ECON 102: Fall 2024

Tyler Porter
Derivative Rules (Proofs)

Here I outline some basic rules of differentiation that you’ll find useful in economics courses. I provide proofs of
all of these rules (except for the chain rule). You don’t need to be able to prove these rules in this class, you only need
to apply them. Whenever I say that a function is differentiable, this just means that the derivative of the function
exists. Throughout, I use the two forms for the derivative of a function f (x):

d
f (x) = f ′ (x)
dx
interchangeably.
1. (Derivative of a Constant) Suppose that f (x) = a, where a is some constant. Then:

d
f (x) = 0
dx
So the derivative of a constant is 0. This makes sense because the graph of f is horizontal at y = a. So its slope
is 0 everywhere.

2. (Re-scaling) Suppose that f is differentiable at x. Then if a is some constant:

d h d i
[a · f (x)] = a · f (x) = a · f ′ (x)
dx dx
One way to think about this is that multiplying f by the constant a will just stretch or shrink the function f by
a factor of a. This then streches or shrinks the slope of f at every point by that same factor a.

Proof. Let g(x) = a · f (x). Then we use the definition of the derivative for g:

d g(x + h) − g(x) a · f (x + h) − a · f (x)

g(x) = lim = lim
dx h→0 h h→0 h
Then we just factor a out of the entire expression and get:

f (x + h) − f (x)
= a · lim = a · f ′ (x)
h→0 h

3. (Derivative of a Sum) Suppose that f and g are both differentiable at x. Then:

d d d
[f (x) + g(x)] = f (x) + g(x) = f ′ (x) + g ′ (x)
dx dx dx
Here I’ll give a proof of the sum rule:

Proof. Let J(x) = f (x) + g(x). Then:

d J(x + h) − J(x) (f (x + h) + g(x + h)) − (f (x) + g(x))

J(x) = lim = lim
dx h→0 h h→0 h
f (x + h) − f (x) + g(x + h) − g(x) f (x + h) − f (x) g(x + h) − g(x)
= lim = lim + lim
h→0 h h→0 h h→0 h
d d
= f (x) + g(x)
dx dx

1
4. (Product Rule) If f and g are differentiable at x, then:
d h d i h d i
[f (x) · g(x)] = f (x) · g(x) + f (x) · g(x) = f ′ (x) · g(x) + f (x) · g ′ (x)
dx dx dx
Proof. Let J(x) = f (x) · g(x). Then:
d J(x + h) − J(x) f (x + h) · g(x + h) − f (x) · g(x)
J(x) = lim = lim
dx h→0 h h→0 h

we’ll use an “add-subtract” trick here:

f (x + h) · g(x + h) − f (x + h) · g(x) + f (x + h) · g(x) − f (x) · g(x)
= lim
h→0 h
and then split this up into two terms
f (x + h) · g(x + h) − f (x + h) · g(x) f (x + h) · g(x) − f (x) · g(x)
= lim + lim
h→0 h h→0 h
then do some factoring:
 g(x + h) − g(x)   (f (x + h) − f (x)) 
= lim f (x + h) · + lim g(x) ·
h→0 h h→0 h
and then use the fact that we can split the limit of a product into the product of two limits (as long as each of
them exists):
   g(x + h) − g(x)     f (x + h) − f (x) 
= lim f (x + h) · lim + g(x) · lim
{z h {z h
h→0 h→0 h→0
| {z } | } | }
=f (x) d d
= dx g(x) = dx f (x)

and thus:
= f (x) · g ′ (x) + f ′ (x) · g(x)

There are definitely some limit rules being used here that we never proved. We just don’t have the time to give
a full and proper treatment of limits (and in any case we would not do much with them in this class aside from
prove this exact statement).
5. (Power Rule) For a function of the form f (x) = xp with x > 0:
d p
x = p · xp−1
dx
1 √
The reason that we assumed x > 0 here is to avoid a case like (−1) 2 = −1 = i, which is a complex number.
You can still take the derivative of a function with complex values, but it’s beyond the scope of this class. This
is probably the most important derivative rule in the sense that you’ll use it the most. Note that this still
d −1
holds when p < 0, so for example dx x = −1 · x−2 = − x12 ; and it also holds if p is a fraction so for example
d √ d 2 1 1 −21
dx x = dx x = 2 · x = 12 · √1x
This turns out to be a little harder to show for arbitrary p, but we could prove it when p is a positive integer by
repeatedly applying the product rule as follows:
d
x=1
dx
d 2 d  d   d 
x = (x · x) = x ·x+x· x = 1 · x + x · 1 = 2x
dx dx dx dx
d 3 d  d   d 
x = (x · x2 ) = x · x2 + x · x2 = 1 · x2 + x · 2x = 3x2
dx dx dx dx
One could continue this process in the same fashion to show that it’s true for any positive integer. Note that we
can also do the same thing when p is a negative integer.

2
6. (Multiplicative Inverse) Suppose that f is differentiable at x and f (x) ̸= 0. Then:

d 1 1  d  1
=− 2
· f (x) = − · f ′ (x)
dx f (x) (f (x)) dx (f (x))2

There are many ways to show this! One route is the direct way (shown below), and another involves applying
both the chain rule (Rule 8) and power rule (Rule 5).

1
Proof. Let J(x) = f (x) . Then the derivative is defined as:

1 1
d J(x + h) − J(x) f (x+h) − f (x)
J(x) = lim = lim
dx h→0 h h→0 h
Note here that since f (x) ̸= 0 and f is continuous at x, f (x + h) ̸= 0 for small enough h (try to sketch this out).
To make things easier to factor, let’s multiply the first term in the numerator by ff (x)
(x)
, and the second term in
f (x+h)
the numerator by f (x+h) :

f (x) f (x+h)
f (x)·f (x+h) − f (x)·f (x+h)
= lim
h→0 h
Now we’ll factor out a (-1) from the numerator:
f (x+h) f (x)
f (x)·f (x+h) − f (x)·f (x+h)
= lim −
h→0 h
And then factor out the denominators in the terms on top:

1 f (x + h) − f (x)
= lim − ·
h→0 f (x) · f (x + h) h

and then split this up into two limits:

 1   f (x + h) − f (x) 
= lim − · lim
h→0 f (x) · f (x + h) | h→0 {z h }
d
= dx f (x)

and note that since f (x + h) → f (x) as h → 0, we have:

1
=− · f ′ (x)
(f (x))2

7. (Quotient Rule) Suppose that f and g are both differentiable at x and that g(x) ̸= 0. Then:

d f (x) f ′ (x) · g(x) − f (x) · g ′ (x)

=
dx g(x) (g(x))2

We’ve basically already proved this. Indeed we can just apply the product rule (Rule 4) and Rule 6. This is
done as follows:

Proof. Note that:

f (x) 1
= f (x) ·
g(x) g(x)

and hence we can apply the Product Rule:

d  1   d  1  d 1 
f (x) · = f (x) · + f (x) ·
dx g(x) dx g(x) dx g(x)

3
 and apply Rule 6 to the term on the far right side to get:
1  1 
= f ′ (x) · + f (x) · − · g ′
(x)
g(x) (g(x))2

After simplifying this a little bit, we get:

f ′ (x) f (x) · g ′ (x)

= −
g(x) (g(x))2

Now we multiply and divide the first term by g(x) to get:

f ′ (x) · g(x) f (x) · g ′ (x)

= −
(g(x))2 (g(x))2

which can then combine over the common denominator:

f ′ (x) · g(x) − f (x) · g ′ (x)
=
(g(x))2

8. (Chain Rule) The chain rule deals with the composition of two functions g and f . In particular, if f is differen-
tiable at x and g is differentiable at y = f (x), then:

d
g(f (x)) = g ′ (f (x)) · f ′ (x)
dx
where the first term on the right is the derivative of g evaluated at the point f (x), and the second term on the
right is the derivative of f evaluated at the point x.

The chain rule turns out to be harder to show than the rest of the rules on this list, so we will leave out a
proof of it.