We let $g(z)$ be a strictly monotonous function so: $$\frac{dg(z)}{dz}>0$$ Consumer 1 has preferences given by the utility function $u(x_1,x_2)=ln(x_1)+2ln(x_2)$, while consumer 2 has preferences given by n $v(x_1,x_2)=g(x_1x_2^2)$. Then I have to show that consumer 2 got same preferences as consumer 1. I think I have to use MSR on $x_1x_2^2$ and on $v(x_1,x_2)$. For MSR on $x_1x_2^2$ I get: $$MRS=-\frac{\frac{\partial }{\partial x_1}}{\frac{\partial }{\partial x_2}}=-\frac{x_2^2}{2x_1x_2}$$ But How can I find MSR on $v(x_1,x_2)$ (I think I got the same MRS if I use the chain rule, but I'm not sure?), and how can I use this to conclude that the ranking of the two indifference curves is the same when $g(z)$ is monotonous? I hope that someone can help me?

  • $\begingroup$ Note that ln(x1)+2*ln(x2) = ln(x1)+ln(x2^2) = ln (x1 * x2^2) and then verify that MRS is invariant to any positive monotone transforation. $\endgroup$
    – Bayesian
    Nov 26, 2020 at 16:41

1 Answer 1


Note that $$ln(x_1)+2*ln(x_2) = ln(x_1)+ln(x_2^2) = ln (x_1 * x_2^2),$$ and note that $$MRS_v = \frac{g'(x_1 * x_2^2) x_1 * 2 x_2}{g'(x_1 * x_2^2) x_2^2}$$ such that the derivative of $g$ cancels out.

For more intuition, see here.

  • $\begingroup$ Nice that makes sense. Thank you $\endgroup$
    – Lifeni
    Nov 26, 2020 at 16:53
  • $\begingroup$ You should also see that if you prefer bundle x1,x2 to y1,y2 then it must be that u(x1,x2) > u(y1,y2). If you now apply a positive monotone transformation g, you can see that u(x1,x2) > u(y1,y2) if and only if g(u(x1,x2)) > g(u(y1,y2)). $\endgroup$
    – Bayesian
    Nov 26, 2020 at 16:55
  • $\begingroup$ @Lifeni if you think this answer answered your question consider accepting it. $\endgroup$
    – 1muflon1
    Nov 26, 2020 at 16:56
  • $\begingroup$ Hi Bayesian that makes sense and the intuition in the link makes sense but have can I formally show that the ranking of the two indifference curves is the same when 𝑔(𝑧) is monotonous? $\endgroup$
    – Lifeni
    Nov 26, 2020 at 17:06
  • $\begingroup$ See my comment if you compare any two bundles then you have u(x1,x2) > u(y1,y2) if and only if g(u(x1,x2)) > g(u(y1,y2)) by definition of g being a positive monotone transformation. Therefore the ranking is not changed. $\endgroup$
    – Bayesian
    Nov 26, 2020 at 17:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.