In the standard expanding variety model where the representative household faces the following CRRA utility function

$$ \int_{0}^{\infty} \exp(-\rho t) \frac{C(t)^{1-\theta}-1}{1-\theta}dt $$

and final goods that is produced competitively with the following production $$ Y(t) = \frac{1}{1-\beta} \left( \int_{0}^{N(t)}x(v,t)^{1-\beta}dv\right)L^{\beta} $$

with the following resource constraint

$$ C(t) + X(t) + Z(t) \leq Y(t) $$

where $X(t)$ is the total spending on machines, and $Z(t)$ is the total expenditure on research and development.

The innovation possibilities frontier takes the form

$$ \dot{N}(t) = \eta Z(t) $$

The intermediary goods firm faces the following problem where

$$ r(t) V(v,t) - \dot{V}(v,t) = \pi(v,t) $$

The thing I don't understand about the model is how the Euler Equation is derived. In almost everywhere I look the Euler Equation is the standard

$$ \frac{\dot{C}(t)}{C(t)} = \frac{1}{\theta}(r(t)-\rho) $$

I don't understand how you would derive this equation.

  • $\begingroup$ Hint: set up the Hamiltonian $\endgroup$
    – ChinG
    Jul 5, 2020 at 17:19

1 Answer 1


The way you would go about solving this problem is as the ChinG said is by setting up the Hamiltonian. In this case this is:

$$\mathcal{H}:e^{-\rho t} \frac{C(t)^{1-\theta}-1}{1-\theta}+\mu(t)\left[Y(t)-C(t)-X(t)-Z(t)\right]$$

Taking the first order condition for this problem we get: $$\frac{\partial \mathcal{H}}{\partial C(t)}:e^{-\rho t}C(t)^{-\theta}-\mu(t)=0 $$ or $$e^{-\rho t}C(t)^{-\theta}=\mu(t)$$

Taking the natural logs of this equation and differentiating with respect to time we get: $$\ln\left[\frac{\partial \mathcal{H}}{\partial C(t)}\right]:-\rho t -\theta \ln [C(t)]=\ln[\mu(t)]$$ $$\frac{\partial\ln\left[\mathcal{H}_{C(t)}\right]}{\partial t}:-\rho-\theta\frac{\dot{C}(t)}{C(t)}=-\frac{\dot{\mu}(t)}{\mu(t)}$$ Rearranging the above and noting that $\frac{\dot{\mu}(t)}{\mu(t)}=r(t)$ we get: $$\frac{\dot{C}(t)}{C(t)}=\frac{1}{\theta}(r(t)-\rho)$$ Which is the Euler equation.

Hope this helps.

  • $\begingroup$ Hi, thanks for the response! On thing I don't quite understand is why $\dot{\mu}/\mu = r(t)$? $\endgroup$
    – Finalblue
    Jul 8, 2020 at 14:55
  • $\begingroup$ @Finalblue Theres a ton of math to get that result so I tried to be sneaky and handwave a bit. I could tell you that intuitively we think of $\mu$ as the marginal utility from output over time. We interpret the growth in that utility from output over time $\dot{\mu}/\mu$ as the rate of return on the market $r(t)$. $\endgroup$
    – EconJohn
    Jul 9, 2020 at 1:37

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.