Skip to main content

A help wanted sign hangs in a bar window along Queen Street West in Toronto on June 10.CARLOS OSORIO/Reuters

Is the Bank of Canada driving the economy toward a painful recession or steering it toward a “soft landing” in which inflation declines without a surge in unemployment? The answer could come down to a debate about job vacancies that has captivated economists over the past month.

In the past, sharp increases in interest rates have typically produced a spike in unemployment. Central bankers are betting that things will be different this time around, arguing that the high number of unfilled positions in today’s ultratight job market will provide a cushion for falling labour demand.

This idea is not without critics. Several prominent economists have attacked central bank optimism in recent weeks, suggesting that forecasts of a soft landing – in which businesses take down help wanted signs but don’t start laying off employees – are based on theoretical errors and a disregard for history.

“In all episodes when the vacancy rate came down in a meaningful way, the unemployment rate increased substantially,” former U.S. treasury secretary Larry Summers and former IMF chief economist Olivier Blanchard argue in a paper published last month that criticizes the U.S. Federal Reserve’s analysis. “There was no free lunch, and there is no reason to expect one today.”

This prompted a response from Federal Reserve board member Christopher Waller, who published his own paper in late July defending the Fed’s position and criticizing Mr. Summers and Mr. Blanchard’s analysis as “non-standard.”

The tit for tat, which has captured the attention of Wall Street and Bay Street, revolves around something called the Beveridge curve. Named after mid-20th-century British economist William Beveridge, the graph illustrates the relationship between unemployment and job vacancies: When unemployment is low, job vacancies are high, and vice versa.

Usually the relationship is relatively stable. But sometimes the entire curve shifts outward on the graph, indicating there are more job openings than before for any given level of unemployment.

This has happened in a dramatic way during the pandemic. Economists are still debating the exact reasons for this change, but it broadly suggests that labour markets have become worse at matching would-be employers and employees over the past 2½ years – something economists call a decline in “matching efficiency.”

The Beveridge curve in Canada shifted

outward during the pandemic

Per cent

6.0

5.5

Oct. 2020 -

May 2022

5.0

Job vacancy rate

4.5

4.0

2015 Q1-

2019 Q4

3.5

3.0

2.5

2.0

5.0

6.0

7.0

8.0

9.0

10.0

Unemployment rate

the globe and mail, Source: td economics;

Statistics Canada

The Beveridge curve in Canada shifted

outward during the pandemic

Per cent

6.0

5.5

Oct. 2020 -

May 2022

5.0

Job vacancy rate

4.5

4.0

2015 Q1-

2019 Q4

3.5

3.0

2.5

2.0

5.0

6.0

7.0

8.0

9.0

10.0

Unemployment rate

the globe and mail, Source: td economics;

Statistics Canada

The Beveridge curve in Canada shifted outward during the pandemic

Per cent

6.0

5.5

Oct. 2020 -

May 2022

5.0

4.5

Job vacancy rate

4.0

3.5

2015 Q1-

2019 Q4

3.0

2.5

2.0

5.0

6.0

7.0

8.0

9.0

10.0

Unemployment rate

the globe and mail, Source: td economics; Statistics Canada

This shows up clearly in the data. Canadian employers were recruiting for more than a million vacant positions in May, roughly 300,000 more job openings than a year before. Moreover, in the first quarter of the year, there were only 33.6 hires per 100 job postings, down from 47.8 a year before and far below the 81.1 in the first quarter of 2016, according to Statistics Canada.

What does this have to do with the probability of a recession and whether central banks are pushing their economies toward a hard or soft landing? It comes down to whether these labour-market changes are temporary or permanent.

Both sides of the debate agree that high inflation is being driven in part by tight labour markets. When demand for workers outstrips supply, positions go unfilled and companies bid up wages to compete for scarce workers, feeding into consumer price inflation, particularly in the service sector. The pace of wage growth in Canada is picking up – average hourly wages increased 5.2 per cent year-over-year in both June and July – although it remains below price inflation.

For central banks to get a handle on inflation, they need to take some steam out of the job market by raising interest rates and lowering demand for workers. But how much pain do they need to cause to accomplish this?

This is where the theoretical debate comes in. If the position of the Beveridge curve has permanently changed because labour markets have become structurally worse at matching workers and employers, then a significant decline in vacancies will imply a significant rise in unemployment. On the other hand, if the Beveridge curve shifts back to its prepandemic position, you could theoretically get a large drop in vacancies without a large rise in unemployment.

“This is the soft landing we are projecting. Interest rate increases can cool demand and inflation without choking off growth or causing a surge in unemployment,” Bank of Canada Governor Tiff Macklem said at a news conference last month, shortly after the central bank announced its largest interest rate hike since 1998.

“Some sectors will be more affected by interest rate increases than others, but the very tight labour market means there is room to reduce the number of job vacancies without having a big impact on overall employment,” he said.

The Bank of Canada declined to make anyone available for an interview.

Skeptics such as Mr. Summers and Mr. Blanchard aren’t convinced by this scenario. Every decline in job vacancy rates since the 1950s has been accompanied by a rise in unemployment, the economists point out in their July paper.

Moreover, they argue that the decline in matching efficiency means the so-called “natural rate of unemployment” is higher than previously thought, at least in the United States. That would imply the labour market is even tighter than people think.

So far the evidence suggests that job vacancies are beginning to decline in Canada and the U.S., while both countries remain at record-low levels of unemployment: 4.9 per cent in Canada and 3.5 per cent in the U.S. in July.

Brendon Bernard, senior economist with Indeed Canada, said Canadian job postings on Indeed’s website peaked in May at 74 per cent above prepandemic levels. They have since come down, although they remain elevated at 63 per cent above prepandemic levels.

“Our data does suggest that things have cooled down a bit, but not to a degree that’s necessarily associated with a major shift in job market conditions,” Mr. Bernard said in an interview.

Canadian Imperial Bank of Commerce chief economist Avery Shenfeld said he comes down on the side of the central bankers in the Beveridge curve debate. He said the deterioration of matching efficiency during the pandemic is likely the result of waves of industry lockdowns and reopenings – dynamics that should fade over time.

“Part of the reason vacancies are so elevated, or job matching has been so difficult, is that we’ve switched people’s industries so radically over the past two years back and forth. So a snapshot taken at any point in time is going to find jobs going vacant in positions that don’t really have huge skill requirements,” Mr. Shenfeld said in an interview.

“What’s going to help that is some of the other booming sectors are slowing, so they’re going to create a pool of labour. And these aren’t such high-skilled jobs that it’s going to be tough to reposition those people.”

Mikal Skuterud, an economics professor at the University of Waterloo, echoed this view, arguing that it’s easy to overestimate how permanent labour-market disruptions will be. At the same time, he cautioned that the situation in Canada and the U.S. is not identical, so the Fed’s analysis may not hold for the Bank of Canada.

“They’ve got twice as many job vacancies per unemployed worker, per job seeker, as we do,” Prof. Skuterud said in an interview.

“That’s actually probably not a good thing.” If there’s less of a job vacancy cushion in Canada, “then the trade-off in unemployment might be less forgiving, so we might have a harder landing.”

Whether or not Canada is heading toward a recession will depend on a range of factors. Central bankers are eyeing things such as commodity prices, consumer spending and saving patterns, as well as people’s beliefs about future inflation – all of which could influence how high interest rates need to go to get inflation under control. Last month, Mr. Macklem said the path to a soft landing is narrowing because inflation has remained so stubbornly high.

As for the labour market, Mr. Waller of the Federal Reserve acknowledged that history is not on the side of central bankers in the Beveridge curve debate. Nonetheless, he remained optimistic about a soft landing.

“We recognize that it would be unprecedented for vacancies to decline by a large amount without the economy falling into recession,” he wrote in the paper responding to Mr. Summers and Mr. Blanchard. “As a result, we are, in effect, saying that something unprecedented can occur because the labour market is in an unprecedented situation.”