Assorted links

1. Stephen Williamson on AD-AS models, read the bottom of his post.

2. How and why regulators multiply.

3. What do they read in prison?

4. Will religion die out in some countries?  Dubious claims based on extrapolation and then called a “model.”

5. Passing the Turing test on Twitter?

6. Markets in everything: eleven creative pencil sharpeners.

7. Tax cuts do not starve the beast.


"...and what amazed me was the sheer number of regulators I dealt with in the course of applying for approval."

In many analogs I think one step would be to internalize the coordination function. When I say analogs, an anecdote: I've relayed this story before, but when my child was born we chose to bank the cord blood, and while our experience was almost identical to that in Michael Lewis's "Home Game" with seemingly thousands of nurses bugging the shit out of us constantly, I was on my own to find the shipping/receiving dock so that I could sit and wait for my child's precious stem cells to be picked up by the Fed Ex man.

Tyler: Are you aware of an analogous paper to Romer's that tests whether tax increases feed the beast? (My basic prior of pessimism says that they probably do as well, but it may well be that there's no such thing as an exogenous tax increase)

With respect to n.4: so they just draw the curve, fit a trend and extend it?? God, physicists should stop doing economics.

No, they come up with a slightly sophisticated logistic model and test whether it fits the data. This is not the same as drawing a curve, fitting a trend, and extending it. Did you even read the paper before you criticized it?

"God, physicists should stop doing economics." and statistics. Listening Dr. Mann?

The people in the fourth link are the same ones that in 2005 said house prices would continue to rise.

>Dubious claims based on extrapolation and then called a “model.”

Works for the climate guys. Why stop there?

"Larry Sullivan, a John Jay College of Criminal Justice professor and expert on prison reading. “It’s easy. It’s like candy. But reading almost anything is better than reading nothing. You’d rather have someone reading a novel than hanging around the yard stabbing someone.”"

I'm glad we got an expert to make that claim.

#4 is a great example of journalists who don't comprehend the subjects they're writing about. I've read the paper, and -- as one might expect -- it's much more restrained and much less bombastic than the article makes it out to be.

Advice to future commenters in this thread: before you criticize, read the paper so you don't erect a strawman.


The paper relates past tax rate changes to past government size changes and concludes there's no relation. It is fatally flawed, just like the MBS models that used 10 years of data in 2007. It's pretty clear that there is a threshold somewhere: below it you're fine, above it everything crashes. The beast isn't starved gradually: the beast is starved when a treasury auction fails (or maybe when inflation is running at ~50%). The "starve the beast" technique has been gloriously shown to be wildly successful in Greece, and there's no reason to think it's going to be less successful elsewhere.


This begins to get close to my pedestrian notion of the successful Turing test: when a computer/bot/software program is able, without programmed instruction or guidance, to cheat or lie--by which time it will be doubtless far too late for pedestrian truthtellers . . . .

They are avoiding taxes. In Germany, you pay a tax that go to fund your church unless you left the Church

Comments for this post are closed