Авторы: 147 А Б В Г Д Е З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я

Книги:  180 А Б В Г Д Е З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я


7.11.2 Monetary regimes and inflation targeting

If a consensus of economists agree that inflation is damaging to economic

welfare, it remains to be decided how best to control inflation. Since it is now

widely accepted that the primary long-run goal of monetary policy is to

control inflation and create reasonable price stability, the clear task for economists

is to decide on the exact form of monetary regime to adopt in order to

achieve this goal. Monetary regimes are characterized by the use of a specific

nominal anchor. Mishkin (1999) defines a nominal anchor as ‘a constraint on

the value of domestic money’ or more broadly as ‘a constraint on discretionary

policy that helps weaken the time-inconsistency problem’. This helps to

solve the inflation bias problem inherent with the use of discretionary demand

management policies (Kydland and Prescott, 1977). In practice, during

the last 50 years, we can distinguish four types of monetary regime that have

operated in market economies; first, exchange rate targeting, for example the

UK, 1990–92; second, monetary targeting, for example the UK, 1976–87;

third, explicit inflation targeting, for example the UK, 1992 to date; fourth,

implicit inflation targeting, for example the USA, in recent years (see Mishkin,

1999; Goodfriend, 2004). While each of these monetary regimes has advantages

and disadvantages, in recent years an increasing number of countries

have begun to adopt inflation targeting in various forms, combined with an

accountable and more transparent independent central bank (see Alesina and

Summers, 1993; Fischer, 1995a, 1995b, 1996b; Green, 1996; Bernanke and

Mishkin, 1992, 1997; Bernanke and Woodford, 1997; Bernanke et al., 1999;

King, 1997a, 1997b; Snowdon, 1997; Svensson, 1997a, 1997b, 1999, 2000;

Artis et al., 1998; Haldane, 1998; Vickers, 1998; Mishkin, 1999, 2000a,

2000b, 2002; Gartner, 2000; Muscatelli and Trecroci, 2000; Piga, 2000;

Britton, 2002; Geraats, 2002; Bernanke and Woodford, 2004; see also the

interview with Bernanke in Snowdon, 2002a, 2002b).

Following Svensson (1997a, 1997b) and Mishkin (2002), we can view

inflation targeting as a monetary regime that encompasses six main elements:

1. the public announcement of medium-term numerical targets for inflation;

2. a firm institutional commitment to price stability (usually a low and

stable rate of inflation around 2–3 per cent) as the primary goal of

monetary policy; the government, representing society, assigns a loss

function to the central bank;

3. an ‘information-inclusive strategy’ where many variables are used for

deciding the setting of policy variables;

4. greater transparency and openness in the implementation of monetary

policy so as to facilitate better communication with the public; inflation

targets are much easier to understand than exchange rate or monetary


5. increased accountability of the central bank with respect to achieving its

inflation objectives; the inflation target provides an ex post indicator of

monetary policy performance; also, by estimating inflationary expectations

relative to the inflation target, it is possible to get a measure of the

credibility of the policy;

6. because the use of inflation targeting as a nominal anchor involves comparing

the announced target for inflation with the inflation forecast as the

basis for making monetary policy decisions, Svensson (1997b) has pointed

out that ‘inflation targeting implies inflation forecast targeting’ and ‘the

central bank’s inflation forecast becomes the intermediate target’.

The successful adoption of an inflation targeting regime also has certain

other key prerequisites. The credibility of inflation targeting as a strategy will

obviously be greatly enhanced by having a sound financial system where the

central bank has complete instrument independence in order to meet its

inflation objectives (see Berger et al., 2001; Piga, 2000). To this end the Bank

of England was granted operational independence in May, 1997 (Brown,

1997). It is also crucial that central banks in inflation targeting countries

should be free of fiscal dominance. It is highly unlikely that countries with

persistent and large fiscal deficits will be able to credibly implement a

successful inflation targeting strategy. This may be a particular problem for

many developing and transition economies (Mishkin, 2000a). Successful inflation

targeting also requires the adoption of a floating exchange rate regime

to ensure that the country adopting this strategy maintains independence for

its monetary policy. The well-known open economy policy trilemma shows

that a country cannot simultaneously maintain open capital markets + fixed

exchange rates + an independent monetary policy oriented towards domestic

objectives. A government can choose any two of these but not all three

simultaneously! If a government wants to target monetary policy towards

domestic considerations such as an inflation target, either capital mobility or

the exchange rate target will have to be abandoned (see Obstfeld, 1998;

Obstfeld and Taylor, 1998; Snowdon, 2004b).

As we noted in Chapter 5, Svensson (1997a) has shown how inflation

targeting has emerged as a strategy designed to eliminate the inflation bias

inherent in discretionary monetary policies. While Friedman and Kuttner

(1996) interpret inflation targeting as a form of monetary rule, Bernanke and

Mishkin (1997) prefer to view it as a monetary regime that subjects the

central bank to a form of ‘constrained discretion’. Bernanke and Mishkin see

inflation targeting as a framework for monetary policy rather than a rigid

policy rule. In practice all countries that have adopted inflation targeting have

also built an element of flexibility into the target. This flexible approach is

supported by Mervyn King (2004), who was appointed Governor of the Bank

of England following the retirement of Eddie George in June 2003. King

identifies the ‘core of the monetary policy problem’ as being ‘uncertainty

about future social decisions resulting from the impossibility and the undesirability

of committing successors to any given monetary policy strategy’.

These problems make any form of fixed rule undesirable even if it were

possible to commit to one because, as King (2004) argues,

The exercise of some discretion is desirable in order that we may learn. The most

cogent argument against the adoption of a fixed monetary policy rule is that no

rule is likely to remain optimal for long … So we would not want to embed any

rule deeply into our decision making structure … Instead, we delegate the power

of decision to an institution that will implement policy period by period exercising

constrained discretion.

The need for flexibility due to uncertainty is also emphasized by Alan

Greenspan, who became Chaiman of the US Federal Reserve in August 1987

(he is due to retire in June 2008). The Federal Reserve’s experiences over the

post-war era make it clear that ‘uncertainty is not just a pervasive feature of

the monetary policy landscape; it is the defining characteristic of that landscape’

(Greenspan, 2004). Furthermore:

Given our inevitably incomplete knowledge about key structural aspects of an everchanging

economy and the sometimes symmetric costs or benefits of particular

outcomes, a central bank needs to consider not only the most likely future path for

the economy but also the distribution of possible outcomes about that path. The

decision-makers then need to reach a judgement about the probabilities, costs, and

benefits of the various possible outcomes under alternative choices for policy.

Clearly the setting of interest rates is as much ‘art as science’ (Cecchetti,

2000). The need for flexibility can in part be illustrated by considering a

conventional form of the loss function (Lt) assigned to central bankers given

by equation (7.16).

Lt = Pt − P + Yt − Y > 1


[ ˙ ˙ )2 φ( )2 ], φ 0 (7.16)

In this quadratic social loss function ˙Pt is the rate of inflation at time period

t, P˙ * is the inflation target, Yt is aggregate output at time t, and Y* represents

the natural rate or target rate of output. The parameter φ is the relative weight

given to stabilizing the output gap. For strict inflation targeting φ = 0, whereas

with flexible inflation targeting φ > 0. As Svenssson (1997a) notes, ‘no

central bank with an explicit inflation target seems to behave as if it wishes to

achieve the target at all cost’. Setting φ = 0 would be the policy stance

adopted by those who Mervyn King (1997b) describes as ‘inflation nutters’.

Thus all countries that have introduced inflation targeting have built an element

of flexibility into the target (Allsopp and Vines, 2000).

What should be the numerical value of the inflation target? Alan

Greenspan, currently the most powerful monetary policy maker in the world,

has reputedly defined price stability as a situation where people cease to

take inflation into account in their decisions. More specifically, Bernanke et

al. (1999) come down in favour of a positive value for the inflation target in

the range 1–3 per cent. This is supported by Summers (1991b, 1996),

Akerlof et al. (1996), and Fischer (1996b). One of the main lessons of the

Great Depression, and one that has been repeated in much milder form in

Japan during the last decade, is that it is of paramount importance that

policy makers ensure that economies avoid deflation (Buiter, 2003b;

Eggertsson and Woodford, 2003; Svensson, 2003a). Because the nominal

interest rate has a lower bound of zero, any general deflation of prices will

cause an extremely damaging increase in real interest rates. Cechetti (1998)

argues that the message for inflation targeting strategies is clear, ‘be wary

of targets that imply a significant chance of deflation’. It would therefore

seem unwise to follow Feldstein’s (1999) recommendation to set a zero

inflation target. Akerlof et al. (1996) also support a positive inflation target

to allow for relative price changes. If nominal wages are rigid downwards,

then an alternative way of engineering a fall in real wages in order to

stimulate employment is to raise the general price level via inflation relative

to sticky nominal wages. With a flexible and positive inflation target this

option is available for the central bank.

Following the UK’s departure from the ERM in September 1992 it became

imperative to put in place a new nominal anchor to control inflation. During

the post-1945 period we can identify five monetary regimes adopted by the

UK monetary authorities, namely, a fixed (adjustable peg) exchange rate

regime, 1948–71; a floating exchange rate regime with no nominal anchor,

1971–6; monetary targets, 1976–87; exchange rate targeting (‘shadowing the

Deutchmark’ followed by membership of the ERM), 1987–92; and finally

inflation targeting, 1992 to date (Balls and O’Donnell, 2002). The credibility

of the inflation targeting regime was substantially improved in May 1997

when the Bank of England was given operational independence. This decision,

taken by the ‘New Labour’ government, was designed to enhance the

administration’s anti-inflation credibility by removing the suspicion that ideoThe

new Keynesian school 417

logical or short-term electoral considerations would in future influence the

conduct of stabilization policy (see Chapter 10).

The current UK monetary policy framework encompasses the following

main features:

1. A symmetrical inflation target. The targets or goals of policy are set by

the Chancellor of the Exchequer.

2. Monthly monetary policy meetings by a nine-member Monetary Policy

Committee (MPC) of ‘experts’. To date, current and past membership of

the MPC has included many distinguished economists, including Mervyn

King, Charles Bean, Steven Nickell, Charles Goodhart, Willem Buiter,

Alan Budd, John Vickers, Sushil Wadhami, DeAnne Julius, Christopher

Allsopp, Kate Barker and Eddie George.

3. Instrument independence for the central bank. The MPC has responsibility

for setting interest rates with the central objective of publication of

MPC minutes.

4. Publication of a quarterly Inflation Report which sets forth the Bank of

England’s inflation and GDP forecasts. The Bank of England’s inflation

forecast is published in the form of a probability distribution presented in

the form of a ‘fan chart’ (see Figure 7.13). The Bank’s current objective

is to achieve an inflation target of 2 per cent, as measured by the 12-

month increase in the consumer prices index (CPI). This target was

Figure 7.13 Bank of England inflation report fan chart for February 2004:

forecast of CPI inflation at constant nominal interest rates of

4.0 per cent

1999 2000 01 02 03 04 05 06







Percentage increases in prices on a year earlier

announced on 10 December 2003. Previously the inflation target was 2.5

per cent based on RPIX inflation (the retail prices index excluding mortgage

interest payments).

5. An open letter system. Should inflation deviate from target by more than

1 per cent in either direction, the Governor of the Bank of England, on

behalf of the MPC, must write an open letter to the Chancellor explaining

the reasons for the deviation of inflation from target, an accommodative

approach when confronted by large supply shocks to ease the adverse

output and employment consequences in such circumstances (Budd, 1998;

Bean, 1998; Treasury, 1999; Eijffinger, 2002b).

Since 1992 the inflation performance of the UK economy has been very

impressive, especially when compared to earlier periods such as the 1970s

Note: a Implied average expectations from 5 to 10 years ahead, derived from index-linked gilts.

Source: Bank of England, www.bankofengland.co.uk.

Figure 7.14 UK inflation and inflation expectations, October 1991–

October 2003










Oct. 91 Oct. 93 Oct. 95 Oct. 97 Oct. 99 Oct. 01 Oct. 03

Implied inflation

from IGsa

RPIX inflation rate






and 1980s when inflation was high and volatile. Figure 7.14 clearly illustrates

the dramatic improvement in the UK’s inflation performance since 1992,

especially compared to earlier periods (see King, 2004).

While it is too early to tell if this monetary arrangement can deliver lower

inflation and greater economic stability over the longer term, especially in a

more turbulent world than that witnessed during the 1990s, the evidence from

recent years at least gives some cause for optimism, a case of ‘so far so good’

(see Treasury, 1999; Balls and O’Donnell, 2002). However, Ball and Sheridan

(2003) argue that there is no evidence that inflation targeting has improved

economic performance as measured by inflation, output growth and interest

rates. They present evidence that non-inflation-targeting countries have also

experienced a decline in inflation during the same period as the inflation

targeters, suggesting perhaps that better inflation performance may have been

the result of other factors. For example, Rogoff (2003), in noting the fall in

global inflation since the early 1980s, identifies the interaction of globalization,

privatization and deregulation as important factors, along with better

policies and institutions, as major factors contributing to disinflation.

7.11.3 A new Keynesian approach to monetary policy

In two influential papers, Clarida et al. (1999, 2000) set out what they consider

to be some important lessons that economists have learned about the

conduct of monetary policy. Economists’ research in this field points towards

some useful general principles about optimal policy. They identify their

approach as new Keynesian because in their model nominal price rigidities

allow monetary policy to have non-neutral effects on real variables in the

short run, there is a positive short-run relationship between output and inflation

(that is, a Phillips curve), and the ex ante real interest rate is negatively

related to output (that is, an IS function).

In their analysis of US monetary policy in the period 1960–96 Clarida et

al. (2000) show that there is a ‘significant difference in the way that monetary

policy was conducted pre-and post-1979’, being relatively well managed

after 1979 compared to the earlier period. The key difference between the

two periods is the magnitude and speed of response of the Federal Reserve to

expected inflation. Under the respective chairmanships of William M. Martin,

G.William Miller and Arthur Burns, the Fed was ‘highly accommodative’. In

contrast, in the years of Paul Volcker and Alan Greenspan, the Fed was much

more ‘proactive toward controlling inflation’ (see Romer and Romer, 2002,


Clarida et al. (2000) conduct their investigation by specifying a baseline

policy reaction function of the form given by (7.17):

rt r E Pt k t P E yt q t

= +β[ ( ˙ | ) − ˙]+ γ [ | ] , Ω , Ω (7.17)

Here rt

* represents the target rate for the Federal Funds (FF) nominal interest

rate; ˙

Pt,k is the rate of inflation between time periods t and t + k; P˙ * is the

inflation target; yt,q measures the average deviation between actual GDP and

the target level of GDP (the output gap) between time periods t and t + q; E is

the expectations operator; Ωt is the information set available to the policy

maker at the time the interest rate is set; and r* is the ‘desired’ nominal FF

rate when both ˙P and y are at their target levels. For a central bank with a

quadratic loss function, such as the one given by equation (7.16), this form of

policy reaction function (rule) is appropriate in a new Keynesian setting. The

policy rule given by (7.17) differs from the well-known ‘Taylor rule’ in that it

is forward-looking (see Taylor, 1993, 1998a). Taylor proposed a rule where

the Fed reacts to lagged output and inflation whereas (7.17) suggests that the

Fed set the FF rate according to their expectation of the future values of

inflation and output gap. The Taylor rule is equivalent to a ‘special case’ of

equation (7.17) where lagged values of inflation and the output gap provide

sufficient information for forecasting future inflation. First recommended at

the 1992 Carnegie-Rochester Conference, Taylor’s (1993) policy formula is

given by (7.18):

r = P˙ + g(y) + h(P˙ − P˙ * ) + r* (7.18)

where y is real GDP measured as the percentage deviation from potential GDP;

r is the short-term nominal rate of interest in percentage points; ˙P is the rate

of inflation and P˙ * the target rate of inflation; r* is the ‘implicit real interest

rate in the central bank’s reaction function’; and the parameters g, h, ˙P * and

r* all have a positive value. With this rule short-term nominal interest rates

will rise if output and/or inflation are above their target values and nominal

rates will fall when either is below their target value. For a critique of Taylor

rules see Svensson (2003b).

In the case of (7.17) the policy maker is able to take into account a broad

selection of information about the future path of the economy. In standard

macroeconomic models aggregate demand responds negatively to the real

rate of interest; that is, higher real rates dampen economic activity and lower

real rates stimulate economic activity. From equation (7.17) we can derive

the ‘implied rule’ for the target (ex ante) real rate of interest, rrt

. This is

given by equation (7.19):

rrt rr E Pt k t P E yt q t

= * + − − +



, ( )[ (˙ β 1 |Ω ) ˙ ] γ [ |Ω ] (7.19)

Here, rrt rt E Pt k t P

−[ ( ˙ | ) − ˙ ], ,

Ω * and rr* r* − P˙ * is the long-run equilibrium

real rate of interest. According to (7.19) the real rate target will respond

to changes in the Fed’s expectations about future output and inflation. HowThe

new Keynesian school 421

ever, as Clarida et al. point out, the sign of the response of rrt

to expected

changes in output and inflation will depend on the respective values of the

coefficients β and γ. Providing that β > 1 and γ > 0, then the interest rate rule

will tend be stabilizing. If β ≤ 1 and γ ≤ 0, then interest rate rules ‘are likely

to be destabilising, or, at best, accommodative of shocks’. With β < 1, an

increase in expected inflation leads to a decline in the real interest rate, which

in turn stimulates aggregate demand thereby exacerbating inflation. During

the mid-1970s, the real interest rate in the USA was negative even though

inflation was above 10 per cent.

By building on this basic framework, Clarida et al. (2000), in their examination

of the conduct of monetary policy in the period 1960–96, find that

the Federal reserve was highly accommodative in the pre-Volcker years: on average,

it let the real short-term interest rate decline as anticipated inflation rose. While it

raised the nominal rate, it did so by less than the increase in expected inflation. On

the other hand, during the Volcker–Greenspan era the Federal Reserve adopted a

proactive stance toward controlling inflation: it systematically raised real as well as

nominal short-term interest rates in response to higher expected inflation.

During the 1970s, despite accelerating inflation, the FF nominal rate tracked

the rate of inflation but for much of the period this led to a zero or negative ex

post real rate. There was a visible change in the conduct of monetary policy

after 1979 when, following the Volcker disinflation via tight monetary policy,

the real rate for most of the 1980s became positive. In recognition of the lag

in monetary policy’s impact on economic activity, the new monetary regime

involved a pre-emptive response to the build-up of inflationary pressures. As

a result of this marked change in the Fed’s policy, inflation was successfully

reduced although as a consequence of the disinflation the USA suffered its

worst recession since the Great Depression. Unemployment rose from 5.7 per

cent in the second quarter of 1979 to 10.7 per cent in the fourth quarter of

1982 (Gordon, 2003).

In their analysis of the change of policy regime at the Fed, Clarida et al.

compare the FF rate with the estimated target forward (FWD) value for the

interest rate under the ‘Volcker–Greenspan’ rule for the whole period. According

to Clarida et al. the estimated rule ‘does a good job’ of capturing the

broad movements of the FF rate for the post-1979 sample period.

There seems little doubt that the lower inflation experienced during the past

two decades owes a great deal to the more anti-inflationary monetary stance

taken by the Fed and other central banks around the world. DeLong (1997)

suggests that the inferior monetary policy regime of the pre-Volcker period

may have been due to the Fed believing that the natural rate of unemployment

was lower than it actually was during the 1970s. Clarida et al. (2000) suggest

another possibility. At that time ‘neither the Fed nor the economics profession

understood the dynamics of inflation very well. Indeed it was not until the midto-

late 1970s that intermediate textbooks began emphasising the absence of a

long-run trade-off between inflation and output. The ideas that expectations

matter in generating inflation and that credibility is important in policymaking

were simply not well established during that era’ (see also Taylor, 1997a;

Mayer, 1999; Romer and Romer, 2004). To understand the historical performance

of an economy over time it would seem imperative to have an understanding

of the policy maker’s knowledge during the time period under investigation.

Since a great deal of policy makers’ knowledge is derived from the research

findings of economists, the state of economists’ knowledge at each point in

history must always be taken into consideration when assessing economic

performance (Romer and Romer, 2002). Although it is a very important task of

economists to analyse and be critical of past policy errors, we should remember

that, as with all things, it is easy to be wise after the event.

While a consensus among new Keynesian economists would support the

new Keynesian style of monetary policy outlined above, there remain doubters.

For example, Stiglitz (1993, pp. 1069–70) prefers a more flexible approach

to policy making and argues:

Changing economic circumstances require changes in economic policy, and it is

impossible to prescribe ahead of time what policies would be appropriate … The

reality is that no government can stand idly by as 10, 15, or 20 percent of its

workers face unemployment … new Keynesian economists also believe that it is

virtually impossible to design rules that are appropriate in the face of a rapidly

changing economy.

7.11.4 Other policy implications

For those new Keynesians who have been developing various explanations of

real wage rigidity, a number of policy conclusions emerge which are aimed

specifically at reducing highly persistent unemployment (Manning, 1995;

Nickell, 1997, 1998). The work of Lindbeck and Snower (1988b) suggests

that institutional reforms are necessary in order to reduce the power of the

insiders and make outsiders more attractive to employers. Theoretically conceivable

power-reducing policies include:

1. a softening of job security legislation in order to reduce the hiring and

firing (turnover) costs of labour; and

2. reform of industrial relations in order to lessen the likelihood of strikes.

Policies that would help to ‘enfranchise’ the outsiders would include:

1. retraining outsiders in order to improve their human capital and marginal


2. policies which improve labour mobility; for example, a better-functioning

housing market;

3. profit-sharing arrangements which bring greater flexibility to wages;

4. redesigning of the unemployment compensation system so as to encourage

job search.

Weitzman (1985) has forcefully argued the case for profit-sharing schemes

on the basis that they offer a decentralized, automatic and market incentive

approach to encourage wage flexibility, which would lessen the impact of

macroeconomic shocks. Weitzman points to the experience of Japan, Korea

and Taiwan with their flexible payment systems which have enabled these

economies in the past to ride out the business cycle with relatively high

output and employment levels (see Layard et al., 1991, for a critique).

The distorting impact of the unemployment compensation system on unemployment

is recognized by many new Keynesian economists. A system

which provides compensation for an indefinite duration without any obligation

for unemployed workers to accept jobs offered seems most likely to

disenfranchise the outsiders and raise efficiency wages in order to reduce

shirking (Shapiro and Stiglitz, 1984). In the shirking model the equilibrium

level of involuntary unemployment will be increased if the amount of unemployment

benefit is raised. Layard et al. (1991) also favour reform of the

unemployment compensation system (see Atkinson and Micklewright, 1991,

for a survey of the literature).

Some new Keynesians (particularly the European branch) favour some

form of incomes policy to modify the adverse impact of an uncoordinated

wage bargaining system; for example, Layard et al. (1991) argue that ‘if

unemployment is above the long-run NAIRU and there is hysteresis, a temporary

incomes policy is an excellent way of helping unemployment return to

the NAIRU more quickly’ (see also Galbraith, 1997). However, such policies

remain extremely contentious and most new Keynesians (for example,

Mankiw) do not feel that incomes policies have a useful role to play.