In A Contribution To The Empirics of Economic Growth found here. The authors perform an econometric regression on the Solow-Swan model, they also do a restricted regression to find if the null hypothesis that the coefficient on the log of savings $\hat{\gamma}_1$ is equal and opposite to the coefficient of the log of depreciation $\hat{\gamma}_2$, that is:

$$ \hat{\gamma}_1 = - \hat{\gamma}_2 $$

Where in the Solow model this is:

$$ \hat{\gamma}_1 = \frac{\alpha}{1-\alpha} $$

I see how they calculated the implied $\alpha$, $1.43 = \alpha/(1-\alpha) \Rightarrow \alpha = 0.59 $

But why did they use $ln(s) - ln(n+g+\delta)$? Trying to solve this myself ...

$$ 1.43 = ln(s) - ln(n+g+\delta) \simeq \hat{\gamma}_1 - \hat{\gamma}_2 = 0 $$

$$ = \frac{\alpha}{1-\alpha} + \frac{\alpha}{1-\alpha} = 2 \frac{\alpha}{1-\alpha} $$

Which is too large by factor of 2. Also, and I don't know if I should post this as a separate question is that I have no idea how they calculated the implied $\alpha$ and $\beta$ in the next model which includes schooling as a proxy for measuring human capital.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.