• Focus on job market data is set to intensify in the coming months, with job growth being a key factor for the sustainability of the current recovery. A repeat of the jobless recoveries in 1991 and 2002 would be a serious blow to outlook. However, for several reasons we believe this is less likely to happen.
  • First, the current recovery will see sufficiently strong growth during the initial three to four quarters to kick-start the labour market. In fact the very weak recoveries in 1991 and 2002 were the main factor behind the lack of job creation.
  • Second, the economy and in particular businesses saw renewed (in some instances intensifying) headwinds during the 1991 and 2002 recoveries. So far this is not the case. Although businesses remain under pressure, most headwinds have eased rather than intensified.
  • Third, because of the extreme pressure on the corporate sector during this crisis businesses are leaner now compared with exits from the 1991 and 2001 recession. In the two previous recessions labour hoarding and a subsequent rise in productivity explains some of the weakness in job growth. While a further strong rise in productivity this time cannot be ruled out, the moderate investment growth prior to this crisis does not support this.
  • Finally, hours worked have not declined more than usual in this downturn. Hours will go up as activity rises but it is not likely to fully crowd out jobs.
  • In absence of new unexpected headwinds job growth is likely to return by year-end and to reach the 200,000-250,000 range by the middle of next year, with the unemployment rate peaking in Q1 at 10.1%.
  • This would be a positive surprise and a very important signal for further risk-taking in the market. This should also pave the way for the next leg up in bond yields and increase market speculations about Fed tightening.

Strong growth to kick-start the job market

While the initial recovery phase, driven by recoil in the manufacturing and residential construction sectors, has sufficient steam to drive the economy for the coming two to three quarters, it is very obvious that a recovery in final demand – in particular private consumption – will eventually be needed for a sustained recovery to materialise. Hence, job growth is needed to return rather sooner than later. If job growth does not reach close to 200,000 per month by the middle of next year there is an increased risk that the recovery will fizzle out. This is exactly what the markets have begun worrying about.

With the outlook for around 4% growth in the current and coming two quarters, our main expectation is that job growth will return by late this year. Moreover we believe it is likely that pass-through from growth to jobs will be more normal in the current recovery, than following the past two recessions (see the following sections). Hence we forecast job growth to reach about 200,000-250,000 people/month by mid-2010.

Indeed this would be sufficient to take nominal aggregate compensation growth back to around 3% by the middle of next year, which in turn should make room for consumption growth to reach about 2% given relatively subdued inflation and some slight accommodation in the savings rate. This should provide the green light for further risk-taking, add renewed upward pressure on bond yields, and increase speculation about monetary normalisation.

Alternatively, if job growth is not able to keep up with the business cycle, as in the past two recessions, it spells problems. In that case the unemployment rate will continue to climb and job growth will be too weak to secure sufficient nominal income growth for consumption to return to a pace, which is consistent with a sustained recovery. That would be bad news for risk taking and put central banks on hold for longer.

The roadmap below depicts different scenarios for the economy and the job market based on: a) A normal pass-through from GDP to jobs; and b) A “jobless” pass-through as seen during the jobless recoveries in 1991 and 2002 where job annualised growth was about 1-1½% lower than what should be expected given the level of GDP growth. The scenarios are based on a model, which estimates non-farm payrolls on current and lagged GDP growth and the underlying trend in labour productivity.