Organizations. Don't. Have. Goals. (A rant.)

It's a truism, but it's true.  Firms and other organizations are arenas of conflict as well as cooperation; witness any salary negotiation.  To assume that a firm has a particular goal is to flaunt reality from the first; it is to obfuscate meaning rather than illuminate (and the third time someone does it, we should question his motives).  Sure, there may be some goal that everyone classified as part of a particular organization shares, but that would have to be empirically demonstrated (which may not even be technically possible; what do you do to prove it, take a survey?).  Even if we want to assume, for the sake of argument or model-making, that everyone in an organization shares a particular goal and many people in the organization share additional goals...well, can't we, shouldn't we, mustn't we say exactly that?

There is an economist joke that amuses me:  Two economists walk by a car dealership; one gazes longingly through the window at a car and says "I want that," and the other economist says "No, you don't."  In that sense, I don't believe that economists want to emulate practice in the physical sciences -- for perhaps the most fundamental way to emulate physicists, biologists, etc., would be to adopt jargon and speak with precision, at least about things we know to be true.  Instead, economists seem more bent on emulating the man in the street; the word "profit", as used by economists, is pretty much how it is used by the guy talking to his god on the street corner.  The sad part is that virtually every trained economist knows better -- and yet, there is hardly anyone in the profession that would dare mount the following response to a speaker:

"Making our standard assumption that firms maximize profits...yes, question from the back row?"

"Are we talking about accounting profits?" 

"No, of course not."

"Then what do you mean by profits?" 

"You know...economic profits."

"And what is that, exactly?  A sort of summed-up change in total cardinal utility across the shareholders, managers, and workers?"

"Uh...I would be glad to talk about this after the presentation, but I am not sure we have time for that discussion at the moment.  Moving on, the dependent variable in our regression is, um, return on investment..."

If I were a journal editor, I probably wouldn't print any empirical work that incorporated the profit-maximizing assumption (I'd allow the vote-maximizing assumption, because it is at least possibly accurate).  I'd circle the assumption in red and write in the margin:  "If we can't get the easy, obvious stuff straight in our heads, we will never answer the hard questions."