Policy Implications of Politico-Economic Models: An
Independent Central Bank?
In introducing ‘The New Monetary Policy Framework’ for the UK economy
on 6 May 1997, which established ‘operational independence’ for the Bank
of England, Chancellor Gordon Brown, in an official statement, provided the
following rationale for the government’s strategy (Brown, 1997, emphasis
added):
We will only build a fully credible framework for monetary policy if the longterm
needs of the economy, not short-term political considerations guide monetary
decision-making. We must remove the suspicion that short-term party political
considerations are influencing the setting of interest rates.
Chancellor Brown’s decision to grant much greater independence to the Bank
of England had its origins in a 1992 Fabian Society paper entitled ‘Euro
Monetarism’, written by Ed Balls, Brown’s economic adviser. As a former
student of Larry Summers at Harvard, Balls was familiar with the empirical
work on central bank independence produced by Alesina and Summers (1993).
In a visit to the USA in March 1997, Shadow Chancellor Brown and his
economic adviser met both Alan Greenspan and Larry Summers. And so was
born the strategy to go for immediate greater central bank independence if
elected.
The general debate on the relative merits of rules versus discretion in the
conduct of fiscal and monetary policy was given a new stimulus by the
research surveyed in Chapters 5 and 7. Since the non-optimal use of monetary
and fiscal instruments lies at the heart of the various strands of the
political business cycle literature, most of this work points towards the desirability
of establishing a policy regime which curtails the incentives policy
makers have to engage in destabilizing policies. The new classical contributions
of Kydland and Prescott (1977) and Barro and Gordon (1983a), which
highlighted time-inconsistency, credibility and reputational issues, have provided
extra weight to the case for monetary rules associated with the work of
Friedman (1968a). The politico-economic literature has also shown how strong
partisan or opportunistic behaviour can generate a non-optimal outcome for
aggregate variables. However, in order to make policy rules credible, some
sort of enforcement mechanism is required. For this reason many economists
in recent years have argued in favour of institutional reform involving the
establishment of an independent central bank (see Goodhart, 1994a, 1994b).
The assumption lying behind this argument is that such an institution (at least
in principle) is capable of conducting monetary policy in a manner free from
opportunistic and partisan influences. In addition, fiscal policy will also be
subject to a harder budget constraint, providing the independent central bank
is not obliged to monetize deficits (see Alesina and Perotti, 1995a). Greater
central bank independence was also one of the objectives contained in the
Maastricht Treaty, which sought to bring about some fundamental changes in
national banking legislation in anticipation of European monetary union (see
Walsh, 1995a, 1995b).
The case for central bank independence is usually framed in terms of the
inflation bias present in the conduct of monetary policies. Such an inflation
bias is evident from the relatively high rates of inflation experienced in
industrial countries in the 1970s and early 1980s. Because a majority of
economists emphasize monetary growth as the underlying cause of sustained
inflation (see Lucas, 1996), it follows that prolonged differences in countries’
rates of inflation result from variations in their rates of monetary expansion.
Any plausible explanation of these ‘stylized facts’ must therefore include an
understanding of central bank behaviour (Walsh, 1993). In particular, we
need to identify the reasons why monetary policy is conducted in a way that
creates a positive average rate of inflation which is higher than desirable.
There are several reasons why monetary authorities may generate inflation.
These include political pressures to lower unemployment in order to influence
re-election prospects, partisan effects as emphasized by Hibbs and
Alesina, dynamic inconsistency influences and motivations related to the
financing of deficits. The last is particularly important in economies with
inefficient or underdeveloped fiscal systems (see Cukierman, 1994).
The theoretical case for central bank independence in industrial democracies
relates to a general acceptance that the long-run Phillips curve is vertical
at the natural rate of unemployment. This implies that although monetary
policy is non-neutral in the short run, it has little effect on real variables such
as unemployment and output in the long run. With no exploitable long-run
trade-off, far-sighted monetary authorities ought to select a position on the
vertical Phillips curve consistent with a sustainable objective of price stability
(see Goodhart, 1994b; Cukierman, 1994).
The empirical case for central bank independence is linked to cross-country
evidence which shows that for advanced industrial countries there is a
clear negative relationship between central bank independence and inflation
(see Grilli et al., 1991; Cukierman, 1992; Alesina and Summers, 1993;
Eijffinger and Keulen, 1995; Eijffinger, 2002a). It should be noted, however,
that this negative correlation does not prove causation and fails to hold for a
larger sample of countries which includes those from the developing world
(see Jenkins, 1996). Poor countries with shallow financial markets and unsustainable
budget deficits are unlikely to solve their inflation problems by
relying on the creation of an independent central bank (Mas, 1995). However,
at least as far as advanced industrial democracies are concerned, the theoretical
and empirical work suggests that monetary constitutions should be designed
to ensure a high degree of central bank autonomy.
Although the success of the independent German Bundesbank in delivering
low inflation over a long period of time inspired other countries to follow
its example, it is also clear that some important problems emerge with the
measurement, form and consequences of central bank independence. First,
the whole question of independence is one of degree. Although before May
1997 the Federal Reserve had much more independence than the Bank of
England, it is clear that legal independence does not (and cannot) completely
remove the influence of ‘monetary politics’ (see Mayer, 1990; Havrilesky,
1993; Woolley, 1994). For example, Chappell et al. (1993) show how partisan
influences on the conduct of monetary policy can arise through presidential
appointments to the Board of Governors of the Federal Reserve. Through
these ‘political’ appointments and other forms of presidential signalling (moral
suasion) the Federal Reserve’s monetary policy making can never be totally
independent of political pressures. Nevertheless, considerable research effort
has been made in recent years to measure the extent of central bank independence
in a large number of countries (see Cukierman, 1992; Eijffinger
and Keulen, 1995; Healey, 1996). According to Cukierman, four sets of
indices can be used to identify the degree of independence of a central bank:
(i) legal indices; (ii) questionnaire-based indices; (iii) the turnover of central
bank governors; and (iv) the political vulnerability of the bank. However,
such studies and their implications have come in for considerable criticism
(see Jenkins, 1996).
In discussing the form of central bank independence, Fischer (1995a,
1995b) introduces the distinction between ‘goal independence’ and ‘instrument
independence’. The former implies that the central bank sets its own
policy objectives (that is, political independence) while the latter refers to
independence with respect to the various levers of monetary policy (that is,
economic independence). Using this framework the Bank of England was
granted instrument (economic) independence but not goal (political) independence
in May 1997. The distinction between goal and instrument
independence can be used to illustrate the difference between the two main
models of independent central banks which have been developed in the
theoretical literature. The first model is based on Rogoff’s (1985) ‘conservative
central banker’. In this model an inflation-averse conservative central
banker is appointed who ensures that inflation is kept low in circumstances
where it would otherwise be difficult to establish a pre-commitment to low
inflation. Rogoff’s inflation-averse central banker has both goal and instrument
independence. The result is lower average inflation but higher output
variability. The second model, associated with Walsh (1995b), 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, and the central bank’s rewards and
penalties are based on its achievements with respect to inflation control.
However as Walsh (1995a) notes:
An inflation-based contract, combined with central bank independence in the
actual implementation of policy, achieves optimal policy outcomes only if the
central bank shares social values in trading off unemployment and inflation. When
the central bank does not share society’s preferences, the optimal contract is no
longer a simple function of inflation; more complicated incentives must be generated
to ensure that the central bank maintains low average inflation while still
engaging in appropriate stabilization policies.
While the New Zealand Federal Reserve Bank, which was set up following the
1990 reforms, resembles the principal–agent model, the German Bundesbank,
before EMU, comes close to the conservative central banker of the Rogoff
model. In Fischer’s (1995a) view, the important conclusion to emerge from this
literature is that ‘a central bank should have instrument independence, but
should not have goal independence’.
In recent years many countries have set about reforming the institutions of
monetary policy. Such countries include those of the former ‘Eastern bloc’ of
communist economies as well as those from Latin America and Western Europe.
Most have adopted some variant of the principal–agent approach whereby
the central bank is contracted to achieve clearly defined goals, provided with
the instruments to achieve these desired objectives, and held accountable for
deviating from the chosen path (see Bernanke and Mishkin, 1992; Walsh
1995a). In the New Zealand model the central bank governor is accountable to
the finance minister. This contrasts with the German Bundesbank, which was
held accountable to the public. In both Canada and the UK, emphasis has been
placed on inflation targeting. Failure to meet inflation targets involves a loss of
reputation for the central bank. In the case of the UK, the approach adopted
since 1992 has involved inflation targeting combined (since May 1997) with
operational independence for the Bank of England (see Chapter 7).
One of the main theoretical objections to central bank independence is the
potential for conflict that this gives rise to between the monetary and fiscal
authorities (see Doyle and Weale, 1994). An extensive discussion of the
problems faced by policy makers in countries where monetary and fiscal
policies are carried out independently is provided by Nordhaus (1994). In
countries where this has led to conflict (for example in the USA during the
period 1979–82) large fiscal deficits and high real interest rates have frequently
resulted. According to Nordhaus this leads to a long-run rate of
growth which is too low. The tight monetary–easy fiscal mix is hardly surprising
given the predominant motivations driving the ‘Fed’ and the ‘Treasury’
in the USA. Whereas independent central banks emphasize monetary austerity
and low stable rates of inflation, the fiscal authorities know that increased
government expenditure and reduced taxes are the ‘meat, potatoes and gravy
of politics’ (Nordhaus, 1994). In this scenario the economy is likely to be
locked into a ‘high-deficit equilibrium’. Self-interested politicians are unlikely
to engage in deficit reduction for fear of losing electoral support. These
coordination problems which a non-cooperative fiscal–monetary game generates
arise inevitably from a context where the fiscal and monetary authorities
have different tastes with respect to inflation. At the end of the day these
problems raise a fundamental issue. Should a group of unelected individuals
be allowed to make choices on the use of important policy instruments which
will have significant repercussions for the citizens of a country? In short,
does the existence of an independent central bank threaten democracy (see
Stiglitz, 1999a)? What is clear is that independence without democratic accountability
is unacceptable (Eijffinger, 2002b).
There is now an extensive literature related to the issue of central bank
independence. The academic literature has pointed to the various reasons
why industrial democracies have developed an inflation bias in which governments
are allowed discretion in the operation of fiscal and monetary
policies. In contrast to Rogoff (1985), Alesina and Summers (1993) point to
the empirical evidence which, for advanced industrial countries, shows that
inflation is negatively correlated with the degree of central bank independence
without this having significant adverse effects on real growth and
employment in the long run. Central bank independence seems to offer a
‘free lunch’! However, this issue is made more complicated by the causes of
output variability. Alesina and Gatti (1995) distinguish between two types of
variability which can contribute to aggregate instability. The first is economic
variability resulting from different types of exogenous shocks to aggregate
demand and/or supply. The second type is political or policy-induced and has
been the subject of this chapter. Rogoff’s conservative central banker does
not react much to ‘economic shocks’. However, Alesina and Gatti (1995)
argue that an independent central bank will reduce policy-induced output
variability. Hence the ‘overall effect of independence on output induced
variability is, thus, ambiguous’. It follows that in cases where policy-induced
variability exceeds that resulting from exogenous shocks, a more independent
central bank can reduce inflation and the variance of output, a result consistent
with the Alesina and Summers (1993) data. To the critics, central bank
independence is no panacea and the Alesina–Summers correlations do not
prove causation (Posen, 1995). In addition, the numerous domestic and international
policy coordination problems which independence can give rise to
could outweigh the potential benefits. In the face of powerful ‘economic’
shocks a conservative and independent central bank may not be superior to an
elected government.
Independent Central Bank?
In introducing ‘The New Monetary Policy Framework’ for the UK economy
on 6 May 1997, which established ‘operational independence’ for the Bank
of England, Chancellor Gordon Brown, in an official statement, provided the
following rationale for the government’s strategy (Brown, 1997, emphasis
added):
We will only build a fully credible framework for monetary policy if the longterm
needs of the economy, not short-term political considerations guide monetary
decision-making. We must remove the suspicion that short-term party political
considerations are influencing the setting of interest rates.
Chancellor Brown’s decision to grant much greater independence to the Bank
of England had its origins in a 1992 Fabian Society paper entitled ‘Euro
Monetarism’, written by Ed Balls, Brown’s economic adviser. As a former
student of Larry Summers at Harvard, Balls was familiar with the empirical
work on central bank independence produced by Alesina and Summers (1993).
In a visit to the USA in March 1997, Shadow Chancellor Brown and his
economic adviser met both Alan Greenspan and Larry Summers. And so was
born the strategy to go for immediate greater central bank independence if
elected.
The general debate on the relative merits of rules versus discretion in the
conduct of fiscal and monetary policy was given a new stimulus by the
research surveyed in Chapters 5 and 7. Since the non-optimal use of monetary
and fiscal instruments lies at the heart of the various strands of the
political business cycle literature, most of this work points towards the desirability
of establishing a policy regime which curtails the incentives policy
makers have to engage in destabilizing policies. The new classical contributions
of Kydland and Prescott (1977) and Barro and Gordon (1983a), which
highlighted time-inconsistency, credibility and reputational issues, have provided
extra weight to the case for monetary rules associated with the work of
Friedman (1968a). The politico-economic literature has also shown how strong
partisan or opportunistic behaviour can generate a non-optimal outcome for
aggregate variables. However, in order to make policy rules credible, some
sort of enforcement mechanism is required. For this reason many economists
in recent years have argued in favour of institutional reform involving the
establishment of an independent central bank (see Goodhart, 1994a, 1994b).
The assumption lying behind this argument is that such an institution (at least
in principle) is capable of conducting monetary policy in a manner free from
opportunistic and partisan influences. In addition, fiscal policy will also be
subject to a harder budget constraint, providing the independent central bank
is not obliged to monetize deficits (see Alesina and Perotti, 1995a). Greater
central bank independence was also one of the objectives contained in the
Maastricht Treaty, which sought to bring about some fundamental changes in
national banking legislation in anticipation of European monetary union (see
Walsh, 1995a, 1995b).
The case for central bank independence is usually framed in terms of the
inflation bias present in the conduct of monetary policies. Such an inflation
bias is evident from the relatively high rates of inflation experienced in
industrial countries in the 1970s and early 1980s. Because a majority of
economists emphasize monetary growth as the underlying cause of sustained
inflation (see Lucas, 1996), it follows that prolonged differences in countries’
rates of inflation result from variations in their rates of monetary expansion.
Any plausible explanation of these ‘stylized facts’ must therefore include an
understanding of central bank behaviour (Walsh, 1993). In particular, we
need to identify the reasons why monetary policy is conducted in a way that
creates a positive average rate of inflation which is higher than desirable.
There are several reasons why monetary authorities may generate inflation.
These include political pressures to lower unemployment in order to influence
re-election prospects, partisan effects as emphasized by Hibbs and
Alesina, dynamic inconsistency influences and motivations related to the
financing of deficits. The last is particularly important in economies with
inefficient or underdeveloped fiscal systems (see Cukierman, 1994).
The theoretical case for central bank independence in industrial democracies
relates to a general acceptance that the long-run Phillips curve is vertical
at the natural rate of unemployment. This implies that although monetary
policy is non-neutral in the short run, it has little effect on real variables such
as unemployment and output in the long run. With no exploitable long-run
trade-off, far-sighted monetary authorities ought to select a position on the
vertical Phillips curve consistent with a sustainable objective of price stability
(see Goodhart, 1994b; Cukierman, 1994).
The empirical case for central bank independence is linked to cross-country
evidence which shows that for advanced industrial countries there is a
clear negative relationship between central bank independence and inflation
(see Grilli et al., 1991; Cukierman, 1992; Alesina and Summers, 1993;
Eijffinger and Keulen, 1995; Eijffinger, 2002a). It should be noted, however,
that this negative correlation does not prove causation and fails to hold for a
larger sample of countries which includes those from the developing world
(see Jenkins, 1996). Poor countries with shallow financial markets and unsustainable
budget deficits are unlikely to solve their inflation problems by
relying on the creation of an independent central bank (Mas, 1995). However,
at least as far as advanced industrial democracies are concerned, the theoretical
and empirical work suggests that monetary constitutions should be designed
to ensure a high degree of central bank autonomy.
Although the success of the independent German Bundesbank in delivering
low inflation over a long period of time inspired other countries to follow
its example, it is also clear that some important problems emerge with the
measurement, form and consequences of central bank independence. First,
the whole question of independence is one of degree. Although before May
1997 the Federal Reserve had much more independence than the Bank of
England, it is clear that legal independence does not (and cannot) completely
remove the influence of ‘monetary politics’ (see Mayer, 1990; Havrilesky,
1993; Woolley, 1994). For example, Chappell et al. (1993) show how partisan
influences on the conduct of monetary policy can arise through presidential
appointments to the Board of Governors of the Federal Reserve. Through
these ‘political’ appointments and other forms of presidential signalling (moral
suasion) the Federal Reserve’s monetary policy making can never be totally
independent of political pressures. Nevertheless, considerable research effort
has been made in recent years to measure the extent of central bank independence
in a large number of countries (see Cukierman, 1992; Eijffinger
and Keulen, 1995; Healey, 1996). According to Cukierman, four sets of
indices can be used to identify the degree of independence of a central bank:
(i) legal indices; (ii) questionnaire-based indices; (iii) the turnover of central
bank governors; and (iv) the political vulnerability of the bank. However,
such studies and their implications have come in for considerable criticism
(see Jenkins, 1996).
In discussing the form of central bank independence, Fischer (1995a,
1995b) introduces the distinction between ‘goal independence’ and ‘instrument
independence’. The former implies that the central bank sets its own
policy objectives (that is, political independence) while the latter refers to
independence with respect to the various levers of monetary policy (that is,
economic independence). Using this framework the Bank of England was
granted instrument (economic) independence but not goal (political) independence
in May 1997. The distinction between goal and instrument
independence can be used to illustrate the difference between the two main
models of independent central banks which have been developed in the
theoretical literature. The first model is based on Rogoff’s (1985) ‘conservative
central banker’. In this model an inflation-averse conservative central
banker is appointed who ensures that inflation is kept low in circumstances
where it would otherwise be difficult to establish a pre-commitment to low
inflation. Rogoff’s inflation-averse central banker has both goal and instrument
independence. The result is lower average inflation but higher output
variability. The second model, associated with Walsh (1995b), 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, and the central bank’s rewards and
penalties are based on its achievements with respect to inflation control.
However as Walsh (1995a) notes:
An inflation-based contract, combined with central bank independence in the
actual implementation of policy, achieves optimal policy outcomes only if the
central bank shares social values in trading off unemployment and inflation. When
the central bank does not share society’s preferences, the optimal contract is no
longer a simple function of inflation; more complicated incentives must be generated
to ensure that the central bank maintains low average inflation while still
engaging in appropriate stabilization policies.
While the New Zealand Federal Reserve Bank, which was set up following the
1990 reforms, resembles the principal–agent model, the German Bundesbank,
before EMU, comes close to the conservative central banker of the Rogoff
model. In Fischer’s (1995a) view, the important conclusion to emerge from this
literature is that ‘a central bank should have instrument independence, but
should not have goal independence’.
In recent years many countries have set about reforming the institutions of
monetary policy. Such countries include those of the former ‘Eastern bloc’ of
communist economies as well as those from Latin America and Western Europe.
Most have adopted some variant of the principal–agent approach whereby
the central bank is contracted to achieve clearly defined goals, provided with
the instruments to achieve these desired objectives, and held accountable for
deviating from the chosen path (see Bernanke and Mishkin, 1992; Walsh
1995a). In the New Zealand model the central bank governor is accountable to
the finance minister. This contrasts with the German Bundesbank, which was
held accountable to the public. In both Canada and the UK, emphasis has been
placed on inflation targeting. Failure to meet inflation targets involves a loss of
reputation for the central bank. In the case of the UK, the approach adopted
since 1992 has involved inflation targeting combined (since May 1997) with
operational independence for the Bank of England (see Chapter 7).
One of the main theoretical objections to central bank independence is the
potential for conflict that this gives rise to between the monetary and fiscal
authorities (see Doyle and Weale, 1994). An extensive discussion of the
problems faced by policy makers in countries where monetary and fiscal
policies are carried out independently is provided by Nordhaus (1994). In
countries where this has led to conflict (for example in the USA during the
period 1979–82) large fiscal deficits and high real interest rates have frequently
resulted. According to Nordhaus this leads to a long-run rate of
growth which is too low. The tight monetary–easy fiscal mix is hardly surprising
given the predominant motivations driving the ‘Fed’ and the ‘Treasury’
in the USA. Whereas independent central banks emphasize monetary austerity
and low stable rates of inflation, the fiscal authorities know that increased
government expenditure and reduced taxes are the ‘meat, potatoes and gravy
of politics’ (Nordhaus, 1994). In this scenario the economy is likely to be
locked into a ‘high-deficit equilibrium’. Self-interested politicians are unlikely
to engage in deficit reduction for fear of losing electoral support. These
coordination problems which a non-cooperative fiscal–monetary game generates
arise inevitably from a context where the fiscal and monetary authorities
have different tastes with respect to inflation. At the end of the day these
problems raise a fundamental issue. Should a group of unelected individuals
be allowed to make choices on the use of important policy instruments which
will have significant repercussions for the citizens of a country? In short,
does the existence of an independent central bank threaten democracy (see
Stiglitz, 1999a)? What is clear is that independence without democratic accountability
is unacceptable (Eijffinger, 2002b).
There is now an extensive literature related to the issue of central bank
independence. The academic literature has pointed to the various reasons
why industrial democracies have developed an inflation bias in which governments
are allowed discretion in the operation of fiscal and monetary
policies. In contrast to Rogoff (1985), Alesina and Summers (1993) point to
the empirical evidence which, for advanced industrial countries, shows that
inflation is negatively correlated with the degree of central bank independence
without this having significant adverse effects on real growth and
employment in the long run. Central bank independence seems to offer a
‘free lunch’! However, this issue is made more complicated by the causes of
output variability. Alesina and Gatti (1995) distinguish between two types of
variability which can contribute to aggregate instability. The first is economic
variability resulting from different types of exogenous shocks to aggregate
demand and/or supply. The second type is political or policy-induced and has
been the subject of this chapter. Rogoff’s conservative central banker does
not react much to ‘economic shocks’. However, Alesina and Gatti (1995)
argue that an independent central bank will reduce policy-induced output
variability. Hence the ‘overall effect of independence on output induced
variability is, thus, ambiguous’. It follows that in cases where policy-induced
variability exceeds that resulting from exogenous shocks, a more independent
central bank can reduce inflation and the variance of output, a result consistent
with the Alesina and Summers (1993) data. To the critics, central bank
independence is no panacea and the Alesina–Summers correlations do not
prove causation (Posen, 1995). In addition, the numerous domestic and international
policy coordination problems which independence can give rise to
could outweigh the potential benefits. In the face of powerful ‘economic’
shocks a conservative and independent central bank may not be superior to an
elected government.
No comments:
Post a Comment