Good and Plenty

Cosma Shalizi has a book review called "In Soviet Union, Optimization Problem Solves You" on Red Plenty by Francis Spufford. This is not a review of his review. This is a response to his invitation to play.

having said that, I should point out that Shalizi ruins the Yakov Smirnoff reference by using the term "In Soviet Union..." as opposed to the historically correct "In Soviet Russia..." set up for the joke. This is a minor quibble, but the small, crabbed pedant in me insists it be noticed, even though Smirnoff jokes are not particularly funny, and therefore any adulteration of them is really a trivial thing.

I should also note that the faux cyrillic use of the backwards R (which in Russian is pronounced "YA" and sounds nothing like a backwards R) to indicate some kind of Soviet reference also irritates the shit out of me. So thank goodness the publishers resisted the urge to title Spufford's book "Яed Plenty". See what I mean? It adds nothing to the context or content.

In any case, I picked up the book from the socialist book collective just this past Tuesday, and have only gotten through the introduction, but I plan on reviewing it as soon as I am done.

Actually, I got turned on to the book because it is currently being discussed at Crooked Timber, and that's where I saw Shalizi's review. Quickly, the book Red Plenty covers that period of history within the Soviet Union, roughly the mid-1950s through the 1960s when it looked like they just might kick the Western world's ass. That magic time (and Spufford does call his story a fairy tale) is one when

"people believed that the state-owned Soviet economy might genuinely outdo the market, and produce a world of rich communists and envious capitalists.  Specifically, it’s about the last and cleverest version of the idea - central planning via cybernetics". 

Now, Shalizi's attention conservation notice (and a cute device it is, considering he provides them as a warning, but which I, invariably, halfway through one of his essays and hip deep in saw grass and gators, finally realize was not meant as an enticement), can be further condensed to:

"Over 7800 words about optimal planning for a socialist economy and its intersection with computational complexity theory... and uses Red Plenty mostly as a launching point for a tangent".

Shalizi does not disappoint. Vaguely recalling through a blue haze of undergrad dope smoke a day or two learning the minimax method of linear programming, and, yes, some distant thought bauble of actually solving simultaneous equations, I was able to skim through the denser sections. And his tangent (forget physically possible, is it computationally possible to optimize a centrally planned economy as large as the Soviet Union's using the crude vacuum tube and occasionally transistorized clunky old room-sized computers of the day? Short answer: probably not, or rather, not in a timely fashion.

But his tangent provoked my own tangents.

He points out that any system usually has a goal, some objective function the powers that be work on optimizing. What would that objective function look like? What would be the goal? Perhaps to maximize the potential of every individual within the society? It is what a eusocial system would do, and paradoxically, it would probably involve maximizing the degree of freedom allowed each individual, or then again, maybe not.
I'll avoid the whole circus geek trap of debating whether a clever use of physical sampling and observation, statistics, human computers (back when the term referred to a profession instead of a machine), electronic calculating machines, and some form of hierarchical administrative calisthenics could not have worked out to produce an optimal Five Year Plan in a month or six. Although the idea, the visuals, of what type of 1950s style dieselpunk electromechanical sensory equipment required just to get a whole society's worth of useful metrics sounds interesting (and conjures up pictures of Terry Gilliam's Brazil, with copper wires and ferromagnets, pneumatic tube delivery systems, and ducts, lots of ducts). I doubt very much, though, that the New Soviet man, or woman, would sit still for having their poo weighed and inventoried by protein content, or their sexual proclivities monitored and categorized, much as is done nowadays voluntarily through smartphones, and drones, and surveillance cameras, and Apple, and Facebook, Twitter, and Tumbler, and... who needs 1984, when you got Web 2.0? Why, who needs RFID tags, micro- and millimachines latched onto your clothes and skin, when we've got your credit card buying history? And banks and banks of Cray XK6 supercomputers to process it all? In 21st century America, machines milk you!

In any case, it was nice that Shalizi, being a smart young kid, doesn't write off central planning entirely, but realizes you need the proper tool for the job.

Speaking of central planning, and not that I cheer it's every incarnation, but I do feel it gets a bum rap a lot of the times. One of the things that annoys me is that stupid, unthinking, thoughtless mantra "Central planning doesn't work!" that religious types like libertarians use. We should all call bullshit on it. We should all tell such religious types to fuck off. Okay, actually, what really annoys me are religious types like libertarians, but in any case...

Clearly, central planning - when the goal is well-defined, set, and limited, and manageable - works like a fucking champ. I can find plenty of examples where it works in a spectacularly successful manner.

Don't believe me? Apollo Moon Project? The Japanese economy,  post-WWII? How about the command economy of the United States of America during WWII? You really thought it was free enterprise what whupped Hitler and Tojo? Nah. We didn't have time for market inefficiencies. There was a war on.
There is no such thing as a 1942 model of an American motor car. Why? In January 1942, the Office of Production Management banned the sale of new cars and trucks in favor of tank, airplane, and other munitions productions. There was a whole alphabet soup of regulatory agencies that cropped up during the war years, which pretty much had authoritarian control over every switch and lever of the economy. The Office of Price Administration introduced freezes on commercial, farm, and commodities prices. Widespread rationing instituted, wages and rents controlled, taxation used to control salaries, and consumer credit tightly monitored and reduced. War profiteering was aggressively combated.

Indeed, only the fact that most industries, financial institutions, and resources were not nationalized kept the United States identified as a capitalist state, as opposed to that other dreaded C-word. Of course, the success of the war-time planned economy can only be compared to what Shalizi would call the "objective fucntion", the goal of the acitivity. Objective function for US economy from 1941- 1945? Mass produce weapons, materiel, and warriors. Some 40% of GNP from 1942 -1945 were war-related outputs.


Of course, this type of obejctive function is not conducive to a general prosperity. One look no further than North Korea. Or the Soviet military buildup of the 1970s, when Breshnev and his generals convinced themselves that the US was seeking an overwhelming first-strike capability.

I find it frankly amazing how you much a regime can get away with skeletonizing an economy before people complain and do something about it. I mean, you got people eating roots and digging up corpses in North Korea while their soft, fat, well-fed psychopath of a leader builds useless nukes and intercontinental ballistic missiles. Clearly, this type of objective function is something we could all do without.

But then, I also have to wonder what was the objective function of the free market economy of the US, circa 1870-1900? (And please, yes, don't fall into the trap of platonic derangement. Real world examples of ideological systems are the real deal. You will never get closer to ideal than real, so don't bother with the "But the Soviet Union wasn't a real communist state", or "but the US has never enjoyed a real free market economy" whining bullshit on me. It will not fly. So shut the fuck up).

What was the objective function of that 1870s-1900 Gilded Age economy, that system of system of graft, corruption, kick-backs, usery, debt peonage, theft, piracy, favoritism, all embedded within a cradle of government largesse, which is the default form of free market capitalism, and any and all other forms are merely aberrations thereof? Oh, did I go off on a tangent?

Fine. What was the objective function of the Gilded Age economy? Full employment? Maximize potential of citizens? Oh come on, you know the answer. Them that gots, got more. Simple as that. So, the question I've got is, is there a better way, a more decent world, a world that everyone can, maybe not enjoy, but at least find tolerable? Is there some combination of methods that can work out so that everything is a positive sum game? Shalizi wonders the same thing.

Hmm. The right combination of methods? Sounds like a job for linear programming!