The debate over tracking Canada’s productivity performance continues.
CIBC World Markets chief economist Avery Shenfeld dumps on new research released this week that suggests Statistics Canada has badly underestimated so-called multi-factor productivity since the early 1960s.
Mr. Shenfeld joked in a research note Friday that “New Productivity Measure” might be the most boring headline since “Worthwhile Canadian Initiative.”
The source of Mr. Shenfeld’s derision is a paper released this week by Erwin Diewert, a leading Canadian economist and world-renowned productivity expert, who concluded that multi-factor productivity has grown an average of 1.03 per cent per year since 1961, or much faster than the official Statscan figure of 0.28 per cent per year.
“The new measures may not, in fact, be more accurate,” Mr. Shenfeld said. “The authors make a case that there are methodological issues with the prior estimates, but they have to use less aggregated data than Statistics Canada is able to tap into. In sum, it’s not clear who’s right.”
Mr. Shenfeld argued that multi-factor productivity, or MFP, is a less useful measure than the more familiar labour productivity, or output per hour worked. “Faster growth in output per hour is key to being able to pay our workforce more, particularly if commodity prices don’t soar to cover for the lack of production increases,” he said.
MFP is considered a proxy for a country’s innovation performance, encompassing technological change and other efficiencies. Many economists swear by it because it’s much more comprehensive than labour productivity, incorporating all the factors that contribute to growth – capital, energy, materials, services as well as labour.
The problem is that the numbers are finicky and hard to estimate accurately.
Mr. Diewert, who co-wrote the paper with Emily Yu, an economist at the Department of Foreign Affairs and International Trade, insists he is much closer to the truth. The Statscan data are flawed due to “erratic” rates of return on assets among Canadian businesses, he said.
“Our estimates are certainly not the best but they are much more accurate than the Statistics Canada estimates,” Mr. Diewert said.
Other experts who’ve looked at Mr. Diewert and Ms. Yu’s work say the methodology is also much closer to what the U.S. does, making the numbers more suitable for international comparisons.