The Bank of England Ten Years On - a view from Frankfurt

Hans-Helmut Kotz, of the Executive Board of the Deutsche Bundesbank and honorary Professor at Freiburg University offers a personal view of ten years operational independence at the Bank of England.

Monetary policy - unexciting on purpose
Keeping inflation at bay has become the (almost) unequivocal target of most central banks. This is an objective which has been largely achieved by the Bank of England — with inflation (as measured by the RPIX) hovering about 2.5 per cent ever since the mid 1990s. Thus, the BoE has met its objective as defined in the Bank of England Act of 1998, that is: to maintain price stability. Keeping promises generates reputation, in this case of delivering a stable monetary environment. And the return on such an ac-cumulation of reputational capital translates into lower inflation premia in long-term nominal interest rates. This is what we saw as well in sterling bond prices: expecta-tions have been rather firmly anchored at the level of the inflation objective the BoE is striving for — at least until recently. Clearly, monetary policy has become much less exciting and therefore the BoE delivered as well on its at times (and tongue in cheek) made promise: namely, to become as unexciting as Keynes' humble, competent dentist, possibly even boring.

Against this background, we could just as well end here. But there are at least three dimensions, in which some excitement can be detected. First, from an historical per-spective, this change of tack of the UK towards macro moderation is rather remarkable. And it would be interesting to know where the ‘nice’ (non-inflationary consistently expansionary) economy features come from. Second, if, as the domi-nant answers to the first question imply, it is the independence of the BoE as well as its strategy which explain to a significant extent this remarkable performance this would have obvious lessons for monetary policy in general. Fortunately, in the de-bate on the optimal approach to monetary policy there are still some controversies left — at any rate at a theoretical level. Finally, one might wonder about potential challenges to this gentle boredom. Are there developments, possibly emanating from financial markets, which threaten the prevailing tranquility?

Involuntary search for new guiding principles
Exchange rate crises have been instrumental in two prominent cases in the search for monetary strategies: money supply targeting and inflation targeting. The unravel-ing of the Bretton-Woods-System of fixed exchange rates led the Bundesbank in the mid 1970s to opt for an intermediate target, a two-staged strategy, that is, to control money supply - pragmatically. The BoE’s decision to go instead for the direct con-trol of the ultimate target — inflation containment — was the result of the UK leaving the European exchange rate system in September 1992. Interestingly, and with the benefits of hindsight, both approaches produced rather remarkable results. This is because both approaches have apparently rather diverse conceptual backgrounds. Appearances can, however, be deceiving, a point to which we return.

Two defining characteristics are held responsible for the success of the BoE’s monetary policy. The first is (instrument) independence, which was given to the BoE by the incoming Labour government on May 6 1997.1 The second defining institutional innovation was the new strategy to be pursued - namely, inflation targeting. Of course, IT had been introduced earlier, in October 1992. While all of this is unremarkable from today’s angle, in the mid-1990s both innovations were publicly controversial. Essentially, the case had to be made that macro stability is conducive to medium-run growth. This was forcefully reasoned by a young journalist in a pamphlet of the Fabian society — Ed Balls.2 Ultimately, this argument, frequently raised by economists, meandered its way onto the platform of the Labor party.

Focusing primarily on inflation control was not generally accepted in the mid 1990s. At that time the dominant view held that the mandate of central banks also included providing for maximum output and hence employment. Seeing the purpose of central banks mainly in underwriting monetary stability therefore was, in the words of Stanley Fischer, a recent phenomenon: ‘(C)entral banks have increasingly come to emphasize the fight against inflation and to deemphasize the possibility that mone-tary policy can affect the level of output.’3 Indeed, this is a recurring topic. The US Fed still has such a dual mandate. And during the French presidential campaign it became evident that similar views on reducing output fluctuations (and fostering trend growth) have a positive resonance in EMU also.

Neither was it conventional wisdom 10 years ago to render monetary policy to an autonomous institution, be it only for instrumental purposes. However, influential academic papers produced convincing evidence on the correlation between degrees of independence and price stability since the late 1980s only. It is interesting to note that in the UK’s case IT was introduced to support the BoE in its advisory capacity to the Treasury which ultimately bore responsibility for the conduct of monetary policy.4 Since, as the French say: comparaison n’est pas raison, it is nonetheless important to understand where the support for a stability oriented policy actually comes from. After all, correlation is not causation. Thus, while independence is helpful without any doubt, the public perception of what monetary policy can really achieve is of the essence. Adam Posen correctly made the point that in the German case an important underlying explanatory variable was the strong dislike of inflation. This inflation aversion of the general public supported the Bundesbank substantially — over and above its legal independence.

Two procedural innovations
Nonetheless, the innovations have born fruit rather rapidly: Inflation risk premia fell out of bond prices within a short period of time. The anchoring of inflation expectations was rapid and tight. And this most plausibly had to do with procedural innovations as well. Two are particular relevant.

Aiming at the ultimate objective directly was — as the debate about intermediate target had shown — prone to substantial difficulties. Therefore, somehow out of frustration about the controllability of the final objective(s) directly, intermediate targets were introduced. By exploiting the information in these intermediate variables the monetary authorities should achieve their ultimate goals more effectively.5 In actual practice, the BoE did not dispense with the intermediate information variables. But, through its Inflation Report it intended to give a disciplined and structured evalua-tion of the transmission mechanism - including uncertainties in evaluation. Thus one could state that ‘(t)he Bank’s inflation projection — when taken alongside the other intermediate variables which make up the inflationary assessment — influences monetary policy decisions in much the same way as does any other intermediate variable’.6

Moreover, monetary policy was to be decided on by a committee of individually accountable members. This being an innovation in at least two ways: the MPC was not assumed to decide by consensus (which means, by the way, not unanimity) but majority — to lay bare opposing views in light of possibly diverging perspectives on the same data (the publishing of, for sure, edited minutes and voting records in-cluded). Moreover, in the Inflation Report, of which the MPC took ownership, uncertainty about the possible trajectory of inflation was frankly admitted. The fan chart made clear the fragility of the central forecast, which was ultimately guiding an explicitly forward-looking monetary policy.

Best practice?
All of the above would suggest that the BoE's approach has become the best prac-tice way of doing things central banking. But not all share this view. The ECB does not conduct monetary policy in light of an IT approach. And the US Fed is still rather reluctant.

The ECB’s two pillar approach, with a particular emphasis on money supply, has been developed explicitly in response to the experience with money supply targeting on the one hand and inflation (forecast) targeting on the other. Issing et al. emphasize that ‘in spite of the rigidity of theoretical monetary targeting and inflation tar-geting, judgment has been a crucial component of both monetary strategies’. And they add that the ECB’s two-pillar strategy (a short-horizon, broadly-based econmic analysis, cross-checked against a long-term oriented monetary analysis) has been conceived to face ‘the high degree of uncertainty and imperfect knowledge prevailing at the beginning of Stage Three of EMU.’7 Flexibility here means that the mone-tary framework should allow for supply (or financial market) shocks in order to con-tain output instability. This is why the ECB’s inflation norm is to be honored not at any moment in time but over the medium-run.8

In the US, which comes from exactly the opposite direction, IT on the other hand has been advocated as ‘a framework, not a rule’ to allow for ‘constrained discretion’. Proponents of IT highlighted its flexibility, in particular by pointing out that IT is not ‘falling on the rule side of the traditional dichotomy’ (between rules vs. discretion).9 This point has been stressed not only in order to allow for the inevitable ‘judgmental content’ of monetary policy (the ‘art’ part). Highlighting the discretionary option was (is) important in the US context to suggest that IT would not prevent appropriate responses to shocks. Otherwise, IT’s compatibility with the dual mandate, which Congress has given to the Fed, would be in doubt, making it clearly politically unpalatable. Almost a quarter of a century ago James Tobin wrote: ‘…monetary policy cannot be governed by irrevocably fixed rules blind to actual outcomes, …policies responsive to events cannot be prescribed fully in advance but ultimately depend upon discretion, monetary authorities cannot escape responsibilities for real economic outcomes of significance to the society’.10 Interestingly, Bernanke and Mishkin dealt in their article with the question whether nominal GDP targeting, ‘which can be thought of as “velocity corrected money growth”, could not as well be a sensible goal variable. And they answered in the affirmative -—namely that nominal income targeting, which was at that time incidentally the favor-ite of many US economists, was 'generally consistent with the overall strategy for monetary policy as discussed in this article’.11

Does IT and independence matter?
We write these lines (on July 5th) a few hours before the BoE decides on policy rates. In putting themselves into the MPC members' shoes analysts as well as inves-tors (cf. implied interest rates) are betting on a tightening move of 25 bp. These expectations are anchored in the BoE’s communication, which has been highlighting the underlying dangers of inflation remaining stubbornly above the medium run ob-jective. Markets, in other words, take their cue of the economic environment as if they were reading the world through an Inflation Report perspective. Though there have been two surprises of late, the BoE is so transparent that interest-rate expecta-tions are fairly robust — the low-level of implied volatility testifies to this point.

Now, while conceptual differences between approaches are frequently sharply stressed, it is interesting to see whether policies, as actually practiced, do really show such a variance in actual implementation. This is perhaps the point where a postmodern linguist (à la one of David Lodge’s characters) could become helpful. In deconstructing and re-interpreting the wording one could, of course, rather easily render for example the Bundesbank’s policy approach as implicit inflation targeting — as have done Bernanke et al. The ‘unavoidable inflation’ (or later: the ‘price norm’) as a core part of monetary policy formulation as well as implementation make distinctions in some eyes arbitrary. In fact, on the conceptual level one could assert that the Bundesbank’s ‘pragmatic monetarism’ has stressed more the qualifier than the noun. In pondering an explanation by Helmut Schlesinger, Chief Economist and later President of the Bundesbank, of the Bundesbank's money supply targeting (in those days central-bank-money, not M3) Alan Blinder came to the conclusion that ‘(i)t is hard to imagine a clearer description of a purely discretionary re-gime…(y)et no one doubts the Bundesbank's anti-inflationary zeal’.12

Some even claimed that the Fed under Alan Greenspan followed the IT approach, quoting his famous (though characteristically vague) remark from 1989 that the Fed would try to control monetary policy in order to produce ‘price levels sufficiently stable so that expectations of change do not become major factors in key economic decisions.’ Such a proposition, however, would somehow amount to a reduction of IT ad absurdum. Nonetheless, when actual monetary policy, is located on a spectrum between the poles of art and science, conceptual differences can easily be over-rated. Jürgen von Hagen and Manfred Neumann, in our view correctly, hold that 'IT, like other monetary policy strategies, must be seen in the context of (economic) culture and traditions…Giving the central bank’s commitment to price stability and its willingness to bind its policy to an intermediate target…the choice between an inflation target or a monetary policy aggregate then is probably more a question of culture than economic principles.’13 Empirically, in other words, it’s impossible to tell the difference. In the dark of the night all cats look grey.

In any case, IT and instrument independence, the two 1990s institutional innova-tions, did matter in the UK’s case. They contributed, over and above, structural factors to the great macro moderation. As a result of the observational similarities between the pragmatic monetarism, IT and the two pillars they do face, however, the same challenges.

We conclude by alluding to two: first, Benjamin Friedman forcefully argues against IT because, in the US context, it would, in his eyes, dilute the dual mandate, ‘the objectives beyond price stability’. Moreover, Friedman holds that words should matter and complains 'that under inflation targeting policymakers normally reveal to the public only one of their multiple targets — which would obviously undermine transparency. Friedman once remarked that he mainly liked what the ECB does, but does not at all appreciate how she phrases it. This is the whole issue of communication and transparency, which, of course, exists for a while. In any case, Friedman posits that predictability would be clearly buttressed if all the reasons un-derlying policy formation were laid out. In former, ‘opaque’ times, central bank watchers have been less exacting. Paul Samuelson, for example, argued ‘For fifty years the Fed has not seemed to me mysterious or perverse. Like the old farmer who found his donkey by asking himself, “Where would I go if I were a jackass?” I could mostly guess in which direction the Fed would move. This despite its own talk.’ 14

Second, in trying to gauge the transmission of its impulses, monetary policy still focuses very much on its impact on bank balance sheets. With the ever increasing importance of financial markets, in particular markets in pricing of credit risk, accounting for the consequences of endogenously created liquidity becomes ever more important. This is, admittedly, a problem in particular for a money supply oriented concept. On the other hand, carefully looking at growth rates of money and credit or, in a more general sense, broad liquidity is of the essence for ‘anyone who believes that default, risk aversion and income constraints matter…such interpretation will be an art. Nevertheless it is an art worth attempting.’ 15

Meanwhile, and this brings us in a way back full circle. One hears voices out of the BoE that money and liquidity might again bear some consideration. If appearances are not completely deceiving, here we are talking about ‘pragmatic monetarism’, somehow reminding us of the Frankfurt implicit IT variety.


1. The BoE is however not goal independent: The Chancellor of the Exchequer, in giving the BoE its remit with a precise content, called upon the BoE to strive for an annual rate of increase in the RPIX of 2.5 % ‘at all times’. Since December 2003, the RPIX has been substituted for by a CPI target of 2 %, in recognition of the fact that the CPI increases are on average 0.5 percentage points below the RPIX. Moreover, mindful of the hierarchical ordering of priorities, the remit challenges the BoE to support growth and employment.

2. E Balls (1992): Euro Monetarism: Why Britain was ensnared and how it should escape, London: Fabian Society.

3. S Fischer (1995): Modern Approaches to Central Banking, NBER Working Paper no. 5064, 2.

4. See A Haldane (1995): ‘Inflation targets’, BoE Quarterly Bulletin, August, 250-274 and Charles Bean (2004): ‘Inflation targeting: the UK experience’, Perspektiven der Wirtschaftspolitik, vol. 5, 405-21.

5. B Friedman (1984): ‘The Value of Intermediate Targets in Implementing Monetary Policy’, in Fed Kansas, Price Stability and Public Policy, Jackson Hole, 169-171 and V Chick (1973): The Theory of Monetary Policy, Basil Blackwell: Oxford, 8-13.

6. Haldane, ibid., p. 252.

7. See O Issing et al. (2001): Monetary Policy in the Euro Area. Strategy and Decision-Making at the European Central Bank, CUP, Cambridge, 104-105.

8. See for a very concise and convincing explanation of the ECB’s strategy L Papademos (2006): The role of money in the conduct of monetary policy, ECB.

9. B Bernanke and F Mishkin (1997): ‘Inflation Targeting: A New Framework for Monetary Policy?’, Journal of Economic Perspectives, vol. 11, no. 2, 97-116, here p. 104.

10 James Tobin (1983): ‘Monetary Policy: Rules, Targets, and Shocks’, Journal of Money, Credit and Banking, vol. 15, no. 4, 517.

11. Ibid., p. 112. Bernanke and Mishkin confessed a mild preference in favour of IT (relative to nomi-nal income targeting) for mainly three reasons: timeliness and availability of inflation data, substan-tial inherent flexibility of IT and the public's better understanding of the concept of inflation.

12. A Blinder (1987): ‘The Rules-versus-Discretion Debate in the Light of Recent Experience’, Weltwirtschaftliches Archiv, vol. CXXIII, 399-413.

13. J von Hagen and M Neumann (2002): ‘Does inflation targeting matter?’, Fed St Louis, Quarterly Review, July/August, 145.

14. P Samuelson (1994), Panel Discussion, in: J Fuhrer (ed., 1994): Goals, Guidelines, and Con-straints Facing Monetary Policymakers, Fed Boston, 231.

15. C Goodhart (2007): ‘Whatever became of the Monetary Aggregates?’, Lecture in honour of Maurice, Lord Peston.

Page Options