Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!darkstar.UCSC.EDU!news.hal.COM!olivea!uunet!hobbes!earth.armory.com!rstevew
From: rstevew@armory.com (Richard Steven Walz)
Subject: Re: Does Automation Take Jobs Away?
Organization: The Armory
Date: Mon, 25 Jul 1994 15:54:05 GMT
Message-ID: <CtI6u7.51v@armory.com>
References: <30s4ne$3ro@tribune.usask.ca> <30ub72$bvp@mailer.fsu.edu>
Sender: news@armory.com (Usenet News)
Nntp-Posting-Host: deeptht.armory.com
Lines: 58

In article <30ub72$bvp@mailer.fsu.edu>,
Edward Flaherty <eflahert@garnet.acns.fsu.edu> wrote:
>Does automation take jobs away?
>
>The answer to this question is easy with an appeal to basic mircoeconomic 
>theory.  Suppose you have a profit maximizing firm which is a price taker 
>in both the input and output markets.  If the firm wants to produce an 
>output level, q, then it will want to minimize the costs of doing so.  
>That is, it will want to solve
>
>	min
>	L,K      L = wL + rK + lambda[q - f(L,K)] 
>
>by choosing the amounts of L and K is hires, where L is labor, K is 
>capital, w is the wage, r is the rental rate of capital, f(L,K) is the 
>production function, and lambda is a LaGrangian multiplier.  Solving this 
>yields the optimal conditions of 
>
>	w   MPL  (marginal product of labor)
>	- = ---
>	r   MPK  (marginal product of capital).
>
>This could also be written as
>
>	w = lambda * MPL
>and 	r = lambda * MPK.
>
>Because lambda measures the change in the objective function (total cost) 
>when there is a small change in the constraint (total output), lambda 
>represents marginal cost.  Thus,
>
>	w = MC * MPL
>and 	r = MC * MPK.
>
>Suppose the price of capital decreases such that the firms wants to have 
>more automation.  The optimizing condition shows that the firm changes 
>its input mix to include more capital and less labor.  This is the 
>substitution effect.  However, at the same time the lower capital price 
>makes the marginal cost of production decrease, increasing the firm's 
>profit maximizing level of output.  This will create additional demand 
>for labor.  This is the output effect.
>
>	Thus, automation can decrease employment if the substitution 
>effect is stronger than the output effect, but it might increase 
>employment if the output effect prevails.  Isn't microeconomics cool?
>
>	It seems intuitive (to me, at least) that the output effect 
>prevails most of the time.  Otherwise, we would expect that capital would 
>eventually become the only input into production, which has clearly not 
>been the case in the 150 years of American industrialization.
-------------------------------
Math is so pretty. Too bad it has never worked in this field!! You cannot
find two economists, not currently paid by the same government or corproate
administration, who agree. And then they'll stop agreeing if they don't get
paid!!! Amazing. It's kind of like psych majors who want to diagnose you
their first week in class. Kind of pitiful, really.
-Steve Walz   rstevew@armory.com

