Name Dropping
Paul Krugman writes about “Vanishing Dark Matter.”
Robert Waldmann waxes nostalgic.
Ah memories. Brad DeLong and I recall an impromptu debate on dark matter at 1050 Mass ave (NBER) featuring Paul Krugman and Larry Summers (apparently just back from advising Dukakis — some memories aren’t as pleasant as others).
googling (krugman, summers, nber, … “anything at all”) at Brad’s blog is totally pointless. I can only hope that in my whole life I write as much as Brad has written on (krugman, summers, nber, … “anything at all”) .
The scene is that, according to the official statistics, the US has just become a net debtor (or the world’s number 1 net debtor, the US went from 0 to deeper in the hole than any nation had ever been so fast that my head spun (was net debtor see below)). Summers argued that the book value net position was irrelevant since net US capital income was still positive.
Krugman said (and I quote from memory) “well you have to consider that the reason that we have a higher return on our assets is that many of them are FDI so it’s not clear if it is really capital income or fees for insurance and management services. And if people are smart enough to understand that, then they are smart enough to know that zero isn’t an especially important number.”
Ah yes. “zero isn’t an especially important number” up there with “accidental theorist” as a demolition of people who don’t want to bother with math (including arithmetic).
Other good lines. On the other side of the discussion
Larry Summers — “Real analysis — isn’t that mostly about the difference between less than and less than or equal to ?”
Larry Summers — “You do know about the concept of present value and its implications” (subtle but it means that knowing dynamic programming is just knowing what “present value” means).
Robert Barro — “Sometimes a negative cite is more useful than a positive cite — just speaking from casual empiricisms” (means I was Fukayaming when I wrote “Are government bonds net wealth”)
Robert Barro — “Isn’t there an existence result that there is a second best argument for everything.” (means we should assume that the market outcome is the first best or we would have to admit that economic theory will not be useful to policy makers until it is empirically grounded. That is, better bad advice than no advice. Also a true statement. The proof is trivial).
By popular demand by commenters I try to slow it down by farking the Profs after the jump.
Krugman said (and I quote from memory) “well you have to consider that the reason that we have a higher return on our assets is that many of them are FDI so it’s not clear if it is really capital income or fees for insurance and management services.
[sounds pretty clear to me. It is hard to tell if profits sent to the USA from a foreign subsidiary are return on capital or the fruit of the brilliant guidance coming from the US CEO (this was a longggg time ago). In general risky returns are higher, but they are risky. The US has been getting net income from the rest of the world, because our investments are risky. Krugman recently quotes one Gian Maria Milesi-Feretti (who had the dubious pleasure of being a TA along with me once) who claims the USA just took a 2 trillion net capital loss. You can think you are doing fine if you are bearing risk and lucky so far. The point was that returns on US investments abroad are highe than on foreign investments in the USA for a reason and it isn’t that they love us so much that they are throwing money at us]
And if people are smart enough to understand that, then they are smart enough to know that zero isn’t an especially important number.”
[Many people think they can get away with thinking of quanties as “Positive negative and zero” in this case net debtor, net creditor or neither. Sometimes the difference between -1 and 1 is about as important as the difference between 1,000,000 and 1,000,002. This is very confusing to people who don’t like numbers. I discuss an extreme example here
http://tinyurl.com/cuecq6. very very often “zero isn’t an especially important number” is a devastating counter argument to an innumerate argument.
Ah yes. “zero isn’t an especially important number” up there with “accidental theorist” as a demolition of people who don’t want to bother with math (including arithmetic).
Other good lines. On the other side of the discussion
Larry Summers — “Real analysis — isn’t that mostly about the difference between less than and less than or equal to ?”
[the difference between open and closed sets is very important in real analysis. An example of an open set is the set of real numbers less than zero. An example of a closed set is the set of real numbers less than or equal to zero. That difference is of very little importance if the number in question (here zero to remind people of Krugman) is really important for some reason. However, the concepts are very useful in doing math which is very useful to physicists, chemists and Engineers and which has had a huge influence on economists.]
Larry Summers — “You do know about the concept of present value and its implications” (subtle but it means that knowing dynamic programming is just knowing what “present value” means).
[I said it was subtle. The point is that Fresh Water macroeconmists think that a mathematical technique called dyanmic programming is the belly button of the universe or at least of macroeconomics. They think the point of the first semester of a graduate macro course is to teach about this technique (and complain if they have to teach the second semester and I taught the first). The key idea is the “value function” which gives the maximum present value of some stream of income or pleasure or whatever you want as a function of the state (or maximum expected present value if there is uncertainty).
So one can argue that if you know about present value, you can handle dynamic programming.
Frankly, I didn’t notice any difficulty in dynamic programming when I first encountered it. In fact the hard part is proving the existence of a value function and the natural thing to do is to just assume one exists. This works for all problems which can be solved by dynamic programming.
Robert Barro — “Sometimes a negative cite is more useful than a positive cite — just speaking from casual empiricisms” (means I was Fukayaming when I wrote “Are government bonds net wealth”)
The “casual empiricism” means look at me, I’m a superstar because so many people lined up to criticize my papers. Also I am hinting that I knew that they would and wrote the papers to make sure that famous people would feel obliged to criticize me and I would get lots of citations. This is called Fukayaming because the most successful effort ever was Francis Fukayama’s “The End of History” which made everything historic that happened everywhere a refutation of Fukayama who was cited continuously and became very famous.
OK Barro wrote that paper on Ricardian equivalence. The paper included 10 or so footnotes explaining why Ricardian equivalence is not likely. For each footnote there was at least one paper in a top journal presented as a counter argument to Barro. He got a citation and could note that he had already said that.
When someone praises you, it might mean your friend has cited you. When someone criticizes you it means you have scared your adversaries.
A reasonable scam is for two friends to stage a furious academic debate citing each other along the way. ]
Robert Barro — “Isn’t there an existence result that there is a second best argument for everything.” (means we should assume that the market outcome is the first best or we would have to admit that economic theory will not be useful to policy makers until it is empirically grounded. That is, better bad advice than no advice. Also a true statement. The proof is trivial).
[in fact the core asumptions of economics like say rational utility maximisation can’t have any implications without some assumptions about utility functions. The trivial proof is insane policy would be optimal if people are made very very happy by the knoweledge that it was implimented (not the effects just knowing it was done). Barro’s point, which is also valid is that there are people like Stiglitz who can basically always fit the assumptions around the policy so the fact that they can write a model where it is optimal tells us nothing. I agree, but I also think that the mathematical result that lump sum transfers then laissez faire is optimal in a general equilibrium model with asymmetric information complete markets perfect competition and no spillovers is optimal is equally worthless.]