05 February 2017

I Still Hate Neil Irwin, part the sixth

He's done it again, slurped my mission, but other reporting during the week reflecting the asymptote of progress means he'll be lastly discussed in this missive.

Here and elsewhere I've argued that the future of mobile compute devices is squarely (huh?) on the levelling arc of the asymptote since power available to such devices is limited to what can be packed into a battery of such devices' form factor. The Note 7 is merely the latest example of the results of trying to fool Mother Nature. Well, Nova this week reported on the "Search for the Super Battery", narrated by the NYT's David Pogue. It tickled my ovaries to see one of the battery scientists play with a lego-ish version of the periodic table to demonstrate that lithium is as far as we go. Until someone invents a better atom, of course. Kim Jong-Don's unleashing of innovation by giving away the rest of our GDP to billionaires will certainly accomplish that, of course. One point I had long forgotten, but is vital to understanding the issue was the segment on lithium-as-metal battery and lithium-as-ion battery. The first attempts were the former, but way too dangerous. The latter have become the method for exploiting lithium, but provide about half the energy density of metal based batteries. And can be a tad dangerous, as well. There was some discussion of efforts to find a way to use lithium-as-metal in a safer way, but nothing imminent. In any case, power available to mobile compute devices doesn't jump by anything like an order of magnitude. That pesky asymptote.

Friday morning, Adam Feuerstein linked this essay on possible reduction of FDA's efficacy responsibility with quotes such as
What makes drug development long and expensive is the need to prove, beyond statistical doubt, that your damn drug works. That's not on the FDA -- that's on the underlying biology of the diseases we're trying to crack.
-- Michael Gilman
Anyone can come up with safe snake oil, but, if that becomes our regulatory standard, that's what we're going to get. Will it create a vibrant industry? No, it will obliterate it.
-- Bernard Munos

Again, that pesky asymptote of progress: it's the real world science/engineering that determines whether a new widget really is better, not financial engineering. It was this latter that has driven our economy's worse collapses. Kim Jong-Don is determined to make that happen all over again. And with drugs, we get the added benefit of super expensive snake oil. I guess we're not in Kansas any more.

Speaking of the Devil, Maher's show Friday night, without any acknowledgement in Real Time, offered up another asymptotic moment. For those who don't do Maher, the main body of an episode is Maher sparring with three guests. Most often, it seems to me, is two sorta liberal and one token from the lunatic right. Friday was, sorta kinda, the other way round. In particular was Tomi Lahren, whom I'd never heard of until then. But, then, she was titled as a "TheBlaze" talking head, so that's no surprise. Off and on, the discussion dealt with "regulation" during which Lahren repeatedly demanded that coal miners were being abandoned due to regulatory burden. In other words coal miners, unlike say auto workers facing mass unemployment during the auto "bailout", deserved life-long sinecure. Which is not to say that traditional underground coal mining is an easy life. But it does seem that a goodly percent of the 77,000 in PA, MI, and WI voted for Kim Jong-Don on the assurance that they'd get their jobs back. Right now. But that pesky asymptote, again. First, electric power plants have been converting to gas as fast as they can, just because it's cleaner and cheaper. One might argue that this conversion effort is "burdensome" EPA regulation, but said regulation is just implementing statute. Without such conversions, the USofA must inevitably have cities as toxic as China's today or London in the early 1950s. Take your pick. And, second, coal mining in the East has turned to hill topping, which dis-employs underground miners entirely.
In Kentucky, for example, the number of workers has declined over 60% from 1979 to 2006 (from 47,190 to 17,959 workers). The industry overall lost approximately 10,000 jobs from 1990 to 1997, as MTR and other more mechanized underground mining methods became more widely used. The coal industry asserts that surface mining techniques, such as mountaintop removal, are safer for miners than sending miners underground.
In other words, coal miners have been replaced by machines and methods long before EPA began to clean the air. And, of course, the House added more profit (unpaid negative externalities) to coal companies emboldened by Kim Jong-Don.

Clean coal?

So, after this leisurely stroll down the Yellow Brick Road, we get to Irwin's latest, which discusses whether US labor is finding life better or worse. It's his usual decent essay, but he elides the two most important points.

First, employment dislocation remains analyzed in the century old context of farm-to-factory, and Irwin doesn't get beyond that. Farm-to-factory worked, in the sense of "new jobs replace old jobs", just because farm work was at least as skilled as factory work and abundant, so the segue was easy. The only real difficult part was getting from shit kicker states to urban ones. That context is simply duplicitous these days. Lahren, see above, argues for coal miner sinecure (but Northern auto workers get shafted) just because there's really no other choice. A 50 year old ex-coal miner isn't going to be retrained to be a 55 year old financial analyst; not that we need any more folks skilled at deceptive finance, of course. Having a GED, or less, is one impediment. The other, of course, is that education is the main part of what used to be called "human capital" and as with any capital expenditure, the cost/benefit decision rests on the pay back period. So, coal miners are either treated as specially deserving of sinecure, as opposed to Northern unionized factory workers, or they're not. Fair is fair. Well, it's supposed to be. Income redistribution is the only efficient path.
"When firms invest in technical change that disrupts employment structures," Mr. Johnson said, "their decisions focus solely on private profit and neglect the costly side effects that society must bear. When one small firm adopts a new technology displacing workers, this may not be a societal crisis. When many firms do this at the same time, the changes in the nature of production and employment across the nation become a profound social problem."

Second, neither Irwin nor his interviews admit to the asymptote problem. But they do, implicitly.
[The Economic Innovation Group] cite federal data showing that in 1977, more than 16 percent of firms in the United States were less than a year old, a figure that had fallen to half that by 2014. New businesses have similarly done less to power new jobs than they once did, while the biggest, oldest firms account for a rising share of economic activity. Market concentration increased for two-thirds of industries between 1997 and 2012, the report found. That coincided with a steady rise in corporate profits as a share of gross domestic product, and in a decline in the share going to workers' wages.

I guess that lunatic right meme, "small business drives the country", is just more bullshit. Why am I not surprised?

That nasty asymptote, yet again. New business depends on true innovation, and Gordon's book, if you read carefully, demonstrates that the driving force is that asymptote.

Finally, I'll admit that the notion of the asymptote is old hat among quants. So far as I know, the term is my invention, but the quant has been around for centuries. It's called the Gompertz curve.

No comments: