Wednesday, May 16, 2012

There are many things I disagree with in the comment section of my optimization post... I thought it was worth collecting them.

PrometheeFeu writes: "I think part of the problem is that this may not be emphasized enough in economics college courses. My international relations courses made a big deal of the fact that our models were just tools to understand the world from the very first day of IA101. My economics courses felt like they were making a big deal that people are rational agents."

I have a couple thoughts. The first is agents are rational in the colloquial sense of the term. Some people try to change this into meaning that they're emotionless or that they don't have diverse goals, but I think that's a misleading use of the word. Plus economists are usually not using it in the colloquial sense and have something specific in mind, and that's worth keeping in mind too. As for the point generally I of course can only fight anecdote with anecdote - but I've racked up quite a few anecdotes over the years and my experience is completely the opposite. Economists in my experience are very good at emphasizing that their models are just useful abstractions to illustrate a particular mechanism.

S/he continues: "Maybe it's just anecdotal, but then there is the fact that in many studies, econ students do behave much more rationally than their peers, happily violating social norms in the name of optimization."

Selection effects are very important here. Certain types of people choose to take economics. I'd also wager the anecdote is an overstatement. And finally, a lot of the empirical evidence on this comes from experimental economics games. To me, that just says that economics students know the theory behind the games and they know how powerful some anonymous, unrepeated defection can be. Insofar as that's the evidence, I'd be wary of generalizing that to behavior in the real world (which, I understand, kind of shoots a hole into the whole point of experimental economics).

Elliott had the most bizarre comment of all: "Policy setters attempting to "de-complicate" the system based on recommendations stemmed from mainstream economic models -- I imagine this is what pisses everyone off -- the reality of "uncertainty" and "entrepreneurship" by themselves should shine a light upon delusional recommendations stemmed from the notion of state enforced common good"
My experience is that it's usually precisely we economists who are the only ones telling policymakers that they are not capable of achieving this. Come on guys - it's the old Hayek line about how the task of economists is "is to demonstrate to men how little they really know about what they imagine they can design"

Charles Rice says: "I agree right up to the part where you said "to develop testable predictions""
This gets to my long-standing point about explanation versus forecasting (and "prediction" is admittedly an ambiguous word). Forecasting a complex system is very hard, whether it's the weather or the economy. We can do OK at forecasting using a process that Charles describes: throwing a bunch of lags and polynomials of a bunch of variables into a model and seeing what gets spit out. That's often what forecasting amounts to. If you're a trader that has its place, but its not economic science because you're not really learning anything about the processes of the economy. I would not be as caught up with forecasting as Charles's comment is. I'm more concerned with testable predictions about the behavior of data. He goes on to say we should bet on our conclusions... I've never understood the fascination with prediction markets. A few interesting bets here and there always makes things interesting, but I don't see what's necessary about that.

Unlearningecon says: "I think the difference is that there is only one variable that the algae adjust to here, whereas when humans make decisions there are a lot of different influences."

I disagree with the first part of the sentence and agree with the second part. Then he goes on to essentially blame the financial crisis on mainstream economics, which is nuts because mainstream economics is giving us the best explanation now of what all happened.

 Min provides another head-scratcher: "First, evolution is a satisficer, not an optimizer. It's survival of the fit, not the fittest. Herbert Spencer did not get it right. In fact, some of the best evidence for evolution is the lack of optimality."

Right. But Min, you seem to have an odd view of the objective function of a maximization problem. It doesn't have to be "maximize my strength subject to these constraints" - that would be the (false) optimizer view of evolution (as you put it). If the objective function is simply to pass on genes subject to constraints - and particularly if those constraints involve a degree of path dependence - you have a satisficer version of evolution. This is what bugs me about critics of mainstream economics who hate the idea that everyone is motivated by profits, for example. Fine! Put something else in the objective function! Mainstream economics is a framework for approaching a problem. I've never heard anyone say "if you don't think people are solely interested in maximizing profits you can't be a mainstream economist". Dierdre McCloskey, great as she is, makes this mistake a lot.

Min goes on: "When the environment changes at the same rate as the organism, the result can be chaotic. Human systems in which agents are adapting to each other can be chaotic, as well."

Well, right Min! Min is a lot like Charles Rice above... I think half the reason why s/he disagrees with me is because s/he thinks I'm saying something that I'm not.

What takes the cake is that Min shares this link about how Russ Roberts butchers a discussion of biology and economics:

Clearly Min is unaware that I harangued Russ over EXACTLY this post and I made most of the points that Mike makes here. Mike argues that economists should be careful when they talk about biologists. The same is true in reverse (although physicists are the absolute worst when it comes to thinking they know something about social science). Clearly Min is assuming I'm saying things that I'm not saying.


  1. OK, let's not do the financial crisis thing here.

    Actually, I don't know a lot about biology either so I won't bother. What I *will* say is that even if the model the biologists were using worked well, they'd adjust it if they had new evidence as to how people *actually* behaved. Economists seem reluctant to put some basic behavioural constraints into their core models (some, like loss aversion, would be incredibly easy).

    1. That should read: 'how the bacteria* actually behaved.'

  2. a handsome brave bearded man -- or just just unusual or unexpected -- either way thank you DK

    now more to the point

    if economics was kept separate from government (only in the absence of gov't) then I would think you would have a point -- it is not --

    look you do have a right to flip out if Ron Paul became president -- if he was he could put his "minority" economic observations which you think would be disastrous into effect -- a Ron Paul white house in essence does not have any more "right" to encourage and pursue aspects of "deflationary reform" any more than a Obama white house has a right to pursue and encourage "inflationary policy" -- I trust here you get my point

    this being said I personally may have cast vote for Ron Paul because of the sense of control over my own life would exceed the sense of control I previously felt under a Obama Presidency and you may feel the opposite in that situation -- what neither of us will dispute is the level of direct immediate control upon your life that a government can and does exert

    as a economist if you make your observations without denouncing the institutions (perverting your observations by way of direct force) than don't be surprised if the "laymen" counts you as part of the problem rather than part of the solution

    this is why I suspect so many are drawn to Austro-libs not because they agree with block -- hoppe -- murphy and more on every econ observation/recommendation they make but because they say -- we think the state should be abolished and every position we take inherently recognizes a stateless ideal -- it seems people are increasingly coming to the conclusion that in order to increase the control they feel over their own personal situation they need less state -- a trend I personally enjoy and hope to see continue -- "before we explain why people commit mistakes, we must first explain why they should ever be right"

    the I didn't write this bill or when I talked to that senator he ignored this possibility though truthful -- seems like hiding -- lets be honest way less people would care that krugman and murphy may argue about 1920-21 -- Great Depression -- or ABC theory if the consequences of the positions didn't impact the students later doing the research affecting policies encroaching on their lives

    "demonstrate(ing) to men how little they really know about what they imagine they can design" is only helpful if those officials are listening to you beyond using your observations to more than justify their own personal ends -- economics today to many I suspect appears eerily similar to a home -- a asset increasingly looking like a liability

    one more point to be honest I don't know the history of the sameulson soviet PPF -- he probably rationalized his position and it may have made sense he was obviously a smart guy -- that being said I can't even begin to guess what impact his economic views especially concerning the USSR had on US policy -- look economics may be a positive science but you are not positive scientist and that is what pisses everybody off


All anonymous comments will be deleted. Consistent pseudonyms are fine.