Flexible Inflation Targeting Beats a Dual Mandate: Lessons for Canada’s 2026 Framework Renewal

Introduction

The Bank of Canada and the Government of Canada will announce the renewal of their agreement concerning the monetary control framework towards the end of 2026. In 2021, the last agreement explicitly mentioned the desirability of maintaining high employment for the first time. The Bank of Canada (2021) stated that:

The Government of Canada and the Bank of Canada believe that the best contribution of monetary policy to the well-being of Canadians is to continue to focus on price stability. The Government and the Bank also agree that monetary policy should continue to support maximum sustainable employment, recognizing that maximum sustainable employment is not directly measurable and is determined largely by non-monetary factors that can change through time. Further, the Government and the Bank agree that because well-anchored inflation expectations are critical to achieving both price stability and maximum sustainable employment, the primary objective of monetary policy is to maintain low, stable inflation over time.

The agreement was thus a departure from past frameworks in that it softened the focus of the Bank of Canada on inflation as its mandate and elevated the importance of employment. However, it did not explicitly commit the Bank of Canada to any goal other than the existing 2 percent inflation target. It therefore represented a continuation of flexible inflation targeting (FIT) – a framework first adopted in 1992.¹The text of the agreement also mentioned some other desirable goals of policy, such as financial stability, climate change, and income distribution, without making any of these a formal part of the Bank of Canada’s mandate. See Murray (2021) for a discussion on the worthiness of these goals, but the fact that this kind of mission creep will undercut the Bank’s ability to fulfill its primary responsibility of low and stable inflation.

In the run-up to the renewal in 2026, there has been a discussion about whether the Bank of Canada should take on the explicit dual mandate of both low and stable inflation and maximum output and employment.²See, for example: Senate of Canada (2024). The news release of this report can be found at https://sencanada.ca/en/ newsroom/banc-interim-canadas-monetary-policy/. According to its proponents, under this mandate, the central bank would better balance variables such as employment and output than under its existing mandate, which focuses more exclusively on price stability. This argument has gained additional traction in a world of more supply shocks – such as tariffs – which may force the central bank to more frequently choose between stabilizing inflation and supporting output and employment.

In this E-Brief, we demonstrate that both theoretical and practical considerations give a clear advantage to FIT over a dual mandate – even in the face of ongoing supply shocks. As such, we view the commitment to “low and stable inflation” within the FIT framework as a central pillar for the upcoming renewal of the Bank’s monetary policy framework.³We do not consider variations of a policy framework with a sole focus on price stability but with a different policy target, such as price level targeting (see Ambler [2009] for a detailed discussion) or nominal income level targeting (see Ambler [2021] for a review). Williamson (2021) evaluates those, among others, in comparison to an inflation target in the context of the 2021 renewal and finds that there are good reasons to question whether Canada’s economic performance would have been better under any of the proposed alternatives to FIT. We also take Canada’s flexible exchange regime as given. In future work, we will look at other aspects of the policy framework: possibly strengthening the Bank of Canada’s operational independence and accountability, the appropriate price index for the inflation target, and the role the housing market and housing prices should play in inflation targeting.

Renewal 2026: Key Questions Answered

Q: Is there a long-run trade-off for monetary policy between higher inflation and higher output?

A: No. Low and stable inflation leads to better economic performance.

There is broad consensus among economists that any attempt to systematically increase output and reduce unemployment through monetary policy will simply lead to more inflation, leaving the real economy potentially worse off.⁴This proposition was first asserted by Friedman (1968) and Phelps (1968) and is often referred to as the Phillips curve being vertical in the long run. The Phillips curve relates inflation in the short run to expected future inflation and to the output gap. What monetary policy can control in the long run, however, is inflation, mainly by anchoring expectations about future inflation. By doing so, the central bank allows households and businesses to make economic decisions with more certainty so that low inflation benefits economic activity.

There is, therefore, no long-run trade-off between higher inflation and higher output. In fact, research has shown that high inflation tends to lead to lower output growth, and when inflation is low, there is no clear relationship.⁵See, for example: Bruno and Easterly (1998) and Ascari, Bonomolo and Haque (2023). Moreover, permanent shocks to inflation do not lead to permanent changes in output.⁶See, for example: Bullard and Keating (1995). Finally, inflation volatility tends to affect growth negatively. Since inflation volatility tends to rise with higher inflation, inflation negatively affects both economic growth and the volatility of economic activity.⁷See, for example: Judson and Orphanides (1999). From a long-run perspective, this makes it clear that maintaining low and stable inflation delivers better economic outcomes than any other goal for monetary policy.⁸Ascari and Sbordone (2014) summarize and synthesize much of the empirical and theoretical literature on the relationship between trend inflation on the one hand and the level and rate of growth of output on the other. Ascari, Phaneuf and Sims (2018) find that even a modest increase in trend inflation, from 2 to 4 percent, could plausibly lead to a loss in economic welfare equivalent to a 4 percent reduction in consumption.

The record for Canada is also unambiguous in terms of inflation and the real economy. From January 1996, when the target officially became 2 percent, until March 2025, inflation averaged 2.1 percent. By comparison, in the same number of months from October 1961 to December 1990, it averaged 5.7 percent (see Figure 1).⁹The 1960-1990 period contains two major supply shocks (the two oil price shocks), while the 1996-2025 period also contains a major supply shock (the COVID-19 lockdowns). We thus view these as comparable periods. The standard deviation of inflation also decreased from 3.2 percent over the latter period to 1.3 percent over the former period. Overall, since the introduction of inflation targeting, consumption spending has been 50 percent less volatile, GDP growth 30 percent less volatile, and unemployment more than 40 percent less volatile.¹⁰Note for consumption and GDP growth we removed the 2020-2022 COVID period due to the wild swings in each because of the once-in-a-century shutdown and reopening of the economy. We did not need to remove this period for the unemployment calculation as the swings were not as severe. See an earlier version of these results in Ambler and Kronick (2024). Ravenna and Ingholt (2021) attribute these changes to the introduction of inflation targeting that successfully stabilized inflation expectations. These results are consistent with Parkin (2016), who compares economic variables for countries before and after becoming inflation targeters, using as controls countries that made no change to their central bank’s mandate during the period under investigation (1980-2013). He finds that implementing inflation targeting leads to lower and less variable inflation, higher GDP growth, and a lower variability of GDP and unemployment. Judging central bank success based on such volatility measures is critical since, as we discussed above, central banks have no control over long-run real variables.

This success of Canadian monetary policy over the last three decades seems to already imply a clear endorsement for a renewed and exclusive focus on “low and stable inflation” in the forthcoming agreement in 2026. In what follows, we further support the positive experience with such a framework by showing why FIT has a clear advantage over a dual mandate, where the central bank also has an explicit mandate in the form of maximum employment or output.

Q: Is there a short-run trade-off for monetary policy between inflation and employment?

A: Yes. The Bank of Canada should aim for a compromise when stabilizing inflation and output.

In the short run, the economy faces shocks that lead to fluctuations in inflation and real economic activity. Fluctuations in real economic activity give rise to what’s called the output gap: the difference between actual output and the output the economy could potentially achieve. When the economy is operating at its potential, it is operating with maximum employment, and the output gap is closed.

Models¹¹See, for example: Woodford (2003) and Galí (2015). used by central banks capture these fluctuations in the context of a so-called Phillips curve, where inflation in the short run is determined by expected future inflation and the output gap. When expected inflation is constant, there is an inverse relationship or trade-off between inflation and the output gap (more economic activity leads to more inflation). However, inflation expectations will not remain constant if central banks continue to allow excess activity to increase inflation, leading to an exacerbation of the effect on prices.¹²As noted in the previous section, there is no trade-off between inflation and the output gap in the long run. In addition (as noted in the next section), the output gap is a theoretical concept which depends on knowing or estimating the fullemployment level of output.

The importance of this trade-off depends on the precise nature of the shocks the economy is facing, which are difficult to identify in real-time. Shocks that primarily influence demand tend to move inflation and output in the same direction, leading to a phenomenon known as the “divine coincidence,” where stabilizing inflation and the output gap become complementary for the central bank. Think of a recession induced by falling aggregate demand. This produces a negative output gap and falling inflation. The central bank loosens monetary policy to stimulate consumption and investment. Demand re-ignites, pushing the output gap closed and inflation back up.

Temporary supply shocks tend to be different. A negative shock – such as a spike in oil prices or the imposition of tariffs – typically leads to higher prices and lower output. Central banks will look through these shocks if they believe they are one-off. The concern is that these shocks may not be one-off events, raising the risk that inflation expectations become de-anchored. For central banks, the challenge is that tightening monetary policy to bring down inflation can further reduce output in the face of supply shocks. Loose monetary policy will boost output but at the expense of a further increase in prices. The divine coincidence breaks down, and this is where the idea of a dual mandate has gained some traction.

The Bank of Canada has identified the “4Ds” – decarbonization, deglobalization, demographic trends, and digitalization – as medium-term challenges that could trigger supply shocks, potentially leading to both inflationary pressures and job losses in the Canadian economy. Take demographics, for example: as babyboomers retire, the labour force shrinks, and the economy’s potential output is reduced. This negative supply shock will be inflationary. A similar effect arises from deglobalization. As countries become more isolationist, as we see today with the United States, costs will increase for consumers who have only domestic producers to buy from, while slowing global demand will shrink the customer base producers can sell to.

Such changes may affect the short-term trade-offs between inflation and employment or output. We note, however, that these changes are structural, and there is little the Bank of Canada – and monetary policy more generally – can do about them. Regardless, this raises the question of whether this trade-off implies putting more weight on output and employment by switching to a dual mandate.

Q: Should there then be a dual mandate for monetary policy?

A: No. Such a framework relies too heavily on hypothetical concepts that cannot be directly measured and evolve over time.

A genuine short-run trade-off between stabilizing output and returning inflation to target could be an endorsement of a dual mandate. Indeed, the Federal Reserve Board in the United States has operated under an explicit dual mandate since 1946, when the Employment Act specified that the Fed should pursue both price stability and maximum employment. The Full Employment and Balanced Growth Act or Humphrey-Hawkins Act reiterated this dual mandate in 1978.

A dual mandate is usually interpreted as achieving maximum sustainable output or employment while stabilizing inflation. The employment or output component is often defined using concepts like the “non-accelerating inflation rate of unemployment” (NAIRU) or “potential output” – the level of output the economy could sustain in the absence of shocks. These are hypothetical concepts that cannot readily be observed and, worse, can change over time. Households can relate much more directly to the costs of inflation than to abstract concepts like the output gap and the NAIRU. Everyone experiences inflation first-hand, while output losses are often not directly felt by the general population.

These quantities must also be estimated,¹³See Gnocchi et al. (2024) for a discussion of this challenge. and the central bank will never know for certain that it has attained the maximum sustainable level of employment or output because shocks are constantly hitting the economy. As Svensson (2004) succinctly notes, “We can say that the output target is subject to estimation, but it is certainly not subject to choice.” Higher uncertainty concerning the value of the output gap or the NAIRU strengthens the case for placing less emphasis on these real-economy variables, and for discouraging central banks from relying on subjective judgments of them when setting policy.

The necessity of estimating potential output or the NAIRU is compounded by the fact that they are subject both to short-run fluctuations and to structural change. Estimating them amounts to estimating a moving target, meaning the degree of uncertainty surrounding estimates of potential output or the NAIRU will not diminish over time. This is a daunting task, especially in times when significant changes – such as those driven by the “4Ds” identified by the Bank of Canada – are likely to cause major disruptions to the Canadian economy.

The Federal Open Market Committee (2024) recognizes this challenge in a document stating its long-term goals and strategy:

[T]he maximum level of employment is a broad-based and inclusive goal that is not directly measurable and changes over time owing largely to nonmonetary factors that affect the structure and dynamics of the labor market. Consequently, it would not be appropriate to specify a fixed goal for employment; rather, the Committee’s policy decisions must be informed by assessments of the shortfalls of employment from its maximum level, recognizing that such assessments are necessarily uncertain and subject to revision.

Not surprisingly, then, the Federal Reserve, while keeping the dual mandate, eventually moved towards a single quantitative goal in 2012 in the form of 2 percent annual inflation. And, while its 2020 renewal put increased emphasis on maximum employment, empirical evidence suggests this move was responsible for the Fed falling behind the curve in the battle against the post-pandemic inflation surge. From Romer and Romer (2024):

Our key finding is that the elevation of the maximum employment side of the dual mandate played a crucial role in limiting the Fed’s response to inflation… By the fall of 2021, policymakers believed that inflation was above the objectives they had set out in their forward guidance and expected that the economy would reach maximum employment soon, but they refused to act until that actually occurred.

Pang and Shiamptanis (2024) show that the Bank of Canada’s increased emphasis on the state of the economy also caused it to fall behind the curve in fighting the inflation surge in Canada. Nevertheless, inflation decreased more rapidly in Canada than in most other advanced economies, including the United States. Recent work from the Bank of Canada (Feunou and Kumar 2025) shows that the inflation risk premium in the United States has increased, meaning investors are demanding more compensation for inflation, suggesting a greater risk of de-anchoring inflation expectations.¹⁴This is consistent with earlier work from Gürkaynak et al. (2007): “For the United States, we find that far-ahead forward inflation compensation has reacted significantly to macroeconomic data releases, suggesting that long-run inflation expectations have not been completely anchored. In contrast, the Canadian inflation compensation data have exhibited significantly less sensitivity to Canadian and U.S. macroeconomic news, suggesting that inflation targeting in Canada has helped to anchor long-run inflation expectations in that country.”

We see the drawbacks of the dual mandate elsewhere as well. In late 2023, the Reserve Bank of New Zealand backtracked from a dual mandate introduced in 2018 under the post-COVID inflation surge. The government passed legislation – consistent with the advice of the central bank – instructing the central bank to return to a sole inflation focus (Craymer 2023). Further, with respect to the recent COVID-19 inflation surge, the government argued that “with no hierarchy of objectives, the introduction of a dual mandate heightened the risk of a future policy error – with monetary policy led in multiple directions, even as inflation embedded itself in the economy.”¹⁵See: Reserve Bank of New Zealand (2023).

Beyond the issue of trying to target an unobservable, moving variable, there are further reasons not to rely on an explicit dual mandate. The Bank of Canada, like most central banks, uses a single instrument – the overnight rate – to implement its policy goals. Other tools like quantitative easing or forward guidance influence the entire term structure of interest rates and are useful if executed appropriately, but they are only a variation on the overnight rate in that they simply change expectations about its future level.¹⁶See: Koeppl et al. (2024). Therefore, the combination of conventional and unconventional monetary policy remains but one tool. And this matters, as Nobel Laureate Jan Tinbergen (1952) pointed out many years ago: successfully achieving multiple independent policy objectives requires at least an equal number of policy instruments. Otherwise, a trade-off – in this case, between employment or output and inflation – is inevitable.

Next, under an explicit dual mandate, the Bank of Canada would have to report on its degree of success in achieving maximum sustainable employment. It would have to explain its strategy for achieving the goal and explain to the public why it wasn’t successful in those instances. As mentioned, this is difficult to do with an unobservable, moving target. This could, then, undermine public confidence in the Bank.

There is one final concern. An explicit dual mandate also requires the framework to specify the relative importance of the two goals, especially if there is a trade-off between them. Because maximum employment (output) is hard to define, this would potentially leave a central bank open to pressure from the government to concentrate on boosting output at the cost of inflation. The Bank of Canada Act allows the government of the day – specifically the Minister of Finance – to give explicit instructions to the Bank of Canada.¹⁷It is generally understood that this would prompt the resignation of the Governor. See: Thiessen (2001, p. 32). Also, see Siklos (2008) for details on the events leading to the amendment of the Bank of Canada Act which allows for this possibility. An explicit dual mandate could give the government added justification to intervene in the conduct of monetary policy, threatening the operational independence of the Bank of Canada.

Q: Should there be, then, a narrow focus on inflation in the explicit mandate for monetary policy?

A: Yes. It facilitates communication, ensures accountability, and minimizes political interference, anchoring long-run inflation expectations.

An explicit target for inflation – such as 2 percent in Canada since 1996 – avoids a lot of the issues we have just discussed. The realized inflation rate is readily observable, with only a short lag of two to three weeks. It is a trivial matter to calculate the variance of the inflation rate around the target level to judge the performance of the central bank. This directly leads to better accountability for the central bank’s conduct of monetary policy.

Communication with the general public is also relatively straightforward for the central bank. The Bank of Canada can explain how shocks and structural changes have influenced the realized inflation rate and relate how the perceived output gap and employment are putting upward and downward pressure on inflation; this is precisely its current communication strategy. Better communication and accountability help build public trust in the central bank. A dual mandate, however, would constantly require the Bank to explain the trade-offs involved, adding pressure to its public messaging.

For example, after the renewal in 2020, the Federal Reserve felt the need to clarify that it is the “shortfalls of employment from its maximum level – not the deviations” that matter and that low unemployment itself in the absence of price or financial pressures will not trigger policy actions (see Clarida 2020). This could be too much for the public to try and parse.

Finally, with a clear framework focused on inflation, the central bank can resist political pressure. The economic literature and experiences from the 1970s and 1980s clearly established that short-run monetary stimulus to increase output will have to be paid in long-run inflation¹⁸See, for example: Goodfriend (2007). – a link we were all reminded of again post-COVID. A well-measured inflation target can act as a shield for the central bank from political interference.

Another difference today from the inflation experience in the 1970s and 1980s is the improved anchoring of long-run inflation expectations through inflation targeting. Anchored inflation expectations make the goal of stabilizing inflation much simpler. In the face of an unexpected shock, which leads inflation to increase, any increase in expected inflation (in the short, medium, and long run) would make it harder to bring down inflation because of the impact of higher expected inflation on price increases by firms and on wage settlements.

The recent experience of high inflation post-COVID reinforces this point and must be seen as a success for the Bank of Canada and its FIT framework. While the Bank was late raising interest rates, inflation fell relatively quickly from its peak of 8.1 percent back to the target of 2 percent, with remarkably little volatility in between, keeping real costs of this disinflationary period limited. Canada’s inflation fell back to target faster than in many other advanced countries – including the United States, which has not reached the 2 percent mark since March 2021. At the same time, inflation expectations remained fairly well-anchored, especially for the medium and longer term.¹⁹See, for example, the quarterly Business Outlook Survey of the Bank of Canada. Data can be found on the Bank of Canada’s site as well: https://www.bankofcanada.ca/rates/indicators/capacity-and-inflation-pressures/expectations/.

Q: What is, then, the best framework for monetary policy that balances theoretical and practical considerations?

A: FIT puts inflation at the centre, while allowing for flexibility due to other policy considerations when necessary.

The Bank of Canada’s current policy framework (FIT) affords the Bank flexibility in two distinct ways. First, it recognizes that the Bank is concerned with goals that are auxiliary to an inflation framework. For example, the Bank may engage in more accommodating policy in the aftermath of very large shocks. A financial crisis may require additional measures to re-establish financial stability even if there are upward inflationary pressures in the long run. Natural disasters – if large enough – may also require monetary stimulus for an extended time.

Second, the Bank of Canada cannot keep inflation exactly at the 2 percent target at all times because of unexpected shocks and the delayed effects of interest rate changes on the economy. When a shock causes inflation to deviate from the target, the Bank must always decide how quickly it aims to get inflation back.

This means, in practice, that when inflation is above target, the Bank may decide to take a little longer to get it back down to 2 percent so that its monetary policy is not so restrictive that it throws the economy into a recession or unnecessarily adds to the economic costs of large shocks. Conversely, when inflation is below target, the Bank may decide not to stimulate the economy so much in the short run that inflation risks overshooting the target in the medium run.

The key distinction from a dual mandate is that this flexibility does not introduce other goals that are considered equal to the primary goal of achieving the inflation target. Low and stable inflation remains the goal.

To look at the trade-offs and outcomes of the different possible frameworks, the Bank of Canada, in the context of the 2021 renewal, ran a simulated “horse race.” They found that the dual mandate generated modest improvements with respect to some measures and weaker performance in others. Specifically, inflation (albeit mildly), the output gap, and real GDP growth were more volatile under the dual mandate, while unemployment and interest rates were less volatile.²⁰This is the case where unconventional monetary policy at the effective lower bound is ineffective. For the polar opposite case, where unconventional monetary is a perfect substitute for conventional monetary policy, thus abstracting from the effective lower bound, the results are the same except for, interestingly, inflation, which is mildly less volatile under the dual mandate. For details, see Dorich et al. (2021). Consequently, there is not much there that suggests the need to change the Bank of Canada’s framework.

It is common to describe how a central bank should operate with the so-called “Taylor rule” (Taylor 1979). The rule stipulates that the central bank adjusts its policy interest rate in response to deviations of inflation from the target rate as well as deviations of output (real GDP) from its estimated full-capacity (“potential”) level. We can then think of an explicit dual mandate as caring equally about inflation deviations from target and actual output (or employment) deviations from potential output (or maximum employment), with the goal of minimizing the combined economic losses from both. On the contrary, under FIT, a central bank maintains a principal emphasis on inflation but still takes into account other considerations – often real output – to achieve low and stable inflation at a minimum cost in terms of excessive output fluctuations.

The economic literature has emphasized that FIT is a good approximation for the optimal conduct of monetary policy.²¹See Koeppl (2009) for a recommendation of the framework in the context of the renewal in 2011 and Williamson (2021) in the context of renewal in 2021. In addition to the Bank of Canada, other central banks (e.g., the European Central Bank, the Bank of England, and the Reserve Bank of Australia) similarly operationalize FIT under a hierarchy of goals, placing an inflation target first and above all other mandates, including maximum employment. The conclusion these monetary policy authorities have all reached is that FIT balances secondary policy considerations with the superiority of a singular focus on low and stable inflation as the optimal framework for a central bank.

Conclusion

“Low and stable inflation,” operationalized using FIT, was a success story for the last three decades. It was relatively easy for the Bank and the public to understand both. The Bank achieved its goal, and the public could readily verify its success and – short-lived – misses. These are clear advantages over a dual mandate. Indeed, if anything, the most recent period of high inflation only reinforces its value. It was the anchoring of inflation expectations at 2 percent that allowed us to bring inflation back to target as quickly as we did and without a recession. In the current climate of uncertainty surrounding trade policy, a monetary policy focused on price stability is the best contribution it can make to facilitate the structural adjustments this may bring.

Let’s not endanger a framework that ensures the Bank of Canada’s accountability and independence, allows for clear communication about monetary policy with the Canadian public, and has led to superior economic outcomes for households and businesses.

The authors extend gratitude to Daniel Schwanen, Mawakina Bafale, Pierre Duguay, Paul Jenkins, David Laidler, David Longworth, Angelo Melino, John Murray, and several anonymous referees for valuable comments and suggestions. The authors retain responsibility for any errors and the views expressed.