In the usual CRRA $\frac{c^{1-\sigma}-1}{1-\sigma}$ function we have that the intertemporal elasticity of sustitution $\partial\frac{c_{t+1}}{{c_{t}}{\partial r}}$ is $\frac{1}{\sigma}$.

But how can i calculate the IE of S when the CRRA function is the composite:


Thanks in advance

  • $\begingroup$ you can still take derivative of such function by chain rule $\endgroup$
    – 1muflon1
    Nov 16, 2020 at 12:59
  • $\begingroup$ Can you explain that in an answer? $\endgroup$
    – mmendina
    Nov 16, 2020 at 13:15

1 Answer 1


Suppose $c_t$ is some composite function of interest rate $r$, e.g. $c_t(G(r_t))$.

First the intertemporal elasticity of substitution is actually (IES) given by $\frac{\partial \ln(c_{t+1}/c_{t})}{\partial r}$ (or also $\frac{\partial \ln(c_{t+1}/c_{t})}{\partial \ln( u'(c_{t+1})/u'(c_t))}$).

You can take the derivative above using chain rule for composite functions which says that $dF(G(x))/dx = \frac{dF}{dG}\frac{dG}{dx}$.

So in the above case we would get:

$$\frac{\partial \ln(c_{t+1}(G(r))/c_{t}(G(r)))}{\partial r} = \frac{\partial \ln(c_{t+1}(G(r))/c_{t}(G(r)))}{\partial G(r)} \cdot \frac{\partial G(r)}{\partial r} $$

You can directly apply this to the problem you mention in your question.

  • $\begingroup$ Perhaps I expresed myself not clearly enough. MY problem is the fact that the function G() enters substracting consumption in the traditional CRRA function not as a composite one. $\endgroup$
    – mmendina
    Nov 16, 2020 at 17:27
  • $\begingroup$ @MartinMendina I don’t understand what you mean. Why would the fact that it is subtracted from consumption function matter at all? The above formula would work even with negative utility or consumption regardless if that makes economic sense. $\endgroup$
    – 1muflon1
    Nov 16, 2020 at 17:30

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.