When $4,001,483 Is Less Than $4,000,000

What's bigger, the distance between 1 and 2 or 3,987 and 3,988?

If you answered that they're both the same, congratulations. You can subtract.

But Cornell researchers claim that people process large precise numbers (those that aren't rounded off such as 43,497) differently than small numbers or large round numbers (like 2 or 3,000 ).

This has implications in the world of prices for if this short-hand process does exist (as suggested in this book), then people could be incorrectly judging precise prices to be lower than round prices, leading them to raise the amount they're willing to pay for something.

Why might this happen?

In our daily lives we're used to encountering exact numbers, but only in small magnitudes (for example: our weight, favorite cable channel, the time the train leaves, etc). For larger numbers we're accustomed to rounding off. You might know that about 3,000 people died in the terrorist attacks of September 11, 2001, but how many people know the exact number? It's just not practical. (The number is 2,976)

To test for the existence of this "precision heuristic," the Cornell researchers ran a series of experiments. The more interesting of these I'll describe below:

In one test, college students were asked to judge list prices of a house on sale in a nearby city. Students were randomly shown one of two sets of the following prices: $390,000, $395,000, $400,000, $501,298, $505,425, $511,534 and $391,534, $395,425, $401,298, $500,000, $505,000, $510,000. The results are eye-opening: students felt that the precise prices, even though they were all larger than their closest round number pair, were smaller:

The researchers also looked at list and final sales prices of homes in South Florida and Long Island to see if the heuristic manifested itself in the prices people paid for homes. They only looked at the cases were the final price was lower than the list price on the idea that when the final price is above list, there are likely multiple bidders involved which drowns out the effect of the heuristic.

And they found that rounder prices are associated with lower prices:

These results are consistent with the notion that the precision heuristic influences buyers' magnitude judgments; buyers judge more round prices to be higher, and this judgment reduces their [willingness to pay] for the house.

But the amounts we're talking here are very small. In Florida, round numbered list prices reduced the final sales price by 0.72 percent. In Long Island the effect was 0.27 percent.

Another explanation for this phenomenon does exist: When a buyer sees an exact price he might think there is less room for negotiation, hence leading to paying more. The researchers say that they couldn't control for this effect.

The obvious conclusion of all of this? When it's time to sell a home or a car, the nice price is the precise price.

Related:

• Frequency of occurrence of numbers in the World Wide Web (pdf)


• An Empirical Analysis of Price Endings with Scanner Data (pdf)