Also in town for last week’s International Symposium on Forecasting was Bill Gavin from the Federal Reserve Bank of St. Louis. I had an interesting discussion with him about changes over time in U.S. employment dynamics that I wanted to share with our readers.
In the earlier postwar recessions, the unemployment rate began to fall very quickly once the expansion began. By contrast, the unemployment rate continued to climb even after the recovery had begun for the 1990-91 and 2001 recessions. No one is predicting a rapid drop in the unemployment rate this time around, either.
 |
Bill called my attention to the contribution of temporary layoffs to this changing behavior in the unemployment rate. He noted that the Social Security Amendments of 1958 explicitly exempted unemployment insurance from income taxation, and recalled a 1976 paper by Martin Feldstein which proposed that this gave firms a strong incentive to use temporary layoffs in response to a business downturn. By temporarily laying workers off rather than asking them to work shorter hours, the firm could deliver maximal after-tax compensation to its labor force, intending to hire those same workers back as soon as business improved. Temporary layoffs accounted for up to a quarter of those unemployed at the worst of the 1973-75 recession.
 |
Bill believes that the key developments that changed this dynamic were the Revenue Act of 1978, which subjected unemployment benefits to partial taxation under the income tax law, and the Tax Reform Act of 1986, which made unemployment
benefits taxable as ordinary income. Since the mid-1980s, the above graph shows that temporary layoffs have become a much less important feature of recessions.
Suppose you adopted Feldstein’s perspective that those individuals who fully expected to get back their old jobs soon are not really unemployed, but should instead be viewed the same way we treat workers who remain employed but work shorter hours. If you subtract temporary layoffs from the number of unemployed, here’s what the adjusted unemployment rate would look like. The earlier recessions look much more like the recent jobless recoveries.
 |
That adjustment also makes all previous post-WWII recessions much, much, much smaller than our current episode. It’s now twice as big as any other post-WWII recession..,
Very informative. Tax incentives at work, though perhaps not as intended.
How does this fit in with new openings/severance? I remember seeing some discussion that severance was running at a normal rate at this point but rehiring was subpar. It would seem to cast doubt on Feldstein’s idea, as would a larger than normal ratio of permanent layoffs vs temporary ones.
Plus anecdotally I think employers have no intention of rehiring workers, and laid off folks I know are not expecting to return.
This also suggests that conventional analysis may be misinterpreting long-term unemployment in certain respects. Usually, it is said that the short-term unemployed retain skills and therefore prevent the NAIRU from rising, whereas long-term unemployed lose skills and cause the NAIRU to rise. But if the short-term unemployed are largely people on temporary layoff that are only nominally unemployed, then their prevalence actually increases the NAIRU, since it means that the “true” unemployment rate is lower than the reported rate to which the NAIRU is compared. By contrast, the prevalence of long-term unemployment in the aftermath of the recent recession suggests that most of the unemployment has been “real” and therefore not the sort that would artificially increase the NAIRU (although the mechanism of skill loss still exists).
Eyeballing the two series, my impression is that the tax change may have played some role but falls short of a comprehensive explanation.
The adjusted time series shows unemployment on a plateau for some time after the recessions up to 1980. But for any recession after that the unemployment rate continues to drift up for quite some time.
I have to disagree with you. To me the culprit is the major difference in government attitude: after WWII the goal was not to engineer economic recoveries, real factories were being built, an arms race with the USSR was going on, alot of urbanization was going on as well. Neither gov’t nor employers, nor employees were looking for ways to squeeze 0.5-1% extra tax benefit.
Fast forward to 1990 and you have one severe crisis after another, gov’t engineered growth, massive debt accumulation and naturally debt induced ‘jobless recoveries’.
The economy of today is not the economy of half a century ago, gov’t today is not what it used to be 50 years ago. Shall we expect similar results when the ingredients and the cooks are quite dissimilar?
…”Suppose you adopted Feldstein’s perspective that those individuals who fully expected to get back their old jobs soon are not really unemployed, but should instead be viewed the same way we treat workers who remain employed but work shorter hours….”
Feldstein wrote the paper in 1976. Hasn’t structural unemployment been *increasing* since that time? I don’t see how altering taxation of unemployment benefits is going to affect whether the company that you work for is going to still be in existence or not after a layoff occurs.
Chris, Baychev, and Doc: Perhaps I did not emphasize strongly enough the implication of the second graph. Temporary layoffs were an important factor in unemployment in the 1970s but are not a significant part of what is going on right now.
I read this article as:
Unemployment numbers in the past might be exagerated by temporary layoffs, which counted as unemployment instead of underemployment.
Adjusting the charts for that effect paints a grim picture of today’s recession.
I’d really think about Mr green smiley. He’s looking backwards.
The reason the unemployment definition is constructed as it is — you have to be actively seeking work to be counted as unemployed — is to deliberately exclude the unemployed worker who is not seeking work because they expect to soon return to their old job.
So because of the change in the tax law you are saying their is a change in the unemployment rate that did not include these workers either before or after the tax law change.
I can not buy this argument.
There have been too many structural changes in the economy for this explanation to hold water.
The old concept of a temporary lay-off just does not impact most unemployed in the modern economy.
KevinM: The Econbrowser Emoticon is neutral, not smiling.
I agree that today’s recession looks much worse when we take layoffs into account. However, besides the level effect, the dynamics of the unemployment rate during a recovery seems to me quite similar in both pictures. So, is this really evidence that earlier recessions were followed by jobless recoveries as well?
Perhaps the issue is that there is little to lead to higher employment in a “post industrial” United States. Without the pull from large corporate employers, the service/support sector is stagnating or devolving.
Increasingly, employment will be self-employment rather than corporate employment. Increasingly, the highly aware and highly motivated individuals will be the source of wealth. Conversely, the growing proportion of the U.S. population that is poorly educated, underskilled, and under motivated will find themselves unemployed or underemployed.
The days of getting by with assembly line, union protected jobs are numbered. The haven for those jobs now appear to be in government.
If my argument that most of the people on temporary unemployment are not seeking work and thus are not unemployed is correct to make the adjusted unemployment rate in your second chart you need to add the temporary lay-offs to both the numerator and the denominator.
Great to put this important graph on even basis. Now its clear that things proceed as they should.
spencer:
“The reason the unemployment definition is constructed as it is — you have to be actively seeking work to be counted as unemployed — is to deliberately exclude the unemployed worker who is not seeking work because they expect to soon return to their old job.”
But the data are collected via a survey in which workers self-report their labor force status. Unemployment insurance requires recipients to be actively seeking work, so people on temporary layoff who wish to receive unemployment insurance will actively — if perhaps half-heartedly — seek work. At the very least, they will make diligent efforts to convince the unemployment insurance office that they are actively seeking work. I’m guessing that they are also likely to report in the CPS interview that they are actively seeking work, rather than acknowledging that they are engaged in a ruse to fool the unemployment office. And I would guess that most of them really are actively seeking work, since (1) I don’t think most of them want to be quite so dishonest, (2) being out of work even temporarily can be significant problem, and (3) there is no guarantee that they will be rehired.
Andy, the point is you are guessing. You do not know.
My guess is that people on temporary layoff tell the unemployment survey — conducted by the federal government while the unemployment compensation is done by state governments –the truth. Remember, the sample is a rotating sample where they go back to the same people for several weeks.
If your guess is correct no one would report that they are only temporarily unemployed. They would report that they do not know. For unemployed union members, auto workers for example their union would probably tell them how to answer the questions.
So why is one guess any better or worse than the other guess?
But the thesis behind the chart requires assuming something we do not know.
So why my guess any worse or better than your guess?
We all know that the nature of unemployment has changed over the decades from largely employees of durable goods manufacturers who were usually a temporary lay-off to more evenly spread among all industries. If you look at the data you see a clear trend of the number of people on temporary is on a long term downward slope while the number of people on permanent lay-off is on an upward slope. In the latest cycle the number on temporary lay-off was half the number in the 1975 cycle while the number on permanent lay-off in 2009 was almost double the number in 1975.
The ratio of temporary to permanent lay-offs directly contradict the thesis that tax changes lead to a change in behavior.
I just did a chart of the share of job losers on temporary layoff. From 1967 to 1983 it bounced from 25% at cyclical lows to some 40% at cyclical highs.
After the 1980 recession it started trending down so that the latest peak was about 30 and it is now at 15% — an all time low about 10 percentage points below the lows in the 1970s.
This directly contradicts the thesis that tax changes lead to an increase in the number, or share, of people on temporary layoffs.
Well, then we have the big recession that could probably be renamed depression. No temporary layoffs and an important part of the manufacture jobs have -irreversibly?- fleed to other countries.
It is time to see if the hi-tech, knowledge-driven economy is compatible with low unemployment rates.
Spencer: Let me use the seasonally unadjusted data (for which the accounting identities hold exactly), and use the numbers as reported today for May 2010 (these may be different from the May 2010 numbers if you look them up on the site tomorrow). BLS reported 14.369 million Americans unemployed in May 2010. From their Table A-11, this breaks down as follows: 1.192 M on temporary layoff, 7.620 M who lost their job for reason other than temporary layoff, 0.922 M who chose to leave their job, and 4.635 M who were previously not in the labor force but are now looking for a job. Notice that (1.192 + 7.620 + 0.922 + 4.635) exactly sums to the 14.369 reported unemployed, and that the 14.369 unemployed divided by the 153.866 M labor force exactly equals the reported 9.3 % (seasonally unadjusted) unemployment rate.
It thus makes perfect sense to subtract 1.192 from 14.369 to calculate the number unemployed for reasons other than temporary layoffs, and to then take the ratio of that difference to the 153.866 labor force to construct an adjusted unemployment rate.
Spencer at 2:48 p.m. The claim was that the tax changes over 1978-1986 led to a decrease in the share of temporary layoffs in the recessions after 1986 compared to the share of temporary layoffs in recessions prior to 1978.
That is certainly confirmed by the second graph above which sounds very similar to the graph you have created. I conclude that you must have misunderstood the point Bill Gavin and I were making.
You said:
a 1976 paper by Martin Feldstein which proposed that this gave firms a strong incentive to use temporary layoffs in response to a business downturn. By temporarily laying workers off rather than asking them to work shorter hours, the firm could deliver maximal after-tax compensation to its labor force
I’m sorry, but how can I read this except as to say that tax changes lead to firms increasing the
ratio of temporary layoffs?
Yes, later you said:
Since the mid-1980s, the above graph shows that temporary layoffs have become a much less important feature of recession
Which statement am I to respond to:
first you say it leads to increased temporary layoffs.
But then you reverse yourself and say that layoffs have become less important.
Actually, we are talking about a tax change in 1958, but we are using data that only goes back to 1967 to debate if there was a change in the early 1960s.
We have no data on temporary unemployment before 1958. So one of our data can really address the question of whether or not there was a change in behavior before and after the 1958 change in tax law.
Spencer: You’ve got to be kidding.
I described one change in tax law in 1958 and explained why that led to an increase in the role of temporary layoffs. I described two subsequent changes in tax law in 1978 and 1986 and explained why those led to a decrease in the role of temporary layoffs.
If you think my claim was that all changes in tax law, no matter of what type, always lead to an increase in the role of temporary layoffs, then I assure you that you have misunderstood my position.
OK. I’m sorry. I read the part of your analysis before the charts and did not read the part after the charts.
I thought you were talking about one change in 1958 — I did not read the part about later reversals of that change.
Authors and economists wag on while the housing industry rests in a graveyard, never to live again. There is no shortage of pulpits for authors and economists that have never broken a sweat swinging a hammer or crushed their knees laying shingles, carpet, or tile. They chatter away their lives while home builders have no place to be counted or heard. The doublespeak I read here does not comfort me after two years of unemployment and bankruptcy. A career of building homes has left me unemployable. If economists and pundits will not call it…. then I will be direct: This is the “Greatest Depression Ever”. A jobless recovery is a lie and an insult. A bubble which is as destructive as this one is deserves a more sinister and deadly moniker. I pray for you, all you heartless hacks.
I have been out of work longer during this “Greatest Depression Ever” than the total sum of all of my layoffs in the last 40 years. Now, how will you spin that?
So you’re 60 or so and unemployed. Write your Republican congressman or senator and point out you really don’t want to be that way for another 10 years before getting SS benefits.
And I agree with you. This is a Depression and it’s going to be so for an extended period of time — at least for 15% of the population. If not more.
A note on “temporary layoff” – if the employer notifies the the unemployment office a list of workers are on temporary layoff, they are not expected to look for [full time] work to qualify for unemployment benefits. The insurance rates and such, account for such practices, which are far less common today than in the 60s. The availability and terms of such unemployment insurance varied from state to state and over time.
So, the “temporary unemployed” is a clear definition, not a state of mind.
This article by former Intel CEO Andy Grove is anecdotal but makes an effective point that there seem to always be a lot more manufacturing than knowledge based jobs in even innovative high tech industries. Therefore, he argues, giving up manufacturing competitiveness inevitably leads to an economy with precarious employment prospects.
Perhaps it is simply that the grey bars are incorrect. Recession ‘ends’ were called too quickly based on government goosing of the indicators rather than on underlying growth.
Going beyond Andy Groves’ points on social and technological change, I’ve put together some thoughts on the issue of moving beyond a jobless recovery on Google Knol, including heterodox economics perspectives and possible cures, with ideas from Marshall Brain, Bob Black, Jane Jacobs, E.F. Schumacher, James P. Hogan, Mahatma Gandhi, Gene Roddenberry, James S. Albus, Jacque Fresco, Daniel P. Moynihan, and many others.
Paul, one kind of issue you didn’t seem to touch on is the idea that if demand for various types of labor is elastic while costs of various forms of increased social welfare have been folded into the cost of delivered labor, while the social welfare benefits are not included in “real wages”, then the demand for labor at a given measured “real wage” would show a decrease in the absence of other factors.
Examples of costly constraints aimed at improving social welfare that are sometimes folded into the cost of delivered labor include costs of medical insurance, workers compensation insurance, unemployment insurance, medicare, compliance with OSHA regulations, compliance with EPA regulations, compliance with ADA regulations (in places of business), compliance with local codes and regulations, credentialing costs, background checks, accounting and tax compliance costs, etc. My sense is that the impact of these costs has increased over time.
Where I live, in the Twin Cities, MN area, delivered cost of skilled carpentry and delivered cost of skilled plumbing seem to have remained in excess of $60/hour and $100/hour respectively throughout the recession and housing downturn. No doubt there are people capable of doing these skilled trades that one can find on e.g. CraigsList.org for less. but reputable businesses and the costs of doing reputable business still apparently dominate the market. I’d be interested in reading a study of how delivered cost and “real wages” for employees have varied in the skilled trades over the last several decades.
These charts have me wondering about the speed of economic structural adjustment and how that might be evolving over time. I would have thought that the speed of structural adjustment has increased over time with information technology, new inventory techniques and the increased use of outsourcing. The chart of permanent unemployment contradicts that impression.
Josh, thanks for the comment. I agree with you that compliance costs have probably increased over time. While you go more in depth, part of the issue you raise (from the employment creation side) was touched on in this section of the document on “Increased Social Benefits”.
The suggestion there is that when costs per employee are fixed at a high rate (like with medical insurance) there is an incentive not to create more jobs when there is more work but instead to work the employees you have as hard as you can until they collapse (and then replace them, subject on to the cost of turnover and retraining). Another factor is that every new employee requires more management attention and training, another cost companies want to minimize. If everyone in the USA had single payer health care, I would predict we would see more jobs. That is not to disagree with you that rising costs might reduce some people’s incentive to hire others; it’s a complex issue, as you point out. Other incentives that have been tried include Government subsidies to the private sector (for new job creation).
I’d agree that market position (including having made it through costly barriers of entry) is a major determination of how much you can charge for a service (you can afford to advertise, you have built a reputation over years, you went up the learning curve of compliance, etc.). But another aspect of what you touch on is “sticky wages”, in that wages for old timers in a position rarely fall in absolute terms, but when people are displaced, new jobs may be at much lower wages. So, we see some increasing wage divides among those old timers who have kept their jobs vs. those who got laid off, so even when new jobs are created during a downturn, they will likely pay a lot less than jobs created during a boom.
Where I live, a fairly rural area in NY with few local jobs, from what my neighbors tell me, there are skilled people around who will work for US$20 an hour, which is barely enough full time to make ends meet in our society (considering downtime, taxes, sick care, etc.) but many people in this area are frugal, own their own homes free-and-clear, drive old cars, and so expenses may be lower than near a major city. Also, I’ve met several people working at local places like Home Depot or Lowes (which I doubt pays more than US$10 an hour) who have been ready to hand out a card saying they do plumbing or other things. What would be of great interest is what are the wages of newly hired apprentice plumbers in larger companies? Still, I won’t disagree that in some places, like Silicon Valley, people were saying a decade ago that the real money was in plumbing, not software. 🙂
Ultimately, if we continue to have large numbers of unemployed people, the natural level of wages eventually will be the level at which people are willing to slowly descend into debt and burn up their health and other personal physical capital like cars (so, less than a living wage). If the jobs don’t came back, without broad social changes, I predict that is what we would see. Or maybe worse in some places: “1,500 farmers commit mass suicide in India”.
Here is a video presentation by Marshall Brain from a year ago on why many more people are going to be hurting soon in the USA from job loss:
“Marshall Brain – Automation & Unemployment”.
As I see it, most economists are clueless when it comes to the implications of modern technology on the workforce — in part because divide by zero errors from plunging labor costs would mess up all the pretty and elegant equations. 🙂
Google search on: “divide by zero” and economics
My feeling is that we are finally seeing the long-predicted “Black Swan” as far as the declining value of most human labor in the information age. Even for most of the carpenters and plumbers you mention, this will happen at some point, such as when new tools or new materials let one plumber or carpenter do the work of two or more. And if you look around, there are various initiatives to essentially print houses complete with plumbing in one day (I’m not saying any of them are a big success yet, just that people are working towards that). Given the dynamics of capitalism, automation and better design only has to reduce the need for human labor a little to see a race to the bottom on wages over the long term. Unless demand for new junk keeps pace with rising productivity, we will see net jobs losses; but we now live in an age of “Reduce, Reuse, Recycle”. The only way past that is IMHO through some mix of things like a basic income, a gift economy, stronger local subsistence economies, and better resource-based planning, as well as more mainstream ways of job creation.
Problem with these “official” unemployment figures is that they are too low. The real figure is more at 20% because it doesn’t take into people who just decided to retire, quit looking for work, etc.
Paul, several of the costs I mentioned , including workers comp, unemployment insurance, medicare, insurance, and social security, mostly scale with hours worked or wages paid rather than being a fixed cost per employee.
Whether or not technology will substantially change the business of carpentry and plumbing in the future, it’s interesting to understand why the delivered cost of these labor intensive services at present seems to be so little affected by factors that would seem to dramatically affect the supply of willing workers, and why this is true even though its clear that thequantity of demand for these services is an elastic function of price – e.g. people wouldn’t DIY and would carry out more projects if it was cheaper to hire.
Various commentators have described the current business conditions as showing a shortage of “final demand” – for example. So it is interesting to reflect on situation where it seems that society has implicitly chosen to accept a lower level of final demand in exchange for the benefits of greater workplace safety and social insurance.
I wonder how international trade would be affected if importers and retailers were required to prominently display information about the social/working conditions surrounding the production of particular imported goods at the POS.
Josh, this is a little out of date, but health care costs have only risen since: How the Payroll Dollar Is Spent
Wages 61.0 Cents
Medical benefits 11.0 Cents
Time not worked (vacation and other time off) 10.9 Cents
Legally required payments (payroll taxes, workers camp, unemployment, etc.) 8.2 Cents
Retirement and savings 8.0 Cents
Other benefits and costs 1.0 Cents
Total not exactly $1.00 because of rounding.
Source: U.S. Chamber of Commerce survey of employer practices in 2001.
I’m not sure how much retirement is tied to hours worked, and this also does not include recruitment, training, and management costs (as well as increased risks of various lawsuits from every new employee who you take a chance on). Taken together, the fixed overhead per employee can be significant, even though, as you say, there are also variable costs.
Here is a better explaination for short-term stickiness of wages:
Why Wages Do Not Fall in Recessions
“Why, then, are wages sticky? Mr Bewley concludes that employers resist pay cuts largely because the savings from lower wages are usually outweighed by the cost of denting workers’ morale: pay cuts hit workers’ standard of living and lower their self-esteem. Falling morale raises staff turnover and reduces productivity. Cheerier workers are more productive workers, not only because they work better, but also because they identify more closely with the company’s interests. This last point is crucial. Mr Bewley argues that monitoring workers’ performance is usually so tricky that firms rarely rely on coercion and financial carrots alone as motivators. In particular, high morale fosters teamwork and information-sharing, which are otherwise difficult to encourage. Firms typically prefer layoffs to pay cuts because they harm morale less, says Mr Bewley. Pay cuts hurt everybody and can cause festering resentment; layoffs hit morale only for a while, since the aggrieved have, after all, left. And whereas a generalised pay cut might make the best workers leave, and a selective one damage morale because it is seen as unfair, firms can often lay off their least competent staff.”
I agree with you that there is a big competitive disadvantage for US workers relative to workers in other countries with less worker protections (like China). More notifications might help with that, but in the end, US trade policy needs to reflect our political preferences (like with tarrifs to level the playing field regarding working conditions).
Guessing, given the discount contractors get at places like Home Depot, DIY may actually be more expensive for many projects where that discount exceeds the hourly labor costs (especially when you add in the cost of tools and mistakes). Although if the job displaces paid entertainment or other idle purchases, it might be a wash. 🙂 But this issue of contractors having cheaper access to materials, which may dominate the cost of many projects, is not very obvious. For example, if you have a plumber put in $300 worth of new faucets, and they get a 20% discount on them, that pays for an hour of plumbing right there compared to DIY, and they have the tools and will probably do the job quickly. Established plumbers are probably going to have better deals with suppliers too, etc.. So, the economics of someone new coming into an area and offering to do plumbing cheaply is going to be complicated by factors like these. Ultimately, as with many professions, what you end up paying for is things like trust or reduced uncertainty. Also, its not clear how elastic demand for some home repairs are; if your plumbing does not work, your house in not liveable. Like with some medical crisis involving clogged arteries, when your toilet doesn’t work, you probably can’t take the time to shop around much– even though both clogged arteries and clogged toilets can generally be prevented (a heart healthy diet for one, care in what you put in it and/or septic pumping for the other).
Again sticky wages are part of the issue. But so is the fact that many companies with strong market positions will just pocket the difference from cheaper or more productive labor. For most workers in the USA, real wages have essentially not risen in three decades in the USA (even as family income has risen from women moving into the workforce), while productivity per worker has more than doubled during that time. Who got those gains? University of Massachusetts Economics Professor Richard Wolff’s take on that:
Capitalism hits the fan.
Ultimately, I think you are right to point to limited demand, both for reasons of monetary policy and for reasons of cultural change. Companies are not going to hire people if they don’t think the demand is there, regardless of, say, hiring subsidies. As I mention in that document I organized, by the end of this decade, with population growth, the USA has to create about thirty million net new jobs (that is, jobs beyond replacements for newly lost jobs from automation) to see employment return to where is was in 2000. I just can’t imagine that happening given limited demand, foreign competition, increasing automation, better design, and voluntary social networks. That’s why I think this current curve discussed here is going to at best level off and more likely continue to rise (maybe with a dip here and there like with a few million green jobs). Since the US economic ideology is based mainly on a link between the right to consume and having a job, something has to give. Short term unemployment benefits have smoothed this over for a bit, but they were not designed as a lasting solution to fundamental structural change, nor do they cover the large number of independent contractors or small business people out there also hurt by the downturn. But in the long term, we probably will see lower plumber costs per job as unemployment persists or rises, but most of us may not be able to afford even that unless our society makes other changes (like a basic income) since our own labor will be worth less too and the cost of materials may remain high due to rent seeking (same as in the Great Depression, with warehouses full of goods people wanted, but no one able to buy them). Ultimately though, things like 3D printing and other DIY technology will break through that logjam (like in old days with seeds and farmers, a friend might print you a small 3D printer and some solar panels just to be neighborly), but by then our 21st century economy will not be recogizable by 20th century macroeconomic standards. What bothers me is that these issues are, for the most part, not on the table in our national policy discussions. Of course, I can just stand in line behind all the other people saying their pet issues (like serious banking regulation) are not on the table either. 🙂
Paul, if wages are 61% of payroll costs, then they are obviously much lower than 61% of the costs of a contracted service (which includes profit, advertising costs, transportation, etc.). Let’s call that lower percentage X. Say for example that X is 40%. At that level, a 5% change in wages takes X either up to 42% or down to 38%, so it is only a 2% change in the cost of the delivered service.
The lower the number X is, the more likely it becomes that changes in the cost of other factors have a greater effect on overall cost than changes in wage levels, even if wage levels remain the largest single component of the cost. That is, the lower the percentage of cost that is represented by wages, the less relevant changes in wage levels are likely to be in affecting levels of final demand for a given service (whether they are sticky or not). So in case of either a negative shift in the demand curve itself or an increase in other components of the price, it becomes harder to adjust wages to levels that result in the previous number of service contract agreements. Medical insurance costs, compliance costs, and, in some industries, profit margins, have risen over the last few decades.
I was trying to give a relatively clean example of the above via reference to the contract market for plumbing and carpentry services. It seems that you didn’t believe my example because, IMO, you have many misconceptions about that market. But rather than argue those details (e.g. explaining why contractor discounts on material are both small and generally available to everyone) it now seems more relevant to focus on the abstract point.
Josh, there well may be things I don’t understand about the plumbing services market (although, I might suggest knowing the ins and outs of discounts is something experienced people in an area know, even if they might in theory be accessible to everyone who knew them or had enough chutzpah).
Your leverage point provides yet another insight into the situation to the question you raise. As you suggest, let’s assume a 5% reduction in wages translated into a 2% reduction in retail price to a customer (given other fixed costs, though it might be a bit better since, as you say, there are other variable costs like FICA). Then a company could decide to, say, reduce the wages of its plumbers by 15% to reduce retail costs by 6%, which is a difference that most established customers might not really notice or even expect, and also it might make the company look desperate. On the other hand, you would, as Bewley’s point cited above, almost certainly have a lot of very unhappy plumbers on staff to manage now, with a related loss in morale, and so quality would suffer, and you might lose customers or have other costs to pay for rework. It’s more acceptable psychologically in an existing firm to just tell your staff that times are tough and give no raises, or to fire the least cost-effective performers performers (whether those with the least skills or those at the highest salaries).
However, this does suggest that there would be a long-term opportunity in a capitalist economy for new firms with lower costs to startup in the market (“Joe’s Discount Roboplumbing”? :-), whether due to low wages (new trainees, H1Bs, guest workers, illegal immigrants, teleoperated robot plumbers run from Mexico as in a recent fictional movie?) or different practices (house printing, new hand tools to deskill tasks?) or new supply sources (direct import of cheap plastic parts, like if Walmart supplies the plumbers as well as the parts?). Of course, starting a new business often requires access to capital like from loans, and those may be in short supply right now. Still, even if someone (Walmart?) is going to make another fortune in plumbing using teleoperated robots run by people in Mexico installing parts from China in US homes, it’s not going to be the average “Joe the Plumber” from the USA who makes that fortune. The existing plumbers paid high wages are just going to be put out of work and push that unemployment graph line above upward (and many will look for something like their old job for a couple years before becoming bitter and disheartened and maybe then voting for whatever demagogue says they will increase employment whatever it takes).
As Marshall Brain suggests, the days when most humans can supply valuable labor to the market place are numbered. We already see that now, with, say, US radiologists being replaced by radiologists from India (since the images and results can be sent over the internet). Eventually, we’ll see Indian radiologists displaced by AIs (where one practicioner and better software can do the work of several, until the AI analysis software just gets built into the X-ray machine itself and is just checked by the technician operating the device for routine situations).
Fixing plumbing in existing homes is harder because it is more unstructured and physical, but I predict we’ll continue to see better tools and materials lowering plumbing costs. And, like with those amazing teleoperated robot submarines doing plumbing a mile underwater in the Gulf of Mexico, maybe we’ll even even see teleoperation of robots in the USA by Mexicans to get around wage issues and limited robotic intelligence. Then there will be increasing automation as there is a push for one person in a low wage country to manage more teleoperated robots. The US military is pushing many robotic and AI technologies to minimze US military casualties in war(-of-choice) zones, so there are other dynamics too. (Note that wars-of-choice are both ways to increase employment and to, sadly, remove young people permanently from the labor pool.)
I was told a decade ago by a leading roboticist that the technology existed then to totally automate mining (including via teleoperation) so there were no people in mines, but this has been resisted by unions. Some existing unions may still be good at maintaining private welfare states inside corporations for their current members (who would, say, rather run the risk of being crushed by falling rock in a mine than face their spouses if they lost their jobs to a machine). But there are less and less union members as time goes by and new non-union businesses spring up (like new non-union auto manufacturing in the South displaced union manufacturing in Destroit) facilitated by the easy movement of goods and information these days. Local unions are becoming obsolete it seems; perhaps they will be replaced by national or global movements at some point?
Even the adjustment of other currencies relative to the US dollar won’t stop these trends related to permanent structural changes, since ultimately who would want to hire a person to do a job if a machine (or better design) can do the job without lawsuit risks or much management or health insurance costs? If the USA has not had China and India and Mexico, we might have seen robotics develop even faster. These are the sorts of trends that I feel will continue to effect that graph above. As Marshall Brain suggests, and many others before him, without substantial social changes that rethink an “income-through-jobs link”, it is only “make-work” at this point that will otherwise maintain full employment in our current system, given limited demand and rising productivity, Mainstream economic assumes unlimited demand, a flawed assumption IMHO given a law of diminishing returns for most goods and services and a desire for more materially affluent people to volunteer or develop non-material sides of their lives. Make-work can be things like more wars-of-choice, endless schooling, security theater, expanding our prisons, a focus on treatment of disease not prevention — essentially, just ramping up what we already have been doing in the USA for a couple decades to create jobs at the cost of a lot of needless human suffering IMHO. 🙁 A healthy USA, at peace with itself and the rest of the world, full of independent frugal self-starters learning on their own and helping their neighbors voluntarily — the kind of place the USA was more like some decades after 1776 for at least part of its non-slave population — that place would need very few paid jobs given today’s technology, which is an enormous amplifier of human ability. To me, that’s a big irony of the USA in the technologically advanced 21st century — conscously or unconsciously, we’ve chosen a path of make-work instead of happiness and enlightenmnent in order to deal with all the innovation our predecessors (like Benjamin Franklin) have gifted to us and which, ironically, has turned out to be a problem ideologically and economically. It’s very funny in that ironic sense when you think about it. 🙂