"Why does a negative exponent mean reciprocal?" is one of the most common questions a student asks when the exponent laws are first introduced. The formula a^{-n} = \dfrac{1}{a^{n}} feels like a rule pulled from a hat — until you see this one picture.

Drag the exponent slider below from +3 down through 0 and into the negatives. At the instant the exponent crosses zero, the value of 3^x flips: it was on top of a fraction bar, and now it is on the bottom. The rest of the "shrinking" is just the denominator growing. Nothing is pulled from a hat — the whole rule is a consequence of the smooth descent of the curve 3^x through x = 0.

The picture

Negative-exponent slider showing value of three to the x flipping across zeroA draggable point on the curve y equals three to the power of x. To the right of zero the value is greater than one. At x equals zero the value is exactly one. To the left of zero the value drops below one and approaches zero. Readouts at the top show the current exponent, the value three to the x, and a fraction form showing one over three to the positive n. x y 1 3 9 -1 -0.5 0 0.5 1 ↔ drag from +3 down through 0 to −3
Drag the exponent from $+3$ down through zero. At $x = 3$ the value is $27$. At $x = 2$, it's $9$. At $x = 1$, it's $3$. At $x = 0$, exactly $1$. Just below zero, at $x = -1$, the value becomes $\tfrac{1}{3}$ — the reciprocal of $3^1$. At $x = -2$, it's $\tfrac{1}{9}$. At $x = -3$, it's $\tfrac{1}{27}$. The curve never jumps; the "flip" is smooth.

Why crossing zero feels like a flip: the curve is continuous, so the value changes smoothly. But the number line has a natural split: to the right of 1, values are bigger than 1; to the left, they are fractions less than 1. As the exponent crosses zero going left, the value crosses 1 going down — and you can rewrite every value below 1 as "1 over something bigger than 1." That rewriting is the reciprocal rule. Same point on the curve, written with a fraction bar for clarity.

The formula, feel-first

Reading from the slider positions at x = -1, -2, -3:

3^{-1} = \tfrac{1}{3}, \qquad 3^{-2} = \tfrac{1}{9} = \tfrac{1}{3^2}, \qquad 3^{-3} = \tfrac{1}{27} = \tfrac{1}{3^3}.

The pattern is unmistakable: each time the exponent decreases by one, the value is divided by 3. This is the same rule as "each time the exponent increases by one, multiply by 3." You can walk the exponent axis in either direction, and each unit step is "multiply by base" one way and "divide by base" the other. The reciprocal formula is the closed-form answer to the question "what do you get after n divisions starting from 1?"

a^{-n} = \underbrace{\tfrac{1}{a} \times \tfrac{1}{a} \times \cdots \times \tfrac{1}{a}}_{n \text{ times}} = \tfrac{1}{a^n}.

Why the formula is forced on us

Here is the consistency argument that makes the formula unavoidable.

If you know the rule a^m \cdot a^n = a^{m+n} (the product law), you can figure out what a^{-n} must be without a separate definition.

a^{n} \cdot a^{-n} = a^{n + (-n)} = a^{0} = 1.

So a^{-n} is whatever number, when multiplied by a^n, gives 1. That number is the reciprocal of a^n — by the very definition of "reciprocal." So

a^{-n} = \frac{1}{a^n}.

This is not a new definition; it is the only value of a^{-n} consistent with the product law. The slider picture is the visual companion to this algebraic argument: it shows you that the value on the curve at x = -n is indeed the reciprocal of the value at x = +n, geometrically.

Three standings to try

Drag the point to these three positions and read the value.

x = 0.5. The value is \sqrt{3} \approx 1.732. This is a fractional exponent — also in scope for the exponent rules. 3^{1/2} = \sqrt{3} because, by the product law, 3^{1/2} \cdot 3^{1/2} = 3^{1/2 + 1/2} = 3^{1} = 3, and the only positive number whose square is 3 is \sqrt{3}.

x = -0.5. The value is \tfrac{1}{\sqrt{3}} \approx 0.577. By the reciprocal rule, this is \tfrac{1}{3^{1/2}} = \tfrac{1}{\sqrt{3}}. The slider shows you that negative fractional exponents behave exactly like negative integer exponents — they flip the positive version to the other side of a fraction bar.

x = 0. Exactly 1, for every base. This is the dividing line between "value greater than 1" and "value less than 1" on the curve y = a^x — and the reason a^0 = 1 is exactly that it is the crossover point where multiplying and dividing by the base leave you fixed.

A quick computational reflex

Once the slider is in your head, every negative exponent becomes a two-step move.

  1. Ignore the minus sign. Compute the positive version: a^{|n|}.
  2. Put 1 over it.

Example: 2^{-5} = \tfrac{1}{2^5} = \tfrac{1}{32}. Example: 10^{-3} = \tfrac{1}{10^3} = \tfrac{1}{1000} = 0.001. This last one is worth noting — negative powers of 10 are the mechanism of the decimal system for small numbers. When a physics textbook writes 1.6 \times 10^{-19} C for the electron charge, the 10^{-19} is exactly \tfrac{1}{10^{19}}, which is what makes the electron's charge tiny.

Reading scientific notation with negative exponents

The mass of a proton is about 1.67 \times 10^{-27} kg.

Using the reciprocal reflex, 10^{-27} = \tfrac{1}{10^{27}}, so the mass is

\frac{1.67}{10^{27}} \text{ kg} = 0.00000000000000000000000000167 \text{ kg}.

That is twenty-six zeros after the decimal before the digit 1. The negative exponent is the compression that makes this number readable. Without it, you would need a paragraph to write the proton mass; with it, a single symbol.

Carry-away

Related: Exponents and Powers · Exponent Slider: Watch 2^x Sweep Through 1/8, 1/4, 1/2, 1, 2, 4, 8 · Fractions and Decimals · Operations and Properties