Employment Theory Needs Work

Frank Wisdom
August 15, 2014


…we have involved ourselves in a colossal muddle, having blundered in the control of a delicate machine, the working of which we do not understand.

– John Maynard Keynes, The Great Slump of 1930



Conventional Thinking About Unemployment

During the Great Depression, failure of prevailing economic theory to explain the persistent downturn led Keynes to develop a more sophisticated understanding of the economy. His General Theory of Employment, Interest, and Money (1936) explained that in such a serious slump increased government spending is needed to increase production and employment. Not only did Keynesian economics make sense, it was empirically supported by the downturn following premature austerity in 1937, and confirmed again by the recovery stimulated by World War II. How is it possible, given modern understanding of how to manage economic declines, that we again suffer from persistent unemployment? Didn’t we learn from Keynes how to fix that?

Economists used to assert confidently that nothing like the Great Depression could happen again. After all, we know far more than our great-grandfathers did about the causes of and cures for slumps, so how could we fail to do better?

– Paul Krugman, Knowledge Isn’t Power (2014)

There are two prevailing views about the adequacy of Keynesian economics to address today’s slow growth and unemployment. The conservative view rejects Keynesianism; the liberal view accepts it. Neither sees Keynesian economics the way Keynes saw classical economics, as a theory that works in some circumstances but needs major revision to cope with current events.

Conservatives think Keynesian economics fails to address today’s slow growth and unemployment because it is fundamentally wrong. Keynesian stimulus was tried in response to the Great Recession and didn’t work, they say, and the reason it didn’t fix the unemployment problem is that government spending does not create jobs:

Presidential candidate Mitt Romney:Government does not create jobs.

Republican National Committee Chairman Michael Steele:Not in the history of mankind has the government ever created a job.

Texas Governor Rick Perry:And the other thing this president [Obama] has done, he has proven for once and for all that government spending will not create one job. Keynesian policy and Keynesian theory is now done. We’ll never have to have that experiment on America again.

On the conservative view, high unemployment persists not because of insufficient government spending, but due to impairment of economic activity by excessive taxes and regulations, plus coddling of lazy people who subsist on handouts rather than working, the sort of people House Budget Committee Chairman Paul Ryan had in mind when he said that “we don’t want to turn the safety net into a hammock that lulls able-bodied people to lives of dependency and complacency, that drains them of their will and their incentive to make the most of their lives.

Like many dangerous misunderstandings, this conservative view gains plausibility from some underlying truths. There really are people who prefer to rely on disability benefits and other welfare support rather than work. Such parasitism may be more evident in European countries with more generous welfare systems, but is not entirely unknown in America. If the alternative were starvation, some fraction of currently unemployed people would be sufficiently motivated to get to work. In addition, taxation and regulation do interfere with economic growth and employment. For instance, calculation and payment of employment taxes, the need to adhere to employment regulations, and the possibility of being sued for alleged discrimination based on age, race, sex, religion, or national origin, are all burdens that discourage hiring. Such deterrents result in less employment whenever there’s a sufficiently attractive alternative such as outsourcing or automation.

Nevertheless, conservatives’ wholesale rejection of Keynesian economics is unjustified. Arguments against stimulus such as University of Chicago finance professor John Cochrane’s claim that “Every dollar of increased government spending must correspond to one less dollar of private spending.” are naive. As Krugman explains, increased government spending increases GDP. The higher GDP is associated with higher income, which generates higher tax revenue, so even if we rule out the option of printing more money, the added government spending need not be financed by subtraction of an equal amount of private spending. Moreover, higher income tends to increase private spending.

Suppose, for example, that the government agrees to spend an additional $100,000 to buy fish from an otherwise unemployed fisherman, who then catches and sells the fish (employing himself and increasing GDP), pays $10,000 in taxes on the income received, and spends $85,000 of the rest. Assuming the government covers its net payment to the fisherman by taking from other taxpayers an additional $90,000 which they otherwise would have spent, the private spending decrease corresponding to the added $100,000 of government spending will be only $5,000 ($90,000 – $85,000).

Thus increased government spending can increase employment. Despite comments to the contrary such as those of the Republican luminaries quoted above, even conservatives recognize increasing government spending can create jobs — or that reducing government spending can eliminate them, the other side of the same phenomenon — at least when the spending is military. Presidential candidate Mitt Romney’s campaign ads, for example, warned that looming defense spending cuts threatened the loss of hundreds of thousands of jobs.

On the liberal view, Keynes was correct but high unemployment persisted after the Great Recession because the stimulus was too small. In a replay of the misguided response to the Great Depression, the Great Recession was followed by an ill-timed drive to cut government spending. Seeking to balance the budget, conservatives demanded austerity, thereby ensuring spending was insufficient to get the economy back on track, either because they did not accept Keynesian economics or possibly because they wanted to make sure the Democrats would not get credit for economic recovery. The latter explanation may be too cynical, and may overestimate Republican intelligence, but it is striking how much less conservatives objected to deficit spending a few years earlier under a Republican administration.

Liberals understand that there is high unemployment when insufficient demand requires too few workers. A ratio of three people looking for work for each job opening, as was recently seen in America, makes it impossible for all the unemployed to get hired. Thus it appears it is not the fault of the unemployed that they are out of work, but the fault of the economy.

On behalf of the liberal view, it should be noted that the Great Recession was in fact followed by far less stimulus than Keynesians such as Krugman called for. Moreover, the stimulus did increase employment, though it is impossible to know precisely how much, for we cannot determine just how much worse the economy would have been without it. However, the fact that the liberal view is consistent with recent experience does not prove it is correct, any more than persistence of an illness following treatment with less than the prescribed dose of medicine guarantees the full dose would have cured it.


Historical Replacement of Lost Jobs

The history of agricultural employment illustrates how reduction of employment in one sector can be offset by growing employment elsewhere. From 1900 to 1960 the U.S. agricultural labor force declined from 11,680,000 to 5,970,000. Thus agricultural employment was cut nearly in half, while the population more than doubled. During this time the fraction of the American labor force engaged in farm work declined about 80%, from around 40 per hundred in 1900 to just over 8 per hundred in 1960. (It continued declining to about 2 per hundred by 2010.) The cause of the reduction in agricultural employment is no mystery. Advancing technology, including development of synthetic fertilizers and pesticides and internal combustion engine improvements, together with increasing availability and use of petroleum, enabled the industrialization of agriculture. Human workers and draft animals were increasingly replaced by machines. From 1900 to 1960, U.S. motor vehicle registrations increased from 8,000 to nearly 74,000,000, and American petroleum consumption increased from .2 quadrillion BTUs to 19.9 quadrillion BTUs. Fast growth in the use of petroleum-powered machinery transformed farming along with other parts of American life.


employment history needs work - farm employment

Data from Stanley Lebergott’s Labor Force and Employment, 1800-1960   


employment theory needs work - vehicle registrations

Data from Federal Highway Administration


employment theory needs work - oil consumption

Data from U.S. Energy Information Administration


Why didn’t the disappearance of millions of agricultural jobs create enormous unemployment of farmers or would-be farmers? The general answer is that rapid growth of the economy provided enough new work opportunities to absorb those freed from farming. The industrialization that reduced the need for agricultural employment created ample alternative work suitable for displaced farm workers. For instance, industrialists organized assembly lines in which humans worked with tools and machines to create motorized vehicles. In such work the worker is typically told what to do and repeatedly performs a defined set of routine tasks. In general these workers need at most moderate skill; the more important basis for their productivity is industriousness. Consequently, such industrial work offered a ready alternative to farm work. Former farmers and farmers’ children who lacked agricultural employment opportunities could easily be trained to fill assembly line positions.


Factors Facilitating Job Replacement

Many similar examples of industrial jobs replacing obsolete ones throughout the history of industrialization encourage taking such easy relocation of workers for granted, and make it possible to think such absorption of displaced workers is somehow guaranteed, so that there is no need to worry about advancing technology causing unemployment, however disconcerting it may be to those who lose their jobs.

Throughout history, aspiring Cassandras have regularly proclaimed that new waves of technological innovation would render huge numbers of workers idle, leading to all manner of economic, social and political disruption. … So far, of course, they’ve all been wrong. But that has not prevented a cascade of shrill new proclamations that — notwithstanding centuries of history — “this time is different”: The technology revolution will impair the livelihoods of millions of Americans. … Of course, I can’t prove that the impact of some new wave of technological innovation won’t ever upend thousands of years of history. But it hasn’t happened yet.

– Steven Rattner, Fear Not the Coming of the Robots, 2014


Assuming that this time is different is a common mistake. But assuming this time is not different can also be a mistake.

During the industrial era so far, absorption of workers who lost jobs due to new technology was facilitated by the following factors:

  1. Rapid growth of the population and economy created abundant new work opportunities.
  2. There were enormous resource rich new lands, including North America, South America, and Australia, free or nearly free for taking and exploiting. Lacking guns and immunity to European diseases, the natives scarcely hindered the European takeover (and after conquest could serve as an additional resource, exploited as slaves or low cost laborers). As recently as the first half of the 20th century, during the great shift away from agricultural employment, plentiful American resources were still essentially free. The Homestead Act offered free farmland, and the General Mining Law allowed Americans to claim and exploit valuable mineral deposits on vast federally owned territories.
  3. Major advances were made in exploitation of fossil fuels, starting with coal, then progressing to oil and gas.
  4. Technology advanced, enabling the harnessing of coal and oil to fuel advancing industrialization.
  5. Industrialization created jobs suitable for employment of those displaced from obsolete forms of work.

Which of these helpful factors still exist today?

  1. With fertility below the replacement rate in much of Europe and Russia, population is declining or growing very little in advanced countries. (America is a bit of an exception, with legal and illegal immigrants adding to its population.) World-wide, population growth is slowing, and must continue to do so as the number of people approaches the limit the planet can support — if it hasn’t already passed it. In developed countries, economic growth has slowed.
  2. There are no new continents to discover and exploit.
  3. Whether due to supply limitations or the need to limit global warming, fossil fuel use cannot continue to accelerate as in the past.
  4. Technology continues to advance, with industrialization transformed by creation of increasingly sophisticated computerized devices.
  5. Though advancing technology still creates new jobs, the new employment opportunities may be too few, and may require too much skill, to absorb all those let go from traditional jobs as demand for less skilled workers declines.

In sum, many factors that contributed powerfully to creation of new work opportunities in the past are now weaker or absent. So it would be unwise to take for granted that job creation will continue as before.


Combotification and Unemployment

In a statement last Wednesday — just hours after the government reported headline-grabbing economic growth of 4 percent in the second quarter — the Federal Reserve said it would continue to stimulate the economy because, despite overall growth, the labor market remained weak. …the employment report for July showed slower job growth, flat earnings, stagnant hours and stubbornly high long-term unemployment.

– New York Times, 8/1/2014, Growth Without Jobs

Basic economics suggests that the labor market will determine a clearing price for labor at which supply equals demand. Consequently there should be no unemployment, apart from a modest amount of “frictional” unemployment occurring while people are between old and new jobs, while new graduates seek their first jobs, and so on.

More refined analysis recognizes that there’s a lower limit below which the price of human labor can’t fall. The ultimate lower limit is the minimal cost of living for a human, essentially the cost of feeding, clothing, and sheltering a worker. In modern America, as in other advanced economies, this limit is buoyed by minimum wage laws and expectations for a modern standard of living, along with social support systems that make it possible to survive without working. (America’s agricultural workers are excluded from minimum wage and overtime pay requirements, so they can be more efficiently exploited.) People will not be employed if the value they produce is less than the minimum possible wage, and there is nothing in the laws of economics to guarantee that the value produced by a human’s work will exceed the minimum wage.

Imagine a maxbot is invented, a robotic machine that can do whatever a typically talented human worker can do. However improbable the development of such a device, there is nothing in the laws of economics to prevent it, the way the laws of physics rule out creation of a perpetual motion machine or a rocket that travels faster than light. Suppose that including the expenses of building and maintaining it, operating a maxbot costs somewhat less than hiring a human worker, say $6 per hour, a bit below America’s minimum wage of $7.25. Once maxbots become available, factories and other businesses will increasingly replace human workers with them. Even managers who consider employing humans more humane will be forced to switch to compete with lower cost rivals.

Consequently, unemployment will grow. As the maxbot inventor and investors grow richer selling maxbots, and other business owners profit from lower labor costs, there will be no effective mechanism to maintain the income of the unemployed. They will receive unemployment compensation for a while, along with food stamps and other welfare benefits, but their income will be far lower than when they had jobs, so they will buy less. Businesses catering to the wealthy may thrive, as may those targeting the poor, but enterprises selling mainly to middle class workers will suffer as their customer base declines. Fiscal stimulus will not create new jobs to make up for those lost, because the new demand created by added government spending will be satisfied by maxbots; humans will remain unemployed.

Though general purpose maxbots do not exist, there is a rapidly growing population of combots, automated systems (computers and robots) that replace unskilled or moderately skilled human labor. We’ve entered an era of combotification, in which human workers are increasingly replaced by machines that perform a growing range of tasks more sophisticated than machinery could perform in the past. It makes little difference that instead of a single type of machine that can do everything we have a wide variety of specialized machines.

For several years now, combots have been increasingly infiltrating and taking over factories. While America’s factory output increased a third in the decade prior to 2012, factory jobs decreased:

Factories have replaced millions of workers with machines. … In the 10 years ending in 2009, factories shed workers so fast that they erased almost all the gains of the previous 70 years; roughly one out of every three manufacturing jobs—about 6 million in total—disappeared.

– Adam Davidson, Making It in America, The Atlantic, Dec. 20, 2011

It might be thought that as traditional industrial jobs are outsourced or given to machines, humans who lose those jobs can find employment elsewhere, especially in the service sector. This happens to some extent, with many Americans working at organizations such as McDonalds and Walmart, perhaps so poorly paid they need welfare assistance to augment their meager earnings, but surviving. However, service jobs are also vulnerable. Consider a few examples that illustrate the wide range of activities now being automated:

Work for waiters is threatened as restaurants increasingly invite customers to order and pay without interacting with humans, using devices such as the tablet-based Ziosk. It was recently announced that the fast food chains Panera and Chili’s are adopting the similar iPad-based Presto.

The Presto aspires to be the food-services version of the airline check-in kiosk or the ATM or the self-checkout at your local pharmacy. It makes a person’s job a computer’s job, and that cuts costs. Each console goes for $100 per month. If a restaurant serves meals eight hours a day, seven days a week, it works out to 42 cents per hour per table—making the Presto cheaper than even the very cheapest waiter. Moreover, no manager needs to train it, replace it if it quits, or offer it sick days.

– Annie Lowrie, This Waiter Doesn’t Need a Tip


As the population ages, caring for the elderly seems a likely growth area for employment of humans, however poorly paid. But even here we find a trend toward replacing people with robots.

Caregiving is hard work. More often than not, it is tedious, awkwardly intimate and physically and emotionally exhausting. Sometimes it is dangerous or disgusting. Almost always it is 24/7 and unpaid or low wage, and has profound adverse health consequences for those who do it. It is women’s work and immigrants’ work, and it is work that many people either can’t or simply won’t do.

Many countries have acknowledged this reality by investing in robot development. … Already, we have robots to assist in surgery and very basic “walking” robots that deliver medications and other supplies in hospitals. Robots are increasingly used in rehabilitation after debilitating events like strokes. …search YouTube and you can watch developmentally delayed children doing therapy with a cute blue-and-yellow CosmoBot that also collects information about their performance. Or you can see older Japanese people with dementia smiling and chatting happily with a robot named Paro that looks like a baby seal and responds to human speech. … In the next decade, robot caregiver prototypes will become much more sophisticated. … Imagine this: Since the robot caregiver wouldn’t require sleep, it would always be alert and available in case of crisis. While my patient slept, the robot could do laundry and other household tasks. … At breakfast, the robot could chat with her about the weather or news.

– Louise Aronson, The Future of Robot Caregivers, New York Times, July 20, 2014


Assuming providing fresh milk to Americans requires that cows be kept in America, the milking of cows can’t be outsourced to China. But technology has advanced sufficiently to construct machines that can do such work, so given the costs and difficulties of using humans the transition to machines is under way:

The Bordens and other farmers say a major force is cutting labor costs — health insurance, room and board, overtime, and workers’ compensation insurance — particularly when immigration reform is stalled in Washington and dependable help is hard to procure. The machines also never complain about getting up early, working late or being kicked. “It’s tough to find people to do it well and show up on time,” said Tim Kurtz, who installed four robotic milkers last year at his farm in Berks County, Pa. “And you don’t have to worry about that with a robot.”

Regarding this automation, Tom Friedman commented:

Overnight, an average farmhand went from knowing how to milk a cow to having to learn how to program and operate the robotic cow-milker — to keep a job. That takes above-average skills.

He’s right: programming and maintaining milking machines takes more skill than milking cows. Also worth noting is that the number of humans needed to program and maintain the machines is far fewer than would be needed to do the milking manually. The new jobs created are too few in number, and require too much skill, to employ the displaced manual milkers.


Mismatch of Workers and Jobs

There are indeed new employment opportunities, but too few that are suitable for the unemployed seeking work. Available jobs and unemployed workers can be incompatible in multiple ways. Some jobs, such as crop picking, are too difficult for Americans, so they are in effect reserved for more desperate foreigners.

How can there be a labor shortage when nearly one out of every 11 people in the nation are unemployed? That’s the question John Harold asked himself last winter when he was trying to figure out how much help he would need to harvest the corn and onions on his 1,000-acre farm here in western Colorado. … Mr. Harold… has participated for about a decade in a federal program called H-2A that allows seasonal foreign workers into the country to make up the gap where willing and able American workers are few in number. He typically has brought in about 90 people from Mexico each year from July through October. This year, though, with tough times lingering and a big jump in the minimum wage under the program, to nearly $10.50 an  hour, Mr. Harold brought in only two-thirds of his usual contingent. The other positions, he figured, would be snapped up by jobless local residents wanting some extra summer cash. “It didn’t take me six hours to realize I’d made a heck of a mistake,” Mr. Harold said… Six hours was enough, between the 6 a.m. start time and noon lunch break, for the first wave of local workers to quit. Some simply never came back and gave no reason. Twenty-five of them said specifically, according to farm records, that the work was too hard. … “It is not an easy job,” said Kerry Mattics, 49, another H-2A farmer here in Olathe…. Still, Mr. Mattics said, he can’t help feeling that people have gotten soft.

– Kirk Johnson, Hiring Locally for Farm Work is No Cure-All, New York Times, October 5, 2011


When the available work is not too physically demanding, it may be too undesirable in some other way, or have non-skill requirements the unemployed cannot satisfy.

Truckers are in short supply, too. The median driver is 47 years old; about 100 retire every day. Many companies would put more trucks on the road if they could find more drivers. Some offer signing bonuses of $1,000 to $3,000. Making pay attractive is only half the battle; the other half is finding candidates without criminal records, who can pass drugs tests and are ready to spend most of their life on the move. This is harder than it sounds.

– Economist, No degree required, May 31, 2014

In many cases, the available work requires skills the unemployed lack. Thus in California’s Central Valley, high unemployment of unskilled workers coexists with unfilled demand for skilled ones.

Despite the high unemployment and willing workers available in the Central Valley, the agribusiness entrepreneurs Lynda and Stewart Resnick, with large operations involving growing, processing and distributing in that region, found that they could not find the skilled workers they needed.

Since 2010 the unemployment rate in Kern County, at the southern end of the Central Valley, has ranged between 10.3% and 17.9%. Despite that high unemployment, the county has had a shortage of workers sufficiently skilled to fill available jobs:

The numbers show that Kern County was one of the fastest growing economies in the state last year which translates to more industry moving into the county. Those new businesses are coming up short finding skilled workers… … “Typically they need folks that understand how to use computers, run machinery. It’s not a question of taking boxes and putting them on a conveyer belt, it’s a question of programing the conveyer belt,” said Richard Chapman, Kern Economic Development Corp.

That the unemployed lack skills needed for employment is supported by the correlation of unemployment with lower levels of education. According to the Bureau of Labor Statistics, in June, 2014 the unemployment rate for Americans over 25 years old with less than a high school degree was 9.1 percent, whereas for those with at least a bachelor’s degree it was 3.3 percent. Since the unemployment figures exclude those who have given up looking for work, it’s also worth looking at the percentage of the available labor force working. Only 39.4% of those with less than a high school degree were employed, compared to 72.7% of those with at least a bachelor’s degree.

In sum, not only are there fewer jobs available than Americans seeking employment, jobs the modern economy produces tend to require skills the unemployed lack. There are still lots of relatively unskilled jobs, but they tend to pay poorly, as is to be expected when there’s a surplus of low skilled workers relative to the number of low skill jobs. Neil Irwin recently noted that “wages of workers without advanced skills have been under downward pressure in the United States and across the developed world over the last generation. The deep recession and slow recovery have only made the trend more pronounced.” From 1979 to 2007, prior to the Great Recession, the real hourly wage for American men who had not completed high school declined 22%; for those with only a high school education it declined 10% (Mishel, Berstein, & Shierholz, The State of Working America, Table 3.16, p. 166).

Stimulating the economy can create more jobs, but the jobs created, at least the ones that pay well, are mostly for highly skilled workers. There is no guarantee that new work requiring only moderate skill will go to people rather than machines.


Distribution of Wealth

Combots, devices that do work formerly requiring humans, make it possible to produce goods and services more cheaply, enriching the relatively wealthy owners of the means of production, while decreasing the need for, and thus the economic value of, unskilled or moderately skilled human workers. Thus it is unsurprising that, as Joseph Stiglitz, Thomas Piketty, and others have noted, inequality in America and other countries is increasing. In the 3 years following the official end of the Great Recession, from 2009 – 2012, the 1% of Americans with the highest incomes captured 95% of the total increase in American income.

In Economic Possibilities for our Grandchildren (1930), Keynes looked forward to a time of abundance, when productivity has increased so much that there is little need for people to work:

We are being afflicted with a new disease of which some readers may not yet have heard the name, but of which they will hear a great deal in the years to come–namely, technological unemployment. This means unemployment due to our discovery of means of economising the use of labour outrunning the pace at which we can find new uses for labour.


But this is only a temporary phase of maladjustment. All this means in the long run that mankind is solving its economic problem. …


Thus for the first time since his creation man will be faced with his real, his permanent problem-how to use his freedom from pressing economic cares, how to occupy the leisure, which science and compound interest will have won for him, to live wisely and agreeably and well.

Missing here, as in many forecasts of abundant leisure resulting from growing productivity, is attention to the issue of who will get the benefit of increased productivity. There is no automatic mechanism to channel increased profits from the wealthy to the workers and unemployed. (Piketty’s proposal to tax wealth looks politically infeasible, and might also have practical problems.)

The rich customarily try to hold on to their wealth and increase it. Prior to the French revolution, French nobles and clergy were exempt from taxation, much as in America the tax system has been structured so the wealthy can pay at a lower rate than middle class workers. Thirty corporations with U.S. profits from 2008 to 2010 totaling $160 billion arranged to pay no federal income tax for those years. When life for the poor in 18th century France became unbearable, inequality was addressed by revolution. In modern America, however, the situation of the poor is less intolerable. So long as they have food and entertainment, impoverished Americans might not revolt. There might instead be an extended era of growing inequality, in which the rich get richer and those in the middle and lower classes survive. High unemployment and a lot of poorly paid employment may persist, the economy of the poor limping along while the economy of the rich drives forward.

In the first half of 2014, the European economy as a whole grew very little, at an annual rate less than 1%.

And now growth has stalled…. It’s really a dismal picture. … Europe will arguably be lucky if all it experiences is one lost decade.
– Krugman, The Forever Slump, Aug. 15, 2014

Southern Europe continued to struggle. In August, 2014, shortly after Italy was recognized to have slipped back into recession, an Economist report on overpaid Italian government workers noted that “more than a decade of economic stagnation has reduced real GDP per person to the level of 1998.” Meanwhile, luxury car sales soared:

Global sales of luxury Rolls-Royce cars, which cost hundreds of thousands of dollars, soared 33 percent to 1,968 vehicles during the first half [of 2014]. The Britain-based manufacturer said that growth was strongest in Europe, up over 60 percent, while sales rose almost 40 percent in the Asia-Pacific region.

Inflation may remain moderate, given low production costs and limited demand for goods consumed by the masses. Meanwhile, the prices of things in limited supply that wealthy people want, like real estate in desirable locations, can be expected to rise. As of 2014 house prices were increasing about 15% per year in San Francisco, and about 19% in London.


Employment Prospects for the Less Skilled

The 20th century replacement of farm work with industrial work is a good example of a recurring phenomenon during industrialization. The repeated relocation of surplus traditional workers into industrial jobs may seem at first glance to support the assumption that it will always be so, that the creative destruction of economic development can be counted on to provide new jobs to replace those lost. But is that assumption justified?

We saw above that some factors that helped provide new jobs in earlier times can no longer be relied on. The current technological transformation differs significantly from past ones, involving combotification rather than traditional industrialization. New jobs created now require far more skill than industrial jobs created in the past. Though a great deal of less skilled work still needs to be done, much of it can be outsourced or performed cheaply with machines, resulting in a shortage of less skilled job opportunities relative to the number of unemployed, and in low pay for such work.

To see how a shortage of suitable new employment for displaced workers might play out, consider what happened to some less adaptable workers who lost jobs on farms during the industrialization of agriculture. Like the humans who supervised them, horses and mules also faced diminishing farm employment. Unlike humans, however, the horses and mules were not suited for the new factory jobs; nor, for the most part, were they suited for any other jobs on offer. Some were employed in racing or recreational riding, but such jobs were a small fraction of those lost on farms.

Another important difference worth noting is that in the case of horses and mules, the bosses control the workers’ breeding. Moreover, the lifespan of these animals is relatively short. So there’s a simple solution to horse unemployment: limit production of new horses to the number needed for suitable available work. Of course the number of unemployed horses can also be easily reduced by sending those who are unemployed to the slaughterhouse. Thus unemployed horses for the most part do not exist (wild horses are a minor exception). In other words, the horse population closely matches horse employment. From 1900 to 1960, the American horse population, and horse employment, decreased about 86%.


employment theory needs work - equine population
Data from The Demographics of the U.S. Equine Population


In the case of horses, the economy failed to produce new employment opportunities for those whose traditional jobs were made obsolete by advancing technology. New jobs were available, but equine workers lacked the skills required to fill them, and training to overcome the skill deficit was not feasible.

Today the situation of ordinary humans is in a crucial respect the same: the satisfactory new jobs created — the ones that pay reasonably well — require skills ordinary people lack. Therefore the Keynesian prescription for achieving full employment won’t work, not only because of unwillingness to undertake sufficient economic stimulus, but also because stimulus will not create enough jobs for people with average skills.

The solution, if there is to be one, must be educational and political as well as economic. Training people to improve their skills can help, but unless education becomes far more effective such efforts will fall short. In America, at least, education is not making the population highly skilled.

In addition to misunderstanding math, American students also, on average, write weakly, read poorly, think unscientifically and grasp history only superficially.

– Elizabeth Green, Why Do Americans Stink at Math?, NYT, July 27, 2014

On the political front, one can imagine radical measures such as shifting from taxing the employment of people (which of course discourages employing them) to taxing the employment of machines, or raising taxes on the wealthy to subsidize employment of the poor. But such schemes are at present fantasies with no prospect of implementation.

Today it is no more guaranteed that the economy will provide jobs for all the unemployed humans than it was guaranteed in the prior century that the economy would produce jobs for all the unemployed horses. The sooner people recognize the profound change in suitable job creation between earlier phases of industrialization and our new era of combotification, the sooner they can begin to think effectively about how to address unemployment.

© Frank Wisdom, 2014