Executive summary

View the most recent version.

The paper examines the level of multifactor productivity (MFP) in Canada relative to that of the United States, using data that are especially constructed for this purpose to provide somewhat more comparability than has existed in the past.

A previous paper (Baldwin et al. 2005) explored the problems that needed to be resolved in order to produce estimates of the level of Canada's labour productivity relative to that of the United States. Existing estimates of the gap in the level of labour productivity were so diverse that an effort was made to provide standardized data that could be used to better estimate the level of labour productivity between Canada and the United States. The same problem exists with regard to different estimates of the level of MFP in Canada compared with that of the United States.

The relative labour productivity level measures the relative efficiency with which the two economies transform labour into output. The relative MFP level compares the relative output differences across two countries, not just to labour input differences but also to capital input differences, and it thus provides a more 'complete' measure of overall efficiency. The relative MFP level captures the difference in the overall efficiency of an economy that arises from the use of superior production techniques, technology, firm organization, firm scale and labour quality.

In addition to providing an estimate of the relative level of Canada–United States MFP, the paper also relates the estimates of relative MFP to the estimates of relative labour productivity. The differences in labour productivity between the two countries can be decomposed into two components: a component that comes from differences in the amount of capital per worker—capital intensity—and a component that comes from differences in MFP—overall efficiency. The decomposition allows us to examine the relative importance of differences in capital intensity and overall production efficiency in accounting for labour productivity differences between Canada and the United States.

Finally, the paper moves beyond the total economy (or the total business sector) that was examined in Baldwin et al. 2005. It examines differences in the relative Canadian-U.S. levels of productivity in different industries in order to estimate the contributions that individual industries make to the overall gap in MFP.

Several questions are posed in this section of the paper.

1. What are the data issues that have been addressed in this paper for comparing relative Canadian-U.S. levels of output, inputs and MFP?

An estimate of the relative MFP levels in Canada and the United States requires comparable data on outputs and inputs. To do so, the paper starts by choosing comparable data for labour, investment and output from the various sources that are produced by the statistical agencies in Canada and the United States. The paper then adjusts each, where required, to improve comparability and applies similar methodologies to estimate capital input, labour input and industry output from the appropriate source data. For capital stock, the paper starts with the investment data by asset type and it then uses the same methodology—the perpetual inventory technique using similar geometric depreciation rates—to estimate capital stock in the two countries. For labour input, the paper estimates hours worked using comparable methods and comparable microdata sources in the two countries (see Baldwin et al. 2005 for details). For output, the paper constructs a measure of output for U.S. industries that is calculated at basic prices that can be compared with the measure of output for Canadian industries that is calculated at basic prices.

The other issue that needs to be addressed is the choice of a price unit that is to be used to compare Canadian and U.S. gross domestic products (GDPs), each of which are produced in their national currencies (Canadian and U.S. dollars, respectively). In order to compare outputs between Canada and the United States that are calculated in their respective dollar currencies we need price deflators to transform relative values of GDP measured in dollars to measures of relative real output. Relative price indices that are referred to as purchasing power parities (PPPs) are used for this purpose. These PPPs are calculated using final or market prices in the two countries. These market-price PPPs are then adjusted to be consistent with the basic-price concept of output used for this paper. The adjustment is based on relative product tax rates between Canada and the United States (or the ratio of GDP at market prices to GDP at basic prices) for the total business and non-business sectors.

2. What are the relative Canadian-U.S. levels of labour productivity, MFP and capital intensity in the total economy?

The aggregate level of labour productivity in Canada was about 89.8% of the U.S. level in 1999. The capital–labour ratio in Canada has been essentially the same as that in the United States since 1999. With capital–labour ratios that are quite similar in the two countries, the relative MFP is quite close to the relative level of labour productivity. The aggregate level of MFP in the Canadian total economy was 87.6% of the U.S. level in 1999.

The similarity in the capital–labour ratio between Canada and the United States masks considerable differences in the type of assets used in each country. In 1999, the ratio of structure capital to labour in Canada was 130% of the same ratio in the United States. The ratio of machinery and equipment (M&E) capital to labour in Canada was 78% of that in the United States. Factor proportions for the total Canadian economy are quite different than for the U.S. economy, at least when it comes to the composition of capital.

3. What are the trends in relative Canadian-U.S. levels of labour productivity, MFP and capital intensity in the total economy for the period from 1994 to 2003?

Before 2000, the relative level of labour productivity in Canada and the United States experienced little change. During the 1994-to-2000 period, the aggregate level of labour productivity in Canada was about 90% of the U.S. level. After 2000, the relative level of labour productivity in Canada declined from 90% of the U.S. level in 2000 to 87% of the U.S. level in 2003.

The aggregate level of MFP in Canada relative to that in the United States increased over the 1994-to-2000 period and then it declined slightly after 2000. It rose from 85% in 1994 to 90% in 2000, and then declined to 89% in 2003. At the same time, the aggregate level of capital–labour ratio declined in Canada relative to that in the United States. The decline was a result of the lower growth in structure and equipment capital stock and a higher growth in hours worked in Canada.

4. What are the sources of the Canadian-U.S. labour productivity level differences in the total economy?

We have decomposed labour productivity differences between Canada and the United States into contributions arising from differences in MFP and differences in capital intensity. The lower level of Canadian labour productivity relative to that of the United States was due to the lower MFP level in Canada. The capital intensity was similar in the two countries, which is a result of higher structural-capital intensity on the one hand and lower equipment-capital intensity in Canada on the other hand.

The aggregate level of labour productivity in Canada was 10% behind that of the United States in 1999. The lower level of MFP in Canada lowered the relative level of labour productivity in Canada by 12%. The lower level of M&E capital–labour ratio lowered the relative labour productivity in Canada by 4%, while the higher level of structure capital–labour ratio increased Canadian relative labour productivity by 6%. Overall, the capital intensity in Canada was slightly higher in 1999, which raised Canadian labour productivity by 2%.

5. What are the relative Canadian-U.S. levels of labour productivity, MFP and capital intensity at the sector level?

Relative Canadian-U.S. labour productivity levels vary considerably in the four major sectors of the total economy: engineering, goods, services and non-business sectors. The first three sectors make up the total business sector. The non-business sector includes the government sector, the health and education sector and owner-occupied dwellings.

The level of labour productivity in the Canadian business sector was about 84% of the U.S. level in 1999. But this aggregate hides considerable differences across the various sectors.

The level of labour productivity in the Canadian non-business sector was similar to that in the United States, as it should be since output is basically equal to inputs in the public sector, by construction, and the purchasing power parity (PPP) is calculated here from relative factor costs.

The service sector in Canada had the lowest labour productivity level relative to that in the United States—at 76% in 1999. The engineering sector, on the other hand, had the highest relative productivity (99%) in 1999. The goods sector in Canada was at 81% of American labour productivity in 1999, above the relative level of the service sector but considerably below the engineering sector.

The engineering sector in Canada was much more capital intensive than its U.S. counterpart in 1999. The goods sector in the two countries had similar capital–labour ratios in 1999, while the Canadian service sector was less capital intensive than its U.S. counterpart. When capital intensity is considered and MFP is estimated for the three sub-sectors of the business sector, the sectors move closer to one another because the relative capital intensity of the engineering and goods sectors in Canada are above that of the service sector.

The service and goods sectors in Canada had lower levels of MFP relative to those of the United States and were behind the United States by about 20% in 1999. The engineering sectors had a higher relative MFP level, behind the United States by 15% in 1999.

6. What are the sources of the Canadian-U.S. labour productivity level differences at the sector level?

The lower level of labour productivity in Canada relative to the United States in 1999 was mainly due to the lower level of MFP in all three major sectors: goods, services and engineering sectors. The lower M&E capital intensity was also a factor behind the lower labour productivity level in the Canadian engineering and services sectors but not in the goods sector. The higher structure-capital intensity in Canada increased the relative level of labour productivity in the three major sectors, and the effect was larger in the engineering sector and was smaller in the goods and service sectors.

7. What are the industry origins of the lower levels of labour productivity and MFP in the Canadian economy relative to the United States?

The lower level of labour productivity and MFP in the Canadian service sector accounted for most of the labour productivity and MFP differences between Canada and the United States in 1999. This is a result of the service sector having the lowest relative levels of labour productivity and MFP in Canada and having the largest share of output and hours worked in the total economy.

8. What is the range of relative Canadian-U.S.MFP levels estimates in the total economy?

The relative MFP-level estimates are derived from constructs that measure output, labour and capital and that are combined analytically in the MFP estimates using weights on both relative capital and labour inputs. Our estimates of each of the variables are subject to error and, therefore, the point estimates of MFP are subject to a degree of uncertainty.

To provide the reader with guidance on the nature of the range within which the true estimate is likely to be found, the paper considers sequentially alternative estimates of PPPs, error margins in our estimates of the relative levels of capital–labour ratios and possible differences occasioned by the choice of different weights in aggregating investments in M&E and structures. The paper finds that the Canadian MFP level in the total economy likely ranges from 86% to 96% of the U.S. level in 1999.

View the publication Relative Multifactor Productivity Levels in Canada and the United States: A Sectoral Analysis in PDF format.