Skip to main content
Heuristics

THINKING STRAIGHT, GO TO AN EXTREME CASE

Test an idea at an extreme value or under an extreme assumption. The result can expose a mistaken answer, clarify how a concept works, or reveal a decision’s limits. Best- and worst-case scenarios are one use of this heuristic.

THE IDEA

THINKING STRAIGHT, GO TO AN EXTREME CASE

Mary and Jim want to paint a room together. If Mary painted alone it would take her 2 hours, and if Jim painted alone it would take him 3 hours. How long would it take to paint the room if they paint together? Many high school students (and adults) quickly answer 2.5 hours when they encounter this kind of problem for the first time. Take a minute to think how you would explain to a high school student why this is not the correct answer.

One effective way to see why 2.5 hours is not the right answer is to imagine that Mary alone paints and Jim just watches. How long will it take? Two hours. If they can work independently at unchanged rates, Jim’s help must reduce the time below two hours. This is an example of going to an extreme to get your thinking straight. In this case, the extreme we used is that Mary paints and Jim does nothing.

Of all the maxims in this book, this is the most popular among Richard’s coauthors and former students. Broadly speaking, this maxim can be especially helpful for two kinds of situations:

  • To understand a concept, problem, or idea better; or
  • To assess the best-case or worst-case consequences of a decision you are struggling to make.

To understand a concept or idea better

Students new to economics sometimes struggle to grasp the concept of the marginal propensity to consume (MPC). MPC is the proportion of additional disposable income (income after taxes and transfers) that an individual consumes. Even harder to understand is what a specific numerical value for MPC implies. In such situations, it helps to go to extreme cases to gain some intuitive feel for what the numbers mean. In this simplified example, consider MPC values between 0 and 1. An MPC of 0 means that if an individual received an extra $100 of income, she would spend none of it. Think of Warren Buffett. He already has plenty of money to consume what he wants. If Berkshire Hathaway were to pay him an extra $100, he would not change his consumption. At the other extreme, an MPC of 1 means that if an individual received an extra $100 of disposable income, he would spend it all. Think of someone in extreme poverty. An extra $100 could go immediately to satisfy his basic needs. These extreme cases can help ground our understanding of MPC.

Understanding a concept sometimes involves addressing and correcting a misconception. Barry Nalebuff, one of Richard’s coauthors, uses this when teaching negotiation. He noticed people often default to proportional division as a fair solution. For example, imagine Division A of a company is spending $20,000 on a software package and Division B is spending $10,000, but they can buy a joint license for $24,000. Most people propose A pays double what B pays. Cooperation saves $6,000, and neither division can obtain the joint license alone. Nalebuff uses that dependence to argue for splitting the savings equally, rather than treating proportional payment as the only fair rule. Barry gives an extreme example between a small iced-tea startup (David) and a large beverage company (Goliath) to show the absurdity of proportional division.

Going to an extreme case can help discover the key insight of a problem. Karen Eggleston used it when studying demographic transition: “Would immortality be bad for the economy?” This reframing helped highlight the imbalance between work lives and retirement.

Another manifestation is learning from outliers. Jonathan Nelson, studying Renaissance artwork prices, used the extreme cases of Raphael’s paintings to understand typical pricing mechanisms.

To assess the best-case or worst-case consequences of a decision

Several practitioners have used this maxim in business and policy:

  • Joachim Neipp (Bayer) built financial models for mergers and drove them to extreme scenarios to assess risk.
  • Xiaochen Fu (Agricultural Bank of China) considered extreme cases of services that cannot be delivered online to decide which bank branches to close.
  • Yinglan Tan (venture capital) uses worst-case scenarios during crises to ground discussions and provide a sense of security.
  • Jack Donahue (U.S. Department of Labor) used an extreme case to propose the Lifelong Learning Tax Credit: accept that most subsidies would go to college tuition and focus on lower-income families.

An extreme case can rule out an answer or expose an assumption. It does not replace estimating what is likely under the actual conditions.

When to use it

  • When teaching a quantitative concept, use extreme values to help students build intuition (e.g., MPC, elasticity).

  • In a negotiation, propose an extreme scenario to test the fairness of a proposed division.

  • Modeling severe downside scenarios to see where a risky business decision could exceed the organization’s capacity to absorb losses.

  • In policy design, consider an extreme version of a proposal (e.g., full subsidy with no restrictions) to identify unintended consequences.

FROM THEORY TO EVERYDAY LIFE

In practice

01

Mary and Jim painting a room

If Mary alone takes 2 hours and Jim alone takes 3 hours, averaging gives 2.5 hours. Mary working while Jim watches would take 2 hours, so independent work at unchanged rates must beat 2 hours. Under those assumptions, adding their work rates gives a completion time of 1.2 hours.

02

Marginal propensity to consume (MPC)

MPC values between 0 and 1 are abstract. By considering extremes: MPC=0 (Warren Buffett spends nothing extra) and MPC=1 (someone in extreme poverty spends all extra income), we ground understanding of the concept.

03

Negotiation between David and Goliath

In the source's example, a small iced-tea startup and a large beverage company could save $4 million by buying bottles jointly. A proportional split based on a 2000:1 size ratio gives almost all the savings to the larger firm. The example supports considering equal division of the additional savings because both parties are needed to obtain them.

04

Raphael's extreme prices

Most Renaissance altarpieces cost around 100 florins, but Raphael's three major works cost 850-1200 florins each. These outliers help understand superstar artist compensation.

05

Bank branch decisions

The source describes a bank with 300,000 staff and 25,000 branches considering a digital transition. Services that required a physical presence, such as collecting physical gold, helped identify limits to closing branches.

06

Policy design for worker training

When designing a tax incentive for worker training, officials struggled to distinguish training from college tuition. Considering the extreme—accept that most subsidies would go to college—led to the Lifelong Learning Tax Credit targeted at lower-income families.

✳

A MOMENT TO REFLECT

Put it to work.

  1. Think of a problem you recently solved. Would considering an extreme case have given you a better intuition?

  2. What is the best-case and worst-case outcome of a decision you are currently facing? How do those extremes inform your choice?

  3. Identify a concept you find difficult. What are its extreme values (e.g., 0 and 1 for a proportion) and how do they clarify the concept?

Connected ideas

Another lens can bring the same problem into focus.

Often confused with

References

  1. Nalebuff, Barry (teaching negotiation example) — Barry Nalebuff
  2. Eggleston, Karen and Victor Fuchs (demographic transition) — Karen Eggleston and Victor Fuchs
  3. Nelson, Jonathan and Richard Zeckhauser (Raphael, Superstar) — Jonathan Nelson and Richard Zeckhauser
  4. Donahue, Jack and Richard Zeckhauser (public-private collaboration) — Jack Donahue and Richard Zeckhauser

CONTINUE THE CONVERSATION

Source texts

  • Maxims for Thinking Analytically The wisdom of legendary Harvard Professor Richard Zeckhauser (Dan Levy)
← More heuristics