Monetary policy, interest rates and the Canadian dollar

Introduction

Like a canoe navigating the shifting currents of Canada’s rivers, the Canadian dollar flows with the rhythms of global markets, risk appetites and economic growth. This floating exchange rate allows for independent monetary policy that can be adapted to the needs of the Canadian economy.

By summer 2024, inflation in Canada returned near the Bank of Canada’s 2% target from a peak of just over 8% in 2022. But in the United States, high inflation was more persistent and hovered above its target for most of the year. Because the economies started to follow distinct paths in Canada and the United States, investors and economists anticipated that a wedge would develop between the interest rates set by the Bank and those set by the Federal Reserve Board.

Unsurprisingly, this wedge weighs on the value of the Canadian dollar against the US dollar. When interest rates in Canada decline more quickly than they do in the United States, the value of the Canadian dollar also declines to equalize the returns from investing in one country or the other. As a rule of thumb, when the 1-year interest rate in Canada drops by 1% below that of the 1-year interest rate in the United States, the exchange rate should drop by 1%. Without that price adjustment, investors would move their investments across the border to earn a higher return.

But the decrease in the value of the Canadian dollar in 2024 was greater than the wedge that had grown between the paths of the interest rates for the two countries. That’s because investors also consider the risks of investing in one currency versus the other—not just the difference between interest rates.

In 2024, uncertainty clouded international trades and financial markets, and market participants embedded a large risk premium across the exchange rates of most currencies against the US dollar. This broad exchange rate risk premium explains roughly two-thirds of the drop in the value of the Canadian dollar in 2024.

Historically, deviations between the path of interest rates in the United States and Canada have tended to dissipate with time, and the major sources of risks and uncertainties have eventually been resolved. Ultimately, other long-term factors play a greater role in determining the value of the currency. They include the following factors, which usually move slowly and are hard to detect:

• productivity and the composition of real activity
• fiscal and trade policies
• long-run inflation rates

Canadian and US monetary policies are on different paths

At the start of 2024, policy interest rates in Canada and the United States were around the same level. In the United States, the policy interest rate had a peak range of between 5.25% and 5.50%; in Canada, it peaked at 5%.

The stance of monetary policy in the two countries was also similar at the start of 2024, but a wedge soon appeared between the economic outlooks of the two countries. Starting in February 2024, prices in the overnight index swap (OIS) market suggested that investors were anticipating that the policy interest rate would remain high in the United States but not in Canada (Chart 1). The wedge between the paths of policy interest rates widened further, and by the end of 2024:

• the Bloomberg Survey of Forecasters shows that the expected policy interest rate at the end of 2025 would be:
  • 2.50% in Canada
  • 3.75% in the United States
• ignoring a potential term premium, the OIS rates on December 31 suggest that the expected policy interest rate at the end of 2025 would be:
  • 2.64% in Canada
  • 3.92% in the United States

Therefore, both sources indicated a divergence of more than 1 percentage point.

Chart 1: The difference between US and Canadian interest rates is widening

In addition, the Canadian dollar depreciated against the US dollar by 7.7% in 2024, reaching below 70 cents US by the end of 2024. This depreciation—combined with the growing wedge between the path of policy rates across the two countries—is drawing a great deal of attention. While this is a relatively large amount of depreciation, it is not extreme by historical standards—the Canadian dollar depreciated by 7.7% or more in one out of five distinct 12-month periods since 1972.

Large asset managers appear to be building short positions in the futures market to protect themselves against the risk that the Canadian dollar depreciates further (Chart 2). Similarly, investors in the options market are paying higher prices to protect against the Canadian dollar depreciating than they would if they anticipated that the Canadian dollar would appreciate. 

Chart 2: More investors are protecting themselves against the risk of further depreciation of the Canadian dollar

Although investors in the exchange rates’ futures and options markets have revealed growing concerns about the value of the Canadian dollar, dealer forecasts imply that it will rise over the course of 2025 (Chart 3). This is consistent with the presence of a foreign exchange risk premium, which we will discuss further below.

Chart 3: Dealer economists expect the Canadian dollar to appreciate slightly by the end of 2025

Policy divergence is common when interest rates are rising

In fact, this pattern where the path for the Canadian policy interest rate evolves below the path of the US policy interest rate is common (Chart 4). Over the past 30 years, the US policy interest rate has been higher than Canada’s over five distinct episodes. And—apart from the episode in 1996—these periods coincided with cycles when central banks were tightening monetary policy.

The current policy divergence started at the end of a cycle of rising interest rates, like the previous three, where the US policy interest rate ended higher than Canada’s (Table 1, bottom panel).

But this current wedge may get wider or last longer if the US policy interest rate remains high while the rate in Canada continues on an easing path, which is what both OIS markets and professional forecasters appear to predict. If this wedge continues until the end of 2025, the current episode would be one of the longest divergences on record. Past episodes of monetary policy divergence lasted less than three years, with peaks occurring around 12 months into each of the episodes.

In other episodes, the Canadian policy rate exceeded that of the United States by as much as 2.25% (Table 1, top panel). This occurred when monetary policy was easing in the United States. However, this did not occur in 1994 when monetary policy was tightening in the United States and Canada was transitioning to a new inflation-targeting regime. The policy interest rates moved in the opposite direction in only two episodes: 1996 and 2001.

This underscores the close economic ties between the two economies because both countries have shared largely common business cycles (Voss 2004). Indeed, when we repeat the same exercise for other economies, such as the United Kingdom, the euro area or Australia, we find the divergences tend to be wider and last longer.

Chart 4: Over the past five tightening cycles, the US policy interest rate exceeded Canada’s

Table 1: Regression model on Securities Repo Operation volumes

Note: Top panel identifies periods where the US policy interest rate was higher than Canada’s. Bottom panel shows identifies periods when the Canadian policy interest rate was higher than the US rate.

How interest rates affect the value of the Canadian dollar

Forecasters and economists agree that the wedge between the interest rates in Canada and the United States tends to lower the current value of the Canadian dollar against the US dollar while raising its expected appreciation. There is a simple intuition for this effect: when the interest rates are lower in Canada than they are in the United States, investors prefer to hold investments denominated in US dollars to earn a higher return. This higher return gives investors an incentive to move their investments to the United States until the Canadian dollar depreciates enough to equalize the expected returns between the two countries.

Consider the real return,\(\displaystyle\, r_{t+1},\) that investors would earn if they borrow US$1 at time t, convert it to Canadian dollars at\(\displaystyle\, q_{t}\) and invest it at the Canadian interest rate at\(\displaystyle\, i_t^*\) until time t+1. Investors then convert the proceeds back to US dollars at\(\displaystyle\, q_{t+1}\) and pay back their US loans plus interest\(\displaystyle\, i_t\) in the United States.

\(\displaystyle\, r_{t+1}\) \(\displaystyle=\, q_{t+1}\) \(\displaystyle-\, q_t\) \(\displaystyle+\, i_t^*\) \(\displaystyle-\, i_t\) \(\displaystyle-\, (π_{t+1}^*\) \(\displaystyle-\, π_{t+1}).\) \(\displaystyle\, (1)\)

These are real returns because they account for the difference in inflation between the two countries. When the interest rate is lower in Canada than it is in the United States, the difference can be offset by lower inflation expectations for Canada than for the United States. However, if the spread between interest rates decreases by more than the wedge between expected future inflation in Canada and the United States—possibly because of the monetary policy stance—then this excess spread can be offset by a depreciation of the Canadian dollar, keeping all else unchanged.

So how big is the impact from interest rates? The following back-of-the-envelope calculation shows that the effect is modest.

Ignoring the differences between inflation rates for now, suppose that the Canadian interest rate decreases by 1% for exactly one year. To equalize expected returns, the Canadian dollar will depreciate by 1% today, and investors will expect a rise in value of the Canadian dollar by 1% over the year: (\(\displaystyle\, q_{t+1} - q_t = i_t - i_t^*\)).

In our example, we assume that investors require no additional returns for the relative risk of holding assets in one currency or the other. However, the intuition carries with risk, which shows that investors care about currency risk. We can rewrite equation (1) as:

\(\displaystyle\, E[q_{t+1}-q_t]\) \(\displaystyle=\, E[r_{t+1}]\) \(\displaystyle+\, (i_t-i_t^*)\) \(\displaystyle+\, E(π_{t+1}^*-π_{t+1} )\) \(\displaystyle\, (2)\)

This equation describes investors’ behaviour.1 Investment flows between countries will keep in check:

• the balance between the difference in inflation rates
• the difference in interest rates
• the additional compensation for risk

This effect of exchange rate risk is the same logic by which riskier stocks have a lower price today and offer a higher expected return to investors on average. The key assumption underlying this decomposition is that investors equalize returns across assets held in different currencies, while adjusting for the level of risk. But investors do not necessarily have to have the correct expectations.

The intuition also works in the opposite direction. If investors require less compensation for risks required for holding assets denominated in Canadian dollars, then the Canadian dollar appreciate to equalize the returns per unit of risk.2

We used one period, say one year between t and t+1, in our examples. But the intuition extends over time across multiple horizons: expected returns should equalize this year, the next two years, and so on. We can show this by iterating equation (2) forward to obtain:

\(\displaystyle\, q_t\) \(\displaystyle=\, E_t[ q_{t+m}]\) \(\displaystyle-\, \sum_{j=1}^m E_t[ r_{t+j}]\) \(\displaystyle+\, \sum_{j=0}^{m-1} E_t[ i_{t+j}^* - i_{t+j}]\) \(\displaystyle-\, \sum_{j=1}^m E_t[ π_{t+j}^* - π_{t+j}].\) \(\displaystyle\, (3)\)

Equation (3) breaks down today’s exchange rate into a stream of future returns and inflation rates over some horizon—take m to be 10 years, say. The last three terms extend the same forces that we analyzed in equation (2). The difference is that we must keep track of differences over the long term.

We can interpret each of the four terms in this decomposition:

• the currency’s long-term or steady-state value: determined by the relative productivity of industries that produce tradable goods and services in each country
• risk premium: the additional returns earned by investors for bearing risks of holding assets denominated in Canadian dollars
• interest rate paths: the spread in the expected path of the policy interest rate in each country
• inflation paths: the spread in the expected path of inflation in each country

Market participants tend to share this intuition

We check whether professional forecasters update their projections of interest rates and exchange rates in a way that reflects the intuition that exchange rates will adjust to equalize returns across the borders—in other words, the uncovered interest rate parity intuition. We find that surveys of market participants validate both the direction and the modest scale of the adjustment to the exchange rate suggested by this intuition.

For example, dealers’ forecasts at the end of 2024 suggest that interest rates would remain lower in Canada than in the United States, which likely weighed on the current value of the Canadian dollar (Chart 2). But these forecasts also suggest the Canadian dollar will appreciate over the next year, which is consistent with the intuition.

This pattern is generally consistent across dealers’ forecasts over the past 20 years, on average. We verify this by running a regression of the one-year forecast of the Canadian dollar returns on the current difference between the one-year yields. Using equation (2), the coefficient should equal one if forecasts are based on the uncovered interest rate parity intuition. In the data, we find that the interest rate difference alone explains a very small share of the forecasts for the Canadian dollar, with a low coefficient estimate of 0.29 (Table 2).

However, this lower coefficient is due to the omitted variable bias. The current interest rate difference may be correlated with other variables that should better predict the Canadian dollar returns. Specifically, the interest rate spread exhibits:

• a positive correlation with differences between interest rates over the long term
• a negative correlation with the exchange rate risk premium

When we conduct more robust regressions that account for these other effects, we find that the coefficient of the one-year interest rate difference is indeed close to one and to the uncovered interest rate parity intuition (Table 2). For example:

• including forecasts of future interest rate differentials beyond the next year only marginally improves our ability to explain the forecasts of the Canadian dollar
• including long-term forecasts of future exchange rate risk premium appears to be especially important to understanding the forecasts of the Canadian dollar—in this case, the coefficient estimate for the interest rate difference is 1.1
• including forecasts of the appreciation of price for crude oil—historically linked to fluctuations in the Canadian dollar—to capture changes in the steady-state value of the currency is also important for understanding the market participants’ forecasts

All these coefficient estimates have the correct sign, and adding these effects increases the precision of our estimates as well as reinforces our conclusion that the forecasts of market participants are consistent with the intuition that the exchange rates adjust to equalize returns across the borders.3 The results also show that the difference in the interest rates plays a relatively small role in explaining the variations in dealers’ forecasts of the Canadian dollar.

Table 2: Regression of forecasters’ one-year expected return for holding Canadian dollars

Note: This table reports regressions of forecasters’ consensus 1-year expected return of the Canadian dollar on current 1-year interest rate differentials (IRD) and forecasters’ consensus expectations: 1-year ahead 10-year IRDs, 1-year ahead excess Canadian dollar returns, and the 1-year ahead return on West Texas Intermediate crude oil price. Differentials are defined as United States minus Canada. The 2-year expected excess Canadian dollar returns and the 1-year ahead returns are inferred from 1-year ahead expectations and 3-year ahead expectations on excess Canadian dollar returns. All variables are expressed in percent. The sample period is September 2005 to March 2024.
Source: Department of Finance Canada and Bank of Canada calculations

Increase in the exchange rate risk premium was the main factor behind the depreciation of the Canadian dollar in 2024

Checking our intuition against the forecasts has helped conclude that we should expect cyclical monetary policy divergences to have a modest impact on the exchange rate. We can also bring this idea to the data and quantify the different factors that led the Canadian dollar’s depreciation in the second half of 2024.

We use an econometric model to construct forecasts for each element of the decomposition. Our approach builds on the fact that models of bond prices can extract both forecasts of inflation and of the path of policy interest rates.

Models of bond prices combine historical data on inflation, interest rates and the exchange rate with assumptions about their dynamics to obtain forecasts that are consistent with surveys of forecasters and the path of interest rates. These dynamic models allow us to obtain the relevant forecasts and combine them to compute the components of the present-value decomposition in equation (3). The specifics of the model, the estimation results and several robustness checks are included in Feunou, Fontaine and Krohn (2022).

We focus on the period after summer 2024. The Canadian dollar fell by 4.5% between August and the end of 2024, reaching a value of 0.696 by the end of the year. The widening gap between the model’s expected interest rates in Canada and the United States explains a little more than one-quarter of this depreciation (Chart 5). This estimated effect aligns with our earlier back-of-the envelope calculation: the interest rate difference is around 1% and could converge within roughly two years.

Variations in the exchange rate risk premium played a larger role in 2024, accounting for most of:

• the rise in value between August and October
• the drop in value to the end of the year

The fact that the interest rate difference and the risk premium drive most of the depreciation is consistent with the fact that dealers expect the value of the Canadian dollar to rise over the course of 2025. The important role played by the exchange rate risk premium is also consistent with the significant uncertainty around the anticipated US tariffs on Canadian goods and the associated risk of the Canadian dollar depreciating.

In addition, the recent fluctuations in the value of the Canadian dollar since the middle of 2024 are common across currencies of advanced economies (Chart 6). Most of the depreciation is explained by the dollar portfolio, which is simply the average of other bilateral exchange rates with the United States. Together, these two observations suggest that the patterns in 2024 across the currency markets largely reflected the broad strength of the US dollar resulting from shifts in the investors’ attitudes toward risk in global financial markets. This is consistent with increasing uncertainty surrounding US trade policy, which is anticipated to affect most, if not all, advanced economies.

Policy divergence alone has a modest effect on the exchange rate

We can use the same decomposition framework to simulate scenarios with wider policy divergence than currently implied in survey forecasts and OIS markets. The historical episodes of policy rate divergences can inform this scenario.

Specifically, we consider a hypothetical scenario in which Canada’s policy rate falls another 125 basis points below the US policy rate (Chart 7, panel a). This would match the widest divergence observed since 1994, around 250 basis points. In our hypothetical scenario, we maintain this spread between the policy rates at this wider level for one year, until the end of 2025. This would match the longest divergence since 1994. Effectively, the scenario corresponds to a sudden event in which investors learn a new path for the policy interest rate in 2025.

In addition, we use the model to keep the risk premium constant in the scenario. In the data, the policy responses occur at the same time as broad economic changes. The exchange rate moves because of:

• the changes in each country’s monetary policy stances
• other simultaneous events in global markets

We use the model to analyze this scenario and isolate the effect of the policy divergence in a counterfactual scenario where only the relative monetary policy path between the two countries changes.

As expected, we find that under this counterfactual scenario, the Canadian dollar depreciates immediately. Quantitatively, the model predicts a depreciation of 1.5% against the US dollar, consistent with the uncovered interest rate parity intuition (Chart 7, panel b). This reflects the size and duration of the divergence that we impose in 2025. It also results from the model’s projection, based on historical data, that the divergence will continue beyond 2025. The divergence then gradually narrows in the model, starting in 2026.

The Canadian dollar would then eventually appreciate as the initial shock dissipates and the monetary policy divergence converges close to its historical average. The path of the predicted appreciation matches the size and the persistence of the divergence between the policy interest rates. The model gradually brings the divergence down starting at the end of 2025 and assumes the Canadian dollar to eventually converge to its sample average. Overall, the scenario is broadly consistent with the modest effect of policy divergence on the Canadian dollar.

Chart 7: The effect of further policy divergence is modest if the exchange rate risk premium is kept constant

Chart 7: The effect of further policy divergence is modest if the exchange rate risk premium is kept constant

Past divergences also had modest effects on the exchange rate

When we overlay the time series of the policy rate divergence and of the Canadian dollar exchange rate, we find a complex picture (Chart 8).

• From 1995 to 2006, the Canadian dollar shows little to no relationship with the two-year spread, which we use to proxy for policy divergence over a longer period. This is also consistent with results in the earlier literature on this topic (see Amano and Van Norden 1995; Baxter 1994).
• Since 2006, policy divergences and the Canadian dollar exhibit a strong positive correlation. However, most of the fluctuations in the Canadian dollar since 2006 were common with other currencies.4

This suggests that the risk premium has been playing a dominant role behind the evolution of the Canadian dollar over the past decade, influenced by broader economic conditions and shifts in the global attitude toward risk. This is similar to our findings for the 2024 episode of monetary policy divergence.

Chart 8: Historically, periods of negative two-year yields spread have not led to a sharp depreciation of the Canadian dollar

Large and lasting change in the exchange rate requires a shift in fundamental factors

The largest fluctuation in the Canadian dollar in recent history occurred in two phases between 2004 and 2014 (excluding the 2008–09 global financial crisis):

• between 2004 and 2008, the Canadian dollar rose from 70 cents US to US$1
• between 2013 and 2015, the Canadian dollar fell toward 75 cents US from US$1

In both periods, the fluctuations were significantly larger in magnitude than those implied from the policy divergence. They were also too large and persistent to be explained by shifts in the exchange rate risk premium.

Exchange rate swings of this magnitude require a shift in fundamental factors, such as changes in:

• productivity across countries
• the composition of economic activities
• other structural features of each economy

In the present-value decomposition that equation (3) offers, the first term captures these long-term changes. However, measuring these changes is difficult because the exchange rates are inherently volatile. Cyclical variations in interest rates and the exchange rate risk premium create a disconnect that researchers have identified in empirical work (Meese and Rogoff 1983; Cheung, Chinn and Garcia Pascual 2005).

Nonetheless, the large swings between 2004 and 2014 support the important role these factors play. During that period, Canada received large foreign investments in the oil and gas sector. These investments contributed substantially to the gross capital formation in Canada (Chart 9), as a share of gross domestic product. They also represented significant innovations that reduced the costs of extracting oil and gas from new fields. This shift had not yet occurred in the United States, which led to an improvement in the terms of trade and a strengthening of the Canadian dollar over several years.

Chart 9: Persistent change in the foreign exchange rate requires a fundamental shift in the economy

Looking forward

Prices in financial markets and responses to surveys of forecasters point to a persistent wedge between policy interest rates in Canada and in the United States. While this divergence will continue to influence the Canadian dollar, we expect this impact to be modest. Instead, foreign investors’ compensation for risk currently plays a more significant role in weighing on the value of the Canadian dollar. Structural changes between two economies could also be influencing the Canadian dollar. Investment and productivity growth have accelerated in the United States compared with other advanced economies in recent years. If these signs of growth turn into a sustained pattern, the US dollar could strengthen further against a basket of currencies, including Canada’s. Therefore, monitoring changes in these long-term factors is essential to understanding the path of the Canadian dollar moving forward.

Acknowledgements

We thank Olivier Gervais, Geoffrey Dunbar, Bruno Feunou, James Ketcheson, Miguel Molico and Yash Chauhan for their helpful comments and suggestions on this analysis. We thank Eugene Trostin for excellent research assistance. Finally, we are extremely grateful for excellent editorial assistance from Nicole van de Wolfshaar.

References

Amano, R. and S. van Norden. 1995. “Terms of Trade and Real Exchange Rates: The Canadian evidence.” Journal of International Money and Finance 14 (1): 83–104.

Cheung, Y.-W., M. D. Chinn and A. Garcia Pascual. 2005. “Empirical Exchange Rate Models of the Nineties: Are Any Fit to Survive?” Journal of International Money and Finance 24 (7): 1150–75.

Fama, E. F. 1984. “Forward and Spot Exchange Rates.” Journal of Monetary Economics 14 (3): 319–338.

Feunou, B., J.-S. Fontaine and I. Krohn. 2022. “Real Exchange Rate Decompositions.” Bank of Canada Staff Discussion Paper No. 2022-6.

Fontaine, J.-S. and G. Nolin. 2016. “The Share of Systematic Variations in the Canadian Dollar—Part I.” Bank of Canada Staff Analytical Note No. 2016-15.

Baxter, M. 1994. “Real Exchange Rates and Real Interest Differentials: Have We Missed the Business-Cycle Relationship?” Journal of Monetary Economics 33 (1994): 5–37.

Meese, R. A. and K. Rogoff. 1983. “Empirical Exchange Rate Models of the Seventies. Do They Fit Out of Sample?” Journal of International Economics 14 (1–2): 3–24.

Voss, G. 2004. “Aspects of Canadian and US Business Cycles.” In Canada in the Global Economy: 411–446. Proceedings of Conference Held by the Bank of Canada, November. Ottawa, Ontario: Bank of Canada.

Endnotes