Recently Robin Hahnel posted the following article on the Participatory Economy Project's blog: Labor Time Accounting: A Postmortem. In this article Hahnel publishes some of his thoughts on the value theories of the classical political economists Adam Smith, David Ricardo and Karl Marx. He also talks about some contemporary political economists, namely Paul Cockshott and Allin Cottrell.

The article is largely about the so-called labour theory of value (LTV). I don't particularly care about the LTV since it is not relevant to planning. The main place where it finds relevance is when interacting with remnants of the market, since having an accurate value theory makes it easier to game the market. It also has relevance in distribution (in the vol II sense).

Hahnel's argument is that there are three different categories of "real" costs in the economy:

• labour costs
• environmental costs
• "time" costs

I will address each of these in turn.

Labour costs

Labour costs are the labour added directly and indirectly in the production of some commodity. In Marx labour is the only cost that a commodity has. Things like gold in the ground, air, sunlight and so on, the so-called free gifts of nature, have no value according to Marx. Only the act of exploiting the free gifts of nature can imbue them with value.

Hahnel makes the point that there is not just an undifferentiated mass of labour power, but specific types of labour power, and limited supplies of each type of labour power. This is an excellent observation, and an important one when it comes to planning. Marx makes much the same point in Capital, pointing out that value abstracts away all the details of each commmodity, and of the labour that went into making the commodity. Instead everything boils down to a single number: the price of the commodity. So, nothing new there. I agree with Hahnel that projecting everything down to a single number is a Bad Thing. It is bad because it leads to bad regulation.

Hahnel appears to believe that Cockshott and Cottrell argue that value should be the only thing that a socialist economy concerns itself with. But if one reads Economic Planning in an Age of Climate Crisis by Cockshott, Cottrell and Jan Dapprich (ISBN-13: 979-8360125563, 2022), one finds that this is not the case. Nor am I aware of any proponent of planning that argues in favor of value being the only thing we should be concerned with. That would be silly. On this issue Hahnel appears to be arguing against straw men.

One thing that I believe Hahnel is trying to say is that disaggregation of labour power for the purposes of planning has not been sufficiently theorized. On this I agree. Cockshott has argued that planning must be carried out in as disaggregated a manner as possible. But unless I've missed something, Cockshott still talks only of abstract labour power in planning, rather than realizing that labour power must also be disaggregated. Planning should not be carried out on the basis of abstract labour power, but on what I will call concrete labour power. Concrete labour power is distinct from concrete labour in that the latter is the actual expenditure of labour on some task, whereas concrete labour power is merely the ability to carry out such labour. Abstract labour power is the ability to carry out any task that a human can do.

As a concrete example of the problem of only considering abstract labour power, consider an economy with only two jobs: plumber and hairdresser. Let's say that we have four people in our economy: two plumbers and two hairdressers. The total amount of abstract labour power is four. A planning system only concerned with abstract labour power would see the following:

The gray area is what the system believes is the feasible region, while the halftone green area is the actually feasible region.

Suppose we need three plumbers and one hairdresser. A planning system based solely around abstract labour would not be able to notice that we do not have enough plumbers to satisfy the demand for them. It would attempt to position itself at the red circle in the figure below.

If we train one of the hairdressers in plumbing we can modify the plumbers constraint and arrive at the following:

Now it is clear that a solution with three plumbers and one hairdresser exists. Other solutions are also possible, for example having the one plumber-hairdresser work half time at either job. The abstract labour power constraint tells us that we can't have more than four people working. It tells us that our three plumbers and two hairdressers are in fact four people and not five.

Finally, it might turn out that we definitely need at least two plumbers, but that hairdressers are deemed "not a need" and therefore we can accept any solution within the green area below:

A feasible solution could be to have the three plumbers work 75%, making 2¼ plumber, while the one not-plumber hairdresser provides the community with mullets whenever there is a want for them.

Environmental costs

When it comes to the environment, Hahnel makes the point that scarcity also constitutes a cost. This is somewhat true in the case of land rent, since land rent can be seen as a cost levied by land owners. Oil is a particularly striking example. But in planning it's a different story.

If something is truly scarce then no production can take place beyond the supply of that thing. It makes no sense to speak of production beyond physical limits, much like it makes no sense putting to work more plumbers than there are. Such production simply cannot happen. In planning we can further say that certain kinds of production shall not happen, or that it shall happen only within a certain narrow set of constraints. The burning of fossil fuels may happen, but only within certain limits, and only until the use of fossil fuels has been made obsolete by the introduction of new fuels or other methods of power delivery, or by the elimination of the need for said power in the first place.

Hahnel is right that we must take into account the fact that we have limited supplies of raw materials, land and so on. But the word "cost" is intimitely linked to exchange. In planning we do not have costs, but constraints, of which cost is but one. When baking a cake, we do not say that the cake costs such-and-such amounts of eggs, flour and so on. There is a cost associated with buying these ingredients, but the use of the ingredients themselves is not a cost, since the baking of cakes is not exchange.

Ricardian rent comes about from the fact that the market must navigate a land constraint. The marginal cost associated with land is in fact the slope of the land constraint. It is the unit increase in labour for each unit of land-constrained commodities to produce. It comes about whenever production butts up against the land constraint. The following set of figures hopefully explain the concept.

In this example there are three plots of increasingly marginal land:

• Land₁ on which can be produced at most 4 units of wheat, with a unit value of ½
• Land₂ on which can be produced at most 2 units of wheat, with a unit value of 1
• Land₃ on which can be produced at most 2 units of wheat, with a unit value of 2

An economy which minimizes labour will first use Land₁, then Land₁ + Land₂, and finally Land₁ + Land₂ + Land₃. In the figure above this is demonstrated by setting the demand for wheat at 2. The green circle indicates the labour minimizing land use to satisfy that demand.

If demand is increased to 5 units of wheat then we get the following:

Here the marginal value (slope) of wheat has increased to 1 rather than ½. In the Ricardian theory of rent this marginal value is taken as the value of all wheat, not just the wheat produced on Land₂. In other words, the 4 units of wheat produced with 2 units of labour on Land₁ are now worth 4 units of labour, rather than the average value of 4*3/5 = 2.4. The extra 2 units of value are land rent. This seems counterintuitive until one realizes wheat is a fungible commodity that must be sold at a price that makes producing it possible in the first place. In other words, it must realize a price equivalent to 1 unit of labour per unit of wheat for Land₂ to be brought into use. The land rent earned from the wheat on Land₁ is taken from labour performed in other industries.

Finally for completeness, if demand is increased 7 then we get the following:

Here the marginal value of wheat is 2, so the total value is 7*2 = 14, of which 6 is value added and 8 is rent. The maximum possible wheat production is 8 units. It makes no sense to talk of producing 9 units of wheat, since there is no land on which to grow it. Beyond 8 units, the unit value of wheat becomes infinite. There is no amount of labour we can add to get more wheat.

If one extends the land constraints so that they intercept the Y axis then the rent extracted from the more productive pieces of land can easily be read off:

Here the land rents of 2 and 8 gained when Land₂ and Land₃ are brought into use are plainly visible. These rents remain constant so long as no more, even more marginal land is brought under the plow.

In convex planning, of which linear planning is a special case, the production of more output when the plan butts up against a constraint also carries with it a marginal increase in the amount of labour that must be used to get that output. In this sense there is a cost associated with these constraints, as Hahnel says, over and above the normal normal cost of producing the desired output. In the above example, when the demand for wheat rises above 4 units, there is an added cost of ½ units of labour per unit of wheat above 4. When demand is above 6, the added cost is 1½ per unit of wheat above 6. Another example is greenhouse gas (GHG) emissions. For lower levels of emissions, no extra cost is incurred since the environment can handle them just fine. But past a certain point, labour must be added to sequester these GHGs. This clearly constitutes an added cost.

In a hypothetical situation where an industry normally subject to rent is entirely planned, rent seizes to exist. In the case of agriculture this requires that the peasantry has been fully dispossessed and that there is no international trade in crops. Only planning. Whether this is actually possible to achieve is an open question.

Hahnel points out that we can calculate the amount of non-labour inputs contained in products. This is something that I have pointed out as well, in the post On vertical integration. We could vertically integrate not just labour time but also for example land and energy. We could then issue land vouchers and energy vouchers, similar to labour vouchers. We could stamp final products with prices not just in terms of abstract labour time, but also land and energy. The latter is suggested by proponents of energy accounting, where vertically integrated energy is known as emergy.

I'd also like to point out that land and resources were not less scarce when Ricardo or Marx were writing, as Hahnel states. Land is less scarce today than it was in Marx' time, at least up here in Norrland. This because the migration of the peasantry into the cities has freed up much of the land, to the point that the rural population is becoming older and sparser with each passing year. The main concern for the rural population is not that we're running out of land, but that the ownership of it is increasingly concentrated. It might be tempting to say that this concentration is equivalent to running out of land, but this seems silly when the land is experiencing afforestation.

With mining, more efficient machinery has brought into use deposits previously considered to be uneconomical. The amount of mineable ores has increased since Marx, not decreased. At the same time, the amount of iron on Earth has not changed significantly since Marx's day. Some iron has been dug out, reduced from its oxides and smelted, then turned back into its oxides again, hanging around as scrap.

What we have "run out of" is the amount of greenhouse gases that can be put into the atmosphere, not the fossil fuels from which these gases originate. This is a qualitative change compared to the fear of running out of coal that Otto Neurath wrote about. If we replace "coal" with "amount of GHGs the atmosphere can hold", then Neurath's argument remains as relevant as ever.

"Time" costs

The section on "time" costs is about what Marx calls "moral depreciation". It is the depreciation due to the adoption of more productive technologies, which render older technologies increasingly obsolete. Hahnel points out that due to these improvements, it can be worthwhile to wait a while, only investing when more productive machinery becomes available. In other words, we can invest real raw materials and labour today to save on raw materials and labour tomorrow. This is why calls for so-called "degrowth" are short-sighted. We may very well have to invest heavily today, growing the economy greatly, performing a large amount of accumulation, so as to satisfy environmental constraints tomorrow. This works because investment planning is non-convex. This is not to suggest that so-called "green growth" is good either. It too suffers from short-sightedness, but of a different kind. What we need are the correct investments in new means of production, and the correct use of existing means of production, in the correct amounts in the correct places at the correct times. Planning is the mechanism by which these amounts are arrived at.

Indicative prices

In the closing section Hahnel talks about the possibility of using indicative prices to deal with these "real costs". On this I am strongly opposed. Why? Because it amounts to indicative planning, the effectiveness of which for dealing with the environmental disaster are felt today in Spain. It is an attempt to indirectly deal with the environmental problem rather than dealing with it directly. It is an attempt to try to get the economy to do the right things by fiddling with distribution, rather than not engaging in undesireable production in the first place.

Final thoughts

What Hahnel calls "costs" seems to correspond to what is called "slackness" in linear programming. When we extract things from nature we reduce the slackness on what can be extracted from nature (modulo investments in new machinery). If the last few years of the planning debate has shown anything, it's that much "disagreement" is just differences in language. It is therefore important that we are very precise with our language, that we use unambiguous words when we talk about things. Else confusion can occur.

I sent a draft version of this post to Hahnel for review before posting, but sadly I got no reply. I hope that I have not mischaracterized his arguments.