Friday, December 28, 2007

The Slow Recession

Post-WWII recessions in the US occurred because of excess inventory accumulation by manufacturers. They came fast and went away pretty fast, too, as plants laid-off and re-hired industrial workers. However, the nature of the business cycle has changed: manufacturing now makes up a much smaller percentage of the economy and employment. Manufacturing jobs in absolute numbers are back to 1949 levels, when the country had half today's population. The process of de-industrialization accelerated sharply after 2000 as China became producer to the world and the US lost an astonishing 3.3 million manufacturing jobs. (see chart below).

Manufacturing Jobs (Chart:BLS)

It is instructive to examine the kinds of jobs lost and created during the full business cycle from 2000 to 2007 (see table below). Almost all net new jobs were in health care, social services, leisure and hospitality, i.e. at the fringes of the service sector. The jobs that replaced manufacturing came in construction, financial and business services.

Data: BLS

The nature of this employment restructuring suggests an increase of economic inertia at the cost of less value-added. I think that such peripheral jobs are slower to be lost during the initial phases of a slowdown, but once gone they will be slow in coming back, too. They don't cost as much to maintain, but they also don't produce as much immediate profit when the cycle turns up.

Therefore, the business cycle may now become more grinding and drawn-out, instead of painful and short. This explains why the current slowdown is like watching paint dry, with limited job losses so far (noting, however, the effects of the controversial BLS birth/death model on job creation since 2003) and mediocre retail sales, ex-fuel and food.


  1. I am unclear as to your focus on the loss of manufacturing jobs in America.

    Are you implying that the loss of these jobs to foreign competitors has been a 'really bad idea'? If so, I wonder if you have fully thought this one through.

    'Yes', America lost Manufacturing jobs at tremendous rates, but this would have occured whether the jobs went overseas or not (it just happened a little faster becasue of the jobs ALSO going overseas). Productivity improvements in manufacturing over the years have been far better than productivity improvements in the services sector. In the past it took 25 workers to make 1 car, today it takes 3. This means that unless we produce an unlimited amount of physical stuff, eventually robots would replace American workers whether the jobs went overseas or not. There are only so many cars we can make!

    Also-- In terms of $ value: US manufacturing as a % of US GDP AND as a % of World Manufacturing is doing quite well. There just aren't any jobs in US manufacturing anymore (but there still is plent of manufacturing and American manufacturers stil make plent of money). Where we have outsourced jobs to foreign manufacturing, the total value of that manufacturing has not been very great.

    This same process is occuring in 'poor countries' today. Robots are replacing their workers such that emerging market productivity is improving quickly at the expense of job layoffs (or less rapid job expansion for unit expansion in manufacturing). This process has always happened and it will continue forever.

    If your point is all this is to complain that Americans are not moving as a group into higher value added positions, I agree with this point, but manufacturing is not the only higher value added job out there.

    This is really a problem with Americans themselves--not only is there a 'debt bubble', there is an 'education bubble'. Students do not study science anymore-- schools are full of interesting but completely impractical majors.

    The US needs to focus much more $ and energy into science and basic science research. This is our ONLY hope.

    Asking former American manufactures to 'return home' is not going to work... robots already took those jobs from Americans. They will soon take them from the Chinese.

  2. Dear thai,

    1..If we do not make the stuff at home we must import it. We are able to run huge deficits because the dollar is still accepted as a reserve currency based on crude oil - but for how much longer? Ten years? Twenty?

    2..Manufacturing is not just robot jobs at the shop floor but also high value R&D, technology and skills. When manufacturing goes abroad they quickly follow. Just look at Japan post 1950s and 1960s..China is now doing it all in less than 10 years.

    3..By contrast, how much value is added by a waiter, a hairdresser or a nursing home attendant? Bread and circuses was not a good recipe for Rome and it isn't for the US, either.

    4..Low-value mfg. cannot and should not come back to the US. But where are the new high value mfg. jobs? Where is the basic research and application of the most important sector for the 21st Century - alternative energy? The largest wind power companies are European, France is #1 in nuclear, Spain and Denmark in wind, Germany in geothermal...etc. We are stuck with the oil-dollar economy, and unless we disengage ASAP we will go down with it. The corn ethanol scam is criminal, to put it mildly.

    5.. Totally agree with your education points. They are symptoms of all the above, however, don't you see? It's not the students' fault that mechanical and chemical engineers got paid 1/10th of what derivatives traders made after 2-3 years. Money is the most powerful incentive. When I graduated college ONE out of 350 of my engineering class went to finance (yours truly). I just happened to speak to the Provost of my old school and the ratio is now 125/350.

    As a finance person I can tell you in no uncertain terms that finance is based ON the economy, not a foundation OF the economy.


  3. Hellasious,

    'yes' and 'no'

    1. Yes, but the real problem with imports is Oil, not manufactured goods. The balance of trade in Manufactured imports are already improving (I know, I know, the dollar weakness has its own problems).

    2. Not really. Scientists working at the NIH here in Bethesda MD are classified as service workers. I beleive most R&D jobs are classified as service jobs. Intel and most of the other 'big boys' in Silicon Valley gave up most of their manufacturing when the Japanese started kicking but in the 80's. They moved into design and R&D and outsourced to the Jabil and Selectron's of the world who moved their manufacturing abroad and then replaced a bunch of phillipino/Taiwanese workers with Robots. Intel employees lots of scientists to this day in the US (most of them were born abroad).

    3. Yes, but this is again making the same mistake. If you look at scientist vs. waiter, they are lumped together when they clearly are not.

    4. To some degree we may be disagreeing based on a definition of terms. I beleive what you mean by high value added manufacturing jobs of tomorrow are classified today as service jobs. I had a friend working on Solar Energy Research here in Maryland and (though I am not sure) I believe his job would have been classsified as 'service'... working in front of a computer doig design work all day.

    5. Yes BUT you are missing your own point. In the
    1950's, kids studied science. They went on to get high paying jobs. Their kids went to school and learned that the ROI for scientific education was not nearly as good as it was for medicine, law, and finance. Their kids took those jobs. Today the ROI for degrees in those fields is dropping (it makes a big difference which school you train at and which subspecialty you go into--- trust me there are tons of MDs/JDs and MBAs coming from second tier schools or going into certain medical specialties (pediatrics/family medicine) that do not make much $).

    Remember, science too has lots of low value added jobs... there is a big difference between someone who processes a machine which rus blood analysis all day (a job which in many ways is akin to a manufacturing job and which pays fairly well, but is classified as a service job) and someone who invents a new machine which diagnoses disease.

    This service vs. manufacturing distinction is a little artificial. Yes waiter jobs are increasing, but they are only increasing becasue Americans eat out more. They eat out more because
    #1-- Americans are all going into debt!
    #2-- They work more hours and are offloading for convenience things like cooking (look at he explosion in stores like Trader Joe's with its 'pull tab gourmet' prepared foods!)

    Remember, a 'service worker' who keeps a manufacturing working alive 10 years longer than they otherwise would have lived so that they can work productively for 8 more years is still very definitley contributing to the economy.

    The largest growth in 'service' jobs is in the healthcare arena (and yes I know a lot of healthcare spending is also wasteful... though tell that to all the enderly who want to have 'golden years' after they retire (clearly the biggest bubble of them all) as this is where most of our American debt is really going).

  4. When lopsided inflation is as rampant as has been as of late. Yes, people will chase the money. They have to. People choose academic pursuits when they don't have to worry about how they will provide food, shelter and clothing for themselves and their families.

    Your government has been enabling Financial Services/Banking corporations to alter the flow of capital from productive uses and into wasteful buckets. This has consequences. The worst of which will be deflationary pressure on wages due to globalization and steady upward/stable bias on the cost of life's necessities.

    All of this on top of the fact that record debt levels/borrowing from future economic activity have been occurred in the recent past. There will be a tough grind ahead.

  5. Finance serves the economy, but should not BE the economy. So true.
    The US's main export thesedays is the Dollar.

  6. Dear thai,

    I am not lumping all service jobs together. The BLS breaks them out into very specific categories. Go to the BLS site and you will see what I mean. It takes some searching but they have a great statistical facility with charts, etc.

    Look at the table in my blog: 4.5 mio net jobs added, 4.4 mio in health care, hospitality etc. And when the BLS says health care services they do NOT mean doctors and highly paid professionals, but much more menial type jobs like attendants, etc. Again, you have to go to the BLS site for all the detail.

    We are dumbing-down our economy and it is reflected in our workforce.

  7. Just a few comments on the education aspect. It takes about 15 years from primary school to third-level graduation. That's a long lead-in time. Attempting to encourage teenagers to choose science, technology or engineering careers is like asking them to have root-canal treatment without anaesthesia!

    Our 'western' economies are approaching the inflection-point peaks of their economic growth curves. Once we are over the top, and begin to descend we shall need increasing numbers of imaginative, skilled technicians - and fewer PhDs! People who can make things, people who can fix things, and most importantly, people who can provide us with clean water and grow our food. Man cannot survive on a diet of cash/debt and fossil fuel. And robots need not apply!

    Brian P

  8. re: our desperate need to develop an alt energy economy

    I was recently at a dinner party with an engineer who is working for a solar start up that has been getting massive VC funding -- even bigger scale than the old days. There is right now a lot of competition among the VCs in the Valley and elsewhere to get their horses funded and out the gate first. The Germans are big now, but depend on massive subsidies and our guest is working on viable ways to leapfrog them. Manufacturing should begin over the next year or so -- and yes, they will be manufacturing here.

    I think that people outside the world of those who made their fortunes in technology tend to forget that techies have a different value set than Wall Steet. Engineers like to solve problems and if those solutions create profits, so much the better. But the great ones do it for the satisfaction of solving the problem -- it's who they are. When you have people like that who've made a few hundred million dollars because of technology, they can indulge that desire to find an elegant solution and end up creating the Alt Energy economy just for the hell of it. They enjoy the money too, but it's not their reason for being. Those geeky boys and girls from the Bay and Austin may save us in spite of ourselves. At least they have the money and talent to try.

  9. The U.S. needs an enormous number of better-paying jobs. The needed number of highly educated and skilled scientists and engineers is a drop in the bucket. (And spending the enormous amount of time, money, and effort to get there is certainly not risk-free: remember all those aerospace engineers?) The necessary jobs at decent wages simply do not exist, and no amount of education or training is going to change that. For good introductions, see Gordon Lafer, "The Job Training Charade," and Louis Uchitelle, "The Disposable American: Layoffs and Their Consequences."

  10. Sean, I agree that creating skills for skill sake is obviously a stupid idea. That having been said, there is a difference between knowledge 'invention' and training. And in particular, I think these issues get confused with the fundamental fractal nature of wealth distribution in society and we get some of the odd blog postings so representative of crowd think.

    We cannot infinitely use oil and coal: we will run out and/or possibly cause major environmental problems even before we do. Similary we cannot infinitely make cars or other manufactured goods (at least cars like we have today) as there are only so many cars needed on this planet.

    Information however is limitless. And there are two kinds of information-- that we know and that we don't. It is the underinvestment in 'that we don't' which is going to get us into trouble. 'That we don't' in general is discovered thru basic science research.

    Not only are we in big doo doo because of all our debt, we are big doo doo becasue we are not spending on basic science R&D. If you look at basic science R&D spending as % of GDP, it is shrinking. We actually spend a fair amount on R&D, but it is to applied research which is different (important but different never the less).

    According to the National Academy of Sciences it takes on average 30 years to see a payoff from basic science research (the average payoff has been 33%for the 20th century).

    No private investor will invest in something that has a nebulous 30 year payoff. And in this world of 'privatize', we are seeing this come back to haunt us. Drug piplines are at recent historical lows if you look at truly novel drugs. This is true for MOST major scientific fields (alternative energy is no exception).

    And private investments in alternative energies want 5-10 year payoffs which are often quite unrealistic.

  11. thai, sean, eva,

    I have a significant difficulty with the idea of a 'knowledge' economy being able to extract us from the inevitable uncoupling of our 'western' economies from our primal energy source - liquid fossil fuel. This change-over may be several decades away - but it will happen.

    The alternative energies are hardly sufficient to support our existing situation, and will be insufficient to sustain even low-level exponential economic growth. At some future time growth will stop.

    This not-so-distant situation should be addressed but it is such an 'appalling vista' that no one wants to discuss it - except in the abstract. It's the equivalent of talking about your own mortality. I repeat - growth must stop at some point; so what plans are we preparing to deal with this situation?

    I know that many are putting their faith, or their hope, in 'new technologies'. Unless the 'techies' discover some way to repeal the Second Law of Thermodynamics new technology will only postpone the inevitable - and only for the wealthy nations, and only for the wealthy of those wealthy nations.

    I have formed the opinion that the future of the developed western nations will be somewhat similar to their historical development since about 1830; only in reverse!

    Brian P

  12. Well, this is an interesting day on the comments column.

    Thai--I could not agree more that we need to put a lot more money into basic research. I started wondering whether American Nobel Prizes in the sciences were a lagging indicator of past funding and would soon peter out, but I'm not sure whether any other nations are pouring resources into basic research even at the rate the U.S. does now. And even at the most successful, basic research won't create many jobs without some sort of industrial policy that keeps the jobs in the U.S.

    I agree with Brian's skepticism that there will be any easy technological solutions. Even in the best case, as Jared Diamond has pointed out, new technology creates new problems as it solves old ones. But incremental technological advances can add up, particularly even they are combined with other societal changes. It would make a lot of sense for the U.S. to become a much more concentrated, urban society, for example, with free public transportation and charges for those who insist on driving. For all the flaws of the U.S., it has historically been a flexible and resilient country--although that may require exhausting the bad ideas, first. Of course, past performance is no guarantee of future results, but I can't think of many other countries that are more likely to muddle through, if that is indeed possible.

  13. The most recent numbers on (a good statistics site I discovered not long ago), for manuf value added, show worldwide %s as follows:

    US 26.9%
    Japan 17.6
    China 10.8
    Germany 6.9

    This is in constant 2000 USD. And nominal, not PPP-adjusted numbers. I've decided that I trust (more-or-less) PPP adjusted numbers for internal national markets and nominal numbers for international trade.

    The actual US published (manuf) numbers are here:

    ( of course).

    This breaks things down by internationally defined industry classes.

    You don't 'see' a lot of US manufacturing, in say Walmart as much of it looks like this:

    Note that number of highly specialized very expensive capital goods manufactured by small US semiconductor equipment companies.

    Also look up Clymer Technologies. In San Diego (and look up what their stock has done).

    Or First Soloar, FSLR, in Phoenix. And look at their stock too, as well as capacity (thin film solar) that's growing by leaps and bounds (yes the next several factories will be in Malaysia - one presumes that the components that are shipped to Malaysia will come from the US).

    Corning owns the market for all the highly specialized glass that goes into wide panel TV screens (you know: Samsung, etc). And I believe the glass is actually the most expensive single component in those TVs. Much of the manufacturing is now done in E Asia, but the key plant that still produces the most important process, etc innovations is the one that out in countryside somewhere in Kentucky.

    Etc, etc, etc.

    The US has lost manufacturing jobs but not heft or skill.

  14. Thai,

    Have you considered that the drive towards higher levels of productivity is inherent to a system of profit maximizing firms which are competitively driven to expand or die, i.e. that capitalism is necessarily expansive. Which, other than for potential resource limits, might seem unproblematic until recognized that same drive towards greater productivity is also an at least relative displacing of living labor from processes of production.
    From which a) higher capital:labor ratios which tends to reduce surplus value relative to the mass of capital, so pressures the rate of profit b) generates an expansive service sector, much of which - from the perspective of capital - engages in unproductive labor hence, as cost, also pressures rate of profit c) downwards pressure on wages, making realization of surplus value as profit more difficult without expansion of credit and an expansion which itself becomes limited by just what it is imagined to overcome: realization difficulties and relative insufficiency of profit.

    The categories and classifications which it seems you rely on are those of neoclassical economics which has never developed a satisfactory understanding of profit and completely ignores the necessary differentiation between productive and unproductive labor. Please note that 'unproductive' does not mean unnecessary since in fact what we've seen is the increasing necessity of unproductive labor as firms find it more difficult to achieve sufficient sales other than through greater reliance upon marketing, advertising and financial activities while the social consequences of prolonged slow growth increases the need for guard-type activities.

    My point? Simply that the capital system contains inherent limits to automation, limits that work contrary to the generalizing of such innovations while also pointing towards what might be.