It seems to me…
“The first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency.” ~ Bill Gates.
The basic premise of several recent books such as The Second Machine Age[i] by Erik Brynjolfsson and Andrew McAfee, professors at MIT, and Capital in the Twenty-First Century[ii], by Thomas Piketty, a French economist who works on wealth and income inequality, is that we now are in a period of highly disruptive economic growth resulting primarily from the introduction of new technology. While the long-term social effects of this technology remain largely unstudied other than mostly for anecdotal predictions, publication of these texts and similar other recent articles have generated considerable topical interest.
Many writers on the subject either are apocalyptic pessimists or optimistically believe technology, especially artificial intelligence (AI), will solve all humanity’s problems. The doomsayers tend to believe technology will result in either a Player Piano[iii] dystopia or Terminator[iv] scenario where artificial intelligent agents exterminate humanity. The future lies somewhere between those optimistic and pessimistic extremes but very little research exists suggesting what course of action would provide maximum future social benefit. Some articles[v] compare the potential societal impact of technology to 1970s-era lack of understanding about atmospheric greenhouse gas increases and recommend multi-discipline research as to probable effects and recommended action to mitigate any possible negative potential. The only multi-discipline report even partially addressing this topic of which I am aware is a memorandum titled The Triple Revolution[vi] prepared in 1964 and sent to President Johnson. Though now quite out of date, its basic premise still remains correct.
It also is becoming increasingly difficult to isolate one aspect of technology from another. Robotics no longer is applicable only to manufacturing assembly lines as 3-D printing, the Internet of Things (IoT), cloud computing… will continue to blur any separation within the field. While robotics is performing an increasing number of structured tasks in manufacturing, what traditionally has been considered “computing” is similarly eliminating employment categories involving mostly repetitive tasks in clerical work and other white-collar occupations. Introduction of technology normally increases GDP and creates employment opportunities but many of the new positions are primarily available only to the technically skilled. (In some of my classes thirty years ago, I preferred to refer to this area of computing as ”cybernetics”, also including some aspects of operations research, but no longer believe even that label is adequate).
There has been considerable concern about off-shoring of manufacturing and other mid-skilled employment positions but a combination of increased transportation costs and 3-D printing has diminished many advantages of low labor costs, especially as technology results in overall employment reductions in those categories. Manufacturing-related positions might show short-term increases but overall sector employment will continue to significantly decrease in coming years. Likewise the IoT and cloud will tend to reduce even many mid-skill positions throughout all employment sectors.
The most significant manifestation of possible problems is wage stagnation for the typical worker. In the U.S., real wages, adjusted for cost of living, have negligibly changed in over four decades as capital increasingly is invested in automation rather than labor. This has resulted in owners of capital capturing an increasing share of income since the 1980s, while labor’s share has fallen[vii].
The primary problem with questions of this type, which most people miss, is that the future with which we are concerned has already arrived and everyone is looking at the world as they know it from past experience (Peter Drucker[viii]). Disruptive technological advancement in the past was constrained by the ability of human societies to adapt to that change; for people to reinvent themselves. Technological progress, which accounted for a long era of rapid productivity growth starting in the 19th century once again is accelerating as typified by advances that allow people to put a computer in their pocket not only more powerful than any in the world 20 years ago but also with considerably better software and access to useful data, people, and devices.
Throughout history, technological progress never has previously failed to generate new employment opportunities. While nothing should be any different now, increasing specialization that has successfully sustained past progress might not be as successful in a new economy increasingly limited by skill and educational attainment.
Since the 1980s when increases in the number of students completing a post-secondary education began to level off, workers have faced increasing competition from both machines and cheap emerging-market workers. As employment tasks become better understood, those tasks can more easily be broken into routine components, technology is then able to automate those components.
Many occupational sectors, especially those currently associated with high levels of education and high wages will survive. There now is an employment availability race between education and technology, many prospective employers report an unavailability of appropriately educated job applicants, but prices of non-widgets, most notably college education and health care, have substantially increased and no longer are affordable to many qualified students.
Assuming a positive long-term outlook with the potential for greater wealth and increased employment opportunities, adaptation to past waves of progress rested on political and policy responses. Past adaptation was dependent upon massive improvements in educational attainment including creation of institutions of universal secondary education and the subsequent rise in university attendance.
Now, successfully encouraging a higher percentage of college students to complete graduate-level work also is becoming increasingly difficult and expensive though innovative online education might increase this possibility. Unfortunately, not everyone has the intellectual capacity for even an undergraduate education. Humans have essentially reached their average maximum intelligence possible. Given that most people are not able to consistently perform creative tasks, there is a limit as to how beneficial encouraging increased educational attainment will be.
Additionally, while growth in areas of the economy not easily automated might provide jobs, that growth will not necessarily result in real wage increases. This has additional possible social implications.
The U.S. pioneering a hyper-unequal economic model in which the upper 1.0 percent of capital-owners and “super-managers” accumulate an increasing share of the national wealth. Development of the middle-class was a 20th-century innovation and an extremely important political and social development across the world. Following World War II, wealthier nation workers secured real prosperity and a large, property-owning middle class came to dominate politics. Current trends resulting in reduction of that middle class potentially could become increasingly politically antagonistic, unstable, and potentially dangerous. If economic inequality is permitted to grow indefinitely, class warfare becomes a distinct possibility. The so-called “Gini index[ix]”, the most commonly used measure of national economic inequality, is already in the U.S. within a range generally considered threatening.
After an era of very low taxes, the U.S. has considerable economic inequality and a high deficit. When this last occurred, the top tax rate soared and the economy boomed even with the maximum rate being high. And it would not be surprised to see history once again repeat itself…
Conservative politicians attempt to gain support by saying people who don’t work shouldn’t be able to eat. As long as right wing electorates blame the poor for their unemployment and ignore the broader economic context, any government attempt to supplement wages or support students will be politically unacceptable.
A so-called “reservation wage”, the wage below which a worker will not accept a job, currently is historically high. If governments refuse to allow jobless workers to fall too far below the average standard of living, then this reservation wage will rise steadily and ever more workers may find work unattractive. And the higher it rises, the greater the incentive to invest in capital that replaces labor.
U.S. society is too puritanical to ever acknowledge having a permanently unemployed class of citizen. We probably are evolutionarily wired to “earn” our way in life. Everyone has a basic need to feel they are important and contributing to the general benefit of society. It also should be assumed that no one will be satisfied with “make-work” employment. Any labor that can be done by a machine should be; substituting a human for what a machine can do is nothing less than a form of slavery and should be equally abhorrent.
From a basic Econ-101 aspect, there are many theoretical alternatives to the dilemma of future employment being dependent upon educational attainment. Realistically, few of those alternatives are practical without major social implications. Most inevitably result in economic inequality that at some point leads to protest and violence.
Much of the current economic consideration is limited to existing employment categories. What hopefully is possible is continuing employment expansion not necessarily in support of current labor opportunities but rather as the result of research and advancement in entirely new fields. While dependent upon highly skilled specialists, many low or mid-skilled workers would be required to translate, construct, test, and maintain new developing concepts into available products or services. If research funding is available, there probably never can be an oversupply of highly educated creative researchers and engineers. One of the unknown quantities is on average how many low-skilled workers will be required to provide the level of support necessary to sustain that research/development.
Regardless of future technological development, there is every reason to be optimistic. As Frédéric Bastiat[x] pointed out two centuries ago, there is no limit to the work that needs to be done or shortage of problems to be solved, the primary limitation is the number of available people prepared to work on them. What is necessary is the political will to address many of the problems facing us: maintaining our national infrastructure and recreation areas, rebuilding our blighted inner cities, restoration of our ecological environment…. The need is obvious; isn’t it time to begin?
That’s what I think, what about you?
[i] Brynjolfsson, Erik and Andrew McAfee. The Second Machine Age, WW Norton, http://books.wwnorton.com/books/The-Second-Machine-Age/, January 2014.
[ii] Piketty, Thomas. Capital in the Twenty-First Century, Harvard University Press, http://www.hup.harvard.edu/catalog.php?isbn=9780674430006, April 2014.
[iii] Vonnegut, Kurt. Player Piano, Dial Press, 1952.
[v] Levy, Frank, and Richard J. Murnane. Researching the Robot Revolution, Communications of the ACM, August 2014, pp33-35.
[vi] The Triple Revolution, Wikipedia, http://en.wikipedia.org/wiki/The_Triple_Revolution.
[vii] The Onrushing Wave, The Economist, http://www.economist.com/news/briefing/21594264-previous-technological-innovation-has-always-delivered-more-long-run-employment-not-less, 18 January 2014.
[x] Claude Frédéric Bastiat, Wikipedia, http://en.wikipedia.org/wiki/Fr%C3%A9d%C3%A9ric_Bastiat.