Au ho : Manuel La a Molina
Su pe iso : Jesús Vázquez Pé ez
Mas e in Economics: Empi ical Analysis and Policies
Uni e si y o he Basque Coun y (UPV/EHU)
2021/2022
The Gibson pa adox
The swing o he in la ion dynamics
ac oss OECD coun ies
1
Acknowledgmen s
Thanks o Jesus Vázquez o guiding me du ing he elabo a ion o his Mas e hesis
and helping me o subs an ially imp o e i ; hanks o amily and iends o being he e
and gi ing uncondi ional suppo .
2
Index
1. In oduc ion
2. Empi ical E idence
2.1. The Uni ed S a es
2.2. Canada
2.3. The Uni ed Kingdom
2.4. Aus alia
2.5. F ance
2.6. The esu gence o he Gibson´s Pa adox
3. A DSGE model augmen ed wi h Money
4. Es ima ion Resul s
5. Model Pe o mance
6. Conclusion
7. Re e ences
Table and Figu es Index
Figu e 1: Time- a ying second-momen s a is ics o in la ion o he U.S.
Figu e 2: Time- a ying second-momen s a is ics o in la ion o Canada.
Figu e 3: Time a ying co ela ion be ween in la ion in Canada and FED und a es
Figu e 4: Time- a ying second-momen s a is ics o in la ion o he U.K.
Figu e 5: Time- a ying second-momen s a is ics o in la ion o Aus alia.
Figu e 6: Time- a ying second-momen s a is ics o in la ion o F ance.
Table 1: C ossed coun y co ela ion o ime- a ying co ela ion be ween in la ion and
in e es a es.
Table 2: P io and es ima ed pos e io s o he s uc u al pa ame e s.
Table 3: Selec ed second-momen s s a is ics.
3
Abs ac
In ecen business cycles, U.S. in la ion has expe ienced a educ ion in in la ion
ola ili y, in la ion pe sis ence and, a se e e weakening o he co ela ion wi h nominal
in e es (Gibson's pa adox). This p ojec concludes ha he empi ical e idence ound
in he U.S. ex ends o o he OECD coun ies such as Canada, Aus alia, he UK and
F ance cha ac e ized by ha ing an independen cen al bank. Fu he mo e, we
examine in la ion dynamics in he U.S. wi h a 4-equa ion DSGE model augmen ed wi h
money. Ou model quali a i ely ep oduces he swings in in la ion s a is ics, bu lacks
sou ces o nominal and eal igidi y o ully cap u e he Gibson's pa adox. In spi e o
hese model´s limi a ions, we ind changes in p ice s ickiness, he mone a y policy ule,
and he pe sis ence o in la iona y shocks as main explana o y ac o s o he Gibson
pa adox.
Keywo ds: in la ion dynamics, he Gibson’s pa adox, DSGE models.
1. In oduc ion
The Fishe Hypo hesis, one o he co ne s one o he neoclassical heo y, s a es ha
nominal in e es a es is closely linked wi h expec ed in la ion (implying a high
con empo aneous co ela ion be ween nominal in e es a es and in la ion). Howe e ,
since oughly a ound 1995, U.S in la ion has expe ienced a weakening in he
co ela ion wi h he nominal in e es a e (Casa es & Vázquez, 2018) (Cogley, Sa gen ,
& Su ico, 2012), his phenomenon is called he Gibson Pa adox and many enowned
economis ha e s udied his issue.
Keynes (1930) coined he e m "Gibson's pa adox" a e he economis Al ed He be
Gibson who was he i s economis o obse e he weak ela ionship be ween
nominal in e es a es and in la ion in a ew his o ical episodes. I was called a pa adox
since no exis ing heo y could explain he eason o his weak ela ionship. Keynes
(1930) de ec ed Gibson's pa adox du ing 1880-1924 ( he gold s anda d pe iod).
F iedman and Schwa z (1982) also concluded ha he Gibson's pa adox clea ly held
du ing he gold s anda d pe iod, bu since he 1960s he co ela ion be ween nominal
in e es a es and in la ion has inc eased. Ba sky and Summe s (1988) showed ha he
4
Gibson's pa adox had anished by he ea ly 1970s and concluded ha his pa adox
migh jus be a gold s anda d phenomenon.
Howe e , i we ocus on he empi ical e idence om he US, Cogley, Sa gen and
Su ico (2012), using an au o eg essi e ec o model wi h d i pa ame e s and
s ochas ic ola ili y, show ha he Gibson's pa adox has e u ned in he las business
cycles and he e o e , his weak co ela ion be ween in la ion and in e es a es may
also be obse able unde a ia mone a y egime. Casa es and Vazquez (2018) ob ain
simila esul s using a 20-yea olling-window o 4- a iables VAR wi h 4 lags and show
ha US in la ion has expe ienced a educ ion in ola ili y and a se e e weakening o
he co ela ion wi h he nominal in e es a e since 1995.
Wha does he li e a u e e iew say abou he sou ces behind o he esu gence o he
Gibson Pa adox? Ba sky (1987) s a ed ha he deg ee o he in la ion pe sis ence is a
key ac o o explain i , he Gibson pa adox appea s in pe iods whe e he in la ion was
weakly pe sis en . Casa es and Vázquez (2018), by es ima ing a DSGE model, conclude
ha he main sou ces behind he appea ance o he Gibson pa adox a ound he mid-
1990s in he U.S a e a la e New Keynesian Phillips Cu e (induced by highe p ice
s ickiness) and a lowe pe sis ence o he wage ma k-up. Cogley, Sa gen and Su ico
(2012) come o he conclusion ha in he las business cycles since he 1990s he e is a
mo e an i-in la iona y mone a y policy ule and a dec ease in he indexa ion o
nominal p ices o pas in la ion.
The main objec i e o his Mas e 's hesis is o s udy whe he he empi ical e idence
on in la ion dynamics obse ed in he Uni ed S a es, mainly he weak co ela ion
be ween in la ion and he nominal in e es a e since he mid-1990s , is also obse ed
in o he OECD coun ies.
We ocus ou s udy on a ew OECD coun ies because each o he coun ies s udied a e
supposed o ha e an independen Cen al Bank moni o ing i s mone a y policy ha is
no in luenced by he p essu es o poli icians who wan o pu sue sho e m policies
o elec o al gain, despi e hei possible u u e losses. Mo eo e , OECD membe s a e
de eloped coun ies wi h a simila le el o pe capi a income and hei economy a e
in e connec ed in oday's globalized economy and he e o e, economic cycles occu in
hese coun ies in simila ime pe iods ( o hese easons could be expec ed o obse e
5
simila swings in hese coun ies). Speci ically, we s udy ou coun ies Anglo-Saxon
oo s: he Uni ed S a es, he Uni ed Kingdom, Canada and Aus alia, and an Eu opean
coun y whose cu ency is he Eu o: F ance.
In o de o conclude whe he he esu gence o he Gibson pa adox is a he obus o
no in he coun ies s udied, we analyse he ime- a ying s a is ics o he second
momen s o in la ion o he coun ies men ioned abo e (mo e p ecisely, we s udy
he co ela ion be ween in la ion and he nominal in e es a es, he pe sis ence o
in la ion measu ed by i s i s -o de au oco ela ion, and he in la ion ola ili y
measu ed by i s s anda d de ia ion .
Once we ha e analysed he empi ical e idence in hese coun ies, we es ima e a 4-
equa ion DSGE model augmen ed wi h money o he U.S.. The pu pose is o assess
whe he a small-scale New Keynesian model ea u ing ewe sou ces o igidi y han
he model o Sme s and Wou e s (2007) can sa is ac o ily explain he second-momen
s a is ics obse ed in he U.S..
This mas e hesis is s uc u ed as ollows. Sec ion 2 shows he empi ical e idence
ound in he coun ies s udied. The ollowing sec ions ocus on he U.S: Sec ion 3
p esen s he DSGE model augmen ed wi h money. Sec ion 4 es ima es he small-scale
DSGE model wi h money using Bayesian app oach and discuss he empi ical esul s.
Sec ion 5 compa es he ac ual and simula ed second-momen s a is ics o assess
whe he ou model eplica es U.S. in la ion dynamics. Finally, Sec ion 6 concludes.
2. Empi ical E idence
In his sec ion, we will analyse whe he he swings in in la ion dynamics obse ed in
he Uni ed S a es, whe e he co ela ion be ween nominal in e es a es and in la ion,
in la ion pe sis ence and in la ion ola ili y ha e declined conside ably in ecen
business cycles (Casa es & Vázquez, 2018) (Cogley, Sa gen , & Su ico, 2012) a e also
obse ed in F ance, he Uni ed Kingdom, Canada, and Aus alia. Fo doing ha we will
es ima e a 20-yea (80 qua e s) olling-window 3- a iable un es ic ed VAR wi h 4
lags.
1
1
We use he me hod discussed in Hamil on (1994, page. 164-266).
6
Ou VAR conside s qua e ly da a om he i s qua e o 1970 o he ou h qua e
o 2019 o wo nominal a iables: he a e o in la ion (calcula ed om he GDP
de la o ), and he sho - e m in e bank in e es a es,
2
wi h he addi ion o a eal
a iable: he economic g ow h a e (calcula ed om he eal GDP). Mos o he se ies
ha we ha e used has been downloaded om he Fede al Rese e Bank o S . Louis,
whe eas he es has been ob ained om he OECD da abases.
F om he es ima ed olling-window VAR, we ob ain he ime- a ying second-momen
in la ion s a is ics o he coun ies s udied (in la ion ola ili y, in la ion pe sis ence
and co ela ion be ween in la ion and he sho - e m in e es a e). These s a is ics a e
showed in Figu es 1, 2, 4, 5 and 6. The ho izon al axes in each igu e indica es he i s
qua e o he co esponding olling-window (e.g. he in la ion s a is ics associa ed
wi h he i s qua e o 1970 a e calcula ed using a sample pe iod om he i s
qua e o 1970 o he ou h qua e o 1989). In he nex subsec ions, we analyse he
empi ical e idence o each coun y.
2.1. UNITED STATES
Figu e 1 shows he U.S. empi ical e idence unco e ed by Casa es and Vázquez (2018),
a educ ion in in la ion a iabili y is accompanied by a all in he co ela ion be ween
Fed unds a es and in la ion and a milde dec ease o he pe sis ence o he in la ion
in he mos ecen windows. This ac is in line as well wi h he esul s ob ained by
Cogley, Sa gen , and Su ico (2012), hey conclude ha since 1995 he e has been a
pe iod o low in la ion pe sis ence and a weak in la ion-nominal in e es a e
co ela ion.
2
These a es a e almos pe ec ly co ela ed wi h he sho - e m a es ypically moni o ed by cen al
banks.
7
Figu e 1: Time- a ying second-momen s a is ics o in la ion o he U.S.
Clea ly, he esu gence o he Gibson pa adox is a he obus in he U.S.. We nex
analyse whe he hese swings in in la ion dynamics occu in o he coun ies.
2.2. CANADA
Figu e 2 shows he empi ical e idence ound o Canada. The swings in in la ion
dynamics a e ai ly simila o hose in he U.S.: In ecen business cycles, he e is a all
in in la ion ola ili y which is ollowed by a weakening o he co ela ion be ween
in la ion and he nominal in e es a e, and a milde dec ease in in la ion pe sis ence.
I is no ewo hy ha he a iabili y o in la ion inc eases in he subsamples om he
1990s onwa ds, and his ac is no accompanied by a change in he co ela ion
be ween he nominal in e es a e and in la ion (i.e. he co ela ion coe icien being
close o 0). This empi ical e idence somewha challenges some iews in he ela ed
li e a u e discussed abo e es ablishing a link be ween low ola ile (s able) in la ion
and he Gibson pa adox.
8
Figu e 2: Time- a ying second-momen s a is ics o in la ion o Canada
The occu ence o Gibson's pa adox is qui e synch onized in he U.S. and Canada, as
he co ela ion o he U.S. and Canadian co ela ion coe icien s be ween in la ion and
he nominal in e es a e compu ed o all he window-subsamples is e y high (0.89)
(see Table 1). This comes as no su p ise since Canadian in e es a es and in la ion a e
highly in luenced by he U.S. mone a y policies. In pa icula , Figu e 3 shows he ime-
a ying co ela ion be ween Canadian in la ion and he Fed's nominal in e es a es
which displays simila dynamics swings as hose depic ed in Figu e 2.
3
3
To calcula e hese coe icien s, we use he same 3- a iable un es ic ed VAR wi h 4-lags es ima ion
ha we ha e used o he o he s coun ies.
15
𝜋𝑡=𝛽𝐸𝑡𝜋𝑡+1+(𝜎+𝜂)(1−𝜔)(1−𝛽𝜔
𝜔)𝑦𝑡+𝑧𝑡
(𝐴𝑆)
The cen al bank con ols he money supply h ough he in e es a e. i se s he
nominal in e es a e wi h he objec i e o s abilizing in la ion and he ou pu gap in a
sys ema ic way. To do so, i will use a Taylo - ype mone a y policy ule (MPR) (Taylo ,
1993). The cen al bank conside s he pas nominal in e es a e (𝑅𝑡−1) when se ing
he cu en in e es a e, hus main aining some pe sis ence in i s mone a y policies.
And inally, he mone a y ule inco po a es a AR(1) shock: 𝑣𝑡.
𝑅𝑡=𝜌𝑅𝑡−1+(1−𝜌)(𝜓𝜋𝜋𝑡+𝜓𝑦
𝑦𝑡)+𝑣𝑡
(𝑀𝑃𝑅)
We in oduce money in ou model as in Casa es and Vázquez (2018). The ole o
money is de ined as being he medium-o -exchange o ca y ou ansac ions. The
s ock o eal money can be used o sa e ansac ion cos s.
6
An inc ease in eal money
holdings educes ansac ion cos s, wi h diminishing ma ginal e u ns. The mic o-
ounded semi-log eal money demand equa ion is desc ibed by:
𝑚𝑡=(𝜆𝑚
𝛾)𝑚𝑡−1 +(1−𝜆𝑚
𝛾)𝑦
𝑡−(1−𝜆𝑚
𝛾)(1−𝑎2)
𝑅𝑠𝑠 𝑅𝑡−𝑎1(1−𝜆𝑚
𝛾)𝜀𝑡
(𝑀𝑜𝑛𝑒𝑦 𝐷𝑒𝑚𝑎𝑛𝑑)
whe e 𝑚𝑡 and 𝑦𝑡
7
a e he logs o eal money balance and ou pu wi h espec o hei
s eady s a e alues.
8
𝛾 is he s eady-s a e ou pu g ow h, and 𝑅𝑠𝑠 is he s eady-s a e
nominal in e es a e.
In sum, ou New Keynesian economy is ep esen ed by ou semi-loglinea equa ions:
agg ega e supply (AS), agg ega e demand (AD), mone a y policy ule (MPR) and,
Money Demand. They de e mine he dynamic beha iou o he ou endogenous
6
We use money as a ansac ion cos echnology ins ead o he way used by I eland (2003), whe e
money balances yield u ili y di ec ly. Ou me hod is mo e lexible o accommoda e he obse ed
in la ion dynamic shi s.
7
Taking in o accoun he de ini ion o ou pu gap, he log o ou pu is equal o he sum o ou pu gap
and he po en ial ou pu : 𝑦𝑡=𝑦𝑡+𝑦
𝑡 whe e po en ial ou pu is gi en by: 𝑦
𝑡= (1+𝜂
𝜂+𝜎)𝜒𝑡.
8
In ou model, o he sake o simplici y, he e is nei he go e nmen no in es men and, he e o e,
ou pu is equal o consump ion.
16
a iables 𝑦𝑡 (ou pu gap), 𝜋𝑡(in la ion), 𝑅𝑡( he nominal in e es a es) and 𝑚𝑡( eal
demand o money).
4. Es ima ion esul s
Ou DSGE model wi h money has been es ima ed o wo subsamples: 1979:3-1999:2
and 1988:3-2008:2. Ou decision o selec hese wo pe iods is because hey a e hose
in which, ollowing ou empi ical analysis in he second sec ion, we ind he highes
and he lowes co ela ion be ween he nominal in e es a e and in la ion (in he
subsample ha s a s in 1979:3, he co ela ion a e be ween in la ion and nominal
in e es a e is 0.753, while in he subsample ha s a s in 1988:2, he same indica o
is 0.05.
The obse ables se ies o ou model will be: 1. Money wi h Ze o Ma u i y (MZM) ( his
de ini ion o money ep esen mo e accu a ely he ole o money as medium o
exchange han o he de ini ions as he Mone a y Base (Casa es and Vázquez, 2018); 2.
The in la ion a e; 3. The ede al und a e; and 4. The ou pu gap. Ou es ima ion
ollows a wo-s ep Bayesian p ocedu e. We se he same p io dis ibu ion as in Sme s
and Wou e s (2007) o he wo subsample pe iods (we do no wan ou esul s being
a ec ed by assuming di e en p io ), which a e shown in he i s columns o Table 2.
In he able 2, we obse e he es ima ion esul s epo ing he mean pos e io alues
along wi h he 10% o 90% c edible se s o he pos e io dis ibu ion o he wo
subsamples analysed. In gene al, he es ima ed pa ame e s a e a he simila in bo h
subsamples. Howe e , we can obse e some ema kable di e ences in he es ima es
o a ew pa ame e s: P ice Cal o's p obabili y has sligh ly inc ease om 0.90 in he i s
pe iod o 0.95 in he second pe iod, implying ha he cons an p obabili y o ecei ing
he ma ke signal ha allows i ms o change p ice has dec eased o e he 1988-2008
pe iod. The e o e, he e is mo e p ice s ickiness and ha may help o explain he lowe
a iabili y o in la ion in his pe iod.
The ine ia pa ame e s in he policy ule (𝜌) has inc eased in he second pe iod, so ha
he cen al bank, when se ing in e es a es, conside s o a g ea e ex en pas
in e es a es and o a lesse ex en he ac ual economic ou look (ou pu gap and
17
in la ion). Mo eo e , we ind ha he money demand ine ia is much lowe in he
second pe iod.
Table 2: P io and es ima ed pos e io s o he s uc u al pa ame e s
P io s
Pos e io s
1979-1999
1988-2008
Log-likehood
-546.833
-521.042
Dis
Mean
S d D.
Mean
10%-90%
Mean
10%-90%
𝑤:
P ice Cal o
p obabili y
Be a
0.75
0.1
0.9043
(0.87-0.93)
0.947
0.82-0.97
𝜌: Ine ia (policy ule)
Be a
0.75
0.1
0.631
(0.55-0.70)
0.801
(0.75-0.84
𝜓𝜋: In la ion, policy
ule
No m
1.5
0.3
1.427
(1.15-1.68)
1.413
(0.99-1.82)
𝜓𝑦
: Ou pu gap, policy
ule
No m
0.125
0.1
0.361
(0.24-0.47)
0.383
(0.27-0.49)
𝜆𝑚:
Money demand
ine ia
Be a
0.7
0.2
0.965
(0.92-1)
0.741
(0.47-0.99)
𝑎1: T ansac ion cos
elas ici y
Be a
0.5
0.2
0.647
(0.38-0.92)
0.665
(0.41-0.92)
𝜌𝜒: Pe sis ence o
p oduc i i y shock
Be a
0.5
0.15
0.529
(0.39-0.66)
0.709
(0.60-0.80)
𝜌𝑧: Pe sis ence o
in la iona y shock
Be a
0.5
0.15
0.845
(0.78-0-89)
0.652
(0.53-0.77)
𝜌𝑣: Pe sis ence o
money demand shock
Be a
0.5
0.15
0.291
(0.14-0.43)
0.5723
(0.46-0.68)
𝜌𝜀: Pe sis ence o
policy shock
Be a
0.5
0.15
0.374
(0.23-0.51)
0.5212
(0.36-0.69)
𝜎𝜒: S d. o p oduc i i y
shock
In gamma
0.15
0.15
3.697
(3.09-4.26)
3.192
(2.63-3.71)
𝜎𝑧: S d. o in la iona y
shock
In gamma
0.15
0.15
0.151
(0.09-0.21)
0.222
(0.14-0.29)
𝜎𝑣: S d o money
demand shock
In gamma
0.15
0.15
0.731
(0.63-0.88)
0.344
(0.29-0.39)
𝜎𝜀: S d o policy shock
In gamma
0.15
0.15
4.301
(2.48-6.35)
7.522
(2.23-12.6)
Many pa ame e es ima es o he shock p ocesses ha e changed in hese wo pe iods:
he pe sis ence o in la iona y shocks is lowe in he second pe iod. Mo e p ecisely,
he pe sis ence o in la iona y shocks has dec eased om 0.84 o 0.65 in he second
pe iod (in pa icula , he oil shocks cha ac e izing he i s pe iod mos ly anish in he
second). Whe eas he pe sis ence o he p oduc i i y shock, he pe sis ence o he
18
money demand shock, and he pe sis ence o he policy shock ha e inc eased in he
second pe iod. Mo eo e , he s anda d de ia ion o some shocks has conside ably
changed in hese wo pe iods: a dec ease in he s anda d de ia ion o he money
demand shock, and an inc ease in he s anda d de ia ion o he policy shock.
5. Model pe o mance
In his sec ion, we e alua e whe he he swings in in la ion dynamics in he mos
ecen U.S. business cycles can be eplica ed by ou 4-equa ion DSGE model
augmen ed wi h money. Table 3 shows he second-momen s a is ics o in la ion o
U.S. da a and he es ima ed model o he subsample wi h he highes in la ion-nominal
in e es a e co ela ion (1979-1999) and o he subsample wi h he lowes in la ion-
nominal in e es a e co ela ion (1988-2008).
Table 3: Selec ed Second momen s s a is ics
S anda d
de ia ion
1979:3-1999:2
1988:3-2008:2
US da a
Model
US da a
Model
𝜎(𝜋𝑡)
2.2454
1.3503
0.8815
0.7908
𝜎(𝑅𝑡)
2.9646
1.6688
1.9318
1.0145
𝜎(𝑦𝑡)
3.0701
3.4805
2.07161
3.3234
Co ela ion
𝜌(𝜋𝑡,𝑅𝑡)
0.7535
0.6761
0.0466
0.3238
Au oco ela ion
𝜌(𝜋𝑡,𝜋𝑡−1)
0.858
0.8323
0.6326
0.6423
In able 3, We obse e ha he model cap u es he educ ion in a iabili y o he
obse ed a iables in he U.S.. In o he wo ds, i clea ly eplica es he dec ease in he
s anda d de ia ion o in la ion and he nominal in e es a es. Howe e , i alls sho in
eplica ing he high a iabili y o in la ion and in e es a es o he i s subsample. In
ega ds in la ion pe sis ence, ou model does a good job in eplica ing U.S. in la ion
19
pe sis ence. Cap u ing he educ ion in he i s o de co ela ion o in la ion om 0.85
(0.83) in he pe iod 1979-1999 o 0.63 (0.64) in he pe iod 1988-2008.
9
Finally, and mos impo an ly, we a e in e es ed in assessing whe he ou model is
capable o eplica ing he e-eme gence o Gibson's pa adox. We conclude ha ou
model can pa ially eplica e a educ ion in he co ela ion, al hough i alls sho since
he ac ual co ela ion coe icien o he 1988-1999 pe iod is 0.05 whe eas ou
es ima ion o ha pe iod is highe a 0.32.
O e all, ou DSGE model can pa ly eplica e he empi ical e idence obse ed in he
U.S. We hen miss some nominal and eal igidi ies, such as he ones conside ed by
Sme s and Wou e s (2007) and Casa es and Vázques (2018) in hei medium-scale
DSGE models o cap u e he s ong weakening in he co ela ion be ween in la ion and
he nominal in e es a e obse ed in ecen U.S. business cycles.
6. Conclusion
Fi s ly, we ha e analysed empi ical e idence om in la ion s a is ics in se e al OECD
coun ies: he Uni ed S a es, Canada, he Uni ed Kingdom, Aus alia and F ance. In all
hese coun ies we ha e obse ed a decline in in la ion a iabili y oge he wi h a
weakening o he in la ion-nominal in e es a e co ela ion and in la ion pe sis ence.
We can he e o e conclude ha he e u n o Gibson's pa adox (a weak
con empo aneous co ela ion be ween he nominal in e es a e and in la ion) is
a he obus in all he coun ies s udied. Howe e , we ha e ound some coun y-
dependen singula i ies in he second-momen s a is ics o in la ion.
Secondly, we ha e ied o assess he d i e s o ecen in la ion swings obse ed in he
U.S. h ough he es ima ion o a ou -equa ion DSGE model augmen ed wi h money.
Ou new Keynesian model augmen s he 3-equa ion ex book New Keynesian
mone a y model wi h demand o money ob ained om a acili a ing- ansac ion
echnology as in Casa es and Vázquez (2018).
Ou model cap u es he educ ion in in la ion a iabili y (al hough i ails o cap u e
he high in la ion ola ili y obse ed in he pe iod wi h high co ela ion be ween
in la ion and nominal in e es a e). Mo eo e , he model success ully ep oduces he
9
Be ween pa en hesis a e he i s o de au oco ela ion o in la ion es ima ed by ou model.
20
all in U.S. in la ion pe sis ence. Bu mos impo an ly, i alls sho in eplica ing
Gibson's pa adox. As shown in Casa es and Vázquez (2018), addi ional sou ces o eal
and nominal igidi ies, such as hose inco po a ed by Sme s and Wou e s (2007) a e
needed, in addi ion o including explici ly money in o he model, o success ully
ep oduce he e y weak co ela ion be ween in la ion and he nominal in e es a es
obse ed in he U.S. in ecen imes.
Due o i s simplici y, ou model does no p o ide a clea in e p e a ion o why Gibson's
pa adox eappea ed. Howe e , we can obse e ha in he pe iod wi h he weakes
co ela ion be ween nominal in e es and in la ion: p ice igidi y has sligh ly inc eased
( he i ms ecei es he signal o change he p ice less o en); mone a y policy will ake
in o accoun o a g ea e ex en he mone a y policies ca ied ou in he pas pe iod
and o a lesse ex en he in la ion and ou pu gap o he cu en pe iod, and he
pe sis ence o in la iona y shocks was weake . Thus, pe sis en oil shocks o he 1970´s
and ea ly 1980´s ha e disappea ed in mos ecen decades. Howe e , he cu en
Russia-Uk aine wa , esul ing in s ong ene gy p ices shocks, may lead o new swings in
in la ion dynamics as we a e al eady obse ing.
7. Re e ences
Ba sky, R. B. (1987). The Fishe hypo hesis and he o ecas abili y and pe sis ence o
in la ion. Jou nal o Mone a y Economics, 3-24.
Ba sky, R. B., & Summe s, L. H. (1988). Gibson´s Pa adox and he Gold S anda d.
Jou nal o Poli ical Economy, 96, 528-550.
Cal o, G. (1983). S agge ed P ices in a U ili y-Maximizing F amewo k. Jou nal o
Mone a y Economics 12, no.3, 383-398.
Casa es, M., & Vázquez, J. (2018). The swings o U.S. In la ion and he Gibson pa adox.
Economic Inqui y 56, 779-820.
Cogley, T., Sa gen , T., & Su ico, P. (2012). The Re u n o he Gibson Pa adox.
Unpublished manusc ip .
Dixi , A. K. (1977). Monopolis ic compe i ion and op imum p oduc di e si y. The
Ame ican economic e iew, 67(3), 297-308.
Fe nandez-Villa e de, J. (2010). The econome ics o DSGE models. SERIEs, 3-49.
21
F iedman, M. & Schwa z, A. J. (1982). Mone a y T ends in he Uni ed S a es and he
Uni ed Kingdom: Thei ela ion o Income, P ices and In e es Ra es, 1867-1975.
Chicago IL: Uni e si y o Chicago P ess.
Galí, J. (2015). A Classical Mone a y Model. In Mone a y Policy, In la ion and he
Business Cycl (p. chap e 2). P ince on Uni e si y P ess.
Hamil on, J. D. (1994). Time se ies Analysis. Pince on: P ince on Uni e si y P ess.
Keynes, J. M. (1930). A T ea ise o Money. London: Mcmillan.
Sme s, F. E. (2007). Shocks and F ic ions in US Business Cylces: A Bayesian DSGE
App oach. Ame ican Economic Re iew 97, 585-606.
Taylo , J. (1993). Disc e ion Ve sus Policy Rules in P ac ice. Ca negie-Roches e
Con e ence Se ies on Public Policy,39, 195-214.
Wood o d, M. (2003). In e es and P ices. P ince on Uni e si y P ess.
