The Secret Sins of Economics

Rating:
Summary: "Cassandra of Economics" and the statistical Trojan horse
Review: Are Economists sinners? And if they are, what are their sins? In this tiny booklet (at 58 pages of text, this must be one of the worst ratios of words per cents available on Amazon), Dierdre McCloskey argues that they are, but that their sins are not what most humanistic critiques think.

I've recently praised a Paul Krugman book as the best written book about economics I've read - I already rue my words, because it hasn't been three months yet, and Professor McCloskey, an artist of prose, surpasses him. Even if you disagree with everything McCloskey says, reading this book is a delight. When satirizing quantification in economics (which she approves, by the way):

"And the economists, oh, the economists, how they counted, and still count. Take any copy of The American Economic Review (Surely you subscribe?) and open it at random. To perhaps Joel Waldfogel, "The Deadweight Loss of Christmas" (No Kidding: December 1993; Waldfogel is arguing that since a gift is not chosen by the recipient it is not worth what the giver spent, which leads o a loss compared with merely sending cash. Who could not love such a science of Prudence?)" (p.6)

You could be mistaken into thinking that McCloskey is against quantification, statistics or mathematics, but she merely cannot resist some highly amusing cheap shots. McCloskey is in favor of Quantification, use of mathematic models, and of the libertarian bias of economists (not very convincingly in the case of the latter, in my opinion).

There are some other, minor sins which are not really the target of McCloskey, but to which she devotes a great deal of her time. So by the time we get to "The Two Real Sins, Almost Peculiar to Economics" it is already page 37. Time's almost up, and the sins are two:

First, McCloskey believes that the theorizing economists do in their theoretical work is really no theorizing - the theories include no place for quantifications, and so remain abstract "If A then B" may be entirely consist in and of itself - but it may be entirely wrong, if A is not true. If the theories are not open to quantification, they can hardly be falsified, nor can we estimate whether their effect is considerable or marginal.

The other sin of economics is the alleged reliance of economists on "statistical significance". A high statistical significance assures us that the result we get is not noise at say 5%, 1% or 0.1% levels, meaning that there's only 1-in-twenty, 1-in-a-hundred, or 1-in-a-thousand chance that the result one gets is accidental. McCloskey's point is that even if a theory fails the 5% level, for example, it does not mean that the theory isn't true (it just means that you're not 95% sure it is true). That point goes the other way around, too - if you measure enough things, you're going to find some which will seem to be correlated even though they clearly aren't - "For a long time in Britain the number of ham radio operator licenses granted annually was very highly correlated with the number of people certified insane. Very funny. So?" (p. 53).

I admit that I lack familiarity with current economic research to comment on how close McCloskey's criticism hits the mark. In the last page of "The Secret Sins of Economics" Deirdre McCloskey compares herself to Cassandra, the prophetess from Troy, whose advice was ignored by the Trojans as they brought the famous horse into their city. Amusing, yet economics is a science, not a grand retelling of an old myth, and McCloskey is not cursed by Apollo. At least some of the guilty, 'Samuelsonian' economists must have responded to McCloskey's critiques, but if you want to hear those answers, you will have to look elsewhere.

Rating:
Summary: "Cassandra of Economics" and the statistical Trojan horse
Review: Are Economists sinners? And if they are, what are their sins? In this tiny booklet (at 58 pages of text, this must be one of the worst ratios of words per cents available on Amazon), Dierdre McCloskey argues that they are, but that their sins are not what most humanistic critiques think.

I've recently praised a Paul Krugman book as the best written book about economics I've read - I already rue my words, because it hasn't been three months yet, and Professor McCloskey, an artist of prose, surpasses him. Even if you disagree with everything McCloskey says, reading this book is a delight. When satirizing quantification in economics (which she approves, by the way):

"And the economists, oh, the economists, how they counted, and still count. Take any copy of The American Economic Review (Surely you subscribe?) and open it at random. To perhaps Joel Waldfogel, "The Deadweight Loss of Christmas" (No Kidding: December 1993; Waldfogel is arguing that since a gift is not chosen by the recipient it is not worth what the giver spent, which leads o a loss compared with merely sending cash. Who could not love such a science of Prudence?)" (p.6)

You could be mistaken into thinking that McCloskey is against quantification, statistics or mathematics, but she merely cannot resist some highly amusing cheap shots. McCloskey is in favor of Quantification, use of mathematic models, and of the libertarian bias of economists (not very convincingly in the case of the latter, in my opinion).

There are some other, minor sins which are not really the target of McCloskey, but to which she devotes a great deal of her time. So by the time we get to "The Two Real Sins, Almost Peculiar to Economics" it is already page 37. Time's almost up, and the sins are two:

First, McCloskey believes that the theorizing economists do in their theoretical work is really no theorizing - the theories include no place for quantifications, and so remain abstract "If A then B" may be entirely consist in and of itself - but it may be entirely wrong, if A is not true. If the theories are not open to quantification, they can hardly be falsified, nor can we estimate whether their effect is considerable or marginal.

The other sin of economics is the alleged reliance of economists on "statistical significance". A high statistical significance assures us that the result we get is not noise at say 5%, 1% or 0.1% levels, meaning that there's only 1-in-twenty, 1-in-a-hundred, or 1-in-a-thousand chance that the result one gets is accidental. McCloskey's point is that even if a theory fails the 5% level, for example, it does not mean that the theory isn't true (it just means that you're not 95% sure it is true). That point goes the other way around, too - if you measure enough things, you're going to find some which will seem to be correlated even though they clearly aren't - "For a long time in Britain the number of ham radio operator licenses granted annually was very highly correlated with the number of people certified insane. Very funny. So?" (p. 53).

I admit that I lack familiarity with current economic research to comment on how close McCloskey's criticism hits the mark. In the last page of "The Secret Sins of Economics" Deirdre McCloskey compares herself to Cassandra, the prophetess from Troy, whose advice was ignored by the Trojans as they brought the famous horse into their city. Amusing, yet economics is a science, not a grand retelling of an old myth, and McCloskey is not cursed by Apollo. At least some of the guilty, 'Samuelsonian' economists must have responded to McCloskey's critiques, but if you want to hear those answers, you will have to look elsewhere.

Rating:
Summary: Truly a must read.
Review: Short, funny, earth shattering. Most thinking people know in their hearts there is something seriously wrong with economics. Dr. McClosky shows what is wrong clearly and with rigor.

In particular any one with a science background will delight in her demolition of the bad science that characterizes so much of economics.

This is one of the best pieces I've read on any topic in the past decade.

Rating:
Summary: Truly a must read.
Review: Short, funny, earth shattering. Most thinking people know in their hearts there is something seriously wrong with economics. Dr. McClosky shows what is wrong clearly and with rigor.

In particular any one with a science background will delight in her demolition of the bad science that characterizes so much of economics.

This is one of the best pieces I've read on any topic in the past decade.