New Zealand’s Thirty-Year Experience with Inflation Targeting: The Origins, Evolution and Impact of a Monetary Policy Innovation

Abstract

Thirty years ago, New Zealand ushered in a revolutionary approach to monetary policy. This approach was formalized by the Reserve Bank of New Zealand Act 1989 which specified price stability as the primary function of monetary policy and provided operational autonomy for the central bank. This institutional innovation spawned the spread of more central banks around the world with a mandate to prioritize inflation targeting. This paper explains the historical origins of the RBNZ Act, its design and the ideas that influenced its design, and how the practice of inflation targeting and choice of monetary policy instruments have evolved. Some parts of this evolution were pioneering, and some parts followed international practice. The paper also reviews evidence of the impact of inflation targeting in New Zealand and discusses recent changes to the RBNZ Act.

Keywords:

• Monetary policy
• inflation targeting
• central bank governance
• accountability
• transparency
• credibility
• sustainability

Notes

Acknowledgements

I am grateful to Alan Bollard, Don Brash, John Creedy, Chris Eichbaum, Norman Gemmell, Arthur Grimes, Viv Hall, Özer Karagedikli, Robert Kirkby, John McDermott, Grant Scobie and Grant Spencer for helpful discussions and suggestions during the preparation of earlier versions of this paper. This version has benefited from suggestions from the Editor and two anonymous referees.

Notes

1 McDermott and Williams (2018) note that several countries had earlier flirted with publishing inflation or price level targets. They note that Italy, Greece and Portugal had single year targets for inflation during the early 1980s and Sweden had introduced price level targeting for a short period in the 1930s. However, none of these had developed the type of frameworks required for sustained inflation targeting that were developed by countries after 1989.

2 Bernanke et al. (2001, 4) defined inflation targeting as ‘a framework for monetary policy characterised by the public announcement of official quantitative targets (or target ranges) for the inflation rate over one or more time horizons, and by explicit acknowledgement that low, stable inflation is monetary policy’s primary long-run goal. Among other important features of inflation targeting are vigorous efforts to communicate with the public about the plans and objectives of the monetary authorities, and, in many cases, mechanisms that strengthen the central bank’s accountability for attaining those objectives.’

3 See for example Alesina (1988) which included analysis of the implications of the ‘independence’ of a central bank from political direction and the potential impact on inflation.

4 Don Brash, the first Governor under the new RBNZ legislation, was active internationally explaining the 1989 legislation. Several examples are in Brash (2014), including his invited speech at a session on inflation targeting at the 2002 American Economic Association annual meeting in Atlanta (Brash 2002) and his membership of an IMF team which recommended a governance arrangement for the central bank of Indonesia like that prescribed by the New Zealand legislation.

5 The focus of this paper is monetary policy and inflation targeting. The RBNZ also has responsibilities for the efficiency and soundness of the financial system, along with the New Zealand Financial Markets Authority. Those responsibilities are beyond the scope of this paper.

6 This aversion to apply monetary instruments and higher interest rates to curb inflation from the late 1970s to mid-1980s when Robert Muldoon was Prime Minister and Minister of Finance is also discussed by Singleton et al. (2006, Chapter 2).

7 For explanations of the reforms see Bollard and Buckle 1987; Silverstone, Bollard, and Lattimore 1996; Evans et al. 1996.

8 Singleton et al. (2006, Chapters 2 and 3) provides a summary of these debates.

9 Roger Douglas was the Labour government Minister of Finance after the 1984 election and until 1988; Robert Muldoon was Prime Minister and Minister of Finance in the National government during the decade prior to 1984 and loomed large over economic policy during that period. Chapters 2 to 5 of Singleton et al. (2006) provide an insight into the interplay between officials and politicians in the lead-up to the financial crisis in 1984 and the subsequent design and implementation of the RBNZ Act 1989.

10 Forder (2005) reviews some of the explanations for why there emerged a strong interest in adopting ‘independent’ central banks during the 1990s. Some of the arguments he dismisses as valid explanations for this trend in the 1990s, such as the previous inflation experience, establishing policy credibility, and the relevance of time-inconsistency literature and institutional design, were important influences for introducing the New Zealand legislation in 1989 and inflation targeting.

11 Hunt (2017) shows there are many dimensions to this term and they can vary across central banks. It is problematic as a description of the RBNZ governance arrangements. Although the Bank has discretion over the choice of monetary policy instruments, their use is subject to meeting other conditions specified in PTAs and the Act. Moreover, the objectives of monetary policy are determined by legislation and agreement between the government and the Bank Governor via PTAs and now Remits. In addition, governance arrangements require monitoring of the Bank and Governor’s performance by the Board. Furthermore, the RBNZ does not have financial independence (its expenditure budget is subject to a five-year agreement between the Minister of Finance and the Governor), and the Governor is appointed by the Minister of Finance for a fixed term. ‘Independence’ in the context of the RBNZ Act 1989 aligns closely with Bernanke’s (2017, 29–36) four elements of central bank independence.

12 There have been four Governors appointed since the RBNZ Act 1989 was introduced: Dr Don Brash (appointed in 1988, and first Governor under the new Act from 1990 to 2002), Dr Alan Bollard (2002–12), Graeme Wheeler (2012–17), Adrian Orr (appointed in March 2018). Two Deputy Governors have been appointed Acting Governors: Dr Rod Carr in 2002 and Grant Spencer in 2017.

13 Operational autonomy involved a marked shift of responsibility from the Minister of Finance to the Governor. Brash (2002, 3) considers that political acceptance of this shift was helped by the provision that the objectives of monetary policy and price inflation targets remained the prerogative of elected governments and were to be determined by agreement between the Minister of Finance and the Governor. Brash (2002, Chapter 5) argued this meant that the situation which often occurred previously whereby confidential recommendations to the Minister about what changes should be made to official interest rates to control inflation were ignored whenever it was politically inconvenient, could no longer prevail.

14 The OCRAG is now called the Monetary Policy Advisory Group (MPAG) and external input is through external members of the MPC.

15 McDermott and Williams (2018, 12) explain the process of consultation and decision-making by the Bank during the period up to 2018.

16 Grimes (2001) summarizes the Svensson Review and the initial response by the RBNZ.

17 The additional employment objective was included in the first PTA signed by Governor Orr in 2018.

18 Snively, Edey, and Karacaoglu (2018, Appendix 1, 33–9) contains a description of the goals of monetary policy for a selection of developed economies and the dates they were established. Wadsworth (2017) also compares inflation-targeting frameworks for a selection of developed economies.

19 The Advisory Panel also considered that the simultaneous calibration of employment and output objectives may be difficult to achieve. This is evident from Hall and McDermott (2016) who show that the dynamics of output and employment in New Zealand differ. Although not discussed in the Advisory Panel Report, another reason for specifying maximum sustainable employment could have been because it may be easier to estimate than potential output (Spencer 2017).

20 Svensson (2001) had earlier proposed a formal committee-based decision process for the RBNZ.

21 Governance issues are to be considered, along with the prudential role of the RBNZ, in the second phase review to be undertaken in 2019 (Robertson 2017).

22 See for example Singleton et al. (2006, Chapters 3 and 5).

23 The MCI was introduced prior to a major demand shock arising from the Asian Financial Crisis and supply shocks arising from successive severe droughts that impacted pastoral production in New Zealand. The nominal exchange rate fell as would be expected, which increased the value of the MCI. But although a fall in domestic interest rates would be required to help moderate the fall in domestic economic activity, financial markets believed that an increase in interest rates was required to maintain the MCI at its target level. Engelbrecht and Loomes (2002) found that, contrary to its stated intentions, the MCI regime failed to improve the Bank’s communication of its monetary policy stance to financial markets. The Bank has acknowledged that the way it implemented monetary policy in this period accentuated output and interest rate volatility (Reserve Bank of New Zealand 2000b).

24 After a few years, SCBs were supplemented by Reserve Bank Bills and the Bank targeted the combined aggregate referred to as ‘primary liquidity’, which was still a quantity-based instrument.

25 However new liquidity facilities, such as residential mortgage-backed securities (RMBS), were introduced during the GFC to provide liquidity to commercial banks for financial stability purposes.

26 The inflation-output volatility trade-off is one of the relationships that underpins the RBNZ approach to inflation targeting (see Bollard and Karagedikli 2006, Figure 4, 14).

27 Grimes (2014) shows that targeting nominal GDP rather than inflation does not avoid this trade-off.

28 These rigidities help explain why the large increases in the relative price of oil during the 1970s contributed to stagflation in many countries dependent on imported oil for production.

29 See note 7 for other references that discuss these reforms.

30 See also Bollard and Karagedikli (2006, 15–17) and references therein for a discussion of other influences contributing to lower inflation and changes to inflation dynamics in New Zealand during the 1990s and 2000s and which could have contributed to shifting the efficiency frontier.

31 See Chen and Rogoff 2003; Munro 2004; Buckle et al. 2007.

32 These differential sector effects are like ‘Dutch disease’ effects when relative prices change, and resources shift to the ‘booming sector’. In the context of monetary policy and inflation targeting, the situation described here is akin to Corden’s (1997) ‘by-product distortions’. An implication that applies to any policy assignment decision is that it is important to understand these by-product effects because the optimal choice of policy instrument can change if these side-effects are sufficiently costly. Corden’s approach to the assignment of instruments in the trade policy context is represented by a ‘hierarchy of policy instruments’ which acknowledges that secondary distortions from each instrument (when it is used to hit its primary target) may also be important. In the context of Mundell’s (1962) assignment principle this would mean the optimal assignment of policy instruments would change.

33 The first PTA was signed on 2 March 1990. Since then, there have been 12 further PTAs either because of the appointment of a new Governor, a change of government or Minister of Finance, or an agreed change in the parameters following changing circumstances (Reserve Bank of New Zealand 2000a, 2018). See also Grimes (1996, 261), Reserve Bank of New Zealand (2007), and Lewis and McDermott (2016, 344–7).

34 For example, Archer (1997, 4) described the consequences of the framework laid down by the Act as being ‘to focus the implementation of monetary policy solely on price stability to the exclusion of real economic objectives’. Bryant (1996) gives a similar interpretation. It is evident from successive PTAs that this was not strictly the case.

35 See for example Singleton et al. (2006, Chapter 6) for a summary of the public debates during this period.

36 See Chen and Rogoff (2003) and Munro (2004). Buckle et al. (2007) also show that the exchange rate tends to act as a buffer and after controlling for the impact of the effects of changes in government demand, climatic conditions, and international commodity, business cycle and financial shocks, the residual effects of the exchange rate in inducing further volatility of New Zealand’s GDP were trivial after the exchange rate was floated.

37 This scheme is distinct from the use of foreign reserves by the Bank for intervention in the foreign exchange market to preserve the functioning of the market in times of a foreign exchange crisis.

38 Munro (2004) provides a discussion of what the RBNZ considered at that time to be the principal determinants of movements in the New Zealand exchange rate.

39 This provoked criticism that monetary policy exacerbated the New Zealand recession from 1990 to 1992. However, research estimating the contribution of monetary policy to inflation and output volatility suggests that recession was caused by international factors (including commodity price changes) and non-financial domestic factors, and monetary policy moderated these effects (Buckle, Kim, and McLellan 2003).

40 This point is also made by Dalziel (1997, 291) in his discussion of the choice of mid-point for the inflation band.

41 Brash (2012) and Archer (1997) offer examples of monetary policy responses in the first decade of inflation targeting which in their opinion suggested monetary policy was more focused on the inflation outlook, more immune to the election cycle, and involved more forward-looking policy decisions than was the case prior to the 1989 legislation.

Additional information

Notes on contributors

Robert A. Buckle

Robert A. Buckle, ONZM, is Professor Emeritus at Victoria University of Wellington and an External Member of the Reserve Bank of New Zealand Monetary Policy Committee. Research areas include: macroeconomics, business cycles, fiscal, taxation and monetary policy and tertiary education policy, including the impact of performance-based research funding schemes.