
Calculus Calculus Math Forum 
 LinkBack  Thread Tools  Display Modes 
December 20th, 2016, 11:42 PM  #1 
Newbie Joined: May 2015 From: UK Posts: 8 Thanks: 0  Differentiate both sides of equation (labour demand and supply)
Hi all I am looking at the simple economic theory behind the impact of wage subsidies and I'm struggling with a simple differentiation of the labour demand and labour supply functions. In Page 13 of the linked pdf (extracted image attached), the author differentiates Equation (2) to get Equation (3). It appears to be a simple differentiation but I'm not able to work through the steps to get (3), and I'm not sure how to differentiate both sides of an equation that has different function arguments. Any help will be greatly appreciated! https://ink.library.smu.edu.sg/cgi/v...t=soe_research 
December 21st, 2016, 02:56 PM  #2 
Global Moderator Joined: May 2007 Posts: 6,788 Thanks: 708 
$\displaystyle s^w$ is the independent variable. Everything needs to be expressed as functions of it.

December 21st, 2016, 06:10 PM  #3  
Newbie Joined: May 2015 From: UK Posts: 8 Thanks: 0  Quote:
Am I suppose to differentiate the left hand side totally like the attached image?  
December 22nd, 2016, 05:13 PM  #4 
Global Moderator Joined: May 2007 Posts: 6,788 Thanks: 708 
As a math question, too much is undefined. What is the relationship (if any) between w^f, w^h, and s^w?
Last edited by mathman; December 22nd, 2016 at 05:17 PM. Reason: typos 
December 22nd, 2016, 06:10 PM  #5  
Senior Member Joined: May 2016 From: USA Posts: 1,310 Thanks: 551  Quote:
I am in no mood to plow through a bunch of implied math written in weird notation. Basically, what the author is saying is: $p = P(h) = labor\ provided,\ where\ h = wage\ taken\ home\ by\ employee;$ $u = U(w) = labor\ used,\ where\ w = wage\ paid\ by\ employer;\ and$ $h = w(1 + s),\ where\ s = subsidy\ paid\ by\ government.$ At a neoclassical equilibrium, $p = u \implies P(h) = U(w) \implies P\{w(1 + s)\} = U(w).$ So far so good, though the notation is less than clear. Next we jump into elasticities, an admittedly useful technical concept in economics. If the author thinks he is writing for anyone except fellow economists, he might think to define: $\eta = \left  \dfrac{dp}{dh} * \dfrac{h}{p} \right\ and\ \epsilon = \left  \dfrac{du}{dw} * \dfrac{w}{u} \right .$ I am good up to here because, for my manifest sins, I once long ago studied some economics. I have no idea what the author is doing when he differentiates with respect to s, which up to now appeared to be a constant such as 0.1. Last edited by skipjack; December 24th, 2016 at 03:23 AM.  
December 23rd, 2016, 01:50 PM  #6 
Global Moderator Joined: May 2007 Posts: 6,788 Thanks: 708 
Using the new notation, what is the question?

December 23rd, 2016, 06:44 PM  #7 
Senior Member Joined: May 2016 From: USA Posts: 1,310 Thanks: 551 
Mathman I am not quite sure. A standard neoclassical argument would posit a demand curve U(w) for labor depending on w, the wage to be paid by the employer, and a supply curve P(h) depending on h, the wage taken home by the employee. We define s = (h  w) / w, entailing that h = (1 + s)w. So redefine P(h) as P(s, w). Absent taxes or subsidies on wages, s would equal 0 by definition, and equilibrium would occur at w = w_1 such that U(w_1) = P(0, w_1). The equilibrium would be stable if $\dfrac{dU(w_1)}{dw} \le 0\ and\ \dfrac{\delta P(0,\ w_1)}{\delta w} \ge 0.$ This is standard Marshallian theory, taught in first semester microeconomics. I think where the argument is headed is to show that if s > 0, then there $\exists\ w_2\ such\ that\ w_2 < w_1 < (1 + s)w_2,\ P(s,\ w_2) = U(w_2) > U(w_1),$ $\dfrac{dU(w_2)}{dw} \le 0\ and\ \dfrac{\delta P(s,\ w_2)}{\delta w} \ge 0.$ But I am lost with respect to taking the derivative with respect to s. If both s and w are variables, then I do not grasp why we are not dealing with partial derivatives. I completely lose the thread here. I may not even get what is trying to be demonstrated. I am not at all saying that there is any error in the text, but the notation is so unnecessarily convoluted and the presentation so diffuse that I do not have the patience to decipher it. I think the OP should send an email to the author and ask for clarification directly. Last edited by skipjack; December 24th, 2016 at 01:41 AM. 
December 24th, 2016, 01:58 PM  #8 
Global Moderator Joined: May 2007 Posts: 6,788 Thanks: 708 
I'm just guessing, but it looks like the analysis is for fixed s.

December 24th, 2016, 11:09 PM  #9 
Global Moderator Joined: Dec 2006 Posts: 20,835 Thanks: 2162 
If $w^f$ is treated as a constant, $\dfrac{dw^h}{ds^w} = \dfrac{d}{ds^w}(w^f(1 + s^w)) = w^f$. On that basis, $\dfrac{d\log w^h}{ds^w} = \dfrac{1}{w^h}\cdot\dfrac{dw^h}{ds^w} = \dfrac{\dfrac{dw^h}{ds^w}}{w^f(1 + s^w)}= \dfrac{1}{1 + s^w}$. That supposedly equals $\dfrac{\eta}{\eta + \epsilon}$, i.e. $\dfrac{1}{1 + \epsilon/\eta}$, which would require that $\epsilon/\eta = s^w = w^h/w^f  1$. As $\epsilon$ and $\eta$ are respectively the wage elasticity of labor supply and the wage elasticity of labor demand, and labor demand equals labor supply in labormarket equilibrium, one might suppose that $\epsilon/\eta = \dfrac{dw^f}{dw^h}$ or something like that, but I don't see how he gets $w^h/w^f  1$. Maybe someone else will do better. 

Tags 
demand, differentiate, equation, labour, sides, supply 
Search tags for this page 
Click on a term to search for related topics.

Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
ti83/84 supply and demand function  blimper  Algebra  4  August 25th, 2016 04:48 PM 
Help: Demand and supply equation  HungerGames1  Economics  0  April 10th, 2012 01:23 AM 
Supply and Demand  inequality  Economics  3  September 20th, 2009 08:33 AM 
Supply and demand...  johnnyboy12  Algebra  3  June 11th, 2009 09:08 AM 
Supply and Demand  inequality  Applied Math  1  December 31st, 1969 04:00 PM 