December 31, 2004
The Socialist Calculation Debate
Tyler Cowen has the good thought that we should reopen the socialist calculation debate:
Marginal Revolution: The socialist calculation debate: Here is one of my (very short) essays that nobody ever read; it is now on my home page. We all know that communism and socialism fail because planners do not have access to market price signals. They therefore cannot calculate the best means of producing goods and services. But is the argument so simple? I start with a few questions: 1. How does rational calculation take place within the firm? Keep in mind that some corporate giants are larger in economic terms than the smaller socialist economies. 2. If one person owned (privately) all the firms in the economy, would rational calculation be possible? 3. If one dictator controlled all the firms in the economy, would rational calculation be possible?
Posted by DeLong at December 31, 2004 09:15 PM
TrackBack URL for this entry:
Listed below are links to weblogs that reference The Socialist Calculation Debate:
» The socialist calculation debate: what's the relevance? from Idealistic but not naive, realistic but not cynical
Technorati.com invites everyone to enter resolutions for 2005. To some extend I already articulated them in a post in July The dilemma of the independent blogging politician: there is something [Read More]
Tracked on January 2, 2005 05:18 PM
Prices don't change. Products change.
Posted by: Paul G. Brown at December 31, 2004 09:30 PM
I think economists often bark up the wrong tree on understanding all the benefits of captalism via the efficiences of markets and the detriments of socialism via poor market choices.
I'm not entirely schooled in the most recent Behavorial economics issues but I would point to the effects of market systems with private property ownership in weeding out inneficiences of corruption.
I am of the opinion though, that its not just markets and capitalism that accomplish that but the dynamics of the creative destruction of shcumpert.
However, if left to its own, I am of the opinion that large commercial enterprizes would form beurocratic and cartel sorts of cartels over markets without a socialist sort of rule making for rules of engagement and socialist sorts of extra aid by states enabeling folks to gain property and unnaturally aid bargaining, pay and work rules for employees.
The independant home builder could be easily denied access to concrete in a world of payment for exclusivity, perhaps at the national level of the largest builders forcing a handful of mines only to sell to readymix plants that wouldn't sell to small competitors.
And I'm of the opinion that without the social welfare state that Marx may have been proved correct in that the price of any input, including labor, would decrease to the lowest cost of producing the next group of workers.
Where a line is drawn is of course problematic, but certain socialist activities involved in spreading wealth and accessing capital enable the market systems to function in a superior manner because of the freedom that such a sytem gives folks to spend marginal time finding needs and filling them with the time and tool resources that socialistic guarantees provided them (and the sense of the saftey net of being able to regroup by taking a job if their endevour failed)
Broad markets keep corruption in check much more easily. I understand the pricipals behind the notion of monopolies naturally being broken but I don't think that the added "stickieness" or "fat tail" or whatever connected to the behaviorial economics concepts really is fully counted.
Markets are biased to the recent past and won't react as quickly as they rationally should. This sort of attitude can help monopolists and cartels stay attop in a truer market system.
When you've got a federal reserve, school connections greasing classmates pockets, and government pork being able to put wind behind different proposals in an UNCapitalistic way, the acess to capital, albeit corrupt, also functions like socialism in magnifying a created destruction and broadening the number of suppliers (and lining previously unlined pockets) so that the money slushes around a bunch more in similar ways to socialist sorts of aids.
Long and short...Markets are great but socialist (or facist?) influences help the markets stay fluid and work towards producing a wealthier socitety.
Of course, other external events like the Plague also proved helpful in granting wider empowerment to people influencing the markets. Conquests and expansion worked well too in times when land could be settled or taken and worked gainfully in ways that basically widened the pool of independent players.
In a bearocratic, tecnocratic, corporate state I'm not sure that such wide abilities will naturally occur without public policies that are distinctly enabling of idividuals.
Public health care, a socialist proposition, might actually enable budding capitalists from freeing the percieved risk of losing a job, and discouraging free rider anomolies where one company can soak up folks who already have health care from elsewhere because a benefit (that is required for all or for none by most state laws) that other firms offer isn't worth anything to them.
The capitalist gripe would be that the anomalies are caused by government interference in the market place that disadvantages some companies through govenment interference.
A public healt care voucher to all is a market socialist notion that might enable a wider capitalist marketplace than a true capitalist system with no such government involvement.
Posted by: Tom Norian at December 31, 2004 11:14 PM
I've always figured that the entire advantage of decentralized economies over centralized economies is that the person with the most information about a particular decision a) makes that decision b) pays the biggest share of whatever cost there is from making the wrong decision
One private person owning the entire economy loses that advantage. I wouldn't expect that person to do a better job allocating resources than a politburo.
Posted by: Michael Brummit at December 31, 2004 11:32 PM
To simplify my ramble....I think "markets" are great but markets really don't require "capitalism". Markets require trade but how invidual parties come to their trade goods doesn't necessarily require that the goods were gained through notions of saving and investment and non fiat currency.
The "pureness" of incentives doesn't translate linearly with motivations to take endevours. There is a relation, and there are behavorial issues.
People might work less ala Art Laffer, if they felt the government was taking 70% of every incrementatl dollar they made. But if the government had an excise tax across a whole industry the successful business person might be making the government three dollars more for every buck he earned but the tax might look like any other cost input in his product that he sold at 10% gross margins.
If that money was taken and distibuted to people per head at various levels of discretion of the peoples priorities and weighing of benefits and sourcing by council people or representives as if they were owners, you might have a primarily socialist system that used markets.
I just doubt there is much "capital" in capitalism these days, and I'm not sure "capitalism" has a greater claim to the word markets than subsitance, tribal, military economies. I can envision economies where right to hold assets were absolute but the owership was so focused that it inhibited trade compared to more socialsit forms of government where citizens could make claims upon societal captial that "wasn't theirs"
Posted by: Tom Norian at January 1, 2005 12:26 AM
If ownership is concentrated in a relatively small number of large corporations which are operated centrally (that is, not using robust profit-center methodologies), then at a certain point of concentration, the net effect is similar to the problems faced by socialist economies.
Actually that seems fairly obvious. The question is, how close might we be to the threshhold of hyper-concentration, and what does the curve of inefficiency versus concentration look like?
I suspect those questions are very difficult to answer except in retrospect, which really means, not at all.
Maybe a simulation would help. When lack of diversity of suppliers starts to affect pricing and quality, the effects become generalized.
Posted by: Ralph at January 1, 2005 01:14 AM
A good subject for New Years Day, Brad!
One way in which the calculation problem could be approached is by a computer network distributed throughout society. But, one may reasonably ask, why would people participate in such a network? Wouldn't a parallel cash economy tend to become the pricing mechanism? Perhaps not, if the right network externalities come into play. Thus, for instance, people use Microsoft software, even though it is often considerably more expensive in cash than alternatives. As open software improves, it seems likely that it, in turn, will displace proprietary software. For open software production, a donation and government subsidy model seems to be workable. Thus, in open software, "price" is replaced by some different thing.
If, for a moment, we imagine Drexler's nanotechnological revolution, it becomes plain that participation in the information networks that control nanotech production will be so important to wealth, and so there would be a very strong incentive to participate in whatever price calculating network was worked out.
I do not think we will achieve the strong artificial intelligence necessary to Drexler's revolution. But it seems likely that, in the 21st century, it will become possible to produce most products using locally-extracted materials and locally-gathered energy. So I doubt that the economic system of the 23rd century will not be industrial capitalism. And so, it seems to me, the market price calculation will be superseded by some new calculation mechanism, which will be as much superior to the calculation system of industrial capitalism as that system was itself superior to that of agriculural feudalism.
Posted by: Randolph Fritz at January 1, 2005 02:20 AM
A simulation could be interesting but the algorithm used would produce junk in junk out problems I'm afraid.
Somehow human biases against change and irrational human behaviors such as personality conflicts and personal grudges are probably underestimated in economic models.
I don't know how you build relative power lust and vengance into economic profit models.
Say Scott McNeally or Ellisonhad an oppotunity to poke a major whole in Microsfots franchise even if the IRR for their own company would favor them paying out a huge dividend or something.
I can envision billions of dollars being allocated on a personal dream. Of course in that instance the human bias would lead to a dissapation of control.
I suspect a more common human trait is the florentine Meddicci's desire to grip power.
It would be nice to answer the question, if left alone, would capitalism gather wealth and capital into a few hands, and thus as you noted,if that happened lassiz faire capitalism might lead to simlar problems associated with state control.
I do believe there is a natural correction to such a state of affairs...its called Revolution.
The incremental socialism we have so much a heart of our system actually might make markets function more efficiently than would a pure lassis faire capitalism.(I'd think even in the US given the large percentage of money collected and spent by government in a non user tax way a person would need to concede that we were a fairly socialized government)
I don't think the concept is new by any means but by changing the semantics and saying socialist policies help markets be markets it causes a mental inversion that free's different ways of thinking.
The customary method is talking about "regulating capitalist excesses" rather than using socialist policies to Enhance competition.
The coining of the terms changes the argument. (and frustrates the heck out of people who love markets and think they hate socialism)
Posted by: Tom Norian at January 1, 2005 02:32 AM
Of course, early socialist activities of our government might have included the postal service, the homestead act, and road subsidies? To a greater and lesser extent those are examples of acts where goverment saw that giving away services and property might lead to greater long term good....I suppose you could call that a capitalist cost benefit analysis from the entitty called "government" or perhaps "society" (who owns the government...government property could be seen as jointly owned private property?).
Of course if you looked at any cost benefit analysis leading to greater long term wealth as "capitalist" decision you might see social security as a rational "capitlist" decision. But thats a stretch.
Its not capitalism vs socialism (which I believe are more philisophical positions of who "owns" property and property "rights" but notions of where "markets" and where "planned economies" or the middle of where to interfere and change the parameter in which markets function where the true decisions are made. Thats a matter of studing the streghts of where markets function best and limited use of planned coraling (diking?, damning?, routinging?, guiding?) of the markets run does more good than harm.
The "socialist vs capitalist" debate seems so tinged with notions of "moral rightness" and I might say antiquated notions of what money and wealth are. I'd drop the terms.
Posted by: Tom Norian at January 1, 2005 03:05 AM
Paul Cockshott and Allin Cottrell have done some work on this question: http://www.ecn.wfu.edu/~cottrell/socialism_book/.
Posted by: Kieran Healy at January 1, 2005 03:40 AM
Ralph suggests "Maybe a simulation would help. When lack of diversity of suppliers starts to affect pricing and quality, the effects become generalized."
I think that one could look at the concentration of ownership in the mass media to see what happens to "price" and "quality".
"Quality" has gone down, while "prices" have increased---and as a result a plethora of alternative media has emerged. However, with a infintesimal percentage of exceptions, these alternatives are not economically self-sustaining, and none of them are likely to ever be in a position to compete economically with the "concentrated" media. (i.e. Duncan Black and Josh Marshall may be able to make a living blogging, but they will never be able to compete with Rupert Murdoch.)
Posted by: paul_lukasiak at January 1, 2005 05:04 AM
Companies which make inferior products, or
good but high-priced products, will lose
revenue, lose access to low-cost capital, shrink
and often cease to exist.
So there is not much more "calculation" than
in Darwinian natural selection - but that turns
out to be a very effective mechanism for
creating good designs.
Having worked over 20 years in various
corporations, I've never seen much evidence of
rational calculation in the management of
them, other than killing failed projects
usually about 18 months after the failure
has become obvious :-)
Posted by: Richard Cownie at January 1, 2005 06:59 AM
Real companies behave differently than theoretical companies. For example many industries are characterized by an industry leader and many sub-leaders. It's commonplace for non-leaders to price their products based not on their costs but relative to the industry leader's products. So they're basing their calculations on someone else's putative calculations. It's a form of rational calculation but not what most people think of.
Posted by: Dave Schuler at January 1, 2005 07:20 AM
Brad quotes: We all know that communism and socialism fail because planners do not have access to market price signals.
No, we all know they fail for entirely different reasons, namely because few people will produce for no extra reward. So you either (1) have to force them (Stalin) or (2) they produce little, move to underground economy or run away. (Even Chomsky beleives garbagge collection should be the citizen's duty rather than a job - http://blog.zmag.org/index.php/weblog/entry/more_on_the_draft/)
[I had thought that really existing socialism had no problem with the idea of material incentives...]
Posted by: a at January 1, 2005 07:24 AM
That's a mischaracterization of Chomsky's views on garbage collection. He is saying that some people shouldn't be forced into exclusively being garbage collectors due to lack of alternatives. Rather, he is saying that scutwork and undesirable jobs should be equitably distributed, without having a few people who have "nice" jobs and a lot who have "dirty" jobs. He also certainly believes that those who do such work should be well compensated--ie, not a duty. You have to follow the Parecon discussions are little bit to understand the position (though you don't necessarily have to agree that Parecon will work, but Albert and Co. put up a pretty thorough defense of it).
Posted by: Mandos at January 1, 2005 08:17 AM
New blog format is much less appealing visually and harder to read on the screen than old. Find myself reading your site much less frequently as a result.
Posted by: Martin at January 1, 2005 08:24 AM
My feeling on it is that the only competition worth mentioning when it comes to pricing, is the ability to say "no thanks, I'll not purchase X good at all". That's the only thing that seems to lower prices.
In any case, the real concern I'd have with...well..anything monoplist, is that without anti-trust laws, you'd basically have complete economic control of the economy, if you controlled one of a few key sectors.
The two in mind I can think of, is energy and bandwidth. If a local power company decided, we're turning off all the power to the supermarkets not owned by us, forcing them out of business then giving a local supermarket monopoly. They could do the same thing for practically everything. You could never start a business because you can't really count on the things you need to run your business.
Posted by: Karmakin at January 1, 2005 09:03 AM
I don't see that anyone has commented on this, yet, so I'll bring it up... It strikes me that, contrary to Mr. Cowen's paper, there is a price signalling mechanism, even in a planned economy: inventories. Inventories go below a target value, the planners can see that in the statistics and raise prices. When inventories are too high, lower the prices... I'm not sure how this relates to intellectual property, where inventories can be essentially unlimited, but I imagine that problem can be resolved as well.
Posted by: Aidan at January 1, 2005 10:09 AM
Of course, "rational calculation" does still
take place - but mostly in the minds of
individuals who have personal choices both
as consumers (which car should I buy ?) and
as workers (which job offer should I accept ?).
The historical weakness of socialist systems
seems to be that they have a fuzzy view which
ignores details - and details *really* matter -
such as whether your tractor breaks down after
2 years or 10 years.
Posted by: Riichard Cownie at January 1, 2005 11:12 AM
"...communism and socialism fail because planners do not have access to market price signals. They therefore cannot calculate the best means of producing goods and services. But ... 1. How does rational calculation take place within the firm? ..."
No. Communism and socialism -- and many private firms as well -- fail because even when market price signals are present, most people will ignore them if at all possible, or rig the market to distort the price signals.
Within private firms, "rational calculation" based on price signals constrains decisions only to the degree that market feedback is prompt and unambiguous. In many businesses, it's not, and firms operate nearly "open loop", with the market functioning only as an agent of Darwinian selection on whatever heuristic strategies and tactics the firms happen to employ. And even that Darwinian selection is less than perfect in weeding out the unfit -- inept managers may by dumb luck hire subordinates skilled enough to save their bacon (they may even inherit them from competent predecessors who were fired by incompetent upper management).
Any system that is not robust enough to function without failure in the presence of corruption and incompetence is doomed to failure. Both democracy and free markets work better than the alternatives because they allow the consumers to vote rascals out.
Posted by: jm at January 1, 2005 11:34 AM
I said: Even Chomsky beleives garbagge collection should be the citizen's duty rather than a job.
Mandos said: That's a mischaracterization of Chomsky's views on garbage collection.
Chomsky said (http://blog.zmag.org/index.php/weblog/entry/more_on_the_draft) : In a decent society it shouldn’t be “volunteer” in the sense that it’s undertaken only by people who are driven to it by need. Rather, it should be equitably distributed—which one can call “conscription” if one likes.
Posted by: a at January 1, 2005 11:35 AM
I’m not sure if I have a point or not, but wouldn’t a capital gains tax of more than 50% essentially result in a socialist economy while preserving all the current decision making structure. This is clearly politically untenable in any current market economy, but it seems superior to taxing labor in a time when labor is being moved overseas.
Posted by: ExLurker at January 1, 2005 11:51 AM
Longer notes on Random Thought:
I've worked for a couple of medium to large companies. Decisions about our product's prices were almost always made with little regard to market data. There was some flexibility at the point of sale -- sales staff were permitted to give steep discounts. But whether or not the firm's management signed off on a particular deal had more to do with how well the sales team within the manager's organization was doing during the period in question. All of this a socialist centralized planner might do just as well.
However, prices shifted when new products came into the market; when an entrepreneur saw an opportunity to use technology in new ways and created a new product with a different value proposition. Coming up with successful innovative products is, as other posters have suggested, subject to 'natural selection'. Lots of stuff gets tried. Few have any impact. Some of the junk you just couldn't give away!
Innovation along these lines is extremely wasteful. Lots of time and energy that might be spent honing production and distribution instead gets diverted into "makin' stuff up", and very few of the things created have any value. But I have no idea how an economy run on a centralized planning model would replicate this. It depends on individuals making local decision--most of them bad.
So. It isn't clear to me that improved access to market signals would, in the long haul, make much difference to socialist planners. Market signals aren't much used by planners within capitalist firms, either. The 'socialist calculation' debate, therefore, misses the point.
Posted by: Paul G. Brown at January 1, 2005 12:19 PM
Brad said: [I had thought that really existing socialism had no problem with the idea of material incentives...]
Yes - but
1. The range is much narrower than in the capitalist society. You have to be Stalin to maintain the range (Stalin's take on rags-to-riches: Gulag-to-CPSU Central Committee).
2. Also, the magic of competition. In the capitalism you have to outrace the other guy who does his best to outrace you. Everybody keeps paddling faster as a result - even those who do it all the way to the bottom.
Posted by: a at January 1, 2005 12:21 PM
Hey, I like garbage collection, if the wages were high enough. You get to be outside, have meaningful work, see places progress over time.
Well, it might get boring but so would selling the same insurance policies over and over again.
From a Studs Terkle perspective I think you could do worse than being a garbage man.
The pay of the job and the benefits seem to be more important to the issue of whether its a particlualry exploitive job. I'd take it to most cubicle postions anytime.
Posted by: Tom Norian at January 1, 2005 01:00 PM
a: This quote (in the overall context from which it emerges) clearly does not imply that they shouldn't do it for "no extra reward" which is what you were accusing Chomsky of proposing. Again, Chomsky is discussing in terms of Parecon, which does not envision a world where everyone is necessarily forced to be a garbage collector regardless of discussions about the desirability of the job...
Posted by: Mandos at January 1, 2005 10:48 PM
Mandos said: a: This quote (in the overall context from which it emerges) clearly does not imply that they shouldn't do it for "no extra reward" which is what you were accusing Chomsky of proposing.
Please do not attempt to twist my words. I did not accuse Chomsky of proposing that doing garbagge collection duty should carry no extra reward. I said very explicitely "Even Chomsky beleives garbagge collection should be the citizen's duty rather than a job." Duty in exact sense Merriam-Webster has for it: "obligatory tasks, conduct, service, or functions that arise from one's position (as in life or in a group)". In fact, in many armies conscripts receive pay for doing their duty so your reference to "extra reward" does not make any sense.
Posted by: a at January 2, 2005 12:22 AM
I had a friend in the UK in the early 1980s who
was very bright, fully qualified as a teacher, and
chose to work as a garbageman. He liked to keep
his brain for his own thoughts, didn't mind getting
up early in the morning, and loved being finished
with work by noon every day.
Of course that was when it was a council job with
decent benefits - the big wave of privatization
and outsourcing may have made it much more
unpleasant since then. Still, the principle
holds true, collecting garbage is not necessarily
a terrible job, and for a person with the right
temperament it can even be quite appealing.
Posted by: Richard Cownie at January 4, 2005 12:43 PM