May 14, 2024 Reading Time: 4 minutes
Enhancements in automobile safety are among the best-known examples of the substitution effect.

Economics undergraduates learn early on — in their second microeconomics course if not in their very first — the importance of distinguishing substitution effects from income effects. Although jargony, this distinction is important. Without it, a great deal of real-world economic phenomena are misdiagnosed.

Non-economists readily understand income effects. These effects are those caused by changes in purchasing power — that is, by changes in wealth or income. Even seven-year-olds correctly realize that if their parents give them more money, they can buy more M&Ms. Conversely, seven-year-olds also are aware that if they lose some of their money, they can buy fewer M&Ms.

Substitution effects are more subtle, although hardly difficult to grasp. These effects are those caused by changes in the relative attractiveness of different options. If the price of gummy bears rises while that of M&Ms doesn’t change, candy buyers — including seven-year-olds — will purchase fewer gummy bears and, quite likely, more M&Ms. This change in relative prices causes consumers to substitute out of gummy bears and into M&Ms.

The amount of M&Ms people seek to purchase, therefore, isn’t determined only by people’s incomes; It’s determined also by the ease or difficulty of obtaining other goods and services that are related, in consumers’ minds, to M&Ms. Change the availability of gummy bears and you change consumers’ interest in buying M&Ms even without changing consumers’ income. Likewise, change the popularity of Halloween and you also change the demand for M&Ms.

Most commentary on public policy by competent economists focuses on substitution effects. The reason is not that income effects aren’t real or that economists think these to be less important than substitution effects. The reason is that the general public, including politicians, naturally understand income effects but seem unaware of substitution effects. It’s left to economists to explain the reality of these effects. Tracing out substitution effects is one of the great services performed for the public by competent economists.

Although I didn’t use the term, my last column focused on substitution effects. I pointed out the error in the commonly heard argument that hikes in minimum wages won’t cause “rich” companies to change their employment practices. Rich companies, we are told, can afford to pay the higher wage. People who offer this argument take account only of income effects. But when substitution effects are brought into the account, the story changes dramatically. Because raising the minimum wage increases the cost of employing low-skilled workers relative to other options — for example, using more labor-saving machinery — employers substitute away from employing low-skilled labor into other options that are made relatively less costly by the rise in the minimum wage.

Another example of the substitution effect comes from a 1981 study by economists Sidney Carroll and Robert Gaston. These researchers found that the more restrictive is the occupational licensing of electricians, the greater is the incidence of accidental electrocution. To many people, this finding is counterintuitive, but not to people who understand substitution effects.

Without substitution effects, accidental electrocutions would fall as the requirements to practice the occupation of electrician rise. After all, the stricter the licensing requirements, the higher the average quality of professional electricians. But substitution effects are ever-present. States with very strict requirements on working as an electrician naturally have fewer electricians than they’d have if their requirements were less strict. And so while the average quality of licensed electricians in these states is higher than in states with less-stringent requirements, the cost of hiring a professional electrician is higher in the strict-requirement states. Homeowners and small-businesspeople in strict-requirement states are thus more likely than are their counterparts in states with less-strict requirements to substitute either into do-it-yourself electrical work or into sticking longer with older electrical wiring — and, therefore, raise the risk of accidental electrocutions.

What economists now call the Peltzman effect is yet another example of substitution effects. Named after University of Chicago economist Sam Peltzman, the Peltzman effect occurs when the changing riskiness of engaging in some activity causes people to change their behavior in ways that offset, partially or fully, the change in riskiness.

My late, great George Mason University colleague Gordon Tullock offered what is perhaps the most vivid example of the Peltzman effect. Gordon famously observed that government could, without outlawing driving, immediately reduce the number of traffic fatalities to near-zero by taking just one simple step, namely, mandate that a steel dagger be fitted onto every steering column and pointed at each driver’s heart. Outfitting automobiles with these daggers would so raise the riskiness of driving that less driving would occur, and that which did occur would be done with the utmost care.

The beauty of Gordon’s steel-dagger hypothetical is that, in its vividness, it serves as a springboard for revealing what is equally true, but less obvious, for lesser and more realistic changes in risk. While everyone sees that drivers drive less cautiously without daggers pointed at their hearts than with the daggers installed, too few people naturally see that drivers also drive less cautiously when other, more modest improvements in automobile safety are put in place. The steel-dagger hypothetical positions the economist to ask: “If making driving more safe by removing steel daggers will cause drivers to drive less cautiously, won’t making driving more safe by, say, installing shoulder harnesses and airbags have a similar effect on drivers?” I can attest from many years of using this example with my undergraduate students that the point is made effectively.

And the point here is not that technology or government regulation that reduces the risk of serious injury while driving is a bad idea. Instead, the point is that, because of substitution effects, we should always be aware that outcomes quite different from those that seem most obvious are possible. Mandating greater automobile safety might reduce traffic fatalities. Or it might not — or not by enough to justify the cost of the additional safety.

Society abounds with substitution effects. In our daily lives we routinely do such substitutions without thinking about them, as when the child buys more M&Ms after the price of gummy bears increases, and when the homeowner personally installs her new electrical outlet when the price of hiring a professional electrician rises.

Unfortunately, the reality of substitution effects is too often ignored by politicians and regulators, as when they raise minimum wages under the mistaken assumption that ‘rich’ companies will respond by doing nothing other than pay the higher wages by dipping into their cash reserves.

Donald J. Boudreaux

Donald J. Boudreaux

Donald J. Boudreaux is a Associate Senior Research Fellow with the American Institute for Economic Research and affiliated with the F.A. Hayek Program for Advanced Study in Philosophy, Politics, and Economics at the Mercatus Center at George Mason University; a Mercatus Center Board Member; and a professor of economics and former economics-department chair at George Mason University. He is the author of the books The Essential Hayek, Globalization, Hypocrites and Half-Wits, and his articles appear in such publications as the Wall Street Journal, New York Times, US News & World Report as well as numerous scholarly journals. He writes a blog called Cafe Hayek and a regular column on economics for the Pittsburgh Tribune-Review. Boudreaux earned a PhD in economics from Auburn University and a law degree from the University of Virginia.

Get notified of new articles from Donald J. Boudreaux and AIER.