Hacker News Reader

Comment by JumpCrisscross | Hacker News Reader

> Economists are known for constructing theories that sound plausible to the typical person, but which are subtly geared to concentrate wealth at the top

Milgrom wrote the book on auction theory, a subcategory of game theory, a domain of mathematics. Your criticism is with a consulting firm started by an economist. Not economics per se let alone mathematics.

Even then, your critique is unsustained. Milgrom is saying Google ran fair auctions. I don't believe anyone has put forward evidence to suggest it didn't. The point isn't that he's wrong; it's that he's irrelevant. Google's antitrust problem isn't in it running unfair auctions, but being the only auction house.

> "trickle-down economics"

Not economics [1]!

> Economists core moral value is maximum efficiency

Karl Marx was an economist.

[1] https://en.wikipedia.org/wiki/Trickle-down_economics#History

View on HN

> Milgrom is saying Google ran fair auctions. I don't believe anyone has put forward evidence to suggest it didn't.

...and the idea that running fair auctions are subtly geared to concentrate wealth at the top is silly. Yes, there is a well understood principle that fair auctions will out-compete unfair auctions in the long run, but whether that concentrates wealth at the top is a consequence of a lot of other factors that are entirely external to the auction dynamics.

> there is a well understood principle that fair auctions will out-compete unfair auctions in the long run

In what sense? If anything, the market failure of non-competitive markets almost implies the reverse: power prefers unfair auctions.

> whether that concentrates wealth at the top is a consequence of a lot of other factors

Correct. (And as an aside, sealed-bid auctions are particularly good for small bidders or bidders in an adversarial market, the latter for the same reasons a secret ballot is better for small parties.)

> No, there isn’t. If anything, the market failure of non-competitive markets almost implies the reverse: power prefers unfair auctions.

I do not follow your reasoning here. What do you mean by "the market failure of non-competitive markets"?

Seriously, this stuff is econ 101. Ultimately auctions are about bringing together as many buyers and sellers as possible to produce as much liquidity as possible. Fair auctions naturally attract more buyers and sellers than unfair auctions. If the auction is unfair to say, buyers, then you'll have fewer buyers using them than auctions that are fair or favour buyers, which in turn means that sellers will get access to fewer buyers, so they'll go somewhere else.

> What do you mean by "the market failure of non-competitive markets"

You’re going off about econ 101 but do not recognise the term market failure [1]?

> Fair auctions naturally attract more buyers and sellers than unfair auctions

Citation needed. (Plenty of examples of open-bid ascending-price auctions beating better systems due to familiarity, for instance.)

More fundamentally, auctions are subject to network effects. They benefit from anchor participants, i.e. repeat participants of substantial size. Those anchors tend to want perks for participating. The entire domain of market microstructure and ECNs rebating market makers is a story of broad participants’ desires for a fair system competing with narrow participants’ desires for a biased one. This effect is exacerbated when one side of the market is repeat or big while the other is ephemeral or small.

[1] https://en.m.wikipedia.org/wiki/Market_failure

One can recognize a term but be confused by a statement in which the term is used and how it is relevant to the discussion.

> Plenty of examples of open-bid ascending-price auctions beating better systems due to familiarity, for instance.

Yeah, I think we're talking past each other. I wasn't making a statement about "better" systems, and there are certainly other factors that can influence outcomes. I was only speaking of the principle that it is advantageous for an auction to be fair, and open-bid ascending-price auctions can be fair. I am quite certain that if you were to murder anyone who participates in a fair auction, then unfair auctions will be more successful, but that's outside the scope of the principle I'm speaking to.

Reply on news.ycombinator.com