Political Independence of Central Banks

The issue of political - or goal - independence of central banks, i.e. independence in setting their own goals,^[38] was first discussed within the realm of positive economic policy.

Bade and Parkin (1978) measured the impact of indepen­dence in twelve advanced countries, finding that most showed an association of higher independence of monetary policy with a smaller average government budget deficit. A number of later studies (ranging from Alesina (1988), to Grilli, Masciandaro and Tabellini (1991), Cukierman (1992) and Alesina and Summers (1993), to name only some of the first contributors) tried to refine the measurement of the degree of independence.

More recently, after studies by Fry et al. (2000) and Crowe and Meade (2008) and some previous investigations of their own, Dincer and Eichengreen (2014) measured trends in independence through time up to 2010 for more than 100 central banks, basing their measure on various indicators. These include political transparency as to policy objectives and their priorities, economic transparency referring to the information used for policy action (economic data, the model of the economy, etc.), procedural transparency and opera­tional transparency concerning specifically indications of methods for correcting imbalances between targets and achievements. The authors found that independence is high­est in the European Monetary Union (EMU) and some small countries and lower in the United Kingdom, the United States, Australia, India and other countries.

goals. Definitions are from Grilli, Masciandaro and Tabellini (1991) and Debelle and Fischer (1994). co-movement of transparency, which has risen through time in both advanced and less developed countries, and central bank independence. According to them, the former should tend to rise as a consequence of the need to ensure account­ability of independent policy authorities in a democratic society. However, while the need for transparency is parti­cularly evident in countries with strong political institu­tions, this is not rigorously true for independence.

In normative theory, as a first approximation, the status of independence and the issue of the identity of the central banker were not discussed until the 1980s. During that dec­ade, the literature generally favoured separation of the autho­rities governing monetary and fiscal policy. Blinder (1982a) suggested a reason in favour of independent authorities, stating that dispersion of power can be a defence against its misuse. Then Rogoff (1985a) applied the concept of indepen­dence of central banks to the issue of time inconsistency and suggested appointment of an independent central banker to cope with it. The possibility of setting monetary policy with­out interference or restriction could at least partially solve the inflation bias problem, since an independent central bank could decide the monetary policy stance without any political calculation of its effects on voters.

In addition, Rogoff (1985a) faced the issue of the optimal degree of conservatism of a central bank from the point of view of a social-welfare function. We deal with this in the next subsection. He kept the two issues, conservativeness and independence, separate, even if claiming that both are necessary to pursue a higher welfare. In contrast to Rogoff, almost all following papers have not made this distinction and treat central bank independence and conservatism as a joint or indistinct variable. In empirical or theoretical analyses, the two are usually represented by a unique parameter expressing the relative weight attributed to inflation and stabilisation in the central bank's objective function.

Possibly also under the influence of Rogoff (1985a) and the subsequent literature, in the 1980s and 1990s the number of central banks that were given a statute of political indepen­dence increased. But this was not an effect of free political decisions. Cukierman (1994) suggested the possibility of political and economic factors influencing the degree of legal independence. According to Posen (1995), there were underlying forces pushing towards independence, taken as an indicator of a relatively lower preference for inflation. Hayo (1998) found empirical evidence in favour of the idea that central bank independence alone cannot explain macro­economic performance and found some proof for the idea that the culture of people in support of low inflation might have favoured it, even if the possibility of reverse causation cannot be excluded and data availability did not allow appli­cation of the test of ‘Granger causality'. Political indepen­dence does not come at no cost, as it can involve a lower inflation rate and a higher unemployment rate, at least if monetary action is effective. Political independence was indeed supported by the financial sector, which has a low preference for inflation (Santoni 1986). This is a short way to saying with Posen that both political independence and inflation are the outcome of structural economic and social factors that result in the central bank statutes having no impact of their own - i.e., independently of these factors - on inflation. In fact, as we will see in Chapters 5 and 6, (vested) interests play an essential role in the formulation of policy action.

More recent papers have confirmed the previous dichot­omy in the orientation of the theoretical literature, with a prevalence of sceptical views on the advantages of independence.^[39] On the one hand, some maintain the idea of a positive effect of monetary policy independence on inflation and macroeconomic performance. Berger, de Haan and Eijffinger (2001) corroborate robustness of the negative correlation between the degree of central bank independence and inflation.

Others (e.g. Hayo and Hefeker 2002; Mishkin and Schmidt- Hebbel 2007) have confirmed the conclusions and empirical findings of Posen and Hayo. In particular, according to Hayo and Hefeker (2007), central bank independence does not encourage better monetary policy performance, as it is a condition neither necessary nor sufficient for monetary stability. Beyond all this, many central banks are only instru­ment independent, or it is very difficult to define whether instruments or political independence holds.^[40] Hayo and Hefeker (2007), after distinguishing between independence and conservativeness, suggest that if there is an issue of time inconsistency, there are also other and less costly solutions to it, such as inflation targets, fixed exchange rates and infla­tion contracts. Central bank independence is not a necessary condition for price stability, even if it can be judged as an appropriate solution for some countries.

On empirical grounds, some studies find a high positive correlation between independence and monetary stability, but others contradict this result. Moreover, correlation says nothing about causality. The correct way of dealing with the relationship between central bank independence and infla­tion would be to inquire at historical and political levels about the process leading to the decision about the impor­tance of price stability as a major economic policy objective, if this has been the case for a country.

The relevant topics on which reflection must be enter­tained also concern issues that are different from the simple impact of independence on inflation. On the one hand, Demertzis, Hughes Hallett and Viegi (2004) find that the pursuit of different targets by an independent central bank and elected governments generates a conflict, and both will be unable to reach their respective objectives, the more so the larger is the difference between their preferences. These con­flicts will drive fiscal policy to become more expansionary than is appropriate because of the need to offset/overcome the effects of more conservative monetary policies. This has consequences for efficiency and the long-run debt position, yet the endogenous voting extension of the model shows that the electorate will naturally vote for such outcomes.

Benigno and Woodford (2004) also underline this need for complementarity of monetary and fiscal policy. On the other hand, Woodford’s (2000) recognises the tendency of monetary policy to overreact to cost-push shocks, a kind of stabilisation bias under discretion relative to commitment arising even when no inflation bias is present, as inflation is not higher than the average inflation under commitment. Albanesi, Chari and Christiano (2002) show that a stabilisa­tion bias arises when monetary policy is the only policy tool to react to shocks. In these conditions, with discretionary monetary policy, expectation traps arise because the private sector tends to protect itself from inflation, which induces monetary authorities to react by supplying the expected level of inflation.

As seen in the preceding section, Di Bartolomeo, Tirelli and Acocella (2015) strengthen Benigno and Woodford’s (2004) argument that the optimal fiscal and monetary stabi­lisation policies should be seen as complements by referring to a different, novel mechanism, acting when public transfers are introduced into the analysis. Thus, the monetary author­ity should consider the consequences of its actions for the government budget. In this regard, a substantial amount of inflation volatility is indeed desirable to deflate nominal debt and to limit the accumulation of real debt and stabilise debt/ GDP ratios in the long run.

A final reason for coordinating fiscal and monetary action derives from the mixed nature of some innovative unconven­tional monetary policies. We deal with this in Chapter 5. Arestis (2015) suggests coordination of fiscal policy not only with monetary policy but also with financial stability policies in order to reduce unemployment and income inequality.

3.7