Wednesday, May 23, 2012

Private Market Bubble

Every time a company goes public and the price pops, you get a long line of economists talking about how this is inefficient and that the company has left money on the table.

Well, Facebook just IPO'd and there was no "pop". This is not being trumpeted as a triumph of efficiency.
It may be part of an underwriter’s responsibility to support an IPO at the offering price, but the fact that Morgan Stanley and its partners were forced to wage such an epic battle to do so in the final hour of trading on Friday indicates a pricing failure. Simply put, the artificial demand distorted the market — until Monday, when the shares tumbled 11%. With Morgan’s Stanley backstop gone, the market priced Facebook at a more authentic level, $34, or about $10 billion less by market capitalization than the offering price. “The underwriters completely screwed this up,” Wedbush analyst Michael Pachter told the Wall Street Journal. The offering “should have been half as big as it was, and it would have closed at $45.”
Looks like if shares go up you lose, and if shares go down you lose as well. I think this reveals something quite different.

1. Facebook has been actively traded on private markets for quite some time now, so this wasn't an IPO in the traditional sense as there had been extensive price discovery long before the company officially went public. I don't know how options were prices for recent pre-IPO hires, but the usual stuff that gets done prior to an IPO (valuation, price discovery, finding buyers) had all been taken care of.

2. There is a pre-IPO internet bubble on, with angels, VCs, and companies paying too much for private concerns.
This group needs a robust post-IPO bubble so they can make money on top of the inflated prices they are already paying. That may not happen. If so, the pricking of this bubble will be very interesting, and very different, from the late 90s asset bubble.But the truth is that Facebook’s valuation had grown so large — thanks to several huge venture-capital rounds totaling a record-breaking $2.2 billion — that by the time the offering reached the public, it was already overpriced. In other words, insiders (and others, like Goldman Sachs, which invested $500 million last year at a $50 billion valuation) bid up the company’s stock price, leaving little upside for public investors. “The I.P.O. system only works if it preserves a balance between public and private investors,” writes the New Yorker‘s John Cassidy. “If this balance is upended, and virtually all of the rewards are reserved for insiders, ordinary investors will refuse to play the game. A dearth of I.P.O.s would hurt insiders along with everybody else."
Why? Why should a corporation give unearned upside to any entity, whether it's Goldman Sachs or Joe Sixpack? The article seems to contradict itself in the next paragraph:
But the various problems with Facebook’s IPO reinforce some of the worst stereotypes about Wall Street: That it’s skewed toward insiders and top banks to the detriment of average Americans.
 I'm not sure how average Americans, staying away from an expensive valuation, did anything to their detriment while helping out insides and top banks who over payed and were looking for a greater fool but didn't find one.

Tuesday, May 15, 2012

Big Data, Big Blindspots

I really liked this post on some of the more subtle problems with sampling.
Now here's where we get to the math. The logician, computer scientist, and fellow UCLA faculty Judea Pearl uses a graph theoretic approach to logic that emphasizes using counter-factual understandings to get at the underlying structure of causation. (His magnum opus is Causality. For an introduction relevant to the social sciences see Morgan and Winship.) One of Pearl's most interesting deductions is the idea of conditioning on a collider. If a case being observed is a function of two variables then this will induce an artifactual negative correlation between the variables. This is true even if in the broader population there is no correlation (or even a mild positive correlation) between the variables.
Totally true. And that's assuming you have clean data to begin with. I have another great example of that, which shows how correlation really is not any kind of causation at all.

Tuesday, May 01, 2012

Real wages, nominal debt

SRW says that, in a balance sheet recession,
The only way out of a post-Keynesian depression is to increase real wages relative to the real burden of debt. In the post-Keynesian story, inflation is helpful only if real incomes hold steady, or, at very least, fall more slowly than the real value of prior debt.
I'm not sure if this is true.

Debt is nominally denominated, so its real burden goes up in a deflationary environment. This means that, even if real wages remain flat or fall, real debt burdens will fall so long as nominal wages increase regardless.

So, even if my old $100 salary bought two loaves of bread and my new $200 salary buys one loaf, I am still better off carrying the burden of my $1000 debt. Here, my real income is worse, my nominal income is better, but my debt burden is lighter.

If the Government uses its monopoly power as a producer of NFA(e) and writes everyone a check, then the non-Govt sector has more nominal wealth, but real wealth is unchanged (it may be redistributed). If this increase in nominal wealth boosts AD and factories start humming and the unemployment line shortens, then the real wealth of the economy increases as well as there is more sweat and atoms producing real output.