Monetary Policy: How It Works, and What It Takes
At the heart of the Bank's monetary policy is a commitment to maintain low and relatively stable inflation—in particular, to keep the rate of inflation close to the 2 per cent midpoint of the 1 to 3 per cent target range. The Bank's commitment is essential for influencing firms' and households' expectations of inflation. Faced with a shock that threatens to push inflation either above or below the inflation target, Canadian firms and households are confident that the Bank will act to bring inflation back to the 2 per cent target. This confidence in the Bank's policies is a result of the Bank's past record of doing what it claimed it would do—keep inflation low and stable.
When the Bank of Canada has clearly stated objectives and takes policy actions that affirm those objectives, the result is an increase in its credibility. This credibility, in turn, helps to keep expectations of future inflation close to the inflation target—what is sometimes called an anchoring of inflation expectations. The importance of well-anchored inflation expectations is best illustrated by recalling what happens when such anchoring is not present, as in the 1970s and 1980s. During those years, the inflation-control process was difficult because economic shocks led to adjustments in expectations which, in turn, led to behaviour that influenced actual inflation. An important lesson learned since the early 1990s, in Canada and elsewhere, is that keeping inflation expectations well anchored is an important part of keeping actual inflation low and relatively stable.
With these comments as background, it is now possible to examine a simple analytical framework illustrating how the policy actions of the central bank influence many macroeconomic variables and, ultimately, help to keep inflation low and relatively stable.
4.1 The transmission mechanism of monetary policy
The transmission mechanism is the complex chain of cause and effect that runs from the Bank of Canada's actions to changes in asset prices, aggregate demand, the output gap and, eventually, inflation. Among economists, there is some debate about the nature of the transmission mechanism. Engert and Selody (1998), for example, emphasize the important distinction between the "passive-money" and "active-money" views of the transmission mechanism and argue that the possibility of making policy errors can be reduced by paying attention to both views. 1 Even among those who agree on the broad nature of the mechanism, there is recognition of considerable uncertainty regarding the timing and quantitative importance of specific linkages. A collection of speeches and research papers published by the Bank of Canada (1996) provides a mainstream view of the transmission mechanism. 2 Chart 6 is a simplified illustration of the transmission mechanism. The dotted line in the chart shows how the Bank's policy commitment, and thus its acquired credibility, helps to anchor firms' and households' expectations of future inflation.
The transmission mechanism is best understood by tracing through the effects of a hypothetical policy decision. For example, consider a situation akin to that in the autumn of 2004, when the Bank had good reason to expect that the solid economic recovery occurring both in Canada and in the global economy would create pressures for Canadian inflation to rise over the coming months. In this case, the Bank's policy response was to raise its target for the overnight interest rate. How does such a policy action help to contain inflationary pressures?
Following the announcement of the Bank's policy action to increase its target for the overnight rate, the actual overnight interest rate adjusts almost instantly. As the overnight interest rate rises, two responses are observed. First, the hike in the overnight rate leads to an increase in longer-term interest rates in Canada. This increase occurs because there is an entire spectrum of financial assets, ranging from overnight loans to 30-year bonds, and their rates tend to move together. The second response is that, as Canadian interest rates rise, financial capital from around the world flows into Canada in pursuit of higher yields. This capital inflow leads to an appreciation of the Canadian dollar. At this point, the stage is set for a change in quantities, specifically investment, and net exports.
The rise in longer-term interest rates increases the cost of borrowing and thus tends to dampen the demand for interest-sensitive expenditures. In particular, the rise in interest rates will reduce the growth rate in firms' demand for investment goods and households' demand for residential housing and other big-ticket items such as automobiles, furniture, and appliances. The appreciation of the Canadian dollar increases the prices of Canadian products relative to foreign ones and thus leads to adjustments in spending. Specifically, the demand by foreign consumers for Canadian products will be dampened, and the demand by Canadian consumers for foreign products will be stimulated. The growth of Canadian exports will decline, and the growth of imports from abroad will increase—a decline in the growth of Canadian net exports. Taken together, the reduced growth rates of household consumption, firms' investment demand, and net exports imply a reduction in the growth rate of Canadian aggregate demand. 3
Over short periods of time, perhaps up to a few years, output in the economy is determined by the level of demand for all of the goods and services produced in Canada—that is, output is determined by aggregate demand. With aggregate demand now growing more slowly than before the Bank's policy action, there will soon follow a slowing of the growth rate of actual output. With some given underlying growth rate of potential output, this reduction in the growth rate of actual output implies the opening (or enlarging) of a "negative" output gap, in which actual output is less than potential output.
The final step in the transmission mechanism is the link from the output gap to the rate of inflation. The output gap is a convenient measure of the amount of excess demand or excess supply in the aggregate economy. With actual output below potential output, firms are producing below their capacity. This situation of excess supply can continue for a while, but eventually, the economic slack leads to a reduction in the growth rate of wages and the prices of other inputs. This reduction in the growth of firms' costs then leads them to be less aggressive in increasing their prices, and inflationary pressures subside.
4.2 Time and uncertainty
Thus, we see how the Bank of Canada's action to raise its target for the overnight interest rate sets in motion a complicated sequence of cause and effect that helps to keep inflation from rising above its target. 4 But how much time elapses between the Bank's policy action and the other effects, especially the final effect on the rate of inflation?
Economists usually think of the transmission mechanism as containing "long and variable lags," indicating not only that central bankers must be patient while waiting for the results of their policy actions, but also that they must be prepared to accept a few surprises while they're waiting. The workings of the economy are sufficiently complex, and our understanding of the economy sufficiently incomplete, that the various lags in the process may turn out to be either longer or shorter than initially expected.
The Bank of Canada's policy actions relating to the overnight interest rate have almost immediate effects on the exchange rate and interest rates, but current estimates suggest that it takes between 12 and 18 months for most of the effect on aggregate output to be observed. Most of the effect on inflation is not apparent for between 18 and 24 months (Duguay 1994). And even these estimates are subject to considerable variation.
The time lags inherent in the transmission mechanism make it difficult to conduct monetary policy. In particular, these long time lags mean that central banks must be forward-looking in their policy decisions. Consider two examples to illustrate the point. If, on 1 January 2005, the Bank of Canada observes an event in the world economy that is likely to reduce aggregate demand beginning in June of the same year, there is nothing the Bank can do in January to fully offset that shock. Even if it responded immediately and lowered its policy rate in early January, there simply would not be enough time for its policy to stimulate aggregate demand sufficiently to offset the effects of the shock by June.
This example illustrates why the Bank places more emphasis on long-lasting shocks than on shocks believed to be short lived: monetary policy takes so long to work that it makes little sense to respond to shocks that will have disappeared by the time the policy takes effect. A related difficulty, however, is that it may take the Bank a few months to fully determine the nature of the shock, as well as its likely duration. Thus, a forward-looking central bank must do its best to anticipate what events are likely to occur in the future.
A second example is more familiar in recent Canadian experience. Suppose the current rate of inflation is low, but that economic developments expected to occur in the near future will create an undesirable increase in future inflation. In such a situation, a central bank that recognizes the importance of the long lags in the transmission mechanism will increase its policy interest rate now even though current data suggest that there is no inflation problem. This example illustrates why vigilant central bankers are often accused of fighting demons that are not present. The problem is that if central bankers delay their policy response until inflation actually appears in the data, it would be too late to have the desired impact. Being forward-looking means anticipating where the demons will surface and acting in a pre-emptive manner. Although a central bank that is successful in anticipating future inflationary pressures, and also in taking the appropriate actions to prevent their full realization, can easily be accused of seeing dangers that are not genuine, the truth may be that the dangers were avoided only because of the central bank's vigilance.
Although many central bankers may do a good job of predicting future events, no central bank has a crystal ball—the future is only clearly visible once you are there. Indeed, it is actually worse than this in economics. Because of imperfect information that is often revised, even several months after the fact, economists have a difficult time knowing with precision what is happening in the current quarter until they are two or more quarters down the road. This mention of imperfect information leads us to discuss the informational requirements for monetary policy. In a nutshell, the successful conduct of monetary policy requires good information, and plenty of it.
4.3 Monetary policy faces two types of uncertainty
In order to understand the informational requirements for monetary policy, it is helpful to reconsider the nature of the transmission mechanism. Chart 7 shows the transmission mechanism again, but adds two types of uncertainty. The first type involves uncertainty about the details in the transmission mechanism itself; that is, uncertainty about the precise nature of the linkages between key macroeconomic variables. This uncertainty is shown by the pink numbered balloons. The second type is uncertainty about current and future economic developments in the domestic and world economies, as shown by the yellow starbursts.
Six instances of the first kind of uncertainty are shown in Chart 7, each referring to a different aspect of the transmission mechanism. The nature of the uncertainty in each case is as follows:
1. Term structure. How do the Bank's changes in the target overnight interest rate lead to changes in longer-term interest rates? Are the changes always in the same direction? What magnitude of changes are observed for longer-term interest rates?
2. Foreign exchange market. How do the Bank's changes in the target overnight interest rate lead to changes in the exchange rate? How big a change in the exchange rate typically follows a change in the policy rate by the Bank of Canada?
3. Interest sensitivity of spending. How much, and over what timeframe, do aggregate consumption and investment respond to changes in longer-term interest rates? Do different components of consumption and investment have different responses to changes in interest rates?
4. Sensitivity of net exports. How much do exports respond to a change in the exchange rate, and with what time lags? How quickly and in what magnitude do imports respond to the same change in the exchange rate?
5. The multiplier. How big is the "multiplier" that connects initial changes in aggregate demand to the overall change in aggregate output? Over what time period are the full effects on aggregate output observed? 5
6. Excess demand or supply. How quickly does the excess demand or supply associated with any given output gap cause changes in the growth rate of wages and the prices of other inputs? How quickly do these changes show up in inflation?
Eight examples of the second type of uncertainty are shown in Chart 7, with the yellow starbursts each referring to a different kind of shock that can affect the economy. A brief description of each is as follows:
A. Portfolio adjustments. For several reasons, creditors may decide to adjust their holdings of short-term and long-term Canadian securities, leading to changes in Canadian interest rates.
B. Foreign exchange market. Changes in exchange rates occur daily and for many reasons, including changes in the growth of the global economy, changes in world commodity prices, and changes in international asset portfolios.
C. Consumption and investment. Households change their spending, and firms change their investment plans, often in unpredictable ways. Expectations regarding future economic conditions are important.
D. Government expenditures. Canadian governments (federal, provincial, territorial, and municipal) change their spending on an annual basis, sometimes in unexpected ways.
E. Net exports. Changes in foreign income lead to changes in the demand for Canadian exports. The rise of specific countries in the production of certain goods frequently leads to changes in world demand, either away from or towards Canadian goods.
F. Potential output. The economy's production capacity is not directly observable, and therefore must be estimated. Its growth depends on labour-force growth, the accumulation of physical and human capital, and the growth of productivity. Changes in potential output often cannot be detected until well after the fact.
G. Inflation expectations. Large and sudden changes in the prices of specific products frequently lead to changes in inflation expectations. However, the central bank's commitment and credibility help to anchor expectations in the face of such shocks.
H. Inflation shocks. The rate of inflation is regularly affected by changes in indirect taxes, sharp changes in the prices of specific products, and by changes in the exchange rate that alter the Canadian-dollar prices of imported products. Not every change in measured inflation is caused by excess demand or supply in the Canadian economy. 6
This collection of uncertainties—about the economic linkages and the economic events—is crucial to the conduct of monetary policy, not least because of the long and variable lags that we discussed earlier. For the Bank of Canada to set its policy interest rate now in order to keep inflation within its target range in the future, it is necessary for the Bank to anticipate the likely changes in the economy that will occur over the next two years. It is also necessary for the Bank to anticipate how its actions will be transmitted through the economy. Since no central bank has the ability to foretell the future or has perfect knowledge of the various linkages in the economy, this is a difficult task. But knowledge of the transmission mechanism, simplified as it is in Chart 7, permits the Bank to be systematic about which questions it asks, and to be analytical about interpreting some of the answers.
This discussion underscores why monetary policy is best viewed as a problem of policy-making under uncertainty. Faced with such uncertainty, the Bank needs to be forward-looking, aware of many possible shocks that may occur in the near future. It must also be aware that economic developments shown to be present by current data may not persist for long, or may in the near future be revealed, through a revision of the data, never to have existed at all. 7 Thus, the Bank is forced to perform a precarious balancing act—sometimes taking action in anticipation of what is likely to happen while at other times waiting to see what new data are confirmed as genuine. Not surprisingly, good judgment based on considerable experience is an essential part of good monetary policy.
4.4 Good information, and plenty of it
In addition to judgment and experience, consider what the Bank of Canada—and any other central bank—requires in order to conduct the best possible monetary policy in the face of uncertainty.
The Role of Research
To deal with the uncertainty regarding the various linkages between macroeconomic variables (the pink balloons in Chart 7), the Bank needs to conduct a significant amount of economic research, both theoretical and empirical, and to subject the results of this research to ongoing testing. The nature of modern economies is such that this job will never be finished, and the complete set of answers will never be known. For example, even if the collection of economic researchers were able to estimate the precise relationship between changes in longer-term interest rates and firms' planned investment spending, there is no reason to believe this relationship will be stable over time, or that factors that proved to be important in the past will remain of central importance in the future. Economic relationships depend in important ways on human behaviour, which itself depends on the specifics of time, place, and circumstance.
In short, the economic relationships that are central to the conduct of monetary policy are difficult to pin down and are constantly changing. This simple fact requires the Bank to be continuously conducting research on the nature of the transmission mechanism. To do otherwise would be to abrogate its central responsibility.
The Importance of Current Analysis
Dealing with the uncertain developments in the domestic and world economies (the yellow starbursts in Chart 7) requires information of a different kind. In order to know what events are occurring, and what events are likely to occur in the near future, the Bank needs to collect and analyze a great deal of current data—a process that is often called current analysis. Although the relatively small number of yellow starbursts in Chart 7 may suggest that the required effort in this direction is commensurately small, this suggestion would be misleading. In fact, the large quantities of variables that feed into each yellow starburst, and the inherent complexity involved in understanding each individual variable, mean that the task of current analysis for any central bank is Herculean. Thus, a great many people at the Bank are assigned the task of collecting and analyzing data on hundreds of variables, from employment and exports to commodity prices and housing starts, from government spending and exchange rate regimes to domestic steel production and foreign crop failures. Only when the various shocks to the economy are observed and understood can the Bank hope to incorporate that information fruitfully into its overall decision-making.
A relatively new and important example of data collection and analysis that the Bank carries out in an attempt to better understand the emerging trends in the economy is the Business Outlook Survey (BOS). Four times a year the Bank's regional offices survey approximately 100 firms, the overall sample chosen to be roughly representative of the Canadian economy. A number of issues are explored, including the firms' views on likely future demand for their own products, capacity pressures in their specific sectors, any emerging labour shortages, and the firms' own plans for hiring or expansion. By analyzing these data carefully, the Bank is able to better understand how Canadian firms respond to the various shocks affecting the Canadian economy.
Economic research and current analysis are not independent activities. In order to conduct thorough empirical economic research, knowledge of the data is essential, and such knowledge typically comes from experience in current analysis. Conversely, the ability to interpret current data—what is going on and why?—requires a thorough knowledge of economic relationships that comes from experience in research. This ongoing interaction between research and current analysis explains why many economists at the Bank of Canada are in positions that require a regular transition between current analysis duties and research projects.
The best example of how the insights gleaned from economic research are combined with the knowledge embodied in current analysis is in the Bank's regular projection, or forecasting, exercise, based on its large and complex statistical model of the Canadian economy, the Quarterly Projection Model (QPM). Embodying the knowledge of economic relationships gained from many years of research, QPM is a mathematical representation of the interaction of the various agents in the Canadian economy—households, firms, and governments—and shows how these relationships must evolve over time to be consistent with the underlying assumptions of agents' behaviour. The model then incorporates past and current data from the Canadian and world economies and projects the most likely future path of Canadian macroeconomic variables—including output, employment, wages, and prices.
The world rarely turns out as the model projects, for two reasons. First, the model itself, though extremely complex, is nonetheless a highly simplified description of the real economy. It lacks the remarkable and changing complexity that actually characterizes any modern economy. Second, the data that are fed into the model, as good as they are, are also imperfect, and the Bank's best predictions regarding what is actually happening in the Canadian and world economies may well turn out to be wrong in some way. Nonetheless, the economic projection provides the Bank with a logically consistent and well-articulated starting point regarding the future evolution of the Canadian economy, as well as a starting point for analyzing the likely future impact of its policy actions.
These three exercises in information creation—research, current analysis, and economic projections—are imperfect and, necessarily, ongoing. Research will never be entirely "correct" and thus will never be complete. Current analysis, by its very nature, must be an ongoing process, with constant effort expended to improve data definitions and accuracy. And the art of model building and producing sensible and consistent macroeconomic projections is, perhaps unbelievably, still in its infancy. Such failings, however, in no way suggest that these activities can be forsaken. It would be impossible to conduct prudent monetary policy without the creation and provision of such information, and it is thus not surprising that central banks the world over invest considerable resources in these three key activities.
- Laidler and Robson (2004, especially Chapter 3) offers a balanced and very readable description of both views of the monetary policy transmission mechanism. [←]
- See in particular, Thiessen (1996) for a description of the transmission mechanism and Duguay (1994) for empirical estimates. [←]
- Government purchases of goods and services are the final component of aggregate demand. In this analysis, we focus on the effects of monetary policy actions and assume that government spending (fiscal policy) is unchanged. [←]
- The same logic, working in the opposite direction, applies to the situation in which the Bank lowers its target for the overnight interest rate in an attempt to keep inflation from falling below its target. [←]
- The "multiplier" describes how changes in aggregate demand are multiplied in size in the short run. An initial increase in demand leads to more output and income which, in turn, leads to a further increase in demand, a further increase in output, and so on. [←]
- This is one reason why the Bank of Canada monitors movements in "core" inflation. By removing eight particularly volatile elements (see the list in footnote 7) from the measure of CPI inflation and adjusting for the effect of changes in indirect taxes, the resulting measure of inflation provides a better indication of excess demand or supply pressures in the economy. See Macklem (2001). [←]
- This third type of uncertainty, uncertainty about the accuracy of data, is discussed in Bean (2005). [←]