write a critique for other classmate
WHY ARE THERE SO MANY JOBS?
The Impact of Automation on Employment in the United States
Draft 1
by
Jashn Sardana
Economics 322, Economics Seminar
Professor McCain
Drexel University
Sardana 2
Introduction
“Everyone can enjoy a life of luxurious leisure if the machineproduced [robots] wealth is shared, or
most people can end up miserably poor if the machineowners successfully lobby against wealth redistribution.
So far, the trend seems to be toward the second option, with technology driving everincreasing inequality.”
― Stephen Hawking
In the last few decades, specifically after the end of World War 2, there has been a surge
of technological innovation in the World. Technological advances have not only made our lives
much easier, but have introduced the concept of automation in our workplace. Even though
workplace automation has made firms more efficient and profitable, it has done so by
eliminating a huge proportion of jobs, predominantly jobs which require a medium level of skill.
In addition to the number of jobs lost by automation, experts are also worried about automation
preventing the creation of new jobs. In the past, industries hired a lot more people in comparison
to lay offs, however this is not the case for newer industries. The newer industries seem to have a
smaller workforce however are able to produce a much higher level of output.
Argument: Technological advances are not always employmentincreasing or Paretoimproving
because of the following reasons:
1. Complementarity
2. Demand Elasticity
3. Labor Supply
Sardana 3
However some ‘technologically lagging’ industries appear strongly income elastic. Examples:
personal fitness, and the restaurant industry.
The Relationship between Automation and Unemployment
It may seem that automation and technological advances should have decreased the
demand of human labor, however this has not been the case in the past. In the 20th century, the
participation rate increased as more women started working. Figure 1 below shows the
increasing trend of the EmploymentPopulation Ratio from 1965 to 2000. Even though the
Financial Crisis of 2008 caused a deep drop in the Ratio, there has been an increasing trend,
which has only been possible due to increases in the demand of labor.
Figure 1, Source: BLS
Sardana 4
The EmploymentPopulation Ratio supports the notion that automation does not have a
negative correlation with employment. In addition, there is much evidence that automation has
not eliminated majority of jobs over the last few decades. It is true that automation has
substituted labor in numerous occupations however it has also complemented labor in numerous
occupations by raising output, productivity and wages. This substitution effect has been
disproportionate to the skill and income levels, causing a polarization of the labor market. The
top and bottom income and skill level workers have seen a rise in jobs however there has been
decline in jobs for the middle income and skill level jobs. This has one big implication; the
decline of the middle class. The decline of the middle class is creating a wealth inequality
problem which might create problems in the future. Figure 2 below shows the decline of the
middle income households between 1971 and 2015.
Sardana 5
Figure 2, Source: Pew research center
Employment Polarization
“Job polarization refers to the recent shrinking concentration of employment in
occupations in the middle of the skill distribution. Jobless recoveries refers to the slow rebound
in aggregate employment following recent recessions, despite recoveries in aggregate output.”.
(Source: National Bureau of Economic Research). The decline in the middle income and skill
jobs is because these are the jobs that are being replaced by automation. These jobs are more
‘routine tasks’ like processing a company’s payroll, tabulating the age distribution in a county
and other clerical tasks. This ‘routine tasks’ can be fully codified as they follow repetitive
procedures, hence they can be codified by the latest computer softwares and then performed by
computers. Higher income level jobs require skills such as flexibility, judgement, high reasoning,
logical deductions, creativity, intuition and language which cannot be automated. On the other
end of the income scale, lower wage occupations such as food preparation, serving, janitorial,
maintenance, hairdresser and security need a different type of skills that do not always require
high education such as physical skills, situational adaptability and communication which need to
be ‘manual’ because making them automated would decrease the quality of the work and would
not be efficient.
Sardana 6
Figure 3, Source: Katz and Margo (2014)
Figure 3 above shows the average change per decade in US occupational employment for
two periods: 19401980 and 1980 to 2010. We can see that professional/technical jobs that are
complemented by automation have been constantly growing since 1940’s, and operative/laborer
occupations have been constantly declining since the 1940’s. Agricultural occupations declined
vastly before 1980 because of a wide range of automation technology flooding the agriculture
industry. Even though there are new jobs being created as some jobs become automated, there is
Sardana 7
still a problem of wealth inequality that is arising. This is not only a problem in the US, but also
a problem in other countries. The 16 countries in EU show similar results, as shown in Figure 4
below.
Figure 4, Source: Goos, Manning, and Salomons (2014)
Wage Polarization
Employment Polarization and Wage Polarization are not always positively correlated. Just
because we have deduced the presence of Employment Polarization from the parabolic shape of
the figures above, it would be wrong to assume the presence of wage polarization from thouse
Sardana 8
figures. If the demand for highskill jobs were inelastic, gains in productivity would have an
effect on wages. Nonetheless all evidence points towards our hypothesis that automation has in
fact, encouraged productivity as demand has kept up. In case for low skilled jobs like janitors,
servers and drivers would not be affected by technology as automation does neither complement
nor substitute it. They might be weakly complemented by technology, for example, in the case of
Uber, customers can order a cab from their mobile phones and pay through their cell phones too.
As these occupations do not need training or education, these occupations are income elastic.
Even though automation has a strong positive correlation with employment polarization, it does
share the same correlation with wage polarization.
The Recent Slowdown in the Growth of HighSkill Occupations
Polanyi’s Paradox
“We can know more than we can tell... The skill of a driver cannot be replaced by a thorough
schooling in the theory of the motorcar; the knowledge I have of my own body differs altogether
from the knowledge of its physiology.” (Michael Polanyi, 1966)
Will it overcome?
Environmental Control
Machine Learning
Conclusions
Sardana 9
Bibliography
David H. Autor, “Why Are There Still So Many Jobs? The History and Future of
Workplace Automation”
Joel Mokyr, Chris Vickers, and Nicolas L. Ziebarth, “The History of Technological Anxiety and
the Future of Economic Growth: Is This Time Different?”
Gill A. Pratt, “Is a Cambrian Explosion Coming for Robotics?”