Monday, January 16, 2017 11.58PM / Mark Carney, Governor of the Bank of England
Good evening. It is a pleasure to be at the LSE and an honour to share the podium with Amartya Sen.
Professor Sen is rightly recognised for his many contributions, not least to welfare economics and social choice theory. He has posed, and in many cases answered, some of the most fundamental questions facing economists. For example, given the diversity of people’s preferences, is it possible to arrive at coherent aggregative judgements about how society is arranged? If there are as many preferences as there are people, is reasonable social choice possible at all? And how do we value public goods?
His work underscores the value of empirics – informational broadening – to help make the interpersonal comparisons necessary to understand, and act upon, the force of public concerns about poverty, inequality, even tyranny. And his insights have been applied to the most pressing economic and ethical questions, such as the prevention of famines.
His insights are also relevant to social choices about macroeconomic stabilisation, including inflation control, and what society is prepared to do to achieve it.
What level of inflation does society wish to achieve? How aggressively should inflation stabilisation be pursued when doing so imposes costs in terms of lost output and higher unemployment?
Low, stable and predictable inflation is a public good. It is not merely that rising prices mean households have to shop around or businesses have to update their prices periodically. High inflation hurts those, particularly the worst off in society, who don’t hold equities or property as well as those whose incomes are fixed in nominal terms. It distorts price signals, inhibits investment, and can ultimately damage the productive potential of the economy.
Equally, deflation can imperil growth and employment. In a highly indebted economy, deflation raises real interest rates, increases debt burdens, lowers wages, and reduces growth. In the extreme, these can morph into debt deflation, causing very high and persistent unemployment and financial collapse.
The happy medium is low, stable, predictable inflation over the medium term. A little inflation ‘greases the wheels’ of the economy, for example by helping real wages adjust more smoothly.1 Moreover, a positive inflation rate gives monetary policy space to deliver better outcomes for jobs and growth when shocks hit, without the distortionary costs of high and volatile inflation.2
Recognising the social value of inflation control is one thing, delivering it is quite another.
In the past, many societies could not. This is because the instrument that affects inflation most powerfully – monetary policy – also affects output and employment, at least in the short run. That influence tempted authorities to promise low inflation in the future, but then to renege in order to boost activity. Electoral cycles reinforced this predisposition.3 Firms and households began to anticipate these incentives, however, and eventually pre-empt them. The economy ended up in a worse equilibrium with higher inflation and unemployment.4 Such time-inconsistent policies contributed to excessive inflation and higher structural unemployment in the UK during the 1970s and 1980s (when inflation averaged 9.5% and unemployment over 7.5%).
In light of similar experiences, many societies came to recognise that macro outcomes can be improved by having society first choose the preferred rate of inflation and then delegate operational responsibility to the monetary authority to take the necessary monetary actions to achieve that objective.5 By “tying the hands” of authorities, time inconsistency is resolved and better outcomes for both inflation and unemployment become possible.
Inflation targeting, as practised in the UK, represents the most comprehensive adoption of these insights. By delegating monetary policy to an operationally independent central bank, inflation control becomes a more technical, ‘engineering’ problem, as described in Professor Sen’s work on ethics and economics.6 The committee responsible is given a clear remit – with a lexicographic preference for inflation control – and is charged to do what is necessary to achieve the inflation target over the policy horizon. For example, the Monetary Policy Committee (MPC) of the Bank of England must deliver price stability over the medium term, as defined by 2% CPI inflation. The inflation target is symmetric (meaning we care as much about returning inflation to target from below as from above), and it applies at all times.
To be clear, although the method by which the Bank of England achieves the inflation target may be a technical exercise, the UK’s monetary policy framework is grounded in society’s choice of the desired end. These ethical determinations are encoded in the monetary policy remit, from which the MPC takes its orders and against which it is accountable. Society chooses the ends, and, within pre-set boundaries, the MPC determines the means to achieve them.
Yet even in this framework monetary policymaking will at times involve striking short-term trade-offs between stabilising inflation and supporting growth and employment. In other words, the monetary policy problem cannot be fully contracted ex ante. But to re-emphasise, decisions about the precise trade-off that the MPC pursues are subject to the limits provided by its statutory objectives and the Chancellor’s remit. And they are disciplined by the transparency and accountability mechanisms including Inflation Reports, the minutes and transcripts of its meetings, the Parliamentary testimonies of its members, the open letter process triggered if the inflation target is missed by more than one percentage point, and discussions and debates in public fora such as this one.
The balance of my remarks this evening will concentrate on the experience with such trade-offs, which have become more common of late for two reasons. First, since the global financial crisis, the UK economy has been subject to a series of major supply shocks, which, for reasons that I will explain, create a tension between short-term inflation and output stabilisation. And second, there has been a renewed recognition that, in extreme circumstances, the monetary authority may have to take into account financial stability considerations in setting monetary policy with potential consequences for the short-term path of inflation control.
These are big decisions, decisions subject to considerable uncertainty, and decisions which entail potentially large welfare costs, arising from high unemployment and unevenly distributed costs of inflation. They deserve scrutiny to ensure that they are consistent with society’s preferences as expressed in statute and remit.
1. Monetary policy remit
For most of the Bank’s history, an assessment of whether it was acting in a way consistent with society’s view of the public good could not be made. As former Governor Eddie George remarked, during the half century that followed its nationalisation in 1946 “the Bank operated under legislation which, remarkably, did not attempt to define our objectives or functions.” They were, instead, “assumed to carry over from [the Bank’s] earlier long history.”7 In that regard, the Bank’s ‘constitution’ resembled that of the United Kingdom more broadly, comprising a rich history of law, principle and convention.
All changed with the passing of the Bank of England Act in 1998, which made specific “provision about the constitution, regulation, financial arrangements and functions of the Bank.” The Act clarified the Bank’s responsibilities and granted independence to the Bank for the operation of monetary policy. In delegating authority to an independent body, the MPC, in this way, the Act ensured the Bank would operate under ‘constrained’ rather than ‘unfettered’ discretion.8 It would be accountable to Parliament for operating the instruments of monetary policy to achieve the objectives of monetary policy, as determined by the Government.
The operational independence of the Bank of England is an example of power flowing from the people via Parliament within carefully circumscribed limits. Independence in turn demands accountability for the Bank to command the legitimacy necessary to fulfil its mission. By publishing its analysis, giving testimony, and delivering speeches, the Bank explains how it is exercising its powers to achieve its clearly defined policy remits.
Today, the MPC’s monetary policy remit requires it to achieve price stability, defined by the Government to mean 12-month CPI inflation of 2%. The remit recognises that inflation may deviate temporarily from the target on account of shocks, however. Since 2013, the remit has explicitly recognised that in these circumstances, bringing inflation back to target too rapidly could cause volatility in output and employment that is undesirable. The remit requires the MPC to consider, balance and explain such short-run monetary policy trade-offs.
Specifically, in exceptional circumstances, “shocks to the economy may be particularly large or the effects of shocks may persist over an extended period or both”. When this is the case, the challenge facing the MPC can be more significant, and the remits directs that “[i]n forming and communicating its judgements, the Committee should promote understanding of the trade-offs inherent in setting monetary policy”, including, importantly “the horizon over which the Committee judges it is appropriate to return inflation to the target”.
2. Achieving price stability
A simple framework for thinking about how to manage the monetary policy trade-off is shown in Figure 1. The red line in the figure illustrates a policymaker’s preferred trade-off: mapping the size of the shortfall of output below potential – the output gap – that the policymaker is prepared to tolerate for a given overshoot of inflation from target, and vice versa. The flatter the line the less the weight the policymaker places on output stabilisation and the more they are willing to tolerate large output gaps in order to eliminate small overshoots in inflation.
Figure 1: A stylised depiction of the monetary policy trade-off
The blue line in the figure is the Phillips curve, which summarises the structure of the economy and, in particular, how changes in demand, via the output gap, affect inflation.9 The economy’s equilibrium is at the intersection of these two lines. In the case of an inflationary shock that induces a trade-off, the Phillips curve shifts up, as shown in the figure. Monetary policy needs to tighten to lower pressure on resources and reduce inflationary pressures, such that the economy ends up on the red line, consistent with the policymaker’s preferences (as shown by the dotted arrows in the figure). The slope of the Phillips curve is crucial in determining in the size of the fall in output needed to reduce inflation to an acceptable level – the sacrifice ratio.
Figure 2: Stylised policy responses
Frequently, the economy experiences shocks that drive inflation and output in the same direction. These shocks to aggregate demand can include variations in government consumption, households’ desire to consume, or business’ desire to invest. Increases in demand put pressure on the use of resources, causing prices to rise. Because monetary policy can also influence demand it can lean against such shocks. If successful, it can stabilise inflation. In this case, no output-inflation trade-off arises. This is the so-called “divine coincidence”.10
In Figure 2, negative demand shocks would create a tendency for inflation and the output gap to move into the south-west quadrant. Similarly, for positive demand shocks, inflation and output gap would move to the north-east. Because in these cases, there is no trade-off between stabilising inflation and the output gap, it is relatively straightforward for monetary policy to keep inflation and the output gap close to the origin.
Chart 1: For much of the great moderation period, data were consistent with predominance of demand shocks…
In practice, data in these quadrants reflect lags in the transmission of monetary policy and other uncertainties: it takes time for monetary policy to affect output, and for output in turn to affect inflation. But if monetary policy has been effective, these deviations would not be expected to be either large or persistent. This state of affairs characterised much of the “Great Moderation” in the United Kingdom, from 1993 to 2007 (Chart 1), which was effectively a period of demand management.11
Things are different when shocks drive inflation up or down independently of demand. Exogenous changes in firms’ pricing power are one example – so-called cost-push shocks. Shocks to the exchange rate, the economy’s supply capacity, or commodity prices also have this flavour.12 Because monetary policy’s influence on inflation is predominantly an indirect one, via demand, in such circumstances inflation can only be controlled by delivering an opposing movement in aggregate spending. If something pushes up on inflation directly, monetary policy can only bring inflation back down by causing a reduction in spending via higher interest rates. The speed with which this adjustment is delivered is determined by the monetary policy maker guided by their remit.
Such circumstances have characterised the period in the UK since the global financial crisis (Chart 2), which entailed a large adjustment to the supply side of the economy, meaning a lower exchange rate, lower growth, and higher inflation.
Chart 2: … whereas after the financial crisis, the economy experienced a challenging monetary policy trade-off
In contrast, the US and the euro area have seldom faced a trade-off between output and inflation stabilisation, even since the global financial crisis (Charts 3 and 4).
A simple way to represent more formally how the policymaker optimises the trade-off is in “linear-quadratic” form – a set of linear constraints describing the behaviour of the economy, and quadratic preferences that penalise deviations of inflation from its target and output from its potential. The relative weight the policymaker places on output stabilisation, relative to inflation stabilisation, is often denoted 𝜆 – or ‘lambda’. The policymaker’s “loss function” in any particular time period is: 𝐿𝑜𝑠𝑠𝑡≡(𝜋𝑡−𝜋∗)2+𝜆(𝑦𝑡−𝑦𝑡∗)2
where 𝜋𝑡 is inflation, 𝜋∗ is the inflation target, 𝑦𝑡 is output and 𝑦𝑡∗ is ‘trend’ output, so that 𝑦𝑡−𝑦𝑡∗ is the output gap, and 𝑡 subscripts denote time periods.13 The policymaker’s objective is to minimise the discounted sum of these losses over time. In this formulation, a lambda of zero would imply no weight on the stabilisation of real activity – so-called “inflation nutter” preferences. When lambda is positive, in contrast, the policymaker is willing to strike at least some trade-off between output and inflation stabilisation, as directed by the MPC remit.
This objective function is optimised subject to the constraints implied by the aggregate behaviour of the economy, including the relationship between interest rates and activity; and the relationship between activity and inflation – the Phillips curve shown in Figure 1. In essence, these suggest lower interest rates raise activity;14 and higher activity generates higher inflation.
Under this simple framework, in certain circumstances, the optimal policy balances inflation overshoots with shortfalls of activity relative to potential, and vice versa for inflation undershoots. (A derivation appears in the Annex.) The relative size of these two deviations is governed by the strength with which higher output translates into higher inflation; and the preferences of the policymaker, or: 𝜋𝑡−𝜋∗=−𝜆𝜅(𝑦𝑡−𝑦𝑡∗) where 𝜅 is the effect of the output gap on inflation. From this, it is clear that the higher is lambda, the greater the weight placed on output stabilisation and the more a given shock is allowed to flow through to inflation. As lambda shrinks to zero, the policymaker becomes an inflation nutter, with all of the adjustment to a shock forced through the output gap in order to keep inflation very close to the target.
The slope of the trade-off is also affected by that of the Phillips curve. When the Phillips curve is flatter, output gaps of a given size result in larger inflation over- or under-shoots.15 Intuitively, a flat Phillips curve makes the output costs associated with moving inflation around higher. As a result, a flatter Phillips curve means the policymaker will optimally choose to allow inflation to be further away from the target than if the Phillips curve were steeper.
3. Lambda, lambda, lambda
What does past behaviour reveal about the preferences of policymakers – the value of lambda?
Kindly download the full speech here
1 See Tobin (1972), “Inflation and unemployment”, American Economic Review, 62.
2 Inflation can have large redistributive effects. Volatile inflation arbitrarily re-distributes wealth across borrowers and savers. And most recently, there has been a debate about how the stance of monetary policy has affected the distribution of wealth and income. In a series of recent analysis and speeches, members of the MPC have shown that, while distribution is not an objective of monetary policy, the evidence strongly suggests that policy since the crisis has been associated with better outcomes.
3 Nordhaus, W D (1975), “The political business cycle”, Review of Economic Studies.
4 See, inter alia, Kydland, F and Prescott, E, (1977), “Rules rather than discretion: the inconsistency of optimal plans”, Journal of Political Economy, 85(3), 473-492; Barro, R, and Gordon, D B, (1983). "A Positive Theory of Monetary Policy in a Natural-Rate Model". Journal of Political Economy. 91 (4): 589–610.
5 See Rogoff, K (1985), “The optimal degree of commitment to an intermediate monetary target”, Quarterly Journal of Economics, November, pages 1,169–90.
6 The ‘engineering’ view is primarily concerned “with logistic issues rather than with ultimate ends”, including their ethical characteristics. “The ends are taken … as given, and the
7 George, E. (2000), “Central bank independence”, speech at the SEANZA Governors’ Symposium, 26 August 2000.
8 See, for example, King, M. (2000), “Monetary policy: theory in practice”, 7th January 2000. King attributes the ‘constrained’ versus ‘unfettered’ distinction originally to Ben Bernanke and Frederick Mishkin.
9 Originally formulated by Phillips (1958), albeit as a relationship between wage growth and unemployment. See: Phillips, A. W. (1958). "The Relationship between Unemployment and the Rate of Change of Money Wages in the United Kingdom 1861-1957". Economica, 25 (100): 283–299.
10 See Blanchard, O., and J. Galí (2007). "Real Wage Rigidities and the New Keynesian Model". Journal of Money, Credit, and Banking, 39(1), pp. 35-65.
11 See Broadbent, B (2013), “Conditional guidance as a response to supply uncertainty”, speech at the London Business School.
12 Exchange rate shocks also have demand effects including via net trade. For example, a higher exchange rate lowers imported price pressures, reducing inflation directly. It also makes domestic goods less attractive on world markets, lowering the contribution of net trade to demand, which has second-round effects on inflation via the reduced demand of exporters for labour. At the same time, real household income rises with a higher exchange rate, providing some countervailing support to demand and domestically-generated inflation.
13 A “micro-founded” loss function resembling this can be derived in New Keynesian models, in which lambda is a function of the underlying deep parameters in the economy (such as those governing the degree of price stickiness). These are typically based on the representative agent assumption. A newer literature drops representative agent assumption. In this case, the “micro-founded loss function” contains additional terms capturing some aspects of heterogeneity between households. For example, in a model with borrowers and savers, a term in the loss function containing the “consumption gap” arises – measuring the gap between the consumptions of the two types of household. The reason terms like this appear is that dropping the representative agent assumption tends also to introduce additional frictions, like borrowing constraints. In the example given here, the “consumption gap” represents the welfare loss due to imperfect risk-sharing (borrowers can’t smooth consumption as easily). Strictly speaking, this is something the “social planner” should care about, even if the central bank, with a narrowly-defined inflation mandate, does not.
14 Lower interest rates, that is, relative to the equilibrium or “natural” interest rate.
15 In New Keynesian models, the slope of the Phillips Curve is related to the deep parameters governing the frequency with which firms are assumed to be able to re-optimise their prices, following the formalisation of Calvo (1983). See Calvo, G (1983), “Staggered prices in a utility-maximising framework”, Journal of Monetary Economics, 12.