Somewhere in the middle of the first day of a fictional Public Policy graduate economics class:

"Are you saying markets are always efficient?" asked Melanie Garcia. "I just don't believe that. Capitalist firms manipulate consumers for their own ends. Our national poverty rate is a disgrace. That's what markets give us. Is that part of your 'efficiency'?"

Professor Weiss raised his hand to his chin, pondering how to respond this challenge. Melanie clearly thought that a trap had been laid for them, and she was not going to fall into it.

"I like the connection you make, Melanie, between this simple model and the real world. I do not know if I can give you a satisfactory answer today, but let me try. Do you recall the analogy I made in the beginning of class, comparing this introductory discussion with showing someone a half-built model of a car? Do you recall that I did not show it in order to convince anyone that this particular vision is desirable, but rather to get you interested in economic modeling? I meant that. As the course goes on, I may well argue that a market or marketlike allocation process is better than some alternative. In other particular circumstances, I will argue that markets do not serve us well at all compared to government-guided allocation. In either case, I will tell you up front what I am going to argue and why.

"I am not saying that markets are always efficient," he continued. "For example, and this is just my personal opinion, I do not think the national labor market with an involuntary unemployment rate of 10 percent is very efficient. The particular market of this model is efficient, but I have not claimed that the model is a good representation of any real market.

"I am glad that you are willing to entertain the idea that a model can represent a real situation. But every good analyst"—Weiss paused, looking around the room for responses to this implicit challenge—"has the responsibility for critically assessing the appropriateness of a particular model to a particular real situation. This is a matter of evidence and of judgment. My bet is that many of you, unlike Melanie, have been accepting this model without thinking enough about why."

Now Barbara was truly puzzled. Rather than defending himself against Melanie's criticisms, he seemed to be agreeing with her. Am I guilty of accepting this model uncritically? Each part seems to have passed my own "tests" based on common sense; I did think about airfares and use of land for corn growing. And I want to trust this professor. Doesn't he want us to trust him? What is the point of this lesson then, and why should I struggle with all the model's intricacies if it doesn't apply to anything real?

Weiss continued as if he had been reading Barbara's mind. "There are many good reasons why you should understand this particular model backwards and forwards. For one, it is the most common model used in microeconomic policy analysis. For another, different pieces of this model will often be useful, even if the model as a whole is not. Finally, I think there are situations in which the model as a whole is, in fact, better than other alternatives that analysts can reasonably contemplate.

"Getting each of you to the point where you have developed the skills necessary to make these assessments yourself will take some time. Therefore, I hope I can earn and maintain your trust in a limited way. I do not want you to trust me to tell you when a specific model applies; I want you to develop your independent critical abilities. I do hope that you trust me to do my best to help you achieve mastery of microeconomic policy analysis."

He stopped and looked directly at Melanie. She looked satisfied, at least for now.