Since the concavity of the Engel Curve determines whether it is a necessity or luxury (i.e. how fast quantity demand changes in relation to changes in income), and since the second derivative of a Cobb-Douglas Engel Curve is 0, does that mean it is neither category?

Edit: In simple terms, if an Engel Curve is a straight positively-sloped line, it is obviously a normal good. But if the curve is represented by a function like so:

$I = 10 * P_x * x.$

Then the curve is ambiguously sloped. If the price of $x$ ($P_x$) happens to be greater than 1/10 then the good is a luxury, and vice versa for a good with $P_x < 1/10$. But if it's ambiguous like this, then the second derivative (which indicates concavity and therefore what direction the curve is increasing/decreasing in) then is it impossible to tell?

For reference, this Engel Curve was derived from the Cobb-Douglas utility function:

$U(x,y) = x^{(1/10)}y^{(9/10)}.$


1 Answer 1


Recall the following equivalent definitions for luxury goods and necessities:

  • A good $x$ is considered a necessity if $e_{(x,I)}<1$.

  • A good $x$ is considered a luxury good if $e_{(x,I)}>1$.

    As you can see, these definition do not encompass all possible scenarios, so any specific good does not have to be either a luxury or a necessity.

In the case of a Cobb-Douglas utility function $U(x,y)=x^\alpha \cdot y^\beta$ we get $x^* = \frac {\alpha I}{P_x}$. One can easily verify that $e_{(x,I)}=1$. In other words, the demand does for $x$ does not change with $I$. This means that in this case, $x$ is neither a luxury good nor a necessity.

  • 1
    $\begingroup$ I would add that for necessity good the elasticity is more precisely between 0 and 1 otherwise for a negative elasticity it would be a Giffen good ! $\endgroup$
    – Alexis L.
    Commented May 29, 2016 at 15:00

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.