When should we expect jobless claims to stop falling? At some point,
the trend will end for this cycle. That's not necessarily the end for
growth, at least not in the short run. When initial jobless claims
stopped trending lower in 2005-2006, for instance, the economy didn't
peak until December 2007. The lag time isn't unusual. Jobless claims
tend to stop falling well before the onset of recessions. Sometimes
claims bounce along a bottom for months, sometimes years, before the
economy tanks. Weekly claims issue a warning sign only when the numbers
are routinely rising, particularly on a year-over-year basis. For now,
at least, there's no sign of that dark state of affairs. But the
possibility that claims have hit a floor, while still debatable, is
certainly open for discussion these days.
[Related -Thoughts on MetLife and AIG]
New filings for unemployment benefits increased 20,000 last week
to 362,000 on a seasonally adjusted basis. Looking at recent history
shows that the ongoing decline in claims is slowing and perhaps
bottoming. As you can see in the chart below, this leading indicators
seems to be stuck in neutral lately. It could all be noise, of course.
It was only a few weeks ago that new claims dropped to a post-recession
low of 330,000. It's unclear if we'll reach that level again for this
cycle. Until (or if) we learn otherwise, recent history suggests that
we're stuck in the mid-300,000 range.
[Related -A 2016 Recession Would Be Different]
If the trend has run it's course for this cycle, it may or may not be
a warning sign for the immediate future. Much depends on whether claims
simply remain trapped in a range vs. a full reversal of fortunes, i.e.,
weekly filings start trending higher. For now, the worst you can say
about claims is that they've stopped trending lower, and even that's
still premature. Still, looking at unadjusted weekly data on a
year-over-year basis suggests this possibility.
As the second chart reveals, last week's filings were virtually
unchanged from the year-ago level before seasonal adjustments. One
weekly report doesn't mean much, of course, but it's clear that claims
are no longer falling consistently by 5% to 10% on an annual basis.
For the last several months, we've seen raw claims data post
year-over-year changes closer to zero and sometimes well above zero.
Some of that's noise. But it seems as though the downward trend of late
has weakened. Is that a sign that jobless claims have reached the lows
for this cycle? The jury's still out.
At some point, the decline in jobless claims will cease. Reaching
that point isn't necessarily a warning sign for the economy. History
suggests that jobless claims can move sideways for as long as several
years while growth rolls on. It'll be time to worry, however, when
claims are rising on a year-over-year basis on a persistent basis. So
far, the numbers don't look alarming on that front.
The transition to treading water, on the other hand, looks more
plausible. But even this possibility will take many more weeks to sort