Carpe Diem

The US economy is now producing 2.2% more output than before the recession, but with 3.84 million fewer workers

The chart above is an updated version of one that I’ve featured several times before on CD, and helps to graphically capture several important trends in the U.S. economy over the last ten years: real GDP and civilian employment. It generated more than 100 comments when I posted it about a year ago, so I thought it was time for an update.  Here’s what the chart shows:

1. Measured by real output (GDP), the U.S. economy has made a complete recovery from the 2007-2009 recession now that real output in Q3 of this year at $13.6 trillion (2005 dollars) was 2.2% (and $290 billion) higher than the $13.32 trillion of real GDP in Q4 2007 when the recession started (blue line in chart).

2. While real output has completely recovered to 2.2% above its pre-recession levels, the current U.S. employment level of 142.4 million jobs in Q3 is still 3.84 million jobs (and 2.62%) below the 2007 peak of 146.27 million jobs (red line in chart), and that translates into the current “jobless recovery.”

3. The recovery of real output to an historically high level that is 2.2% above pre-recession levels  with 2.6% fewer employees has also translated into record-level after-tax corporate profits, which are now 30% above pre-recession levels.

4. The recovery of both output and profits to above 2007 levels with 3.84 million fewer workers could explain the sluggish job growth that will probably continue for several more years.  If companies can produce more output now than in 2007 with fewer workers and earn record profits, where’s the incentive to hire more workers?

The Great Recession stimulated huge productivity and efficiency gains as companies shed marginal workers and learned how to do “more with less (fewer workers).”  Technological advances may have also contributed to the productivity gains. The surge in productivity since the recession started has been significant (see chart below of real GDP per worker) and may have long-lasting effects, e.g. an extended period of time with a jobless rate above 7%. Even with real GDP, real GDP per worker, and corporate profits at all-time historical highs, we can expect sluggish job growth and the “jobless recovery” to continue for the foreseeable future.  But unless we start to see sharp declines in real GDP and/or corporate profits, it’s likely that the sub-par economic expansion will continue.

41 thoughts on “The US economy is now producing 2.2% more output than before the recession, but with 3.84 million fewer workers

  1. We need to recall that there are now two classes of workers those based on the carbon cycle, and those based on silicon that just use electricity. Since the silicon workers are cheaper they will replace humans in all repetitive tasks over time. With the research about driverless tractors, so that a farmer could control a fleet of tractors from a hub, how long till we need to do bread and video games (as an update from bread and circuses). Of course this is the society seen by Arthur C Clarke in the City and the Stars. Folks live in virtual reality a lot of the time and get their adventure there. Just like I wonder why given that continous coal miners are contolled by 20 foot cables right now that could not become a 2 mile cable and humans don’t go down the mine for example.

    • “Of course this is the society seen by Arthur C Clarke in the City and the Stars. Folks live in virtual reality a lot of the time and get their adventure there.”

      A better explanation at this time is that the data is screwed up. While it is clear that some workers were not necessary it is very clear that understating inflation and overstating the ‘production’ by looking at a flawed measure like GDP are a much better explanation. Imagine what the BLS would have done with pyramid building.

      • Lets look at one industry where the bulk of the changes are now historic. Ag in 1850 employed 90% of the folks in the country, in 1900 it was 36% and now is 4% of the total work force to feed the US and a good bit of the world and produce a good bit of fuel to boot. In that case it was iron horses (tractors) that made a lot of the difference, plus the disappearance of the small family farm which tried to be all things, in favor of more specialization. Other blogs have point out that if a job is repetitive in nature t will be automated away. For another long term example consider how we have all become telephone operators until recently while the job of telephone operator with the panel and plug was abolished. Of course just recently our silicon workers have gotten smart enough that they can do what the old operator did, and connect you by a voice request.

        • Lyle,

          In my lifetime of 61 years, many millions of jobs have been eliminated through automation. At the same time, total jobs in the economy have continued to grow. All that is happening is that machines and computers continue to increase the productivity of workers – meaning that it always takes fewer workers to do the same work. But the demand for goods and services continues to grow. So more workers are producing many more goods than six decades, and the standard of living of all Americans has sharply increased.

        • I do not dispute history. I dispute your interpretation of what has happened in the past four years. During those years there is little evidence of a major type of change as was seen in the agricultural or telecom sectors during the last century. The easier explanation is a change in the methodology of calculating inflation and the shift of components that go into GDP.

        • The article has a better insight on the effect on GDP. When you have algorithms generate friction fees from massive speculative trading it is easy to claim that GDP is higher. The problem of course is what happens when the same algorithms get out of hand and drive the markets sharply lower and break what is left of the real economy. When so much of GDP is make believe coming from the financial sector it is difficult to pretend that things are improving as much as the promoters say they are.

  2. The worst part of the economic train wreck is in front of us, from the fiscal cliff, expiration of the Bush tax cuts, Obamacare taxes, inflationary Fed, regulations, the Alternative Minimum Tax (which now affects 30 million Americans, i.e. the middle class, because it hasn’t been indexed to inflation), a recession in a depression, etc., while we continue to pile-up trillions of dollars in debt, which may soon lead to another downgrade.

  3. Mark,

    I agree that some companies have learned to produce more output with fewer workers. But I’m not sure that’s the whole story. Some of the gain in GDP per worker can be attributed to the relative growth of high productivity industries and the decline of low productivity industries. GDP for construction, accomodation/food services, and auto manufacturing declined significantly from 2007 to 2010, and are probably still depressed. At the same time, GDP for high tech manufacturing, computer services, ambulatory health care (doctors’ offices), and oil/gas extraction increased.

    GDP by industry is not yet available for 2011. I think those figures will show a continuation of the shift in share of GDP from low productivity industries to high productivity industries. Auto manufacturing, food service, and construction may have come back, but not enough to regain their 2007 share of overall GDP.

    • But when you have GM stuffing the inventory channel talking about GDP growth coming from the auto sector is misleading. The same is true when you have an increase in the production of military equipment and supplies that will be wasted on unproductive activities that create no wealth or help with the accumulation of productive capital. And when election year spending ‘stimulates’ there is little in the way of a true return for that spending no matter how the effects show up in GDP over the short term.

      • Vangel, who was talking about GDP growth coming from the auto industry? or military equipment? My statement about growth was:

        “GDP for high tech manufacturing, computer services, ambulatory health care (doctors’ offices), and oil/gas extraction increased.”

  4. When you say “real GDP”, are you including the money SPENT by the various levels of Government? The US Government can increase GDP simply by running bigger deficits, which is not what most businessmen would call “production”.

    The last I saw, the federal “contribution” to US GDP was on the order of 23%, a huge number. And to this must be added (actually substracted…) a similar % of money SPENT (not “invested”) by state and local governments. So the REAL US GDP is perhaps 50% smaller than any number calculated by the US Government.

    For example, if payouts of unemployment insurance (and the uninsured extension periods) are counted as GDP, then OF COURSE we need fewer workers. Otherwise the unemployment “production” would go DOWN.

    Once we get the 47% up to 100%, everything will be perfect. Although the USA will be bankrupt.

    What is US GDP after government SPENDING is substracted?

    • Vinnie,

      GDP is a measure of the goods and services produced by the U.S. economy in a given period. Any goods and services produced by the government – such as the education services provided by public schools or defense services provided by the U.S. armed forces – are a part of GDP. Transfer payments are not.

      What screws up a lot of thinking about GDP is that the government doesn’t always measure it directly. One method for measuring GDP is to use the familiar expenditures equation:

      GDP = C + I + G + (Ex – IM)

      where:

      C is Consumption
      I is Investment
      G is government spending (but not transfer payments)
      Ex is Exports
      Im is Imports

      That equation confuses many because they do not understand that it is just a way of measuring GDP, and not an equation to express the relationships between GDP and any one of the factors. (For example, some incorrectly argue that the equation shows that increasing government spending will increase GDP.)

      Note that GDP does not include transfer payments such as social security or all the various forms of welfare payments.

      • Those transfer payments show up in consumption and if a portion are taxed away in taxes. Transfer payments can boost GDP through increased consumption.

        • I should have added that contributions for social benefits from individuals are subtracted from personal income. The ratio of contributions to payments has shifted from over 1 to now under 1 and during the recession the ratio fell further and faster so that excess showed up almost immediately in consumption which increased GDP by some amount.

        • Steven,

          The consumption of all persons shows up in GDP, regardless of how that consumption is funded. What is being estimated, though, is not consumption but rather the production of the items or services being consumed.

          If you are arguing that transfer payments increase total U.S. consumption, then I disagree. The money which is transferred reduces the consumption of those persons from whom the money was transferred.

          Keynesians might argue that the transfer payments can increase current GDP at the expense of future GDP if the transfer is from future taxpayers to current transfer recipients. IMO, they are mistaken because they ignore the impact of anticipated future taxes on current consumption and investment.

          • steven: “The ratio of contributions to payments has shifted from over 1 to now under 1 and during the recession the ratio fell further and faster ”

            Not sure if my point was clear. I think the ratio you refer to includes only the direct impact on current GDP – the current period transfers and the current period taxes. The indirect impact on current GDP of such transfers would include the negative effects of anticipated future taxes.

          • I was talking about how personal income is calculated in the NIPA. The seven NIPAs are used to estimate GDP. The personal income account is used to in part determine the consumption component of GDP.

            What is being argued is the deficit component of the budget. If the budget is in balance increasing transfer payments has no accounting effect on GDP. Your point only holds for budgets that are in balance. A budget surplus may have contractionary effects a budget deficit may have expansionary effects.

          • steve: “Your point only holds for budgets that are in balance.”

            I’m not sure you understand my point. A budget deficit represents a transfer from future taxpayers to current consumers. Even in such cases, that transfer still does not increase current GDP. That’s because the anticipation of future taxes affects the current behavior of the future taxpayers.

            For example, businesses which anticipate tax rates to increase in the future will invest less than they otherwise would have. At least, they will invest less in the U.S.

          • John, If a deficit did not have expansionary effects or provide support to aggregate demand in a counter cyclical manner during a recession then why do it? If all deficits are offset by current anticipations of higher taxes then the most recent recession would have been much deeper than it was.

            The additional deficit spending shows up in GDP proprotionally throughout the budget from government consumption expenditures to transfer payments. I don’t know how you could separate the recessionary dampening effects on aggregate demand from the expectations effect of higher taxes on current demand. It seems to me that there has to be a long lag in the expectations effect if it is present at all. It just doesn’t seem to be in the data where there is a clear small multiplier effect from deficit spending.

            In the past the net interest component of the federal budget (the only way that expectations of higher taxes from a budget deficit can be communicated) was very small in relation to NGDP and when it has grown and fallen it doesn’t seem well correlated with changes in consumption. The FED seems to have anchored expectations of changes in the net interest component far enough into the future for the budget deficits to have expansionary effects.

            I do agree that the trade off between consumption and investment influences future output and that deficits represent a claim on future output communicated to the consumer through expectations of higher taxes or higher tax rates (re: tax rates: which would be a permanent shift in expectations) but that effect is likely small as long as economic growth resumes and tax revenues rise as a result of that growth.

          • John, If a deficit did not have expansionary effects or provide support to aggregate demand in a counter cyclical manner during a recession then why do it?

            That is an easy question. If you are a politician you ‘do it’ because you have political goals and need to satisfy your sponsors. The vast majority of politicians are driven by libido dominandi, the lust for power and control over others. They do it by using the power of the state; in this case the power to borrow.

          • steven: ” If all deficits are offset by current anticipations of higher taxes then the most recent recession would have been much deeper than it was.”

            I disagree, and so do many economists. Of course, those “economists” who favor increased government intervention – during booms and during recessions – are going to argue that Keynesian stimulus can increase GDP.

            Do you really believe that Keynesian silliness?

          • John, I have my doubts about Keynesian stimulus but the growth we have seen in real GDP as MP illustrated above has to be in some part due to deficit spending. But the ill effects of the $250 billion increase in entitlement spending vs. pre-recession trend are being seen in falling labor force participation.

          • Steven Hales: “but the growth we have seen in real GDP as MP illustrated above has to be in some part due to deficit spending.”

            Why? How do you come to the conclusion that such growth “has to be” due to deficit spending?

            Government spending crowds out private sector consumption and investment. All resources – dump trucks, PVC pipe, computer scientists, civil engineers, mechanics, financial capital, etc – are limited in the short run. When government projects use those resources, prices for those resources rise above what they would otherwise be. At the margin, private sector projects are not available and private sector financing is curtailed.

            For anyone reading this comment who might believe Steven Hales argument, please read Keith Riler’s explanation at American Thinker:

            http://www.americanthinker.com/2011/06/crowding_out.html

            Riley shows that:


            A positive multiplier on government spending has been discredited by numerous reputable economists, who have demonstrated that increased government outlays substantially reduce private GDP. A dollar of government spending is destructive and produces less than a dollar of net economic benefit. Through spending, government destroys value and good jobs. “

          • John, I want to thank you for a great exchange. The government spending multiplier is so deeply embedded in the macro models and is so good a justification for social justice policies that arguing against it is a psychological assault against a view of world without which whole political parties might collapse.

            But if I am not mistaken, without it, the models underforecast. Sort of light the aerosol fix of climate models.

          • Steven

            But if I am not mistaken, without it, the models underforecast. Sort of light the aerosol fix of climate models.

            That sounds like a good argument for not taking either economic or climate models seriously.

  5. This +1.2% productivity growth is about one-half of what we had been running. MP, I guess, thinks this is good because its going up but it isn’t good enough and still shows an underlying weakness. And to brush aside the fact that maybe 8 million Americans don’t have a job and the toll that it is on their families misses a big point.

    • Norman, I didn’t read anywhere in Mark’s post that he thought 1.2% productivity growth was good enough. In fact, he referred to the economy as a “sub-par economic expansion”

      Mark certainly did not “brush aside” the fact that millions of Americans don’t have a job. Rather, he explained the cause for those out of work. Mark is doing his job as an economist showing why GDP can grow faster than jobs grow.

      You seem to resent that Mark chose to look at GDP and jobs from a different perspective than you think he should have taken. But Mark’s post is very legitimate, and a very valid point to make. I can understand that you want to add to Mark’s points. But why attempt to disparage his arguments?

      • His graphs and comments are meant to be optimistic which didn’t show the whole story. If he would have included a long term productivity graph showing that it was in a decline and way below the average this piece would have been complete. It wasn’t complete and that’s what I wanted to point out.

        Futher, MP is my absolute favorite economic blogger (too weak of a term for his work). His graphs are great. But that doesn’t mean I swallow everything whole.

        • Norman, please consider that your comment would be better received if you had first offerred your suggestion:

          “If he would have included a long term productivity graph showing that it was in a decline and way below the average this piece would have been complete.”

          rather than incorrectly suggesting that Professor Perry thought 1.2% GDP growth was good. Further, you accused him of “brushing aside” the fact that millions of Americans are still employed, when he did no such thing.

  6. The U.S. economy remains on life support from an additional $1 trillion a year in federal spending (compared to 2007), endless quantitative easings, and generous imports.

    • Hydra, there’ve been only four major economic revolutions in human history: Agricultural-Industrial-Information-Biotech.

      There will be hundreds more, and there’s lots of work to be done.

      The potential speed of economic revolutions are constrained by resources, e.g. labor and capital.

  7. Labor is not a constraint in a robotic economy. Hence the increase in the ecinomy with fewer workers.

    Capital does not seem to be an issue since fewer and fewer people are retaining more and more if it. There are fewer opportunities for them to invest it with fewer workers earning less to spend.

  8. New employment down, far larger people dropping out, therefore, lower unemployment number! Government inserting in every way possible more dollars.

    The age of Depression is coming. If and when we have to raise rates for Government and dollar to survive, because we must at some point to stop the erosion of the dollar on necessary goods, (Food, Fuel, Med canter costs have all increased 100% averaged since Fall of 2008. It’s getting worse! All hidden but, not contestable!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>