(By
Robert Johnson, CFA) While
investors focused on April's employment report and cried, I was looking
at a lot of other data that points to better days ahead. Auto sales hit
their second-best month of the entire recovery, the PMI that everyone
had been fretting over hit its highest level in 12 months, weekly
initial unemployment claims posted a meaningful drop, and weekly retail
sales growth is now approaching the 4% mark again.
In fairness, to the untrained eye, a fall in employment growth looks
dismal at just 115,000 jobs added in April versus a recovery high of
275,000 in January. Economists reached for their handy rulers (their
favorite piece of fancy equipment) and reasoned that if job growth
decelerated by 160,000 in three months and the current job growth rate
was 115,000, it was only a matter of another two or three months before
the remaining 115,000 would be lost. Weather, the auto industry, and
seasonal factors are playing havoc with this data set, especially the
month-to-month data. The report looks a lot less confusing if the data
is examined on a year-over-year basis. Unfortunately, that same analysis
suggests that January's 275,000 jobs were probably a bit of a mirage
and the economy is more likely to settle into a pattern of job growth of
150,000-220,000 for the rest of the year. Far from Panic City, but no
boom, either.
Employment a Lot Steadier, but Slower Than Many Believe
I continue to believe that employment remains on a slow but
steady path up despite the monthly up-down, roller-coaster ride
delivered by the month-over-month employment data. Monthly data suggests
that the economy has added 277,000 private-sector jobs in January, only
to collapse to the 130,000 jobs that were added in April. Sounds like
devastation, with job destruction likely within the next three months,
right? That just doesn't square with the other data that I see. For the
very same month, auto sales are near record levels, housing data
continues to improve, and retail sales are moving ahead at a slow but
steady pace. The data starts to make more sense if we look at the
figures on a year-over-year basis, and strip out seasonality.
Seasonality continues to fluctuate, and tends to remove month-to-month
anomalies such as strikes or weather-related booms. As the chart below
shows, growth has remained rock-steady for most of the past year when
employment growth is compared on a year-over-year basis.
The miniscule blip up we saw in January and February has now reverted
to its mean of 1.4%-1.5%. That's not great growth (it equates to about
160,000 jobs a month) but it is slowly denting the rate of unemployment
(the rate actually fell from 8.2% to 8.1%). We aren't booming and we
aren't falling apart. In light of my GDP forecast of 2.0%-2.5%,
employment growth is likely to remain stuck in the 1.5%-2.0% range
(typically employment growth trails GDP growth by 0.5%-1.0% due to
productivity gains). By the way, the 275,000 growth in non-farm payroll
for January equates to 2.5% annual employment growth, a rate that
clearly wasn't sustainable without a lot higher GDP growth.
Job growth by sector has been anything but uniform. Growth compared
with past recessions demonstrates a lot of stark contrasts, too. The
table below shows total employment growth over the last year and then
compares it with the average of the compound annual growth rates in
employment during the three most recent recoveries (from 1982, 1990, and
2000).
Durable Goods and Professional Services Drive Recovery
The data shows this is a far from typical recovery. Interestingly, it
is not a subpar recovery for all. The manufacturing and extraction
sectors are performing way above average. Meanwhile government is in
sharp decline, an unusual turn of events. Also, though the construction
industry is small, it is also performing way below trend.