As 2010 progressed, it became widely acknowledged that the British economy was over the worst of the Great Recession of 2008-09 and that a ‘double dip’ sequel was increasingly unlikely ( The Economist , 2010b). Despite negative growth figures for Quarter 4 of 2010, most continue to hold this view. Use of fiscal stimulus is now being scaled back to the point at which the United Kingdom is beginning to confront its large budget deficit, whilst leaving to the monetary authorities the job of supporting the recovery through loose monetary policy ( The Economist , 2010a). In light of these developments, it is now time to look to the medium term and how macroeconomic policy should be conducted in light of the seismic events experienced since 2007.
This paper does just that, and in doing so advocates a new mandate for the Bank of England in which responsibility for sustainable growth in the banking system is added to their existing target for inflation. The author shall first outline his assumptions before considering the performance of inflation targeting since its introduction in 1992. The paper shall then note the merits of the M4 indicator of money supply and make a recommendation for the new dual mandate mentioned above. A number of objections to such a mandate are then considered.
The author should make clear early on the key assumptions that underpin his analysis. Firstly, the analysis is conducted within what may be called a ‘broad money monetarist’ framework, most prominently represented in the United Kingdom by Tim Congdon. That is to say that the growth in broad money, with broad money made up of notes and coins in circulation and the banking sector’s deposit liabilities, is assumed to have a strong causal impact on the growth of nominal income in the medium to long term. Within the context of the British economy the principal measure of money supply that captures this has been, since a change in the behaviour of building society deposits during the 1980s, the M4 measure. 1
This broad money outlook is underpinned by two significant assumptions, whilst a third assumption moderates it somewhat. Firstly, given that that there is a causal link running from broad money growth to national income growth and that growth in real output cannot be stimulated through monetary expansion in the medium and long term, it can therefore assumed that excessive growth in broad money will inevitably lead to higher inflation. Secondly, and consequently, the Phillips Curve is vertical in the long run, as first proposed by Friedman (1968) and Phelps (1968). Given that present monetary arrangements throughout most of the world indicate that policymakers tacitly accept these two assumptions (Goodhart, 1995: 60-63), it is hoped that those reading this paper will do likewise.
The third assumption should also be relatively uncontroversial, though some monetarists may object to it. Whilst the growth of money, broadly measured, is taken to be the cause of inflation, the author believes the velocity of circulation to be too unstable in the short term for a money supply target, whether M4 or otherwise, to be effective (Goodhart, 1995: 62). If velocity of circulation were invariably stable then there would be a strong case for a return to monetary targeting. But monetary targets were given ample time to bed down in a large number of developed economies during the first half of the 1980s. With the partial exception of Germany they proved unsustainable in the long term, despite their key role in ending the Great Inflation of the 1970s (Goodhart, 1989: 300-308). The author has no wish to advocate winding back the clock twenty five years or more.
The Testing of Inflation Targeting
1992-2007 saw a remarkable improvement in the economic performance of the United Kingdom, with low and stable inflation allied to consistent expansion in the economy. Whilst the role of the microeconomic reforms of the 1980s in making product markets freer and labour markets more flexible should not be overlooked, the macroeconomic framework used throughout the period has also been an object of praise (Budd, 2005: 29). Following the limited successes of money supply targets and Britain’s calamitous exit from the Exchange Rate Mechanism (ERM) system of fixed exchange rates, policy was reconstructed in September and October of 1992. An inflation target range was introduced, and policymaking was opened up considerably for the purposes of establishing market credibility (Lamont, 2006: 163-164). Progressive reform from 1992 to 1997, culminating in independence for the Bank of England, strengthened the framework further (Lamont, 2006: 163-164).
In fairness a number of people, including the politician most responsible for their introduction in the United Kingdom, warned long before 2007 that inflation targets did not mean ‘the end of history’ in macroeconomic management (Lamont, 2006: 164). But a certain amount of hubris nevertheless built up around inflation targeting, and its nemesis came in the credit crunch of 2007 and the subsequent recession of 2008-09 which proved to be the deepest since the horrors of the 1930s. In the search for scapegoats inflation targeting can be charged with failing to give an appropriate warning with regards to the massive pre-2007 asset boom. The asset boom was an early symptom of the problems to come, and its collapse would later play an important role in undermining the balance sheets of the financial sector.
The essential problem with inflation targeting is not a complicated or a new one. Inflation is a variable that lags greatly, and the link between it and monetary policy is an uncertain and complex one (Friedman, 1968: 16). It has proved a highly effective nominal anchor when acceptable levels of inflation and inflation expectations have already been established, but in more unstable times inflation has provided nothing better than a rear view mirror by which to steer. The activist policy responses of the Bank of England from 2008 onwards, which included deep cuts in interest rates and the buying up of bonds with newly created money (quantitative easing – QE), were certainly not being driven by reference to an inflation rate that remained stubbornly above target for most of the period. Had the policy response been dictated by reference to the inflation target alone, as it seemed to be for much of 2008, an even more disastrous collapse in output would likely have resulted.
If the inflation rate failed to see the crisis coming and provided a poor anchor for policy after 2007 this should not surprise us. The introduction of inflation targeting in developed economies, which began in New Zealand in 1988 (Bernanke et al., 1999, 86-114), came at a time when the Great Inflation of the 1970s had already been tamed by most of the developed economies. It should be noted in accordance with this that inflation in Britain had already been reduced to negligible levels by the time inflation targeting was introduced in 1992. In the specific case of the United Kingdom, inflation had been broken by the blunt but successful (in disinflationary terms, at least) tool of ERM membership (Budd, 2005: 26-27). A polemical critique of inflation targeting would consider it little more than a ‘flat track bully’ which had the good fortune to be introduced at a time when conditions for successful counter inflation policy had never been better. In this context, almost any sane framework for macroeconomic policy could have achieved similar results.
The author should not like to go so far as that. Inflation targeting has its problems, like all macroeconomic policy frameworks. But as a framework for ‘constrained discretion’ in the post-monetary target world it has genuine merits (Bernanke et al., 1999: 10-25) and, despite 2007-10, a strong record. To abandon it completely would be folly. But on its own, and particularly if a less than auspicious macroeconomic climate remains the norm for some time to come, it does not provide the effective lodestar that it once appeared to. The remainder of the paper is therefore devoted to developing a more robust framework for macroeconomic policy in light of this.
M4 and Broad Money
As highlighted in the previous section, a potential criticism of inflation targeting relates to its failure to force a reaction by policymakers to the asset boom that developed in the run up to the Credit Crunch. Asset booms have almost always spelt trouble for the United Kingdom economy in recent times, with the 2005-07 episode now added to the 1972-73 and 1986-89 experiences (Congdon, 2005b: 56-70) in leading to economic grief. But asset booms are generally a symptom of the underlying problem as opposed to its direct cause.
The most likely culprit, in terms of the direct cause, has usually been a rapid growth in money broadly measured (Congdon, 2007: 281). This has been a feature of all three of the asset booms mentioned above. Congdon (2006) was among those who gave advance warning of such problems in the period leading up to the 2007 crisis, in his case with reference to the rapid growth of financial sector money holdings. Low and stable growth of M4 was a feature for all but the later stages of the 1992-2007 ‘Great Moderation’ (Greenwood: 2009, 37-43), and it is difficult to believe that economic outcomes would have been significantly different during this period had an M4 target rather than an inflation target been made the centrepiece of macroeconomic policy in 1992.
Broad money also has the advantage, from the perspective of the theorist, of having a credible ‘transmission mechanism’ by which monetary growth leads to growth in nominal income. If demand for money is relatively stable among economic agents, people will seek to hold a fixed proportion of their income in money. It was long ago laid out how, if people’s holdings of money rose above this desired ratio, they would seek to return to this desired ratio by spending their excess money balances (Fisher, 1912: 242-247). Since money is only transferred within a closed circuit as people spend it, the only thing that can adjust and allow for a return to monetary equilibrium is a rise in national income. It has since been highlighted how only a broad measure of money is relevant within this transmission mechanism (Congdon, 1992: 182-183). In today’s British economy, the relevant broad money indicator is M4.
Despite its obvious qualities both theoretically and empirically, M4 should nevertheless not be held up as a magic bullet for conducting future macroeconomic policy. In the early 1980s £M3 (then the relevant broad money indicator) was the nominal anchor for policy at a time when the authorities were seeking to conduct a gradual disinflation (Goodhart, 1989: 302-305). Despite an explosion in £M3, the government largely ignored the implied message that monetary policy was too loose. Yet a rapid rather than gradual disinflation was nevertheless carried out successively (Goodhart, 1989: 302-305). In this case, a rapid increase in the demand for bank deposits had occurred that was related to the abolition of exchange controls and the ending of quantitative credit restrictions on banks (Congdon, 1992: 83-94). This episode warns against putting too much faith in M4 or any other monetary aggregate within the context of deregulated financial markets.
Congdon (1989) has argued that the early 1980s episode constituted a ‘one-off’ step-change in people’s demand for deposits as a proportion of their income due to unusually rapid financial deregulation, and that a return to a broad money target after 1985 would have resulted in superior macroeconomic management throughout 1986-92. This is a proposition that cannot be tested, but Congdon does emphasise that any broad money target needs to be seen in a ‘medium term’ perspective with overshoots and undershoots of such targets being placed in context and not acted on mechanically if policy errors are to be avoided (Congdon, 2005a: 52).
The author by and large agrees with Congdon, but is far less optimistic that policy can be run in such a manner when money supply targets can be overshot and undershot by vast amounts in the short term due to the instability of the velocity of circulation within a short timeframe. Inflation targeting in the United Kingdom could become problematic in the near future with consistently above target inflation eventually to higher inflation expectations, whatever the Bank of England may say about inflation returning to target in the medium term. The problems of an M4 target with regards to inflation expectations should M4 grow above its prospective target rate, with the Bank pleading that this is merely down to more rapid-than-usual financial innovation, can only be imagined. In light of this, this article rejects the idea of returning to a broad money target in spite of M4’s obvious qualities.
A New Dual Mandate
A development that should be noted is the return to the Bank of England of the powers to supervise banking. This follows the widely acknowledged failure of the tripartite structure in which authority was split between the Financial Services Authority, the Treasury and the Bank of England. In light of this development, what is required is a new mandate for the Bank of England that ties together supervision of the banking sector, an inflation target and a more privileged role for M4. This could help to avoid the sort of asset booms and the resulting problems that have bedeviled macroeconomic management in the United Kingdom in recent times.
The author will firstly state upfront that an inflation target should remain as part of the mandate from Parliament. The inflation index used and the specific rate targeted are technical matters which need not concern us in this paper. The reasons for the maintenance of the inflation target were highlighted in an earlier section. But the caveat given earlier was that an inflation target alone would no longer suffice in light of the weaknesses displayed by the inflation targeting regime in the run up to, and since the beginning of, the crisis of 2007-10.
The corollary to the existing mandate should be an obligation on the Bank of England to ‘ensure sustainable growth in the banking system’. This can be done through the use of interest rates to restrict the speed at which banks can expand their balance sheets. There are two reasons for expanding the Bank’s mandate. Firstly, this will provide a suitable obligation on the Bank in relation to its reacquired role as supervisor of the banking sector and it can answer to parliamentary committees accordingly. But the second reason is perhaps more important. A banking sector that grows too slowly relative to the output of the economy is analogous to a contraction in the money supply, and is likely to result in deflation as was feared in 2008-09. In contrast, a banking sector growing at a much more rapid rate than the economy will inevitably become associated with asset booms and inflation, with the sort of disaster that occurred in 2007-10 a possibility. As such, sustainable growth in the banking sector would undoubtedly lead to better economic outcomes (Greenwood, 2009: 41).
M4, consisting mostly of the banking system’s deposit liabilities (notes and coins in circulation form only a tiny proportion of M4), provides a good proxy for growth in the banking system. A mandate for ensuring sustainable growth in the banking system therefore constitutes an informal target for M4, with all the benefits of such a target but without the drawbacks highlighted in the previous section. The unique qualities of M4 in warning of asset booms and inflations as well as asset price collapses and deflations (Congdon, 2005b: 87-107) could then take a more central role in policymaking. But short term instabilities resulting from the likelihood of variations in the velocity of circulation in the short run can be safely ignored, with policy decisions instead taken within the medium term framework called for by Congdon (2005a). Additionally, the maintenance of the inflation target will continue to provide a numerical target by which to hold the Bank to account.
There are two immediate objections to such a mandate which I will now confront. Firstly, that additional M4 (and, indeed, M0) monitoring ranges to buttress inflation targeting were tried in the United Kingdom during the early part of 1992-2007 but failed. M0 and M4 were widely disregarded and were soon dropped as a result. This objection ignores the fact that a mandate for ‘sustainable growth in the banking sector’ should acquire much more attention from politicians, the media and the public in light of the fact that the recent crisis was caused by too rapid an expansion of the banking sector prior to 2007 (Greenwood, 2009: 41) followed by an uncalled-for contraction during 2007-09. These sorts of violent swings can be avoided in future if the proposed framework is adopted.
A second objection is based on the idea that dual mandates are likely to lead to a conflict between the two aims, with one inevitably falling by the wayside. The author does not disagree that a mandate that involved numerical targets for both growth and inflation would lead to one being privileged over the other. But in the case of a dual mandate consisting of a quantitative inflation target and a mandated demand for sustainable growth in the banking system such a conflict does not exist, as the two aims are mutually reinforcing. Sustainable growth in the banking system is a precondition of, not an impediment to, low and stable inflation in the medium to long term.
A Brief Detour: Overfunding and the Constitutional Implications
Although the section above rejects two objections to the proposed framework, two more pressing ones are acknowledged briefly here. Central banks around the world have become more constitutionally privileged in recent times with the Bank of England, among others, only becoming independent in the last fifteen years (Bernanke et al., 1999: 169-171). Part of the reason for this shift has been the increasing acceptance that there is no long term trade-off between inflation and unemployment, and hence the best that can be done through macroeconomic policy is to secure a low and stable rate of inflation (Goodhart, 1995: 60-63). In light of this, monetary policy should therefore be handed to technocrats (Goodhart, 1995: 60-63).
This paper does not challenge that thinking, but it should be noted that independence in setting interest rates has recently had QE added to it so as to counteract the recent recession. QE has so far been pursued only with the express permission of the government. But in an era of very high deficits and macroeconomic instability, funding policy is likely to prove more important than in recent times. This could also include overfunding (issuing more gilt-edged securities than required to fund the deficit) should high inflation re-emerge, as was practiced in the United Kingdom during the first half of the 1980s. If funding policy is to play a more active role than hitherto, then the present ad hoc arrangements between Parliament and the Bank could prove awkward.
A transfer of responsibility for funding policy to the Bank in addition to its existing roles in financial supervision and the setting of interest rates under the mandate proposed in this paper could, in the opinion of the author, prove the most appropriate framework for future macroeconomic policy. But he does not deny the constitutional implications of such a move, as the Bank would become remarkably powerful by the standards of unelected institutions. Whether this is good for democracy is an open question that the author will not go into now, as it would risk going wildly off topic. It is nevertheless an issue that should be openly acknowledged.
Alongside this is a potential for a conflict of interest between the mandate to ensure stable growth in the banking sector and the role of the Bank of England in prudential supervision. Provided the Bank is able to demonstrate to Parliament that its performance in neither area is being undermined by its responsibility for the other then this could prove to be an advantage, as the Bank will have first-hand knowledge of conditions in the banking sector and can consider its interest rate responses in light of these to ensure sustainable growth in the banking sector. Nevertheless this is a significant issue that merits further study.
For those who reject the assumptions laid out in the introduction in this paper the proposals that follow are clearly of no value. For those who do accept them the author hopes that the proposals that follow do at least provide an alternative to the present and, in the author’s view, inadequate framework for macroeconomic policy. Inflation targeting was and remains a system of ‘constrained discretion’ (Bernanke et al., 1999: 22), as is the macroeconomic framework proposed in this paper. In the uncertain world of economic management in which a ‘magic’ policy rule remains chimerical, a consistently successful system of constrained discretion remains the best we are likely to do and evolution remains our best hope for finding it.
‘Ouch!’, The Economist (2010a), 23 rd October 2010, 13.
‘The growth surprise: So far, so good’, The Economist (2010b), 30 th October 2010, 34.
Bernanke, Ben S., Laubach, Thomas., Mishkin, Frederic S. and Posen, Adam S. (1999) Inflation Targeting: Lessons from the International Experience . Princeton: Princeton University Press.
Budd, Alan. (2005) Black Wednesday: A re-examination of Britain’s experience in the Exchange Rate Mechanism . London: Institute of Economic Affairs.
Congdon, Tim. (1989) Monetarism Lost, and why it must be regained . London: Centre for Policy Studies.
Congdon, Tim. (1992) Reflections on Monetarism: Britain’s Vain Search for a Successful Economic Strategy . Cheltenham: Edward Elgar.
Congdon, Tim (2005a) ‘Commentary’, in Alan Budd, Black Wednesday: A re-examination of Britain’s experience in the Exchange Rate Mechanism . London: Institute of Economic Affairs, 43-55.
Congdon, Tim (2005b) Money and Asset Prices in Boom and Bust . London: Institute of Economic Affairs.
Congdon, Tim. (2006) ‘The blunders of boom and bust’, Sunday Telegraph , 3 rd September 2006.
Congdon, Tim. (2007) Keynes, the Keynesians and Monetarism . Cheltenham: Edward Elgar.
Fisher, Irving. (1912) Elementary Principles of Economics (New York: Macmillan).
Friedman, Milton. (1968) ‘The Role of Monetary Policy’, American Economic Review , 58 , 1-17.
Goodhart, C.A.E. (1989) ‘The Conduct of Monetary Policy’, Economic Journal , 99 , 293-346.
Goodhart, C.A.E. (1995) The Central Bank and the Financial System . Basingstoke: Macmillan.
Greenwood, John. (2009) ‘The Successes and Failures of UK Monetary Policy, 2000-08’, in Philip Booth (Eds.), Verdict on the Crash: Causes and Policy Implications . London: Institute of Economic Affairs, 37-43.
Lamont, Norman. (2006) ‘Out of the Ashes’, in Howard Davies (Eds.), The Chancellors’ Tales: Managing the British Economy . Cambridge: Polity Press, 147-178.
Phelps, Edmund. (1968) ‘Money-Wage Dynamics and Labour-Market Equilibrium’, Journal of Political Economy , 76 , 678-711.
©Richard Wade. This article is licensed under a Creative Commons Attribution 4.0 International Licence (CC BY).
M4 is technically defined as: Cash outside banks (i.e. in circulation with the public and non-bank firms) + private-sector retail bank and building society deposits + Private-sector wholesale bank and building society deposits and Certificates of Deposit. ↩