Jacquino , Vale ie Ja is, Vincen Labha d, Desisla a Rusino a, and Béla
Szö i. 2014. “Po en ial Ou pu om a Eu o A ea Pe spec i e.” Occasional Pape
Se ies No. 156. F ank u : Eu opean Cen al Bank (ECB).
A nold, Robe W. 2009. “The Challenges o Es ima ing Po en ial Ou pu in Real Time,”
Fede al Rese e Bank o S . Louis Re iew. July/Augus 2009.
Aue bach, Alan, and Daniel Feenbe g. 2000. “The Signi icance o Fede al Taxes as
Au oma ic S abilize s.” Jou nal o Economic Pe spec i es 14 (3): 37–56.
Aue bach, Alan, and Yu iy Go odnichenko. 2013a. “Fiscal Mul iplie s in Recession and
Expansion.” In Fiscal Policy A e he Financial C isis, edi ed by Albe o Alesina,
and F ancesco Gia azzi, 63–98. Chicago: Uni e si y o Chicago P ess/Na ional
Bu eau o Economic Resea ch (NBER).
_____. 2013b. “Measu ing he Ou pu Responses o Fiscal Policy.” Ame ican Economic
Jou nal: Economic Policy 4 (2): 1–27.
Ball, Lau ence, Da ide Fu ce i, Daniel Leigh, and P akash Loungani. 2019. ”Does One
Law Fi All? C oss-Coun y E idence on Okun’s Law.” Open Economies
Re iew 30: 841–74.
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
ESTIMATING STRUCTURAL
BUDGET BALANCES
IN DEVELOPING ASIA
João To a Jalles
ADB ECONOMICS
WORKING PAPER SERIES
NO. 719
Ap il 2024
Es ima ing S uc u al Budge Balances in De eloping Asia
This pape e iews he cu en discussions, me hods and p ac ices su ounding he es ima ion o easonable
p oxies o he unde lying iscal posi ion, a use ul ancho o iscal policy. An empi ical applica ion o
de eloping Asian economies is ca ied ou . The e is no one-size i s all ype o app oach and he sensi i i y
and disce nmen ega ding speci ic economies a e impo an in a ious s ages.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 68 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.
