0
$\begingroup$

In a two-good world, if one good is a Giffen good then the other is a good luxury.

This is a false statement. Reason: If X is a Giffen good then Y must be a normal good. But we cannot get any conclusion about luxury good. Am I correct?

$\endgroup$

1 Answer 1

3
$\begingroup$

If $X$ is a Giffen good, then it must an inferior good.

Next, as the share-weighted sum of the income elasticities of all goods sum to 1 (Engel aggregation), and given that the income elasticity of $X$ is negative, the income elasticity of good $Y$ must be bigger than 1, which means that $Y$ is a luxury good.

To see that the latter, start from the budget constraint: $$ p_X X(p,m) + p_Y Y(p,m) = m. $$ Take the derivative with respect to income, $m$: $$ p_X \frac{\partial X}{\partial m} + p_Y \frac{\partial Y}{\partial m} = 1. $$ This gives: $$ \frac{p_X X}{m} \frac{\partial X}{\partial m} \frac{m}{X} + \frac{p_Y Y}{m} \frac{\partial Y}{\partial m}\frac{m}{Y} = 1. $$ Let $s_X = \frac{p_X X}{m}$ be the budget share of good $X$ and $s_Y = \frac{p_Y Y}{m}$ be the budget share of good $Y$ and denote by $\varepsilon^X_m$ and $\varepsilon^Y_m$ the income elasticities of goods $X$ and $Y$ respectively. Then: $$ s_X \,\,\varepsilon^X_m + s_Y \,\,\varepsilon^Y_m = 1. $$ The first term is negative (as $X$ is inferior, $\varepsilon^X_m$ is negative). This means that $s_Y \varepsilon^Y_m \ge 1$. As $s_Y \le 1$, we obtain: $$ \varepsilon^Y_m \ge \frac{1}{s_Y} \ge 1. $$ This shows that $Y$ is a luxury good.

$\endgroup$
3
  • $\begingroup$ I have another question. Could you help me check that? Thank you! economics.stackexchange.com/questions/57779/… $\endgroup$
    – Jonathen
    Commented Feb 28 at 14:15
  • $\begingroup$ Another question which is still relevant on income and price elasticity: But this time this have "no money illusion" condition, which confuses me a lot. I don't know how to use that condition. economics.stackexchange.com/questions/57778/… $\endgroup$
    – Jonathen
    Commented Feb 28 at 14:43
  • $\begingroup$ This is a very good answer. I forgot to consider their mathematical connections. Thanks! $\endgroup$
    – Jonathen
    Commented Feb 28 at 14:44

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.