Why Do MPG Figures Mislead Car Buyers
The fuel economy figure on a label counts distance per gallon, and fuel saved does not rise in step with it. A 2008 experiment, a federal label rule and an economist's estimate show what that changes and what it costs.
Answer to the question
Why do MPG figures mislead car buyers?
Because MPG measures distance per gallon, while a buyer trying to save fuel cares about gallons used. Fuel used falls along a curve as MPG rises, so a gain from 10 to 20 MPG saves more over 10,000 miles than a gain from 20 to 50. In a 2008 study, 75 per cent of participants shown MPG chose the plan that saved less.
Key takeaways
- Larrick and Soll (2008): 171 people chose a fleet replacement plan. Of 78 shown miles per gallon, 75 per cent picked the plan that saved 6,684 gallons rather than 14,035. Of 93 also shown gallons per 100 miles, 64 per cent picked the larger saving.
- Lee (2023, pp. 48 to 49): over 10,000 miles, a truck going from 10 to 20 MPG saves 500 gallons and a car going from 20 to 50 saves 300. In the first Larrick and Soll study, 60 per cent of 77 students ranked replacements by linear MPG gain and 1 per cent by fuel saved.
- A 2011 Federal Register rule required gallons per 100 miles on labels from model year 2013, with reduced prominence relative to MPG.
- Allcott (2013) estimated that the illusion reduces welfare by less than $4 per new vehicle. Camilleri and Larrick (2014) found preference for efficient vehicles highest when economy was shown as gas cost over 100,000 miles.
Fuel economy labels on new liquid-fuelled vehicles in the United States now carry two figures. The familiar one is miles per gallon. The second, gallons per hundred miles, appears with less prominence, and the federal rule that required it called it a complement to the traditional MPG number. The second figure was added because of a finding about how people read the first, and the finding rests on arithmetic that fits in a short table.
Why Ten to Twenty Beats Twenty to Fifty
The second figure exists because MPG misleads in a predictable direction. A large gain on an efficient car looks more valuable than a small gain on a thirsty one, although the thirsty one's gain can save more fuel. Fuel used over a fixed distance is that distance divided by miles per gallon, so plotted against the label figure the result is a curve. The first miles per gallon gained remove a lot of fuel, and each further gain removes less.
The error is easy to make because the number on the label is in the wrong unit for the decision. A buyer who wants to cut fuel use cares about gallons consumed, and the label counts distance per gallon. Richard Larrick and Jack Soll, then both at Duke University's Fuqua School of Business, named the resulting habit the MPG illusion in a paper in Science. It is one instance of a wider tendency, linear bias, which reads a curve as a straight line. A neighbouring failure of proportion is covered in [Denominator Neglect and the Bigger Bowl](https://thewhyweproject.com/articles/denominator-neglect-and-the-bigger-bowl).
The Art of Clear Thinking, by Hasard Lee (2023), sets out the consequence at pages 48 to 49 with two vehicles that each travel 10,000 miles a year. Replacing a truck that does 10 miles per gallon with one that does 20 saves 500 gallons. Replacing a car that does 20 with one that does 50 saves 300. The book adds that even a 100 miles per gallon car would save less than the truck swap, and the arithmetic agrees: that upgrade saves 400. The table sets that pair beside the pair used in the 2008 experiments described below, all per vehicle over 10,000 miles.
| Swap (MPG) | Gallons before | Gallons after | Gallons saved |
|---|---|---|---|
| Truck, 10 to 20 | 1,000 | 500 | 500 |
| Car, 20 to 50 | 500 | 200 | 300 |
| Fleet type A, 15 to 19 | 667 | 526 | 140 |
| Fleet type B, 34 to 44 | 294 | 227 | 67 |
A Fleet, Then a Federal Label
The paper reported several studies. In the first, 77 college students ranked pairs of vehicle replacements. Sixty per cent ordered the pairs by the linear improvement in miles per gallon, and 1 per cent ordered them by the actual improvement in fuel used.
The third study moved the question to a policy choice. A total of 171 people from a national subject pool read about a town with 100 vehicles that got 15 miles per gallon and 100 that got 34, each driven 10,000 miles a year, and chose between two replacement plans. Option 1 replaced the first group with vehicles that got 19 and would save 14,035 gallons. Option 2 replaced the second group with vehicles that got 44 and would save 6,684. Among the 78 participants who saw the plans in MPG, 75 per cent chose option 2.
A second group of 93 saw the same plans with the vehicles also described in gallons per 100 miles, 6.67 for the old type A and 2.94 for type B, and with the outcomes restated in that unit. Of these, 64 per cent chose option 1. Across the two conditions, the share choosing the plan that saved more fuel rose from 25 per cent to 64 per cent. The authors recommended that the United States express fuel efficiency as gallons per 100 miles.
The revised fuel economy label rule appeared in the Federal Register on 6 July 2011. It required gallons per 100 miles for combined city and highway operation of liquid-fuelled vehicles, with reduced prominence relative to the traditional figure. It applied from model year 2013, with voluntary use in model year 2012. The rule's own text uses the term MPG illusion and, in a footnote, cites Larrick and Soll's 2008 paper for it. Duke's Fuqua School of Business describes the 2013 label change as following the authors' research.
Converting by hand takes one division. One hundred divided by 15 is 6.67 gallons per 100 miles, the figure used in the fleet study, and one hundred divided by 34 is 2.94. Supermarket unit prices are a version of the same conversion, and [Why the Big Box Is Not the Cheap One](https://thewhyweproject.com/articles/why-the-big-box-is-not-the-cheap-one) looks at when the bigger pack costs more per unit.
Is the Second Figure Enough?
Larrick went on to ask whether the per-distance figure was the best replacement. With Adrian Camilleri, he ran two online experiments that varied the metric, gas consumption or gas cost, and the scale, 100, 15,000 or 100,000 miles. The paper's title calls metric and scale choice architecture tools (Journal of Public Policy and Marketing, 2014).
Preference for the fuel-efficient vehicle, which the authors attribute to compatibility, anchoring and familiarity effects, was highest when fuel economy was expressed as the cost of gas over 100,000 miles, whether or not the higher purchase price was recouped through gas savings over the distance driven. How a cost is presented also changes how it is felt at the till, as [Why Paying by Card Feels Cheaper](https://thewhyweproject.com/articles/why-paying-by-card-feels-cheaper) shows.
A separate question is what the mistake costs. The economist Hunt Allcott, using a new consumer survey, the Vehicle Ownership and Alternatives Survey, documented consumers reasoning as if fuel costs scale linearly in MPG rather than in fuel used per mile. His counterfactuals suggest that the illusion reduces welfare by less than $4 per new vehicle (American Economic Journal: Economic Policy, 2013). Even a severe but plausible underestimation of the financial benefits of fuel-economy gains, he found, could not account for the consumer welfare gains attributed to fuel-economy standards.
Common questions
Why do MPG figures mislead car buyers?
Because MPG measures distance per gallon, while a buyer trying to save fuel cares about gallons used. Fuel used falls along a curve as MPG rises, so a gain from 10 to 20 MPG saves more over 10,000 miles than a gain from 20 to 50. In a 2008 study, 75 per cent of participants shown MPG chose the plan that saved less.
What is gallons per 100 miles?
It is the fuel a vehicle uses over 100 miles. Dividing 100 by the MPG figure gives it: 15 MPG is 6.67 gallons per 100 miles and 34 MPG is 2.94. United States labels have shown it alongside the older figure since model year 2013.
Does a higher MPG figure always save more fuel?
A higher figure always means less fuel over the same distance, but the saving depends on the starting point. Over 10,000 miles, moving from 10 to 20 miles per gallon saves 500 gallons and moving from 20 to 50 saves 300.
Sources
Larrick, R. P., & Soll, J. B. (2008). 'The MPG illusion.' Science, 320(5883), 1593-1594. doi:10.1126/science.1154983
Lee, H. (2023). The Art of Clear Thinking: A Fighter Pilot's Guide to Making Tough Decisions. London: Torva, pp. 48-49.
U.S. Environmental Protection Agency & National Highway Traffic Safety Administration (2011). 'Revisions and Additions to Motor Vehicle Fuel Economy Label.' Federal Register, 76 FR 39477. federalregister.gov/documents/2011/07/06/2011-14291/revisions-and-additions-to-motor-vehicle-fuel-economy-label
Allcott, H. (2013). 'The welfare effects of misperceived product costs: Data and calibrations from the automobile market.' American Economic Journal: Economic Policy, 5(3), 30-66. doi:10.1257/pol.5.3.30
Camilleri, A. R., & Larrick, R. P. (2014). 'Metric and scale design as choice architecture tools.' Journal of Public Policy & Marketing, 33(1), 108-125. doi:10.1509/jppm.12.151
Duke Today Staff (2008). 'Gallons Per Mile Makes More Sense.' Duke Today, 19 June 2008. https://today.duke.edu/2008/06/gpmfuqua.html
Duke University Fuqua School of Business. 'Change in Labels Could Increase Interest in Fuel Efficiency.' Duke Fuqua Insights. https://www.fuqua.duke.edu/duke-fuqua-insights/labels-fuel-efficiency