Hi guys,
I have read Mises' book Socialism and I think I have a decent understanding of the Economic Calculation problem, but I am not an economist by trade. I am a computer geek learning economics. A friend (a Socialist) has posted a lengthy response to my request that he read Mises' paper. I can take his paper apart pretty easily, and I'll point out some things I've noticed The thing is, he has anarcho-socialist leanings (and very anti-Mises, thinking it's all propaganda), and I really want to show him where he has gone wrong and I can't explain it very well. Rather than send a half ass reply, I thought I would ask for help.
First, his review of Mises' paper:
============================================================================
A major misconception by the free-market fundamentalists is that central planning is mathematically impossible. So they consider the Soviet example, and imagine that if only they look at Gosplan's "source code" hard enough, they will inevitably discover some bug. So they write esays like this (http://www.mises.org/pdf/econcalc.pdf), failing to point to any real bugs, but making general claims that the source code must be full of them. Theis basic argument is this: programming is hard, and they are trying to write programs, so their code must be buggy and therefore their system must crash. And when the system does crash, they are quick to claim victory, and the fact that the crash can actualy be attributed to a fire in the server room never bothers them.
However, if you look for bugs in the right place, you will quickly see that actually it is Mises's paper that needs debugging. Because in its current form, it just can't withstand any serious criticism. Let us see, for example, how Mises' analysis falls short already in the very beginning of his "proof".
In Part 1, Mises makes a series of pretty obvious observations, pointing out, correctly, the importance of money in a socialist economy. (It bears mentioning that, for some reason, many Americans imagine that a Soviet-like system rejects the use money - this is patently untrue, even if you listen to Mises). Here, however, Mises takes his first shot at finding some fatal error in Socialism. While grudgingly admitting the possibility of assigning value to consumption goods, he chooses to attack the idea of appraising the value of production goods. To quote Mises, "Money could never fill in a socialist state the role it fills in a competitive society in determining the value of production goods. Calculation in terms of money will here be impossible."
Let us pose here and think about this argument. On a first glance it looks that Mises actually has a point. If there is no free market for production goods, then how do you value, say, a tractor vs. a combine? In a capitalist economy companies will decide how much they are willing to pay for either item, but how will government bureauctats arrive at the correct valuation? Mission impossible, huh?
Well, not so simple. First of all, note that even under capitalism, there is no such thing as "objective valuation" of production goods. Upon a little thought we can immediately identify two possible prices for any piece of equipment. One price is determined by the labor theory of value: it is the number of man-hours required to produce the tractor or the combine. A capitalist would call it "self-cost" or "operating expence". Provided the market is highly competitive and there are no major production bottlenecks, the market price of the production good will approximate its self-cost. The other price is determined by the "marginal utility". For example, the extra amount of grain that can be harvested by the most efficient farmer if he purchases an extra tractor, will give you the tractor's marginal utility. If the market is monopolized by John Deere or if there are some production bottlenecks (e.g. shortage of iron), then the market price will approach the marginal utility of the production good. In a real economy the price will fluctuate between these two extremes.
The United States economy is a mixture of government-created monopolies and competing private businesses. As a result, the price of one and the same production good can vary enormously depending on the degree of government involvement. By and large, however, whenever the market is allowed to operate more or less freely, the price of a tractor is much closer to the self-cost than to the maximum price the highest bidder would pay. But this immediately destroys Mises' argument against socialism. For if the "correct" price of a tractor is equal to its self-cost plus a few percent profit, then a central planner who assigns prices based on the labor theory of value, will never be very far off the mark. Using different processes, the capitalist and the socialist have arrived to rougly the same valuation.
Mises's argument will still be valid if the productivity of labor in a given country is too low to produce the tractor as cheaply as John Deere does in America. Let's say, the country lacks automated production lines, and as a result each tractor produced locally costs a million dollars. This, presumably, is but a temporary condition. Eventually technological bottlenecks will be removed, but while resources are still limited, it is essential that the country makes the best use of what's currently available. So maybe we can say that Mises' objection is valid at least for undeveloped countries where every tractor is as valuable as gold?
Alas, here again Mises misses his target. Any honest Marxist would explain to Mises that the necessity of achieving high productivity of labor is recognized by Marxism as an essential prerequisite for Socialism. One cannot start building Socialism in a feudal economy. That's why Marx, who was well aware of this difficulty, has always insisted that capitalism must be allowed to run its course before one even begins to think of socializing production. The Marxist theory was intended for modern, technologically advanced economies; it is dishonest to critisize it for its failures outside its realm of applicability.
However, Mises still has one objection. If capital goods are not really in such short supply that they couldn't be efficiently distributed without the help of some marginal utility mechanism, then maybe some shortage of natural resources will still save the "proof"? Here is the way he puts it: "For, over and above the actual labor, the production of all economic goods entails also the cost of materials. An article in which more raw material is used can never be reckoned of equal value with one in which less is used."
This point merits a separate discussion. Above all, we must here must a clear distinction between an objective shortage and poor availability of row materials. Needless to say, to the ultra-capitalist Mises, it appears that the market price of a commodity is sacred and beyond dispute. But the reality is not very Mises-friendly. In most cases, what free-market fundamentalists describe as "scarcity" is just another word for a high self-cost. There is no physical shortage of oil on Earth. There is enough oil to last us a thousand years, but some of that oil will require many man-hours to extract. A $100 oil simply means it takes too many workers to assemble the rig and to drill the well. In these cases, the labor theory of value will work every bit as well as it works for any consumption good. Thus the iron ore that is needed to produce a tractor, and the oil that is needed to produce the iron ore will all be assigned a value in terms of man-hours, which will then appear in the socialist price of the tractor. Again, this method of valuation will work as well as any capitalist one.
If, on the other hand, the nature of the constraint is a physical one (a river that only has so much water, for instance), then a socialist planner will have to decide on its optimal use without the labor theory or the marginal utility theory to guide him. Actually, the latter point is not really true. A planner can still estimate the marginal utility by running a calculation along these lines: "if I give the resource to factory A, they can produce this quantity of consumption goods, and factory B could only produce half that much, but factory C could produce 20% more than factory A. So factory C gets the resource". No calculation is perfect. The planner will not find the best solution, but he can at least take a shot at finding the second-best one.
It looks like Mises has finally got something to chew on. Maybe if we zero in on this (admittedly unlikely) scenario, we can "prove" the inefficiency of socialism at least on that one count?
Alas, it will be like the pot calling the kettle black. When you ask a free-market fundamentalist how his beloved capitalism will handle this situation, you wind up with a solution that is much worse than any inefficiencies in the socialist model. Turns out that under capitalism, the water right will belong to a private monopoly which will then be selling its water to the highest bidder - here you got your marginal utility theory in action. Eventually one company will outbid every competitor and obtain the resource. The diference is that where a socialist company could use the resource right away, its successful capitalist counterpart has emerged from the bidding process totally exhaused and saddled with a huge debt which will then have to be passed to the consumers. The socialist company that won the contest may have been 5% less efficient; in contrast, under capitalism you've got a company that is fully efficient and is squezed dry by the private owner of the limited resource. What a sure way to outpace a centrally planned economy!
I couldn't take apart the whole essay, but I hope this already illustrates my point that one should always take it with a grain of salt when he hears that this or that essayist has "proved" some economic theory. More often than not it just means that the guy has tried to bring up some point or raise some objection to promote his chosen point of view. Then, under close scrutiny, it will all turn out to be very subjective...
First off, I've never heard of any Mises supporter citing Gosplan's source code as the problem. That's just stupid. I'm a programmer, and the simple reason complex models for Socialist programming don't work is because humans have to assemble the data. It's not the code or the act of inputting the data, it's the data itself that is the problem. Where does it come from? Who determines the parameters for the programs? Etc. Everything comes from the planning board, and where do they get the data? We are right back at the EC problem.
Next, he seems to be unaware of Lenin's attempts to destroy money through massive debasement in the first few years of Bolshevism. Anybody got a source for early USSR monetary policy?
When you get into the crux of the argument, he has the entire pricing system backward. He doesn't understand the role of the consumer in the pricing process or the role of the entrepreneur in the innovation process.
Much of the rest is just ramblings. However, I want to give him a fair and honest critique and present him with a couple of questions to get him thinking. Any help would be appreciated.
Thanks,
David
As with any debate regarding the calculation debate I suggest reading this:
http://mises.org/journals/qjae/pdf/qjae10_4_1.pdf
He sets up his theory on purely essentialist grounds from Rothabardian monopoly theory. All systems are planned the question is what type of planning: competitive or non-competitive.
The atoms tell the atoms so, for I never was or will but atoms forevermore be.
Yours sincerely,
Physiocrat
Skimread it. He really has no idea what he's talking about (e.g. he repeatedly uses "marginal utility" as a term - does he have any idea what this notion refers to?), yet like many socialist brats he still has an arrogant, dismissive tone to his writings. Perhaps he should shut it and read up before posting such off-the mark ramblings? Maybe give him a primer in economics so he can dispel his confusion and actually understand what Mises is saying? I might go through it a bit later when I've got more time, unless someone else beats me to it, but there's so much to set straight...
Freedom of markets is positively correlated with the degree of evolution in any society...
Search for War Communism.
He doesn't have an idea of that marginal utility is. The price of a good is always equal to its marginal utility, not only in a single-seller condition or bottleneck. With private property, no producer can decide "freely" the price of the goods (or services) he offers, because his decision is constrained by other "propietors" (capitalists, landowners, workers, etc.). In socialism, every enterprise is a monopoly, at least in the sense that they are not accountable to "profit and loses". They will just operate, and the prices they assign are purely arbitrary. When your friend says that in a very competitive environment the price of a good approximates its "labour cost", he is missing the mark. There is not such think as "labour cost", as any good, is, in essence, determined by the cost of other goods, all interconnected in a market economy. It's impossible to assign an objective value to anything, despite our desire to do it. The idea of socialism is all rooted in this wrong concept, the labour theory of value.
He definitely can't let go of Labor Theory of Value, so I will look up some focused critiques on this subject.
I'll also try to help him out on Marginal Utility.
Also, I noticed that his argument was very Walrasian, as if there is some general equilibrium of prices - just a few pennies of profit on every step of production . I'll introduce him to the rebuttals of that I found in Rothbard's End of Socialism paper.
He's in way over his head, but I don't want to degrade him. I think I can show him "the light" and if not, I'll learn from it too!
To combat the belief that production can be mathematically planned according to centrally planned formulae, I find Hayek's version of the calculation problem to be far more efficient, "The Use of Knowledge in Society" is a good article by Hayek about how knowledge is utilized in society.
Abstract liberty, like other mere abstractions, is not to be found.
- Edmund Burke
Fellow computer geek/wannabe economist here
I'd say you're right on with your criticism of his programming metaphor. Keep in mind that not all socialists worship Lenin, and many despise him. Many socialists do support the use of money, so pointing out that Lenin didn't really isn't very productive.
I'd say the major flaw with his argument is his use of the labor theory of value. If you spend all day making apple pies and I spend all day making mud pies, are our products equally valuable? If you make ten apple pies and I only manage to make five, are mine twice as valuable? Who would pay twice as much for an apple pie that was produced less efficiently but is otherwise indistinguishable from the cheaper pies? Value is subjective.
First of all, note that even under capitalism, there is no such thing as "objective valuation" of production goods. Upon a little thought we can immediately identify two possible prices for any piece of equipment. One price is determined by the labor theory of value: it is the number of man-hours required to produce the tractor or the combine. A capitalist would call it "self-cost" or "operating expence". Provided the market is highly competitive and there are no major production bottlenecks, the market price of the production good will approximate its self-cost.
The value of the labor is determined by the worker's marginal productivity. The market price takes into account other factors, such as materials and transportation costs.
For if the "correct" price of a tractor is equal to its self-cost plus a few percent profit, then a central planner who assigns prices based on the labor theory of value, will never be very far off the mark.
The "correct" price of the tractor is whatever the market will bear, i.e. whatever someone is willing to offer in exchange for the tractor. If all that could be had for the tractor was the labor put into making it plus a few percent, the tractor would likely not be made as this fails to cover the costs of capital and its upkeep, management costs, the risk taken by anticipating future demands, etc.
In most cases, what free-market fundamentalists describe as "scarcity" is just another word for a high self-cost. There is no physical shortage of oil on Earth. There is enough oil to last us a thousand years, but some of that oil will require many man-hours to extract. A $100 oil simply means it takes too many workers to assemble the rig and to drill the well.
I don't think he understands the concept of scarcity as used by economists. A good that is available in high quantity is still scarce. In addition, oil rigs were not created equal. More expensive, technologically advanced oil rigs and refineries are often needed in addition to the labor.
If, on the other hand, the nature of the constraint is a physical one (a river that only has so much water, for instance), then a socialist planner will have to decide on its optimal use without the labor theory or the marginal utility theory to guide him. Actually, the latter point is not really true. A planner can still estimate the marginal utility by running a calculation along these lines: "if I give the resource to factory A, they can produce this quantity of consumption goods, and factory B could only produce half that much, but factory C could produce 20% more than factory A. So factory C gets the resource". No calculation is perfect. The planner will not find the best solution, but he can at least take a shot at finding the second-best one. It looks like Mises has finally got something to chew on. Maybe if we zero in on this (admittedly unlikely) scenario, we can "prove" the inefficiency of socialism at least on that one count?
Why is it unlikely for a good to have several uses, some more efficient than others? Also, this might work if the central planner is dealing with factories that are all producing a single product, but how is the planner supposed to decide which use is most efficient for making different products? This cannot be done efficiently without knowing the desires of the intended consumers--knowledge that the market provides.
Assuming property rights could be construed as to give a single company rights to all the water flowing through a certain river, there are still other rivers so this would not be a monopoly. The water rights need not be sold entirely to a single company either. I could quite easily sell x gallons per day to one buyer, y gallons a day to another, etc. Even if somebody obtains a monopoly on river water, the high prices fetched by water would provide an incentive to use alternative sources of water (purification of rainwater, desalinization of ocean water, etc) which would drive prices back down. The assertion that somebody who outbids another is too exhausted to immediately use the resource is ridiculous. When I buy something off of ebay, I am capable of using it as soon as I can get my hands on it. There's nothing about price negotiations that prevents this. The "huge debt" that must be passed on to consumers reflects the costs of providing any good, regardless of the method of allocating resources. Socialists like to pretend that the need to produce wealth in order to receive wealth is a uniquely capitalist invention. Don't let them.
Edit: making nonsensical claims due to lack of coffee.
Physiocrat: As with any debate regarding the calculation debate I suggest reading this: http://mises.org/journals/qjae/pdf/qjae10_4_1.pdf He sets up his theory on purely essentialist grounds from Rothabardian monopoly theory. All systems are planned the question is what type of planning: competitive or non-competitive.
That's one excellent piece of reading there. Thanks very much!
Actually, Physiocrat, whilst that article is pretty damn good, I'm not sure it doesn't leave out the possibility of socialist systems that don't organise under one single central planner. Could they not argue that if society was organised into a number of self contained collectives, each run internally along socialist lines with a single owner of it's means of production but externally they traded, not just consumer goods but factors of production too. Does that not sufficiently provide exchange so that prices for the factors can be worked out? Could such a model be the way forward for socialists in the face of the "Calculation Problem"?
In other words, does the "Calculation Problem" only apply when there is one single owner/planner. If society devolves into at least two collectives, there will be exchange and therefore meaningful prices for the factors of production?
Fried Egg: Actually, Physiocrat, whilst that article is pretty damn good, I'm not sure it doesn't leave out the possibility of socialist systems that don't organise under one single central planner. Could they not argue that if society was organised into a number of self contained collectives, each run internally along socialist lines with a single owner of it's means of production but externally they traded, not just consumer goods but factors of production too. Does that not sufficiently provide exchange so that prices for the factors can be worked out? Could such a model be the way forward for socialists in the face of the "Calculation Problem"? In other words, does the "Calculation Problem" only apply when there is one single owner/planner. If society devolves into at least two collectives, there will be exchange and therefore meaningful prices for the factors of production?
From what I remember from article, I haven't read it for a while, the calculation problem to the extent of the absence of private property. Crusoe is on his island and owns everything; he imputes values on his property and can assign prices to them. This is arbitrary of course but so are all values and prices yet since there is freedom of entry into the market his prices are competitive and are not deficient a priori. Now supposing the whole world was Crusoe's island no calculation problem arises. The problem is not with a single owner but of an absence of competition in pricing. So a world state would reek havoc since there would be no competition whatsoever in pricing however an individual who privately owns the whole world does not cause a problem a priori- his pricing would be rational.
So in fact is clouds the issue to call it a calculation problem but a competition/ property problem: Legally restricting the competition in pricing causes systematic mistakes whereas in competitive pricing mistakes are only incidental.
Physiocrat, I don't think you have addressed my question.
Imagine a socialist society organised not under one single planning authority, but into ten different planning authorities. Each planning authority took under it's control one tenth of the available capital and populations to run as collectives. If each collective in this community could buy and sell factors of production from/to other collectives in that society, there would exist a competative market for the factors of production, one in which meaningful prices would emerge and on which each collective could use internally to plan it's production.
The arguments in the article you supplied apply only in the case where there is one single central planning authority and therefore no competition of ideas as to how best to use the factors of production. Could this not be solved by breaking down socialist society into a collection of competing collectives? Could they not, in this way, abolish private ownership of the means of production without falling foul of the calculation problem?
Normal 0 false false false EN-US X-NONE X-NONE
dvdburns:Here, however, Mises takes his first shot at finding some fatal error in Socialism. While grudgingly admitting the possibility of assigning value to consumption goods, he chooses to attack the idea of appraising the value of production goods. To quote Mises, "Money could never fill in a socialist state the role it fills in a competitive society in determining the value of production goods. Calculation in terms of money will here be impossible."
Well, what system is your friend talking about? Is he talking about a soviet type system? One where the government completely owns the means of production? Where it produces every product from start to finish? Or a mixed type of system where there are markets and exchange is somewhat permitted, though heavily regulated? If his system is the complete collective ownership of the means of production then there is indeed no use for money. If the government produces all goods, and merely redistributes these goods based on the amount of labor undertaken by an individual, then what is the purpose of exchange? How could there be arbitrage? If it's the latter, then the works of Hayek, Mises, Rothbard, and others provide the explanation; namely that prices will not accurately represent consumer demand, leading to shortages or surpluses, overproduction and underproduction. Where the free exchange of goods and services is entirely unknown, there is no use for money whatsoever.
dvdburns:One price is determined by the labor theory of value: it is the number of man-hours required to produce the tractor or the combine. A capitalist would call it "self-cost" or "operating expence". Provided the market is highly competitive and there are no major production bottlenecks, the market price of the production good will approximate its self-cost. The other price is determined by the "marginal utility".
He's completely contradicting himself here; again, which one does he plan on using? If he's using the labor theory of value, then what will determine the value of labor? I understand that the value of commodities will be determined by the value of labor; but what determines the value of labor? How do you have a "highly competitive" market if the government controls everything and doesn't compete at all? If he's using marginal utility, then individuals determine their own subjective valuations of every single commodity/service. Again, which system is your friend talking about?
dvdburns:By and large, however, whenever the market is allowed to operate more or less freely, the price of a tractor is much closer to the self-cost
Huh? Tell that to hummer producers when the price of gas is $10 a gallon.
dvdburns:Alas, here again Mises misses his target. Any honest Marxist would explain to Mises that the necessity of achieving high productivity of labor is recognized by Marxism as an essential prerequisite for Socialism. One cannot start building Socialism in a feudal economy. That's why Marx, who was well aware of this difficulty, has always insisted that capitalism must be allowed to run its course before one even begins to think of socializing production.
Interesting. Someone should tell your friend that the USSR tried to extend the structure of production and make it more capital intensive in order to increase the productivity of the workers. Unfortunately, they over-extended the structure of production, and didn't produce enough current goods to satisfy current consumption demand; the result? Millions of people starved to death. But this is the inevitable result of central-planning: without money time-preferences cannot be expressed, that is, the rate of interest, and thus, the required level of capital investment can never be known.
dvdburns:This point merits a separate discussion. Above all, we must here must a clear distinction between an objective shortage and poor availability of row materials. Needless to say, to the ultra-capitalist Mises, it appears that the market price of a commodity is sacred and beyond dispute. But the reality is not very Mises-friendly. In most cases, what free-market fundamentalists describe as "scarcity" is just another word for a high self-cost.
There is one Mona Lisa, the cost to produce this piece of art was very low, and yet it is very expensive; why? Because marginal utility is right, and the LTV is wrong. Your friend could arbitrarily assign x labor dollars to the skill of the artist in creating this painting; but how did he determine this figure? Furthermore, how does your friend know that there is enough oil to last us for a thousand years? That's news to me! Your friend is trying to deny scarcity, which is a fool’s errand.
Your friend really needs to make up his mind and decide which system he's talking about, and which method of value will be used. If he's going to have a collective ownership of the means of production and use the labor-theory of value, then he needs to explain how he will determine or differentiate between different skill levels of labor. What will be the distribution to a doctor, and what to that of a coalminer? If he pays them differently than a class system is inevitable, and what then would be the point of socialism? Who gets to be the doctor, and who gets to be the coalminer? What if no one wants to be the coalminer?
If he's going to have markets, and competition, while applying marginal utility as the measure of value (STV), then you don't have socialism; you have Sweden, the United States, the United Kingdom, ect, ect. Your friend hasn't addressed any of Mises' points whatsoever.
"If we wish to preserve a free society, it is essential that we recognize that the desirability of a particular object is not sufficient justification for the use of coercion."
Fried Egg: Physiocrat, I don't think you have addressed my question. Imagine a socialist society organised not under one single planning authority, but into ten different planning authorities. Each planning authority took under it's control one tenth of the available capital and populations to run as collectives. If each collective in this community could buy and sell factors of production from/to other collectives in that society, there would exist a competative market for the factors of production, one in which meaningful prices would emerge and on which each collective could use internally to plan it's production.
There would be competition to some extent and prices would emerge but that's in an analogous way to how the current British railway service operates- there are x number of franchises and when you own one you can do whatever. But the size and number of these franchises is not open to competition and thus not truly competitive. This is the same in your example so the prices are deficient a priori. There's further problem though of competition in teh creation of the 10 planning authorities and the appropriation of the initial resources.
Fried Egg: The arguments in the article you supplied apply only in the case where there is one single central planning authority and therefore no competition of ideas as to how best to use the factors of production. Could this not be solved by breaking down socialist society into a collection of competing collectives? Could they not, in this way, abolish private ownership of the means of production without falling foul of the calculation problem?
As I've tried to show above in the absence of 100% private property true competition does not exist. All statist societies have the property problem because there's no competition in law of the defence thereof. You idea ameliorates the problems somewhat but they're still there until state ownership of ALL resources is abandoned.
I guess then that a socialist might content themselves with this form of market-socialism, accept the deficiencies of not having complete and free competition over the determination of the prices of the factors of production as a "price" worth paying in order to eliminate the evils they see as associated with the private ownership of the means of production (i.e. the exploitation of the working classes).