One of the most notable features of the Great Recession
relative to that of all previous recessions is that in this one, once
you lose your job, it has been incredibly difficult to find a new one.
As a result, people have been pounding the pavement looking for work
for periods of time that would have been unheard of in any other
downturn.
There are two basic measures of
unemployment duration provided in the monthly employment report — the
average (mean) duration and the median duration. The average has a
longer history, but since it is impossible to be unemployed for less
than zero weeks, it will always be higher than the median, and is a
number that is much more prone to be skewed by outliers.
However,
as the first graph below shows, regardless of measure, this downturn
has been FAR worse than anything else the nation has experienced since
the Great Depression. Prior to this downturn, the worst the average
duration of unemployment had ever reached was 21.2 weeks back in July
of 1983. The worst median duration of unemployment had ever hit was
12.3 weeks hit two months earlier in May 2003. The long term average
for the mean unemployment duration (since 1947) is 13.7 weeks. The long
term average for the median is 7.4 weeks.
In
January, the signals were mixed. We got a little bit of good news in
that the median duration of unemployment dropped to 19.9 weeks from the
all-time record high set in December. The average duration of
unemployment, however, set yet another new record of 30.2 weeks, up from
29.1 weeks in December.
Let’s take a closer
look. The BLS breaks down the unemployed into four categories based on
how long people have been looking for work. Note that none of the
figures are adjusted for population growth, so you would expect all of
them to have a slight upward tilt over the long term. However, that has
not generally been the case with respect to short term unemployment
(less than 5 weeks).
Yes short-term unemployment
tends to go up a little bit when the economy gets soft, but generally
it has been trending down since the early 1980’s. For the most part,
though, short-term unemployment is pretty stable. There will always be
short-term unemployed, even in the most robust of economic booms.
Generally, short-term unemployment is no big deal, sort of like an
unplanned vacation — unless it turns into long-term unemployment.
Regular
state unemployment benefits don’t fully make up for the lost paycheck,
but usually expenses are lower as well. People might have to dip into
their savings a little bit, or run up their credit cards, but savings
are unlikely to be totally depleted, or cards maxed out.
If
you think you are going back to work soon, you will continue with many
expenses. You don’t quit the country club (and forfeit the initiation
fees) just because you are out of work for a few weeks — hey, you
finally have a chance to play a few rounds! You don’t cancel little
Julie’s ballet or little Johnny’s piano lessons just because you got a
pink slip. However, as time wears on, the difference between the old
paycheck and the unemployment benefits starts to add up. Also, your
contacts start to get stale and your skills deteriorate.
When
you go past 26 weeks, or six months, in normal times regular state
unemployment benefits expire. During economic downturns, the Federal
government will usually step in with extended benefits. Currently the
number of people on extended, Federally paid, unemployment benefits far
exceeds the number on regular state based benefits — 5.855 million to
4.602 million.
If it were not for the extended
benefits, which are a major part of the ARRA, or Stimulus act, those
people would be left with no income whatsoever, their savings would be
completely depleted and their credit cards maxed out. With no income,
they would not be able shop at the Salvation Army, let alone at
Wal-Mart (NYSE:WMT) or J.C. Penney’s (NYSE:JCP). They would have to rely
on the local food bank instead of getting their groceries at Kroger’s
(NYSE:KR).
While people on extended benefits
will not be shopping at Saks (NYSE:SKS), at least they can still go to
Big Lots (NYSE:BIG). Those retailers (and the companies that make and
transport the goods) would then be forced to lay off even more people.
Now,
one can argue that it has cost too much to save these millions of
fellow citizens from Calcutta-style poverty. However, only someone who
for political purposes wants to see as much suffering in the country as
possible could argue that the ARRA has had no positive effects. Those
retail jobs are some of the ones “saved” by the ARRA.
One
can also legitimately worry about extended benefit programs turning
into a back-door welfare program, and fostering a culture of
dependency. Clearly, extended benefits are not a full substitute for
real jobs. However, the alternative of cutting people off completely
after six months would be, quite frankly, cruel in this economic
environment.
What really makes a recession a
recession is when the number of longer-term unemployed starts to grow.
Take a look at the black line, which is the number of people who have
been out of work for more than six months. It is far more volatile than
the shorter-term unemployed groups, rising during recessions (and the
immediate aftermath of recessions) but then plunging as economic
expansions take hold.
Note that in the last
expansion, the number of long-term unemployed fell, but not nearly as
far as it had in previous expansions. At the best point of the last
expansion, the number of long-term unemployed was almost as high as at
the peak following the deep mid-1970’s recession and significantly
higher than at the peak of any downturn prior to that (again the numbers
are not population-adjusted, so one would expect in general to see
higher highs and higher lows over time).
However,
the high starting level sure didn’t stop long-term unemployment from
soaring in the Great Recession. There are now over 6.3 million people
who have been out of work for more than 26 weeks — that is almost a
five-fold increase from when the recession started in December 2007.
The long-term unemployed now make up 41.2% of all the unemployed, up
from 39.8% in December, and just 22.4% a year ago. The long-term
average of long-term unemployed is just 13.3%.
Another
way to look at this data is to examine the ratio between the long-term
(over 26 weeks) and short-term (under 5 week) unemployed. Prior to the
Great Recession, the worst the ratio had ever hit was 0.78, in March
1983. Since 1947, the ratio has averaged 0.34.
The
first time we ever had more long-term unemployed than short-term
unemployed was in April of 2009. In January, the ratio was at 2.10,
just slightly below the record 2.12 set in November. In the absence of
the Stimulus Bill, it is likely that there would be far more
unemployed, both short and long term.
I would
argue that the problem with the Stimulus Bill was that it was too
small, not that it did not work. I would also point out that at the
time the bill was being debated I was arguing that it was insufficient
to do the job. That point of view, even though shared by several Nobel
Laureates in Economics, was almost entirely shut out of the debate at
the time.
The imbalances in the economy that
caused this downturn were roughly on the same order of magnitude with
the imbalances in the late 1920’s prior to the Great Depression. The
financial panic was by many measures just as bad, if not worse, although
much fo the data from the late 1920’s and early 1930’s is spotty.
However,
where there was good data — for example the spread on yields between
AAA corporate bonds and BBB corporate bonds (a good measure of the
markets perception of the probability of widespread bankruptcies) — the
situation was actually worse than the spreads in the early 1930’s. The
big difference is that this time around, we acted quickly, both under
the past administration and even more so under the current one. It was
if action were taken to fight the Great Depression in 1930 rather than
waiting until 1933.
That is the reason we are
only in the Great Recession (and based on GDP coming out of it) rather
than being in the Second Great Depression. While the situation is
generally better than it was a year ago, or at least headed in the
right direction when things were going at warp speed in the wrong
direction back then, the situation today is far from satisfactory. More
action is needed, and the Jobs bill currently under consideration in
the Senate is a good start, but also probably not enough to really get
the job done.
Yes, a new jobs bill will add to
the deficit, but the biggest cause of the increase in the deficit over
the last year has not been increased spending, but lower tax revenues.
To some extent, that is due to tax cuts that were part of the ARRA (tax
cuts which, by the way, went to 95% of families).
However,
the bulk of the decline in tax revenues has been because of the weak
economy. With profits down, so are corporate income tax revenues. People
with no income do not pay individual income taxes. Unemployment
benefits are taxable (so some of what was spent is recouped), but people
getting unemployment benefits will be in a lower tax bracket than they
were when they were employed.
In other words,
the money spent on the Stimulus Bill did not add dollar for dollar to
the deficit relative to what the deficit would have been in its
absence. There is some truth to the supply-sider argument that dynamic
scoring of tax cuts can lead to economic stimulus, but anyone claiming
that tax cuts actually increase tax revenues has not looked at reality.
With
a top marginal rate of 35% right now, we are firmly on the left hand
side of the Laffer curve. The Kennedy tax cuts were from top marginal
rates of over 90%, so in that case we were probably very far on the
right side of the curve. Of course, if you are going to generate the
same amount of revenue with two different tax rates, the lower one is
going to be preferable, and the higher one just stupidly punitive.
