Accessible navigation:

  1. Main page text
  2. Main navigation
  3. Section navigation

Bank of Canada

Regular page >>
      

Monetary Policy

Measuring Economic Growth

Economists use many different methods to measure how fast the economy is growing. The most common way to measure the economy is real gross domestic product, or real GDP. GDP is the total value of everything—goods and services—produced in our economy. The word "real" means that the total has been adjusted to remove the effects of inflation.

There are at least three different ways to measure growth of real GDP. It is important to know which is being used, and to understand the differences among them. The three most common ways to measure real GDP are:

  • Quarterly growth at an annual rate
  • The four-quarter or "year-over-year" growth rate
  • The annual average growth rate

Quarterly growth at an annual rate shows the change in real GDP from one quarter to the next, compounded into an annual rate. (This process is often called "annualizing.") For example, in the second quarter of 2001, the economy grew 0.1 per cent from the first quarter. If the economy had grown at that pace for an entire year, the annual growth would be 0.4 per cent. So the quarterly growth at an annual rate was reported at 0.4 per cent.

This measure is often used by the media. It does a good job of showing recent economic developments. But it also tends to be volatile (see bars in Chart). This is because the effects of any one-time-only factors during the quarter, labour disputes for example, become compounded when the rate is annualized.

The four-quarter, or "year-over-year" growth rate, compares the level of GDP in one quarter to the level of GDP in the same quarter of the previous year. For example, in the second quarter of 2001, GDP was 2.1 per cent above that in the second quarter of 2000. This measure is popular among businesses, who generally present their own quarterly earnings results on that basis to avoid seasonal variations.*

The year-over-year growth rate tends to be somewhat less volatile than quarterly growth at an annual rate (see line on Chart). That is because the effect of any special factors does not get compounded. But it is also less timely, since it looks at what happened to the economy over the entire previous year, not just the past three months.

Finally, the annual average growth rate is the average of year-over-year percentage changes reported during a year. The November Monetary Policy Report indicates that the Bank expects the annual average growth rate for 2001 to be about 1.5 per cent. For the first half of 2001, the year-over-year growth rates as published by Statistics Canada are 2.5 per cent in the first quarter and 2.1 per cent in the second quarter. For the third and fourth quarters, a profile that is consistent with the expectations described in the November Report (say -0.5 per cent and 0 per cent, respectively at annual rates) yields year-over-year growth of 0.9 per cent in the third quarter and 0.5 per cent in the fourth quarter. Averaging the four year-over-year growth rates in 2001 gives the annual average growth rate of 1.5 per cent (dashed bar in Chart).


* Projected growth rates incorporate "First Scenario" values.

Each measure has strengths and weaknesses. But mixing up the measures can lead to results that may look confusing at first glance. The Table below provides some examples that illustrate this. In the Table, the numbers for 2001Q1 and Q2 are as reported by Statistics Canada. For the next six quarters from 2001Q3 to 2002Q4 the numbers provide two illustrative scenarios designed to make a point. The illustrative scenario in the top panel is broadly consistent with the economic outlook described in the November Report: zero to slightly negative growth in 2001H2, 2 per cent growth in 2002H1, and 4 per cent growth in 2002H2.** The annual average growth rate for 2002 is 1.5 per cent. This sounds low, but as the quarterly growth at annual rates illustrates, to achieve this annual average requires a considerably stronger quarterly profile through 2002. The reason for this is that the annual average growth for 2002 is pulled down by the very weak growth in the second half of 2001.

To illustrate this point, the lower panel of the Table puts the quarterly growth at annual rates in 2001Q3 and Q4 arbitrarily at 3 per cent, while leaving the profile for quarterly growth at annual rates for 2002 unchanged. With this change to the second half of 2001, 2002 begins from a higher starting point so, while the quarterly profile in 2002 is the same as in the upper panel, the annual average growth rate is a full percentage point higher at 2.5 per cent.

The Table also illustrates another point. Note that in the upper panel the annual average growth rates for 2001 and 2002 are the same but the quarterly profiles in the two years are very different. Through 2001 growth decelerates, while in 2002 growth picks up through the year.

The Bank of Canada uses average annual growth as a summary measure of broad trends. Annual averages are also useful when comparing to other forecasters. However, the Bank uses the other measures to focus on shorter-term developments.

FIRST SCENARIO
  2001 (illustrative after Q2) 2002 (illustrative)
  Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Quarterly growth at an annual rate: 2.0 0.4 -0.5 0.0 1.0 3.0 4.0 4.0
Four-quarter or year-over-year growth: 2.5 2.1 0.9 0.5 0.2 0.9 2.0 3.0
Average annual growth rate:     1.5       1.5  
SECOND SCENARIO
  2001 (illustrative after Q2) 2002 (illustrative)
  Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
Quarterly growth at annual rate: 2.0 0.4 3.0 3.0 1.0 3.0 4.0 4.0
Four-quarter or year-over-year growth: 2.5 2.1 1.8 2.1 1.9 2.5 2.7 3.0
Average annual growth rate:     2.1       2.5  

* GDP data published by Statistics Canada are already adjusted for seasonal variations.

** The Bank of Canada's practice of citing average quarterly growth in each half of the year is an attempt to seek a balance between volatility and timeliness. The results are less volatile than quarterly growth at an annual rate, and more timely than year-over-year growth rates.