Moving Beyond Markets

R. Salisbury

Thanks to the downfall of the Soviet Union and other nominally communist countries, decades of red scare propaganda, and a poorly-fought intellectual war against socialism, people today widely believe that capitalism, or just markets, are indispensible for determining how we should provision resources. This belief has given rise to alternative systems such as "parecon", market socialism, distributism, market anarchism (which is a leftist stance not to be confused with "anarcho"-capitalism), and so on. There is a diverse ecosystem of these ideas, but they are all largely motivated by the same red scare and neoclassical propaganda. There are a few basic ideas at the heart of this belief:

  1. Allocation is so complicated that no mechanism could possibly accomplish it as well as markets: This is the most common, and is based on ideas initially espoused in "Economic Calculation in the Socialist Commonwealth" by Ludwig von Mises. There are numerous and manifold absurdities with the economic calculation problem, which I will not get into here. To briefly touch on it, the idea that we could not, in the information age, deal with a complex, information-based problem, except by using a millenia-old technology invented for the purposes of provisioning standing armies, is a bold claim that requires substantially more support than the double-standards and arguments from incredulity that the Miseans and their ilk present.
  2. Markets are naturally emergent and it will be impossible to get rid of them, so we must embrace them: This is also a very common argument, and often makes a point of the existence of black markets in the aforementioned "socialist" countries. Unfortunately for this argument, markets do not emerge naturally, and even the Soviet Union thoroughly used markets and money in its state-driven functions. Without using currency, without taxes being imposed on citizens, and without a plethora of goods available for purchase, there is no reason for markets to form.
  3. People will not cooperate without markets/markets best deal with our natural self-interest or greed: This is the human nature argument. This argument has been addressed numerous times before, so I will simply restate that humans have many natures, proven by the fact that not all humans act greedy or self-interested.

On the contrary, markets do not solve any allocation problem; decentralization of decision-making simplifies allocation, and markets decentralize decision-making to a certain extent. However, they also introduce additional problems, primarily serve the ends of money creators, and are based on coercion and rule, controlling human behavior through a system of sanctions and privileges. Allocation without markets is indeed possible, desirable, and necessary. Using modern concepts in mathematics, logistics, and computer science, we can come up with a basic idea of how to accomplish it.

The Duty of the Commonwealth

In markets, obtaining survival needs and luxuries is contingent on fulfilling the whims of someone with money. This is a social duty that forces the masses to do the bidding of those in control of money creation and distribution. In a society free of rulers, we still need some manner of coordinating social activity to achieve social needs, such as building and maintaining infrastructure and caring for the elderly and disabled. For obvious reasons, it is important to share resources and help one another, and to be included in markets, there is a social duty to fulfill. This is a necessary component of any society.

Citizens of a society organized without rulers and without markets would still have certain duties or obligations expected of them. Though there may not be rulers, there are still rules, and there are still consequences to violating them. There are some large-scale projects that will be necessary for basic social function; some of these would be done by volunteer, but it's unlikely that all would be. In that case, the community political body would need to enact a mandate, which would select by some mechanism (vote, lottery, etc.) enough qualified candidates to complete the project. Those selected would have a duty to make a good faith effort to work on the project through completion, or to provide a good reason that they cannot. There is no reward for this; we should, as much as possible, strive not to create institutions in the shape of Skinner boxes, reducing human agency to control behavior through sticks and carrots.


The thing most people are concerned about today when discussing a system of provision is how new products are created. Production of something new requires choosing the size of an initial production run. Businesses cannot predict how people will respond to a new product. Markets do not have magical predictive powers, so businesses have to do market research to discover this. Non-market processes to decide how to allocate resources would have to do something similar. In other words, no special new process would have to be created for this step.

Crowdfunding sites highlight another way to predict demand for a new product. We can simply create a listing describing the product, and collect preorders for it. This gives us a good starting figure without any guesswork. The clever market enthusiast would say that this will not work the same way, that some of this demand will be illusory, since without paying money, they aren't risking anything by saying yes. However, this is a fairly weak argument: People get buyer's remorse and products often become "vaporware" even if they get funded. There is not a 1:1 match either way. Because the distribution process happens over time, it doesn't matter much if the first guess is exactly right. If there are leftovers, they can be held for later demand, and the production target can be adjusted downward; if there is more demand than expected, the production target can be adjusted upward, and those who didn't get the product can wait until more are made.

Others are concerned with innovation and how it could be engendered outside of market competition. For one, competitions of ideas do not also need to be competitions for survival. The need for market participants to compete for their lives or for the existence of their organizations, rather than stimulating innovation, stimulates risk aversion. The quid pro quo nature of the market leads investors to demand results-based funding for research, which stifles innovation and again stimulates risk aversion. Markets demand some form of intellectual property, whether something akin to patent law or simply secrecy regarding a business's developments. Without it, there is little stopping a company from monopolizing a market by using everyone else's ideas rather than developing their own. However, when it comes to innovation, adopting one another's ideas is the wiser choice. Software has evolved at a blistering pace in large part because developers started a movement aimed at doing exactly that.

Product Modeling

In order to have a rational method of provision without the use of money and prices, we need to perform calculation in kind, which simply means "economic calculation" using actual, material units, as opposed to using prices. When creating a product, service, or project, its specifications will be written in a computer-readable standard. This should be considered a requirement for receiving the materials needed from the commonwealth. The particular standard doesn't matter a whole lot, since computer systems are easily able to deal with multiple standards. The standard will include the measured properties of the product (for example, a steel specification would include its tensile strength, hardness, etc.), a name, ID, and description, unit size, and necessary inputs. Each of those inputs would themselves be modeled in the same manner.

The last item is important because it's what allows calculation in kind to take place. Those familiar with programming concepts will see that there is a recursive relation between products: Each product contains a map of the products that compose it. This mapping creates a tree from each product to the most basic materials that compose it (technically they would not be trees, since there would be cycles, e.g. energy must be used to produce energy), which means two products will be comparable in terms of common units. These common units are multidimensional, making them more complicated to deal with than the one-dimensional price, but they are also actually measuring something meaningful, unlike a price. The comparison can be made on the basis of the differences between the two. In other words, if we're comparing two ways to achieve the same result, we can ignore the similarities between the two and analyze only the differences.


When economists speak of the necessity of economic principles, they are typically talking about opportunity cost. In other words, they are concerned with how to decide what to do with scarce resources when there are competing ends for those resources. The only rational way to decide this is to set some priorities to resolve conflicts. In capitalism, the higher priority item is whatever will generate more revenue. In a command economy, it would be whatever the state believes is more important. These priorities are used to decide which of the competing ends should receive the scarce resources they need.

While artificial scarcity is heavily created in the capitalist system, and most probably in an alternative market system as well, real scarcity does exist. That makes it important to have some sort of method to deal with scarcity when it does happen. Our non-market system needs to have its own set of priorities. In our system, this can be done very impartially. We can measure the priority of a resource based on how important it is to the measured demands of society. I mentioned earlier that proceeding from a product through its components generates a tree. Taking all these trees together would produce a very large graph, which is increasingly connected (meaning there are more paths to the same node) from the products to the basic components.

In graph theory, centrality is the measure of a node's influence in a graph. Centrality is used, for example, in Google's PageRank algorithm to measure the importance of a page to a search phrase. Its influence is, roughly, the number of paths that go through it. If we were to examine all possible paths through a graph, the highly-central nodes would appear most often in those paths. Applying a centrality measure to the resource graph will give us a figure for each node (resource) we can use to resolve conflicts between competing ends--the end with the greater priority wins. To give an obvious example, if we had limited water and were deciding what to do with it, food crops would be prioritized over tobacco, because food is more important to the functioning of the overall system than tobacco. Priorities help us connect costs and outcomes, so that a dirty factory that produces cigarettes would not be produced instead of a clean factory that produces insulin, even if the cigarette factory is more highly demanded.

Tracking demand

For its initial production run, a business needs to make a guess based on the measurements it took. It will then track the rate of consumption of these initial products, and adjust the amount it needs to subsequently produce based on this measurement. Again, there is no need to reinvent the wheel here. This guess, check, and adjust process can occur without the use of prices or markets. We could even use something like a store, without the part where you give money for the thing you want. This would make demand tracking a matter of keeping a ledger on the amount in the store by taking inventory of deliveries and of the items in the store. For products not intended for end users, as in factors of production, it is less likely they would be put in a store. In this case, just-in-time manufacturing can be used. Just-in-time (JIT) manufacturing is a modern logistical strategy, where products are only manufactured when they are demanded. Tracking demand would be a duty of the manufacturer.

For services, we can track the utilization rate of the service, as well as the material needs for the typical utilization rate. Utilization rate, to put it simply, is the ratio of use to non-use. If a service is available for 8 hours a day and it is used for 6 hours a day, the utilization rate is 6/8 or 0.75. Making a service available to multiple users is a well-studied problem called queuing theory. Using measurements of the average rate that users arrive to use the service, the average time it takes to serve the user, and the desired level of service (the likelihood that a user will have to wait in a queue), queuing theory can tell you exactly the number of servers to use.

Periodic Budgeting

Markets lack any way to actually budget real materials. The cost of a material is the money required to get humans to extract it. This includes the money paid to the workers, money paid to the landowner, as well as the instutional fees paid for the damage done through the process of extraction. There is no connection to the actual (long-term) availability of the material--capitalist businesses care only about the rate at which it is extracted and sold. Even when a survey is done to find out how much of some resource there is, it only goes to a certain point--businesses don't know the total reserves, they only know the proven reserves.

An ecological budgeting system needs to be based on the replenishment rates of the resources it's budgeting. If a community can replace 100 trees per year, then it cannot sustainably harvest more than 100 per year. Periodic budgeting makes this idea the basis of the accounting system. Budgeting is done on a fixed period, such as a month. The replenishment rate for resources, normalized to this period, are the absolute limit for the budget. Following our tree example, the monthly budget for trees would be 8. This should be considered a limit, not a guideline; ideally, consumption should be less than this. There is no borrowing from future periods and no substituting one resource for another to get around the limits.

An indispensible quality of money, according to the economic calculation problem, is that it can be used as a common basis for comparison. To compare to different things with different properties and different compositions, we need a common factor between the two. The need for this is exaggerated in the calculation problem, as things must still be compared for their specific properties even if they have prices attached to them, but it's still useful to be able to compare two different things using a common basis. Calculation in kind already has a common basis of comparison, which is energy. Energy is required for all activity, so any activity, such as a production process, can be compared to another in terms of its energy consumption.

Using periodic budgeting, we establish a second common basis of comparison. We can now compare things using the resources they consume in terms of units of time. This allows us to examine the benefit of expending resources to affect the amount of time something takes. For example, we could find out when expending additional energy and materials to speed up a transportation system no longer produces any benefit. Using time as a common basis unit, it may even be possible to use existing process scheduling algorithms that computer operating systems use to provision computing time to provision potential projects with commonwealth resources. This would eliminate much of the work in implementing an unproven system. Time value will be covered in more depth later.

Priority Queueing

Put all these elements together, and we get a basic system of allocation without markets. Production and other social activities are done on the basis of voluntary initiative, unless the community decides that there is a project that is necessary and requires conscripted labor. These projects will have members selected according to some mechanism such as lottery. Each budgetary period, groups that need resources will submit proposals into a queue. This will get easier over time as the data model is developed further. When the submission deadline hits, the queue will be emptied, in priority order (see: priority queue or heap), until the budget limit is hit. Proposals that leave the queue will have the consent of the community to use resources from the commonwealth for their specific ends. The community can institute rules for being considered worthy of inclusion in the queue, such as preventing proposals to use resources of the commonwealth for commercial gain, and producing a machine-readable representation of the result.


Allocation without markets is both conceivable and possible. This particular concept has been developed over the years and is by no means complete, but is more than enough of a starting point to build a working system. Some parts of the concept, such as periodic budgeting, would be useful in any system of allocation. Others are strongly-coupled and would have to be implemented simultaneously in order to be useful.

While advocates of markets allege or imply that markets are an apolitical system, in reality there are no apolitical systems of provision. The system described above is no exception--it is based in anti-authoritarian politics, ecological principles, libertarian muncipalism, and the open-* movements of the 21st century. Like any political system, it will require constant evaluation and criticism to ensure that it adheres to the anti-authoritarian principles that are important to us as anarchists. Like any political system, speculating on how it could be is much easier than doing the work of building it and imbuing it with political agency. However, conceptualizing such a system is important to counteracting pro-market propaganda and to set goals for ourselves to strive towards.

The system may not be perfect, but that's not the end of the world. As long as it functions correctly for the most important things, i.e., food, water, shelter, things can still work out. Whereas economists criticizing alternative allocation systems presage certain doom for any system that fails to fulfill the slightest whim of its participants, in reality people are able to go without a lot before they start rioting. After all, the majority of people on Earth are little more than slaves, and work tirelessly just for their basic needs. What's really important for a system of provision is that it fulfill those basic needs and free people from toil. Only after that is done for all people do we need to begin thinking about doing more.