51 pages • 1 hour read
A cause is something that leads to something else. Causes of people’s actions often are hard to discern because it can be tough to distinguish between a cause and a correlation—a connection between two things that might merely be a coincidence or might be a signal that a third thing is causing the first two. For example, the US crime rate dropped sharply in the late 1990s, and many experts believed the cause was more and better policing, or stricter sentencing, or the aging of a population of impulsive young men. The authors, however, believe the crime drop occurred partly because legalized abortion removed a large cohort of unwanted children (who often end up as criminals) from a population that, when it reached maturity, contained fewer lawbreakers. This cause thus was hidden behind several correlations that merely appeared to be causative.
The book defines correlation as “nothing more than a statistical term that indicates whether two variables move together” (163). Correlations don’t mean one variable causes the other, just that they tend to occur alongside each other. Sometimes two variables are correlated because one variable does cause the other, but not always, and proving it can be difficult. The public, however, often confuses correlation with causation.
A cheap alternative to powdered cocaine, crack cocaine was developed in the 1970s by South American drug producers, who discovered that “mixing powdered cocaine in a saucepan with baking soda and water, and then cooking off the liquid, produced tiny rocks of smokeable cocaine. It came to be called crack for the crackling sound the baking soda made when it was burned” (107-08). The high was quick but short; buyers quickly came back for more. US gangs sold the drug in local communities, causing vast new social problems. Crack was so addictive that damaged Black neighborhoods, which had otherwise made great strides in the previous 40 years. The murder rate among young Black men spiked, mainly because drug-selling gangs fought over prime sales territories.
Incentives are rewards or penalties for doing something. Examples include bonuses for good work, fines for running red lights, price changes, and changes in the availability of resources. Economists believe people make decisions based on incentives, of which there are “three basic flavors […] economic, social, and moral” (17). Businesses, governments, and institutions often design incentives to get people to change their behavior, but these attempts can backfire, as in the example of the fines for tardy child pickups at Israeli daycare centers that caused the lateness to get much worse.
Regression analysis is a method of teasing out variables from large data sets, comparing them one at a time to each other, and graphing out their relationships. The purpose is to find variables that cause changes in other variables, and to discern variables that don’t affect others. This is especially useful in large data sets with lots of variables: “Regression analysis is the tool that enables an economist to sort out these huge piles of data. It does so by artificially holding constant every variable except the two he wishes to focus on, and then showing how those two co-vary” (163). The power of regression analysis in economics is the ability to unearth often-surprising connections between behaviors and incentives, connections that might otherwise go unnoticed.
Plus, gain access to 8,800+ more expert-written Study Guides.
Including features: