Charting Matters: Jobs Edition

Earlier this month, New York Senator Kirsten Gillibrand voted “no” on the budget/debt limit deal and said:

The fact is, there is nothing in this deal that will address the significant jobs crisis we are facing…. today we could have gone further in reducing America’s debt with a sensible compromise that both cut discretionary spending and raised revenues.

This was a gutsy decision, and Gillibrand has taken a lot flak from the usual right-wing bloviators. Only 6 Democrats voted against the deal (plus Vermont’s Bernie Sanders, an Independent). The rest of the 26 “no” votes were by Republicans who wanted more budget cuts which, by any credible logic, would cause more unemployment and lost jobs. Gillibrand is right about the jobs crisis outweighing debt concerns. Jobs are the immediate problem, while debt can and should be handled long-term.

What follows is a review of the historical record for job losses and recovery. At the same time, I want to explore how we look—literally look—at the graphs and charts we use to represent the numbers. Charting is a vital part of economic analysis, and it is important to understand what a chart tells us and what it doesn’t tell us.

Here is a picture of the current job-loss situation as of July, along with job losses during earlier recessions (Figure 1):

Figure 1

Anyone who follows Bill McBride’s Calculated Risk blog will know the model for this chart. When I began studying the economy in late 2008, early 2009, I was lucky in discovering Calculated Risk, and I learned to make charts by working to reproduce McBride’s excellent examples. But with some differences. Even more than McBride does, I prefer a spare, uncluttered view of the data. So, the plotting surface is almost always plain white, with no gridlines, and for this chart I plot fewer recessions in order to make comparisons easier and clearer—also to focus on what I see as significant comparisons. A good chart provokes and guides intuition. Experience with good charts provokes and guides further questioning and study. So, I want to focus attention as much as possible on the patterns and relationships that a chart reveals. Also, since charts like this one plot time series, running from left to right, I’ve chosen to place the vertical “y” axis on the right-hand edge of the chart, because that way it’s easier to gauge where things are “now” or at the end of the series, which is where current interest usually is.

Rather than a picture being “worth a thousand words,” I’ve found that it takes a thousand words—sometimes a good deal more—to convey what a picture is “saying.” This chart is no exception, though I hope to be more frugal here.

The first thing one sees in this chart is the large difference between the current “Great Recession” and earlier post-WW2 recessions. In this one, losses reach a maximum 5.69% in February, 2010—more than twice the 1981-82 maximum. The difference in numbers is even more striking: 8.7 million jobs lost this time compared to a maximum 2.8 million in December, 1982.

The second thing I’d notice is the difference in the number of months it takes to recover jobs to the peak level where they were before losses began. (Note that the period of job losses usually begins a few months later than the official start and continues after—often long after—the official end of a recession.) I have emphasized the three longest recessions by making their lines thicker than the others (the red, blue, and green lines). They just happen to be the three most recent recessions, and one wants to know why they are significantly longer than earlier recessions. I call them Post-Reagan recessions, and that points to a likely explanation. (Paul Krugman calls them “post-modern” recessions, here and here.) A quick look at all recessions since WW2 highlights the difference (Figure 2):

Figure 2

Before 1990 the average time to full job recovery is 19.6 months, or just over a year and a half. For the three most recent recessions the average is 40.7 months, on the way to becoming three and half years. The longest recovery (2001) took a full four years, and recovery this time could take as long as six years or more, given how little progress we have made so far.

Bill McBride also publishes another version of the job-loss chart (scroll down to 2nd image). This one aligns recessions together at the point of deepest losses. Here’s my version of it (Figure 3):

Figure 3

The advantage of this chart is that it separates and highlights the two stages, loss and recovery, to the left and right of the zero line. Pre-Reagan recessions still register the shortest time to recovery. But the 2000 recession now appears to take less time than the 1990-91 recession. For the recovery stage that’s true, but not for the loss stage. Losses tied to the 2000 recession accumulate for 30 months, or two and a half years. This is the longest time on record and a sign of how flawed and misdirected George W. Bush’s fiscal policy was—specifically, the Bush tax cuts, which were tilted toward the rich rather than toward moderate- and lower-income taxpayers, and are still a drag on the economy after being extended past their original expiration date in December, 2010.

How is the current recovery faring, really? One answer is to compare the average slope of each of the chart’s job recovery lines for the three Post-Reagan recessions. A steeper line indicates a faster recovery, as is easy to see in comparing the series for 1990-91 and 2000. Clearly, from months 0 through 9 or 10, the 2000 series is a lot more vigorous than 1990-91. It is harder to make a comparison with the current recession, because its losses are so much deeper than the others, putting it some distance away from them on the chart. Still, we can easily make an effective visual comparison by normalizing the start of each recovery on the same number, as in this next chart (Figure 4):

Figure 4

The three series are normalized at -100 (negative because it represents losses), and now the recovery of job losses tied to the current recession starts from the same place as the earlier ones. The news is not good. Earlier, commenting on Figure 3, I said the current job recovery could take six years or more from its start in early 2008, and the charts confirm this. Things can change, although maybe not in today’s hostile political environment.

So, again, Kirsten Gillibrand is right about jobs trumping debt as our major concern. But why is debt the object of such frantic outcries when interest rates remain at historical lows? Interest rates are low because the markets are less worried about U.S. debt than about our political failure to do anything sensible about the weak economy. Turmoil in the stock markets has prompted a rush to safety. Where? To the U.S. Treasury market, where prices are up sharply since the end of July, and interest rates down. Last Friday (August 12), the Treasury could have borrowed a billion dollars for ten years at 2.24%. But, then, the real 10-year rate adjusted for inflation was -0.02%. That’s negative two one-hundredths of a percent: it’s like getting paid to borrow money. So why can’t we borrow more to get the economy moving now, and plan for debt reductions later when the economy is in a healthy expansion and tax revenues are back to normal?

The obvious reason is that Republicans and right-wing Tea Party zealots are jealously guarding a Reaganomic fantasy which they are heavily invested in—the fantasy that regulation is wrong and free markets self-correcting, that a supply-side tax code funneling wealth to the already wealthy is what creates jobs (and not asset bubbles?), that “big government” is another name for (gasp!) socialism. Republicans adamantly refuse to acknowledge any responsibility for the mess we’re in. But the difference between Pre-Reagan recessions and the current Post-Reagan recession is that a Republican administration followed the Reagan script, and an over-leveraged, under-regulated financial sector, driving an out-of-control housing bubble, crashed the world economy. But, to be fair, this is not just a Republican problem. Enough Democrats embraced deregulation to produce big majorities for measures that did away with Glass-Steagal banking regulations in 1999 and prohibited the regulation of over-the-counter derivatives in 2000. And it was a Democratic President, Bill Clinton, who signed them into law.

Sad to say, it is going to take us a long time to recover, socially and politically, as well as economically.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s