By John Kemp
LONDON Nov 2 David Stockton's gentle review of
forecasting at the Bank of England, published on Friday, does
not hide the cultural problems which have caused systematic
errors in the central bank's inflation predictions and undercut
the credibility of its forecasting regime over the last five
Despite employing dozens of the brightest economists in the
country, and spending more than 66 million pounds ($106.5
million) on monetary analysis last year, the Bank has
consistently over-predicted growth and under-predicted inflation
Its performance is worse than private forecasters, as
Stockton -- once of the Federal Reserve -- noted.
"Fundamentally, the forecast process and the associated
forecasting tools employed by the Bank in support of its
monetary policy decision-making are sound," according to
Stockton, a former director of research and statistics at the
Federal Reserve in Washington engaged to conduct a review of the
Monetary Policy Committee's forecasting capability.
"Potential for improvement should not be read as an
indictment of the current forecast process at the Bank of
England and the MPC," Stockton wrote.
The evidence presented in the report suggests a much deeper
The report contains a useful review of the factors the Bank
believes have thrown its forecasts off course, and a welter of
sensible recommendations for improving the forecasting process
and presenting the outcome to the public. In many ways, the
Stockton report is a model "lessons learned" exercise. But it
skates around the two fundamental problems:
(1) Did over-centralisation of power in the Bank, especially
the dominant role played by its intellectually brilliant
governor, Mervyn King, promote a culture of "group think" within
the MPC and among Bank employees that discouraged anyone from
challenging the views of the governor and the "best collective
judgment" of the MPC even when proved (repeatedly) wrong?
(2) Were the Bank's inflation and output forecasts adjusted
to justify policy, rather than the other way around? Did the
Bank massage its predictions to show its strategy would work,
rather than reconsider its strategy in the light of incoming
data and changes in the likely outlook?
Stockton's review contains some useful recommendations. But
the real problems stem from the Bank's culture. If the Bank is
to improve its forecasting performance, it needs an extensive
shake-up of senior personnel and a new, much more open and
Stockton explains that the Bank's forecast errors have been
much worse in the five years following the eruption of the
financial crisis than in the first decade after it was given
operational independence in 1997, and significantly worse than
private sector forecasters.
Insiders blame a series of unusually large shocks. Stockton
accepts those explanations as "persuasive" though he questions
whether "the serial persistence of the MPC's recent errors
represents only a string of bad luck, points to some slowness in
responding to these errors as they became apparent, or reflects
a deeper flaw in the analytical framework".
But the really significant thing about the errors is that
they were all in one direction (too much growth, too little
inflation). If the Bank's forecasts were really unbiased, upside
and downside errors on growth and inflation should have been
much more evenly distributed.
At times, senior officials have suggested the over-run in
inflation was due to a series of unforeseen shocks. At other
times they have suggested the Bank has deliberately "looked
through" price increases and been prepared to "tolerate" a
limited overshoot to pursue other goals.
The first set of explanations suggests the overshoot was an
accident, the second that it was a deliberate by-product of
The inconsistent explanations have contributed to a sense
the Bank's flawed forecasts were a cover for a strategy
policymakers had already settled on: massive monetary stimulus
and a willingness to accept an inflation overshoot to balance a
programme of tax rises and spending cuts.
All the Bank's errors were in the same direction because
officials believed their own strategy would succeed in restoring
growth, while inflation would quickly return to target without
the need for any rise in interest rates or lessening of asset
Would a genuinely independent forecaster, who had no
responsibility for formulating policy but only in assessing
likely outcomes, have made the same errors as consistently, and
been so slow to change their predictions?
In the area of tax and spending, policymakers have accepted
the need for greater separation between policy formulation and
the assessment of its likely effects, with the creation of the
independent Office for Budget Responsibility.
But in monetary policy, forecasting and policy choices
remain uncomfortably commingled.
Reviewing the process among the major central banks,
Stockton notes: "One nearly uniform feature of the published
forecasts is that they are "owned" by the policy makers and not
by their staff (the one major exception is the ECB).
"This seems generally as it should be -- policy actions
ought to be conditioned on the views of the policymakers."
However it is not just politicians who may be tempted to
tweak forecasts to justify their strongly held beliefs.
The MPC seems to have been slow to adjust its forecasts when
its strategy did not deliver intended outcomes, instead arguing
the plan would work eventually and should be given more time.
Stockton's recommendations would build up more professional
expertise among the Bank's forecasting staff by slowing the pace
of rotations "to develop a staff with enough experience and
analytical depth on forecast-related issues to challenge members
of the MPC in a constructive manner".
The unstated implication is that the staff are not
sufficiently robust in challenging the MPC at the moment.
More tentatively, he suggests the staff might be encouraged
to produce its own forecasts, which might be published "with a
lag of sufficient length so as not to distract from the
forecast" of the MPC itself.
"Knowing that the staff forecast would eventually be
published could very well sharpen the focus of both the staff
and the members of the MPC," Stockton explains. It would be a
form of internal discipline.
"Some cultural adaptations would be required by the Bank,"
according to Stockton, which is an understatement. "The
development of a staff forecast would be a very big step."
Therein lies the problem. It would imply a uncomfortable
decentralisation of power away from the MPC and towards the
professional staff, and away from the governor and to more
junior but specialised officials. It would reverse the strong
centralising trend evident over the course of King's
Dissenting votes within the MPC have fallen sharply,
especially among the five Bank insiders, and especially in the
last few years. Despite the huge uncertainties over the outlook,
the Bank's insiders voted as a block as never before. King's two
deputies appear to have been particularly unwilling to challenge
If the top of the Bank is affected by group think, staff in
more junior positions can hardly be expected to demonstrate more
The selection of a new governor to replace King when he
retires next year was outside Stockton's remit. The basic choice
lies between continuity (promoting Bank Deputy Governor Paul
Tucker) and change (appointing any of the outside candidates for
the job, led by Adair Turner, currently chairman of the
Financial Services Authority).
Tucker has given King unswerving support over the past five
years. Supporters will argue that the sometimes uncomfortable
job of the deputy is to support the boss. Detractors will
counter that Tucker has not sufficiently challenged the governor
and is deeply implicated in the forecasting errors of the last
Tucker remains the front-runner. But if he is promoted, it
is hard to see how he can bring about a radical improvement in
the forecasting process, given how closely he is associated with
the failures of the past.
It would be better to use King's retirement to conduct an
extensive reshuffle of the Bank's leadership and change its
overly conformist culture in a bid to improve its performance in