The Economics of Artificial Intelligence

Home > Other > The Economics of Artificial Intelligence > Page 40
The Economics of Artificial Intelligence Page 40

by Ajay Agrawal


  these new technologies.

  Appendix

  Derivations for the Basic Model

  Suppose that assumption (A1) holds. We fi rst derive the demand for

  factors:

  • Denote by p( x) the price of task x. Assumption (A1) implies

  R

  if x

  N

  1, I

  ( x)

  (8A.1)

  p( x) =

  M

  W

  if x

  I , N

  (

  .

  ( x)

  L

  • In addition, the demand for task x is given by

  y( x) = Y .

  p( x)

  • Thus, the demand for smart machines in task x is

  Y

  if x

  N

  1, I

  k( x) =

  R

  ,

  0

  if x

  I , N

  (

  and the demand for labor in task x is

  0

  if x

  N

  1, I

  ( x) =

  Y

  if x

  I , N

  (

  .

  W

  • Aggregating the demand for machines from this expression and set-

  ting it equal to the supply of capital, K, we have the following market-

  clearing condition for capital:

  K = Y ( I

  N + 1).

  R

  Similarly, aggregating the demand for labor and setting it equal to its inelas-

  tic supply, L, we obtain the market- clearing condition for labor as

  230 Daron Acemoglu and Pascual Restrepo

  L = Y ( N

  I ).

  W

  • Rearranging these two equations, the equilibrium rental rate and wage

  can be obtained as

  (8A.2)

  R = Y ( I

  N + 1) and W = Y ( N

  I ),

  K

  L

  which are the expressions used in the text.

  We next turn to deriving the expression for aggregate output.

  • Because we normalized the price of the fi nal good to 1 as numeraire,

  we have

  N

  ln p( x) dx = 0.

  N 1

  • Plugging in the expressions for p( x) from equation (8A.1) yields

  I

  N

  ln R

  ln

  ( x) dx +

  ln W

  ln

  ( x) dx = 0.

  M

  L

  N 1

  I

  • Substituting the expressions for R and W from (8A.2), we obtain

  I

  ln Y

  ln K /( I

  N + 1)

  (

  ) ln ( x) dx

  M

  N 1

  N

  + ln Y ln L/( N I)

  (

  ) ln ( x) dx = 0.

  L

  I

  • This equation can be rearranged as

  I

  K

  N

  L

  ln Y =

  ln

  + ln

  ( x) dx +

  ln

  + ln ( x) dx

  I

  N + 1

  M

  N

  1

  L

  N 1

  I

  I

  N

  =

  ln

  ( x) dx + ln

  ( x) dx

  M

  L

  N 1

  I

  K

  L

  + ( I N + 1)ln

  + ( N I)ln

  ,

  I

  N + 1

  N

  I

  which, after taking exponentials on both sides of the equation, yields the

  expression for aggregate output in equation (1) in the text.

  Assumption (A1)

  We now show that assumption (A1) is equivalent to the capital- labor ratio

  of the economy taking an intermediate value. In particular, there exist two

  positive thresholds < such that assumption (A1) holds whenever

  Artifi cial Intelligence, Automation, and Work 231

  K

  (A2)

  ( , ).

  L

  Equation (8A.2) shows that

  W

  N

  I

  = K

  .

  R

  L I

  N + 1

  Defi ne

  ( I )

  ( N )

  = I N + 1 L

  , and

  = I N + 1

  L

  .

  N

  I

  ( I )

  N

  I

  ( N

  I )

  M

  M

  Then equation (A2) is equivalent to assumption (A1).

  Derivations in the Presence of Technology- Skill Mismatch

  • Denote by p( x) the price of task x. Assumption (A1) together with the fact that W > W (see footnote 12) implies

  H

  L

  R

  if x

  N

  1, I

  ( x)

  M

  W

  p( x) =

  L

  if x

  ( I , S )

  .

  ( x)

  L

  WH

  if x

  S, N ]

  ( x)

  L

  • Following the same steps as in our baseline model, we obtain the

  market- clearing condition for capital,

  K = Y ( I

  N + 1).

  R

  • The demand for low- skill labor in task x is given by

  0

  if x

  N

  1, I

  Y

  ( x) =

  if x

  ( I , S )

  .

  WL

  0

  if x

  S, N ].

  • Aggregating the demand for low- skill labor and setting it equal to its

  inelastic supply, L, we obtain the market- clearing condition for low-

  skill labor as

  L = Y ( S

  I ),

  WL

  which implies the expression for W given in the main text.

  L

  232 Daron Acemoglu and Pascual Restrepo

  • The demand for high- skill labor in task x is given by

  0

  if x

  N

  1, I

  0

  if x

  ( I , S )

  h( x) =

  .

  Y

  if x

  S, N ].

  WH

  • Aggregating the demand for high- skill labor and setting it equal to

  its supply, H , we obtain the market- clearing condition for high- skill

  labor as

  H = Y ( N

  S ),

  WH

  which implies the expression for W given in the main text.

  H

  Derivations for the Model with Distortions

  • Denote by p( x) the price of task x. The variant of assumption (A1) introduced in section 8.5 implies

  R(1

  )

  if x

  N

  1, I

  ( x)

  M

  W (1 + )

  p( x) =

  if x

  ( I , J )

  ( x)

  L

  W

  if x

  J , N ].

  ( x)

  L

  • Following the same steps as in the model with no distortions, we obtain

&n
bsp; the market- clearing condition for capital,

  K =

  Y

  ( I

  N + 1).

  R(1

  )

  • The demand for labor in task x is

  0

  if x

  N

  1, I

  Y

  if x

  ( I , J )

  ( x) =

  W (1 + )

  .

  Y

  if x

  J , N ]

  W

  • The expression for ℓ( x) implies that the total amount of labor employed

  in tasks where labor gets rents is

  L =

  Y

  ( J

  I ).

  A

  W (1 + )

  Artifi cial Intelligence, Automation, and Work 233

  The total amount of labor employed in tasks where labor does not get rents is

  L

  L = Y ( N

  J ).

  A

  W

  To derive the expression for (gross) output we proceed as follows:

  • Again from our choice of numeraire, we have

  N

  ln p( x) dx = 0.

  N 1

  • Plugging in the expressions for p( x) we obtain

  I

  J

  ln R

  ln

  ( x) dx +

  ln W + ln(1 + ) ln

  ( x) dx

  M

  L

  N 1

  I

  N

  +

  ln W

  ln

  ( x) dx = 0.

  L

  J

  • Substituting for factor prices using the expressions for K, L , and A

  L – L , we obtain

  A

  I

  ln Y

  ln K / ( I

  N + 1)

  (

  ) ln ( x) dx

  M

  N 1

  J

  +

  ln Y

  ln L / ( J

  I )

  (

  ) ln ( x) dx

  A

  L

  I

  J

  +

  ln Y

  ln ( L

  L ) / ( N

  J )

  (

  ) ln ( x) dx = 0.

  A

  L

  I

  • This equation can be rearranged as

  I

  K

  J

  L

  ln Y =

  ln

  + ln

  ( x) dx +

  ln

  A

  + ln ( x) dx

  I

  N + 1

  M

  J

  I

  L

  N 1

  I

  N

  L

  +

  ln

  + ln ( x) dx

  N

  J

  L

  J

  I

  N

  K

  =

  ln

  ( x) dx + ln

  ( x) dx + ( I

  N + 1)ln

  M

  L

  I

  N + 1

  N 1

  I

  L

  L

  L

  + ( J I)ln

  A

  + ( N J )ln

  A

  ,

  J

  I

  N

  J

  which yields equation (12) in the text.

  234 Daron Acemoglu and Pascual Restrepo

  References

  Accenture PLC. 2017. “How Companies Are Reimagining Business Processes

  With IT.” https:// sloanreview.mit .edu/ article/ will- ai- create- as-many- jobs- as-it

  - eliminates/.

  Acemoglu, Daron. 2009. Introduction to Modern Economic Growth. Princeton, NJ:

  Princeton University Press.

  Acemoglu, Daron, and David Autor. 2011. “Skills, Tasks and Technologies: Implica-

  tions for Employment and Earnings.” Handbook of Labor Economics 4:1043– 171.

  Acemoglu, Daron, David Autor, David Dorn, Gordon H. Hanson, and Brendan

  Price. 2014. “Return of the Solow Paradox? IT, Productivity, and Employment

  in US Manufacturing.” American Economic Review: Papers & Proceedings 104

  (5): 394– 99.

  Acemoglu, Daron, and Pascual Restrepo. 2016. “The Race between Machine and

  Man: Implications of Technology for Growth, Factor Shares and Employment.”

  American Economic Review 108 (6): 1488– 542 .

  ———. 2017. “Robots and Jobs: Evidence from US Labor Markets.” NBER Work-

  ing Paper no. 23285, Cambridge, MA.

  ———. 2018a. “Excessive Automation: Technology Adoption and Worker Displace-

  ment in a Frictional World.” Unpublished manuscript.

  ———. 2018b. “Low- Skill and High- Skill Automation.” Journal of Human Capital

  12 (2): 204– 32 .

  Aghion, Philippe, Benjamin F. Jones, and Charles I. Jones. 2017. “Artifi cial Intelli-

  gence and Economic Growth.” NBER Working Paper no. 23928, Cambridge, MA.

  Allen, Robert C. 2009. “Engels’ Pause: Technical Change, Capital Accumulation,

  and Inequality in the British Industrial Revolution.” Explorations in Economic

  History 46 (4): 418– 35.

  American Enterprise Institute (AEI). 2008. “Taxing Capital.” Report by the

  American Enterprise Institute. https:// www .aei .org.

  Autor, David H. 2015. “Why Are There Still So Many Jobs? The History and Future

  of Workplace Automation.” Journal of Economic Perspectives 29 (3): 3– 30.

  Autor, David H., David Dorn, and Gordon H. Hanson. 2013. “The China Syn-

  drome: Local Labor Market Eff ects of Import Competition in the United States.”

  American Economic Review 103 (6): 2121– 68.

  Autor, David H., Frank Levy, and Richard J. Murnane. 2003. “The Skill Content of

  Recent Technological Change: An Empirical Exploration.” Quarterly Journal of

  Economics 118 (4): 1279– 333.

  Ayres, Robert, and Steven M. Miller. 1983. Robotics: Applications and Social Implications. Pensacola, FL: Ballinger Publishing Company.

  Bessen, James. 2016. Learning by Doing: The Real Connection between Innovation,

  Wages, and Wealth. New Haven, CT: Yale University Press.

  Bloom, Nicholas, Charles I. Jones, John Van Reenen, and Michael Webb. 2017.

  “Are Ideas Getting Harder to Find?” NBER Working Paper no. 23782, Cam-

  bridge, MA.

  Boston Consulting Group. 2015. “The Robotics Revolution: The Next Great

  Leap in Manufacturing.” https:// www .bcg .com/ en- us/ publications/ 2015/ lean

  - manufacturing- innovation- robotics- revolution- next- great- leap- manufacturing

  .aspx.

  Brynjolfsson, Erik, and Andrew McAfee. 2014. The Second Machine Age: Work,

  Progress, and Prosperity in a Time of Brilliant Technologies. New York: W. W.

  Norton & Company.

  Artifi cial Intelligence, Automation, and Work 235

  Chandler, Alfred D. 1977. The Visible Hand: The Managerial Revolution in American

  Business. Cambridge, MA: Harvard University Press.

  Deloitte and The Manufacturing Institute. 2011. “Boiling Point? The Skills Gap

  in U.S. Manufacturing.” Report. http:// www .themanufacturinginstitute .org/ ~

  / media/ A07730B2A798437D98501E798C2E13AA.ashx.

  Ford, Martin. 2016. The Rise of the Robots: Technology and the Threat of a Jobless

  Future. New York: Basic Books.

  Goldin, Claudia, and Larry Katz. 2010. The Race between Education and
Technology.

  Cambridge, MA: Harvard University Press.

  Gordon, Robert J. 2016. The Rise and Fall of American Growth: The U.S. Standard

  of Living Since the Civil War. Princeton, NJ: Princeton University Press.

  Graetz, Georg, and Guy Michaels. 2015. “Robots at Work.” CEP Discussion Paper

  no. 1335, Centre for Economic Performance.

  Groover, Mikell. 1983. CAD/ CAM: Computer- Aided Design and Manufacturing.

  Englewood Cliff s, NJ: Prentice Hall.

  Groover, Mikell, Mitchell Weiss, Roger N. Nagel, and Nicholas G. Odrey. 1986.

  Industrial Robotics: Technology, Programming and Applications. New York:

  McGraw- Hill.

  Herrendorf, Berthold, Richard Rogerson, and Ákos Valentinyi. 2013. “Two Perspec-

  tives on Preferences and Structural Transformation.” American Economic Review

  103 (7): 2752– 89.

  Kolb, David A. 1984. Experiential Learning: Experience as the Source of Learning

  and Development. Englewood Cliff s, NJ: Prentice Hall.

  Kuznets, Simon. 1966. Modern Economic Growth. New Haven, CT: Yale University

  Press.

  Landes, David. 1969. The Unbound Prometheus. New York: Cambridge University

  Press.

  Mantoux, Paul. 1928. The Industrial Revolution in the Eighteenth Century: An Outline of the Beginnings of the Modern Factory System in England. New York: Harcourt.

  Manuelli, Rodolfo E., and Ananth Seshadri. 2014. “Frictionless Technology Diff u-

  sion: The Case of Tractors.” American Economic Review 104 (4): 1368– 91.

  McKinsey Global Institute. 2017. “Jobs Lost, Jobs Gained: Workforce Transitions

  in a Time of Automation.” Report, McKinsey & Company. https:// www .mckinsey

  .com/ mgi/ overview/ 2017-in-review/ automation- and- the- future- of-work/ jobs

  - lost- jobs- gained- workforce- transitions- in-a- time- of-automation.

  Mian, Atif, and Amir Sufi . 2014. “What Explains the 2007– 2009 Drop in Employ-

  ment?” Econometrica 82 (6): 2197– 223.

  Minsky, Marvin. 1967. Computation: Finite and Infi nite Machines. Englewood Cliff s, NJ: Prentice- Hall.

  Mokyr, Joel. 1990. The Lever of Riches: Technological Creativity and Economic Pro-

  gress. New York: Oxford University Press.

  Nilsson, Nils J. 2009. The Quest for Artifi cial Intelligence: A History of Ideas and Achievements. Cambridge: Cambridge University Press.

  Olmstead, Alan L., and Paul W. Rhode. 2001. “Reshaping the Landscape: The

  Impact and Diff usion of the Tractor in American Agriculture, 1910– 1960.” Jour-

  nal of Economic History 61 (3): 663– 98.

  Pew Research Center. 2017. “Automation in Everyday Life.” Online Report. http://

  www .pewinternet .org/ 2017/ 10/ 04/ automation- in-everyday- life/.

  Rasmussen, Wayne D. 1982. “The Mechanization of Agriculture.” Scientifi c

  American 247 (3): 76– 89.

  Simon, Herbert A. 1965. The Shape of Automation for Men and Management. New

 

‹ Prev