May 7th, 2010

It's Probably Just Me

Book #2 in the Will Buy Immediately category for me is now unavailable on the kindle and $9.99 in hardcover. Not having these hardcovers stacking up around the house was a very compelling feature of the kindle, and it turns out I'm willing to wait at least a full month in the case of Jim Butcher's _Changes_ and some unknown amount of time in the case of Charlaine Harris' _Dead in the Family_ to read the book, rather than have a physical copy to deal with.

Also, I persist in not buying _The Checklist Manifesto_ (feeling really good about that one) and _The Big Short_ (feeling even better about that one, as I'm reading _Econned_). Just so you publishers know, I thought _13 Bankers_ was worth picking up in hardcover (I would have bought _Econned_ hardcover, too, but couldn't find it on the shelf at Willow the two times I looked -- I might have missed it), because I intend to pass it along to the Mayberry library after I'm done with it. I don't mind a physical book, if I want to proselytize. I _do_ mind being forced to deal with paper if it is Just For Me.

I am shopping less on Amazon, but that has very little to do with the kindle to-do and a lot to do with unrelated issues (sun, exercise).

How About That Housing Market?

The Coldwell Banker website for our area includes historical sales for the last 18 months. I decided to take a look at what had sold during that time in my town. 361 sales with a median price of about $450K. A little less than 20 of those sales appear to be ineligible for the first time or repeat buyer tax credit (because the sale closed outside the date boundaries of eligibility) and a mostly non-overlapping almost 30 sales would have failed to qualify because the purchase price exceeded $800K. With a total number of households in my town of between 7K and 7.5K, just under 5% of the housing stock in my town turned over in the previous 18 months, which is on the low side:

For comparison purposes, see:

The graph towards the bottom of the post, labeled: Annual Existing Home Sales and Year End Inventory as Percent of Owner Occupied Units

I have seen some commentary suggesting that the move-up buyer is non-existent; the median price here of $450K would seem to suggest otherwise _in my town_ -- anyone buying a first home in that range probably isn't eligible for the tax credit because of income limitations. The distribution of sales in the above $800K range was a little thin, obviously, but not ridiculously different from the overall distribution (that is, they weren't all very recent and they weren't all back in November 2008, and they showed seasonality similar to the overall group).

Interestingly, March and April of 2010 didn't look as good YOY. Either I screwed up and tabulated these things wrong, or that interesting weather people speculated would impact sales .... did.

Judging by the sold/listed at/previously listed at, price capitulation occurred sometime before the end of 2009.

So what do I think? Without looking at some other town, it would be impossible to say anything about anywhere but here. But it sure _looks_ like we're seeing real activity not driven by the tax credit. I don't have any numbers, but when I was looking at recent historical sales in this town before we bought here, well, there was less happening in 2007/8 than in 2009. It looks like sellers/agents are figuring out how to price stuff so it sells pretty close to what it is listed at. There's decent quality inventory on the market currently (I looked at that, too), and buying something now comparable to what we bought a little over a year ago would cost about the same as what we paid, perhaps slightly more. Definitely not a _lot_ more.

I would expect that if I looked in more outlying areas, things would be a lot more depressing. R. tells me that prices are already climbing in Back Bay. This fits in well with our thesis that one of the primary location drivers is commute distance and pricing is reflecting that more strongly this time around than it has in the past.

The 19 month old is paying attention

She was crawling around on my lap late this afternoon/early this evening, and I sang "Snugglepuppy" to her (Sandra Boynton). She promptly got down, ran over to the bookcase, dug out the book and brought it back to me for an encore performance.

_Econned_, Yves Smith (kindle)

Subtitled: How Enlightened Self Interest Undermined Democracy and Corrupted Capitalism

Before getting to the text, I want to point out a few things about this book. First, the typography of the single word main title: ECONned. I _sort of_ understand the point: it's a pun on ECON as an academic discipline/course and being taken in a confidence game. The cover of the book (haven't seen a physical copy so I'm going off pictures online) is fairly lame, too: a silhouette of a guy on a tightrope, in a suit, with a briefcase, IIRC. All of this adds up, to me, to a huge vote of no confidence. Second, the publisher is Palgrave Macmillan; regular readers know my recent experience with a book published by Macmillan's academic imprint and it was not pretty. Again, an additional vote of _no_ confidence.

So why did I get this book?

Because Rortybomb recommended it and _13 Bankers_ as the go-to books to read to understand the financial crisis and make sense of regulatory approaches. All righty then.

Additional reason to love, love, love the idea of this book? It's another blogger turned author (and that actually turns out _really_ well, at least at this point in time) -- and she's a woman. She's a woman with depth and breadth of experience in economics and financial markets and that is so freaking rare it's difficult to convey to us ordinary humans.

Was it worth it? Yes. Absolutely and definitely. There are problems with this book, and with Ms. Smith's prose style in particular -- that's good, because it means I failed to find major problems with the content. I found while reading this book and having recently finished it that I often hear "news" stories that are ripped wholesale from its pages. And it has a release date of two months ago; textual indications suggest the bulk of the writing was completed by the fall of 2009.

The book's structure is excellent. Smith describes a little of the history of economics as a discipline and how its position within the social sciences and larger society morphed as it "mathed up" and started taking an active role in public policy decision making. She describes what she calls "neoclassical economics" more or less interchangeably with the phrase "mainstream economics", and then describes its failure to do a good job either describing reality, predicting the future, or even making much sense internally, relegating some of the math-ier bits to an appendix. She mentions behavioral economics and informational asymmetry as attempts to modify the theoretical structure and asserts that these really cannot be wedged in as they are too substantive a change to it.

Moving from economic theory to "free markets" and policy derived from "free markets" as an ideal, she spends time both on theoretical problems with approximations to "free markets" and what-actually-happens when people who are big believers in "free markets" get a disproportionate say in public policy. Chile comes up, obviously.

The remainder of the book zeroes in on the financial markets, mostly but not exclusively in the US. There's some history here on derivatives and creepy organizations like Bankers Trust (failing to escheat stuff to the state? What were they thinking?). She does the best job I've seen yet explaining why compensation is important -- I had no idea the bonus calculation works the way she says it does, but I can't find any indication that she's wrong. Freaky. Obviously, inappropriately aligned incentives get treated extensively, but she probably spends more time on the interaction between the shadow banking system and the people who probably should have been working a lot harder to regulate them, but actually seemed quite clueless as to what was going on, that it might possibly be a problem, and, in fact, _was_ a really big problem. Needless to say, Greenspan comes in for a beating here, but he's got a lot of company. I'll always cheer on anyone who can describe in detail why Larry Summers is a Bad Guy.

The final chapter, inevitably, includes her (depressed) perspective on the prospect of reform: she doesn't think it is particularly likely but fortunately that does not stop her from detailing what to look for and what might sound good but won't do jack.

Smith is being credited in some places for covering Magnetar and the Constellation CDS, and this is appropriate (the only problem being that she isn't getting _enough_ credit, when clearly the publication of this book was a major impetus to other coverage that has popped up in the last month or two). She does a much better job of explaining not only how the deals worked, but also laying out formulaically just how much money was involved (in terms of how much money was made available to be loaned out in the form of truly sucky subprime mortgages that would pay for a little while and then blow up). It is quite breathtaking to contemplate. I am mildly tempted to lay hands on a copy of Lewis' book, to see how his treatment of people shorting subprime compares. Certainly the secondary coverage of his book suggested sort of a, wow aren't these people bright, rather than the, holy fuck these people took a problem and turned it into a disaster, which would be a more reasonable response.

What's my prose style beef? Here's an example from relatively early on in the book (and this is an example that both displays the issue _and_ works fairly well -- unlike other examples later on):

"Yet the Great Moderation was, as bodybuilders describe steroid-abusers, a Cadillac body with a Chevy underneath. Its rate of expansion was lower than previous postwar growth phases. Inflation-adjusted worker wages had been stagnant. Dampened swings in the real economy were accompanied by more frequent and severe financial crises. The supposed better timing of central bank intervention merely led financial market participants to believe they could count on the authorities to watch their backs, encouraging more risk-taking. But perhaps the biggest danger was that blind faith in the virtues of markets converted regulators from watchdogs to enablers."

It is quite clear what she is saying. From the 1980s through the 2000s, moderate growth in the economy was accompanied by stable real wages. Financial crises had returned, but the full business cycle (in terms of a very tight job market at the peak, and high unemployment in the trough) had not. Financial market participants expected the government to help them through the crises and that made the whole situation more and more dangerous.

Both the way she said it, and the way I say it, make use of metaphors, some of which are shared: cycles of expansion and contraction in both, for example, and the word crisis is an embedded metaphor for a time when important decisions must be made. But she's _also_ got people using hormones inappropriately, a couple cars, standing water (that's the stagnant part), a pendulum, a possible fight requiring assistance in defense, some religious ideas and a 12 step program. That's a lot of imagery which suffers from being a little too vivid (crisis does not suggest a disease process for most people and stagnant probably does not call to mind still, oozing water to most people; the cars and dogs and so forth bring up images quite readily) and a little shopworn. Worse, the images are not closely related to each other and do not function cooperatively.

Smith gets big points for having an authorial voice that is understandable, recognizable and engaging. But wow. A little work on the mixing of the metaphor and the cliche action. I am _so_ happy that's the worst thing I can say about _Econned_. Run right out and buy it now, or get it from the library, or whatever. I'd loan you my copy...but I can't.

About That Certainly Untrue Story Yesterday

Planet Money has this to say on the subject today:

Short form: when things get hinky at the NYSE, the NYSE takes a wee break (about a minute) to let humans look at what is going on. But the other exchanges (NASDAQ and on down) do not.

The entities placing orders usually do so in a hierarchy, depending on where the item in question is listed natively (that's not the word they use), which exchanges are currently open for business, cost, turnaround time, blah, blah, bleeping, blah. Historically, the NYSE has tended to have the really high volume stuff, altho R. contends this is no longer true. So investigators are exploring the possibility that the other exchanges weren't ready for the wall of orders that hit them when NYSE took a wee break on some stocks. The proposed solution? Maybe if we take a wee break at NYSE we should take the same break everywhere else, too. I foresee technical difficulties in getting that to happen quickly, but never mind that now. I want to speculate wildly.

So let's say that wall of volume from high frequency traders (who some coverage is saying took a break during the flash crash -- but these are the same people who think a 'b' vs. 'm' illion is a plausible explanation) moves from NYSE to NASDAQ. At least some of that software must know about the wee break -- and know that it means trouble, which triggers some strategy which probably means even faster trading. Bigger wall of volume. And let's make one more big assumption: that wall of volume is, in effect, a denial of service attack on the exchange.

Here's terrible coverage of high frequency traders: some guy, possibly co-lo'd with an exchange.

Here's a decent description of a denial of service attack:

"One common method of attack involves saturating the target (victim) machine with external communications requests, such that it cannot respond to legitimate traffic, or responds so slowly as to be rendered effectively unavailable."

I've screwed up with Teh Computers before and brought big iron to its knees (oops with the shell programming and recursion and a simulation). If I were a high frequency trader whose trading strategy went all freaky, jumped from NYSE to NASDAQ and make a zillion requests to trade P&G, none of which were responded to before I decided I'd waited way longer than I usually do so it had probably gotten lost or refused or no one wanted to be the other side of my trade so I reduced my price and tried again. If I'd done that, and brought NASDAQ to its knees, yeah I'd _for sure_ say, no, man, I turned the program off when it got all weird. I had _nothing_ to do with _that_. And I'd hope that none of it got logged or, at least, none of it got found until some other sap got blamed for it. And I'd _really_ be hoping at least 10 other guys had done the same thing so we could all point at each other and say hey, it wasn't all my fault.

ETA: Some people are way more paranoid than me:

I read this _after_ posting the above. And I think these people are wacky. But hey. Fun wacky, right?

"If you could fill up the bid queue with orders all which are way above the last trade then you could disable the bid queue. None of these trades will be filled but could you effectively block others from entering bids. Then, in an instant (electronically), you could kill all your bogus bid orders and enter an oder for one cent and also enter a market order to buy. That order would be filled at one cent. I'm not saying this is exactly how it happened, but this is more plausable than all of the other reports out there."

ETA still more:

NASDAQ upped its bandwidth requirements in February. Up from the 2007 peak.

ETA even more:

Then in March, they made it so instead of deleting and resubmitting an order, you could _replace_ an order. They did this to try to reduce traffic. Which suggests there was an ungodly amount of this activity going on as "ordinary business".