Unemployment: the Good, the Bad, and the Ugly

I’ve been working on this post for some time, trying to identify a good way to introduce the subject. The graphic above, kindly furnished by The Big Picture, is as good a place to start as any. The unemployment situation is different in this recession than in other post-war economic downturns. Jobs are simply not returning at the rate that would have been expected. That unemployment remains a significant story is borne out by President Obama’s Thursday remarks on the subject:

President Obama today called the job numbers released this morning “positive news” but said there still is work to be done on the economy. The numbers showed a drop in the nation’s unemployment rate to 9.4 percent in December.

“Overall, the decline in the unemployment rate is positive news, but it only underscores the importance of us not letting up on our efforts,” President Obama said from the floor of Thompson Creek Manufacturing factory in Landover, Maryland.

“We know these numbers can bounce around from month to month, but the trend is clear,” Obama added, noting that this marks 12 straight months of private-sector job growth for the first time since 2006. “The economy added 1.3 million jobs last year, and each quarter was stronger than the previous quarter, which means that the pace of hiring is beginning to pick up.”

or the NYT op-ed from Michael Powell and Sewell Chan:

The year 2010 ended on a disappointing note, as the economy added just 103,000 jobs in December, suggesting that economic deliverance will not arrive with a great pop in employment.

Signs still point to a long slog of a recovery, with the unemployment rate likely to remain above 8 percent — it sits at 9.4 percent after Friday’s report — at least through the rest of the president’s four-year term.

That the president, Congressmen, newspaper editors, and bloggers like me pore over each of these reports like the entrails of an ox underscores the ongoing concern. President Obama’s re-election prospects are likely to rise and fall on the employment picture. As should be self-evident the economy as a whole and unemployment are in a deadly embrace. There’s no way we’ll see a robust recovery without a reduction in unemployment and there’s no way we’ll see a reduction in unemployment without a robust economic recovery.

In this post I’m going to reflect on a number of unemployment scenarios we might see.

There is little doubt that something is different. Over what is different and why it is different there is significant disagreement. The Keynesian or Neo-Keynesian view is that the present unemployment is cyclical in nature, there is presently inadequate aggregate demand, employment won’t grow to its former levels until aggregate demand returns to its former levels, and that the federal government as the consumer of last resort and unburdened (as state and local government in general are) by the constraints that balancing its budget might bring should be spending at a level high enough to boost aggregate demand to its former heights.

Paul Krugman is the dean of those who see the situation in this light. I won’t bother citing a specific column—that’s the message of practically every column he’s written over the period of the last eighteen months. It may even be a majority position among economists.

However, it is not the only view. It’s hard not to see the loss of jobs in the residential construction business including some of the associated jobs, e.g. realtors, plumbers, electricians, and so on, as not being structural in nature. With the enormous inventory of unsold new and existing houses (not to mention the significant number of foreclosures we should be expecting) is it likely that we will we see that many jobs in residential construction for the foreseeable future?

Additionally, propelled by the housing bubble and the attendant use of home equity for ordinary retail spending, the retail sector surely must have been larger as a consequence of that bubble than it otherwise would have been. Under the circumstances it seems to to me that the return of many of those retail sector jobs in the foreseeable future is pretty unlikely.

I think there’s another way we should be thinking about structural unemployment. More about that later.

I mentioned the good, the bad, and the ugly in the title of this post and it’s about time we get around to it. First, consider this graph.

The graphic above is from David Leonhardt’s year-end review for 2010 in the New York Times and it illustrates how long it would take to recover the jobs lost during the recession with three different rates of job increse: 300,000 additional jobs per month, 250,000 additional jobs per month (the pace during the mid-1990s, the period of the dot-com bubble), and 200,000 additional jobs per month.

Calculated Risk, among others, envisions job growth over the next several years at the rate of 250,000 additional jobs per month (the number of jobs added last month was about 100,000). That would appear to me to be the good, the optimistic scenario. If jobs grow at that pace and taking into account the roughly 125,000 new workers that come into the market each month, we would put those who are unemployed back to work some time in 2017, a period of about 72 months.

According to the NBER, the official scorekeeper for such things, the median post-war expansion was 45 months. Since the recession has been determined to have ended in June, 2009, the present expansion has been going on for 19 months. The last three expansions, the expansion of the 1980s, the expansion of the late 1990s, the the expansion of the Aughts, were 92 weeks, 120 weeks, and 73 weeks in length, respectively, and the last two of those at least were fueled by bubbles. During the last expansion we saw job growth at roughly 150,000 per month.

How long would it take to recover the jobs lost at that rate of increase? We’d never recover them. The next downturn would surely take place before the jobs had returned.

Let me repeat this: if you believe we are going to create 250,000 jobs per month for long enough to put those who are unemployed as of the start of the recession back to work you believe that we are going to experience the equivalent of another bubble economy and it will persist for at least five years. I don’t find that credible.

That brings us to the bad: the prospect that jobs are created at too low a rate to accomplish a return to full employment and we maintain chronically high unemployment. This is the situation that much of Europe has experienced for decades. The European solution to the problem has been to put much higher government support in place for those who are unemployed than has typically been the case in the U. S. This would certainly be bad, at least if you value the United States as it has been.

However, the alternative, chronically high unemployment without a system of government support for those who are unemployed because there are no jobs to be had, could be pretty ugly.

12 comments… add one
  • Icepick Link

    If jobs grow at that pace [250,000 a month] and taking into account the roughly 125,000 new workers that come into the market each month, we would put those who are unemployed back to work some time in 2017, a period of about 72 months.

    I crunched a few numbers along that line using what has happened in Florida over the last month for which data is available (November) and for the previous year. The results weren’t encouraging.

    Llet’s look at the broader picture of the whole state of Florida. From an article in the Orlando Sentinel last Thursday:

    The statewide figure [of 12.0% U-3 unemployment] represents about 1.1 million jobless in a labor force of about 9.2 million. Total non-agricultural employment grew by 300 jobs from the previous month.

    Three hundred jobs! Let’s say we want the unemployment rate in Florida to get down to 5%, which is actually higher than it was pre-recession. That means that 640,000 currently unemployed Floridians (on net) need to find a job. At 300 jobs a month it will take approximately 2,133 months for the U-3 rate to return to normal. (That’s over 177 years.)

    Now you can accuse me of looking at the worst case. You would be wrong (we could easily lose jobs), but I can find a better case from the same article.

    Since last year at this time, Florida has added 36,200 jobs – an annual growth rate of 0.5 percent. The national growth rate over that time has been 0.6 percent.

    Okay, so the monthly average job gain of the last year has actually been a little over 3,000. So that means that it would take only about 213 months, or over 17.5 years, for the employment numbers to improve. And none of that takes increases in population into account.

    I’m picking UGLY as the end scenario even if the government starts adding European style social safety nets. It’s not like we can afford them anyway. Nor can the Europeans, for that matter.

    I’m just trying to figure out how to buy the dip for this situation.

  • With the enormous inventory of unsold new and existing houses (not to mention the significant number of foreclosures we should be expecting) is it likely that we will we see that many jobs in residential construction for the foreseeable future?

    No, jobs in that sector will be a long time in coming back. It is also foolish to try and stimulate that sector since all you risk doing is restarting yet another bubble which will not only prolong the pain, but make it worse. The residential housing market needs to clear that excess supply. The one sure fire way is to let housing prices decline so that the market clears. But housing prices are not flexible downwards. Also, it is not a pleasant thought, but I don’t see any other solution that would help quickly. So in the end we are probably stuck with a much longer recovery this time because the bubble was in the worst sector of the economy to have a bubble…real estate/housing.

    It is this kind of view that makes the Krugman et. al. position untenable. If the above is true, that we wont see these jobs return quickly, it is a structural problem. We’ve over invested in housing and now we have this huge amount of capital we need to draw down on over time. Even if there was another sector of the economy that could pick up the slack we wouldn’t see these jobs coming back. To cling to the notion of cyclical unemployment is to think that if the government spends enough money long enough we can get back to where we were before the bubble burst…but bubble economies are not sustainable and hence Krugman and those who agree with him and their advice should be looked at with considerable skepticism.

    And you might want to look at this paper by Edward Leamer. I think you’ll like it.

  • Certainly chatty, isn’t it? I’ve scanned through twenty or thirty pages. His complaint that macroeconomics relies too heavily on theory and not enough on data warmed the cockles of my heart.

    It’ll take me a while to plow through it. One thing I noticed in his “contributions to GDP by sector section” is that I strongly suspect that we’re overstating the contributions of the healthcare sector, three-fifths of which is funded by tax revenues and government borrowing. I think that were we to re-calculate to include only the portion that isn’t funded by the government it would produce a picture of the economy that comported with experiential reality a little more closely.

  • There was also this from the UCLA Anderson Forecast

    In a report titled “A Homeless Recovery,” UCLA Anderson Forecast director Edward Leamer notes that while expansive, free-spending consumers fueled past economic recoveries, today’s “frugal consumers” cannot be counted on to power a strong recovery in the foreseeable future.

    “If the next year is going to bring exceptional growth,” Leamer writes, “consumers will need to express their optimism in the way that really counts — buying homes and cars. And that is not going to happen if businesses continue to express their pessimism in the way that really counts — by not hiring workers.”

    The result is an economic Catch-22.


  • Leamer is an economatrician not a macro/theory guy (well except for econometric theory, but that deals with things like estimators and what not, where statistics meets data), so yeah it will put a different spin on things.

  • It is this kind of view that makes the Krugman et. al. position untenable. If the above is true, that we wont see these jobs return quickly, it is a structural problem.

    But Steve, Krugman has an answer to that – a 1930’s style WPA program!

  • steve Link

    Read Mandel. Advertising for new jobs has picked up quite a bit.

    The structural vs cyclical argument gets bogged down by how you look at people like construction workers. When the inventory is worked down, will there be jobs for them again? Yes, but if that takes many years, at some point it becomes structural. I think that Krugman also misses the importance of what recessions do in a capitalist economy. It lets companies sort out which jobs were not really needed. A lot of old jobs will not come back.


  • steve:

    Read the JOLTS report. Absent a bubble we haven’t created openings at a level much above layoffs and terminations for a generation and to bring the unemployment rate down to anything that resembles full employment we’re going to need to create a lot more new jobs than layoffs and terminations and we’re going to need to do it for the better part of a decade.

    I could believe that we might create a lot of jobs (for a short period of time) or that we might have a long expansion. Given the history I find it hard to believe we’ll do both at the same time.

  • steve Link

    I hope you realize that I largely agree with you on this being a very slow recovery. However, I do find Mandel’s findings interesting. Konczal’s piece yesterday was also interesting claiming that unemployment hit every sector at about the same rate, but some started with much lower rates of unemployment.


  • I don’t think that unemployment being spread throughout the economy signifies that it’s cyclical as Dr. Krugman would have it. IMO that can only be maintained if you think that history began in 1995.

    In my view it merely means that successive bubbles have resulted in more growth throughout the economy than would have been seen otherwise. Our problems have been a long time coming and IMO will be here for a long time as well.

  • Drew Link

    I read just this morning another dour report on housing. Wow, what an inventory overhang. And as Steve V observes, with sticky downward price acceptance (but inevitable capitulation) that’s more latent wealth destruction in the pipeline.

    Perhaps the only positive indicator I’ve seen recently was the ADP correlation, cited by Dave here, and elsewhere. Unfortunately, there is another long term correlation that is looking very bad right now: that between consumer confidence and retail sales. Implicitly it makes Nov – Dec appear to be a holiday blip.

    I’m sure Dave will comment on it at one point, but the state of IL is going in the exact wrong direction. Last night the state income tax rate was raised by 2 points to 5%, along with a dramatic increase in business taxes. Its supposed to come with a 2% spending growth cap, but if you believe that I’ve got a bridge I’d like to sell you. This will be chilling, and the Dems will be left wondering – when the tax revenues come in short, and the jobs start fleeing – “what happened? My question is “when will the voters wake up?” This is a self inflicted wound.

  • Drew Link

    Oh, and I think steve makes an important and often overlooked observation: recessions naturally cause businesses to re-evalute the necessity of certain positions. Some will be deemed obsolete.

Leave a Comment