Vol. 13 (May 2013), pp.17-31 © 2013 Central Bank of the Republic of Turkey https://www3.tcmb.gov.tr/cbr/
T
HEC
ASE FORF
OURP
ERCENTI
NFLATION Laurence M. Ball∗ABSTRACTMany central banks target an inflation rate near two percent. This essay argues that policymakers would do better to target four percent inflation. A four percent target would ease the constraints on monetary policy arising from the zero bound on interest rates, with the result that economic downturns would be less severe. This benefit would come at minimal cost, because four percent inflation does not harm an economy significantly.
JEL E52, E58, E31
Keywords Inflation, Monetary policy, Inflation target
ÖZBirçok merkez bankası yüzde ikiye yakın enflasyonu hedeflemektedir. Bu çalışma, politika yapıcıların yüzde dört enflasyonu hedeflemelerinin daha iyi olacağı görüşünü ileri sürmektedir. Yüzde dört enflasyon hedefi, para politikası üzerindeki faiz oranlarındaki sıfır sınırından kaynaklanan kısıtlamaları hafifletecek ve neticesinde iktisadi bunalımların şiddeti daha az olacaktır. Bu fayda, yüzde dört enflasyon ekonomiye önemli ölçüde zarar vermediği için minimum maliyetle elde edilecektir.
ENFLASYON HEDEFİNİN YÜZDE DÖRT OLMASI DURUMU JEL E52, E58, E31
Anahtar Kelimeler Enflasyon, Para politikası, Enflasyon hedefi
∗ Department of Economics, Johns Hopkins University, Mergenthaler Hall, 3400 North Charles Street,
1. Introdu A centra inflation, b choose a ta on average in monetary
No econ inflation ra inflation ta Reserve (w European C to, 2%”), an This essa clear what because the the past. If percent, the The prim problem, th nominal int the long-ru before the for a centra occurs.
In the U dampened zero bound rates away the econom rates during In contra nor eviden significantl A numbe notably Bla monetary e
uction al bank doe but it does c arget for the . What is th y economics nomic rese ate. Yet man arget near tw which calls Central Bank nd most oth ay argues th
target is id e United Sta f central ban e economic b mary reason he constrain terest rates un levels of zero bound al bank to r United Stat the Great R d. Looking
from zero d my would be
g past recess ast, the costs nce from re ly lower with er of econom anchard et a economists,
es not perfe control infla e inflation ra e optimal in s.
earch has c ny central ba wo percent. T
two-percen k (which aim er central ba hat a two per eal, but fou ates has live nks raised t benefits wou
to raise inf nt on mone cannot be n f nominal ra d becomes b restore full tes, a four Recession o
ahead, four during future e substantial,
sions.
s of four per esearch sugg
h four perce mists have s al (2010). Y and it is ana
ectly contro ation in the ate and keep nflation targe convincingly anks have c These centra nt inflation ms for infla anks in adva rcent inflatio ur percent is
d comfortab their inflatio uld exceed t flation targe etary policy negative. A h
ates, allowin binding. Thi
employmen percent in of 2008–09,
r percent in e recessions , based partl rcent inflatio gests that a ent inflation suggested a Yet the idea
athema to ce
l the short- long run. P p inflation c et? This is a y determin chosen a com
al banks inc a “longer- ation rates “ anced econom
on target is s a reasonab bly with that on targets fr the costs.
ets is to ease y arising fro higher inflat ng larger de is flexibility nt when an nflation targ
when inter nflation wou s. I argue tha ly on the beh on are small an economy
than with tw four percent
is widely u entral banke
-run behavio Policymakers lose to this a classic que ned the op mmon polic clude the Fe -run goal”) below, but mies.
too low. It i ble guess, in t inflation ra from two to e the zero-b om the fact tion target r ecreases in y makes it e economic s get would rest rates hi uld keep int at the benef havior of int l. Neither hi y’s efficienc wo percent.
t inflation ta unpopular am
ers. Accordin or of s can level estion timal y: an deral , the close is not n part
ate in four ound t that raises
rates easier lump have it the terest fits to terest story cy is arget, mong ng to
Bernanke opposes an greater cost If one opposition this essay research ha zero-bound to inflation digit inflati 2. The Ben
Economi above two target reduc 2.1. The Z
Many ce their instru unemploym demand. In rate substan 1994). The increases in
When a constraint:
constraint return of ze and receive because the The zero economy in rates, but ra sufficient unemploym it through f
Keynes w his Genera considered monetary p
(2010a), th n increase in ts than bene believes th to this poli explores th as underesti d problem. I n, a result of ion in the 19 nefit from H
ists have m percent. Y ces the risk t Zero-Bound entral banks ument is a s ment rises, p n the United
ntially in ev ese actions s n unemploym
central ban it cannot arises from ero. A negat es less than e lender cou o bound on nto a recessi ates may fal stimulus. I ment can dra further intere warned of t al Theory (19
the liquidity policy. The
he Federal n its inflatio efits.”
hat four pe icy is a puz he sources imated the In addition, f both econo 970s.
Higher Infl made severa
et one argu that interest d Problem
s pursue co short-term in policymaker d States, the very recessio spurred econ
ment during nk seeks to
reduce no m the existen tive interest
$100 in th ld do better interest rate ion. The cen ll all the way In this situ ag on indefin est-rate cuts this situation
936). Yet fo y trap a theo reason was
Open Mar on goal, whi ercent inflat zzle. In addi of opposit danger of l policymake omic theory
lation al arguments ument stand
rates hit zer
ountercylical nterest rate:
rs cut the i Federal Res on since Wor nomic recov g recessions.
o increase a ominal inter nce of an a
rate would m e future. Su
by putting c es can bind i ntral bank ca y to zero be uation, an nitely, with
.
n, which he or decades a oretical idea s that nomin
rket Comm ich “would tion is des ition to adv tion. In my low inflation ers have an
and the exp
s for raisin s out as co ro.
al monetary when a rec interst rate t serve has cu rld War II ( veries that u aggregate d rest rates asset, cash, mean that so uch a loan w cash in a saf
if an advers an respond b efore the eco economic the central e called the after Keynes a that didn’t nal interest
mittee unani likely entai sirable, the vocating the
y view, the n arising fr excessive a perience of
ng inflation mpelling: a
policy. Ty cession occu to boost ag ut the federa Romer and usually rever demand, it below zero with a gua omeone lend would never fe deposit bo se event pus by lowering onomy has r
slump an bank unable
“liquidity tr s wrote, econ
matter for p rates were
imously il much strong policy, oretical rom the aversion
double-
targets a higher
ypically, urs and ggregate
al funds Romer, rsed the faces a o. This aranteed
ds $100 r occur, ox.
shes the interest eceived d high e to end
rap,” in nomists practical usually
well above central ban without hitt
This situ bubbles in slump. In re 6% in 199 point, Japa 2006, when
Then the Federal Res August 200 remains in rose from u The United counting.
As the U banks redu Bank. With rose in muc the Fed a
“quantitativ purchases o nudged the the effects employmen In econo turn worse from a new bound wil unemploym 2.2. The R
The dang bank’s cho starts in lon rate, which target. The Now sup the central
e zero, eve nks were abl ting the zero uation chang Japanese st esponse, the 92 to 0.1%
an was in a n a brief reco e financial c serve reduce 07 to a rang
that range under 5% in d States has U.S. recessi ced interest h policymak ch of the wo and ECB, ve easing”:
of securities e American a of quantita nt.
omies with n e. Another a w financial c
ll prevent ment will ris Role of Infla
ger of hittin ice of an inf ng run equil h is indepen
long run lev ppose an ad l bank want
en during r e to cut rate o bound.
ged in the 1 tock and hou e Bank of Ja
in 1999 --y liquidity tra overy began crisis of 20 ed its target ge of 0 to 0.
in early 20 2007 to 10%
s been stuck ion spread
rates to 1%
kers unable t orld and stay
sought to they expan s. Generally and Europea ative easing
near-zero in adverse sho risis in the U
central ban e from its al ation ng the zero i
flation targe ibrium. Let ndent of mo vel of the no dverse shock ts to respon
recessions. W es by enough
1990s, starti using prices apan reduced
yet output r ap. Interest n.
07–2008 str for the fede .25% in Dec 13. Meanwh
% in 2009, a k in a liquid around the
% or less, inc o cut rates m yed high. So
stimulate nded the m y, economis an economie
have been nterest rates, ock might o U.S. or Euro nks from lready high l
interest-rate et. To see th
∗ be the lo onetary polic ominal intere k pushes the nd by lowe
When rece h to restore ing in Japan s produced a d its policy i remained de rates stayed ruck the Un eral funds rat cember 200 hile, the une and it is still dity trap for world, man cluding the E
much farther ome central their econ monetary bas
sts think the es in the rig
far too wea , bad situati occur, arisin ope. If this h
offsetting levels.
bound depe his point, sup ng run level cy, and let est rate is ∗ e economy ering interes
ssions occu full employ n. A collap a deep econ interest rate epressed. At d near zero nited States.
te from 5.25 8, and the t employment l high --near r four years ny other ce European Ce r, unemploy
banks, inclu nomies thr se through ese actions ght direction ak to restore ions could e ng, for exam happens, the
the shock,
ends on the ppose the ec l of the real
∗ be the in
∗ ∗
into recessio st rates. Sin
urred, ment se of nomic
from t that
until The 5% in target t rate r 8%.
s and entral entral ment uding rough large have n. But e full easily mple, zero and
central conomy interest nflation on, and nce the
nominal ra ∗ ∗ be reduce the but ∗ is 4 target impl policymake To see t levels of th zero, the r inflation ra greater mon
The histo did not co recessions 1980s. Infl central ban For exampl the nomina
In contra 2008–2009 inflation r recessions as policyma
How dif targets? Su not 2%. As the state o inflation an When the f reduced int
How wo do a back calibrated b
where is year. Based during 200 constant aft
ate starts at efore it hits nominal rat 4%, the nom
lies that ra ers can resto the same po he real inter real interest ate reduces netary stimu ory of inflat onstrain po between 19 lation and n nks could cut le, during th al federal fun ast, the Japa 9 occurred d rates. Nomi struck than akers respon fferent wou uppose that t
ssume neutra f the econo nd nominal financial cri terest rates b ould a larger k-of-the-env by Ball (199
the log of o d on this e 09, output in fter that, the
∗ ∗, t s zero. If te by up to minal rate c ates can fal ore full empl oint in a slig rest rate. Sin
t rate, this lower ulus.
ion helps ex olicy in the 945 and 20 nominal inte
t rates drasti he U.S. rece nds rate by 1 anese slump during an er inal interes they were in nded to the r uld recent h the Federal R ality of the t omy in 2008
l interest ra isis explode by two point r interest-rat velope calcu 99):
1.0 output, is th quation, if n 2010 wou
effect would
the central
∗ 2% and four percen can fall by ll by more, loyment.
ghtly differe nce the nom , cann bound on t xplain why t e past. In 08 occurred erest rates w ically withou
ssion of 198 10 percentag p of the 199
ra when cen t rates we n the 70s an recessions.
history hav Reserve’s ta target away 8 would hav ates were tw d at the end ts more than
te cut have ulation usin
0 0.
he real inter interest rate uld have be d have grow
bank can r d ∗ 2%, ntage points
six points.
, making it ent way, co minal rate, i, not fall belo
the real inte the zero bou many coun d during th
were high d ut approach 81–82, the F ge points, fro 0s and the G ntral banks w ere much l nd 80s, so ra
ve been wit arget during from the ze ave been the wo percent d of 2008, t n it actually d
affected the ng a dynam
8
rest rate, and es had been een 2% high wn because o
reduce this policymak . If ∗ is ag A higher in t more like onsider the f , cannot fall ow . A erest rate, a und on intere
ntries, the e 1970s an during that
ing the zero Federal Rese
om 19% to 9 Great Reces were targeti lower when
tes quickly h th higher in the 2000s w ero bound, m e same exce age points the Fed cou did.
e economy?
mic IS cur
d a time per n two points
her. If rates of the lagged
rate by ers can gain 2%
nflation ely that
feasible l below higher llowing est rates deepest nd early
era, so bound.
erve cut 9%.
ssion of ing low n these
hit zero nflation was 4%, meaning ept that
higher.
ld have
? Let us rve, as
riod is a s lower stayed d output
term. The o 2010–2013 coefficient 8.2 percent 3. Future R
Recent h rates, but forward. Th of future z historical b substantial theoretical less credibl 3.1. Histor
The fina an unusual this experie
Mishkin has surely remembere have not ex for over se benefits to benefits wil
In my v economy w once per se
First, if w unique in r Japan to th banks bega economies years.
Second, severe. The interest rate zero bound
Rudebus
output gain f 3 would be of one half, tage points.
Risks from history illust this experi he benefits o zero-bound behavior of
if central b research on le than the h rical Eviden ancial crisis
lly deep rec ence does no
(2011), for been a m ed that episo
xperienced a eventy years a higher in ll only be av view, Mishk with 2% infl
eventy years we look bey recent histo he zero bou an to target
that adopte in these two e shocks to
es to zero. S d to bind.
sch (2009) i
for 2013 wo 16.4% of a , the cumula
m the Zero B trates the co
ience does of a higher i
episodes. T U.S. interes banks targe n the zero bo historical evi nce on Inte of 2007–20 cession. As ot justify a h example, a ajor proble odes like th a negative s s. If shocks nflation targ vailable infre kin underst lation is like . Several asp yond the Un ory. A comp und in 1997
t inflation d this policy o episodes,
the econom Smaller sho llustrates th
ould be 5.9%
annual outpu ative reducti
Bound sts of hitting
not determ inflation targ To gauge th
st rates, whi et two perce ound, which
idence.
erest Rates 009 was an u
a result, ma higher inflati argues: “Alth
em in this his do not c shock to the s of this ma
get will not equently.”
tates the ris ely to hit the pects of hist nited States,
pletely sepa . It was onl rates of 2%
y both hit th the distortio my were mor
ocks would h he severity o
%, and the cu ut. Assumin ion in unemp
g the zero b mine optima
get depend o his risk, I fi
ich suggests ent inflation suggests sm
unusual eve any econom ion target.
hough this [ recent epis come very of e economy of agnitude ar t be very la sk of the z e bound far tory support the crisis o arate financi ly around 1
% or less. T he zero boun ons of mone re than large have been s of the zero-b
umulative ga ng an Okun ployment w
bound on int al policy g on the likeli first examine
s that the ri n. I then re maller risks b
ent and prod mists believe [the zero bo ode, it mus often. Indeed of this magn
e rare, then arge because zero bound
more often this predict f 2007–09 i ial crisis pu 1990 that ce The two la nd within tw etary policy
e enough to sufficient fo bound const
ain over n’s Law ould be
terest going hood e the isk is eview
but is
duced e that ound]
st be d, we itude n the
e the . An than ion.
s not ushed entral argest wenty were push or the traint
in the Unit before 200 that interes employmen rate to -2%
bound.
More ge Interest rat Fed were t Table 1, wh 1960. The (defined as twelve mon the recessio the nomina Table 1. Rec
Reces
1960:4 -19
1969:12 -
1973:11 -
1980:1 -19
1981:7 -19
1990:7 -19
2001:3 -20
2007:12 -2
We can d Recessio inflation ra most direct of the three hit the zero occurred in
ted States. H 8 implies a t rates were nt. A smalle
% or -3%, neg enerally, U.S tes would li
targeting 2%
hich present Table lists s the percen nths before) on and the f al and real fe cessions and I
ssion Infl a
961:2
1970:11
1975:3
980:7
982:11
991:3
001:11
2009:6
divide the re ons with Lo ate between t evidence o e is the Grea o bound. Th n 1960–61, a
He finds that federal fund 500 basis p er shock mi gative enoug S. history su ikely have h
% inflation.
ts key statis the core in ntage chang ); the highe following re ederal funds Interest Rate
lation Rate at Start
2.00
5.91
4.73
11.97
11.14
5.11
2.61
2.44
ecessions in ow Initial In n two and th
on the zero-b at Recession e other two and the last o
t the Taylor ds rate of -5
oints above ight have p gh to imply uggests that hit zero dur
We can se stics for each nflation rate ge in the CP
st level of u ecovery; and
rates.
es
Maximum Unemployment
Rate 7.10 (1961:05)
6.10 (1970:12)
9.00 (1975:05)
7.80 (1980:07)
10.80 (1982:11)
7.80 (1992:06)
6.30 (2003:06)
10.00 (2009:10)
Table 1 into nflation: Thr hree percent bound probl n of 2008–0 are the first one before th
r rule that fi 5% in 2009.
the level ne pushed the T
substantial t the zero bo ring previou ee this poin h of the eig at the start PI less food unemploym d the lowest
Minimum F Nomina 1.17 (1961:07
3.29 (1972:02
4.61 (1977:01
9.03 (1980:07
5.85 (1986:10
2.92 (1992:12
0.98 (2003:12
0.07 (2011:07
o two groups ree recessio nt. These ep lem at low i 09, when the t recession i he Great Re
its monetary . This sugge eeded to rest Taylor-rule costs from t ound is dan us recession nt with the ght recession
t of each re d and energ ment reached t levels reac
Federal Funds R
al Rea
7)
-0.1 (1961:
2)
-2.3 (1971:
1)
-5.8 (1975:
7)
-4.1 (1980:
0)
1.71 (1986:
2)
-0.5 (1993:
2)
-0.9 (2001:
7)
-2.1 (2012:
s:
ons began w pisodes prov inflation rat e federal fun in the Table
cession, in 2
y policy ests that tore full interest the zero ngerous.
s if the help of ns since ecession gy from d during ched by
Rate al
4 07) 1 02) 0 03) 3 06) 1
09) 5 02) 6 12) 9 01)
with the vide the es. One nds rate , which 2001.
In additi two notew unemploym Table. Seco came close recession, a the funds r fall farther Let us co of 1960–61 rate, ignorin mild recess these cases push the op typical rece
Recessio since 1960, interest rate like to kno the recessio during the f As discu as a bound bound on t likely to p actually sta before the inflation of
For the gauge the r value reac recovery. T 1. In two c below -4%
monetary p minimum r been probl was -0.6%;
of -1%. On real rate ab
ion to low in worthy feat ment peaks o
ond, the fed e. The nomi
and 1.0% fo rate reached if the econo ompare the 1 and 2001.
ng the zero sions reduce s, it seems ptimal rate ession startin ons with Hig , inflation be es never app ow what wo
ons began. W five episode ussed above, of on th the real rate push inflatio arted with 2 economy r f 2% will pro
five recessi relevance of hed by the These minim cases --the r
%. A lower policy in th real rate was ematic. For
; this episod nly in one c
ove zero.
nflation, the tures. First of 7.1% and deral funds inal funds ra ollowing the d 1.0% in 20 omy remaine severe reces . Rudebusch bound, fell ed the federa
likely that below zero ng at 2% inf gh Initial In egan above proached the
uld have ha We can get es.
, we can int he real rate.
e is -2% at on down so 2–3% inflati recovered. H oduce a bou ions that sta f a real-inter e real rate mum values
recessions o r bound of hese episode
s -2.3%, so a r the recessi de would hav
case, the rec
e recessions , they we d 6.3% are tw
rate did no ate fell to 1 2001 recess 003, many e ed weak.
ssion of 200 h argues tha to -5% durin al funds rate a recession o -that the z
flation.
nflation: In 4%. With h e zero bound appened if in
an idea by e erpret the lo If is 2% a t that point.
omewhat. In ion, the infl History sug
nd of -1% o arted with i rest-rate bou during the are shown i of 1973–75 a
f -1% wou es. For the again a lowe on of 1990–
ve been a ne cession of 1
of 1960–61 ere mild r
wo of the th ot hit the ze .2% follow sion. In the economists 08–09 to the
at the optim ng the sever e to about + n of average zero bound
five of the high initial in d. For these nflation had examining r ower bound at the start o
. However, n the three flation rate f ggests, there
on the real in inflation abo und by exam e recession in the final and 1980-- uld have se
recession o er bound of
–91, the mi ear-miss wit 1981–82, w
1 and 2001 recessions:
hree lowest i ero bound, b
ing the 196 latter case, w thought it m e mild reces mal federal f
re recession +1%. Comp e severity w
would bind eight reces nflation, nom
cases, we w d been 2% w real interest
on interest f a recession the recessio e recessions
fell to abou efore, that i nterest rate.
ove 4%, we mining the lo and subseq column of T the real rate everely dist
of 1969–70 -1% would inimum real th a lower b as the minim
have their n the but it
0–61 when might sions funds . The aring would in a sions minal would when rates rates n, the on is
that t 1%
nitial e can owest
quent Table e fell orted 0, the
have l rate ound mum
To summ percent inf constrain m bound will the frequen two percen 3.2. Theor A growin Many auth economy, o concludes t percent inf find that in such as Re that zero-b are small.1 I am du evidence on looking IS doubt that t do not cap 2000; Rudd
More sp determine function of zero, firms future, and equilibrium immediatel questionabl Some ec problem w four percen the price le
1 Williams (200 target. In his m macroeconomic
2 Some analyses Phillips curve, describe their in As discussed by
marize, the p flation targ monetary po keep intere ncy of recess nt inflation ta retical Rese ng literature hors analyze
or simulate that the zero flation. Som nterest rates eifschneider ound episod ubious of th
n the zero b and Phillips these equati
ture the ine d and Whela pecifically, o output as a f future outp expect that d policy w m. In the m
ly, reducing le that expec conomists a with a two p
nt inflation.
evel (Eggert
09) finds that the model, however,
volatility and a l s of the zero bou yet these specifi nflation equation y Mankiw (2000),
past behavio et, the low olicy in a lar est rates seve sions --eight arget has lar earch e seeks to qu
this issue i larger vers o-bound pro me papers, su
will rarely h and William des occur w hese results bound. In ad s curves of N
ons are relia ertia in real- an, 2006; Ba one can solv a function al put levels. Ty
the zero bou will ease to model, this
g the impa cted future p acknowledg ercent infla Instead, the sson and W
costs of zero-bo this result is n low value of the n und use inflation ications still do
as a mix of the N , neither of these
r of interest wer bound o
rge fraction eral points a t in a period ge costs.
uantify the r in small “N ions of the oblem is not
uch as Schm hit zero if in ms (2000) a with some fre because the ddition, the r New Keynes able tools fo -world inflat arnes et al., 2 ve forward ll future int ypically, eve und will cea
bring the expectation act of the z
policies have ge that the tion target, ey advocate Woodford, 20
und episodes can not robust: it de
neutral real intere specifications w not fit the data.
New Keynesian m models captures
t rates sugge on interest n of recessio
above the op d of 50 year
risks of hitti New Keynes models. Ge t serious eno midt-Grohe nflation is tw and Coibion
equency, bu ey conflict results depe sian models or policy ana
tion and ou 2011).2
the New K terest rates,
en if the cur ase to bind a economy n raises ou
zero bound e such stron
zero boun but still rej a different 003; Coiboin
n be substantial w epends on assum est rate (see Wood with greater inerti
Reifschneider a model and the Fu the degree of ine
ests that, wit rates is lik ons. Sometim
ptimal level s-- it appear
ing the zero ian” models enerally, thi ough to justi and Uribe wo percent.
n et al (2012 ut the welfar
with the hi end on the fo s. Many econ alysis, becau utput (e.g. M Keynesian m
and inflatio rrent interes at some poin
back to lo utput and in d. In reality
g effects.
nd is a sign ject the solu
policy: a tar n et al., 2012
with a two percen mptions of a hig
dford, 2009).
ia than the New and Williams, for uhrer-Moore (199 ertia in real-world
th a two kely to mes the
. Given rs that a
bound.
s of the is work ify four (2011), Others, 2), find re costs istorical orward- nomists use they Mankiw, model to
on as a t rate is nt in the ong run
nflation y, it is
nificant ution of rget for 2). This
nt inflation h level of
Keynesian r example, 95) model.
d inflation.
policy prod a zero-boun In this reg inflation ris out of a slu
As Wood whether it expectation than-averag inflation is It is diff central ban policymake seen prom announced credibility bank indep would allo employmen
Generall fell, but o Expected policymake targets did inflation r Countries ratios (Deb
In these ability to r interest ra expectation might be ev to raise infl 4. Opposit To justif against the modest lev Economist 2000). Yet
duces low in nd episode h gime, if the ses, which re ump.
dford (2009
“can actuall ns are affect
ge inflation, coming. Wi fficult to pre nks have not
ers promise mises of lo
disinflation by adopting pendence. T ow actual in
nt.
y, these effo only because inflation o ers made a not achieve reduction--
with highly belle and Fis
episodes, i reduce infla ates. Yet d ns. The prom
ven less effe lation at zero ton to High fy a four pe e costs. Man
vels, is hig at the ECB policymake
nflation on a has pushed e nominal educes the r ) emphasize ly be made ted in the de
the central ill that work edict the ef tried this po to raise inf wer inflatio ns during g explicit in The goal wa nflation to forts to shift
e monetary only fell a announceme e lower sac than other y independe cher, 1994).
it was gene tion if they disinflation
mises of hig ective, becau
o interest rat her Inflatio
ercent inflat ny central b ghly undesi B called infla
ers’ aversion
average, but the price le interest rate real interest r es, the succe
credible to esired way.”
bank must k?
ffects of pr olicy. We ha flation signif on. Policym
the 1980s nflation targ as to reduce
fall without expectation contraction fter actual ents. Countr crifice ratios
r countries ent central
.
erally clear y were deter announcem gher inflatio use policyma
tes.
n
ion target, w bankers bel irable. Ottm ation “a betr n to moderat
inflation ris evel below it
e hits zero rate and boo ess of this po the public,
” During a convince pe rice-level ta
ave not seen ficantly. We makers in and 1990s gets or stren e expected t large cost ns did not su
ns caused d inflation ries with e s --output lo (Bernanke banks had
that centra rmined to d ments had on that Woo
akers may n
we must we lieve that in mar Issing, rayal of the te inflation is
ses temporar ts long run o, then expe
osts the econ olicy depend so that infl period of lo eople that h
rgeting, bec n what happe e have, how many coun s, often see ngthening ce inflation, w ts in output
ucceed. Infl deep recess fell, not w explicit infl osses per un
et al., 1 higher sacr l banks had do so, by ra
little effec odford envi not have the
eigh the ben nflation, eve the first C people” (Is s unwarrante
rily if path.
ected nomy ds on ation ower-
igher cause ens if wever, ntries eking entral which t and
ation sions.
when ation nit of 999).
rifice d the aising
t on sions
tools
nefits en at Chief ssing, ed.
4.1. Are T Economi (2011) give the financi aggregate p and difficul But how by the exp inflation. P arising from compelling of the dirty universally up with emb
Krugman 1970s. Sin empirical st
A numbe growth. A reduce gro threshold v Easterly, 19 4.2. Is 4%
Some ce harm the ec Some argu inflation to In rejecti would be h
“inflation e Accordin period of ti inflation lo 4%, we wo say, well, if expectation Similar Mishkin sa than at a 2
There Costs ists have po es a typical ial sector;
price level; d lties in finan large are th erience of t Papers such a
m relative-p g case that in y little secret y regarded a barrassingly n is talking
ce it is diff tudies have er of cross-c common f wth, but lo vary consid 996), but 4%
Inflation D entral banke
conomy. W ue that a d rise above 4 ing a 4% inf
higher and expectations ng to Berna ime, has esta ow, around ould risk a l
f we go to 4%
ns at 4%.”
arguments ays: “If it we
% level, the s?
ointed out m l list: distort greater un distortions o ncial plannin hese effects?
the 1970s, h as Fischer (1 price variab nflation is ha ts of econom as a terrible y small num about rese ficult to ide even tried to country stud finding is th
wer levels erably, from
% is clearly b Destabilizin ers acknowl Why then do decision to 4%, or at lea flation targe
probably m s would also
anke (2010 ablished a g 2%. If we w lot of that h
%, why not g appear in ere no more n the case fo
many advers tions in cas certainty ab of the tax sys
ng.
? An empiric has tried to 1981), for ex bility. This armful. As K mic analysis
scourge, eff mbers.”
earch on the entify costs o find costs dies ask whe hat inflation
of inflation m 8% (Sare
below the th ng?
ledge that 4 they oppose accept 4%
ast create ex et, Bernanke more volati likely becom b): “The F great deal of
were to go t hard-won cr go to 6%? I
Woodford difficult to s for raising th
se effects of sh holdings;
bout relativ stem; redistr cal literature measure so xample, exa research ha Krugman (19 is that even fforts to mea e double-dig
of inflation of single-di ether inflatio n rates abov n are neutral
el, 1995) to hreshold.
4% inflation e an increas
inflation m xpectations o e (2010a) ass ile under su me significa Federal Res f credibility to 4% and s redibility, be It’d be very d (2010) and stabilize infl he inflation
f inflation. M overinvestm ve prices a
ributions of e, motivated ome of the c amine ineffic as not prod 997) remark n though infl asure its cos git inflation n at that lev git inflation on affects ec ve some th l. Estimates o 40% (Bru
n does not se from 2%
may actually of that outco serts that “in uch a polic antly less sta serve, over in terms of k say we’re g ecause folks difficult to ti d Mishkin flation at a 4 target to 4%
Mishkin ment in and the
wealth;
largely costs of ciencies duced a ks, “one
lation is ts come n of the vel, few
. onomic hreshold s of the uno and
greatly to 4%?
y cause ome.
nflation cy” and able”.
a long keeping going to s would ie down (2011).
% level
% would
be much st that this is
“If infla believe that and then th not 6%, or the 3% leve We mig alcoholic’s is the first s The ratio Mishkin ar explain this recent expe the zero-bo inflation to inflation ro policymake
An incr transitional greater than harm the e not tried ve the casual a History expectation measured b They follow followed in expectation Mishkin.
4.3. Under Central Under Cha digit inflati about 4% f
“conquest”
Volcker an tightened p
tronger. How not the case tion rises to t price stabi he question 8%, and so el, it tends to ght call this
first drink, step in a dan onale for thi
rgue that a s policy to t erience that
ound proble o 4%, and k ose to 4%
ers repeat th ease in a l period of n usual. But
conomy sig ery hard to arguments th
does not s ns” if infla by surveys, h
wed inflatio nflation dow ns will over
rstanding th bankers hav airman Paul
ion of the 1 from 1985
” of inflation nd his collea
policy at the
wever, the h e.”
o 4%”, Mis ility is no lo
arises, ‘if a o on.’ We ha o keep on ris view the 4% inflation ngerous, unc
s view is no cental bank the public, a 4% inflation em. Why c keep it ther
and the F hat mistake?
central ban f learning,
nobody has gnificantly. I make that c hat I have qu suggest that ation rises
have genera on up during wn. If inflat
rshoot this
he Attitudes ve not alw l Volcker, t 1970s, but i through 198 n (e.g. Sarg agues did no
e end of 19
history of th shkin conten
nger a cred a 4% level o ave seen tha
sing.”
addictive th n may not d controllable p ot clear. In o
k should de and carry it n is better th can’t policy re? Mishkin Fed let it k
nk’s inflatio during whi s demonstrat Indeed, oppo case; they si uoted.
t it would modestly.
ally followed g the 1960s tion rises to level, as s
s of Policym ways found
he Federal it allowed th 88. This ex
ent, 1999).
ot try to redu 988, when in
he inflation p nds, “the pu dible goal of
of inflation at when infla
heory of in do great harm
process.
other context etermine its t out. We ha
han 2% infla ymakers ex n points to t keep rising, on target m ich inflation ted that this onents of 4%
imply dismi be “difficu Inflation e d actual infl
and 70s, an o 4%, it see
uggested by
makers 4% inflatio Reserve en he inflation xperience is
Once inflat duce it furthe
nflation star
process sug ublic is like f the central is OK, then ation rises a nflation. Lik m by itself, b
ts, Bernanke optimal po ave learned
ation, becau plain this, the 1960s, w
but why might invol n uncertain transition w
% inflation iss the idea ult to tie d expectations lation with a nd after that ms unlikely y Bernanke
on unaccept nded the dou n rate to sett
often called tion reached
er. The Fed rted rising a
ggests ely to
bank n why above ke an
but it e and olicy, from use of raise when must ve a ty is would
have with down s, as a lag.
they y that e and
table.
uble- tle at d the d 4%, only again
(Romer and Support Canada and and pushed were partly down as a reached 2%
run.
Why do did not? Th Instead, po proportion One is t inflation of thinking of were consi Policymake inflationary Volcker’s c dangers of
It is instr 1960s and 7 Depression memory of that might accelerate.
digit inflati are determi The sec research by that centra excessive in mitigated if than is jus described, justification 5. Conclus A worldw The centra
d Romer, 19 for 2% infl d New Zeala d inflation t y accidental side-effect
%, however, today’s cen he answer is olicymakers to its true co the tendenc f the 1970s f central ban idered the w ers believed y episode:
conquest. T inflation.
ructive to c 70s. DeLon n” influence
f high unem t slow the Since then ion replacin ined not to r ond factor y academics al banks fa nflation. Ro f policymak stified by th
the Kydland n for hawkis sion
wide consen al banks of
994).
ation started and, many c to that level l. In the Un of recessio policymake ntral bankers
not that res have develo osts. There a y for policy s was a sca nkers since t worst mone d that their m
nobody wa This mind-se
ompare the g (1997) des ed the Fede mployment m
economy, the pendulu ng the Depre
repeat.
that has i s. Kydland a ace a dynam ogoff (1985)
kers are high he true cost d-Prescott a sh monetary
nsus holds th most advan
d to grow d central banks
l. In other c nited States, ons in 1990–
ers decided t s oppose 4%
earch has id oped an ave are two reaso
ymakers to arring exper
then. Before tary-policy most importa
anted to b et has led p monetary p scribes how eral Reserv made policy
with the r um has swun ession as the nfluenced c and Prescott
mic consist convinced e hly inflation
ts of inflati and Rogoff y policy.
hat the optim nced econom
duirng the 19 s have adopt countries, d
for exampl –91 and 20 to lock in th
% inflation w dentified new ersion to infl
ons.
fight the l rience that e the crisis
disaster sin ant job was be remembe olicymakers policy of rec w “the shadow
ve in the e ymakers fea result that ng the other
e nightmare central bank t (1977) con tency probl economists n-averse --m
ion. As Be papers prov
mal inflation mies target
990s. Startin ted targets n eclines in in le, inflation 001. Once in hat rate for t
when Paul V w costs of in
lation that is last war. Th has domina of 2008, the nce World W
to prevent ered for re s to exagger cent decades w cast by th earlier perio arful of any they let in r way, with e that centra kers is the nvinced econ
lem that pr that this pro more averse- rnanke (200 vide an inte
n rate is abo inflation ne
ng with near 2%
nflation drifted nflation he long Volcker nflation.
s out of he high ated the e 1970s War II.
another eversing rate the s to the he Great od. The y action
nflation double- al banks
oretical nomists roduces oblem is -, even, 04) has ellectual
out 2%.
ear that
level. Som central bank
This ess inflation. R on monetar result that benefit wou an economy Raising i 4% target m employmen the zero-bo sufficient b Reference
Akerlof, G.A and the O Activity, 1 Ball, Laurenc
83.
Barnes, M. L Looking R Curve,” F Bernanke, Be Symposiu Bernanke, Be April 14, Bernanke, Be Monetary Bernanke, Be
“Inflation Press.
Blanchard, O Macroeco Bruno, M. a Relationsh Coibion, Oliv
Rate in N Light of th Debelle, Guy
Goals, Gu Bank of B DeLong, J. B Romer an of Chicag
e emerging ks have a lo say argues t Raising the i ry policy ari economic uld come at y significant inflation tar might reduce nt (Akerlof
ound proble by itself to ju es
A., W.T. Dicken Optimal Rates o 1(2000): 1-60.
ce, 1999, “Effic L., F. Gumbau- Relationships Federal Reserve
en. S., 2010a, um, August 27, en. S., 2010b,
2010.
en. S., 2004, “W y Policy 25 Yea
en S., Thoma n Targeting: Le Olivier, Giov onomic Policy,
and W. East hip,” Federal R vier, Yuriy Gor New Keynesian he Zero Lower y and Stanley F uidelines, and Boston Confere Bradford, 1997 nd David Rome go Press.
economies ong-term goa that central inflation targ ising from th downturns t minimal co
tly.
gets might h e the effects
et al., 2000 em because ustify a 4% t
ns and G.L. Pe of Inflation an cient Rules for -Brisa, D. Lie in Closed For e Bank of Bost
“The Econom , 2010.
Testimony be What Have W ars after Octob as Laubach, F essons from th vanni Dell’A
” IMF Staff Po terly, 1996, “ Reserve Bank rodnichenko an n Models: Sho
r Bound?" Rev Fischer, 1994, d Constraints
ence Series No 7, “America’s
er (eds.), Redu
have highe al of reducin banks wou get from 2%
he zero boun would be l ost, because have additio s of downwa 0). I have st e it is so c
target.
erry, 2000, “N d Unemploym r Monetary Pol , and G. P. Ol m: An Applic ton Working P mic Outlook a efore the Joint We Learned Sin ber 1979, Feder Frederic S. M he Internationa Ariccia and P
osition Note SP
“Inflation and of St. Louis Re nd Johannes W
uld Central Ba view of Econom
“How Indepen Facing Mone o. 38.
Only Peacetim ucing Inflation:
er targets, bu ng the target
uld do bett
% would eas nd on intere less severe.
4% inflatio onal benefits ard nominal tressed the b clear-cut, an
Near-Rational W ment” Brooking
licy” Internati livei, 2011, “E cation to the N Paper No. 11-3 and Monetary t Economic C nce October 19
ral Reserve Ba Mishkin and A al Experience, Paolo Mauro PN/10/03.
d Growth: In eview 78(3): 13 Wieland, 2012,
anks Raise Th mic Studies, 79 ndent Should a etary Policyma me Inflation: th : Motivation a
ut typically ts.
ter to target se the constr est rates, wit This impo on does not
s. For examp wage rigidi benefit of ea nd because
Wage and Pric gs Papers on E ional Finance 2 Estimation of F New Keynesian
.
Policy,” Jacks ommittee of C 979?” in Reflec ank of St. Loui Adam S. Pose
” Princeton U o, 2010, “Re n Search of
39–146.
"The Optimal heir Inflation T 9(4): 1371–140 a Central Bank akers, Federal he 1970s,” in
nd Strategy, U
their t 4%
raints th the ortant harm ple, a ty on asing it is
ce Setting Economic 2(1): 63–
Forward- n Phillips
son Hole Congress, ctions on is.
en, 1999, University
ethinking a Stable
Inflation Targets in 06.
k Be?” in Reserve Christina University
Eggertsson, G Optimal M Fischer, Stan
Brookings Fuhrer, Jeffre
Economic Issing, Otmar Conferenc Keynes, John Macmillan Krugman, Pau 1990s” M Kydland, Fin Inconsiste Mankiw, N.
Inflation a Mishkin, Fre
Working P Reifschneider
a Low Inf Rogoff, Kenn Target,” Q Romer, Chris
Monetary Romer, Chri Macroeco Rudd, Jeremy Explain In Rudebusch, G FRBSF Ec Sarel, Michae
Paper #95 Sargent, Tho
Press.
Schmitt-Groh Friedman 722, Elsev Williams, Joh Brookings Woodford, M
Activity, 2
ACKNOWLEDGM and Richard
Gauti B. and M Monetary Polic
nley, 1981, “ s Papers on Ec ey C. and Geor cs, 110(1): 127 r, 2000, “Why ce.
n M., 1936, n.
ul R., 1997, “T MIT Press.
nn E. and Ed ency of Optima Gregory, Ma and Unemploy
deric S., 2011 Paper #16755.
r, David L. and flation Era,” Jo neth, 1985, “Th Quarterly Jour
stina D. and y Economics, 3 istina D. and onomics Annua y and Karl W nflation Dynam Glenn D., 2009 conomic Letter el, 1995, “Non 5/96.
omas J., 1999, he, Stephanie, and Woodfor vier Press.
hn C., 2009, s Papers on Ec Michael, 2009 2: 38–49.
MENT I am gratefu Miller.
Michael Woodf cy,” Brookings
“Relative Sho conomic Activi rge R. Moore, 7–159.
y Price Stabilit
“The General The Age of Di dward C. Pre al Plans," Jour ay 2001, “Th yment” Econom
1, “Monetary .
d John C. Will ournal of Mone
he Optimal De rnal of Econom David Romer 4(1): 75–88.
d David Rom al, 9: 13–57.
Whelan, 2006, mics?” America
9, “The Fed’s r 2009-17.
nlinear Effects , “The Conqu
and Martin U rd (eds.), Hand
“Heeding De conomic Activi 9, “Comment
ul for suggestions
ford, 2003, “T s Papers on Ec ocks, Relative ity, 2: 381–431 1995, “Inflatio ty?” in Procee l Theory of minished Expe scott, 1977, “ rnal of Politica he Inexorable mic Journal 11
Policy Strateg liams, 2000, “T ey, Credit and egree of Comm mics, 100(4): 11 r, 1994b, “Mo
mer, 1994a,
“Can Rationa an Economic R
Monetary Pol of Inflation o uest of Ameri Uribe, 2011, “ dbook of Mon eadalus: Optim
ity, 2: 1-37.
on Williams,
s from Olivier Bla
The Zero Boun conomic Activi e Price Varia
1.
on Persistence edings of First Employment, ectations: US E
“Rules Rather al Economy, 85 and Mysterio 1, C45-C61.
gy: Lessons fr Three Lessons Banking, 32(4 mitment to an
169–1189.
onetary Policy
“What Ends al Expectation Review, 96(1):
licy Response on Economic G
ican Inflation”
“The Optimal netary Econom mal Inflation a ,” Brookings
anchard, Daniel L
d on Interest R ty, 1: 139–211 ability, and In
,” Quarterly Jo ECB Central Interest and Economic Poli r than Discre 5: 473–491.
ous Tradeoff rom the Crisis
for Monetary 4): 933–966.
Intermediate M y Matters,” Jo
Recessions?”
ns Sticky-Price 303–320.
to the Curren Growth,” IMF
” Princeton U Rate of Infla mics, Volume 3
and the Zero Papers on E
Leigh, N. Gregor
Rates and . nflation,”
ournal of Banking Money”
icy in the tion: the
between ,” NBER Policy in Monetary ournal of
” NBER e Models
t Crisis,”
Working University ation,” in 3B, 653–
Bound,”
Economic
y Mankiw,
