The annual rate of job growth is in decline. Growth rates in GDP and in hours worked suggest the decline in the jobs growth rate will accelerate. The jobs report Friday May 3, could show a substantial number of jobs lost in April.
Annual job growth started getting ahead of annual GDP growth about two years ago and started back toward alignment after peaking in December 2012.
The annual growth rate can be a useful tool for smoothing out volatile data and giving a clearer picture of the trend. To get the annual rate of job growth you average the number of total jobs for 12 months, then calculate the growth rate from the average of the 12 months prior to that. For example, the average number of non farm jobs for the year 2012 was 133.7 million, which was 1.70% higher than the average for 2011. In March the annual rate had fallen to 1.62%.
The annual growth rate gives a clearer perspective of the trend but is slow to show a change in trend. By comparison the year-over-year growth rate in jobs peaked in February 2012, and has had a haphazard declining trend since.
The chart above shows the annual job growth rate in black and its correlation to annual GDP growth in red. Annual GDP has its best correlation with a five-month lead time. The 2.05% annual GDP growth rate for the first quarter of 2013 suggests the job growth rate should be down at 0.94% in August. Such a sharp decline in the growth rate would imply millions of jobs lost. However, such losses are unlikely unless a sharp recession has begun or begins soon. If the Economic growth rate stays near the 2% range it would probably take a year or two of slow or stagnant job growth to bring job growth back in line with GDP growth.
In some expansions the rate of job growth came to a peak, started down and then resumed going upward making another peak before declining into a recession. In those cases the GDP annual growth rate led the job rate higher. Right now GDP is leading the job growth rate lower at least till August.
If the growth rate declines in April as much as it did in March, we lose 84,000 jobs in April. This is shown as the blue dot on the chart below. The chart below uses the same data and correlation as above, but zooms in on about the last two years of data.
The index of number of hours worked has an even stronger relationship with jobs. Its strongest correlation is with a one-month lead time. Beginning in about 2005 the growth rate in hours compared with jobs sped up from its longer-term relationship. So the chart below fits the relationship only to the period of 2005 to the present. Job growth is in black and growth in hours worked is plotted in red. The growth rate in hours peaked in June 2012 and has declined the last nine months.
To get a clearer picture of the relationship the next two charts will look at the monthly change, or first differences, in the annual growth rates of jobs and hours. In the chart below if the line is above zero the growth rate is rising. If the line is trending down but still above zero it is rising at a slower rate. When the line crosses below zero it means the growth rate is in decline, but it does not give a clear signal of when the number of jobs or hours actually starts a decline.
For the last 12 months the first differences for hours have been continuously below that for jobs. The implication is that the decline in the growth rate of jobs should accelerate. To get a clearer picture the chart below zooms in to the last 15 data points and offers estimates for the April data point.
The point on the black line in the chart representing December 2012, sits just above zero showing the job growth rate was still rising, but just barely. The point for January was below zero showing that the growth rate turned down. The last two black points on the line show that the growth rate accelerated to the downside in both February and March.
The last point on the red line representing hours has a green dot around it and suggests the annual growth rate for jobs will decelerate an additional 0.046 to 1.572% in April. If this were to happen April would lose 260 thousand jobs. On the other hand, if the Briefing.com forecast of 135 thousand new jobs, symbolized by the orange dot, occurs the annual growth rate would only decline 0.021 to 1.597%. If the annual growth rate declines in April at the same pace as March, as shown by the blue dot, there would be 84 thousand jobs lost.
Growth in the labor market has been weakening for some time as confirmed by the declining annual rate of job growth. In fact, every time the annual rate has peaked and declined to the current rate the economy was either already in a recession or one began within a few months. The chart below shows annual job growth for the last 63 years. The red dots show the growth rate where sustained job losses began during recessions.
Annual job growth in March has declined to a level that is below where sustained job losses began in eight of the last 10 recessions.
The economy usually stalls into recession when annual GDP growth slows to about 2%. In 2011 growth dipped below this and was able to rebound to a little above stall speed, but has since fallen two consecutive quarters to 2.05%, perilously close to stall speed.
We are at a point in the business cycle where companies are aggressively controlling costs to meet profit targets. "Aggressive cost control" might as well be a euphemism for squeezing workers on hours, wages and jobs. New unemployment claims suggest we are not at the point where companies are letting lots of people go, but weak growth in hours and GDP may mean we are at the point where companies hire very little and collectively let their work force dwindle by attrition.
The monthly job growth data is variable enough that adding 135 thousand jobs in April, as predicted by Briefing.com, is certainly possible. However job losses look more in line with the declining growth rates of the economy and the number of hours worked. The hours data makes losing 260 thousand in April look plausible. I am not ready to make that as a prediction, but job growth has some catching up to do on the downside.
Last month the weaker-than-expected 88 thousand new non farm jobs set the stock market back for the morning, but the S&P 500 (SPY) rallied to a new all-time high four days later. One weak jobs number is fairly easy to dismiss. A second consecutive weak or negative number this Friday should have a larger and longer lasting downward impact on stocks.
Additional disclosure: There is no guarantee analysis of historical data their trends and correlations enable accurate forecasts. The data presented is from sources believed to be reliable, but its accuracy cannot be guaranteed. Past performance does not indicate future results. This is not a recommendation to buy or sell specific securities. This is not an offer to manage money.