Hacker News Reader

Comment by thehardsphere | Hacker News Reader

Sorry to burst your bubble, but this idea is something that economists have already discussed to death: google "the Socialist Calculation Debate". A very brief summary of its conclusions is: central planning can only perform as well as or better than the market when the economy is completely stagnant.

Also, "centrally planned economy managed by the government" by definition leaves no space for "an open market" for "more productive or talented citizens" to "choose to earn more wealth and purchase from." Unless you're talking about a black market.

View on HN

The "economists have settled this" refrain has become common on hn lately. It is unfortunate.

The reason that conflicting schools of economic can coexist indefinitely in economics (in the contrast to physics, say) is that very little is settled on economics since the question of how an economic system works is predicated

As other have said, your absolute pronouncements about socialism requiring a stagnant economy is coming from a rather specific ideological bent rather than a simple consensus.

It comes from a consensus of people who are skeptical of central planning, which spans more than a single "specific ideological bent" and should be the starting point in any given discussion of the topic since over the decades the planners have produced nothing but failure.

Frankly, the burden of proof ought to be on you to show that it's going to work now just because we have shiny new computers that they didn't have between 1922 to 1991. Which is actually very thoughtfully discussed in the article.

Decades of market failures (and economists claiming that this time they got it right) should also be taken into account.

When you can point out a "market failure" that resulted in famines that killed millions of people, maybe we'll have something on the same scale to account for.

Why can planned and market economy not coexist? Some sectors of the economy are or have been state planned in most countries. Come to think of it, in a slightly more complicated planning model, the cost and benefit of nationalising or privatising certain industries could also be included.

If you're really centrally planning the economy, then all of the decisions are made by a central authority. There's no room for anyone else to do anything, because it messes with the program. The central plan can only work of it knows everything relevant.

If you're running a mixed economy, where some sectors are planned and some are not, thats fine... but it's also not what the post is taking about.

Sure they coexist now, because, as sibling commenter pointed out, while outside of corporations market forces exist, every corporation or government entity itself is a separate world, where most of free market rules do not apply. Most of them are centrally planned in one way or another

Ok, I think we need to define what "central planning" means for people who are going to choose to participate in this discussion.

"Central planning" means "the government (or some other single entity, but usually the government) makes all of the decisions in the economy."

Corporations, despite being "centralized," and despite engaging in "planning," do not meet the definition of what we're talking about when we say "central planning." They emerge in a free market economy in order to reduce the transaction costs that would come with trying to coordinate mulitiple people as independent participants to do some specific activity. They do not exist to decide for the entire country exactly what everyone will be doing all the time to meet the entire nation's needs. That's a very big difference.

>A very brief summary of its conclusions is: central planning can only perform as well as or better than the market when the economy is completely stagnant.

I'd hardly say that was either proven or agreed upon

>"centrally planned economy managed by the government" by definition leaves no space for "an open market" for "more productive or talented citizens" to "choose to earn more wealth and purchase from."

I don't see this a being "by definition" either. You can have a centrally planned economy with people who are independent (or at least as independent as they are in any other economic system) in what job they choose, where they choose to work and what they choose to buy. The government could assess the demand for a product at a given price, determine the resources (material or human) necessary to produce that product as well as all other demands for those resources. When the demands exceed resource availability, you then raise the price for that resource, refactor that into the price of the final product and then assess the demand for it at that new price. You raise the price for each constraining resource until the demand drops to a level that can be fulfilled. This would include specialized labor, so if a job is in high demand and short supply, the wage for that job would go up until the supply increases and/or demand decreases to reach equilibrium.

All of this already happens in an ad-hoc way in a free market, but in a free market it can take a long time for changes in input prices to be reflected in output prices, and it can take even longer to determine how those will affect demand. I don't see why that couldn't be formalized and worked out centrally. I'm imagining an economy where everyone is issued a smartphone through which they make all of their purchases and through which they find and respond to bids for employment. Demand could then be assessed fairly accurately and instantaneously. It seems like it would be way more efficient in terms of reductions in redundancy, overproduction, costs incurred attempting to generate demand that does not exist or siphon demand from competitors' products, and price increases solely for profit-maximization in cases where demand is inelastic and supplies are sufficient (like epi-pens or insulin).

Modeling it would probably be extremely complex and might still exceed our current computational capacity, I dunno, but I don't see any reason to call it impossible.

> The government could assess the demand for a product at a given price, determine the resources (material or human) necessary to produce that product as well as all other demands for those resources.

As the blog post notes, politicians will not leave their hands off sensitive prices, leading to unending systemic distortions. This already exists.

> All of this already happens in an ad-hoc way in a free market, but in a free market it can take a long time for changes in input prices to be reflected in output prices, and it can take even longer to determine how those will affect demand.

Market reactions vary according to the market. Petrol prices at the pump move up extremely quickly, because the station owner is going to pay for the next tank out of this tank. They move down slowly because that's about the only time they turn a profit.

Meanwhile, manufacturing industries can need 1-5 years of lead time to increase capacity. They can't and won't make massive changes without substantial, sustained market signals that it's required.

The difference between the market system -- really, the price system -- and non-market systems is that the price system is sufficient to solve the planning problem. A world of agents who know nothing but prices is still able to allocate resources and production in a way that adapts to changes in demand and supply, including large shocks.

In practice, real markets also make use of other information signals. But they do not require them. And they do not require a central coordinating planner to have perfect or near-perfect information of extremely wide ranges of local information.

Probably the best explanation of this distinction is found in Hayek.

With regards to the innovation question: considering that the soviets we're quite good at military and space innovations, and that defense organizations in general produce a lot of innovation(think DARPA), maybe central planning could work ?

Though of course the required central planning here is quite different, more about smart bets than optimizations, and focusing engineering resources at important problems, which doesn't seem that great at capitalistic economies either(think planned obsolesce, "my generation is optimizing ads", browser in-incompatibilities, etc) .

The Soviets were really good at military and space innovations but:

1. They weren't much better than the West. Consider that the capitalist world surpassed them at pretty much anything they started with an advantage in.

2. They were also really bad at managing the scarcity of very basic things their people needed. Like, food and medicine. The joke was that Russia had "bad weather" for 70 years under communism.

Soviets where good at military and space innovation because there was competition with the West.

There was no competition on their domestic market, hence it sucked.

>With regards to the innovation question: considering that the soviets we're quite good at military and space innovations

Well, space industry was quite suboptimal in USSR. Successes were achieved because of 1) all resources were spent on space and military and 2) great legacy of Russian Empire (especially great minds like Korolev, Tsiolkovsky and Tupolev). So I see it more like stagnation than development.

Maybe central planning works better on big items than it does on little ones. Gorbachev decided that central planning wasn't working when he realized that the USSR could produce military hardware as well as the US, but couldn't supply toothpaste to its population.

Spufford's book mentions this explicitly: during the description of one of the Central Planning Sessions one of the characters muses (i.e. provides an overview to the reader) about the many unwritten rules that guide each of the calculation steps. One (possibly the most important) is that "The Army always gets first pick to anything".

So the Objective Function assigned much higher "value" to Military production (either explicitly or implicitly) and this of course made military production more "efficient" than consumer goods production. (I put quotes around "efficient" because there still was a lot of waste and corruption even in that, of course).

>but this idea is something that economists have already discussed to death: google "the Socialist Calculation Debate"

I wouldn't say it's been "discussed to death" at all; have you read Cockshott's "Towards a New Socialism"? He tries to explain why the problem isn't really a problem at all, with his idea of computerised central planning.

No I haven't read that book, but I have read and heard a lot of people claim that better computers would solve the problems with planning. I've never heard anyone who does so actually make a persuasive case, because they tend to treat the problems as merely being about doing spare matrix math or simply needing more data points to get better answers, which are not sufficient to make the problems of central planning go away.

I will google this book, but could you tell me what you find most persuasive about it?

> google "the Socialist Calculation Debate"

I can't seem to find the document you're referring to and its conclusions, Wikipedia page? Is there a balanced discussion on the subject somewhere that is not from Mises Institute or some other libertarian ideologically-driven piece?

https://en.m.wikipedia.org/wiki/Socialist_calculation_debate

Has references to many of the original essays by Mises and Hayek that brought the problem to broader attention in economics.

That's why I asked about something non-ideological, and neither Mises nor Hayek are that.

Found this though, which brings up good counter-points to libertarian criticism of central planning: https://en.wikipedia.org/wiki/Economic_calculation_problem#C...

> That's why I asked about something non-ideological, and neither Mises nor Hayek are that.

You may have mistaken Mises and Hayek, the fairly thoughtful and insightful market-favouring economists, with Mises and Hayek, the caricatures transmitted via a heavily mutating Murray Rothbard into US Arch-Libertarianism.

In particular Hayek is at great pains that he doesn't think the market is magical or morally superior. He clearly points out that markets assign by prices, not by commendable virtues.

But he does say that markets solve a problem that can't, beyond a certain scale, be safely solved in other ways without causing larger problems to emerge.

Hayek was a Nobel prize winning economist who along with Mises brought this whole issue to the attention of the field of economics. If you want more 'non ideological' than that in economics you're not gonna find it. By the definition you seem to be using a non ideological economist doesn't exist.

I don't see how winning a Nobel prize makes someone less ideological, especially if they're already widely known as particularly ideological.

For example, Soviet economist Kantorovich won the same prize the year following Hayek for work related to central planning, and I'd consider him a lot less ideological, not because of winning the prize, but because of his overall approach to issues: http://www.nobelprize.org/nobel_prizes/economic-sciences/lau...

My point was that if you want a non ideological economist you're going to have a hard time finding one and that Hayek is as much an economist and no more 'ideological' than any other influential economist: from Hayek's contemporary Keynes through Samuelson, Friedman to Krugman they're all indisputably significant figures in economics and Hayek is no more ideological than the others. If you're going to dismiss Hayek as ideological you have to pretty much dismiss economics as a discipline entirely.

Winning a Nobel prize in itself doesn't automatically make a person non-ideological. As others have pointed out, Mises and Hayek have faced (justifiable) criticism for being too ideological, which you can't wave away by simply stating that they're Nobel laureates.

It's worth pointing out, for those of you who are playing the "oh it's libertarian so it must be bad" game, that other socialists also reached the same conclusions about central planning. Central planning is not an explicitly required feature of socialist economics, and critique of central planning is not exclusively the province of the right. Trotsky, Kropotkin, and the Mensheviks would be examples of prominent left wing thinkers who rejected central planning, especially the Soviet style.

You want "non-ideological" yet you also look for "counter-points to libertarian criticism."

I meant something that at least acknowledges the existence of arguments outside of Austrian school, let alone capitalist economics, and then analyzes merits of both/all.

Mises and Hayek were not exactly famous for being unbiased/non-partisans about these problems, though.

Nor methodologically sound, either.

Reply on news.ycombinator.com