Mission Creep and Monetary Policy

The Bank of Canada and the Government of Canada will renew their five-year inflation targeting agreement before it expires later this year. As with the six previous renewals, the Bank will have undertaken extensive consultation and research in advance with a view to identifying possible improvements to the existing framework. Although we have seen only relatively minor changes to the basic framework since inflation targeting was first introduced in Canada in 1991, the Bank regards this process as an important part of its accountability and fiduciary duty to Canadians.¹The term “framework” is used here to denote the core elements of monetary policy. These include its mandate, numerical target and other basic features that are normally agreed with the government as part of the renewal. This is not to suggest that no significant improvements have been made over time to the processes and operational aspects surrounding monetary policy. The Bank’s introduction of regular Monetary Policy Reports, Fixed Action Dates and a number of other innovations related largely to communications have enhanced transparency and accountability, helped to shape expectations and no doubt made monetary policy more effective. That said, it is important to note at the outset that the inflation-targeting framework has performed much better over this period than even its most ardent proponents might have imagined, setting a high bar for any future enhancements.

However, one should not assume from this that the conduct of monetary policy is problem free and that further refinements would not be welcome. Indeed, a few promising modifications to the present framework are briefly touched on in this E-Brief. They are not the primary focus of what follows. Rather, the primary purpose is to push back against those who would expand the mandate of monetary policy in unhelpful ways, threatening its ability to deliver on its core responsibility – price stability.

Serious obstacles to the effective implementation of monetary policy have emerged over the past few years. They include, most importantly, a steady decline in the estimated neutral rate of interest (where the economy’s output is at potential and inflation is on target) and more frequent encounters with the effective lower bound on interest rates. This has made it more difficult for authorities to provide stimulus using traditional monetary policy tools, forcing them to rely increasingly on so-called unconventional tools – with mixed results.²While there is no question that timely injections of emergency liquidity immediately after the Great Financial Crisis and during the COVID-19 pandemic played a critical role in resurrecting market activity, efforts to stimulate aggregate demand through continued use of unconventional monetary policy tools seem to have had limited success. Admittedly, we cannot be sure about what would have happened in their absence; however, empirical support is at best modest. See, for example, Borio and Zabia (2016). The search for a new or modified framework that might prove more effective in overcoming these obstacles has, therefore, assumed greater urgency, with options such as price-level targeting, inflation averaging and nominal income targeting receiving more serious consideration.

Interestingly, and somewhat curiously in the midst of these challenges, efforts to shore up monetary policy’s effectiveness in achieving price stability (and, relatedly, output stabilization) have been accompanied by a push to expand its responsibilities. The proposals have included, among others, suggestions that monetary policy should promote a greener economy, give greater emphasis to full employment, tackle income inequality, reduce the price of housing and accept greater responsibility for safeguarding financial stability.

While these are all good and worthy goals, this E-Brief argues that this sort of mission creep risks undercutting monetary policy’s ability to fulfill its primary responsibility and is neither necessary nor helpful to the attainment of these additional challenges.

Some Guiding Principles: Tools, Targets and the Tragedy of the Commons

Before proceeding with the arguments against mission creep, it is helpful to lay out certain concepts or precepts that guide the analysis.

The first precept is taken from the work of Jan Tinbergen, a Nobel Prize winning economist, who noted that policymakers with multiple objectives were well advised to have at least one tool for each target they hoped to achieve (henceforth known as the Tinbergen Rule)(Tinbergen 1952). While it might be possible to hit two birds with one stone, as the saying goes, such happy coincidences are rare.

The second precept is drawn from Robert Mundell and his “assignment problem.”³The “assignment problem” is an application of Mundell’s “principle of effective market classification.” Mundell observed that it is not sufficient to merely count the number of tools and targets. To arrive at effective, efficient and stable outcomes, tools need to be assigned to policy targets in a way that reflects their comparative advantage in achieving them.

Finally, there is the “tragedy of the commons,” a 19th century insight originating with William Forster Lloyd and later popularized by Garrett Hardin and Anthony Scott (Lloyd 1833, Hardin 1968, and Scott 1995). Simply stated, they observed that an exhaustible resource that is freely accessible to everyone is seldom successfully managed. Applied in modified form to the problem at hand, it suggests that if responsibility for something is shared among different entities, it risks becoming the responsibility of none. More specifically, accountability is critical to successful management and it must be focused and easily monitored. Something that is too diffuse is often lost.

From Early Forms of Mission Creep and the Impossible Trinity to Today’s Pressures

Demanding too much of monetary policy is nothing new. One of the best-known examples concerns the “impossible trinity.”⁴The term is often credited to Mundell, though he denied it. However, he did identify the trilemma. Also known as the “trilemma,” the impossible trinity states that it is impossible to simultaneously satisfy the following three conditions in an international setting: monetary policy independence, a fixed exchange rate, and an open capital market. One of the three must be sacrificed to achieve the other two (Mundell 1963).⁵Hélène Rey has suggested that even this is too ambitious and has put forward the notion of a “dilemma” as opposed to a “trilemma” or “impossible trinity.” See Rey (2013).

The implications of this principle are now widely accepted. Monetary policy alone cannot be asked to fix or peg the exchange rate while at the same time successfully stabilizing domestic prices and output, unless the capital account is closed. While it is possible to achieve all three goals, it requires the application of additional tools. Though this may seem obvious now, it took some time for frustrated policymakers to realize it.

Today, central banks are being pushed to consider an expanded list of responsibilities, including financial-system stability, income inequality and climate issues. However, there are important advantages associated with focusing policy on what is feasible and can be accomplished most efficiently. In the case of monetary policy, achieving low, stable and predictable inflation is regarded as something that is both feasible, subject to certain caveats and the best contribution it can make to economic welfare.

If promoting financial-system stability, income equality and climate sustainability could be added to monetary policy’s “to do” list without prejudicing its primary function, there would not be any reason for concern – with one important additional proviso. Could one ensure that the central bank is held clearly accountable for the success or failure of these supplementary tasks, or would responsibility become increasingly diffuse and, therefore, subject to the tragedy of the commons? As the remainder of this E-Brief will explain, such initiatives are likely to fail on both counts – accountability and effectiveness. Moreover, I will argue that such mission creep is both unnecessary and unhelpful for achievement of these goals, setting aside any damage it might do to monetary policy’s primary mission.

Financial Stability

Preserving financial stability is critical to the functioning of every modern economy, and central banks have an important role to play. This might be limited to the monitoring, reporting and research activities that they undertake, together with the emergency liquidity assistance that they are occasionally asked to provide as a lender of last resort (as we have recently seen in the COVID-19 pandemic), or involve a much wider set of responsibilities (Longworth and Jenkins 2015). While the Bank of Canada has no official responsibility for financial supervision and regulation beyond oversight of systemically important payments processes and retail payments, many central banks do.⁶For example, the Bank of England and the US Federal Reserve System.

This is not the same as expecting monetary policy to bend in a way that guards against financial instability except in so far as greater price and output stability might help to promote it. Neither is it to suggest that financial stability is unimportant for the successful transmission of monetary policy, simply that it has traditionally been regarded as the responsibility of another entity or a separate arm of the central bank.⁷Indeed, those who are concerned about this in a Canadian context might want to give serious consideration to the establishment of a separate entity charged with responsibility for ensuring macro-prudential stability. More recently, however, the idea of “leaning” has received increased support.⁸Often this takes the form of asking central banks to lean against suspected housing bubbles. See, for example, the recent government directive issued to the Reserve Bank of New Zealand (Government of New Zealand 2021).

“Leaning” involves asking monetary authorities to delay achieving full employment and hitting inflation targets in order to reduce the chances of serious financial instability at some, possibly distant, future date (White 2009). It would arise in situations where the real economy is weak, inflation is undershooting its target and monetary easing would normally be required. In this scenario, monetary policy would lean by adopting an interest-rate setting that was less accommodative than otherwise needed to restore inflation to a target within a typical horizon. The resulting higher interest rates and tighter credit conditions should in theory discourage excessive borrowing and asset-price inflation.

The first issue to address is whether such a strategy would work to reduce financial risk, and if it did, what might be sacrificed in the process.

There are a number of reasons to question the efficacy of leaning. To begin, timing is everything. In order to be effective, the higher interest rates must slow the growth of credit before debt reaches a critical level or tipping point. Otherwise, it risks triggering the very event that it hopes to avoid. Turning the proposal on its head, if debt goes beyond the point of no return, it might be better to soften the inevitable collapse with easier, rather than tighter, monetary conditions.

Leaning’s potential damage to employment and income also needs to be considered. Debt/income ratios, a measure of financial vulnerability, decline very slowly, and evidence indicates leaning might require very high interest rates to have much effect. In the interim, debt-servicing costs would rise, perhaps substantially, and employment would suffer, causing output to weaken. The end result could easily be a debt/income ratio that was higher, rather than lower, making the financial system less stable. Research by Lars Svensson, based on the leaning experience of Sweden, suggests the chances of this are reasonably high (Svensson 2016).⁹See also Schularik, ter Steege and Ward (2021). Leaning in the case of Canada is critically examined by Gorea, Kryvtsov and Takamura (2016) and Bank of Canada (2016).

Putting the question of efficacy aside for the moment, and assuming that leaning moves things in the right direction, are there better tools available to handle the situation? The answer is yes. Careful supervision coupled with effective micro- and macro-prudential regulation should be the first and only necessary lines of defence. As well, the danger with spreading responsibility for financial stability too widely is that it will fall between stools, owing to a lack of clear accountability – the tragedy of the commons once again.

Monetary policy is not only of doubtful effectiveness in this regard, it is unhelpful to the extent that people with primary responsibility for ensuring financial stability believe someone else is minding the store, posing serious moral hazard problems.

All of these arguments reinforce the case for a separate entity in Canada explicitly charged with responsibility for preserving financial stability.¹⁰This could take the form of a new department or wing within the Bank of Canada tasked with maintaining macro-financial stability but separate from its monetary policy responsibilities. However, establishing a completely separate agency or institution might provide a cleaner solution.

Full Employment When the “Divine Coincidence” Does Not Hold

Full employment is another worthy goal – indeed, perhaps the most important goal for macroeconomic policy. Happily, achieving a sustainable level of full employment through time (i.e., in the long run) is usually consistent with achieving and maintaining price stability. And to the extent this is true, it is an example of hitting two birds with one stone, or as Olivier Blanchard and Jordi Gali once described it – a “divine coincidence” (Blanchard and Gali 2007).

As is often the case, however, things are a little more complicated than the “divine coincidence” might suggest.

To begin, employment and inflation do not move in a synchronous fashion. Movements in employment and output – or more correctly, movements in the employment gap and the output gap – typically precede movements in inflation. Therefore, the economy is likely to return to full employment well before inflation returns to its target. Moreover, employment and inflation can move in opposite directions for a period of time, depending on the nature of the shocks hitting the economy. Supply shocks and markup shocks are notable examples. Situations such as these force policymakers to decide which variable should receive greater emphasis at a point in time and make it difficult to determine when employment and inflation will start to move in sympathy with one another.

Flexible inflation targeting, as practised by the Bank of Canada, allows considerable scope to adjust the relative weight given to employment versus inflation over the short run. But it does not solve the problem of how exactly the weights should vary or when the Bank should begin withdrawing (or adding stimulus) to bring the economy back to equilibrium. Giving the Bank a dual mandate for price stability and employment, similar to the one under which the US Federal Reserve operates, might have a certain appeal but does nothing to ease the dilemma.¹¹Some observers have suggested that the Bank of Canada should adopt a dual mandate similar to that of the Federal Reserve. They point to the typically lower US unemployment rates as evidence that a mandate requiring the Bank of Canada to focus explicitly on price stability and full employment would yield better macroeconomic results. It is important to note, however, that there are significant structural differences between their economy and ours, as well as significant differences in the way unemployment is measured in the two countries. The introduction of a dual mandate, in other words, is unlikely to deliver any material benefits and risks creating an unhelpful distraction.

Full employment is not a condition that can actually be observed in real time. Neither can you assign it an exact numerical target with any confidence. Rough estimates of full employment can be obtained by tracking a number of proxies, but you can only be sure you have achieved it once you have gone beyond it. Evidence that you have overshot the economy’s capacity limits appears in the form of rising inflation. In addition, because the limits of potential output and employment are reached well before persistent price pressures become evident, policymakers cannot wait until they have seen the whites of inflation’s eyes before they begin to withdraw monetary stimulus. Otherwise, they are likely to overshoot their inflation target by a wide margin. More drastic remedial tightening would then be required with unfortunate consequences for the real economy. Monetary policy, therefore, must be forward looking. (The same would be true, but work in opposite directions, if it were a matter of disinflating and removing monetary tightening.)

To complicate matters further, the short-run trade-off between employment and inflation (i.e., the Phillips curve) is believed to be much flatter now than earlier estimates indicated. The unobservable nature of full employment, together with the flatter inflation-employment trade-off, increases the odds that any strong signal regarding the possibility of going too far with an overly accommodative policy will only be received when it is too late, making a sizable overshoot more likely.

However, the reverse is also true. Though it is generally acknowledged that monetary policy has no effect on the natural or long-run unemployment rate, periods of prolonged unemployment (or under-employment) can have a persistent scarring effect on the economy as displaced workers lose their attachment to the labour force and their skills deteriorate. Therefore, if monetary policy underestimates the limits of full employment and tightens policy too soon, it will impose serious economic costs in the form of unnecessary unemployment and lost output. Inflation would also remain below target.

Yet, if monetary policy remains accommodative for too long, inflation easily could get out of hand and well-anchored inflation expectations would only make the situation worse. In this sense, the existence of a flat Phillips curve is a double-edged sword. On the one hand, it is more difficult to know if you are already in a state of excess demand; on the other hand, a great deal of unemployment might be required to bring inflation under control once you realize you are there.

Getting the balance of probabilities right is clearly difficult. The Bank of Canada has typically taken a cautious approach, trying to avoid any overshooting by relying heavily on estimates of full employment, applying stimulus judiciously and starting to taper it before capacity limits are reached. The Federal Reserve, in contrast, believing that it has historically underestimated how low the unemployment rate can be pushed, has recently adopted a new default strategy (Board of Governors of the Federal Reserve System 2020). Under its new flexible average-inflation-targeting framework, inflation will be allowed to modestly overshoot its 2 percent target in the expectation that prompt action can bring it back, thereby minimizing the fall out. Ideally, overshoots and undershoots would offset one another over time, leaving average inflation much closer to target and behaving more like a price-level target with a 2 percent slope – reducing price uncertainty.

Such a strategy could yield significant benefits, provided inflation can be returned to target in a reasonable time and at reasonable cost. It would speed the return to full employment, reduce the cumulative time inflation spends under or over its target and dampen the variability of the price-level trajectory over time. There are clear risks to such an inflation-averaging strategy, however. Inflation overshooting (undershooting) could dislodge inflation expectations, additional communication challenges could arise and the unemployment costs of bringing inflation back to target could be much higher than anticipated.¹²The new framework announced by the Federal Reserve might encounter additional problems owing to the asymmetric nature of the flexible inflation-averaging strategy that it intends to pursue, the lack of an explicit averaging period for its target and the large amount of discretion that it expects to exercise.

Fighting Income Inequality

An additional challenge that has been put to monetary authorities is to conduct policy in a way that does not aggravate income and wealth inequality.¹³See, for example, the “horse race” that the Bank of Canada conducted comparing the attractiveness of different monetary policy frameworks on the basis of several performance indicators. Ability to deal with income inequality was one of them as noted in Mendes (2020). In this regard, it is worth remembering that monetary policy affects the real economy and aggregate prices through three main channels: changes in interest rates, changes in asset prices and changes in the exchange rate. Conventional policy in an economy with excess capacity and below-target inflation involves lowering the target overnight interest rate with a view to stimulating spending by lowering the cost of borrowing and discouraging saving. Lower interest rates also boost wealth by pushing asset prices higher, as expected future interest income and earnings are discounted at a lower rate. In addition, lower interest rates put downward pressure (i.e., depreciation pressure) on the home country’s currency.

The side effects of this process that most concern some observers are the implications for those people whose consumption basket is more sensitive to the price effects of exchange-rate changes, whose disposable incomes depend importantly on the interest that they receive on bank deposits and bonds, and whose wealth is less likely to benefit from asset-price appreciation.

However, opponents will note that it is impossible to run monetary policy without these transmission channels operating. This is how monetary policy works. The problems just described arise only when monetary policy is responding to a downturn; the effects would presumably be offset by forces working in the opposite direction when monetary policy tightens to dampen overheating. As a result, things would be expected to roughly average out over the business cycle.¹⁴The emphasis here should perhaps be on the word “roughly.” Things might not balance out completely over the business cycle. Recent empirical work by Kronick and Villarreal (2020) highlights a possible asymmetry, suggesting that what is lost in the downswing may not be fully offset on the upswing. People taking this view might also counter that the best contribution monetary policy can make toward reducing poverty is to promote full employment. Therefore, anything that might interfere with timely monetary policy action subverts the alleviation of poverty in a more damaging way.

Whether or not one agrees with these counter arguments, there is a more telling reason to avoid constraining monetary policy in this way. As with financial stability, there are more direct and transparent means with which to tackle inequality issues. Progressive tax reforms, targeted transfers and other fiscal remedies are likely to be far more effective and involve fewer negative externalities than further constraining monetary policy’s room for manoeuvre. Forcing this responsibility on the central bank could once again lead to the tragedy of the commons problem.

Greening Monetary Policy

Suggestions to promote environmental issues with monetary policy are more peripheral to the issue of mission creep but nevertheless deserve mention as they gain momentum.

Although some proponents favour extreme measures such as denying so-called dirty industries access to any credit and worry that today’s low interest rates make it too easy for them to sustain their operations, others would limit lender-of-last-resort operations and liquidity injections more generally to those who can provide green bonds as collateral. They would also limit any quantitative easing involving private-sector securities to purchases of green bonds. A less extreme option would entail having central banks discount any non-green bonds that were offered at a higher rate and granting credit for green bonds at more favourable terms.¹⁵This is the lighter touch currently favoured by the European Central Bank. See Arnold (2021).

But are any of these steps necessary or helpful? Is there any evidence that green and non-green bonds are mispriced?¹⁶One might argue that this represents another opportunity to kill two birds with one stone – correct a market imperfection while meeting the primary monetary policy objective. But can one reliably identify the mispricing and quantify it; if so why stop there? Why not go beyond dirty bonds? A first-best solution would involve remedying the mispricing by introducing whatever fiscal measures, such as carbon pricing, are needed to achieve emissions objectives. Do green bonds need to be subsidized? Even if they do, are there other more effective ways of supporting green solutions?

The answer to the last question is, once again, yes. As with financial stability and income inequality, it is not obvious that there is any shortage of other, more effective, instruments with which to pursue green solutions. These should include, first and foremost, meaningful carbon pricing or a cap-and-trade scheme, supplemented if necessary by direct subsidies, tax penalties and selective regulation. The reader will notice that most of these fall under the umbrella of fiscal policy and none of them fits comfortably with monetary policy.¹⁷A stronger case can be made for attention to climate-change issues by central banks provided they have an explicit mandate for financial stability, a separate division for this purpose, and the micro- and macro-prudential tools needed to address issues such as “stranded assets” and the turmoil that unchecked global warming would produce. Adding it to the role of monetary policy is a different matter.

The list of possible extensions to monetary policy’s mandate does not end here of course. This E-Brief has merely tried to focus on the principal ones. Once you have opened the door, a number of other worthy causes might fight for attention. The problem with directing every tool at every target is that the outcomes are likely to be less than ideal and risk getting in one another’s way. Different horses for different courses is an old and useful adage.

Conclusion – Keep It Simple and Focused

The mission creep currently confronting monetary policymakers is born of good intentions. All of the goals described above are important and deserve support. The issue is one of assigning the right policy to the right goal. To do otherwise risks undercutting the effectiveness of something that works well for its intended purpose and redirecting it toward something for which it is not well suited, necessary or effective.

We should instead focus on changes that might enhance monetary policy’s ability to achieve its primary objective. Broadening its mandate is not the answer. Improvements are more likely to be found by concentrating more narrowly on the challenges that monetary policy currently faces.

Alternatives to the current inflation-targeting framework that are currently under active consideration include: price-level targeting, average-inflation targeting and (perhaps) nominal-GDP targeting. Interested readers who are not familiar with these frameworks can learn more through any number of sources, including the Bank of Canada’s website “Towards 2021: Renewing the Monetary Policy Framework.”¹⁸See, also, an excellent paper by Stephen Williamson (2021) that compares the features of alternative frameworks. They will not be explored here.

Suffice it to say, some of these alternatives, such as price-level targeting and average-inflation targeting, introduce a greater degree of what economists call “history dependence” into the reaction function of central banks. Under certain conditions this may: (1) change agents’ expectations and behaviour in ways that help to stabilize movements in prices and output; (2) lead to more predictable price-level paths through time; (3) limit the chances that the effective-lower bound will constrain the central bank’s room for manoeuvre; and (4) minimize the time spent at the effective lower bound when it is reached.

Other options, such as targeting the level of nominal GDP or adopting a dual employment-inflation mandate for monetary policy, would give greater weight to employment (subject to the reservations that were outlined above), while one of them – targeting nominal GDP – might also help monetary policy deal with negative supply shocks more effectively.

All of these alternatives come with certain potential drawbacks and caveats, however, and require careful examination before they are adopted.

Readers may have noticed that one of the arguments commonly raised against broadening the central bank’s monetary-policy mandate has not directly been featured in this E-Brief – the issue of central bank independence. This is not because independence is unimportant. Whenever central banks venture into areas normally seen as the purview of fiscal authorities or governments more generally, they risk losing their independence and could be criticized for operating outside the bounds appropriate for unelected agents. Attempting to address inequality, climate and financial-instability issues with monetary policy are all cases in point. The reason the risks to independence were not highlighted earlier is that central bank involvement in these areas was already rejected on the basis of ineffectiveness, inefficiency, lack of clear accountability, overly diffuse ownership, possible inattention and lack of focus. Introducing the loss of independence as an additional argument might suggest these other reasons were not sufficient to preclude something that central banks should not be doing in the first place.

To conclude, inflation targeting, as currently practised, has already exceeded expectations and has enough flexibility to incorporate some of the advantages offered by other frameworks without the awkwardness of introducing something less familiar and having to re-educate the public. The bar for any change, in other words, is exceptionally high even if one is talking about something closer to the mission at hand, let alone engaging in a noticeably more ambitious form of mission creep.

The author thanks Jeremy Kronick, Miles Wu, William Robson, Steve Ambler, John Crean, John Crow,
Phil Howell, Paul Jenkins, David Johnson, David Laidler, David Longworth, Angelo Melino, and Avery Shenfeld, as well as anonymous reviewers for helpful comments on an earlier draft. The author retains responsibility for any errors and the views expressed.

References

Arnold, Martin. 2021. “Financial Times. CB Set to Disappoint Campaigners on Climate Change.” Financial Times. 21 February. 

Bank of Canada. 2016. “Renewal of the Inflation-Control Target: Background Document – October 2016.” Ottawa. 

_____________. 2021. “Towards 2021: Renewing the Monetary Policy Framework.” Bank of Canada website.

Blanchard, Olivier, and Jordi Gali. 2007. “Real Wage Rigidity and the New Keynesian Model.” Journal of Money, Credit and Banking. 39 (1): 35-65.

Board of Governors of the Federal Reserve System. 2020. “Federal Open Market Committee announces approval of updates to its Statement of Longer Run Goals and Monetary Policy Strategy.” Press release. Washington. 27 August.

Borio, Claudio, and Anna Zabia. 2016. “Unconventional Monetary Policies: A Re-appraisal.” Bank for International Settlements Working Paper 570.

Gorea, Denis, Oleksiy Kryvtsov, and Tamon Takamura. 2016. “Leaning within a Flexible Inflation-Targeting Framework: Review of the Costs and Benefits.” Staff Discussion Paper 2016-17.

Government of New Zealand. 2021. Direction on Government Policy to the Reserve Bank of New Zealand. Section 68B. 26 February.

Hardin, Garrett. 1968. “The Tragedy of the Commons.” Science. 280: 1243-1248.

Kronick, Jeremy, and Francisco Villarreal. 2020. “Monetary Policy, Income Inequality, and Inflation - What’s the Link.” E-Brief 300. Toronto: C.D. Howe Institute. January. 

Laidler, David E. W. 1991. “How Shall We Govern the Governor? A Critique of the Governance of the Bank of Canada.” Toronto. C.D. Howe Institute.

Lloyd, William Forster. 1833. Two Lectures on the Checks to Population. England. Oxford University.

Longworth, David, and Paul Jenkins. 2015. Securing Monetary and Financial Stability: Why Canada Needs a Macroprudential Policy Framework. Commentary 529. Toronto: C.D. Howe Institute. June. 

Mendes, Rhys. 2020. “2021 Renewal of the Monetary Policy framework: An Update on the Horse Race of Alternative Frameworks.” Workshop Presentation. Bank of Canada. August. 

Mundell, Robert. 1960. “The Monetary Dynamics of International Adjustment under Fixed and Flexible Exchange Rates.” Quarterly Journal of Economics. 74: 227-57. May.

_____________. 1963. “Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange Rates.” Canadian Journal of Economics and Political Science. 29 (4): 475-485.

Rey, Hélène. 2013. “Dilemma not Trilemma: The Global Financial Cycle and Monetary Policy Independence.” Proceedings – Global Dimensions of Unconventional Monetary Policy – Jackson Hole. Federal Reserve Bank of Kansas City. Economic Symposium. pp. 285-333.

Robson, William B.P. 2009. To the Next Level: From the Gold Standard to Inflation Targeting – to Price Stability? Commentary 285. Toronto: C.D. Howe Institute. March.

Schularik, Moritz, Lucas ter Steege, and Felix Ward. 2021. “Leaning Against the Wind and Crisis Risk.” American Economic Insights. Forthcoming.

Scott, Anthony. 1995. Natural Resources: The Economics of Conservation. University of Toronto Press.

Svensson, Lars. 2016. “Cost-Benefit Analysis of Leaning Against the Wind.” Journal of Monetary Economics. 90:193-213.

Tinbergen, Jan. 1952. On the Theory of Economic Policy, New York: North-Holland.

White, William. 2009. “Should Monetary Policy Lean or Clean?” Federal Reserve Bank of Dallas. Working Paper 34.

Williamson, Stephen. 2021. Weighing the Options: Why the Bank of Canada Should Renew Inflation Targeting. Commentary 599. Toronto: C.D. Howe Institute. April.

John Murray is a Senior Fellow at C. D. Howe Institute and a former Deputy Governor of the Bank of Canada.