31 December 2015

Is Life a Cabaret?

There's the signature song from the musical "Cabaret", "Money", as in it makes the world go round. I've had a recent confrontation with a certain Seeking Alpha rightwinger over the notion that there is, or isn't, a profit recession and whether it matters. As I've mused a number of times, financial quants have a problem these days: Beijing does what it pleases with its economy. Math/quant models that so much as touch China tread carefully.

A more recent (today, actually) piece looked at NVIDIA's exposure to China. The comments express a contrary view to the piece. But, the fact is, what Beijing does with the yuan/renminbi is opaque. Which led me to an October Fortune story, which came to the same conclusion at the macro level. Just what these endeavors have been saying for some time. Data is the product of arbitrary human decisions. Not the other way round.

To the extent that profits are down a tad, over-supply of petro and gaming of foreign currencies against the Almighty Buck, are the reason.

This overvaluing of the the U.S. dollar relative to the Euro, yen, and Chinese yuan has led to a huge trade deficit and was one of the drivers of the global financial crisis, as foreign savers piled into overvalued, "safe" dollar-denominated assets, like U.S. mortgage debt. Though global regulators have made strides in making the international banking system safer, the fundamental drivers of global trade imbalances have been largely ignored.

Such a nice quote; I feel the warmth of shared authorship. Coming to a preamble near you real soon now.

28 December 2015

The Most Socialist State in The United States

Sarah Palin, doyenne of the right, was the part-term governor of the most socialist state in the United States. She never, so far as I can see, ever admited to this, of course. But the fact is that a whole generation of Alaskans, today's generation, have lived off the returns of a communal asset. That asset, of course, is oil. The state's ownership rights are not by state fiat. According to this piece:
When Alaska became a state, Congress mandated that the new state could not sell its resources in the ground. Whereas other states have the right to transfer land so private citizens can own and develop mineral wealth, Alaska cannot. The intent was for Alaska to retain ownership of resources, like oil and gas, to fund our government. Thus, Alaska became an "owner state."

That time, by the way, was Eisenhower. I haven't tracked the Congressional history of why that was done, but according to the Wiki, petroleum was known to exist prior to statehood was voted. And, of more import if accurate, 75% of recoverable petro has been shipped.

Which brings us to recent news. The phat life is over.
This is the nation's least-taxed state, where oil royalties and energy taxes once paid for 90 percent of state functions. Oil money was so plentiful that residents received annual dividend checks from a state savings fund that could total more than $8,000 for a family of four -- arriving each autumn, as predictable as the first snowfall.

Socialism from an asset not made by any Alaskan. Sweet way to live.
The energy industry's main lobbying group has vowed to fight Mr. Walker's proposal to collect $100 million in new taxes on oil and gas companies, while reducing by $400 million the tax credits they can claim. But at a recent town-hall event in Anchorage on the budget crisis, it was clear that the energy industry has some image problems, even up here.

"Alaska was a great state before oil came to town," Evan Beedle, 54, an unemployed former school bus driver and technician, told state officials at the meeting. "I remember when neighbors were neighbors and doors were unlocked -- now it's just a skirmish for the dollar."

Mr. Beedle continued, "I realize now that we have become dependent on oil."

What would Ayn Rand recommend? Just askin.

24 December 2015

Outer Limits

Recall Dirty Harry's other signature catch-phrase, "A man's got to know his limitations." One of the minor, occasionally major, themes of these endeavors is that progress, tech in particular, in the future is no longer extensible from past experience. Particularly using a 19th century American view. Now comes reporting that the next step in HDD capacity bump is stalled. Not that it won't happen, but not tomorrow. That bloody asymptote of knowledge of the physical world, again.
Companies like Seagate Technology and Western Digital believe that to hit areal densities beyond 1.5Tb/in², HAMR technology along with higher anisotropy media will be required because of supermagnetic limit (physical "pitches" on optical [sic] media become so tiny that it will not be possible to produce a powerful enough magnetic field in the HDD space to write data into them).

Moreover
Certain principles of heat-assisted magnetic recording were patented back in 1954, even before IBM demonstrated the very first commercial hard disk drive.

The problems now faced are not science, but engineering. Of course, lots of those bytes are never recorded if/when developers embrace Organic Normal Form™, and reduce storage footprint by an order of magnitude or so. And get DRI. And so forth. Someday my Prince will come. If only he weren't so damn drunk.

Santayana's Revenge

"Those who cannot remember the past are condemned to repeat it."

George Santayana, in "Reason and Common Sense", 1905.

Today brings a report that there's at least one Wall Street econ who is willing to pillory the Fed's interest rate move.
The historic rate move on Dec. 16 "was very, very stupid," [Mizuho Securities USA's chief economist Steven Ricchiuto] said.

While his assessment was blunt, his no-hike view had been shared by former U.S. Treasury Secretary Larry Summers, Nobel Prize laureates in economics Paul Krugman and Joseph Stiglitz and noted bond investor Jeffrey Gundlach.

Noting the error made by the ECB,
Despite the slowdown in U.S. domestic manufacturing due to sluggish global demand and modest growth in consumer spending, the Fed may raise again in March or April, according to Ricchiuto, who said the Fed has set the bar fairly low to hike rates again.

"Would the data allow for it to happen? I don't know," he said.

Data are meaningless in the face of countervailing motive and incentive. That notion is at least as important as Santayana's. Nevermind that I'm not Krugman.

23 December 2015

The Moral of The Story

A piece which came through R-bloggers on the subject of AI, caught my attention, and subsequent comment. Of more note, it crystallized a notion which has been floating around my lower brain stem for quite a while. It comes and goes, of course. The notion: is data science, by definition, amoral? Much as laissez faire economics makes no value judgments with regard to fairness or equity of "free market" structures, since it asserts the axiom that, in the end, any market is a fair and balanced arbiter.

Is any application of data/stats/quant disconnected from the use to which it is put? Is it allowable, as I noted in my comment, for the various psych- fields to use data submitted by, or taken unwittingly from, other humans for the sole purpose of directing said humans to behave in ways contrary to their self-interest? Picking which toothpaste, may be who cares. Picking one's government or financial system, may be we should care. Both politicians and marketers have since time immemorial spun narratives to convince those who wouldn't otherwise support their ends, to do so. Thomas Frank's "What's the Matter With Kansas" explored the result a decade ago. The tools of the influence peddlers have only gotten stronger. Vance Packard, in the 1950's, mined the same vein with "The Hidden Persuaders", and that was before transistorized computers even existed.

Is data science just a modern Hessian regiment?

22 December 2015

conSOLIDation

Well, the solid state storage world just got smaller by one player: NetApp has gobbled SolidFire. Frequent reader may recall a few missives here about SF, mostly passing on info from their constant stream of email notices. SF claims great superiority, but as I type, NetApp share had fallen to a 52 week low, but now regained about a Buck of the loss.

This leaves Violin Memory and Pure Storage among the contracting field of public flash companies. Not for long, I expect. Violin (~$90 million market cap) has been the butt of many BK jokes. Pure (~$3 billion, ditto) is still, may be, in IPO honeymoom phase.

The Tyranny of Average Cost, part the seventh

Just a bit of commentary from the AnandTech piece on a 5K monitor from HP. Edited for clarity.

gonzo98x - Tuesday, December 22, 2015
What's the deal with the price of this monitor when Samsung can squeeze a 4k display into a 6 inch screen and sell it for far less?

Doesn't it take a higher level of specialized technology to cram a 4k resolution into such a small display? To me that would mean a phone display should be more expensive or in the case a 27" monitor should cost less.

So where does the cost come from? Or is this simply another example of setting the price for something 'because they can'. We expect it to be expensive so it is expensive?

I'd love to see a teardown and bill of goods for a monitor such as this.

Insert


TheStu - Tuesday, December 22, 2015
I imagine it has something to do with the economies of scale. Whoever is making this panel for HP, they're probably looking at less than 100,000 sales (especially if it is NOT the same panel that's in the 27" iMac). Samsung is going to sell tens of millions of their phone displays.


bryanlarsen - Tuesday, December 22, 2015
A 6" display has an area of 15 in^2. A 27" display has an area of over 300 in^2. 20x in size, ~20x in cost, where's the disparity?


CaedenV - Tuesday, December 22, 2015
less material used, plus higher output capability, plus a more reliable process. It all adds up to far less cost, and far FAR less waste, so the product costs much less.
Think of it this way... a 6" 16:9 display has 15.37 square inches of material. A 27" display with the same aspect ratio has an area of 311.53 square inches.
Lets say that the phone display costs ~$50 (just for the sake of nice round numbers... I have no idea what a high end cell phone display costs). That would break down to $3.25 per square inch... extrapolating that to the larger display it would scale up to $1,012.50.

But there is a difference between a cell phone display and a computer monitor. A cell display is merely the display, and the backlight with minimal controllers and other electronics, and no housing... the monitor has a USB3 hub ($), a stand ($$), a housing ($), a controller supporting multiple inputs, resolutions, frequencies, and scaling ($$$), a power supply ($), higher shipping costs per unit ($), higher storage costs per unit ($) etc.

Plus, lets not forget about issues of manufacturing. Lets say that for every 1000sq" there is a defect that makes a device unusable. That means that for every ~67 cell phone displays, there is one bad apple, so the average cost of each display rises ~1/67th, or 75 cents per unit.
But if that same kind of manufacturing ratio is applied to the larger screen, then that means that one out of every 4 displays is going to be bad, which means a 25% increase in screen price (+$253 per unit! way more than 75 cents!).

TL;DR... smaller things are going to cost less.


Some folks actually get it, even when they don't say so.

21 December 2015

I Go Gaga for Gigi

Oh my!!!
I'm very pleased to announce the release of ggplot2 2.0.0. I know I promised that there wouldn't be any more updates, but while working on the 2nd edition of the ggplot2 book, I just couldn't stop myself from fixing some long standing problems.

An up to date book on the bestest graphing thingee. Be still my heart.

19 December 2015

The Price of Gluttony

Not all that long ago was a missive on China and the IMF. The notion offered was that being a reserve currency might not be the best thing in the world for China and the Renminbi. And so it has come to pass.
The I.M.F. and the Treasury have both urged China to let markets play a greater role in setting the value of the renminbi, which makes it harder for them to object now when market forces push down the currency.

So, now the currency crashes while the Almighty Buck continues to rise. The outflow of capital from China is documented. China is exporting deflation (with its cheap manufacturing) and cratered interest rates (with all that moolah chasing scant real investment). Yet another reason why The Giant Pool of Money won't go away.
A little more than two decades ago, Mexico concluded the North American Free Trade Agreement with the United States and Canada, cementing its position as an emerging market closely tied to the global economy. But less than a year later, as Fed rate increases prompted investors to move money to the United States, Mexico found itself struggling to protect the peso. It ended up letting the peso drop nearly 30 percent in less than a week.

I suppose there's a Chinese word that is pronounced "QE".

Please, sir, I want some more

Good news. DDR4 memory is cheaper than ever. For some definition of ever. Nevertheless, this piece on AnandTech claims prices are dropping. A lot. Well, yes.

Back in June, PC World reported about $2K for 128gb. Today, Amazon has 128 gb for under $1K.

You need a new-ish i7 (near as I can determine, none of the 6th gen i5 will do), and an 8 slot motherboard. Nevertheless, for an increasing number of quant/stat/data problems the complaint about R being memory resident is irrelevant. Yum. A few bucks for hardware, and you'll be super productive in generating the next economic collapse. You'll take your profits before the retail lemmings walk off the cliff, of course.

The Tyranny of Average Cost™ strikes again (from the AT article):
While there are only three major makers of DRAM left on the planet, they continue to fight for market share and profits. In a bid to cut-down costs, manufacturers of memory have to adopt thinner process technologies, which decreases sizes of memory cells and thus increases bit output per wafer. As a result, global supply of DRAM upsurges and affects prices.

Or, as all of my econ professors said at one time or another (but not in any MBA course, of course): "sunk costs are irrelevant to decision making". Once you've bought the plant/machinery, the only thing that matters is marginal cost of production. You've got to cover the amortization (if you're lucky) and direct costs.

Time to engage in Thought Experiment. What might lots more cheap bits in lots less silicon and at lowering voltage mean? Alas, the answer is likely that client-centric computing will get more time on life support. It's clear from reporting that memory makers are banking on phones/tablets/portables as the place to sop up all those cheap bits. Sniff.

18 December 2015

Stupid

Stupid is as stupid does. That's a very old adage. During the various arguments on various sites with regard to Valeant and Turing and KaloBios, I've been adamant that the American public shouldn't be forced to pay remaining life's earnings just to access a life saving/altering generic drug just so that the acquirer, who paid a stupid price for the drug/company, can get back the stupid price and 10% (or so) return on the stupid price. The wonders of our not really free market health industry.

Finally, an industry mogul admits the truth:
Imprimis CEO Mark Baum said that even if Turing's other investors decided to remove Shkreli, that alone would not be enough to change the fortunes of Daraprim. While at Retrophin, Shkreli hiked the price of its key drug, Thiola for kidney stones, to $30 a pill from $1.50. That price has not changed since Shkreli left.

"Once Turing went and paid what it did to buy that drug, they were locked into raising the price," Baum said.

Stupid is as stupid does.

16 December 2015

Those Rubes

Well, it has been a while since the last "I Still Hate Neil Irwin" piece, and not because the opportunity didn't arise, but out of the spirit of The Holidays. Still, the NYT Business Editors have gone deeply into their schizoid egos today.

In my dead trees version, above the fold on an inner page is a great graphic, which the web page doesn't have. It has just the video. Too bad. But, it still has the punchline:
In other words, higher interest rates, after all that, have translated into less inflation.

I'll leave aside, for now, the obvious problem that the interest rate is real return *plus* inflation. What the sentence really means, "jack up interest rates and foment recession, which kills everything; kind of like Agent Orange". Now, anyone with at least a minimally functioning brain knows that the inflation bigots have been crying, "it's here!! it's here!!" for years. But, of course, it isn't. And the reason it isn't is that all that moolah the Fed and ECB and the Bank of Japan have been dumping, or so it is alleged, into the world economy hasn't actually been into *the economy*, but into the financial system. Where it wends its way to corporations and hedge funds and the happy 1%. And, it hasn't trickled down to the 47%; the rich get that way by being tightfisted with free moolah. You would too, I expect, in the same situation. "Fuck you buddy, I got mine!!"
Once again, the sources of inflation:
1) wage push
2) cost push
3) demand pull

All three have been notable by their utter absence. Monetary policy, as being implemented these days, hasn't and can't compel any of these.

The moolah sits on balance sheets. The M&A extravaganza is only just starting. Why invest, and employ, when you can just buy your competitors? Not only less risky at the outset, but also gets you closer to monopoly (or, in a pinch, oligopoly).

Not so obvious, if one just scans the dead trees version, is Eduardo Porter's column. Since he echoes most of my Rational Macro-quant Analysis, which you can find by perusing past entries to these endeavors, I could have titled this missive "I Also Still Hate Eduardo Porter". But I didn't. Here are some quotes, each of which should sound very familiar.

first:
Still, the urgency to head off alleged inflationary pressures seems premature, especially given that the Fed and many economists on and off Wall Street have been crying wolf about inflation for years.

second:
... if the economy falls into a recession when inflation is very low, it might be nearly impossible for the Fed to engineer the negative real interest rates -- after accounting for inflation -- needed to jolt the economy back to life.

third:
There are other tools at the government's disposal to reinvigorate the economy. Notably fiscal stimulus, but with today's Congress, that's doubtful.

fourth:
The previous consensus among economists that we would rarely, if ever, reach this floor was based on analysis of the American economy after World War II, a period of mostly robust, stable growth. Extrapolating from that track, Mr. Williams calculated, a nearly two-year contraction like the Great Recession, which shaved 5 percent off economic activity, could be expected only once every 570 years.

The postwar golden age, though, turned out to be atypical. Basing the analysis on broader historical data -- the experience of 17 developed countries since 1870 -- raises the odds to once every 23 years.
(This one is among my top 1 or 2 errors made by the econ/quant crowd: the post WWII era was the anomaly just because Western leaders, public and private, were of the Greatest Generation and still viewed their world as a social construct. Ayn Rand had not yet polluted their hearts and minds.)

fifth:
Among economists and investors, the problem with the Fed's 2 percent target is that just about everybody believes it is really a ceiling. That makes it even harder for inflation to rise to that level. The market expects the Fed to act pre-emptively to ensure it never goes over that line -- which is what it seems to be doing now.

So, the right wing in DC (and Europe, too) has dug in its heels since before the Great Recession, stopping any meaningful fiscal tools from being used. The simple fact is that supply side machinery has never worked, and still doesn't. What hard headed CEO is going to make more widgets when there isn't unsatisfied demand for said widgets? Nary a one. Give said CEO piles of moolah, for free (or nearly so), and he'll find the most self-serving way to use it. Which may be just to sit on it, waiting for the next recession/depression/deflation to make his pile more valuable. Appeasement never works.

11 December 2015

The Tyranny of Average Cost, part the sixth

Once again, reporting which demonstrates that average cost can't be avoided.

In this case, we find the capex heavy chip business with major agita.
Meanwhile, smaller peers Abu Dhabi-owned GlobalFoundries and China's Semiconductor ManufacturingInternational (NYSE:SMI) are struggling.

"GlobalFoundries (is trying) to grow into existing capacity as its parent deals with sharply lower oil prices," he wrote. And, "while Semiconductor Manufacturing is growing capacity mid-teens, its revenue base is less than 10% of Taiwan Semiconductor's size. . .."

All that capacity, and minimal growth in, or even slack, demand. Soon, the physics and engineering of chips will hit The Wall. It will be world sorta, kinda like a pharma world where every drug is generic; anyone who wants to run a fab has exactly the same machinery, since Newton and Heisenberg say so. I wonder whether all those nascent coders that Bill Gates wants will be happy campers? Plumbers and nail pounders in FL will be making more. There was a time when the American, at least, economy rewarded brains and learning over simple minded brawn. May be not so much now and the future. Just as that shill Dan Gilbert touts Prudential Insurance in the midst of a global savings glut (save more, you'll earn great interest, just buy a Prudential variable annuity), turning out ever more coders just reduces the per coder income. There's not much objective about economics, really, but supply and demand do make a difference. Some times one or the other is manipulated, but there will always be a market clearing price, as you heard in the first week or so of Econ. 101.

10 December 2015

Angela's (risen from her) Ashes

Regular, gentle, reader should recall my musings in the wake of Angela's induction into the Apple pantheon, as uber-marketer. In sum, I said that such an appointment was a tactical and strategic mistake... if Tim and Angela believed, and acted on such belief, that her stint with Burberry, and strategy employed there, made a bit of sense for Apple.

Burberry sells coats to the 1%. Burberry has a built-in plan: sell the Burberry coat appropriate to each occasion. Even the 1% aren't, on the whole, stupid enough to buy the idea that they should have an iPhone for the office, one for at home, one for the Hamptons, one for ... Nonsense, of course. What I proposed, and also in the macro world, was that an economy with massive, and growing, right skew of income and wealth required ever more differentiable bits of bling to entice moolah from those with more of it than brains. Kind of like buying Manhattan, but day after day.

Well, it's widely reported that Apple has annointed another source of bling for the Apple Store. The Times story mentions that the website offers up the execrable B&O products; heavy on odd design, light on good sound. The Apple Way.

Angela&Tim must be reading here, since they've done what I said was needed; ever more different bling items. We've had such high inequality America before, almost the entire history of the country from inception to the end of WWII. Equality and voting democracy are really the exception, just the couple of decades ending with the OPEC oil embargo in 1973, rather than the rule. As to the latter, the Supremes are about to hand the country over to the right wingnuts through the expediency of redefining how population is counted for apportionment purposes. And, once again, the Dems have been asleep at the switch. The case asserts that only eligible voters (and such are not explicitly counted anywhere) should be counted as population in the enforcement of one-man, one-vote. The rural states, being mostly mad, old, exploited white people, are the heart of rightwing-ism. The urban states are younger with larger families, and more semi-legal residents. Nevermind that such has no historical basis. That's never stopped Roberts and the rest in the past. Bring back the good olde days of the Gilded Age and negroes counted as only 3/5 of a person.

Just fur instance:
In 1793, for example, Southern slave states had 47 of the 105 members but would have had 33, had seats been assigned based on free populations.

What all this means to Angela&Tim is clear: while they've been able, so far, to get really, really rich on one device, those days are numbered. China, so far, hasn't moved the iPhone needle all that much. So long as the Angela&Tims of the world focus solely on the upper X% of a market, they've explicitly limited their, well, market. In order to grow in the presence of a stagnant, at best, TAM, they must provide to that TAM evermore intoxicating bling. That's not an easy job, and until the iPhone, one might believe that Apple hadn't yet done it.

08 December 2015

Watch My Pulse

When the Apple Watch was first talked about, and after its release, I noted that the sensor deployment was stupid. There, I said it. Homer Simpson stupid. There's a reason health workers read from the inside of your wrist. That's where the plumbing is most exposed. Well, reports indicate that Apple (may have) has admitted defeat. Here's hoping. OTOH, having the sensors in the band diminishes (and therefore replaceable, if Tim says you can) the planned obsolescence paradigm that Apple so loves: no upgrade, just buy the next version. We'll see.

07 December 2015

Beware of Banksters Seeking Gifts

Gretchen Morgenson has a very long piece on the latest assault by the Banksters, aided and abetted by Obambi. Hold your nose, try to keep your lunch down, and just read it. I'll offer up just one short quote:
But bringing private capital into the mortgage securities market poses perils of its own, other housing experts say: Allowing too-big-to-fail banks to dominate the nation's mortgage market would crowd out smaller lenders and expand the federal safety net, putting taxpayers at greater risk of funding bailouts in a downturn. Relying on mortgage insurers to provide that capital also seems dubious given how badly these companies performed in the 2008 crisis.

Moreover, private capital would probably flee the mortgage market at the first sign of trouble, as it did during the recent credit debacle. This raises questions about the availability of home lending when such a system goes through a rough patch.

In other words, when the shit hits the fan (again) the rule will be (again): "socialize cost and privatize profit". The American way.

Go With The Flow

For those still not getting it, there really is a reason why USofA interest rates are low and P/E rates are high. There's moolah on the move, again and still. Get used to it, and admit that there's a global savings surplus.

Also, now having a reserve currency, Beijing risks greater punishment should it pull more instant devaluations. Be careful what you wish for.

06 December 2015

On The Fish and The Bicycle

If you're old enough, or read history, you'll recognize, "A woman needs a man like a fish needs a bicycle". Woman's Lib from the 70s. There are other mythic cycles out there. Lots of them. I collided with this one and was chastized for my rhetoric. Oh well.

As I've mentioned numerous times, there're more hierarchies in software than there are in the real world. The real world is relational, much of which is mangled into hierarchical datastores. And all the mess that ensues. Equally, there are more math-y models (cycles, of course) of human processes than there are such actual processes.

The comment in the global warming post which sought to equate physical phenomena, which we pretty much understand, to human processes highlights the problem with quant away from hard science. The notion that there have been observed cycles of economic activity in the past means nothing with regard to predicting future conditions.
Sismondi and his contemporary Robert Owen, who expressed similar but less systematic thoughts in 1817 Report to the Committee of the Association for the Relief of the Manufacturing Poor, both identified the cause of economic cycles as overproduction and underconsumption, caused in particular by wealth inequality. They advocated government intervention and socialism, respectively, as the solution.
(And you all thought this Gini index stuff was just the recent maunderings of the soft-hearted?)

One can observe physical science data, fit some function to said data, and go about predicting the future of such a system. Even then, one is not supposed to estimate off the end (either end) of the data. Time series folks do so regularly, of course. Doesn't mean they should. While the arithmetic may be correct, what may not be correct is that the underlying structure has reached stasis and thus continues to exist within the range of the independent variable(s). And so it is for economic data.

We, like it or not, are at the top of that S curve of knowledge. Much of what passes for economics, and more or less the rest of social science, is grounded in previous eras of small population and abundant natural resource. Some retain the notion that "human ingenuity" will solve any difficulties we face, or will face in the future just because the 20th century was more "progressed" than earlier. This Pollyanna view results from looking at the past, the 19th century is a favorite, and applying that course of events to today and our near future. But the 19th century, and even the first 50 or 60 years of the 20th, was a time of filling in many, many gaps in our understanding of the physical world. The periodic table in particular. Most of post-Newtonian physics. Thermodynamics, arguably the most important field in physics affecting our day to day lives, became real in the 19th century. I often wonder what percent of the population realize that semiconductors were understood in the 19th century? We won't have such a frontier again. Organic chemistry is the only field which is more or less infinite, due to the nature of carbon, hydrogen, and oxygen bonding with the rest of the periodic table; one can synthesize to one's hearts content. The speed of light means we're stuck on this blue marble, even if we manage to build rockets with something with more oomph than chemical rockets. Transport powered by Mr. Fusion engines would be nice.

Attempting to analogize human systems, whether economic or socialogical or ..., is foolish. We are not automatons (yet; Blade Runners not in demand) obeying Newton's laws or thermodynamics or ... We, some of us, make up the laws (generally to disproportionately advantage the law makers) which govern our behavior. At times, such as the DotBomb and Great Recession, when moolah is in supply greater than needed by the CxO class for physical investment (M&A don't count as physical investment, either), said moolah flows into fiduciary instruments. Said instruments are constructed out of whole cloth by some of us, to their benefit. They exist, to the extent that they are at all real, based on contract law. Some regulatory oversight exists, SEC and the like, but as both DotBomb and Great Recession demonstrated, clever lawyers will skirt both law and good (beyond short term) sense to make an extra nickel.

In sum, the laws are made, and re-made, for one purpose: socialize cost and privatize profit. Were classical economics, or any of the hard science analogizing, true; such could not happen. All costs, current and future, would appear in prices and there would never be either oligopoly or monopoly (or the demand side -sonies). Chinese cities, London and Los Angeles before them, would not be engulfed in toxic fumes; the true cost of pollution would be explicit in the cost of transport and electricity, etc. Pretending otherwise is just delusional.

01 December 2015

The Treeless Plain

Yet another example of the underlying fact about data in the real world: it's all relational, not hierarchical. Codd and Date are right. Being able to deal with hierarchies inside a relational engine was the best thing ever done. Try doing it the other way 'round. Yeah, you xml kiddies have been attempting that for about a decade and a half.

I hadn't known that SQLite had done so. What's funny? MySql runs on the Oracle (from before they bought it) engine, InnoDB. While Oracle didn't write InnoDB from the start, they've owned it for a decade, so there's no excuse. If nothing else, Oracle could supply CONNECT BY, which Oracle engines have had since long before IBM/DB2 invented formal CTE and got it into the standard.

He He He.

27 November 2015

Billions and Billions of Dollars

I've always suspected, just from acquired memory, that the real reason behind pharma's claim that it cost $X billion to bring a "new drug" to market is that drug companies simply don't pull the plug when the data say the drug has little chance of working and/or getting approved. For those who may not know, in the US, there are three sanctioned levels of trial, not surprisingly called Phase I/II/III. These are trials of the drug in humans. Prior to Phase I there are pre-clinical lab tests to, at least, demonstrate that the drug works chemically, biologically in glass, and biologically in non-humans.

Once FDA is convinced that the compound is non-harmful, or least non-fatal, in non-humans, clinical trials can begin. In general:
Phase I -- safety
Phase II -- safety and dosing, and possibly efficacy, in small trials
Phase III -- efficacy in large trials

As a rule, at least two PIII trials demonstrating statistical efficacy and clinical benefit beyond current therapies are needed to ask FDA for approval. The key points in these trials:
1 -- sponsors (aka, drug companies, mostly) are not required to provide the public, or investors for corporations, all data generated or FDA correspondence during development
2 -- FDA is not allowed to release much, if any, data until such time as it makes a marketing approval decision, and then the non-approval (aka, CRL) may be vague

The result of this is that drug companies often continue pouring money down a rat hole. It's what they do. That's my recovered memory of watching the drug business for the last decade or so. Finding clear data on how many drugs with failed/marginal Phase trials are then sent into the next Phase is difficult. Not the sort of information drug companies want publicized.

Part of the problem may just be a naive` view of stats, in particular what a p-value means. And, no, I don't say that as intro to pumping Bayes in clinical trials. Not even.

Then I found this piece. All is revealed.
And, of course, add to all that the entirely avoidable, but nonetheless remarkably prevalent, tendency to progress agents into Phase 3 that did not actually achieve positive Phase 2 findings (at least without the help of unjustifiable post hoc analyses).

So, here is where all that moolah goes:
If, for example, your primary end-point reaches statistical significance but every secondary end-point suggests no effect, its time to suspect the False Discovery Rate. Put another way, don't let the data from one single experiment (however important) dominate the weight-of-evidence. The attitude "well, the trial was positive so it must work - so lets plough ahead" may well be tempting, but unless the broader picture supports such a move (or the p value was vanishingly small) you are running a high risk of marching on to grander failure.

Leading to his conclusion:
Failures in large, expensive Phase 3 trials are the principle cause of poor capital productivity in pharmaceutical R&D. Some of the reasons for failure are unavoidable (such as, for example, the generalization problem). But the False Discovery Rate is most definitely avoidable -- and avoiding it could half the risk of late-stage trial failures for first-in-class class candidates. That translates into savings of billions of dollars. Not bad for a revised understanding of the meaning of the humble p value.

But, of course, drug companies won't do that, since they get to keep their bloated bureaucracies only if they continue to do trials. Cutting off the losers in PI or PII does nothing to promote that. So, they won't.

19 November 2015

You Say M-eye-cro, I Say M-ah-cro

One of the hallmarks, if not raison d'etre, of microeconomics aka The Corporate Perspective, is that macroeconomics is just the sum of all those homo economici maximizing their use of land, labor, and capital. Such argument has been used for centuries to justify all sorts of zero-sum gaming and short-term decision making. Keynes is the most well known but not first to recognize that the welfare of The Tribe amounted to more than the sum of each member's wealth. Arguing against the macro folks is the 1% argument that "only the little people pay taxes"; if you're of the 1% you directly buy your own cops and schools and such.

Short-term decision making is exemplified by "you don't miss your water until your well runs dry". In California, we find the 1% squandering water on lawns just because, today, there still is some water and today's price (which never seems to calculate depletion) is affordable for them.

Zero-sum gaming is exemplified by the likes of Airbnb, which is subject of Australian hearings. The same old story: we should be allowed to slough off social costs, both current term and long term, because we assert that we expand the larger economy. In the case of American sports teams being gifted with stadiums, often fully gratis, the carry-on effects of bars, restaurants, and memorabilia shops are asserted to bring in more commerce and tax on same than the cost of such stadiums and tax abatements. No unbiased study has ever agreed. What has been found, of course, is that such teams pull up stakes for some other jurisdiction which makes a bigger, dumber offer as soon as, or even before, the lease finishes. Such taxpayer/community gifts are only profitable to the community if customers are imported to the jurisdiction from external places; otherwise the community is simply transferring consumption from a loser (movies and bars and OTB) to an adorned winner (Your NFL Team). Plus, that winner gets extra profit from the fact of not paying substantial cost.

Airbnb has a more difficult case to make: it is a pure replacement for some other form of accommodation. What Airbnb gains, some other facility loses. Since Airbnb runs through Ireland, of course, all other countries see no tax benefit of the corporate cash flow. And, of course, all that happens locally is that Hilton loses a customer to Airbnb's rooming house. Such Airbnb facilities are often sub-rosa, so any local accommodation tax goes unpaid. The argument by the likes of Airbnb amounts to, "we're cheaper than X incumbent, so our customers will spend the difference in the locality". I won't stay at Hilton, but rather some stranger's back room, and I'll have dinner at the Hilton bar?? As if that actually made sense as a justification: the customer base spends the same, so let us avoid taxes because we help the larger, local, economy? So, even if the Airbnb argument were true, the net gain to the community is less than $0; the Airbnb sleeper spends the difference between Hilton and its cot. There is no net increase in the local economy. The only justification for taxpayers subsidizing Airbnb (by letting them skate on regs and taxes and such) is if those who sleep at Airbnb wouldn't otherwise be in the locality spending money.

The bottom line, so to speak: when analyzing micro effects, especially quantitatively, be careful not to get sucked into ignoring macro effects, both immediate term and long term. Nearly always you'll find micro actor(s) seeking to slough off costs to the macro world. It's at best a zero-sum game for the macro economy, while the 99% lose 99.44% of the time.

17 November 2015

Codd Has Risen

The kiddie koders still worship at the feet of flat-files and client-centric transaction control. Largely, it appears, because no one in University bothers to show them the Yellow Brick Road to sanity. There is the recent missive on the MongoDB attempt to dip another toe into the SQL pond, while maintaining that it's really, really a NoSql, client-centric transaction "datastore". "You don't need no TPM, roll your own."

Now comes Intel's parallel memory implementation for Xeon Phi/2. An earlier note.

What we have in a relational engine is SIMD semantics, and thus embarrassingly parallel. Someone is going to see the light (Larry the boatman, perhaps?) and adapt this hardware to Organic Normal Form™ data, mostly if not wholly, in memory.

15 November 2015

Commentariate, part the first

Every now and again I find a thread to which I feel compelled to comment. And, of those, every now and again, I feel the need to document it here.

Thus:
This is all unfortunate. With Xeon, SSD, high bandwidth, fully connected web there isn't much need to view the world as disconnected and client-centric. There just isn't. But adopting a centralized, data-centric world means we don't need gobs and gobs of code. We'd need a fraction number of coders. Just as their COBOL grandpappies cried, "I don't know nuthin bout birthin babies!!!", so too today's web coders discard the RM as irrelevant. We could have a development semantic which looks a lot like a *nix RDBMS talking RS-232 to VT-220s. No muss, no fuss. All coders are needed for is to write screens and input routines; the database takes care of the rest. Scary thought, is it not?

14 November 2015

The Price of Apostasy

No surprise: I've no respect for the NoSql crowd, being as how they're hell-bent on reactionary 1960s client code/file paradigm. Ugh. Before: COBOL. Now: java (and still, in the Fortune X00, COBOL). Before: VSAM. Now: MongoDB (et al).

Comes this bit of irony.
But one thing was missing from that enterprise messaging, perhaps because it went missing from MongoDB's Enterprise product: joins.

That wasn't the plan. The plan was originally to charge for joins (or $lookup, as MongoDB is calling the functionality). Yet MongoDB's ever-watchful (but not always paying) community resisted.

MongoDB's capitulation is, of course, a testament to the company's willingness to heed the voice of its community. However, it's also a testament to just how hard it is to make money on free software: "Here's a new thing but you can't have it!"

Of course, imitation is the sincerest form of flattery, but come on. Stick to your guns. It appears the 20-somethings who actually use MongoDB like the huge flat-file non-structure that enables their infinite employment. Good on them.
This means that MongoDB, like every other open source company, needs to figure out ways to sell something other than open source, and feature-level differentiation, for the reasons stated, won't do. Not for the stuff that really matters, anyway. Otherwise the community build, meant to be an on-ramp to enterprise payola, instead becomes a roadblock to adoption.

Well, PostgreSQL has been doing the Open Source thing for a couple of decades, and supports a number of customization shops. One might also argue that MicroSoft has taken the open-core paradigm with its purchase (and debatable integration with) Revolution Analytics R. Also, I'm among those who deny that java is truly open source. One might argue that applications written with it are, but Leisure Suit Larry controls the language lock, stock, and barrel.

13 November 2015

Dee Feat is in Dee Flation, part the thirtieth

Well, the shit has hit the fan; producer prices continue to tank. All those Austrians keep telling us INFLATION IS HERE!!!!!!! Yet, like Godot, never appears.

Down a record 1.6 over the 12 months. As said so many times, motive and incentive matter more than data to those who pull the puppet's strings

12 November 2015

Linus, Guido, Larry

Whatever one may feel or think about linux, python, and perl, the fact that each has a Benevolent Dictator For Life goes a long way to explain why some really love it. A BDFL means that the application has stated boundaries and norms upon which one can depend.

R, alas, has no BDFL. R is anarchy. I'm certainly not the first to notice this fact, and my particular gripes with the language are not necessarily widely held. An interesting post from an R consultancy takes a direct swipe at the anarchy. It concludes with the following (bold in original):
Data should always be the first argument to a function.

11 November 2015

The Tree of Life

The giant sequoia lives into the thousands of years; ring analysis of all trees shows some wider and some narrower, resulting from level of water, sun, and soil nutrients. More food, thicker rings. Nevertheless, tree growth can be viewed as monotonic increasing in perpetuity. More or less. The same is true of many natural processes.

Unfortunately, the same is not true of human processes, since, unlike trees, we (well, some few of us) control the supporting environment. Sometimes, in some places, the environment is conducive to growth in some aspect of human-kind. Generally, this is due to a self-administered advantage. The banksters, for example, made their $$$ by bending/changing/ignoring the rules of play. They won, and when they were caught cheating, pretty much got away with the filthy lucre. Nice work if you can get it.

Much the same transference of thought goes on in the field of demographics, most often by those who know little about the subject. Josh Barro recently flapped his gums on the subject. Being coy, he framed his analysis with "good news, bad news" meme. Using the title, "Your Kids Will Live Longer Than You Thought" doesn't help.
you need to look at cohort life expectancy, a statistic that adjusts for the fact that death rates tend to decline over time as health and safety improve.

As I noted at the beginning: people live longer today than they did in 1900 not because of some deus ex machina perpetual propulsion, but because some humans figured out how to staunch various methods of early death, notably infant mortality in the period from 1900 to 1960. It wasn't magic, or God, that led to the longer life expectancy at birth. If one goes, as I have urged many times before, and looks at the real data, one sees that life expectancy at age 65 (the metric that matters when discussing Social Security or retirement generally) has increased by a small fraction of at-birth. A really small fraction.

Medicine, these days, is largely concerned with adding a few weeks or months to the terminally ill geezers. Whether we should be wasting so much moolah on geezers is a valid question. The War on Cancer has gone about as well as the ones in the Middle East.
The Technical Panel on Assumptions and Methods established by the Social Security Advisory Board, an independent government agency that advises Social Security's trustees on matters including actuarial assumptions, says Social Security is systematically underestimating future declines in mortality rates, and therefore underestimating the likely life spans of young Americans.

This statement is the Tree Growth Proposal: there's been a continual, but with diminishing incremental effect, decline in early death up to now, so there has to be more decline in the future. We're not trees, and why we die early is mostly up to our behavior. Either we stop doing deadly things (smoking) or we find in medicine (surgeries, drugs, vaccines, etc.) specific therapies to slow or halt specific causes of early death. There is no deus ex machina.

Without specifics as to why, and where from, such future declines will increase I say, "horse shit". We know what causes most early death, and drugs and vaccines have eliminated most of them. Lung cancer has diminished as people are shamed into quitting and not starting. Smallpox is officially gone. Mostly, same for tuberculosis, flu, polio, and so on. The main cause of early death in the Rich West is too many calories, while in the Poor East it is too few.
In the long run, the Social Security Administration assumes the "mortality improvement rate" will be 0.71 percent -- that is, the odds of dying at a given age will fall, on average, that much each year.

Again, more horse shit, without specifics. We are not trees. We won't live longer if you just give us more water and more nitrogen. Gad.

10 November 2015

The Poisoned Apple

Another in the line of missives centered on The Tyranny of Average Cost™ was percolating in my lower brain stem, but not yet putting fingers to keyboard, when what should I see, not tiny reindeer, but this Debbie Downer report on Apple.

Back in the beginning, even before some readers were yet born, Apple was the computer company for the rest of us. An IBM/PC cost in the neighborhood of $5,000 (that's 1981 moolah, btw), while the Apple machines were much cheaper, and less expensive. Then came the Age of the Clone, and the race to the bottom, and so on. With the Mac, Apple became the computer for the upper X%. X being somewhere in the range of 10 to 20. The die, to quote Caesar, was cast.

These days, Apple is largely The High End Smartphone Company. But no one, that I've seen in the popular press, has asked the simple question: can Apple survive without all those low end (and, allegedly, unprofitable) phones flooding the world? And, of course for those following the bouncing ball, not even close. Let's consider a small thought experiment, since dragging out specific data will be arduous if not impossible.

There are a few major fabs: Intel, Taiwan Semiconductor, Samsung, GlobalFoundries.
But $500/wafer is a far cry from the $1,600 or so that a finished memory chip wafer costs, or the $5,000-odd cost of a finished high-end processor wafer. Of that cost half or more is capital depreciation on the equipment that converts the raw wafer to finished chips.

The lion's share of the capital depreciation is lithography equipment, basically cameras on steroids.

Trying to figure out the total shipments of all cpu is beyond my patience. But, if we take the 10 billion/annum shipments of ARM designs as a base, then ask how many of those went into Apple phones, we get some idea of Apple's dependence on foundries' capacity. In 2013, around 150 million were shifted.

So, 150 million out of 10 billion cpu. Apple, as is so often the case with the right wing, is parasitic off the low end. IOW, without all those 9.8 billion other cpu sales, Apple would have to absorb the amortized capex of the foundries all by itself. And this, of course, is the rationale of all those fabless chip companies; get rid of real capital in order to drive up RONA. By the simple expedient of having no assets and charging everything to direct cost. Of course, the assets do have to exist someplace, so from a macro view it's all a shell game. And a zero-sum one, at that.

Ironically, Apple's Watch has been held up as the great disrupter of the Swiss (aka, really really expensive) watch industry; and may not be selling all that well. In fact, it has been the actions of Swatch to stop supplying base movements to the rest of Swiss watch makers. The conflict has been going on for years, but boils down to The Tyranny of Average Cost™: Swatch has the existing capacity to supply all of the boutique watch makers, who themselves individually can't find the capital to create production capacity (which is redundant at the outset). Some of the larger small makers have gone into production of base mechanisms in the last few years, but it is wasted capital, in the macro sense. Swatch's high-end brands now have a cost advantage, in that the capex has been paid for by the boutique makers over the years, while they have had to invest in redundant capacity. The customer and the boutique makers lose, while Swatch wins. Ah, tyranny they name is Rand.

Yeah, I know, this is back of the envelope figuring, but the The Tyranny of Average Cost™ can't be denied. Apple is, through its foundry suppliers, able to cater to the 20% by virtue of the existence of the 80%. Or, as discussed some time ago, if only the 1% get healthcare, they'll demand Obamacare for themselves, since they surely don't want to carry the capex of the healthcare system all by their lonesomes. They may hate, and condemn their oppression by, the 47% but they need the 47% to participate in order to staunch The Tyranny of Average Cost™. If only the 1% get MRIs, how much will an MRI cost? A lot.

04 November 2015

The Tyranny of Average Cost, part the fifth

Regular reader may recall my observation on the folly of Amazon's business model: it's an order of magnitude cheaper to ship by rail than by air. All this building of stores, sorry Fulfillment Centers, was perfect evidence that Bezos sees the problem, but finds the most expensive way to solve it.

Now, of course, Amazon is going into the brick and mortar business. There must be great angst among the pundit class!! Betrayed by Jeff!!

The reason, of course, is it's cheaper to stock a B&M location with the high volume stuff, leaving the high-cost, low-volume stuff to the warehouses. It's only taken a couple of decades for him to figure it out. I suppose the extended interregnum from the declaration of disrupting B&M retail with mail-order only service to now will be enough for folks to forget that point. Just as the large iPhone was solely a decision by Tim to improve the lives of iPhone users, and not the evident demand for same from users, well... Here, too, Jeff will say that he's invented a new sort of B&M.

Yeah. Right. Sears went from being a mail-order only business to being a B&M outfit. Some people just refuse to learn.

29 October 2015

Shave and A Haircut, Two Bits

These are the two bits.

1) The Community Technical Preview of SS with in-engine R connectivity has been released. Runs with the Revolution R version (fee when in general release?). Will the free-as-in-beer version of SS support either version of R in-engine? I have my doubts.
For SQL Server 2016 CTP3, we support ad-hoc execution of R scripts via a new system stored procedure. This stored procedure will support pushing data from a single SELECT statement and multiple input parameters to the R side and return a single data frame as output from the R side.

Not as comprehensive as Joe Conway's PL/R, which lets you create arbitrary functions using R. Well, nearly so at least.

2) An announcement on Planet Postgres that Tom Lane has moved cube. Most folks have no idea who Tom Lane is, and his move isn't germane to this post. The move may be more significant to the PG development community, possibly. But in the text of his new employer's announcement, is this bit:
... as well as the recent of addition of Joe Conway, another highly respected member of the PostgreSQL development community and committer to the PostgreSQL project.

Joe does lots more than PL/R, but moving to an aggresive, by the looks of it, start-up poaching widely, we may well see more robust PL/R support in Postgres. Shut up those MySql folks. Sorry Larry.

28 October 2015

The Global Euro

Let's take a trip through Tim Cook's head. What must be Tim's constant thought? Well, as CEO of the best company on the planet, it must be: how to continue being the best company on the planet over the next decade, or so. Best, of course, from Wall Street's perspective. After that, he'll just pull the rip cord on his golden parachute.

How did Apple get here, one might muse? Was it Steve's genius? Not really. Apple rescued Steve as much as Steve is credited with saving Apple. Next was a failure, pure and simple. Until the iPhone, Apple was just a niche personal computer company. Still is, if you cull out all the non-computer revenue from the filings.

All that wonderful gross margin from iStuff comes from a simple fact: Apple works by cheap goods sold dearly. It has managed to do this by the cult of personality of Steve. The innards of any Apple product are bought in off-the-shelf. Yes, I know: the Ax cpu is wholly Apple's creation of genius engineering, blah blah blah. Fact is, not so much. All the teardowns of the various A chips indicate: added more cache, added more gpus (bought in, of course), and did a good bit of manual layout. The ISA of an ARM chip is set. The node is set by what the foundries can do. If the MO of Apple were not clear before, then the destruction of GT Advanced Technology should make it clear: Apple gives you a pittance, then shoves a telephone pole up your butt. Which fact answers the Big Question: why does Samsung supply Apple? The answer is, they can afford to. Only another economic Gorilla can. It's what the econ types call countervailing power.

It's worth recalling that the first iPhone had a strong advantage for the first couple of years: Apple had captured the supply market for cap touch screens which fact limited competition severely.

(I started this missive while still on the Island, and it's now mid-day Tuesday. I'll wait for the Apple quarterly news before continuing, although the share price isn't moving up.)

Just got back from a teeth cleaning, and Apple is getting cleaned in after hours. I'm going to wait until 8 pm eastern to continue. ... Was down a bit in after hours, now up a bit in pre-market. So.
"The law of large numbers does kick in," he added."Can the iPhone grow another 2-3 years or is it done growing?... I think investors are worried about the next two-year period. 'Can there really be iPhone growth?' Because absent that there's not really going to be much growth from Apple."

In simple terms: China was the last roll-up opportunity. If that sounds alien, check out what's been going on with Valeant; roll-up city. This quarter went OK, and Christmas may as well. Does either speak to the future? No.

Which brings us to Germany, Greece, the Euro, and the Trilateral Commission. One of, if not the largest, paranoid fantasy of the far Right.
On the right, a number of prominent thinkers and politicians have criticized the Trilateral Commission as encroaching on national sovereignty. In his book With No Apologies, former conservative Republican Senator Barry Goldwater lambasted the discussion group by suggesting it was "a skillful, coordinated effort to seize control and consolidate the four centers of power: political, monetary, intellectual, and ecclesiastical... [in] the creation of a worldwide economic power superior to the political governments of the nation-states involved."

Well, perhaps The New World Order is.

The point, of course, is that Germany, through the elegant expediency of forcing the Euro down all those countries' throats, has the benefit of a New European Order, which solely benefits it. While Volkswagen can adjust the price, in Euros, of the Passat across the countries (last issue of report), higher in France while lower in Greece, for example recognizing ability to pay; this is an active step which is totally different from the passive shocks incurred when China, say, pulls the rug out from under the Renminbi. What Apple wants, should Tim ever admit it, is for all countries to do as Bermuda has: a fixed 1-for-1 exchange with the US Buck. Bermuda even calls it a Dollar. That's what Germany got with the Euro. That's what Apple, and the rest of US multis want. Each quarterly they bitch about "currency problems".

A New World (Monetary) Order controlled by Apple and fellow co-conspirators wouldn't upset them in the slightest.

25 October 2015

I Still Hate Neil Irwin, part the third

Spent the last week on the Island, wonky WiFi and just the Wifey's Windoze laptop. No editor or links. Very quiet on the rabble rousing front. So, I get back and The Times does its cynical or ironic or sarcastic two-step. Time to rabble.

First, a long piece on collapsing commodity markets. I'll just leave that for you to peruse at your leisure. No, this missive is all about Neil Irwin and me. Yum.

More than once, I've referenced pieces by the esteemed Mr. Irwin because they provide reportage proving (or, at the least, demonstrating) certain themes of these endeavors. Most often, as this one, they deal with quant (data) issues in macro-economics. Always a thorny problem. Most macro data is generated as sample surveys of business and labor.

Today's piece deals with a subject near and dear to my heart: Dee Feat is in Dee Flation. Not been a missive under that title in some time. Not that there's been a dearth of data, of course, just that the tune remains the same: we're headed down the rabbit (or, rat, as you prefer) hole of deflation. The cash hoarders look likely to get their return on idle moolah. Sigh.

Irwin takes a new path. One I haven't discussed since, I'll guess, grad school. Perhaps undergraduate. The dreaded Phillips curve. He explains, a bit.
How much faith should be placed in a line on a graph first drawn by a New Zealand economist nearly six decades ago, based on data on wages and employment in Britain dating to the 1860s?

The answer, of course, is not much. Phillips, as many macro-economists (esp. of the Right Wing, labour hating, branch), assigns only that sole cause to inflation. But, as the first piece so clearly shows, flation isn't just about wages. Far from it. Just ask your local greasy spoon proprietor why s/he's exploded the price of your 2 over easy.
As the Fed's chairwoman, Janet L. Yellen, put it in a 2007 speech, the Phillips curve "is a core component of every realistic macroeconomic model."

Except it doesn't work. Or at least, it hasn't worked very well in the last few decades in the United States. And it has proved particularly problematic to try to use that historical relationship to predict where inflation is going.

What he neglects, shamefully, is tell the reader about other drivers of flation. That's my task. Again.

To refresh, there are three ways flation (either direction) happens:
- wage push
- cost push
- demand pull

The first happens when workers get paid more than their marginal product (wiki), at least according to even Right Wingnuts who worship at the feet of classical economics. Otherwise, it's just Social Darwinism with capital compressing wages to subsistence, i.e. slavery without the chains. Which worked fine, in a macro sense anyway, during pre-industrial and even early industrial regimes. These days, with such heavy capital requirements, the The Tyranny of Average Cost™ makes it a fool's enterprise. Irwin goes on a long discussion of why Phillips and actual inflation are disconnected, beyond the simple arithmetic.

The second happens when there's not enough widgets to satisfy needs. Petro goes through cycles, which we're always experiencing. And, for the nonce, your 2 over easy. Why the likes of Irwin can't point out this obvious fact is puzzling. Over the decades, and centuries, I'd wager that material shortage is the number 2 cause of inflation.

Finally, that third is what the Right Wingnuts prefer to call "debasing the currency", and why they hate QE and fiscal policy and every other tool used to resurrect an economy. Argentina in the 1980s. But even Left Wing economists know that dropping moolah from airplanes will, if long enough and large enough, cause inflation. This is particularly true when an economy:
- is largely non-self sufficient
- lacks surplus capacity in desired widgets

In other words, the USofA today. Financialized economies, Bermuda is my favorite whipping child, are more susceptible to demand pull inflation, since the moolah dropped from airplanes won't be spent on domestic production. Spent on domestic production, the point of dropping moolah from airplanes, we find more demand for home grown widgets; thus more employment, more investment, more profit and so on. The point of dropping moolah is simply to get folks buying stuff. However, if the stuff isn't created domestically, the follow-on benefits don't occur. And so it was in Argentina.

Simply relying on naive` Phillips is foolish.
If you simply look at the unemployment rate in the United States versus the Consumer Price Index, excluding volatile food and energy prices for every year since 1958, there is nearly no statistical relationship at all, just a jumble of dots. (A best-fit line actually points the wrong direction, correlating higher unemployment with higher inflation, albeit very weakly.)
[my emphasis]

To the rescue comes a Freshwater Economist, Robert J. Gordon of Northwestern. Yes, another nerd who sees the future as distinct from the past. The link is an interesting read, for nerds anyway, but on to the day's topic:
Robert J. Gordon, an economist at Northwestern University, has his own version that he argues explains inflation levels throughout recent decades. But it is hardly simple. Its prediction for inflation relies not just on joblessness but also on measures of productivity growth, shifts in food and energy prices and overall inflation over the six preceding years.
...
or by factors that policy makers have little control over (like what happens to oil prices)

In other words, the sum total effect of those two other, oft ignored, forces. It ain't just the greedy 47%.

The Tyranny of Average Cost, part the third

Once more, The Tyranny of Average Cost™, Volkswagen's emissions venture:
EA stands for "entwicklungsauftrag," or "development order," and signified a major new engine line. Volkswagen deploys its engines and other components not only in Volkswagen cars but also in other brands belonging to the company, like Audi and Skoda. The strategy saves money because costs of development and production are spread across a large number of vehicles. But it means that any problems potentially affect a large number of vehicles.

Once more (the EA189 and EA288 engines): 'we lose money on each unit, but make up for it with volume'. Thus spreads the infection.

17 October 2015

The Walk

No, not a review of some movie about some Freedom Man skipping on a wire between two towers that no longer exist.

This is about walking the walk and talking the talk. More particularly, that Agile doesn't actually do either. The piece is really an exercise in self-promotion, but embodies my experience. It isn't RM-centric, but worth the read. (Disclosure: found via Database Weekly newsletter; you should sign up if you aren't already.)

16 October 2015

The 2 Percent Solution

For some time, both the quant version of these endeavors and the macro-economic version have harped on the quite apparent lack of investing in real assets. Corporations are sitting on $2+ trillion just overseas. QE money has largely gone to financial instruments. True innovation in physical widgets has clearly stalled: the best new thing is Force Touch on an iPhone?? Really?

The CxO class just can't seem to find creative ways to allocate fiduciary capital to physical form. I found this HBR piece recently and was going to write it up. Nah, just go read it. The bottom line, so to speak, is that the Masters of Wall Street are extorting the CxO class to put moolah into fiduciary, rather than physical, investment. The United States of Bermuda.

Well, even Yahoo! Finance (am I the only one to view that as oxymoronic as "happily married"?) now gets it. Real value in an economy comes from physical investment in productive processes. Finance and real estate and insurance (FIRE, as it's known) are just ways to skim moolah off the transfer from savers to borrowers. All their profit comes from extortion. The American Way.

15 October 2015

The Sun Salutation

From the time I exited grad school, I've practiced (and studied when I found an instructor on my wavelength) tai-chi and yoga. Both are Eastern, and tai-chi was originally a martial art. Both have, in Western incarnations, been promoted more as exercise of mind/body. Both are grounded in the same principle: the center (i.e. the space of the pelvic girdle) is all; all energy flows from it and out to the extremities. Golfers have been known to seek guidance from teachers of one or the other.

Tai-chi has a documented history that goes back some centuries, and has evolved into 4 (or 5, depending) schools. Yang is the most commonly taught, and the one I've studied. Again, when an instructor was available. Yoga, on the other hand, goes back millennia, and hatha form is what is mostly taught in the West. Both practices share a common attribute: each instructor popular enough to have a school/academy/franchise modifies "official" forms to better express his/her view of the fundamental principles of the art. And so it is. I first studied tai-chi with a student of William Chen, who taught a Yang form that looks nothing like the standard 24 step Beijing form. The one I teach now came from Art Ziello, from Don Miller (who is the brother of Julian Miller who taught me my first form in Cambridge too many years ago). And it's different from Art's version, which doesn't follow the official 24.

The same process happens in yoga, perhaps more so. Hatha is the main basic school in the West, but each instructor modifies to suit. And, unlike tai-chi, yoga practice is some arbitrary (well, sort of) combination of poses. The most well known documented version is, likely, "The Sivananda Companion...".

"What in Heaven's name does this have to do with quant and normal databases?" I here from the slackers in the back. Well, this. If you've been paying attention to the software world for more than a few days, you're surely aware of the patent trolls, software patentability, and suits from the trolls aimed at various and sundry. The latest (so far as I know) salvo, came last week when the Supreme Court declined to re-open Vringo's loss in the Court of Appeals for the Federal Circuit (CAFC, usually). The CAFC basically told Vringo to shut up and go home by gutting Vringo's "win" over Google. Ever since, the innterTubes has been a Tower of Babel.

The Blue Corner: "All software is algorithms, thus explicitly not patentable under both the Constitution and Federal law. Copyright is appropriate."
The Red Corner: "Software is unique intellectual property and therefore patentable."

Needless to remind, I'm in the Blue Corner and the Vringo, et al, are in the Red Corner. It bears reminding that the first software patents came from a Patent Office that had no experts in software, and were easily gulled (the account goes on a path I don't agree with, BTW).
What we learned is that things that we were doing in previous months are no longer acceptable. It seems that early last week a memo went out from the powers that be to the examiners handling Bilski-related applications, and in the memo it was explained that merely putting "computer implemented method" in the preamble of the claim is not something that will any longer work to overcome a patentable subject matter rejection under 35 U.S.C. 101.

More recently, not so much. The fact that a Supreme Court, deep in the pockets of the Red Corner generally, would refuse to bless Vringo's attempts is of note. Which brings us to today's news.

Bikram Choudhury has made an attempt to copyright his version of yoga, aka "hot yoga". He's just lost. I suspect that lawyers from the Red Corner cabal have already been on the phone offering any and all assistance to right the grievous wrong. All the way to the Supreme Court. You see, the court found that a sequence of yoga poses was not a copyrightable entity!!! This surprises even me. A sequence of words, as poem or novel or essay, certainly is.
"We wanted to establish that you can't own sequences, Mr. Drost, a founder of Evolation Yoga, said. "Traditionally, you take a teacher's training and move forward with it."

I can hear all those Red Corner sphincters sucking half the planet's atmosphere into their lower colons. It is well established that software is a sequence of discrete steps, taken from a previously established set, either the high-level language defined syntax or the cpu ISA. Now, here is a precedent that an real-world sequence, not even the abstract sort of computer languages, isn't even subject to copyright, let alone patent. There's going to be gallons of Chivas downed around Wall Street this week.

The Simple Truth

Up through my stint with CSC, SQL Server was a minor database in my trek. Once, even, the principal datastore. Along about then, I found simple-talk. Then, much of the discussion was driven by database design and development concerns. Celko was a frequent guest essayist.

In recent times, more of the essays have been couched in coder jargon. Sad.

One that's up on the front page, deals with continuous integration of databases. It's interesting, but what I found the j'accuse moment was this bit from a comment by rogerthat:
The worst times we've had with failing database builds were a result of using code first development (program design refactors database design, somewhat automagically). ... Experience leads us to design the database to handle the cited requirements while still keeping the database flexible (and no, this does not include unrealistic character data lengths).

While rogerthat doesn't use the N word, it's clear from the comment that normalization is the key. There is hope for humanity.

09 October 2015

Captain Obvious, part the second

This endeavor's been bleating about the need to use SSD as a matter of course from the beginning. More than half a decade. It is with some amusement that I read this quant/R related post. He who laughs last doesn't get it. Gad. Quants are just as bad as "database developers" who continue to dump flat files in SQL databases. Gad.
a huge increase in performance to the application overall

Well, yeah.

06 October 2015

The Tyranny of Average Cost, part the third

Another tidbit, this time from AnandTech.
Yet more worrying for manufacturers, the costs of preparing chips for new nodes isn't just rising but rising quickly, with mask sets already over a million dollars and expected to grow even further thanks to the high costs of developing masks for current multi-patterning technologies.

The Apple paradigm, selling only to the top X% (thought to be 20-ish), can only work for a slim minority of vendors. Moreover, it won't even work for that slim minority if they become the *sole* vendors from some external production process. In simpler words: the Apple paradigm only works, for Apple, iff there are other vendors to support the capex of said external production process. "We lose money on each widget, but make up for it with volume." Without all those, alleged, money losing smartphone vendors, Apple wouldn't have suppliers to eviscerate.
By splitting up a chip in this fashion, the number of transistors laid down on the leading-edge node would be held to a minimum, resulting in a smaller module that would be cheaper to design and cheaper to produce than a full SoC, all the while the other modules would be relatively cheap to produce and cheap to design (if not largely held over from existing designs to begin with).

There was a time, back in the late 70s and early 80s, when IBM held a massive tech lead on the seven dwarves (may have shrunk by one or two, don't recall exactly) with their "thermal conduction module". This was before VLSI and chips as we know them today. It appears that Marvell has discovered old wine and fashioned a new bottle. Smaller, yes, but exactly the same idea.

There will always be some new way to spread total cost, esp. capex. But, the bottom line, so to speak, is finding expanding demand for cycles. WinTel did that for decades. Now, it's the smartphone vendors and telecoms making the yin-yang of sink and supply. Life changes not so much as it appears.

04 October 2015

Hubris

Volkswagen demonstrated, once again, that The Masters of Universe make money the old fashioned way, they steal it. So, what does the Angela of death have to say?
"I believe the reputation of the German economy and the trust in the German economy has not been shaken by this to the extent that we are no longer considered a good business location," she told Deutschlandfunk, according to the text of an interview due to be broadcast later on Sunday.

Of course, Germany is a good place to do business. Government is the lap dog of business. Germany didn't define fascism, Mussolini did, but they perfected it. Government for, and by, capital. So long as Germany can export, particularly in the fixed rate Euro, it can sit fat and happy. Without export, the German economy collapses. Much like China.

02 October 2015

No Soup For You

Out of curiosity, I went looking at the MS/SQL Server site to check up on the state of the R-in-SS implementation. Not there, yet.
...still work in progress, not available yet.

Sigh.

01 October 2015

Told Ya So, part the second

Gentle reader may remember my musings that the ad blocker spewing was so much baloney. The cure is simple: price based on actual views. Seems Sergei and Larry are regular readers. Google henceforth will do so.

Now, was that so hard?

30 September 2015

I'd Rather Be Lucky Than Good

There's an article on SA, that starts with Abraham Wald, a Brit stat who, during WWII defied the generals. Not a team player, he. The issue was how to re-armor aircraft in order to save them and, more importantly, the flyers. The generals looked at the bullet holes, pointed, and said, "there". Wald also looked and said, "not so much".

What's of interest, beyond the article, is the comments. The gist of which is that, following Wald's thesis, stock picking is foolish, and that inordinant gains are, mostly, just plain luck. You get the market rate of return, period. Unless dumb luck smiles on you.

Which, in turn, raises a question not addressed by either the article or the comments.
Something similar occurs in the investment industry. It claims that some people are extremely skilled, since year after year they've outperformed the market. They'll identify these "investment gurus" and convince you of their abilities. But a simple thought experiment can show that it would be impossible to not have these gurus produced just by luck.

If one goes to the track, and bets on FlapperFluzzy in the seventh, at 40 to 1, and she wins, you get taxed at median income rates, not some capital gains slap on the wrist rate. You got lucky at the track. You don't get an additional tax break.

So, if luck is so much a part of inordinant investment gains, and has even less benefit than playing the ponies, why not revert to the 91% rates of Eisenhower? He of the "fear the military industrial complex"? After all, what else are they going to do with the moolah? Stuff it in a mattress?

29 September 2015

Structured Relations

The McArthur grants hit the news today. Always expecting to see my name. Alas, not this time. Again.

But there was this entry:
Christopher Re, 36, Stanford, California. Stanford University computer scientist, who has created an inference engine, DeepDive, that can analyze data in a way that is beyond the capabilities of traditional databases.

Of course, my first reaction was along the lines of the McArthur folks are pretty stupid, if they think some NoSql mumbo jumbo is stronger than the RM. So off to find DeepDive. This is what I found:
DeepDive helps bring dark data to light by creating structured data (SQL tables) from unstructured information (text documents) and integrate such data with an existing structured database. DeepDive is used to extract sophisticated relationships between entities and make inferences about facts involving those entities.

I didn't get the grant, alas. But the RM won the war. There's satisfaction in that. So, in the end, Codd and Date are right. So, may be, the NoSql folks with all jump off the Golden Gate Bridge; holding hands in failed solidarity.

27 September 2015

Limitless

No, this is not a missive about a soon to premiere TeeVee show. Hardly. No, more musings on the evident reaching of the asymptote of human understanding of the physical world. More specifically, the glass is 99.44% full, not 10% as it was in the 19th century. Don't tell the Rightists, their heads will explode.

More confirming reporting today.
Technologists now believe that new generations of chips will come more slowly, perhaps every two and a half to three years. And by the middle of the next decade, they fear, there could be a reckoning, when the laws of physics dictate that transistors, by then composed of just a handful of molecules, will not function reliably. Then Moore's Law will come to an end, unless a new technological breakthrough occurs.

Some, with whom I've roundly disagreed in various forums, have suggested (not uniquely or first by any means) that some sort of bio based artificial neuron will displace the silicon transistor. Think about this for a second. At 14nm a transistor is composed of 36, more or less, atoms
Not quite, Si is FCC so the lattice constant is greater than the close packed atomic spacing by a factor of √2; you could have up to 36 atoms spanning 14nm depending on crystallographic orientation.
here
How are organic molecules going to top (bottom?) 36 atoms? Not hardly. In any case, as we see with driverless cars and computers that beat the best human chess players, the human brain has its limitations. Even Chinese industry is replacing cheap hands with robots. Oh, where will it end?

How many atoms does a transistor need to be this side of probabilistic? Haven't found a definitive statement, but 36 doesn't leave much room for production error, etc.
In July, Intel said it would push back the introduction of 10-nanometer technology (a human hair, by comparison, is about 75,000 nanometers wide) to 2017. The delay is a break with the company's tradition of introducing a generation of chips with smaller wires and transistors one year, followed by adding new design features the next.

Some say, quoted in the Times' piece, that slowing of Moore is good for competition. On the other hand, it could also mean that the first-to-Xnm will have a solid monopoly, since being second or third will be profitless: we've still not seen the legendary 450mm wafer, since there isn't sufficient demand. Xnm will increase output for that first mover, but without a Windows/Office cycle sink, where will the demand come from? IoT, say some. Yet the Jeep remote hijacking and VW perfidy make IoT a bit more of Big Brother (or, Borg) than might have been previously obvious.

A man's got to know his limitations. And the limits of the physical world. Einstein allowed that the universe is finite but unbounded.

23 September 2015

ZPOP [update]

It's been mentioned in these endeavors a few times: 99.44% of macro-economic data is from surveys, rather than census. The sole prominent number is weekly unemployment; that should be a census of all recipients of UI. Other than that, it's sample data mostly run by various parts of the Census Bureau for BLS or whoever. I've admonished gentle reader, on occasion, to read footnotes to the unemployment report, to see:
1) there's not just one unemployment rate
2) the CI of those rates might make your head spin.

Now, comes reporting that the Atlanta Fed is proposing a new measure of unemploymentthat they're calling ZPOP. In essence: percent of prime age working population who want jobs, with jobs.

The first question which crossed my mind, naturally, when reading the reporting: is the Atlanta Fed Red or Blue?, mostly right or mostly wrong? The regional Feds aren't unbiased, non-partisan think tanks, much as most would have you believe. Here's the list of the Fed banks. Lockhart is president of Atlanta. Here's the WSJ rankings of accuracy. Click the labor tab, and Lockhart floats to second. Not bad.

In any case, the notion of focusing on "prime age" unemployment as the standard bearer makes perfect sense. Note, as has been recently mentioned, that labor force participation decline is not from geezers going out to pasture. It's from the genWhatevers living in Mom's basement because they can't get any kind of job. We can't afford it.

[update]
Upon further chewing, I've got to discuss something in the CBS News piece. The author is an academic economist (self-describes as econometrician). Here's the quote
One final note about this and other measures of resource utilization. Full employment is defined as more than full utilization of resources. Just because a worker is employed doesn't mean he or she is doing what they're best at or employed in their most productive occupation. If an unemployed engineer takes a job waiting tables to feed the family, that worker will be defined as fully employed, but that worker's potential is hardly fully utilized.
[my emphasis]

This point of view tends to be Left Wing; Right Wingnuts would rather that anyone but them live at subsistence: more of the economic pie for themselves, of course. But there's also a real world issue, which is that, over the last few decades, the more education one has the more likely you're employed in the non-productive sectors of finance or health or government. The problem with GDP is that its measurement is fungible. Not too long ago the BEA changed the playing field:
On July 31 [2013], the U.S. Bureau of Economic Analysis will rewrite history on a grand scale by restating the size and composition of the gross domestic product, all the way back to the first year it was recorded, 1929. The biggest change will be the reclassification--nay, the elevation--of research and development. R&D will no longer be treated as a mere expense, like the electricity bill or food for the company cafeteria. It will be categorized on the government's books as an investment, akin to constructing a factory or digging a mine. In another victory for intellectual property, original works of art such as films, music, and books will be treated for the first time as long-lived assets.

The net effect of this, of course, is to further incentivize folks into non-production employment. Not that Joey's going to read the GDP detail before declaring his college major, but the policy wonks (Left and Right) will and are certain to favor such sectors. Gramm-Leach-Bliley came about just for that reason: "we're good at finance, so let's do more!!" Let the Chinese make the stuff. The richest country on the planet in all of human history, but we don't actually make any of the widgets we buy. This makes sense? As the Bloomberg piece says,
Robert Atkinson, president of the Information Technology & Innovation Foundation think tank, criticizes "intangible-capital fundamentalists" who, he says, unreasonably dismiss the contribution to growth of investment in tangible high-tech equipment.

The point, of course, is to reward the non-producing sectors. You know, like financial services which happily "innovated" us into the Great Recession. We really, really need more folks in finance and cartoons.