RDP 2004-04: Inflation Convergence Across Countries 4. Policy as an Explanation for Inflation Convergence

Given the evidence presented in Section 3 suggesting that inflation convergence is not stable property, then perhaps there are particular factors that can explain why the phenomenon occurs at particular points in time but not others. In particular, an alternative explanation for Ball and Sheridan's empirical result could be that strong inflation convergence in the 1990s is a manifestation of the actions of policy-makers. More specifically, it could be that inflation convergence was brought about by the goals of monetary policy becoming more similar across countries. This alternative explanation is discussed in more detail below in the context of industrialised countries.

Figure 4 plots the simple-average inflation rates for the group of countries that eventually adopted inflation targeting and 13 OECD countries that did not adopt the regime, as defined by Ball and Sheridan, as well as the difference between these two averages. The data show that average inflation rates for the two groups tracked each other quite closely through the 1960s, a decade of relatively low and stable inflation. The monetary policy regime in place at that time was the Bretton Woods system of fixed exchange rates, which came to an end in 1971. The first oil-price shock followed shortly afterwards, and the inflation performance of these two groups diverged significantly with the average inflation rate for inflation targeters peaking at over 5 percentage points above that of the non-inflation-targeting group soon after the initial shock. The reason for this divergence is likely to have been a combination of differences in structural factors and the reaction of policy to the oil-price shock between the two groups. What followed the collapse of the Bretton Woods system was a move to generalised floating exchange rates, and a number of OECD countries (including Australia, Canada and the United Kingdom) adopted a form of monetary targeting.^[16],[17] It is obvious from Figure 4 that the eventual inflation targeters had much less success in this regime compared to the non-inflation targeters, although other countries such as Germany, Japan and Switzerland (non-inflation-targeting countries) were successful in producing low inflation through the use of monetary targeting (Argy, Brennan and Stevens 1989; Mishkin 1999).^[18] Also, a large number of the non-inflation-targeting OECD countries considered here took part in the European Community's exchange rate policies, which is likely to have promoted some convergence of their inflation rates with Germany.

While the eventual inflation-targeting countries had some success at closing the gap during the subsequent decade, with the targeters briefly having lower inflation around 1984, this coincided with the OECD-wide recession that occurred in the early 1980s. After the mid 1980s, the unfavourable gap opened up again as output growth recovered, suggesting that the eventual inflation-targeters' policies were not as effective as those of the other group in controlling inflation. By the mid 1980s, it had become clear to policy-makers that the relationship between monetary aggregates, inflation and nominal income was not providing an accurate enough guide for conducting policy, which led to Australia, Canada and the United Kingdom to abandon this regime (Mishkin 1999). What followed for these countries was a period of highly discretionary policy-making, in which the stance of policy was assessed on the basis of a range of indicators. During this time, Germany maintained its monetary-targeting regime with a clear commitment to maintaining price stability, while in the US, the Fed acted to lock in the low inflation brought about in part by early 1980s recession.^[19] Judging by the evidence in Figure 4, the inflation targeters' policies were not very effective in lowering inflation, as evidenced by the eventual inflation targeters having higher average inflation rates compared to the non-inflation-targeting group.

A number of the eventual inflation targeters had either joined or pegged their currencies against the ERM by 1990, which is likely to have been responsible for some of the closing in the gap with the non-targeting group. However, the inflation-targeting countries only managed to close the gap by the end of 1991, associated with recessions that, measured by output gaps, were on average more severe than those experienced in countries that did not adopt inflation targeting. By the end of these recessions, most of the seven that became inflation targeters had adopted the regime. In the following years, the gap appears to have remained closed (the average inflation rates in the two groups are identical after 1991). The evidence here suggests that in terms of inflation control, the inflation-targeting group of countries succeeded in the 1990s (when inflation targeting was introduced) where they had failed in the 1980s. In the 1990s, the policy objectives of both groups, in terms of intolerance of inflation, appeared to become more similar, which is likely to have been at least partially responsible for the similarity in their inflation performance.^[20] In other words, inflation convergence is to some extent caused by convergence in the goals of monetary policy.

Footnotes

In contrast, the economic policies of the European Community (EC) were aimed at maintaining stability in exchange rates. In 1972, the so called ‘snake’ regime was established, whereby bilateral movements of participating exchange rates were constrained to a narrow band. In addition to EC member countries at the time, Denmark, Ireland, Norway, Sweden and the United Kingdom also joined the regime. However, a number of countries (including Sweden and the United Kingdom) exited the ‘snake’ in the following years to leave only five participating countries by 1977. The Exchange Rate Mechanism (ERM) of the European Monetary System (EMS), which superseded the ‘snake’, began operating in 1979 with eight original participants (Belgium, Denmark, France, Germany, Ireland, Italy, Luxembourg and the Netherlands), with Spain, the United Kingdom and Portugal joining about a decade later. Despite numerous realignments (of countries' central rates) during the first eight years of the regime, the ERM provided relative stability to participants until the exit of the Italian lira and the British pound in 1992 (Bladen-Hovell 1994; Kenen 1995; Eichengreen 1997). [16]

Out of the 4 other inflation-targeting countries, Spain also adopted a monetary-targeting regime in 1978 which was in place until it entered the ERM. Sweden and Finland maintained exchange rate anchors right through to their adoption of inflation targeting, while New Zealand had both fixed and crawling-peg exchange rate regimes in place until the float of the New Zealand dollar in 1985, after which there was no nominal anchor until the adoption of inflation targeting (Aurikko 1986; Quigley 1992; Kenen 1995; Bernanke et al 1999). [17]

This is not to say that all of the countries that did not adopt inflation targeting in the 1990s were as successful as Germany and Switzerland with monetary targeting. Bordo and Schwarz (1999) provide a detailed account of US monetary policy, and they argue that in the 1970s the Federal Reserve raised monetary growth to provide employment growth which eventually led to an acceleration in inflation. They suggest that because of the potential political costs, the Fed was unwilling to tighten monetary policy to curb inflation, and high monetary growth and inflation persisted until Paul Volcker took over in 1979. [18]

Bernanke and Mishkin (1997) argue that German and Swiss monetary policy was very similar in practise to present-day inflation targeting, rather than orthodox monetary targeting. Hence, they suggest their policies should be thought of as ‘hybrid’ inflation targeting. [19]

While the other 13 countries have not announced a formal inflation target, or were not inflation targeters until very late in the decade, a number of authors argue that they conduct monetary policy as if they had an ‘inflation target in mind’ (Stone 2003, p 3). Such countries have been called ‘eclectic’ (Carare and Stone 2003) and ‘covert’ (Mankiw 2001) inflation targeters. [20]