The Race between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment

TL;DR: In this paper, a task-based framework is proposed to characterize the equilibrium in a dynamic setting where tasks previously performed by labor can be automated and more complex versions of existing tasks, in which labor has a comparative advantage, can be created.

Abstract: The advent of automation and the simultaneous decline in the labor share and employment among advanced economies raise concerns that labor will be marginalized and made redundant by new technologies. We examine this proposition using a task-based framework in which tasks previously performed by labor can be automated and more complex versions of existing tasks, in which labor has a comparative advantage, can be created. We characterize the equilibrium in this model and establish how the available technologies and the choices of firms between producing with capital or labor determine factor prices and the allocation of factors to tasks. In a static version of our model where capital is fixed and technology is exogenous, automation reduces employment and the share of labor in national income and may even reduce wages, while the creation of more complex tasks has the opposite effects. Our full model endogenizes capital accumulation and the direction of research towards automation and the creation of new complex tasks. Under reasonable conditions, there exists a stable balanced growth path in which the two types of innovations go hand-in-hand. An increase in automation reduces the cost of producing using labor, and thus discourages further automation and encourages the faster creation of new complex tasks. The endogenous response of technology restores the labor share and employment back to their initial level. Although the economy contains powerful self correcting forces, the equilibrium generates too much automation. Finally, we extend the model to include workers of different skills. We find that inequality increases during transitions, but the self-correcting forces in our model also limit the increase in inequality over the long-run.

Summary

1 Introduction

• The accelerated automation of tasks performed by labor raises concerns that new technologies will make labor redundant (e.g., Brynjolfsson and McAfee, 2012, Akst, 2014, Autor, 2015).
• Automation allows firms to substitute capital for tasks previously performed by labor, while the creation of new tasks enables the replacement of old tasks by new variants in which labor has a higher productivity.
• The results in this case are similar, but the conditions for uniqueness and stability of the balanced growth path are more demanding.
• 4 equilibrium incorporating capital accumulation and directed technological change, but also because tasks are combined with a general elasticity of substitution, and because the equilibrium allocation of tasks critically depends both on factor prices and the state of technology.

2.1 Environment

• All tasks and the final good are produced competitively.
• A new (more complex) task replaces or upgrades the lowest-index task.
• Each task is produced by combining labor or capital with a task-specific intermediate q(i), which embodies the technology used either for automation or for production with labor.
• In Section 4 the authors relax this assumption and allow intermediate producers to make profits so as generate endogenous incentives for innovation.

2.2 Equilibrium in the Static Model

• These two special cases ensure that the demand for labor and capital is homothetic.
• The unit cost of production for tasks i ≤ I, on the other hand, depends on min { R, Wγ(i) } reflecting the fact that capital and labor are perfect substitutes in the production of automated tasks.
• In the static model, this will be the case when the capital stock is not too large, which is imposed in the next assumption.
• An increase in I∗—which corresponds to greater equilibrium automation—increases the share of capital and reduces the share of labor in this aggregate production function, while the creation of new tasks does the opposite.
• 14The increasing labor supply relationship, (11), ensures that the labor share sL = WL RK+WL is increasing in ω.

3 Dynamics and Balanced Growth

• The authors then investigate the conditions under which the economy admits a balanced growth path (BGP), where aggregate output, the capital stock and wages grow at a constant rate.
• The authors conclude by discussing the long-run effects of automation on wages, the labor share and employment.

3.2 Long-Run Comparative Statics

• The authors next study the log-run implications of an unanticipated and permanent decline in n(t), which corresponds to automation running ahead of the creation of new tasks.
• Because in the short run capital is fixed, the short-run implications of this change in technology are the same as in their static analysis in the previous section.
• Moreover, the asymptotic values for employment and the labor share are increasing in n.
• The dotted line depicts the case where wI(n) is large relative to R, so that there are significant productivity gains from automation.
• In contrast to the concerns that highly productive automation technologies will reduce the wage and employment, their model thus shows that it is precisely when automation fails to raise productivity significantly that it has a more detrimental impact on wages and employment.

4.1 Endogenous and Directed Technological Change

• This ensures that the unique equilibrium price for all types of intermediates is a limit price of ψ, and yields a per unit profit of (1−µ)ψ > 0 for technology monopolists.
• These profits generate 21 incentives for creating new tasks and automation technologies.
• The authors also assume that this compensation takes place with the new inventors making a take-it-or-leave-it offer to the holder of the existing patent.
• Developing new intermediates that embody technology requires scientists.
• For notational convenience, the authors also adopt the normalization G(0) = κNκI+κN .

4.2 Equilibrium with Endogenous Technological Change

• The authors first compute the present discounted value accruing to monopolists from automation and the creation of new tasks.
• To simplify the exposition, let us assume that in this equilibrium n(t) > max{n̄(ρ), ñ(ρ)}(ρ), so that I∗(t) = I(t) and newly-automated tasks start being produced with capital immediately.
• The condition σ̂ > ζ guarantees that the former, positive effect dominates, so that prospective technology monopolists have an incentive to introduce technologies that allow firms to produce tasks more cheaply.
• Proposition 6 also shows that, for κIκN > κ, the unique interior BGP is globally stable provided that the intertemporal elasticity of substitution is infinite (i.e., θ = 0), and locally stable otherwise (i.e., when θ > 0).
• In summary, Proposition 6 characterizes the varieties of BGPs, and together with Corollary 2, it delineates the types of changes in technology that trigger self-correcting dynamics.

5 Extensions

• In this section the authors discuss three extensions.
• First the authors introduce heterogeneous skills, which allow us to analyze the impact of technological changes on inequality.
• Second, the authors study a different structure of intellectual property rights that introduces the creative destruction of profits.
• Finally, the authors discuss the welfare implications of their model.

5.1 Automation, New Tasks and Inequality

• To study how automation and the creation of new tasks impact inequality, the authors now introduce heterogeneous skills.
• This extension is motivated by the observation that both automation and new tasks could increase inequality: new tasks favor high-skill workers who tend to have a comparative advantage in complex tasks, while automation substitutes capital for labor in lower-indexed tasks where low-skill workers have their comparative advantage.
• When ξ < 1, as the time during which a task has existed tends to infinity, the productivity of low-skill workers relative to high-skill workers converges to γL(i, t)/γH(i) = γH(i) ξ−1, and limits to zero as more and more advanced tasks are introduced.
• If ξ = 1, in the unique BGPWH(t) and WL(t) grow at the same rate as the economy, the wage gap, WH(t)/WL(t), remains constant, and capital, low-skill and high-skill workers perform constant shares of tasks.

5.2 Creative Destruction of Profits

• The authors modify their baseline assumption on intellectual property rights and revert to the classical setup in the literature in which new technologies do not infringe the patents of the products that they replace (Aghion and Howitt, 1992, and Grossman and Helpman, 1991).
• This assumption introduces the creative destruction effects—the destruction of profits of previous inventors by new innovators.
• Here πI(t, i) and πN (t, i) denote the flow profits from automating and creating new tasks, respectively, which are given by the formulas in equations (23) and (24).
• The next proposition focuses on interior BGPs and shows that, because of creative destruction, the authors must impose additional assumptions on the function ι(n) to guarantee stability.
• In their baseline model, the key force ensuring stability is that incentives to automate are shaped by the cost difference between producing a task with capital or with labor—by lowering the effective wage at the next tasks to be automated, current automation reduces the incremental value of additional automation.

5.3 Welfare

• The authors study welfare from two complementary perspectives.
• In particular, suppose that there exists an upward-sloping quasi–labor supply schedule, Lqs(ω), which constrains the level of employment, so that L ≤ Lqs(ω).
• Crucially, the reduction in employment resulting from automation now has a negative impact on welfare, and this negative effect can exceed the positive impact following from the productivity gains, turning automation, on net, into a negative for welfare.
• Interestingly, new tasks increase welfare even more than before, because they not only raise productivity but also expand employment, and by the same logic, the increase in labor supply has a welfare benefit for the workers (since they were previously constrained in their employment).

6 Conclusion

• As automation, robotics and AI technologies are advancing rapidly, concerns that new technologies will render labor redundant have intensified.
• In their model, this takes the form of the introduction of new, more complex versions of existing tasks, and it is assume that labor has a comparative advantage in these new tasks.
• When the long-run rental rate of capital is not so low relative to labor, their framework generates a BGP in which both types of innovation go handin-hand.
• The authors consider their paper to be a first step towards a systematic investigation of different types of technological changes that impact capital and labor differentially.
• Third, in this paper the authors have focused on the creation of new labor-intensive tasks as the type of technological change that complements labor and plays a countervailing role against automation.

Figures

Figure 2: The task space and a representation of the effect of introducing new tasks (middle panel) and automating existing tasks (bottom panel).

Figure 5: Dynamic equilibrium when technology is exogenous and satisfies n(t) = n and g(t) = A∆.

Figure 8: Asymptotic behavior of normalized values. In the left panel, κIvI(n) is everywhere above κNvN (n), and thus there is no effort devoted to creating new tasks in the BGP involves full automation. In the right panel, if κI/κN is sufficiently large, the two curves intersect and we have an interior BGP with both automation and creation of new tasks. The right panel also shows the effect of an increase in the productivity of scientists in automating tasks from κI to κ ′ I .

Figure 6: Dynamic behavior of wages (lnW ), the rental rate of capital (R), the labor share (sL), and the capital stock following a permanent increase in automation.

Figure 7: Varieties of BGPs.

Figure 3: Static equilibrium. The left panel depicts the case in which I∗ = I < Ĩ so that the allocation of factors is constrained by technology. The right panel depicts the case in which I∗ = Ĩ < I so that the allocation of factors is not constrained by technology and is cost-minimizing. The blue curves show the shifts following an increase in I to I ′, which reduce ω in the left panel, but have no effect in the right panel.

Figure 1: Employment growth by decade plotted against the share of new job titles at the beginning of each decade for 330 occupations. Data from 1980 to 1990 (in dark blue), 1990 to 2000 (in blue) and 2000 to 2007 (in light blue, re-scaled to a 10-year change). Data source: See Appendix B.

Figure 4: Behavior of factor prices in different parts of the parameter space.

Figure 9: Left panel: Average years of schooling among employees against the share of new job titles at the beginning of each decade for 330 occupations. Right panel: Change in average years of schooling over the next 10 years (dark blue), next 20 years (blue) and next 30 years (in light blue) against the share of new job titles at the beginning of each decade. See Appendix B for data sources and detailed definitions.

Full text

NBER WORKING PAPER SERIES
THE RACE BETWEEN MACHINE AND MAN:
IMPLICATIONS OF TECHNOLOGY FOR GROWTH, FACTOR SHARES AND EMPLOYMENT
Daron Acemoglu
Pascual Restrepo
Working Paper 22252
http://www.nber.org/papers/w22252
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
May 2016, Revised June 2017
We thank Philippe Aghion, Mark Aguiar, David Autor, Erik Brynjolfsson, Chad Jones, John Van
Reenen, and participants at various conferences and seminars for useful comments and
suggestions. We gratefully acknowledge financial support from the Bradley Foundation and the
Toulouse Network on Information Technology. Restrepo thanks the Cowles Foundation and the
Yale Economics Department for their hospitality. The views expressed herein are those of the
authors and do not necessarily reflect the views of the National Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been
peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies
official NBER publications.
© 2016 by Daron Acemoglu and Pascual Restrepo. All rights reserved. Short sections of text, not
to exceed two paragraphs, may be quoted without explicit permission provided that full credit,
including © notice, is given to the source.

The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares
and Employment
Daron Acemoglu and Pascual Restrepo
NBER Working Paper No. 22252
May 2016, Revised June 2017
JEL No. J23,J24,O14,O31,O33
ABSTRACT
We examine the concerns that new technologies will render labor redundant in a framework in
which tasks previously performed by labor can be automated and new versions of existing tasks,
in which labor has a comparative advantage, can be created. In a static version where capital is
fixed and technology is exogenous, automation reduces employment and the labor share, and may
even reduce wages, while the creation of new tasks has the opposite effects. Our full model
endogenizes capital accumulation and the direction of research towards automation and the
creation of new tasks. If the long-run rental rate of capital relative to the wage is sufficiently low,
the long-run equilibrium involves automation of all tasks. Otherwise, there exists a stable
balanced growth path in which the two types of innovations go hand-in-hand. Stability is a
consequence of the fact that automation reduces the cost of producing using labor, and thus
discourages further automation and encourages the creation of new tasks. In an extension with
heterogeneous skills, we show that inequality increases during transitions driven both by faster
automation and introduction of new tasks, and characterize the conditions under which inequality
is increasing or stable in the long run.
Daron Acemoglu
Department of Economics, E52-446
MIT
77 Massachusetts Avenue
Cambridge, MA 02139
and CIFAR
and also NBER
daron@mit.edu
Pascual Restrepo
Department of Economics
Boston University
270 Bay State Rd
Boston, MA 02215
and Cowles Foundation, Yale
pascual.restrepo@yale.edu

1 Introduc tion
The accelerated automation of tasks performed by labor raises concerns that new technologies will
make labor redundant (e.g., Brynjolfsson and McAfee, 2012, Akst, 2014, Autor, 2015). The recent
declines in the labor share in national income an d the employment to population ratio in the U.S.
(e.g., Karabarbounis and Neiman, 2014, an d Oberfield and Raval, 2014) are often interpreted as
supporting evidence for the claims that, as digital technologies, robotics and artificial intelligence
penetrate the economy, workers will find it increasingly difficult to compete against machines, and
their compensation will experience a relative or even absolute decline. Yet, we lack a comprehen sive
framework incorporating such effects, as well as potential countervailing forces.
The need for such a framework stems not only from the importance of understanding how and
when automation will transform the labor market, but also fr om the fact that similar claims have
been made, but have not always come true, about previous waves of new technologies. Keynes
famously foresaw the steady increase in per capita income during the 20th centu ry from the in-
troduction of new technologies, bu t incorrectly predicted that this would create widespread tech-
nological unemployment as machines replaced human labor (Keynes, 1930). In 1965, economic
historian Robert Heilbroner confid ently stated that “as machines continue to invade society, du-
plicating greater and greater nu mbers of social tasks, it is human labor itself—at least, as we now
think of ‘labor’—that is gradually rendered redundant”(quoted in Akst, 2014). Wassily Leontief
was equally pessimistic about th e implications of new machines. By drawing an analogy with the
technologies of the early 20th century that made horses redundant, in an interview he speculated
that “Lab or will become less and less important. . . More and more workers will b e replaced by
machines. I do not see that new industries can employ everybody who wants a job”(The New York
Times, 1983).
This paper is a first step in developing a conceptual framework to study how mach ines re-
place human labor and why this might (or might not) lead to lower employment and stagnant
wages. Our main conceptual innovation is to propose a unifi ed framework in which tasks pr evi-
ously per formed by labor are automated, while at the same time other new technologies complement
labor—specifically, in our model this takes the form of the introduction of new tasks in which labor
has a comparative advantage. Herein lies our answer to Leontief’s analogy: the difference between
hum an labor and horses is that humans have a comparative advantage in new and more complex
tasks. Horses did not. If this comparative advantage is sufficiently important and the creation of
new tasks continues, employment and the labor share can remain stable in the long run even in th e
face of rapid automation.
The importance of these new tasks is well illustrated by the technological and organizational
changes during the Second Industrial R evolution, which not only involved the replacement of the
stagecoach by the railroad, sailboats by steamboats, and of manual dock workers by cranes, but also
the creation of new labor-intensive tasks. These tasks generated jobs for a new class of engineers,
machinists, repairmen, conductors, back-office workers and managers involved with the introduction
1

and operation of new technologies (e.g., Landes, 1969, Chandler, 1977, and Mokyr, 1990).
Today, while industr ial robots, digital technologies and computer-controlled machines replace la-
bor, we are again witnessing the emergence of new tasks ranging from engineering and programming
functions to those performed by audio-visual specialists, executive assistants, data administrators
and analysts, meeting planners and computer support specialists. Indeed, during the last 30 years,
new tasks and new job titles account for a large fraction of U.S. employment growth. To document
this fact, we us e data from Lin (2011) to measure the sh are of new job titles—in which workers per-
form newer tasks than those employed in more trad itional jobs—within each occupation. In 2000,
about 70% of computer software developers (an occupation employin g one million people at the
time) held new job titles. Sim ilarly, in 1990 radiology technician and in 1980 management analyst
were new job titles. Figure 1 shows that for each decade s ince 1980, employment growth has been
greater in occupations with more new job titles. The regression line indicates that occupations with
10 percentage points more new job titles at the beginning of each decade grow 5.05% faster over
the next 10 years (standard error=1.3%). From 1980 to 2007, total employment in the U.S. grew
by 17.5%. About half (8.84%) of th is growth is explained by the additional emp loyment growth in
occupations with new job titles relative to a benchmark category with no new job titles.
1
-200 -150 -100 -50 0 50 100 150 200
Percent change in employment growth by decade
0 20 40 60 80
Share of new job titles at the beginning of each decade
From 1980 to 1990 From 1990 to 2000 From 2000 to 2007
Figure 1: Employment growth by decade plotted against the share of new job titles at the beginning of each decade
for 330 occupations. Data from 1980 to 1990 (in dark blue), 1990 to 2000 (in blue) and 2000 to 2007 (in light blue,
re-scaled to a 10-year change). Data source: S ee Appendix B.
We start with a static model in which capital is fixed and technology is exogenous. There are
two types of technological chan ges: the automation of existing tasks and the introduction of new
tasks in wh ich labor has a comparative advantage. Our static model provides a rich but tractable
framework to study how automation and the creation of new tasks impact factor prices, factor
1
T
he data for 1980, 1990 and 2000 are from the U.S. Census. The data for 2007 are from t he American Community
Survey. Additional information on the data and our sample is provided in Appendix B, where we also document in
detail the robustness of the relationship depicted in Figure 1.
2

shares in national income and employment. Automation allows firms to substitute capital for tasks
previously performed by labor, while the creation of new tasks enables the replacement of old tasks
by new variants in which labor has a higher prod uctivity. In contrast to the more commonly-used
models featuring factor-augmenting technologies, here au tomation always reduces the labor share
and employment, and may even reduce wages. Conversely, th e creation of new tasks increases
wages, employment and the labor share. These comparative statics follow because factor prices are
determined by the range of tasks performed by capital and labor, and exogenous shifts in technology
alter the range of tasks performed by each factor (see also Acemoglu and Autor, 2011).
We then embed this framework in a dynamic economy in which capital accumulation is en-
dogenous, and we characterize restrictions under which the model delivers balanced growth with
automation and creation of new tasks—which we take to be a good appr oximation to economic
growth in the United States and the United Kingdom over the last two centuries. The key restric-
tions are that there is exponential prod uctivity growth from the creation of new tasks and that
the two types of technological changes—automation and the creation of new tasks—advance at
equal rates. A critical difference from our static model is that capital accumulation responds to
permanent shifts in technology in order to keep the interest rate and hence the rental rate of capital
constant. As a r esult, the dynamic effects of technology on factor prices depend on the response of
capital accumulation as well. The response of capital ensures th at the productivity gains from both
automation and th e introduction of new tasks fully accrue to labor (the relatively inelastic factor).
Although the real wage in the long run increases because of this productivity effect, automation
always red uces the labor share.
Our full model endogenizes the rates of improvement of these two types of technologies by mar-
rying our task-based framework with a directed technological change setup. This full version of the
model remains tractable and allows a complete characterization of balanced growth paths. If the
long-run r ental rate of capital is very low relative to the wage, there will not be sufficient incen-
tives to create new tasks, and the long-ru n equilibrium involves f ull automation—akin to Leontief’s
“horse equilibrium”. Otherwise, however, the long-run equilibr ium involves balanced growth based
on equal advancement of the two types of technologies. Under natural assumptions, this (inte-
rior) balanced growth path is stable, so that when automation runs ah ead of the creation of n ew
tasks, market forces induce a slowdown in subsequent automation and more rapid countervailing
advances in the creation of new tasks. This stability result highlights a crucial new force: a wave
of automation pushes down the effective cost of producing with labor, discour aging further efforts
to automate ad ditional tasks and encouraging the creation of new tasks.
The stability of the balanced growth path implies that periods in which automation runs ahead
of the creation of new tasks tend to trigger self-correcting forces, and as a result, the labor share
and employment stabilize an d may even return to their initial levels. Whether or not this is the
case depends on the reason why automation paced ahead in the first place. If this is caused by
the random arrival of a series of automation technologies, the long-run equilibrium takes us b ack
to the same initial levels of employment and labor share. If, on the other hand, automation surges
3

Frequently asked questions

Q1. What is the key feature of the task-based framework?

The fact that automation may increase productivity while simultaneously reducing wages is a key feature of the task-based framework developed here. 

Q2. What are the contributions in "Nber working paper series the race between machine and man: implications of technology for growth, factor shares and employment" ?

The authors examine the concerns that new technologies will render labor redundant in a framework in which tasks previously performed by labor can be automated and new versions of existing tasks, in which labor has a comparative advantage, can be created. In an extension with heterogeneous skills, the authors show that inequality increases during transitions driven both by faster automation and introduction of new tasks, and characterize the conditions under which inequality is increasing or stable in the long run. Stability is a consequence of the fact that automation reduces the cost of producing using labor, and thus discourages further automation and encourages the creation of new tasks. 

Q3. What are the future works in "Nber working paper series the race between machine and man: implications of technology for growth, factor shares and employment" ?

Incorporating the possibility of such “ middling tasks ” being automated is an important generalization, though ensuring a pattern of productivity growth consistent with balanced growth is more challenging. Second, there may be technological barriers to the automation of certain tasks and the creation of new tasks across industries ( e. g., Polanyi, 1966, Autor, Levy and Murnane, 2003 ). 

Q4. What is the function that determines whether this effect is complete or incomplete?

since the function Γ limits to 1 over time, the parameter ξ determines whether this standardization effect is complete or incomplete. 

Q5. What is the source of non-homotheticity in the general model?

The source of non-homotheticity in the general model is the substitution between factors (capital or labor) and intermediates (the q(i)’s). 

Q6. What is the first-order negative effect of automation on welfare?

But now, because workers are constrained in their labor supply choices, the lower employment that results from automation has a first-order negative effect on their welfare. 

Q7. What are the new tasks that are being replaced by labor?

while industrial robots, digital technologies and computer-controlled machines replace labor, the authors are again witnessing the emergence of new tasks ranging from engineering and programming functions to those performed by audio-visual specialists, executive assistants, data administrators and analysts, meeting planners and computer support specialists. 

Q8. What is the effect of rents on the labor supply relationship?

If the elastic labor supply relationship results from rents (so that there is a wedge between the wage and the opportunity cost of labor), there is an important new distortion: because firms make automation decisions according to the wage rate, not the lower opportunity cost of labor, there will be a natural bias towards excessive automation. 

Q9. What is the function of the assumption that labor has a competitive advantage in tasks with a?

the authors impose the following assumption:Assumption 1 γ(i) is strictly increasingAssumption 1 implies that labor has strict comparative advantage in tasks with a higher index, and will guarantee that, in equilibrium, lower-indexed tasks will be automated, while higher-indexed ones will be produced with labor. 

Q10. how much is the value of a limit price for all types of intermediates?

This ensures that the unique equilibrium price for all types of intermediates is a limit price of ψ, and yields a per unit profit of (1−µ)ψ > 0 for technology monopolists. 

Q11. What is the structure of equilibrium in a task-based framework?

The authors characterize the structure of equilibrium in such a model, showing how, given factor35prices, the allocation of tasks between capital and labor is determined both by available technology and the endogenous choices of firms between producing with capital or labor. 

Q12. What are the reasons why the factor prices are not compatible with the modern office?

Just to cite a few motivating examples for this assumption: power looms of the 18th and 19th century are not compatible with modern textile technology; assembly lines based on the dedicated machinery are not compatible with numerically controlled machines and robots; first-generation calculators are not compatible with computers; and bookkeeping methods from the 19th and 20th centuries are not compatible with the modern, computerized office. 

Q13. How do the authors explain the production of tasks i?

Although tasks i ≤ The authorare technologically automated, whether they will be produced with capital or not depends on relative factor prices as the authors describe below. 

Q14. What is the effect of automation on wage inequality?

As a result, low-skill workers are progressively squeezed into a smaller and smaller set of tasks, and wage inequality grows without bound. 

Q15. What is the need for additional empirical evidence on how automation impacts employment and wages?

and perhaps most importantly, their model highlights the need for additional empirical evidence on how automation impacts employment and wages (which the authors investigate in Acemoglu and Restrepo, 2017) and how the incentives for automation and the creation of new tasks respond to policies, factor prices and supplies.