[lbo-talk] The Financial Times on big data in China

Jim Farmelant farmelantj at juno.com
Tue Sep 5 11:35:37 PDT 2017


https://www.ft.com/content/6250e4ec-8e68-11e7-9084-d0c17942ba93 ---------------------------------------------------------

YESTERDAY by: John Thornhill There are several theories why the Soviet Union collapsed: imperial overstretch, economic inefficiency, ideological bankruptcy. Take your pick or mix all three.

But in his book Homo Deus, the Israeli historian Yuval Noah Harari suggests a more prosaic reason: planned economies (and authoritarian regimes) are rubbish at processing data. “Capitalism won the cold war because distributed data-processing works better than centralised data-processing,” he wrote.

How could any central planner sitting in Gosplan’s offices in Moscow hope to understand all the moving parts of the Soviet economy across 11 time zones?

The lunacy of that system was best explained to me by Otto Latsis, the late Russian economic journalist. He recounted how he once visited a Soviet roof tile manufacturing company, which had big expansion plans. He discovered that all the investment was being sunk into a separate plant to destroy discarded, imperfect tiles to stop them being sold on the black market by the factory’s workers.

As with most Soviet factories, the planners had set demanding targets for the new plant, focusing on quantity rather than quality, creating perverse incentives to smash up near-perfect tiles. It is hard to think of a starker example of the value destruction that wrecked the Soviet economy.

But the explosion of data in our modern world could — at least in theory — inform far better managerial decisions and reduce the information imbalances between a planned and a market economy. Central planners are rapidly acquiring the tools to process data a lot more effectively.

All our connected devices are pumping out oceans of data about our real-time economic activities, demands and desires. Properly harnessed, that data could mimic the price mechanism as a means of matching demand and supply.

For the past few decades, free market economists have been stomping on the grave of the planned economy. Might it rise again in a radically new form?

Some seem to think so in China, which remains wedded to Communist ideology — no matter how loosely applied in practice — and is developing one of the most vibrant data markets in the world.

Last year, Jack Ma, founder of Alibaba, the online platform with some 500m users, argued new technologies provided the means to gather and process data on a near-unimaginable scale.

Applying artificial intelligence to those data sets was deepening our real-time understanding of the world. “As such, Big Data will make the market smarter and make it possible to plan and predict market forces so as to allow us to finally achieve a planned economy,” he told an economic conference.

Some Chinese economists have gone further. In a recent paper Binbin Wang and Xiaoyan Li argue that a hybrid economy could be built on a “market-based, plan-driven” model.

The freer flow of data could counter many of the ills that disfigured planned economies: excessive concentration of power, rent-seeking corruption, and irrational decision-making. The granular detail provided by masses of data could also enable planners to offer consumers more personalised choice.

The authors argue that the online platform monopolies resemble central planning institutions. It would be more “legitimate and rational” for the state to become a “super-monopoly” platform.

Such state-owned platforms could operate like an airport directing market-driven traffic. The airport manages capacity, sets aviation standards, balances the demands of safety, the environment and the movement of goods, and serves the needs of operators, passengers and retailers.

There is no doubt that Big Data can improve the efficiency of managerial systems, both corporate and governmental. It might also lead to more rational resource allocation, whether in terms of planning long-term infrastructure or managing healthcare systems.

But while this hybrid economy may be neat in theory, two big doubts hang over its viability in the real world.

The freer flow of data could counter many of the ills that disfigured planned economies: excessive concentration of power, rent-seeking corruption, and irrational decision-making

The first is that accumulating data and using them in sensible ways are two very different propositions, as shown by any number of failed governmental IT projects around the world.

The second is how innovative such an economy could ever hope to be. It is hard for consumers to signal a data demand for a product that does not yet exist. In some respects, it would be like trying to drive into the future by staring into the rear-view mirror.

As Apple’s co-founder Steve Jobs famously said: “Consumers don’t know what they want until we’ve shown them.”

---------------------------------------------------------

I would note that Oskar Lange was talking about that sort of thing at the very end of his life in 1965. http://econc10.bu.edu/economic_systems/Theory/NonMarx_Socialism/Soc_Contraversy/lange_computer_and_the_market.htm

Lange was thinking in terms of the cumbersome mainframe computers of the 1960's. Any future planned economy is likely to make use of decentralized systems working over the Internet. Right now, big corporations like Walmart already work that way. Market transactions can not only be made online but all the associated phenomena like adjustments of inventories, the setting of prices, etc, can be implemented automatically by computer. Lange queried what would the relative roles of markets versus computers be in a future planned socialist economy, but future technological advances are likely to erode away the distinction between computerized planning and the market IMO.

And here is an oldie by me, where some of this was discussed. https://www.academia.edu/205061/Review_of_David_Laibmans_book_Deep_history_a_study_in_social_evolution_and_human_potential

Some other points:

The conceptualization of markets as information processing systems (which Lange discussed), goes back to Friedrich Hayek's 1945 paper, The Use of Knowledge in Society. http://www.econlib.org/library/Essays/hykKnw1.html

It would be the greatest of ironies, if ideas, going back to von Mises and Hayek, turned out to help provide the basis for a successful planned socialist economy.

Paul Cockshott and Allin Cottrell critiqued Hayek, building upon Lange's arguments. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.24.8724&rep=rep1&type=pdf

Jim Farmelant http://independent.academia.edu/JimFarmelant http://www.foxymath.com Learn or Review Basic Math

____________________________________________________________ After Weeks Of Rumors, Joanna Gaines Comes Clean trecommanews.com http://thirdpartyoffers.juno.com/TGL3141/59aeee93accbe6e930393st01vuc



More information about the lbo-talk mailing list