Reserve Bank BIM: resisting reform

Once upon a time, post-election briefings to an incoming minister actually contained some free and frank official advice on major policy issues.  I have sitting on my desk the 1987 Treasury briefing  –  almost 800 pages of analysis and advice (good and bad).   But as the briefings started being  routinely published, it seemed to start an inexorable trend towards documents of astonishing banality, often containing little more than lists of the official ministerial responsibilities and/or the activities of agencies (I just flicked through the BIM of one major ministry, released yesterday, and it was a startlingly content-free zone).  If there still is free and frank advice offered by the public service, you won’t often find it in these documents.  It isn’t helped by the growing practice of writing (or finalising) the BIMs only after the composition of the new government is known, rather than in advance of the election.

Reserve Bank BIMs over the years have tended to go in the same banal descriptive direction.  So I didn’t expect to find anything interesting when I opened up the document released yesterday.  But I was wrong.    The (unlawfully appointed) “acting Governor” Grant Spencer had used the opportunity to make his case for minimal change to the governance and decisionmaking provisions of the Reserve Bank Act, including an eleven page appendix on that specific issue.

At the time of last month’s Monetary Policy Statement, Spencer –  and his deputy, Geoff Bascand –  went public on their opposition to any substantial change.  But they did that just in response to press conference questions.   I summed up their position –  “just cement in the status quo, and if we really have to have externals make sure they are silenced and that there is no greater transparency” –  this way

[their] approach is that of “the priesthood of the temple” –  we will tell you, the great unwashed, only what it suits us to tell you, in the form we want to present it.  It is simply out of step with notions of open government, or with a serious recognition that monetary policy is an area of great uncertainty and understanding is most likely to be advanced by the open challenge and contest of ideas.

In the BIM, they go further and stress how keen they are that any external appointments should be made by the Governor rather than –  as is the case in most other advanced democracies –  the Minister.   They were keen to keep any review of the Act very narrow, and to control the process themselves as far as possible

we would be happy to prepare a draft terms of reference, in consultation with the Treasury

Fortunately, they lost that one and the review is being led by The Treasury, supposedly supported by an (as yet unnamed – or unannounced – Independent Expert Advisory Panel).

But it was that eleven page appendix I really wanted to write about this morning.  They describe even it as a “brief summary” and refer to a more detailed version of the paper being available. I have lodged a request for it, without any great expectation of it ever seeing the light of day (it might, for example, have been logical to have released it when the Minister of Finance announced his review –  logical, that is, if the government or Bank were interested in open government).

The Bank’s first claim is that the current system –  a single unelected decisionmaker, appointed by other unelected people –  “works well” at least for monetary policy, and that the Reserve Bank’s monetary policy “is highly regarded internationally”.    In principle, those are two separate points.  It is hard to imagine the Reserve Bank being held in “high regard” internationally for its monetary policy over the last decade –  whatever sentimental respect there still is for the Bank as pioneer of inflation targeting almost 30 years ago.    Our central bank remains the only advanced country central bank to have launched into two tightening cycles since the 2008/09 recession, only to have to fully unwind both of them.   Perhaps it is the sort of mistake anyone might have made, but our Reserve Bank did.    And the Bank has no more successful than anyone else in keeping inflation close to target –  even though, with interest rates still well above zero, it faced fewer obstacles than most.

The one recent report the Bank attempts to use in its defence is a brief one written earlier this year by an old friend of the then Governor.  I wrote about it at the time.

When an old uncle or family friend is in town and comes for dinner, the visitor will usually compliment the cook, praise the kids’ efforts on the piano, the sportsfield, or in dinner table conversation, and pass over in silence any tensions or problems –  even burnt meals –  he or she happens to observe.    Mostly, it is the way society works.  No one takes the specific words too seriously –  they are social conventions as much as anything.  One certainly wouldn’t want to cite them as evidence of anything much else than an ongoing, mutually beneficial relationship.

It is a shame the Reserve Bank is reduced to publishing, and touting, a report like this in its own defence.  When good old Uncle Philip, a fan of yours for years, swings by, it must be mutally affirming to chat and exchange warm reassuring thoughts.  But as evidence for the defence his rather thin thoughts, reflecting the favourable prejudices of years gone by, and institutional biases against doing much about inflation deviating from target, isn’t exactly compelling evidence for the defence.    Sadly, getting too close to Graeme Wheeler as Governor seems to diminish anyone’s reputation.  It is a shame Turner has allowed himself to join that exclusive club.

In a way, what strikes me most about the Bank’s appendix is the near-complete absence of (a) any critical self-scrutiny, and (b) any sense of operating in a society that expects scrutiny and accountability of powerful public agencies, and not just on terms set by those agencies.

For example, they organise their thoughts around the notion that “the objective of decision-making design is to create a system that leads to rigorous decision making”.  I’m not sure which management textbook they got that from, but it doesn’t sound a lot like the way we should organise things  for making public decisions in a democratic society (internal management decisions in a private company might be another matter).  “Rigour” in decisionmaking is certainly important, but when those decision are outward-facing and have pervasive effects across a country, with no rights of review or appeal, it is far from being the only relevant criterion.   “Legitimacy” for example –  an ongoing sense of public confidence that the agency is being well run, in the interests of the public not just of officialdom – matters a lot.  So should openness.  For many year now our statute books have contained this provision (part of the purpose clause in the Official Information Act).

to increase progressively the availability of official information to the people of New Zealand in order—
(i) to enable their more effective participation in the making and administration of laws and policies; and
(ii) to promote the accountability of Ministers of the Crown and officials,—
and thereby to enhance respect for the law and to promote the good government of New Zealand:

Around their centrepiece –  the goal of “rigorous decisionmaking” –  they circle “four key components”

  • institutional design,
  • high quality inputs,
  • genuine deliberation, and
  • accountable for decisions

There is nothing of interest under their heading of “institutional design”.  As they note, the Governor is legally responsible for all Reserve Bank decisions, and although all Governors since 1989 have operated with advisory committees (the form and names have changed over time), in a single decisionmaker model there is only one Reserve Bank view.   However, it is worth noting that the Reserve Bank remains intensely secretive about the range of internal views or advice ever becoming known, in ways that do nothing to support good decisionmaking, let alone robust scrutiny, challenge, and accountability.

On ‘high quality inputs’, the Bank claims that its decisionmaking “is supported by a broad range of high quality inputs”.  Perhaps, but (a) despite the inputs they’ve still made some bad policy mistakes, which should at least raise questions about the inputs (recognising that in a field like monetary policy, riddled with uncertainty, some mistakes are inevitable), and (b) we don’t know, because they hold all the inputs very closely, and refuse to release any, even years later.   Just yesterday, we had the refusal to release their background analysis on the aspects of the new government’s policy they’ve incorporated in their latest projections.  I once wrote a paper with Grant Spencer to the then Minister of Finance in which we referred rather scathingly to one particular proposal as involving “trust us, we know what we are doing”.  These days, unfortunately, Grant seems to treat it as a practical guide to running a central bank (whether on most regulatory matters or monetary policy).

They claim, only briefly, that their current system produces “genuine deliberation”.  Again, how would we know?   They refuse to release the minutes of the Monetary Policy Committee, or of the Governing Committee, they refuse to release a summary of the individual recommendations on the OCR, and they refuse to release the background papers.  Not just in the weeks after a decision, but even years later. I know, I’ve asked.   How robust, for example, were the deliberations around the ill-judged tightening cycle than began in 2014?  In my observation –  I was still involved at the time –  not very.

Then there is “accountability”, which sounds good, but isn’t really.  Of course, they cite the role of the Reserve Bank Board, and its reports to the public and to the Minister. But this is the same Board that will have appointed the Governor and which, in history, has never openly said a remotely critical word about any Reserve Bank decision.   Perhaps in private they do a really good job  –  in my experience, at times some of them (usually the more awkward ones) asked some useful questions –  but that isn’t serious public accountability.  The Board seems to see their role primarily as having the back of the Governor –  how else, for example, did they stay silent in the face of Graeme Wheeler’s deployment of his entire senior management to try to silence Stephen Toplis?   The Bank goes on to claim (correctly) that the Board is given the materials used to lead up to OCR decisions, but then (outrageously) claims that “a subset of this information is made available to the general public”.  In fact, none is at all.    All we get is what the Governor chooses to allow us to see scrubbed up and sanitised in his Monetary Policy Statement, even though all the background material is public information, produced with public money.    They have a very strange definition of accountability at the Reserve Bank.

(As they do every few months, they roll out an academic paper from a few years ago suggesting that on that particular measure, the Reserve Bank is one of the most transparent central banks in the world.  As I’ve noted previously, there is a big difference between telling us a lot about what they know (almost) nothing about –  eg where the OCR might be in 2020 –  and telling us stuff they do know about (their own advice, analysis, range of views etc). They do the former well, and the latter is almost non-existent.)

The Bank then turns to potential modifications to the Act.   The (unlawfully appointed)
“acting Governor’s” preference is simply to codify the current committee (the Governing Committee, consisting at present of the “acting Governor”, the Deputy Governor and the chief economist).

In earlier incarnations of this idea, the proposal was that the members of such a statutory committee would all have formal voting rights.  But even that –  weak advance –  has now gone out the window and the Bank is now arguing to protect the single decisionmaker model, while putting into statute simply a requirement that the Governor have an advisory committee.  Under their proposal the Governor “would be responsible for the manner in which the Governing Committee conducts itself”.

Frankly, this is a worse than useless suggestion.  The Bank claims it would increase transparency around the decisionmaking process. In fact, the effect would be quite the reverse.  For a good Governor it might make no effective difference.  For a bad Governor, it would allow him or her the fig-leaf of being able to claim that decisions were made in (statutory) committee even though (a) the other members had no formal vote, and (b) all members would be appointed by, remunerated by, and accountable to, the Governor.  The Bank simply shows no sign of recognising the institutions need to built to be robust to bad appointees (because, in every human institution, they will happen from time to time).

Interestingly, they are open to the idea of separate committees for monetary and financial policy.  I’ve strongly favoured that, but recall that the sort of committees they propose are advisory only.  In fact, there is nothing to have stopped them putting such committees in place already.  Arguably, it was actually the way things worked before the Governing Committee was established: the Governor made OCR decisions in the OCR Advisory Group, and financial regulatory decisions in the Financial System Oversight Committee.  Those specialist committees still exist (OCRAG renamed as the – formal –  MPC).

We get to the real concerns –  they know they’ve lost the fight over keeping the single decisionmaker model –  when they come to the question of external members, the decisionmaking approach, and the communications.

On externals, they argue

The extent to which policy committees benefit from external members depends on the nature and objectives of the committee and the conditions associated with the external members’ appointment. Policy committees which have to interpret political objectives or indeed establish goals to be achieved should benefit from having external members. In New Zealand, the objectives of monetary policy are clear in the Act, and through the PTA. For prudential policy, the objectives of soundness and efficiency are clear, but their interpretation requires complex judgement.

To which my reaction is “yeah right”.  No one thinks the prudential policy objectives are clear –  in the sense of easily operationalisable.  There are big choices to be made about goals and instruments (eg around use of LVR restrictions). But actually it isn’t much different with monetary policy.  No one seriously regards the PTA as a document that avoids trade-offs, or involves no judgements –  indeed, wasn’t the “acting Governor” only the other day arguing for more flexibility in interpreting the goals?   More generally, they offer no support –  none –  for their claim that there is a special case for external members where goals are relatively more fuzzy.  External members can matter for a range of reasons, including minimising the risk of external groupthink, and helping ensure that a full range of models/perspectives are brought to the table.

They go on.

Policy making in monetary and financial policy often involves complex considerations based on multiple indicators, analytic models and competing economic theories. Full-time members with experience and expertise are likely to be better suited to this task than part-time external participants.

So you say.  But there are two separate issues here.  The first is about expertise (do we need subject experts only, or a range of perspectives) and the second is about whether the job is fulltime or part-time.   In Sweden, for example, most of the monetary policy committees members are outsiders (often academics or former market economists), but they are appointed to non-executive fulltime roles while they are on the committee (weirdly, this leads the Reserve Bank to claim they are insiders).  As for expertise, the Bank still seems to have made no effort to show that the issues it deals with an inherently more complex, or in more need of specialist expertise at the decisionmaking phase (as distinct from technical advice and supporting analysis) than many other agencies of the New Zealand government (which are typically run by part-time boards, appointed by ministers, with a range of backgrounds).  Even among executives, I don’t imagine the chief economist actually spends much time on financial regulatory matters on which he helps the Governor make decisions in the Governing Committee (at best, he is a part-timer non-expert in that area).

I’ve covered previously the Bank’s preference to avoid voting in committees, and (especially) to avoid any public revelation of differences of perspective.  They claim that

Research into decision-making practices finds that consensus is the preferred decision approach as it allows for more in-depth discussions, the frank exchange of views, more accurate judgement on average, and higher committee morale.

But they neither provide any references, nor engage with the practical experiences of various other advanced central banks that seem to have found a voting plus openness model works well –  I’ve noted previously the cases of the UK, the US, and Sweden.    And, as I’ve noted previously, you have to wonder how local councils, Parliaments, and higher courts manage?  The Supreme Court –  rather more important on the whole than the Reserve Bank –  seems to manage with both voting and disclosure of individual views.  The public –  a priority, if not for the Bank –  seems to be better for it.

Perhaps most extraordinary is the Reserve Bank’s assertion around the appointment process.

In New Zealand, the Governing Committee is a “technical” committee created to carry out the purpose and objectives set out in the Act, the PTA and the MoU. The Reserve Bank is the government’s agent in carrying out its monetary, prudential and macro-prudential policy objectives, and we consider it critical that policy committee appointments be made by the Reserve Bank for their policy competence and not through a political body.

I could understand if they thought that “on balance, we think it would be better” to have appointments made by the Bank itself.  But “critical”……….really?

You have to wonder what makes New Zealand so different from most other advanced countries.    In Australia, the Governor and the Deputy Governor are both appointed by the Treasurer.  In the UK, the Governor and the (various) Deputy Governors are appointed by the Chancellor.  In the UK, the members of Fed Board of Governors are all appointed by the President (confirmed by the Senate). In Sweden, the key appointments are made by the parliamentary committee that oversees the Riksbank.

It is a good practice that major policy decisions should be made only by elected people –  and the Reserve Bank can’t surely pretend their decisions aren’t “major”, including having significant redistributive consequences –  and that where any such powers are delegated they should be made by people appointed directly by elected officials.  Typically, such decisions –  like those of the Cabinet –  will actually be made (legally) collectively.  But for some reason –  that they refuse to state –  the Reserve Bank thinks it should be different.  It shouldn’t.   There are plenty of different models of how Reserve Bank goverance should be done, but a system that keeps single decisionmaking, or which has all decisionmakers appointed by the Reserve Bank itself, should simply be ruled out from the start.

As a final point, the Bank includes a table which they use to support a claim that “about two thirds of the monetary policy committees of inflation targeting central banks have external members”.  It is a pretty shonky table.  For example, they class Sweden as having no external members, when (as noted earlier) most members are non-executive (but fulltime) externals.  They seem to make the same mistake for the Czech Republic, which appears to have several fulltime non-executives.  Weirdly, they claim that the ECB has a majority of outsiders –  but they appear just to mean the Governors of the constituent central banks, who are insiders if ever there were.  But perhaps as importantly, I’m not sure that Armenia, Peru, Hungary, South Africa, Thailand, Guatemala –  none beacons of good governance –  are the sorts of places I’d be looking to for guidance in structuring a robust and accountable decisionmaking system for the central bank.    Not all of the more advanced countries do have externals on their monetary policy decisionmaking committees, but Australia, Norway, Sweden, the US, the UK, Israel, Iceland, Japan, and Korea do.  And of all those advanced countries, only New Zealand and Canada have the sort of single decisionmaker system that the Reserve Bank wants to maintain.

South Africa –  an embattled central bank, facing the increasing prospect of political interference –  doesn’t have externals, but I did find this nice quote on their website

In monetary policy decision-making processes, committees are preferred above individuals. Not one central bank has replaced a committee with a single decision-maker, a fact that has both theoretical and empirical support; the ability to draw diverse viewpoints from constituent members in committees ensures that there is likely to be some moderation of extreme positions and policies and more even policymaking.

Indeed.  We shouldn’t let the Reserve Bank keep such a flawed system, even gussied up with a statutory advisory committee appointed by the Governor himself.   Other countries don’t do it that way –  and our Reserve Bank has far more power than most central banks because of its regulatory functions –  and hardly any other government functions in New Zealand are run that way.

It was good that the Bank included this material in its BIM, rather than just quietly slipping it to the Minister of Finance in a document we didn’t know existed.  It would be better still if they now released the full version of the document making their case. At present, that case is looking pretty threadbare, not informed by either good comparisons or a strong recognition of what open government should look like, and –  if anything –  designed to serve the interests of career bureaucrats rather than of the public.  That’s not too surprising: bureaucrats typically do what they can to protect their bureau.  But it doesn’t make it a good basis for public policy.

As for the Minister of Finance, perhaps he could take a stronger lead by (a) encouraging the Bank to release the fuller paper, (b) ensuring that Treasury releases the Rennie report on similar issues (and associated supporting documents) and (c) actually named the Independent Expert Advisory Panel supposedly playing a key role in the current Treasury-led review of the relevant provisions of the Reserve Bank Act.