What it means
An arithmetic mean adds the values and divides by how many there are, which works when the underlying quantities simply stack up. The harmonic mean is built differently: it averages one divided by each value, then takes one divided by that result, which is the right approach when the values are rates expressed as something per something else.
The reason it matters is that the arithmetic mean answers a different question. If you buy a fixed dollar amount of shares each month, the average price you paid is not the average of the prices, because your fixed budget bought more shares when the price was low and fewer when it was high.
The same logic applies across finance and operations. Average cost per unit across suppliers when you spend the same amount with each, average price-to-earnings multiple across an index, and average speed over equal distances are all harmonic mean problems rather than arithmetic mean problems.
The harmonic mean is always less than or equal to the arithmetic mean of the same positive numbers, and the gap widens as the values become more spread out. That is exactly why it is used for valuation multiples, where a single very high multiple would otherwise drag a simple average upwards and misrepresent the group.
There is a weighted version too, used when the amounts behind each rate differ. It divides the total weight by the sum of each weight divided by its rate, which is the form most commonly applied to index-level price-to-earnings ratios.
One practical caution is that the harmonic mean cannot handle zero or negative values, because dividing by zero is undefined and negatives distort the reciprocals. In valuation work this is why loss-making companies are excluded from an aggregated multiple rather than included with a negative figure.
In practice
Real-world examples.
Example
A fund manager reports an average price-to-earnings ratio for a ten-stock portfolio using a weighted harmonic mean rather than a simple average. One holding trades on 90 times earnings, and a simple average would have implied the whole portfolio was expensive. The harmonic figure reflects the actual ratio of total portfolio value to total portfolio earnings, which is what the investment committee wanted to see.
Example
A haulage business measures average fuel efficiency across a route that its trucks run in both directions. Because each leg covers the same distance at different average speeds, the operations analyst uses a harmonic mean to combine them. Using a simple average would have overstated efficiency and understated the fuel budget for the year.
Example
A procurement team buys the same $40,000 of packaging from three suppliers whose unit prices differ. To report a genuine blended cost per unit, the analyst divides total spend by total units received, which is the weighted harmonic mean of the three prices. The figure comes out below the simple average of the quoted prices, because the fixed budget bought more units from the cheapest supplier.
Formula
Calculation
The simple form, for n values:
Harmonic mean = n / (1/x1 + 1/x2 + ... + 1/xn)
An investor puts $6,000 into the same share every month for three months. The share price is $50 in month one, $60 in month two and $75 in month three, and the question is what average price per share was actually paid.
Applying the formula: 1/50 = 0.02, 1/60 = 0.016667 and 1/75 = 0.013333, which sum to 0.05. The harmonic mean is 3 / 0.05 = $60.00 per share.
Checking it directly confirms the result. The $6,000 buys 120 shares at $50, 100 shares at $60 and 80 shares at $75, a total of 300 shares for $18,000 spent, and $18,000 / 300 = $60.00 per share.
The arithmetic mean of the three prices is ($50 + $60 + $75) / 3 = $61.67, which overstates the true average cost by $1.67 per share, or $500 across the 300 shares held.Case study
Seen in the real world.
Thornbury Asset Partners is a fictional boutique investment firm invented for this illustrative case. Its quarterly client report showed the average price-to-earnings ratio of its flagship equity fund as a simple average of its 25 holdings, which came out at 31 times earnings and prompted several clients to ask whether the fund had drifted into expensive territory.
The analyst who rebuilt the calculation found that two small holdings on very high multiples were pulling the simple average sharply upward, even though together they made up under 4% of the fund. Recalculating on a weighted harmonic basis, dividing total fund value by total attributable earnings, produced a figure of 19 times earnings, which matched how the portfolio actually behaved.
Thornbury changed its reporting standard, disclosed the change and the reason in the following quarterly letter, and added a note explaining that loss-making holdings are excluded from the multiple entirely. The illustrative lesson was that the choice of average was not a technical footnote but the difference between a client thinking the fund was expensive and understanding what they owned.
Watch out
Common mistakes.
- Averaging ratios with an arithmetic mean out of habit, which systematically overstates the result whenever the underlying values are spread out.
- Including zero or negative values, which either breaks the calculation outright or produces a figure with no sensible interpretation.
- Using the unweighted version when the amounts behind each rate differ, since equal weighting only makes sense when the money or distance behind each value is genuinely equal.
Questions
People also ask.
When should I use the harmonic mean instead of the arithmetic mean?
Use it when the values are rates and the numerator is what stays constant, such as a fixed spend across different unit prices or a fixed distance across different speeds.
Is the harmonic mean always lower than the arithmetic mean?
Yes for any set of positive numbers that are not all identical, and the two are equal only when every value is the same.
Where does it show up outside investing?
It underpins blended unit costs in procurement, average speed over equal distances in logistics, and the F1 score used to combine precision and recall in analytics work.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%