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


5.5.4 Central bank independence

The debate relating to central bank independence (CBI) has been very much

influenced by new classical thinking, especially with respect to inflationary

expectations, time inconsistency, reputation and credibility. If we accept the

Kydland–Prescott argument that discretionary policies lead to an inflation

bias, then it is clearly necessary to establish some institutional foundation

that will constrain discretionary actions. Many economists are persuaded that

some form of CBI will provide the necessary restraint. The theoretical case

for CBI relates to the general acceptance of the natural rate hypothesis that in

the long run the rate of inflation is independent of the level of unemployment

and that discretionary policies are likely to lead to an inflation bias. Hence

with no long-run exploitable trade-off, far-sighted monetary authorities ought

to select a position on the long-run Phillips curve consistent with a low

sustainable rate of inflation (a point near to O in Figure 5.4). The dynamic

inconsistency theories of inflation initiated by Kydland and Prescott and

developed by Barro and Gordon, and Backus and Driffill, provide an explanation

of why excessive (moderate) inflation will be the likely outcome of a

monetary regime where long-term commitments are precluded. Such discretionary

regimes contrast sharply with monetary regimes such as the Gold

Standard, where the underlying rules of the game revolve around a precommitment

to price stability. The emphasis of these models on the importance

of institutions and rules for maintaining price stability provides a strong case

for the establishment of independent central banks whose discretion is constrained

by explicit anti-inflation objectives acting as a pre-commitment device.

Since the problem of credibility has its source in the discretionary powers of

the monetary authorities with respect to the conduct of monetary policy, this

could be overcome by transferring the responsibility for anti-inflationary

policy to a non-political independent central bank. In addition, an independent

central bank will benefit from a ‘credibility bonus’, whereby disinflationary

policies can be accomplished at a low ‘sacrifice ratio’ (Cukierman, 1992;

Goodhart, 1994a, 1994b).

In the debate over CBI it is important to make a distinction between ‘goal

independence’ and ‘instrument independence’ (see Fischer, 1995a, 1995b).

The former implies that the central bank sets its own objectives (that is,

political independence), while the latter refers to independence with respect

to the various levers of monetary policy (that is, economic independence).

The recently (May 1997) created ‘independent’ Bank of England has instrument

independence only. Initially, an inflation target of 2.5 per cent was set

by government, which formed the Bank’s explicitly stated monetary policy

objective. Therefore, in the UK, the decisions relating to goals remain in the

political sphere (Bean, 1998; Budd, 1998).

As noted above, Svensson (1997a) argues that the inflation bias associated

with the time-inconsistency problem can be improved upon by

‘modifying central bank preferences’ via delegation of monetary policy to a

‘conservative central banker’ (for example Alan Greenspan) as suggested

by Rogoff (1985) or by adopting optimal central bank contracts, as suggested

by Walsh (1993, 1995a). Rogoff’s conservative central banker has

both goal and instrument independence and is best represented by the

German Bundesbank, which before European Monetary Union remained

the most independent central bank in Europe (Tavelli et al., 1998). In

Rogoff’s model an inflation-averse conservative central banker is appointed

who places a higher relative weight on the control of inflation than does

society in general (for example President Jimmy Carter’s appointment of

Paul Volcker as Chairman of the Fed in 1979). This is meant to ensure that

the excessive inflation associated with the time-inconsistency problem is

kept low in circumstances where it would otherwise be difficult to establish

a pre-commitment to low inflation. Overall, lower average inflation and

higher output variability are predicted from this model (Waller and Walsh,

1996). However, the research of Alesina and Summers (1993) shows that

only the first of these two predictions appears in cross-sectional data. In

contrast, Hutchison and Walsh (1998), in a recent study of the experience of

New Zealand, find that central bank reform appears to have increased the

short-run output–inflation trade-off. In Rogoff’s model the conservative

central banker reacts less to supply shocks than someone who shared society’s

preferences, indicating a potential trade-off between flexibility and

commitment. In response to this problem Lohmann (1992) suggests that the

design of the central bank institution should involve the granting of partial

independence to a conservative central banker who places more weight on

inflation than the policy maker, but ‘the policymaker retains the option to

over-ride the central bank’s decisions at some strictly positive but finite

cost’. Such a clause has been built into the Bank of England Act (1998),

where the following reserve power is set out: ‘The Treasury, after consultation

with the Governor of the Bank, may by order give the Bank directions

with respect to monetary policy if they are satisfied that the directions are

required in the public interest and by extreme economic circumstances.’ It

remains to be seen if such powers are ever used.

The contracting model, associated with Walsh (1993, 1995a, 1998), utilizes

a principal–agent framework and emphasizes the accountability of the

central bank. In Walsh’s contracting approach the central bank has instrument

independence but no goal independence. The central bank’s rewards and

penalties are based on its achievements with respect to inflation control. The

Reserve Bank of New Zealand resembles this principal–agent type model. An

important issue in the contracting approach is the optimal length of contract

for a central banker (Muscatelli, 1998). Long terms of appointment will

reduce the role of electoral surprises as explained in Alesina’s partisan model

(see Chapter 10). But terms of office that are too long may be costly if

societal preferences are subject to frequent shifts. Waller and Walsh (1996)

argue that the optimal term length ‘must balance the advantages in reducing

election effects with the need to ensure that the preferences reflected in

monetary policy are those of the voting public’.

The empirical case for CBI is linked to cross-country evidence which shows

that for advanced industrial countries there is a negative relationship between

CBI and inflation. During the last 15 years a considerable amount of research

has been carried out which has examined the relationship between central bank

independence and economic performance (see Grilli et al., 1991; Bernanke and

Mishkin, 1992; Alesina and Summers, 1993; Eijffinger and Schaling, 1993;

Bleaney, 1996; Eijffinger, 2002a, 2002b). The central difficulty recognized by

researchers into the economic impact of central bank independence is the

problem of constructing an index of independence. Alesina and Summers

(1993) identify the ability of the central bank to select its policy objectives

without the influence of government, the selection procedure of the governor of

the central bank, the ability to use monetary instruments without restrictions

and the requirement of the central bank to finance fiscal deficits as key indicators

that can be used to construct a measure of central bank independence.

Using a composite index derived from Parkin and Bade (1982a) and Grilli et al.

(1991), Alesina and Summers examined the correlation between an index of

independence and some major economic indicators. Table 5.2 indicates that,

‘while central bank independence promotes price stability, it has no measurable

impact on real economic performance’ (Alesina and Summers, 1993, p. 151).

Table 5.2 Central bank independence and economic performance

Country Average index Average Average Average

of central bank inflation unemployment real GNP

independence 1955–88 rate growth

1958–88 1955–87

Spain 1.5 8.5 n/a 4.2

New Zealand 1 7.6 n/a 3.0

Australia 2.0 6.4 4.7 4.0

Italy 1.75 7.3 7.0 4.0

United Kingdom 2 6.7 5.3 2.4

France 2 6.1 4.2 3.9

Denmark 2.5 6.5 6.1 3.3

Belgium 2 4.1 8.0 3.1

Norway 2 6.1 2.1 4.0

Sweden 2 6.1 2.1 2.9

Canada 2.5 4.5 7.0 4.1

Netherlands 2.5 4.2 5.1 3.4

Japan 2.5 4.9 1.8 6.7

United States 3.5 4.1 6.0 3.0

Germany 4 3.0 3.6 3.4

Switzerland 4 3.2 n/a 2.7

Source: Alesina and Summers (1993).

Source: Alesina and Summers (1993).

Figure 5.6 The relationship between average inflation and central bank


The ‘near perfect’ negative correlation between inflation and central bank

independence is clearly visible in Figure 5.6. However, as Alesina and Summers

recognize, correlation does not prove causation, and the excellent

anti-inflationary performance of Germany may have more to do with the public

aversion to inflation following the disastrous experience of the hyperinflation in

1923 than the existence of an independent central bank. In this case the independent

central bank could be an effect of the German public aversion to

inflation rather than a cause of low inflation. Indeed, the reputation established

by the German Bundesbank for maintaining low inflation was one important

reason given by the UK government for joining the ERM in October 1990. The

participation of the UK in such a regime, where monetary policy is determined

by an anti-inflationary central bank which has an established reputation and

credibility, was intended to tie the hands of domestic policy makers and help

lower inflationary expectations (see Alogoskoufis et al., 1992).

Considerable research has also been conducted into the role played by

politics in influencing economic performance. The ‘political business cycle’

or ‘monetary politics’ literature also suggests that CBI would help reduce the

problem of political distortions in macroeconomic policy making. What is

now known as the ‘new political macroeconomics’ has been heavily influ262

enced by the research of Alberto Alesina. His work has shown that the

imposition of rational expectations does not remove the importance of political

factors in business cycle analysis and in general the political business

cycle literature provides more ammunition to those economists who favour

taking monetary policy out of the hands of elected politicians. Excellent

surveys of the monetary politics literature can be found in Alesina and Roubini

with Cohen (1997) and Drazen (2000a); see Chapter 10.

While CBI might avoid the dynamic time-inconsistency problems identified

by Kydland and Prescott and produce lower average rates of inflation,

many economists doubt that, overall, a rules-bound central bank will perform

better than a central bank that is allowed to exercise discretion given the

possibility of large unforeseen shocks. A central bank which could exercise

discretion in the face of large shocks may be a more attractive alternative to

rule-based policies. This is certainly the view of Keynesians such as Stiglitz,

Solow and Tobin (see Solow and Taylor, 1998; Tobin, 1998; Stiglitz, 1999a).

Other economists who have worked on the inside of major central banks,

such as Blinder (1997b, 1998) at the US Fed, and Goodhart (1994a) at the

Bank of England, are not convinced of the usefulness or realism of the gametheoretic

approach to central bank behaviour. The research of Bernanke and

Mishkin also confirms that ‘Central banks never and nowhere adhere to strict,

ironclad rules for monetary growth’ (Bernanke and Mishkin, 1992, p. 186).

One of the most important theoretical objections to CBI is the potential for

conflict that it generates between the monetary and fiscal authorities (Doyle

and Weale, 1994; Nordhaus, 1994). It is recognized that the separation of

fiscal and monetary management can lead to coordination problems which

can undermine credibility. In countries where this has led to conflict (such as

the USA in the period 1979–82) large fiscal deficits and high real interest

rates have resulted. This monetary/fiscal mix is not conducive to growth and,

during the early period of Reaganomics in the USA, came in for severe

criticism from many economists (Blanchard, 1986; Modigliani, 1988b; and

Tobin, 1987). The tight-monetary easy-fiscal mix is hardly a surprising combination

given the predominant motivations that drive the Fed and the US

Treasury. Whereas independent central banks tend to emphasize monetary

austerity and low inflation, the fiscal authorities (politicians) know that increased

government expenditure and reduced taxes are the ‘meat, potatoes

and gravy of politics’ (Nordhaus, 1994). To the critics CBI is no panacea. In

particular, to say that inflation should be the primary goal of the central bank

is very different from making inflation the sole goal of monetary policy in all

circumstances (Akhtar, 1995; Carvalho, 1995/6; Minford, 1997; Forder, 1998;

Posen, 1998). As Blackburn (1992) concludes, ‘the credibility of monetary

policy does not depend upon monetary policy alone but also upon the macroeconomic

programme in its entirety’.