The wisdom of Berkshire Hathaway's Charlie Munger

May 17, 2007 2:44 PM

“Academic Economics: Strengths and Faults After Considering Interdisciplinary Needs” (.pdf)
posted by Kwantsar (12 comments total)

I am a big fan of Munger's investing history and strategies. He's clearly a successful and intelligent guy. But I have a hard time figuring out what he's driving at in this talk, beyond the simple message "academic economists are ivory tower wankers."
posted by zippy at 3:06 PM on May 17, 2007

Well he's totally right about the Mankiw text. That book is awesome! My intro Econ class in college was the only time I took a boring class that was saved by a great textbook.
posted by grobstein at 3:12 PM on May 17, 2007

Google HTML Conversion
posted by weston at 3:52 PM on May 17, 2007

There's a lot in there to digest. I've never found economics to be of much use in investing, other than in knowing the fundamental principles (supply & demand, tragedies of the commons, etc.). I guess that's part of Munger's point, know the big ideas in every discipline (one of he reasons I like metafilter - eventually you'll come across the big ideas in every discipline).

But applying the more esoteric stuff to the markets just seemed a waste of time. I agree with him that knowing these fundamental concepts helps you see the patterns in the world, and understand how trends might unfold.
posted by Pastabagel at 6:00 PM on May 17, 2007

This is a very nice piece. The large themes are clearly laid out, but (caution) the details seem to be totally screwed up.

I believe this comes from the divergence between what Munger wants you to think is the object of his lecture ("How to be Smart Like Me"), and his actual objective ("What Can I Put into My Listener's Heads that will Profit Me the Most").

He's probably just channeling WC Fields:
"Never give a sucker an even break or smarten up a chump."
posted by hexatron at 7:28 PM on May 17, 2007

Very enjoyable. I always wondered if BH's approach to investing was as simple as they tout - the Washington Post example highlights just how true it is.
posted by datacenter refugee at 9:21 PM on May 17, 2007

This reminded me a little of The Secret Sins of Economics, which I quite liked.
posted by carmen at 9:58 AM on May 18, 2007

The more I study economics, the less I believe in it. I think the worse thing to happen to economics was the to fill it with advanced mathematics and make it seem like a hard science. Being a social science it will never reach the Platonic models of physics:

And, of course, economics is in many respects the queen of the soft sciences. It’s expected to be better than the rest. It’s my view that economics is better at the multi-disciplinary stuff than the rest of the soft science. And it’s also my view that it’s still lousy, and I’d like to discuss this failure in this talk.

and:

Well, Berkshire’s whole record has been achieved without paying one ounce of attention to the efficient market theory in its hard form. And not one ounce of attention to the descendants of that idea, which came out of academic economics and went into corporate finance and morphed into such obscenities as the capital asset pricing model, which we also paid no attention to.

I love Munger! What's most interesting is that this may be the only field in which practitioners and academics differ so completely.

I got all the way to the Niederhoffer section, but I have to run. I have to say a lot of the ideas are similar to Taleb's, in the earlier post I made about his new book, and that I'm currently reading. He calls for greater empiricism in finance/economics and less theorizing. He also spends a great deal of time delving into, what I believe he'd hate me referring to, as behavioral economics. That a lot of decision making is not based in mathematics, but on the over or underestimation of risk based on emotional factors -- more or less.

Great link, thanks.
posted by geoff. at 11:53 AM on May 18, 2007

Not to pick on you, geoff., but I have trouble understanding why popular criticisms of economics so often criticize it as too mathematical. The fact is, you need mathematical models to make sense of the interactions of millions and millions of independently motivated actors. It's not about physics envy. And the other elements of the popular case if anything show that more math is needed. For example, if the advances in behavioral economics tell us something about systematic biases in household decision-making, for example, we still need mathematical models to tell us how each household interacts with all the others -- and the math will be harder, not easier. Likewise, doing good empirical work requires mathematical heavy-lifting (of a different kind, but still). Clever experimental design (like what Levitt does) gets you part of the way, but mostly you need powerful models to make sense of the flood of empirical data.

What I'm trying to say is that the usual popular criticisms of economics should be separated from the criticism that you have to know math to do it right. The math is good.
posted by grobstein at 1:30 PM on May 18, 2007

Please, I would love the chance to feebly explain why I think math is bad. I love math, and wish that math could be used in economics and finance. The problem is that math is too perfect, too much of a platonic ideal. It is akin to using math to construct a house out of bricks. We would use geometric representations to calculate the size of the bricks, how many bricks we need, the space between mortars, etc. The problem is in economics, the slight variations, the indentations and imperfections of each brick has consequences much larger than itself. While chaos theory is a buzzword in finance, I like the findings found in it -- that a very minute change has unpredictable effects. Indeed, it would imply that it is impossible to predict things mathematically. I believe there was an astronomer Renaissance astronomer who once said he could predict all motions of the stars and planets, if he could have data on all the stars and planetary bodies. This is impossible, as is predicting the behavior of every consumer and every supplier. This becomes even more apparent when you look at the asymmetry that one supplier or one consumer can produce on the market as a whole. It takes one Taco Bell in one location with e. coli to send the whole company in the dumps. If cell in our body died, that does not mean the whole system collapses.

It is perhaps best to look at why math is bad when math fails. We can look to the failure of CAPM, Black-Scholes or LCTM as all having great mathematical models, models that worked. Except that the models could not model what they did not know. Models cannot react to regime changes, natural disasters or Wolfowitz stepping down from the World Bank because he got an expensive blowjob. Yet that is the world that we live in, not one of models.

I always liked the Pynchon quote, when asked why he made Gravity's Rainbow so hard, "things shouldn't be easy." If anything math's fault is it is too perfect for a world full of imperfect knowledge. Math can too easily be used in economics to hide or mask large risk.

The examples I used I just made up off the top of my head, and I am sure you can tear them apart with some effort, but I hope the fundamental idea will come across. I feel as if economic papers often hide behind math, perhaps not intentionally, to validate their findings. Perhaps I should start using the term "mathematical sophism", as that probably sums up my feelings with math and economics.
posted by geoff. at 2:18 PM on May 18, 2007

It's clear that, as it stands, economics is a pseudo-science. Mathematical interpretations seem to be very good at predicting what should happen and what did happen, but not so hot on what will happen.

except for these guys, of course, but then they haven't contributed to the knowledge base
posted by Jakey at 3:27 PM on May 18, 2007

except for these guys,

Monkeys on a type writer and such.
posted by geoff. at 3:33 PM on May 18, 2007