25 November 2014

Play to theFront Row

The pundit class has been wondering, in print, how it can be that we have increasing income/wealth inequality, but, if one believes the data, a recovery from The Great Recession and general growth.

The answer, in a nutshell, is Apple. Apple has made a living by ignoring the bottom 80% (or thereabouts) and selling a limited set of models to the top 20% (or thereabouts). A small list of models means what Henry Ford (is said to have) said, "you can have it in any color you want, so long as it is black". Smaller global BoM, smaller unit cost, higher gross margin. You just have to convince your market that they only need that one model. Mo money. As other microeconomists and quants see the light; likely have for some time, what can we expect?

As these endeavors have pointedly asserted, when it comes to data, analysis and decisions based on data is a worthy exercise only when the venue under analysis obeys God's (or Nature's, depending on the -ism you subscribe to) laws. Read the news every day, and you'll find a case where the clever (but not clever enough to avoid getting nabbed, of course) have bent or changed the rules to their advantage. Data on human endeavors is not reliable when used to predict the future. The future's rules won't be today's or history's. When the players are making up the rules as they go along, relying on recent data is foolish. Data driven decision making only works when the decision makers are forced to obey a rule set out of their control. Lower animals and glaciers come to mind.

When the recommendations of data and incentive conflict, follow the incentive. The clever will, like mouses, squeeze through the tiniest crack in the wall. Build a new wall, and they'll find the new cracks. Time series won't tell you where those cracks are.

So, what does Apple have to teach us? First, that by concentrating on the High End, its market in units is limited; as concentration continues apace, the number of individuals near the top X% diminishes as the X%-ers accrue more moolah. Apple counters that fact by limiting models of any device, and thus minimizing corporate BoM. The counter is to increase the number of distinct devices. Whether the iPad (or the don't-call-it-an-iWatch) can survive is an unanswered question. Reporting today reveals geographic/economic differences in mix between iPhone 6 and 6+. Why might this be? The logical conclusion is that lower wealth populations will have more buyers willing to compromise both ends of the size factor by having just one device. The phablet only holder, while the higher wealth populations will have a more reasonable device for each end of the size factor. Should this be a surprise? Not.

What of the macro effect then? It depends. That Burberry marketing gal had an easier time of it peddling coats than she will compute devices. In general, the top X% will have more moolah to spend on bling, but they won't necessarily spend more on bling from any one XYZ, Inc. widgets. Said widgets, unlike fancy coats, may have no value beyond having one. What we should expect then (looking at the incentive, not the old data), is to find more XYZ, Incs. making different sorts of bling for the X%-ers. Samsung was reported in the last couple of days to be reducing its model count drastically. Lowering the corporate BoM to make more moolah. So, what's the next big thing in X%-er bling? I wish I knew.

But, not all is wonderful in X% land. Consider participant sports. Turns out Tiger Woods is thought by some to be key to the golf widget makers' survival. Golf can be an expensive pastime, right up there with skiing. Poor person envy? May be yes, may be no; but it doesn't matter to the analysis. I've never had any interest in either, which I can duly afford if I wanted to. And therein lies the problem with growing the moolah flow to providers of the various bits and pieces associated with them: how to engage the small population that is the X%-ers, because only they can afford the vig. What Americans play tennis anymore?

Professional team sport provides some insight into how it can be done. Players in these stadium sports get ever more tens of millions of dollars per annum to behave like teenagers. How can this happen? Again, look to the incentive. Partly, TV dumps billions of dollars into the pot. But consider the stadiums. Research (I've not got cites to hand, but they weren't hard to find when last I went looking) has shown that most are partly or fully paid for by taxpayers, both directly and indirectly. Most are in large metropolitan areas, where the number of seats is a small fraction of the market population. Since, let's say, the stadium on average needs 50,000 behinds to fill it up, then the team need only convince the increasingly wealthy 50,000 to attend out of a few million in total population. They have no need to attract the lunch bucket crowd. IOW, the Apple crowd has another way to spend disposable income. For the privilege of being special, they can afford to pay yet more. The Washington football team attacked in 2009 when The Great Recession hit. All was not well in Mercedes land (for the record, the real money in the DC area isn't made by civil servants). What's odd about such suits (the piece reports that Washington is not alone in suing) is that, according to legend, season tickets for Washington football (and many others) has a long waiting list. Hard to see where the loss is? Kind of like foreclosing on a house that's still brand spanking new. It should be no surprise that the various leagues expansions into ever smaller markets (hockey in Florida, Arizona, and North Carolina???) have led to failures, and near so. They have to have some minimum population to have the necessary 50,000 X%-ers. Some leagues (3 of 4, not NBA, near as I can find) counter with "revenue sharing" wherein the rich big city owners give to the small hick town owners, aka rich white guys' Socialism.

In general, the incentive for an economy's production under rising concentration of income/wealth is to shift to more varieties of lower volume bling. No one needs two smartphones, drug dealers and con artists possibly excepted, so some other forms of bling must be created to sop up the excess moolah. This happened before in the USofA. The most well known was called The Gilded Age, a coinage and book title from Mark Twain, which, by way of setting context, he wrote before both "Tom Sawyer" and "Huckleberry Finn".

So, the question for macros, micros, and quants: can such an economy thrive? And, of course it can. Except for the period post-WWII to the 1973 oil embargo, that's been the history of the USofA. It was only that rather short period of unspoken Socialism that is the unusual condition. The downside, depending on which side of the X% you sit, is the necessity of an increasingly motivated police state to keep the non-X% from revolting. If you read 19th century American history, it was awash in minor and major insurrections. The fact that there was ever more land to move to and pillage helped keep the lid on, until it didn't. The Civil War is the one remembered, but there were more clearly economic ones across the century. Here is another list, but both don't include Bloody Kansas, more properly the First Civil War. So, of course we can have an economy based on the top X%, but as X gets smaller, the police state must get more motivated. And to think, all those ex-Middle East armored personnel carriers and such have found their way to civilian police under a black president. The X% can't say, "if you don't like it here, move to the frontier and see if it's any better." There is no more frontier. Have a nice day.

Why Not Be Normal? [update]

There is a recent thread on simple-talk dealing with stored procedure as the repository (execution environment) for business logic. One commenter took the opportunity to assert that such a horrible mistake gave rise to NoSql; may be not entirely. And, of course, the implied inferiority of SQL databases, normal forms, and the RM paradigm by implication. Bollocks, again. NoSql is yet another failed, one would dearly hope, attempt by client-side coders to return to the thrilling days of yesteryear, when a lone ranger coder siloed all the data management in his code. And so forth.

The argument against 5NF boils down to: joins are too slow. One might add that most application developers are too dense to understand normalization, and that they prefer the employment security inherent in client-side data management. In other words, it ain't tech it's politics.

The point of these endeavors has always been that tech as it has emerged over the last half-decade or so gives us the tools to implement the RM as Dr. Codd intended. And, of course, that we'd be idiots not to seize the opportunity.

As node sizes have dipped toward 10nm, we get multi-billion transistor cpu/gpu implementations, motherboards with multiple sockets and four or more memory slots. For under $10K, one can build an awful lot of server. That same $10K will support about one month of a client-coding developer. If that doesn't conjure up an image of Jabba the Hutt on one end of the seesaw and Bruce Lee on the other, here's another reason.

Intel has released details of its next 3D NAND, terabytes are now on offer. Fast terabytes. There's just no excuse for all the mess and heartache caused by flat earth folks.


[update]
The comments remind me of the core irony of the conflict from coders toward NF data: the OO crowd always talks about two principles that they hew to, no matter what. First is DRY. The second, related, is that objects and methods must deal with as limited a scope as can be conceived. One java coder of my recent acquaintance used to say that his methods were never "more than a few lines long", as if this were the ultimate virtue. "Do one thing, and do it right" is what we are told. No bloat. No repetition. No duplication. It isn't much of stretch to see that both attitudes applied to data must yield NF schemas and the DRI which results. Yet, they insist on silos of flat-file (or xml, which amounts to the same thing) data.

My Pappy used to tell the old wheeze about the miner, the mule, and the eastern fop. A miner comes into Busted Flats to stock up on supplies. He ties up his horse and the mule to the rail. Or tries to. The mule stops dead in the muddy street. Standing on the walk by the general store is an eastern fop, just off the train. The miner pulls at the mule for a bit, then tires of the effort. He reaches into the pack, pulls out a 2 by 4 and whacks the mule upside the head.
Fop: "You can't do that!!!"
Miner: "Why not? I gotta get his attention somehow."
The mule walks up to the rail, gentle as you please.

19 November 2014

A Dirty Little Secret [update]

The quants operate, oft times unknown to themselves, based on a dirty little secret. "What might that be", called from the Peanut Gallery?? The notion that the risk-free, inflation-free interest rate is equal to homo economicus' "rate of time preference", i.e. homo econ will lend you one unit of fiduciary capital (dollar, bolivar, ruble, whatever) for one unit of time (day, month, year, decade, whatever) on condition that homo econ gets one unit plus some vig. And that such rate is completely independent of everything in the real world but homo econ's demand for the vig.

All those wonderously complex financial engineering models, the quant ones not the reg fiddlers of course, depend on this notion of some exogenous interest rate. Taint so, alas.

The secret, of course, is that this assumption of control, or even choice, is false. The risk-free, inflation-free rate of interest is determined by technological improvement. This risk-free, inflation-free rate is the last chance opportunity cost faced by capitalists. But, here's the thing: if they're unable to conjure productive uses of physical capital, there's no other demand that homo econ can bring to bear on borrowers. Homo econ may want 10% for the use of his moolah, but no one can earn that, so his demand is ignored. And so on down the number line, to 0% where there is no productive use and time preference is irrelevant. No one wants your damn moolah!! In times, such as now, when capitalist overlords can find no way to turn fiduciary capital into physical capital, i.e. (again) they cannot conceive of new plant and equipment improvements which are more productive than existing, it matters not how much holders of units of value wish to be paid. There's no growth in the system to pay them. Full stop. Period.

Unlike usual trade with a supply and demand which are independent and infinite (that last bit is vital), capital return isn't. While a capitalist can make an infinite number of widgets, simply given a high enough price (or consumers take an infinite number given a low enough price), an infinite return on physical investment has to be the result of infinite ingenuity in technology. You have to be organically smarter, you have to know that which is better and was not known before; it's not sufficient to, somehow, "work harder" or have access to more abundant, cheaper inputs. Ah? You need a bigger brain which can divine more about the real world than is universally known, already. Ah? What happens when all that's left to discover is trivial? Cold fusion? Mr. Fusion? Ford Fusion? Movies on a portable phone?

Thus, in this state of the world, we get stories with titles thus, U.S. stocks are soaring, global economy in a ditch: Now what?. Even the business press is befuddled. They really ought to know better. It's just simple arithmetic.
And what's great for American corporations isn't always great for most Americans. Merger activity has picked up to new highs this year, and U.S. profit margins continue to hit record levels, defying forecasts that they are going to come back down to earth. [Yahoo Finance Editor-in-Chief, Aaron] Task says, "The problem for the rest of us is that these corporations... are not spending a lot of money on new development... and they aren't doing a lot of hiring or when they do hiring they are not paying people big wages."

There's a reason that Big Gummint debt, which is auctioned (not set by fiat) just to remind, is going for so little. The alternative is to sit on your moolah when you can't figure out a way to allocate to growing (organically, not just by buying your competitor) your business. Which Big Corps are doing, to a faretheewell. The opportunity cost is, basically, 0.0 (props to Bourdain). They're sitting on trillions of dollars, with nothing to spark their interest. So to speak. Well, they've invented securitized, sub-prime used car loans. Deja vu all over again.

The Wiki article is pretty much what is written in Econ 101 (and subsequent) texts. Quite obviously, it ignores the source of the vig. That source is technology, nothing more. The Dark Ages were a time of stasis because knowledge was so, not the other way 'round. Now that we know pretty much everything about how the real world works, we're near a permanent Dark Age; this isn't 1800 with the periodic table to be filled in. We're in a period, likely infinite, of ignorance, no longer a world of discovery. Most everything that marks us as modern was discovered in the first half of the 20th century or earlier (the understanding of semiconductors began in the 19th century). The technology discovery curve has flattened out; the big deal is whether a 4" smartphone is worse than a 5" smartphone. The population curve hasn't. There is no New World to pillage. And so on. Have a nice day. And, abandon the hope that compound interest will make you, or your spawn, rich just because it worked for your grandpappy. Those days are past.

[update]
To demonstrate how meager "innovation" is these days, we now know that the iWatch is just visual Bluetooth!! Some fool recently compared present day innovation to the steam engine. Gad.

18 November 2014

Make It a Double

When I was growing up there were three categories of lushes: the crude drank bourbon, the smooth drank scotch, and the sissies drank screwdrivers (so Mom couldn't tell; she still could, of course). Seems to be still the case.

There's been a push, and some pushback, for artisinal bourbons. They don't really exist, of course. Here's the short answer as to what makes a bourbon, bourbon. Note, in particular, that aging is not a requirement. In fact, the latest fancy of lushes is white lightning, often called corn squeezins, because it's just the pure alcohol distilled from corn mash. Put that stuff in burned barrels until it gets some color (kind of like white folks wanting a tan), and you can call it bourbon. Doesn't even have to done in Kentucky, although those that are make a big deal of it.

The point being: bourbon is crude liquor made fast. You can leave it in the barrel for a year or more, but it's still just colored corn squeezins.

I bring this up, because Neil Irwin badly tries to analogize bourbon to petro-products in his piece today. For someone who presents as a quant/economist, bad piece. About the only thing they have in common is that they're both liquids. Too bad that Irwin chose such an approach, since the petromarket is totally screwed up. But bourbon production has nothing to aid in fixing it.

Bourbon:
- little physical investment; little barrier to entry
- available nearly instantly, unless you want other wise; the longer you hold the higher you can charge
- typically, but not required to be, made in Kentucky

Petro:
- lots of physical investment; massive barrier to entry
- not available, if at all, for years after initial investment; holding may, or may not, yield a higher price
- can be found nearly anywhere, that's left unexplored

What is true of petro-products, and needs to be dealt with, is that it is, for first world countries, as vital to existence as clean air and water. As such, it should be husbanded for the benefit of all, and all future generations. The cyclic nature of production, in response partly to market price (but also to more efficient, in the short term, technology), is quite real. Do we, of the current generation, have the right to burn off nearly all of it in our Hummers? The Rand-ians say so. May be we should stuff a cabbage in their mouths.

17 November 2014

Sweet Charity, From an Unlikely Source

One of the conundra of our times: Amazon can't manage to make a consistent profit, yet its share price seems to be always rising. Well, until this year where it's been bouncing around the $340 mark. And why is this? Once again: it costs a bunch more to ship one widget by aeroplane then it does to ship that one widget by the tonne in a train. Consumers won't pay the extra if they can avoid it, so the billing price of an Amazon widget has to come down such that the price + transport is in the ballpark of brick&mortar stores. Econ 101.

Today, the good folks at UPS fired yet another shot across the bow of Amazon, et al. Be wary. Be vewwwy, vewwwy wawwwy.
Managing those costs as e-commerce continues to grow is a priority, [UPS CEO David] Abney said. E-commerce accounts for about 45 percent of UPS' business, and should reach 50 percent within five years.

"I think anybody that's delivering directly to consumers is certainly going to be looking at those costs," he said.

It's now some hours later, and this piece came through the innterTubes. Amazon realizes, yet again, that it has to eat the transport costs. For now, at least. Another way to spend years not making any money.
E-commerce professionals polled by MarketWatch were more mixed in their views. Consumer World founder Edgar Dworsky said he was "not aware of that trend," and said ordering a bottle of Tide online could prove an "expensive proposition" if the consumer were responsible for shipping costs.

Yeah, we know: lose money on each widget shifted, but make it up on volume.

14 November 2014

ATM at Home

These endeavors began at the bottom of the Great Recession, and one of the first themes (and continues unabated) was that the use of the Home as ATM was driven by the increasing inequality of wealth/income. The right wingnuts are always complaining that growth is poor because poor folks don't spend money, but that's their point, of course. As the 99% continued to lose out on the growth rewards, they were enticed into burning the (un)earned equity in the Old Homestead. Hard numbers were hard to come by, unfortunately, so the schemers could continue to claim that the Great Recession was the result of Democrats' loosening of the mortgage process to allow their poor folk constituents to get into the game.

One Bethany McLean has talked to the perps, and comes up with the hard data.
According to the financial statements of New Century, the huge lender whose bankruptcy in early 2007 helped kick off the financial crisis, cash-out refinancings were 64.2 percent and 59.5 percent of its business in 2003 and 2004; home purchase loans made up only 25 percent to 35 percent for the two years. A New Century executive told Congress that its customers needed to "tap into their home equity to meet other financial needs, such as paying off higher-interest consumer debt, purchasing a car, paying for educational or medical expenses and a host of other personal reasons." I'll always remember seeing a bank ad blowing in the windy, bleak Chicago winter of 2009. "Let your home take you on vacation," it read.

The right wing just can't admit that it's trying to have its cake and eat it too: keep the 99% poor, but keep the economic merry-go-round spinning at breakneck speed. It can't work that way. Li just made it easier to justify, but not intellectually or arithmetically viable. That link is to a Forbes article; amazing.

The bottom line, so to speak, from McLean's research:
According to Jason Thomas, now the director of research at the Carlyle Group, only about a third of subprime mortgages that were turned into mortgage-backed securities between 2000 and 2007 were used to buy homes.

Makes one wonder what's driving the consumer sector these days? The obvious answer is that the Jobs-ian paradigm is at work: consumer oriented companies are all chasing the 20% with money. The 19th century USofA economy worked that way. Back to the future.

13 November 2014

Talkin' 'Bout My Generation

It's been rather some time since I've explored the world of GUI generation from schema. So far as I'm concerned, it was a solved problem when middlegen (RIP) came out about a decade ago. Nor am I a PHP fan. Nevertheless, here's another shot on goal. Why do we still need so many client-side coders? Hmm???

Booooooiiiiiiinnggg

It appears to be an article of faith, among the New Age capitalists such as Jobs/Cook of Apple fame, that one need only target the few, at a high price, to be success in business (without, of course, really trying). In other words, as you learned in Econ. 101, demand is price inelastic. In fact, one might argue, that the Apple paradigm is to grossly overcharge the BoM, since if the widget's painfully expensive it must be good. Marketing has to convince the few who buy that they've gotten a good deal.

But price inelasticity isn't a given, and most often applies to things like food, water, and clean air. Necessities of further existence, in other words.

Now comes word that M$ has seen the light. Gee. Lower the price, take a smaller cut on lots more shifted, and abracadabra, more moolah. Gad.
Since the price cut was introduced on November 2, Xbox One sales in the U.S. have tripled.

Nobody, pimply faced adolescent males who don't bathe possibly excepted, needs any sort of game console. Go read a book, for crying out loud.

D'oh!!! Get me a Duff's.

12 November 2014

Remember When M$ Was "The Borg"?

It looks like The Federation is close to winning. If only SQL Server (I do like and respect it; Windoze, not so much) can be made to run on linux?

What Are The Odds?

The initial title was going to be something like, "Marty McFly is Grounded", since the subject is Matt Martoma, and yes, it's a bit of a stretch for an allusive pun.
Mathew Martoma, a former portfolio manager at billionaire Steven A. Cohen's SAC Capital Advisors LP hedge fund, on Wednesday lost a bid to stay out of prison while he appeals his insider trading conviction.

As the saying goes, "be careful what you wish for, you just might get it." The wish, in this case, is that cracking down on white-collar financial crime will deter the self-absorbed, greedy Wall Streeters from sinning. It was TNT running "Castle" marathons that got me hooked on the first run on ABC. In the archive shows, Lockwood (a bad guy hired by the Real Bad Guy) says to Capt. Montgomery, "If you and Raglan and McCallister hadn't sinned so spectacularly, then God wouldn't have sent a punishment like me." Soon after Lockwood and Montgomery have a shoot out with both getting killed.

In the case of Wall Street and Martoma, what seems the most plausible outcome? Well, these guys are quants, more or less, so they think in terms of odds and expected value. Well, if we raise the odds of getting caught, losing the moolah, and having to wear Orange for a decade or so, then we've lowered the expected value of schemes currently in play. We, therefore, should expect that the Wall Streeters will dream up schemes which increase in value (to them, not the 99%). I'll all for the guillotine as punishment for such behavior, but that's not enough. The folks in Congress, well the past one not the next one fur shur, also need to anticipate where the wedge will go next. Crooks be clever dudes.

Chasing Unicorns

Josh Berkus, Postgres factotum, has a new post up asking whether there should be a, and what it might look like if there were a, web/PG benchmark app. I kind of noted that web performance in the normal HTTP world doesn't have much to do with RDBMS specific performance, anyway. He wasn't impressed. Not that I mind. The notion did get me to ruminating, though.

This overview is from 2010, but confirms my experience: smart code fiddling dumb data remains the paradigm. Gad.
Most of the social networks also use Sun's MySQL database management system to organize their users' messages and status updates.

They, obviously, would rather keep writing code to keep track of things because, well, they get paid to write code. So, write code they do. There is no try, there is only code or no code.

PG has long been the "Open Source database that isn't MySql and sort of relational", so why would the PG inner sanctum want to toss that all away by jiggering PG into just another sql database which is good at storing flatfiles? Josh mentions that such a benchmark would target social networking. Again, this is the ken of those who are certain that flatfile storage is the bee's knees, at which MySql 3.0 was very good; current releases can be run that way, too. I strongly suspect they are.

Here's a chronicle (or, may be, a superbly written indictment) of a social network app that started life on MongoDB. I find it interesting. The horse punches out Mongo, this time. May be the (social networking/start up/real) world is actually relational, not flat or hierarchical, after all (Date is right, in other words)? Why chase them? Not that I've been harping on it, much.
You can also see why this is dangerous. Updating a user's data means walking through all the activity streams that they appear in to change the data in all those different places. This is very error-prone, and often leads to inconsistent data and mysterious errors, particularly when dealing with deletions.

Kiddies spend too much time learning multiple languages, and not enough time reading Date and Weikum & Vossen.
I learned something from that experience: MongoDB's ideal use case is even narrower than our television data. The only thing it's good at is storing arbitrary pieces of JSON.

Just for context, the author made a reference to Romulans to some of the kiddies she was dealing with, and got nuthin'. If they don't even know about recent history... Kiddies keep making the mistakes our grand pappies made; inventing square wheels. Gad.
Schema flexibility sounds like a great idea, but the only time it's actually useful is when the structure of your data has no value.

She's nicer about it than I am: the flexibility of xml, and its -ish cousins, is just pretty pink bows on chaos, aka illusory, which the coders have to work around since a hierarchy is rigid (while adding a layer, from a typist's point of view, is simple...) and to change it requires walking all the code which deals with the data at the point of change on down (at least). They don't mind, as it gives them more code to type. They like to type code. Real flexibility comes from Organic Normal Form™ schemas; since the data is stored in an orthogonal manner, changes to the schema are transparent to any existing data and code that are disinterested. With the cheapness of white box *nix machines and prosumer SSDs (Intel has just upped the ante, by a tad), joins are basically free. High 5™s all around.

Why chase unicorns?

11 November 2014

Really?

Really. And people continue to argue about Organic Normal Form™. Get a clue.

Worst Study in the World

Olbermann just did the "Worst Persons in the Sports World" bit, and eviscerated the "study" of short NBA refs. The show repeats a couple of times more, on the East Coast at least. A Must Watch.

09 November 2014

Grinding the Hard Candy

Artima has this post up today. The topic is how to approach the problem of making code, basically. The poster discusses the "hardest first" versus "easiest first" paradigms. He, imagine that, falls into the hard first class, and lonely.

To me, at least, this mirrors (or is the origin of) the database as RM engine versus brain dead file dump with easier backup. After all, figuring out the Organic Normal Form™ schema is mostly a thinking exercise, not a typing exercise. And folks still wonder why applications are still garbage. They're mostly made by monkeys whaling away at keyboards looking busy and spewing thousands of lines of code per hour. Swell.
I do hope we've not reached the point where it's automatically seen as wrong to stop and think instead of diving straight into writing a test for some "simple" case.

07 November 2014

Just A Taste

One of the continuing themes of these endeavors is that quant isn't always the best guide to the future; moreover, the data which is ground, finely, through the mill stones isn't always understood for what it is. This is particularly vexing for the macros in the crowd. The micro folks tend to have population data to hand, so aren't really doing stats as a math stat would define the effort.

Chief among these are the government (any one, by the way) economic activity data. Floyd Norris saves me the time to recite the problem(s) once again. The main point is that only the weekly UI numbers are population data. All else is sample data, of varying "quality" and source. Census does most of the actual surveying, for BLS, Commerce, and such. There are periodic censuses for economics, mostly in non-population census years for obvious reasons.

Having been a Fed quant, although not for any agency which produced public macro-economic numbers, I will attest that the grunt quants took no shit from the political appointees. The recent "scandals" that seem to come from the bureaucrats were actually perpetrated by what were known as Supergrades for decades. They're now called The Senior Executive Service, but the issue remains: they're a bit fish and a bit fowl; some of their moolah is scheduled and some arbitrary from the Suits. They sit (or sat, depending on how one looks at it) in the top three rungs of the ladder of civil service. Some argue that this change, which happened under Carter and while I was there, has weakened the Chinese Wall betwixt the grunts who try to be honest and the appointees, who aren't.

Anyway, Norris does a decent job of explaining the situation, without getting into talking about stratified random samples and such. [Aside: with the current hard on for Big Data, one might wonder whether the expertise, across the profession, for sampling might just go poooof???]

My favorite quote:
[Paul Singer, a hedge fund billionaire and top Republican donor], in his denunciation of official figures, questions adjustments that are made for the quality of goods and argues that government indexes understate the costs for the wealthy because prices of things like luxury real estate and art have increased far more than the prices of goods purchased by the middle class.

Oh, the poor little rich kid!! And such folks just don't understand why the French finally had enough and separated some heads from some bodies?? These are the same folks, of course, who bet that house prices would keep going vertical forever, and when they didn't came to the taxpayer begging. And, having gotten their billions, bitch and moan about any small attempt to keep them from desolving the economy, again. I think that's a working definition of ingrate.

As to quant not being the best avenue to understanding the direction of the road ahead, the NYT has another article today which tells you more than any time series. Both Japan and China have forced working folks into mortgages as the sole investment vehicle available. Various reasons why, but the consequence is, mostly, that mortgages aren't productive use of capital. Any return comes from the holder's ability to shift moolah from consumption to paying the mortgage. Psychic income is baloney. The notion of safety in home mortgages was true before the 1% sucked up so much of national income; mortgages were made and held locally, and so on. So olde fashioned. But we didn't get chaos. Change the rules, and you change the incentives, and you change the money flows. Kind of like a switch in a train yard. That Giant Pool of Money is still out there, and continues to grow. American corporations continue to: sit on it, buyback shares, buy each other, and otherwise spend it on non-productive (in the macro sense) allocations. So soon you forget, heh?

06 November 2014

Home Is Where the Heart Is

A couple of interesting occurrences here in West Redneck, CT this past week. First, I got a LinkedIn message, which I ignored. Then, some auto nagger fired up. OK, so I read the message. Turns out that a recruiter was looking for a "SQL Advocate" and did I know of any?? I guess I didn't meet the req. Having a bit of time, and feeling a bit snarky, I typed up, with reduced snark, a reply.

Said reply amounted to:
- I'm the only SQL advocate (really, RM advocate, let's be honest) in this part of the world
- unless the C-suite types enforce it, the coders will obstruct, obstruct, obstruct (kind of like Republicans)
- my experience is that SQL advocates, in orgs run by either pure coders or business analysts, end up with RDBMS as passive filestore
- does your client really, really understand the implications of having a person who's going to be at odds with most of the staff
- what, exactly, does the client want to accomplish; what change is desired
The database wasn't named, but I'll guess SQL Server.

Much to my surprise, I got a message back from her. And it confirmed my concerns; she didn't know what the client wants to do, and doesn't think the client knows. I was shocked, shocked I say!! May be not so much. The recruiter showed a Boston area address, which means there could be some form of intelligent life. Her client, turns out, is not far from West Redneck, but not Hartford or New Haven. Not that either necessarily means a SQL advocate would be warmly welcomed. I took a position at CSC in Hartford a while back, mostly on the assertion (from the hiring person, no less) that "we're going to be doing heavyweight database development". Turns out, all they did was map COPYBOOKS into tables (that link is about a decade after I weathered the nonsense at CSC, the Dark Ages continue). Moreover, mainframe DB2 allows existing COBOL code to be executed as a stored proc. A yukky time for all, but past.

So, then comes a simple-talk piece on where to check the data. Naturally, I didn't keep my mouth shut. But I remain amazed that it's still necessary to slay such octogenarian dragons. Them dragons do endure.

And, it turns out that there are a few coders who recognize that having centralized control is generally a good thing. Allen Holub has been my favorite java guy since the late 90s. He seems to have fallen off the map, largely. I think I know why. Here's a bit from his Bank of Allen series:
The only recourse is to change the ROMs in every ATM in the world (since there's no telling which one Bill will use), to use 64-bit doubles instead of 32-bit floats to hold account balances, and to 32-bit longs to hold five-digit PINs. That's an enormous maintenance problem, of course.

Changing the clients is a waste of time and effort. Bad design, but good for billable hours. Do it his way, and you don't need quite so many coders, don't ya know?

01 November 2014

Time Loves a Hero

Well they say time loves a hero
But only time will tell
If he's real, he's a legend from heaven
If he ain't he was sent here from hell
-- Barrere, Payne, Gradney (from "Time Loves A Hero", Little Feat)

Yesterday was a bit different, schedule wise, in that this is deemed to be a Windy Weekend here in West Redneck, CT. Thus my morning was devoted to moving mucho leaves from their temporary pile on the lawn to permanent recumbence elsewhere. To gird my loins, I made a quick look at r-bloggers, and found that Rob Hyndman had, unusually, posted a job advert. It is from Amazon, and provided me with great mirth as I set about to haul leaves.

His post is mostly a quote from the (unnamed) email he received from Amazon:
... we have found that applied economists compare quite favorably with the machine learning specialists and statisticians that are sometimes recruited for such roles.

The job posting itself is on the AEA site. And provides added humor.
Amazon economists will apply the frontier of economic thinking to market design, pricing, forecasting, program evaluation, online advertising and other areas. You will build econometric models, using our world class data systems, and apply economic theory to solve business problems in a fast moving environment.

Sounds a whole like the description of what those benighted knuckleheads from Wall Street were doing, doesn't it??

All well and good, you say; but where's the laugh out loud part? Well, as I set out to haul leaves I noted that Amazon had provided a job (or, at least, career track) email. Oh goody!! I penned a snarky, but lighthearted, rebuke. But why, you may wonder?

Consider that, among quants, the economists are the most incompetent. While a grad student, most of my profs were the initial wave of New Quant Economists who represented what has been wrong with econ quant for some decades since: they were all flunked out math and science Ph.D. drudges, who found that they could find repentance in the econ department. Thus, we get econ quants who still think in terms of God's Inviolate Laws running the show, rather than the reality that it's Political Economics run by social Darwinists making up the rules as they go along. The rules change without regard to data. The data changes in response to rule changes, which are blithely ignored by said quants. They all believed that a well known ratio (median house price / median income), could go straight up forever. Yeah, sure. They apply the wrong context to a field of which they've not even had 101 training. They consider it just another series of algebra exercises. Political economics ain't algebra.

So I hauled leaves, which meant that I'd missed my mid-morning stroll to get my dead trees NYT. And, if it's Friday, it must be Krugman. I sorely would miss my dose of Krugman, but I really didn't want to re-gather all them leaves. Remember, this is after I'd sent my snarky email (no, they haven't replied). Having done with hauling, I didn't expect there'd be any left, but I got the last copy. Oh joy. He, once again, rubs the mainstream and right wing cabal's collective noses in the Japanese experience.
For now, here's what you should know: Japan used to be a cautionary tale, but the rest of us have messed up so badly that it almost looks like a role model instead.

What, you must be wondering, does Amazon's love affair with econ quants got to do with Krugman and Japan? Glad you asked. It dawned on me that Amazon and econ quants were a perfect match, the totally dysfunctional marriage. Amazon's only hope to be consistently profitable is to gain monopoly status in retailing, since:
The net result of nearly two decades in business is that Amazon's trailing 12-month price-to-earnings ratio stands at an alarmingly high 550. Compare that to consistent profit earners with significant online retail operations such as Google (p/e 29), Wal-Mart Stores (2) or eBay (25), and it's easy to be confused by investors' hunger for Amazon.

Amazon, Bezos in particular, hasn't yet figured out that moving widgets by the tonne on trains is way cheaper, per widget, than by each in aeroplanes. He seems to be getting a clue, what with all these so called distribution centers he's building. They're really just brick and mortar stores, by a different name. In order to compete on price, Amazon has to eat the inflated distribution cost by setting the widget price low enough that, when delivery is added in, the total isn't way out of whack. Some folks may have been stupid enough not to do the arithmetic, but not now. The total price of that widget from Amazon has to compete with the price, ignoring the total cost of driving, at Target or WalMart or Sears.

The only way Amazon gets justifiable profit is when it gets monopoly power to pass on the inflated delivery costs. There is no other option, save bankruptcy. The numbers can only add up one way. Well, unless you're an accountant who can bend or change the rules. Clueless CEO and clueless quants seeking to be "the king of the world!" Jack, and his boat, sank.