Mathematical models create discipline. They ensure that we specify our assumption and that our conclusions then follow from our assumptions. Statistics then provide us with indispensable tests of our theories.
But we need to always remember that data and statistical tests never prove a theory. Typically, many different theories can explain almost any observed phenomenon. Data allows us only to reject a theory. The theories that survive are those that haven’t been rejected yet, and that’s a good reason for humility.
There is a strong predisposition within economics to emphasize individual freedom. Our theories start with the assumption that giving individuals more choices is a good thing — and that assumption leads to the view that people benefit from having more money or lower prices for the goods that they buy.
Edward GlaeserThat assumption doesn’t mean that all regulation is bad or even that, in some cases, people are better off facing fewer soups on a supermarket shelf. Even though people value more choices, they also value information, and an overload of choices can make it hard to figure out which soup is really best. But our assumptions do put freedom first… with a longstanding tendency to view the interests of the government as being distinct from the welfare of the people… Both markets and governments are quite imperfect, and it is important to weigh their failures against each other.