We reported last week how Trimtabs CEO Charles Biderman was concerned investors were misreading initial US non-farm payroll data on account of not fully understanding the methodology that goes into compiling the estimates.
Joining “the payrolls are being misunderstood” brigade on Friday was also Gluskin Sheff’s chief economist David Rosenberg, who noted the much rejoiced-over reduction in the unemployment rate was actually due to a sliding labour force.
Others on Monday were also catching on, among them Stephen Lewis of Monument Securities who noted:
The financial markets last week greeted July’s US employment data as if they marked the end of recession. This was despite the fact that non-farm payrolls were reported to have fallen by 247K on the month. Admittedly, the unemployment rate subsided marginally, to 9.4% from 9.5% in June. But this was not a consequence of a rise in the numbers employed but of shrinkage in the workforce.
So it appears, according to Lewis, that many jobless people are giving up hope on the chances of gaining employment completely. As he notes, this is not a positive for the US economy. Causing yet more confusion, meanwhile, are particularly tricky seasonal adjustments to the July data. Lewis explains:
It is surprising, indeed, that the markets reacted so strongly to the employment report when they were aware, or should have been, from extensive analysis of weekly jobless claims figures that there are serious problems this year with the seasonal adjustments applying to July’s labour data. The annual lay-offs in the auto industry, which usually occur early in July, were not as substantial this year as they usually are, because large numbers of workers in this sector had already been laid off. The payrolls numbers are drawn from the Dept of Labor’s Establishment Survey, which relates to pay periods including the twelfth day of the month.
It is worth noting that this timing would have meant that payrolls were counted at the point where the claims data had been suggesting the distortion from inappropriate seasonal adjustment for auto industry lay-offs was at its height. Since then, the claims series and survey data have pointed to an apparent deterioration in the labour market. It is not entirely clear whether this represents a genuine worsening in labour market conditions or merely the unwinding of previous faulty seasonal adjustments. On either explanation, however, mid-July employment readings are likely to have been giving an unduly favourable view of underlying labour demand. Since the seasonal adjustment applied to July data is the second most extensive of the year (January’s being the largest), it would, in any case, seem unwise to base a view on the economy on these data alone.
Meanwhile, even if you consider the numbers at face value — Lewis still sees no reason for optimism based on the fact that an end to job losses implies an end to cost-cutting, which only puts increased pressure on companies to deliver satisfactory performances from regular operations. As he explains:
In all, non-farm payrolls have already fallen some 4.8% from their peak in December 2007, far exceeding the rate of contraction in any downturn since 1948-49. Private payrolls have fallen even more sharply over this period, by 5.9%. More than 60% of these job losses in the private sector occurred during a six-month stretch within the overall nineteen months of the downturn to date. This six-month period ran from November 2008 to April 2009. This was when US companies resorted to the famous cost-cutting that so much flattered their earnings reports for the first and second quarters of this calendar year.
On average, private payrolls were 1.8% lower in 2009Q1 than they had been in 2008Q4, and were then a further 1.4% down in 2009Q2. Private payrolls in July were only 0.6% below their Q2 level, however. This raises the intriguing possibility that one reason why the payroll declines have slowed is that companies have completed the easy adjustments they could make to the sharp worsening in economic conditions that occurred last autumn and are now having to think harder how to keep the cost cuts coming. A corollary of this might be that US companies’ earnings for 2009Q3 will benefit less from cost-cutting than in the two preceding quarters. US companies may have to achieve an improvement in turnover if they are to carry on delivering earnings surprises that please the markets. It may well take time for them to target and implement further cost cuts.
Related links:
The problem with non-farm payroll numbers – FT Alphaville
Payrolls not so bullish after all, Rosenberg says - FT Alphaville