My coauthor on this edition of Canadian Perspectives is Dr. Vasant Naik, head of PIMCO’s empirical research team and a former associate professor of finance at the University of British Columbia, Vancouver.
In recent weeks, major central banks have turned decisively more aggressive in their nontraditional approaches to easing monetary policy. The QE3 (third round of quantitative easing) announcement by the U.S. Federal Reserve (Fed) on September 13 was a major move: The Fed will buy an additional $40 billion in mortgage-backed securities per month (in addition to the approximately $30 billion per month they currently buy to reinvest cash flows from their existing portfolio). This means the Fed is buying approximately 50% of all the U.S. agency mortgages currently being produced!
The previous week, the European Central Bank (ECB) announced a plan to buy unlimited amounts of European peripheral sovereign debt as long as these countries abide by the conditions set out by the European bailout funds (the European Stability Mechanism (ESM) and its predecessor, the European Financial Stability Facility (EFSF)). This major policy initiative for Outright Monetary Transactions (OMTs) is designed to contain the European sovereign debt crisis. It has many folks in Germany worried about the spectre of future inflation.
It is highly debatable whether these extraordinary monetary policy measures will significantly improve the real economy, but one key trend we see is investors fleeing currencies that are actively printing money to currencies that are pursuing traditional monetary policy. This trend has prompted the Brazilian finance minister to once again accuse central banks of inciting currency wars. It has also caused investors to bid up the prices of hard assets, such as the commodities produced in Canada, which has contributed to an appreciation of the Canadian dollar or CAD (up over 2% this year).
There are three main side effects of QE3 and the rise of the CAD:
- A higher CAD may tighten financial conditions in Canada and cause the economy to slow.
- A higher CAD may also cause a disinflationary impulse to the Canadian economy via lower-cost imports.
- QE3 will likely make it more difficult for the Bank of Canada to hike rates. As Mark Carney, governor of the Bank of Canada, has admitted in the past, there are limits to which monetary policy in Canada can diverge from the U.S.
To assess the potential impact of such macro drivers on Canadian inflation, PIMCO has designed proprietary Canadian inflation models. Our empirical research team continually reviews and refines these models.
A model for forecasting core inflation in Canada
We use a systematic approach for arriving at medium-term forecasts for Canadian inflation. Even though the Bank of Canada states its inflation target in terms of headline CPI, it is useful to break down headline CPI into Core CPI and non-core items (volatile components such as certain agricultural and commodity products) in a forecasting exercise. In order to focus on broad-based and persistent inflationary forces in the economy, our framework seeks to forecast Core CPI alone over a one-year horizon.
Our overall method for forecasting inflation is a combination of signals approach. We start with a few categories of variables organized around different themes, and then combine the predictions from each category into one final forecast. Inflation dynamics are highly complex, and each category of variables captures particular aspects of inflationary pressures in the economy and offers clues about the future path of inflation. Our basic premise is that these different categories are complementary in capturing different aspects of inflationary pressures and jointly provide a useful amount of predictive information about those pressures. The three broad groups of variables are depicted in Figure 1.
Variables in the category Output Gap and Labor Market Slack are meant to capture recent trends in overall capacity utilization in the economy. As the output and employment gap in the economy shrinks, price pressures would tend to mount. Similarly, periods of high unemployment and low capacity utilization will likely see disinflationary pressures and slack in aggregate demand. The category Wage and Cost Pressures includes information about recent dynamics of the overall cost of labor. This category also includes changes in exchange rates and house prices, both of which tend to be associated with overall price levels in the medium term. Again, periods of sustained, broad-based inflationary pressures most likely coincide with periods of increases in factor costs. Finally, we include consensus estimates of inflation from a Survey of Economists’ Forecasts as the third category in our model. Economists’ forecasts tend to be quite responsive to near-term developments in the economy, such as supply and demand factors. These may also benefit from the professional forecasters’ qualitative judgments.
Next, we combine the forecasts from three categories into a single forecast. For this, we first construct an aggregate summary measure of the variables included in each category. Then, for each category, we estimate the predictive relationship between core inflation and this summary measure. Our overall forecast is a weighted average of the three individual category forecasts. Such a combination of forecasts from a range of complementary categories of predictors tends to be more robust than the forecast from any one of them.
Current forecast based on the analytical framework
Currently, output gap remains negative and below its long-term average, showing unused capacity in the economy. However, labor market slack has shrunk and these indicators are close to their long-term averages. Indicators of wage and cost pressures have generally been rising this year (although recently they have eased somewhat). The consensus forecast for the headline inflation 12 months ahead has been on a declining trend but is still tracking a rate close to the Bank of Canada’s target. In the past year, core inflation rose to 2.3% in February and has generally declined since then, reaching a low of 1.3% in the most recent September release. Our proprietary model takes a medium-term view and regards the recent low readings as transitory. Overall, we expect core inflation to rise in the coming year.
Real return bonds likely to outperform
The difference between nominal and real 30-year bond yields (known as the breakeven inflation or BEI rate) is approximately 2%. This is the inflation target of the Bank of Canada. Except when market conditions are severely disrupted, such as after the Lehman bankruptcy, the BEI rate is normally higher than the Bank of Canada’s 2% inflation target (see Figure 2) – logically, nominal investors should be compensated above the target rate for taking inflation risk. However, as yields have dropped to near historic lows, this compensation for inflation has been reduced, making real return bonds attractive relative to nominal bonds.
We believe the ECB’s OMT announcement has cyclically truncated the negative left tail risk of a eurozone breakup; thus, we do not expect a repeat of the 2008–2009 market upheavals. In addition, our Canadian inflation modelling points to higher inflation over the coming year. Combining our top-down macroeconomic view with our bottom-up inflation modelling leads us to favor real return bonds instead of nominal yields.