The statutory goal for monetary policy

The government has a review of the Reserve Bank Act underway at present.    It is an odd beast.    There is a rushed process going on to review and amend the monetary policy bits of the Act, led by Treasury and assisted to some modest extent by an Independent Expert Advisory Panel –  itself chaired by someone with no experience in monetary policy.   And then there might be a second stage review, to be jointly run by the Treasury and the Reserve Bank, which might –  or might not –  have the same Independent Expert Advisory Panel.  If well-intentioned, it is pretty unsatisfactory: lightly-resourced, potentially giving too much influence to the Reserve Bank itself, and not working from a coherent overview of the Bank as an institution (thus, part of the first stage review is looking at the role of the Bank’s Board as regards monetary policy, without having engaged properly with how the whole organisation should be structured, managed, and held to account, or even looking at the interactions between the Bank’s various functions).  How much better it would have been to have had a proper independent review, looking at all the roles/functions of the Bank at the same time, with a view to legislating next year.    That was, for example, what was done in Norway recently with central banking legislation of similar vintage to our own.    It would also reduce the chances that, with a patch-protecting new Governor, the second stage of the review will just peter out and come to nothing.

There are two main bits to the first stage of the review:

  • amending the statutory goal of monetary policy to add a focus on maximising employment (or minimising unemployment), and
  • introducing a statutory committee, including non-executive members, to make monetary policy decisions.

For the time being, much of what the government says they want to achieve around the first bullet  (the objective for monetary policy) could be done through a new Policy Targets Agreement  –  one needs to be signed before Adrian Orr can formally be appointed as Governor (and thus will be necessary even before Stage 1 reforms can be legislated.  I outlined some suggested wording in a post last year.   Sadly, with the rushed review focused on the Act, there is no sign of any public process looking at the content of the Policy Targets Agreement, even though it is the main policy instrument for short-term macroeconomic management.

But how should the statutory goal be restated?  Since 1989 the key clause of the Reserve Bank Act (section 8) has read as follows

The primary function of the Bank is to formulate and implement monetary policy directed to the economic objective of achieving and maintaining stability in the general level of prices.

And 10 years ago, a revised purpose clause was added to the Act, and around monetary policy this is what it said

The purpose of this Act is to provide for the Reserve Bank of New Zealand, as the central bank, to be responsible for—formulating and implementing monetary policy designed to promote stability in the general level of prices, while recognising the Crown’s right to determine economic policy;

Policy Targets Agreements have to be consistent with section 8 –  ie “achieve and maintain” price stability, not just “promote” it.

The Labour Party, in Opposition and now in government, have offered various takes on what they want to do with the statutory goal without (sensibly enough at this stage) specifying a precise set of words.  Presumably they are serious about taking advice.

When the policy was launched last April the suggestion was to broaden “the objective of the Bank from just price stability to also include a commitment to full employment”.

In the terms of reference for the current review, agreed by Cabinet, the phraseology is as follows

recommend changes to the Act to provide for requiring monetary policy decisionmakers to give due consideration to maximising employment alongside the price stability framework;

The Minister of Finance has also talked about looking to the wording used in the central bank laws of Australia and the United States.

As I’ve outlined here previously, I favour changing the wording of section 8, as much as anything because the current wording does not adequately capture why we have a central bank running discretionary monetary policy.   One doesn’t need such an active central bank to maintain a broadly stable general level of prices; indeed, the introduction of fiat money systems over the last hundred years or so has led to less stability in the general level of prices than prevailed previously.

The existing wording of section 8 was an understandable –  and generally helpful –  reaction to the very poor inflation track record New Zealand (and many other countries) had run from the 1960s onwards.  Inflation has been lower and more stable than it was previously –  although not just because of the change in the Act.  But it still didn’t capture what we really wanted the central bank to do.    Active discretionary monetary policy is really about attempting to manage the business cycle –  and especially to be able to respond aggressively to extreme circumstances –  subject to the constraint of not letting inflation get away on us.    To word it that way doesn’t diminish the significance of keeping inflation in check –  any more than saying that fiscal policy should be managed to meet various government goals, subject to the constraint of not letting debt blow-out.

This way of thinking about active monetary policy is quite consistent with what we learned from the Great Depression –  countries that acted earliest to break the golden fetters rebounded earliest and saw resources (including unemployed people) back in work most quickly.  But, more mundanely, it is also consistent with the way in which inflation targeting central banks have (a) run things, and (b) specified their practical mandates.     Thus, from the earliest days of inflation targeting, there have been numerous events –  notably oil price shocks or indirect tax changes –  where the Bank has been expected to let inflation rise (fall) temporarily away from target, to minimise the output and employment consequences of such shocks  (trying to keep inflation on target in face of a big oil price increase would typically be expected to require inducing a recession).  Policymakers recognised that, actually, avoiding unnecessary surges in unemployment –  or recessions –  was locally more important than keeping inflation at, say, 2 per cent all the time.  But that calculus changes if it looked as though inflation was going to drift permanently higher.     The wording of the current legislation is unhelpful if –  and to the extent – it leaves voters fearing otherwise – that policymakers don’t care about unemployment/employment in its own right.

The other thing worth bearing in mind is that no central bank act –  not even something more specific like a Policy Targets Agreement –  can specify everything about how we want a central bank to run monetary policy.  Circumstances change, and the unexpected combinations of shocks happen.    There was a degree of naivete around that in the early days of the current Act, encapsulated in Don Brash’s unconditional answer to a radio interviewer indicating that he would lose his job if inflation went above 2 per cent (the then top of the target range).   But everyone recognises the limits now.

Thus, it is not possible to sensibly write down –  whether in the Act or the PTA – a numerical objective for “full employment”, “maximum employment” or the like (and the government has repeatedly stated that it has no intention of doing so).  But that doesn’t make it less worthwhile writing such considerations into legislation, requiring the Bank to report against them, and expecting those holding the Bank to account to take serious account of them.   And if even that was thought to be impossible, it might sensibly lead to some reconsideration as to whether having an independent central bank set policy –  rather than just advice on it –  was really the best way to organise things.

But how best to word things?  There have been suggestions from the Minister of looking at the specific wording in the Australian and US legislation.  I don’t think that will –  or should –  get people far.   Here is what I had to say on those options from an earlier post.

The Reserve Bank of Australia was set up in 1959, and the section of its legislation relating to monetary policy goals and objectives was in the original.

It is the duty of the Reserve Bank Board, within the limits of its powers, to ensure that the monetary and banking policy of the Bank is directed to the greatest advantage of the people of Australia and that the powers of the Bank … are exercised in such a manner as, in the opinion of the Reserve Bank Board, will best contribute to:

a.  the stability of the currency of Australia;

b.  the maintenance of full employment in Australia; and

c.  the economic prosperity and welfare of the people of Australia.

In 1959, Australia –  like most countries –  had a fixed exchange rate, so that “the stability of the currency of Australia” meant the external value of the currency.  The provision has since been re-interpreted, and as is now taken as meaning the domestic value of the currency (ie domestic price stability), but no one would write the provision that way today.

This wording is also legitimately subject to the criticism made by those who disagree with what Labour is proposing.  It makes no attempt to distinguish between the short and long run, and thus does not recognise that monetary policy cannot affect the longer-term rate of unemployment at all.    The Australian legislation also has nothing like a Policy Targets Agreement (the document that resembles a PTA is informal and non-binding) and provides far too much discretion to the Reserve Bank.  That discretion has not been blatantly misused in recent decades –  a period when the actual conduct of monetary policy in New Zealand and Australia have mostly been quite similar –  but the legislation should not provide any sort of model for New Zealand as to how best to specify the goals of monetary policy.

What of the United States?   Much is made of the “dual mandate” that has guided the Federal Reserve over the decades.   But even that, mostly quite sensible, conduct of policy rests on a rather slender and unreliable legislative footing.    The statutory objectives in the Federal Reserve Act were set out in 1977, around the high tide of monetarism, and read as follows:

Section 2A. Monetary policy objectives

The Board of Governors of the Federal Reserve System and the Federal Open Market Committee shall maintain long run growth of the monetary and credit aggregates commensurate with the economy’s long run potential to increase production, so as to promote effectively the goals of maximum employment, stable prices, and moderate long-term interest rates.

In other words, the Fed is actually mandated to pursue long-term money and credit growth targets, in the belief that doing so will promote (a) maximum employment, (b) stable prices, and (c) moderate long-term interest rates.  Again, no one would write the statutory objectives that way today, and the formulation should offer little or no guide to anyone looking to overhaul the objectives of our own central bank.  In practice, of course, the Federal Reserve works around the statutory formulation, rarely citing it directly.  I think they way they run monetary policy in practice is quite sensible –  and typically not that different to the way the Reserve Bank here has often run policy –  but I bet they wish Congress had written the goal a bit differently in 1977.

In an ideal world, both the Australian and US statutory provisions would be updated and amended.  It isn’t desirable to have powerful autonomous agencies working under the mandates that don’t reflect today’s understandings of policy, leading those agencies to creatively reinvent their own mandates.  Those reinventions probably lead to better policy in the short-run, but the process of doing so undermines confidence in the role of legislatures in mandating, and holding to account, such agencies.

So we need some fresh wording, and as far as I can tell there are no excellent models abroad to look to.

My strong preference would be to focus any new wording on minimising unemployment (or concepts like “full employment” –  a traditional Labour Party focus) rather than on maximising employment.       The concept of unemployment is about avoiding a situation where people who want to work, and are available for work, are unable to find work.   A simple focus on employment suggests that the more employment there is the better, whether people want (or need) to work or not.     I’ve seen political leaders boast that New Zealand has one of the highest employment rates in the advanced world, but to my mind at least that isn’t a sensible or legitimate boast –  or even a matter of public policy interest (speaking as someone who is, entirely voluntarily, not in the paid workforce).  It has Stakhanovite connotations.  By contrast, situations where people who want work, need work, can’t find work should be a matter of real public policy concern, and it is also something that  – in some but not all circumstances –  monetary policy can do something about, since most of the fluctuations in unemployment are a response to demand shocks.

So I’d suggest wording along these lines

“the goal of monetary policy shall be to promote the lowest rate of unemployment consistent with maintaining a low and stable rate of inflation on average over time”

But that might be too stark for some, too radical for others.   Another way of saying much the same thing, while keeping low inflation first in sequence, might be this sort of wording

“the goal of monetary policy is to maintain a low and stable rate of inflation on average over time. In pursuing this goal. the Bank shall do all that can be achieved by monetary policy to avoid unnecessary deviations of the unemployment rate from the rate that would resulting from those regulatory, structural, and demographic factors beyond the influence of monetary policy”

(You will note that I’ve suggested deleting altogether both the suggestion that monetary policy is the “primary function” of the Bank –  these days it is one of two prime functions –  and the references to a “stable general level of prices”.  The Bank has never targeted the price level, and the price level is not stable.    Some argue for a price level focus rather than an inflation rate focus, but there is no easy way for the statute to encompass both options without using even vaguer language.    Arguably, a “low and stable inflation” rate encompasses options for price level targeting, while “a stable general level of prices” does not really encompass a price level growing on average at up to 3 per cent per annum. )

Of course, the new wording for section 8 –  whichever formulation is chosen –  does not (or should not) stand alone.  Appropriate wording for a Policy Targets Agreement would still need to be found.  I’d have no problem with a continuing focus on a 2 per cent target midpoint –  as some sort of medium-term reference point for people to use in their planning.  But there should be a stronger emphasis in the Policy Targets Agreement on the cyclical management responsibilities –  including, in particular, preparation for the next big downturn (when, at present, monetary policy looks to be quite severely constrained in limiting the output and unemployment consequences).

And, as I’ve suggested previously, in both the Policy Targets Agreement, and in section 15 of the Act (governing Monetary Policy Statements) it would be appropriate to add some explicit wording requiring the Bank to report periodically on its estimates of the long-run sustainable rate of unemployment (and/or the NAIRU), on to report regularly on the relationship between actual unemployment rates and those estimates, and one what the Bank is, and is not, able to do about the gap.  These reports would serve several functions:

  • they would help focus the Bank’s attention (policy and research/analysis) on its raison d’etre  –  doing what it can to avoid unnecessary excess capacity, and
  • they would provide a formal and routine vehicle for articulating the limits of monetary policy.  It won’t, for example, be sensible to focus heavily on the unemployment rate for the time being if inflation really looks to be moving permanently much higher, but even in those circumstances, it is useful to require policymakers to contribute to the conversation about the appropriate balance of short-term focus –  what risks can and should be run (or not) and at what cost, even when everyone recognises the constraint that medium-term inflation needs to be kept in check.

One might even provide for the Bank to offer periodic advice to the Minister of Finance on what measures could contribute to lowering the structural rate of unemployment.    There are risks in such a provision –  many of those sorts of choices are almoat inherently political –  but again they help ensure a recognition that if monetary policy can, and should, do a lot about cyclical unemployment, it is unable to do anything much to alter the baseline structural rates of unemployment, which are mostly about political and social choices.

Those opposing change will emphasise the risks aroud these sorts of statutory changes.  And those risks need to be taken seriously.     But it is hardly as if the status quo is without risks or problems –  after all, we’ve spent the entire decade so far with the unemployment rate above domestic estimates of a NAIRU, and if that is no cost to most economists and policymakers, it has been for many ordinary New Zealanders.

Precisely because one can’t write down all that one wants a discretionary central bank to do with monetary policy in all conceivable circumstances, any particularly specification of the goal will be less than ideal.     So whatever the precise specification the government chooses, it is at least as important that an intelligent conversation be maintained about what monetary policy can and can’t do – not just in general, but in particular circumstances –  and that the Reserve Bank be encouraged (and compelled by law where appropropriate) to maintain an open and transparent approach, not just to the final fruit of its deliberations, but to its research, analysis, and the genuine uncertainty everyone faces in making sense of economic developments and the outlook for resource utilisation and inflation.


6 thoughts on “The statutory goal for monetary policy

      • If you go back to 2007 to 2009 in the depths of a NZ recession when 61 finance companies went broke with the loss of $6 billion in mum and dad funds because monetary policy was used by the RBNZ to engineer a severe recession, a lot of people out of work did care with interest rates hovering around 10% because the OCR was set at 9%.

        Farmers and property owners including home ownership burned precious equity by having to borrow even more paying those 10% interest rates.


      • Mortgagee sales and company receiverships and eventually bankruptcy was at record highs. We certainly do not want a repeat of this pure monetary policy ideals at the expense of NZ industries and the loss of 2 segments of the entire NZ economy. The decimation of risk finance and the entire building industry.


  1. I favour NGDP level targeting, but I do like the first formulation you’ve put forward. It would make it hard of the RBNZ to pat themselves on the back in situations where the inflation rate was moderately below target at a time of very high unemployment. Putting unemployment first makes a big difference.


    • At a pinch, “low and stable inflation” could probably loosely encompass NGDP targeting. There is no easy way of being all-encompassing in the Act, except to the point of vacuity – ie “central bank shall pursue whatever target MOF and the Bank agree from time to time”


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s