Why does $\frac{MP_L}{MP_K} = \frac{w}{r}$?

I did a practice problem set. I was given a production function $F(K,L)$ and had to derive the firm's cost function. To do this, I used the following relationship:

\begin{equation} \frac{MP_L}{MP_K}=\frac{w}{r} \end{equation}

I found this formula in a set of lecture notes online. I have no idea where it comes from. I understand what $MP_K$, $MP_L$, $w$, and $r$ are but I don't know why these ratios are equal. I got the problem correct, but I want to understand this point better.

My Question:

Can someone explain why this ratio is true?

This is a result of firm maximization. Consider a problem that a profit maximizing firm faces --it must rent capital, $k$, at rate $r$ and hires labor, $l$, at wage $w$ to produce goods ($F(k, l)$) for which we can sell at a normalized price of 1.

$$\max_{k, l} F(k, l) - w l - r k$$

Then if we just think about the firm's maximization, we can take derivatives w.r.t $k$ and $l$ to get:

$$F_k(k, l) - r = 0$$ $$F_k(k, l) = r$$

and

$$F_l(k, l) - w = 0$$ $$F_l(k, l) = w$$

Notice that by definition $MPL_K = F_k(k, l)$ and $MP_L = F_l(k, l)$. If we take the ratio of the two equations then we get what you initially introduced --that is:

$$\frac{F_l(k, l)}{F_k(k, l)} = \frac{MP_L}{MP_K} = \frac{w}{r}$$

Edit: As mentioned in another answer, this is imposing some assumptions on the behavior of $F(k, l)$. These assumptions are $\frac{\partial F}{\partial k} > 0$, $\frac{\partial F}{\partial l} > 0$ and $\frac{\partial^2 F}{\partial k^2} < 0$ $\frac{\partial^2 F}{\partial l^2} < 0$

• Is it typical to normalize the price? Like how would I know to do that a priori? Or is that something you just learn and remember from experience... – Stan Shunpike Jan 17 '15 at 23:18
• It is something you kind of pick up - - I don't think I would have thought of it without having seen someone else do it first. It makes sense though since everything is in terms of the same good then you can choose whatever price you want because everything can be scaled accordingly. – cc7768 Jan 17 '15 at 23:36

Mathematically it is a necessary condition in problem of profit maximisation and it can be easily derived under usual assumptions. However there is an intuitive explanation.

We make a usual assumption of decreasing productivity of capital and labour (i.e. each additional unit of the input produce some product, but less then previous one, $\frac{\partial^2F}{\partial K^2}<0, \frac{\partial^2F}{\partial KL^2}<0, \frac{\partial F}{\partial K}>0, \frac{\partial F}{\partial K}>0$).

Now, we believe that the firm is maximizing profit (another assumption, better get used to this sort of simplifications). Then the ratio on the right represents how much one additional unit of labor will produce in terms of production of additional unit of capital, while the ratio on the left shows how much more will additional unit of labour cost you in terms of cost of one unit of capital. Also, I will set price of the good the firm produces to be 1 without loss of generality.

Now, if the right side is greater, you may decrease you labour by $\frac{\Delta}{MP_L}$ and increase capital by$\frac{\Delta}{MP_K}$, and you profit will increase by $\frac{w\Delta}{MP_L}-\frac{r\Delta}{MP_K}>0$, so it was not maximised. If the left side was greater you decrese capital, increase labour and increase your profit by$\frac{w\Delta}{MP_K}-\frac{r\Delta}{MP_L}>0$, and so profit, again, was not maximised. It means, that profit can only be maximised at points that satisfy the equation.