Transparency and the effectiveness of monetary policy after the Warsh Review at the Bank of England

Angus Armstrong, Francesco Caselli, Jagjit Chadha, Wouter den Haan

17 March 2015

a

A

Following the Warsh Review of transparency practices and procedures at the Bank of England (Warsh, 2014), the Bank plans to change the structure of its Monetary Policy Committee (MPC) meetings and release its policy decisions, ‘enhanced’ meeting minutes and (once a quarter) the Inflation Report all at the same time (Bank of England, 2014).

There have been critical reactions to the proposals (see Goodhart 2015 and especially Eichengreen and Geraats 2015). In its latest monthly survey of leading UK-based macroeconomists, the Centre for Macroeconomics has canvassed their views on the idea of simultaneously providing the different MPC documents and changing the MPC process.

Background

The Warsh Review, published in December 2014 along with a response by the Bank, describes how transparency could support four strategic policy objectives: “making sound policy decisions; communicating judgments effectively; ensuring accountability for its actions; and creating a fair and accurate historical record”.1

One of the recommendations of the Warsh Review is that “[t]he Bank should make public a concise summary of its policy decision and rationale as soon as is practicable upon the meeting’s conclusion. This summary should encapsulate the MPC consensus. And the votes of individual [Monetary Policy] Committee members should be stated in the policy summary.”2

The Bank responded to this recommendation: “The Warsh Review recommends that the MPC publish a detailed policy statement as soon as is practicable following each policy meeting. The MPC agrees that such a step would improve the effectiveness of its communications and enhance accountability. The [Monetary Policy] Committee believes that there is scope to go further in this direction. It is therefore announcing today its intention to publish both the minutes of its policy meetings and (in the relevant months) the Inflation Report at the same time as its policy decision.”3 To facilitate these changes, the Bank proposes to change the structure of its MPC meetings from two consecutive days to three gatherings spread out over seven days. 

The Bank has also accepted several other proposals contained in the Warsh Review. While this survey focuses only on the timing of the release of information and the change in the structure of the MPC meetings, two of the other accepted proposals have a bearing on the survey’s questions. First, the number of policy meetings will be reduced from twelve to eight.4 Second, the published MPC minutes will be enhanced somewhat so that the discussions on the first day of MPC meetings, which are of a ‘more deliberative’ nature, are reflected more accurately. Currently, the minutes covering the first day are limited to a factual report of economic and financial developments. The Warsh Review recommends that the minutes better capture “the crux” of the deliberations and give “a greater sense of the weight attached by [Monetary Policy] Committee members to competing narratives” to describe recent economic developments.

What should accompany the policy announcement?

Under the Bank’s proposed plan, the MPC will publish a detailed policy statement and rationale on each announcement day.5 This will be accompanied by the simultaneous release of the enhanced MPC meeting minutes and (in the relevant months) the Inflation Report, rather than being published with a two and one week lag, respectively. One possible advantage is that this simultaneous release will be helpful in understanding why certain policy decisions are taken. In the words of Governor Carney, “we believe these arrangements will enhance the effectiveness of our monetary policy communications, making the policy signals we send as clear as possible”.6 Moreover, under the new schedule the market does not have to speculate what new information the minutes and the Inflation Report will contain. A related argument has been put forward by Charles Goodhart (2015), who argues that it may be desirable to reduce the number of monetary news events.

However, one possible concern is that the simultaneous release of so much information may be difficult to absorb and confuse market participants. If true, then this could lead to more volatile market outcomes. Also, the literature on the benefits of public information makes clear that more information is not always better.7 However, here the question is not more versus less public information, but the pace at which the information is released to the public. Indeed, the change in the timing has been discussed in Eichengreen and Geraats (2015) and Goodhart (2015), who question whether policymaking would become more effective under this proposal.  

The first question focuses specifically on the proposed simultaneous release of information. Under the proposed plan for the release of information, the MPC will publish both the minutes of its policy meetings and (in the relevant months) the Inflation Report at the same time as its policy decision. In answering this question, we asked panel members not to take into account that the simultaneous release of information requires a change in the MPC schedule.

Q1: Do you agree that the simultaneous release of the policy decision, the enhanced minutes (including the voting record) of the MPC meeting and (in the relevant months) the release of the Inflation Report will facilitate inference on the likely stance of monetary policy?

Summary of responses

Leaving aside respondents who neither agree nor disagree and the one respondent who did not express an opinion, 65% of the panel members agree or strongly agree that the simultaneous release of all documents related to the MPC process would facilitate inference on the likely stance of monetary policy. When the numbers are weighted with the panel members’ self-assessed confidence level, the fraction of respondents in agreement increases to 71%. Among the members that disagree, nobody strongly disagrees.

Several panel members in support of the simultaneous release argue that a delay between the announcement of the MPC decision and release of the supporting documents implies that there is a period during which it may be difficult to interpret the MPC’s decisions and markets may have to speculate about the reasoning behind its decisions.

Sir Christopher Pissarides (LSE) points out that the simultaneous release ensures that “nothing will be ‘hidden’ by the Bank for a few more days, which could lead to more speculation about its content”. Sushil Wadhwani (Wadhwani Asset Management and former MPC member) observes that “we have had several instances when the Inflation Report press conference has been misleading, because the governor presenting the Report has failed to accurately reflect the weight of opinion on the Committee. Having the minutes appearing simultaneously might well help with this problem”.

Dame Kate Barker (Credit Suisse and former MPC member) mentions that “[p]ast practice often led to the markets interpreting the Inflation Report one way and the minutes a different way. This will now be avoided.” A related argument is given by George Buckley (Deutsche Bank) who comments that “it seems anachronistic that at the Bank’s Inflation Report press conference the governor cannot currently discuss in detail the debate – and the vote – on the Committee at its policy decision a week earlier because the minutes are not published”.

Several of the panel members who do not support the simultaneous release of information are concerned about the possibility that too much information is released and that this could hinder effective understanding of MPC policies and the state of the economy. As mentioned above, the literature provides a rationale for this concern.8

Martin Ellison (University of Oxford) writes that “the most prominent challenge to efficiency [of the ability of markets to process information] is the ‘beauty contest’ idea”, first formulated by Keynes and updated by Morris and Shin (2002). Ellison has “reservations about whether this is really a regular issue for the Bank of England”.

By contrast, other panel members are concerned about the release of too much information. David Cobham (Heriot-Watt University) writes that “there is a real danger of a deluge of information giving the appearance of transparency but in fact just drowning the observers”. However, Richard Dennis (Glasgow) argues that “although the release of greater information may lead to an initial period of learning, market participants are likely to identify quickly the documents and sections of documents that are most relevant”.

Angus Armstrong (NIESR) argues that a drawback of the proposed plan is that “the announcement and enhanced minutes are likely to overshadow the Inflation Report possibly resulting in less analysis and discussion (and feedback to the Bank)”. Jagjit Chadha (University of Kent) emphasises that “the Inflation Report, as a full-blown forecast, can also be thought of as an exercise worthy of scrutiny, per se, and also an opportunity to consider longer-term factors in the policy decisions”.

Finally, more than a few panel members argue that the proposed change to simultaneous release is of minor importance. Andrew Mountford (Royal Holloway) mentions that “it is important that the public and the markets know the rationale and information/data behind policymakers’ decisions. This allows faulty reasoning and analysis to be challenged and corrected and should also make future decisions more easy to predict. Clearly a long delay would reduce transparency but a couple of weeks shouldn’t matter. It would still leave plenty of time before the next interest rate decision.”

Some panel members suggest specific suggestions that would be more effective in enhancing transparency. For example, Simon Wren-Lewis writes that “if the Bank seriously wants to improve its transparency, it should make and publish forecasts of what interest rates will be, as the Fed now does”.

The Bank of England’s proposed MPC process

To enable the simultaneous release of the policy decision, the minutes, and (in the relevant months) the Inflation Report, the Bank proposes to change the MPC meeting process.  Whereas the first question focuses only on the desirability of the simultaneous release of information itself, the second question focuses on the desirability of such a simultaneous release when it is accompanied by the proposed new MPC meeting schedule.

The MPC currently meets on a monthly cycle. This cycle begins with a pre-MPC meeting on a Friday, where Bank staff present to the MPC, followed by two consecutive days of closed meetings on the Wednesday and Thursday of the following week. The first day of the closed meeting is described as “more deliberative” and the second day as “more explanatory and decision-making”. The MPC’s decisions are announced at midday on the Thursday shortly after the MPC meeting has finished.

Writing enhanced minutes and getting this information into the Inflation Report is time-consuming and this means that there has to be a gap between the MPC deliberations and discussions and the announcement.  Specifically, the Bank proposes to change the schedule of the MPC process as follows. First, the deliberating part of the MPC meeting, now held on the day before the announcement, will take place seven days before the Thursday on which the policy decision and accompanying information is published. Second, the explanatory and decision-making part of the MPC meeting, currently held on the morning of the announcement, will be divided into two meetings – the main policy discussion will take place three days before the day of the announcement; the policy vote itself will take place on the evening before the announcement day. Table 1 below provides an outline of the current and the proposed schedule.  

Table 1. Comparison of current and new MPC process

Under the new schedule, MPC meetings will start one full week before the policy announcement day, whereas they start the day before the announcement under the current schedule. The Bank argues that the benefit from this change is that it allows the simultaneous release of a complete set of related documents.  There may be some disadvantages as well. The change in the structure of the MPC meetings from two consecutive days to three gatherings spread out over seven days might affect the coherence of the discussion (and the minutes). Spreading the different parts of the MPC process over a longer time period may create the risk of leaks – actual or perceived – and leave room for a news shock that renders the discussions outdated. However, one could argue that this will not happen very often. Moreover, the actual decisions are made the evening before they are announced so in principle, MPC members can cast their vote depending on the most recent information.

Q2: Do you agree that the Bank’s proposal to release the policy decision, MPC minutes and (once a quarter) the Inflation Report all at the same time justifies a change in the structure of MPC meetings from two consecutive days to a process in which in the MPC meetings are spread out over seven days?

Summary of responses

There is more disagreement among panel members with respect to this second question. Again leaving aside those panel members who neither agree nor disagree and the one panel member who does not express an opinion, 56% of respondents disagree with the statement that the change in the MPC schedule is justified by possible benefits of simultaneous provision of the documents. Among the 44% of respondents in agreement, none strongly agree. Weighted with self-assessed confidence levels, the fraction opposing the proposed change decreases slightly to 53%.

The panel members acknowledge that the simultaneous release of information makes some changes in the MPC schedule unavoidable.  Quite a few of the respondents indicate that these proposed changes carry disadvantages. This is true for respondents who think the simultaneous release of information is in itself beneficial and for respondents who do not think it is.

Among the former, there are panel members for whom the disadvantages of the new schedule are of such magnitude that they no longer support the simultaneous provision of documents.  Specifically, there are five panel members who agree or strongly agree with the first statement but disagree or strongly disagree with the second. In addition, there are three panel members who agree with the first statement but neither agree nor disagree with the second statement.

Some respondents argue that the new schedule has advantages. Dame Kate Barker (Credit Suisse and former MPC member) writes that under the old schedule “there was sometimes less time than I would have liked to reflect on the Wednesday discussion before voting”. George Buckley (Deutsche Bank) suggests that “holding a ‘meeting’ over a week could help focus policymakers’ minds on the broader issues and avoid potentially rash decisions being made on the basis of how policymakers are feeling on any one day – just as holding fewer policy meetings a year should do too”.

One disadvantage mentioned by several respondents is the possibility of leaks. Sushil Wadhwani (Wadhwani Asset Management and former MPC member) writes that “the risk of leaks should not be underestimated. A pretty large number of people will know the likely decision for seven days, something which is pretty dangerous. Remember that the set of people who will know the probable decision will include the prime minister and chancellor (and their special advisers)”.

Michael McMahon (University of Warwick) reminds us that “the push for greater transparency of the Fed in 1993 followed a period of leaked interest rate decisions”.9 But George Buckley (Deutsche Bank) expresses “doubt that the disadvantages related to concerns about leaks are warranted – after all, the Inflation Report itself takes quite some time to put together and by a far larger workforce within the Bank (than the MPC), yet that is never leaked”.

Another disadvantage put forward by the panel is the possibility that decisions do not fully incorporate new information that becomes available during the MPC process. Gianluca Benigno (LSE) writes that “spreading the MPC meeting over seven days seems to me quite a lot and it is likely to make the minutes outdated when new information comes in during that period”. In contrast, Wouter Den Haan (LSE) thinks that “the chance that something completely new and relevant happens during the week of the MPC process is simply not that big”.

Again, several panel members question whether the changes are substantial. Dame Kate Barker (Credit Suisse and former MPC member) writes “having these documents out at the same time as the vote is rather ambitious. I don’t think putting them into the following week would have mattered.” John Driffill (Birkbeck College) writes “financial markets may like these changes, believing they will make more money with the additional information. The public in general and the ‘real economy’ will notice no difference.”

References

Barwell, R and J S Chadha (2014), “Publish or be damned – or why central banks need to say more about the path of their policy rates”, VoxEU.org, 31 August.

Eichengreen, B and P Geraats (2015), “The Bank of England fails its transparency test”, VoxEU.org, 6 January 

Goodhart, C A E (2015), “The UK MPC process in the light of the Warsh Review”, VoxEU.org, 2 March.

Hansen, S, M McMahon and A Prat (2014), “Transparency and deliberations within the FOMC: A computational linguistics approach”.

Morris, S and H S Shin (2002), “Social value of public information”, The American Economic Review 92(5), pp. 1521–34.

Lepetyuk, V and C A Stoltenberg (2013), “Policy announcements and welfare”, Economic Journal 123, pp. 962-997.

Lindsey, D E (2003), “A modern history of FOMC communication: 1975-2002”, Federal Reserve Board of Governors.

Footnotes

1 The Warsh review is available at http://www.bankofengland.co.uk/publications/Documents/news/2014/warsh.pdf. For the full background and documentation please see http://www.bankofengland.co.uk/publications/Pages/news/2014/168.aspx.

2 See page 7 of the Warsh review.

3 See page 2 of the Bank’s response, which is available at http://www.bankofengland.co.uk/publications/Documents/news/2014/warshres....

4 The Bank is minded to move to eight policy meetings a year, but this will require a change in the Bank of England Act (1998).

5 In the past, the Bank did typically not provide a rationale if there was no change in policy.

6 See http://www.bankofengland.co.uk/publications/Documents/news/2014/spnote11....

7 See, for example, Morris and Shin (2002) and Lepetyuk and Stoltenberg (2013).

8 Also, see Barwell and Chadha (2014).

9 See Lindsey (2003, pp. 1001-111) and Hansen et al. (2014).

a

A

Tags:  Bank of England, transparency, Monetary Policy Committee, Warsh Review

Director of Macroeconomic Research, National Institute for Economic and Social Research

Professor of Economics at the London School of Economics

Director, National Institute of Economic and Social Research; Professor of Economics, University of Kent

Professor of Economics and Co-Director of the Centre for Macroeconomics, London School of Economics and CEPR Research Fellow