Interviews Post-Capitalist Vision

Philipp Dapprich Interview on Democratic Central Planning – Part 1 – Opportunity Cost, Environment, Capital Goods

Editor’s Note: In this interview, Dr. Philipp Dapprich talks to After the Oligarchy about his work on refining the model of economic planning first proposed by Cockshott and Cottrell in Towards a New Socialism (1993). Discussion includes opportunity cost, labour cost, calculating opportunity cost in central planning, calculating environmental costs (such as GHG emissions), calculating opportunity cost of capital goods.

[After The Oligarchy] Hello fellow democrats, futurists, and problem solvers, this is After The Oligarchy. Today I’m speaking with Dr. Philipp Dapprich.

Philipp Dapprich is a political economist and philosopher working at the Free University Berlin. His PhD was entitled Rationality and Distribution in the Socialist Economy (2020), and today we’ll be discussing his work on refining the model of economic planning first proposed by Cockshott and Cottrell in Towards a New Socialism (1993).

Today’s conversation is in association with meta: the Centre for Post-capitalist Civilization if you’re not familiar with Towards a New Socialism you can buy the book or find a free PDF online you can also find interviews with Paul Cockshott on this channel and I’ll put links in the description to Philipp Dapprich’s doctoral thesis as well as a relevant paper.

Philipp Dapprich, thank you very much for joining me.

[Philipp Dapprich] Thank you very much for this conversation.

[ATO] Before we begin with the questions, I was talking to Paul Cockshott yesterday and he mentioned that actually you, Paul Cockshott, and Allin Cottrell, have finished a book, a new book, on economic planning called Economic Planning in an Age of Environmental Crisis. And that’s just with the publishers now, and it’s going to come out sometime this year [2022]. So, do you want to say few words about that?

[PD]Yeah. So, what we’re trying to do in in this book is two things.

First of all, we want to demonstrate that you need some kind of economic planning in order to tackle the huge task of transforming the economy away from fossil fuels. Paul Cockshott actually did a calculation, for the book, of the investment that would be necessary in the UK, as an example country, to completely transform the energy system. And the amount of investment that is needed actually exceeds the annual total private investment in the UK. So, if that’s correct then there’s no way that private investment alone will be able to tackle this, and you need the state to step in and take a significant role in this.

The second thing that we’re doing is showing how economic planning techniques can be applied precisely to this problem of transforming an economy towards a completely different energy source. So, one of the things that we’ve looked at is how you can do long-term plans that gradually transform the economy or the basis of the economy. And the other thing, which is something that I worked on in my PhD thesis as well, is to look at how we can consider environmental constraints in planning and also in valuation of goods.

[ATO] Just one more thing on that. It’s a book primarily about long-term planning and about applying that to the environment, or will there be material about relating a long-term plan to, say, a yearly plan?

[PD]The techniques we describe are, of course, generally applicable for long-term planning and they could be applied to any kind of long-term objective that you might have. But what we’re arguing in the book is that this would be particularly relevant when you’re trying to drastically change the way that the economy is structured, and especially the way the production of electricity and energy is done.

[ATO] Well, it sounds like it’ll be very interesting, and I’ll make sure to get a copy when that is released.

But our conversation today is about something else. It’s about your work on introducing opportunity cost valuations into the Towards a New Socialism model. But before we get into what new techniques and methods you introduced, I’d like to situate that in the history of this problem, and also talk a bit about Towards A New Socialism. So, to give the background to viewers, can you frame the issue of economic calculation so viewers can understand why the issue of opportunity cost is important? And we can go on from there.

[PD]Generally speaking in a socialist economy, a similar problem applies as in any other economy, which is how to apportion resources, labour, the means of production, towards various uses. How much labour are we going to use to produce food versus energy, versus other things? And you want to do that in a way that is in some sense efficient.

And there are techniques to do that. There are optimal planning techniques that that can be used to do that, but what they can’t necessarily tell you is which kinds of products are needed. Do we need more food, do we need more laptops, do we need more smartphones? They can’t really tell you that.

So, in a sense, there’s still a choice that has to be made by individual consumers or planners on what kinds of things to produce. And obviously one thing that will factor into that is the usefulness of these products. If we decide how many laptops are we going to produce, we’ll have to take into account well how useful are more laptops going to be?

But of course you also need, on the other hand, to consider what is the cost of producing these laptops. And I think that costs should be understood as an opportunity cost. Basically, when you use resources to produce laptops, you can no longer use these resources to produce other things. Labour that is used in factories to assemble laptops is labour that can’t be used to produce smartphones instead, or food, or something else. So that is really what is meant by ‘opportunity cost’. It is the opportunities that are lost when we’re doing something. When we’re dedicating resources to the production of laptops, we can’t use these resources for something else, and that is the opportunity that’s being lost.

And the difficult thing is trying to capture that in a way that can be measured. Where you have a single scale that would tell you what is the opportunity cost of a laptop versus a smartphone, versus something else.

[ATO] To clarify, you’re saying that, in the context of socialism and socialist planning, there are techniques of optimizing a plan which given the products and the resources that exist today can distribute that in an efficient manner. But that is different from saying next year how that list of ingredients so and products should be distributed.

[PD]When you’re optimizing in mathematics that always means maximizing some function. And what we’re maximizing is the product output. But, of course, you have many different kinds of products; laptops, smartphones, food, and so on. And the way that that we deal with this is by fixing proportions of these products. We’re saying, well, we’re going to produce two smartphones for every laptop, and for so many units of food, and so on. So, you have set proportions, and then we maximize the output at these proportions.

But, of course, you don’t just want to choose any arbitrary proportions, because you want that to reflect the actual need for laptops, and smartphones, and food, and so on. So, you have to adjust these proportions to the actual needs. And this is where opportunity cost plays a role, because when you’re then deciding ‘well, should we maybe increase the production of laptops relative to smartphones?’, then one of the factors besides the usefulness of laptops and smartphones that you have to take into account is ‘well, how many more smartphones could we produce instead of one laptop?’. That’s the opportunity cost: what other things can we not produce if we produce one laptop?

[ATO] Let’s introduce Towards a New Socialism, because, so far, we’ve been talking about economic planning in general, central planning in general. But Towards a New Socialism is a particular model which does things in a particular way, and you decided to take that model and to refine aspects of that.

[PD]What I found the most significant contribution of Towards a New Socialism, and what made me want to work on this model and refine it is that it has this automatic feedback loop which continually adjusts the mix of products being produced in response to consumer demand.

So, if there’s a high consumer demand for laptops then we’ll increase the production of laptops. And the way that this works is that first of all you regulate the prices. So these are token prices in terms of labour vouchers rather than money prices, but we can put that difference aside for now.

You adjust the prices towards the market clearing rates of products, so these market clearing rates are basically the rates at which the supply and the demand for product match. We’re currently producing 1,000 laptops a month and there’s demand for 1,000 laptops a month at the current price. That’s when you have market clearing rates.

And then these prices give you a good idea of how much people are willing to pay for these laptops. And that could be used as a proxy for how useful they find these laptops and how important these laptops are to their lives. Then the idea is, well, if the price is very high then people are willing to pay a lot for it and then maybe that justifies producing more laptops in the future, because people really value them. While if the prices – the clearing prices – are really low, that means people aren’t actually willing to pay that much for that many laptops, and maybe we should be producing less in the future.

But the question now becomes: the price is high or low relative to what? You need some standard of comparison and that will be different for a laptop than for an apple, because a laptop takes significantly more resources, different kinds of resources, but generally more resources. It’s more expensive in some sense to produce.

[ATO] You’re talking about comparing the price at which it sells to the cost of producing it. And then it’s a question of how you define that cost.

[PD]Yes. And what the Towards a New Socialism model does is measure the cost of producing an item in terms of the labour time, the socially average labour time, that is necessary to produce it. This goes back to Marx’s labour theory of value, which is the theory of prices under capitalism but which now applies this to socialism and says this is also an adequate measure of cost, of the cost of producing items under socialism.

And a problem that many people have with this, and I think there’s certainly some truth to it, is that you’re only considering one factor that is necessary to produce an item: labour. And you’re ignoring other factors. You’re ignoring the machinery that is needed for it. Not entirely, because machines take labour to produce, so you could say labour factors into that as well. But you’re certainly ignoring things like fertile land, which is scarce and which can’t be produced by labour.

And what I was particularly interested in, in my thesis, is that you’re ignoring environmental cost. So, the way that I consider this in my thesis is I think we should introduce some kind of cap on the emission of greenhouse gases, and then when you cap the emission rights for greenhouse gases these emission rights become a scarce resource that needs to be economized on. And I think the amount of emissions that the production of your product takes then has to be factored into the cost of that product as well.

[ATO] Let’s address that last example. And I’m going to put myself in the position of somebody defending, who’s advocating for, the Towards a New Socialism model without these modifications. And a person might say: okay, but given that the plan – and, I think it is clear but just for viewers, the plan that we’re talking about is a comprehensive plan for producing all the goods and services for the economy, so everything that appears in in the shops, what resources and inputs go to what production units, ‘projects’ are what they’re called in Towards a New Socialism.

So, surely, say, in the case of carbon emissions, we could decide what our limit for carbon is going to be for this year, in the next 10 years, and then we could just use that, set that as a constraint, and we just make sure that the plan doesn’t exceed that constraint.

To make this completely concrete, let’s say there’s one billion tonne CO2-equivalent that is allowed this year. And so as long as all of the production that happens this year does not produce more than one billion tonne CO2-equivalent it’s fine. So, what do you think are the limitations of that approach?

[PD]I mean, this absolutely works. And even in the original Towards a New Socialism model you can introduce a constraint on emissions and that will make sure that the optimized plan that you calculate in the end doesn’t violate these constraints, towards whatever environmental limits that you think are necessary.

The problem I think is that it will lead to an inadequate measure of the cost of individual products, which will then lead to an inadequate mix of products being produced, with an over-emphasis on products which actually take a lot of these scarce emission rights. Which then means other things can’t be produced anymore.

For example, the cost of fuel. Diesel fuel for cars, or petrol for cars, would be quite low or would represent maybe the labour cost of producing that fuel, but wouldn’t take into account that actually by burning this fuel you’re emitting quite a lot of greenhouse gases. That means that there would be no incentive to say okay well this is actually what’s using up all our emission rights right now, let’s drastically reduce that and maybe even increase production of other things instead which don’t take up that many emission rights.

So that’s the problem. The problem is not that you’ll get a plan that then emits too much CO2, the problem is that we’ll be using all of our scarce emission rights to produce things which maybe we shouldn’t be producing. And we should use these emission rights to produce a lot more of other things which only take a tiny fraction of those emission rights.

[ATO] Let me recap that, and then I’m going to ask you another question as devil’s advocate.

If I’m correct, what you’re saying is that there are really two problems here. There’s an absolute and there’s a relative problem. The absolute problem is: how in absolute terms can we make sure that we do not exceed one billion tonne CO2-equivalent? But there’s a second problem as well. And that problem is: given that we don’t want to exceed that limit, how are we going to distribute these resources the most effectively, the most efficiently, with the greatest social benefit (however you want to phrase it)?

[PD]Exactly. Once we’ve decided to limit carbon emissions, they are a scarce resource and we have to decide how we want to use those resources. And maybe using all of it on car fuel isn’t the best way to use these scarce resources. Maybe other things, where each unit takes less of these scarce emission rights, would be a much much better way to go about it.

And because this isn’t reflected in the labour value costing of these products. If you simply use labour values, I don’t think that’s the result you’ll get. I actually showed this in a computer simulation for a couple of small sample economies, and I got some results that I wasn’t expecting. But generally, what you got was that in the labour value model, what happens is that to stay within the emission constraint you simply produce less overall. Or sometimes, in some really strange cases, you even produce more of the environmentally destructive goods.

Excerpt from Optimal Planning with Consumer Feedback: A Simulation of a Socialist Economy, Philipp Dapprich, 2021. The red arrow shows a mere decrease in how much produced. The green arrow shows a decrease in how much produced and a shift in what is produced.

But in my [opportunity cost] model, generally what you got is a reduction in production to stay within the emission constraint, but you saw a shift. You saw more relative production of environmentally friendly goods, while you saw a significant reduction in the kinds of goods that were actually contributing to carbon emissions. It’s because this was now reflected in their price, and they’d only be produced as long as people would be willing to pay that higher price.

[ATO] We’re going to get to your model, and we’re going to get to your simulations that are very interesting and important.

I’d like to focus on the problem for a bit a little bit longer. To illustrate this issue of staying within the absolute limit but maybe not solving the relative problem. I mean, we could think about that in other cases which would be even more intuitive. I’m going to give a very stupid example, although it’s not really because it’s kind of how the world actually works at the moment.

So, let’s say we have a finite water supply or a finite supply of grain. You might say okay there are 300 million tonnes of grain. We just don’t have more, we don’t have enough arable land, or whatever. An economy could produce that and distribute that in two different ways. One way is that they could make sure that everybody has a nutritious supply of grain, and the other way is that all of the grain could go to one person and everybody else could starve. And neither of these use up more grain than there is.

I just wanted to make it really clear. I said that was a stupid example but actually that is really pretty much how the world works today. Actually, it’s one of the great problems of capitalism. So, we can call a system socialism but if we don’t overcome that problem in socialism then that the problem still exists.

Okay, we’ve made the point clear but I want to drill into that a little bit more. One more question. If somebody were advocating Towards a New Socialism as it is, they might say: what you’re saying is true, but what if we set a long-term plan for a gradual decrease of these carbon emissions? This year it’s one billion tonnes CO2 equivalent, but let’s say in 10 or 15 years it’s going to be zero. So, people are going to have to figure out how to stop using it, effectively. How do you respond to that?

[PD]Once you get to zero emissions – let’s say zero emission really means no emissions, and not that you’re offsetting it. There of course also these ideas that we’re always going to emit some CO2 but we’ll have to capture the same amount of CO2 back from the air. Let’s assume that’s not possible and you really have to have zero emissions.

At that point, you probably don’t need my modification anymore, at least to take into account carbon emissions and costing. Because you could simply ban any products or any production methods which emit CO2 outright. Or you’d have to because you can’t have any of it anymore anyways. And so at that point it probably wouldn’t make a difference anymore.

But even if that’s what we were going to do, and we don’t continue to have small emissions. Because there are some areas where it’s really hard to get rid of these emissions because there aren’t really any feasible alternatives. But even if we were to get to that, at some point, that’s still a few years or decades down the road. So, in the meantime, we will continue to emit some CO2 but we have to drastically reduce that. And then you precisely get these kinds of situations that I’m looking at, where you have a constraint on emissions you still have some emissions, and then the use of emission rights should be reflected in the cost of an item.

[ATO] There’s a trajectory, there’s a journey still, and we can’t say ‘well, we’re going to get there eventually, so it doesn’t matter how inefficient we are in the meantime’.

And the other thing, and this leads naturally into some more issues that you’re trying to address, is that carbon is not the only environmental cost. It’s actually only one type of environmental cost. It’s only one type of natural resource cost. And it’s only one type of cost. Like you were saying, there are issues about the opportunity cost of using capital goods, for instance, as well. So, even if we got to zero emissions, that wouldn’t take care of that. Do you want to just say something about that as well?

[PD] First of all, maybe about environmental constraints. Of course, you can generalize this towards other environmental constraints as well. So, we could as a society decide not just that we want to limit CO2 emissions, but we could also decide maybe we want to reduce the amount of land that is used for agriculture so that we have more land that can be used for natural reserves or something like that. And then you could also say, well no we’re not going to use all available land that we could potentially use for agriculture, we’re actually going to reduce the number of hectares that that can be used in the plan for growing food. And then you have to use the available land more efficiently and so on.

So, you could do this with any kind of environmental concern, or at least a lot of other environmental concerns as well; formulate them as some kind of constraint, introduce them to the optimization problem that’s solved when you calculate an optimal plan, and then in a similar way, in my model, this would be reflected in the cost of producing these items as well.

And then, of course, this this is a general approach. It doesn’t just apply to these additional environmental constraints that you might introduce, but it also affects other constraints which happen to be there in the economy. For example, if there’s limited number of a certain kind of machine available at a time, and it would take perhaps a long time to build up the stock of that of that type of machinery, then that is an effective constraint on the economy as well. And that would also be reflected in the cost of items. Do they need a scarce machine they have limited capacity to produce at the moment? Then they’ll probably have a higher opportunity cost. And this will then be reflected in adjusting the portions of goods that are being produced.

[ATO] Hello, this is After The Oligarchy. A quick message. If you’re enjoying this, please press the Like button – it makes a difference. Don’t forget to visit to see full transcripts of videos, plus other material. And on Twitter I’m @AfterOligarchy if you’re into that sort of thing. As we head back to the show now, let’s keep our heads cool and our critical thinking sharp. Now back …

[ATO] Let’s talk about this issue a bit more, so about the capital goods. This one is a bit different because there’s that embodied labour in capital goods and machinery. What we’re talking about with capital goods – people could think about machine in a production line, could think about a building – they are goods that are used to produce other things. They tend to be longer lasting.

In the case of carbon emissions, there’s no way really to interpret that in terms of labour. It’s just that there is a certain number of carbon emissions and we don’t want to go above that. How do we figure out how we use these?

But with capital there is a way to, at least partially, account for the cost to society of a capital good and that is how much labour went into (a) people actually labouring to assemble it and put it together, and (b) how much labour was put into the things of which the machine is made. So, if the machine, say, has three components – it’s got a belt, an engine, and an electronic interface – how much labour went into those things? And then you can go back, and back, and back, and back.

[PD]That’s precisely how you would calculate the labour value of these machines. And, as you say, there’s a case to be made that machines can be produced by labour. So, they are not as strictly limited as other things might be. We can produce more of them but they might still be limited in the short term. While in the long term you might be able to build up the stock of machinery, in the short term you might simply not have that many machines. And it might take some time to produce more.

And maybe this doesn’t just apply to individual machines but to capital stock in general. You can see – and, I mean, it’s often seen like this in the tradition of the labour theory of value – that capital is in a sense dead labour. This past labour that was put into machines, that was then accumulated and built up over time.

And there might be a limit to that as well, not just to individual machines but to capital stock in general. Because only the labour or the portion of the value produced by labour that is not immediately used up can even be accumulated as capital stock. So, there’s a limit to the capital stock that we have available at any time. And I think that should be reflected in costs as well because that puts a constraint on what we can produce.

[ATO] To develop on that a little bit more, on what this opportunity cost means. The way I think about it is, like you’re saying, firstly about time scales. We might say in 50 years, we might have twice as many, or three times as many, nuclear power stations than we have now. Or, like you’re saying, we might have expanded the total capital stock. It might be twice as big, but on the time scale of, say, this year, it’s essentially fixed. And so the question is, then, how do we distribute those capital goods such that they’re used the most effectively?

And opportunity cost is a way of basically saying how useful is each of these capital goods going to be so that we can make sure that they’re distributed in the right way. It’s quite a different way of thinking than the labour cost method, because that’s thinking about how do we optimize the amount of labour that went into these capital goods and that it’ll save in the future. Whereas this [opportunity cost method] is more about how capital goods are scarce, and given that they’re scarce and people want them for different things, how do we distribute them?

[PD]Exactly. An example that you might give, also to do with ecological transformation, is electricity generation. We have a certain amount of power plants running right now, that are available right now, and we now have to allocate the electricity that can be produced by the available power plants. We might be able to build more of them in the future, but it takes five years or even longer to build a nuclear power plant. So while this might be possible, it doesn’t really affect the available electricity that we have right now and that we need to allocate right now.

[ATO] Yes, exactly.

We’ve given the contours of the problem, we’ve talked about opportunity cost, and planning, and optimizing, and labour costs, so let’s look at what could be the solutions. You’ve talked a bit about that already. So how did you attempt to solve this problem?

[PD]The difficulty is finding a common denominator, a common unit in which to measure the opportunity cost. And there’s a way to do it which is known in the linear optimization literature, it’s called shadow pricing. And the basic idea of shadow pricing is that you make a slight change to one of the constraints of an optimization problem and then see how much does this change the value of the optimized objective function which you’re trying to maximize. That’s in very general terms.

The way that I applied this then to calculate the opportunity cost of consumer products – which is what I was mostly interested in – is I postulated, so to speak, that you get one unit of the product for free. So, let’s say we had one unit of bread for free. I do this by introducing a ‘free bread method’ which is a method of magically producing one unit of bread and without using any labour, any resources, any energy, and no emission rights, and so on. It just magically appears. This method can only be used once, so you can only produce one unit like this.

If I introduce this [production] method and we get this one unit for free, we can then use those resources which would otherwise have been needed to produce that unit of bread to produce other things. And that will lead to a slight increase in the overall production, and the value of the objective function. So, overall, you have a bit more because you get one unit of free.

But this is now measured in some other quantity. It could be could be a quantity of some other product, which I call the objective product. But generally you see this slight increase, and this increase is then used as the value of the product or the relative cost value of a product.

[ATO] Before we go into that I want to say that there’s this paper, ‘Optimal Planning with Consumer Feedback: a Simulation of a Socialist Economy’ [link] and that’s linked in the description. That goes through all of this in the full technical details. There will be people who will want to read through that, so that’s there for them.

If I can recap this, what you’re saying is that we are producing all of these different goods. And let’s say if we take one of them, if we take bread. What would happen if one unit of bread appeared without requiring any inputs?

[PD]Yes. I think one intuition that that some people might have – and I’ve had this question posed to me – at first is ‘well, isn’t what you have then simply one more unit of bread?’. The value of one unit of bread is one unit of bread, it doesn’t tell you a lot.

But you might say well we don’t actually need one more unit of bread. Maybe we want to use some of the resources, which we now don’t need to produce this unit of bread, to produce other things as well. So, you produce a bit more of everything basically with those resources. And this then increases the production of all things, at the proportions which are set by what we call the ‘plan target’. The plant target is what sets the proportions of various products.

[ATO] Can I come in on that just to clarify? What you’re really saying is you’re introducing one unit of bread for free, so that means there’s one unit of bread that you don’t have to make.

[PD] Yeah.

[ATO] And all the stuff that would have gone into producing that one unit of bread now can go to everything else. All of the flour, the electricity to heat up the ovens, the labour, and so forth. Now that we’ve gotten that free unit of bread, these resources can go elsewhere. And you’re seeing what is the effect of using those inputs elsewhere, is it better is it worse.

[PD] Yes. It will be better but better for sure, because you have additional resources available.

[ATO] Well, yes. [I meant relative to other products]

[PD]So you can produce more. The question is: how much more? And so, since we still assume that these proportions are fixed, you’ll produce a bit more of everything. And that means you’ll produce a bit more bread. You’ll have a bit more bread in the end, not a full unit, but some small fraction of it. But you’ll also have a bit more of everything else, and that increase can then be measured and used as the unit of value.

[ATO] You’re talking about optimizing, and you’re saying that whenever you’re optimizing, you’re always trying to maximize or minimize some objective function. To explain that, an objective function could be dollars, it could be labour hours, it could be megawatt-hours, it could be anything. In the case of your simulations and modelling, what is the objective function? How do you decide what the effect of this free unit of bread is?

[PD]I choose an arbitrary consumer product which I call the ‘objective product’. In some of the examples that I ran the simulation on this was grain. And I basically measure the output of grain that is being produced.

Then you could ask: well, why grain, why not any other unit? And the answer is it really doesn’t matter, because the proportions of these products are fixed. So, you’re producing, let’s say, seven unit of coal for every six units of grain. And then if you increase the production of grain, you also have to increase the production of coal, and so on. So, in the end it doesn’t matter which one you use.

You’re just trying to maximize the production of any one product. But because the proportions are fixed, that will also maximize the production of all the other products as well.

[ATO] It’s like there’s a recipe, and we know that in order to make one serving of falafel, you need to have one tin of chickpeas, one head of garlic, and two bundles of parsley. Then, if you’re going to increase the number of tins of chickpeas, well then the others will go up accordingly. It’s fixed.

[PD] Yes.

[ATO] And I suppose somebody might ask, why are these proportions fixed, surely they should vary? And I’m wondering is that just because this is a marginal unit? It’s just changing for one, so you can approximate it as fixed. Is that the answer?

[PD]Well, it is fixed only for the purpose of calculating one production plan. But then it gets continually adjusted. And that’s, I think, precisely the interesting thing here, that you have this automatic feedback loop. That then the proportions for the next planning period will get adjusted depending on the observed behaviour of consumers, and the observed demand of consumers. So, you fix it for one moment in time. But then you continually adjust it for future production periods.

[ATO] Let’s recap quickly, and then we’ll go to how this applies to environmental costs.

You’re saying we introduce our free unit of bread, we see how the inputs that would have gone into making that would be distributed to other products. And then we ask of our objective product – essentially any product, it could be grain, it could be iron – how much does that go up? And you’re saying, well why that? It’s because there’s a recipe. So, you can use any of these goods as a representative for how much production increases.

[PD]Yes, that’s exactly right.

[ATO] Okay. So, let’s look now at how you used this to incorporate environmental costs efficiently into the model.

[PD] You have a constraint on emissions. These emission rights are then scarce. And that means whenever you calculate the value of a product by pretending that you get one unit for free, that means you don’t have to use all the resources needed to produce it. And it also means you don’t need to use the emission rights needed to produce it. And these emission rights are then freed up to be used to produce other things as well.

The emission rights that get used up in the production of a product are now reflected in the cost as well. In the case of a product that uses up a lot of emission rights, when you get one unit of it for free you have loads of emission rights to work with. And you’ll be able to produce a lot more with that, and that’s why it will have a higher cost in the end. And this will then be reflected in how production gets adjusted and so on.

[ATO] You’re able to incorporate the environmental cost like you would how much flour it takes, or how much electricity it takes, to make the bread. We introduced the free unit of bread, and the question is really how efficient is it to use those carbon emissions – or to create those carbon emissions – in making a unit of bread? How much would production increase if they were used elsewhere? And so that’s the way we can actually figure out what is an efficient production plan with respect to distributing emissions.

[PD]In one sense, you already have an efficient production plan simply because you calculate an optimal plan. You can do that anyway, and even the labour value model does it.

I guess the sense of efficiency that you get in my model, that I think you don’t properly get in the labour value model, is an efficiency in terms of actually producing the things using limited emission rights which are actually needed the most or have the most benefit in some sense. That’s where you get the difference.

[ATO] Out of interest, if the constraint is set at one billion tonnes CO2 equivalent for that year, does that mean that the plan will always use up that much? Or could it actually use less using your method?

[PD]It could use less. There are constraints which are not effective constraints. This doesn’t only apply to emission rights. Let’s say we have a million hours of labour time available to produce things, but we only have a limited number of machines or raw materials. Then we won’t be able to actually use all of that labour, because we don’t have the tools, the machines, the raw materials, to do that. And in that case, labour wouldn’t be an effective constraint. You’d still introduce it as a constraint of the problem, but it wouldn’t really matter because you can’t use that much labour anyway, because you don’t have all the other ingredients needed.

And the same can be true for emission rights. It could be the case that maybe labour is the effective constraint on production. Or there could be multiple effective constraints on production, but maybe emission rights isn’t one of them and then you actually don’t use up all of these emission rights. And, in that case, you might as well not have introduced the constraint at all, because it’s not effective. It doesn’t do anything. I mean, you could still do it because you don’t necessarily know whether it’s going to be an effective constraint or not.

[ATO] In chemistry terms this is called the ‘limiting reagent’.


[ATO] Coming back to our falafel recipe, you could have 300 million tonnes of chickpeas but if you’ve only got two bundles of parsley …


[ATO] you’re only going to make one serving.

That’s for environmental costs, so let’s look at capital goods. How can the opportunity cost of capital goods be introduced to the Towards a New Socialism model along these lines?

[PD]This works in just the same way. If you have a product that uses a lot of scarce machinery or capital goods to produce, when you get one unit of it for free suddenly all that machine capacity that that would otherwise have been needed to produce that unit is free to produce other things. And that then means that you’d be able to produce a lot more with that than if you had another product for free which maybe takes the same amount of labour but doesn’t use as much machine capacity. That wouldn’t allow you to produce that much more, and that’s how you get the difference in cost in the final products then depending on how much capital goods are required to produce them.

[ATO] You’re looking at the capital goods that are used, say, to make bread, and that’s treated as another input. So, you might have flour, water, electricity, ovens. It might use three big industrial ovens, but it won’t use a synchronous generator. And so there’s a zero for that and there’s a three for the ovens. And so then it’s the same thing of ‘if the bread wasn’t made, how could those ovens be used?’.

In terms of the unique qualities, characteristics, of capital goods – and this applies to land as well – carbon emissions are a lot more transferable. Anybody can emit carbon emissions, anybody can use flour. However, a machine might be geographically fixed.
So, how does that come into it?

[PD]Yes. That’s probably what people are going to think now. How do you use an oven that can be used to make bread to produce laptops instead? It doesn’t really work that way. This would be relevant when you consider a bit longer time period than this. We have a choice, we can’t produce all the capital goods we would ever want. But we can now choose whether to produce more ovens, or more ion plantation devices needed to make computer chips.

You have a choice at that point. It then becomes relevant whether a product requires a lot of these kinds of capital goods or not. It always depends on the time period you’re looking at. And in the very short term you might not be able to affect the mix of capital goods you have at all, and it might simply be irrelevant. You have to work with what you’ve got and ovens can only produce bread. But, in a bit longer time period, we might have a choice of which kinds of capital goods to produce and then it does matter.

[ATO] We’ll leave it there. There is still a lot more to talk about. We haven’t really dug into your simulations yet. And there’s also the issue of multiple techniques, multiple production techniques. We haven’t really dealt with that at all yet.

That was brilliant I really, I really enjoyed that. Thank you for taking the time.

[PD]Yeah it was good. I was surprised that you really take the time to dig deep into it and make sure people understand. That’s a lot more detail than … I’ve done similar interviews before and I don’t think we’ve gone into that deep into a particular point or anything like that. So that’s really good.

[ATO] That’s great, I appreciate that. That’s exactly what I try to do. I try to make a balance. But basically, with this YouTube channel and the blog that is exactly what I’m trying to do. Effectively, if we’re really going to do this, if we’re really going to change the world – and I know that this is how you feel and think because otherwise you wouldn’t be doing this – we have to have something that will work and that you can actually apply to the real world beyond some generalities. And that means that we need to go into these details.

Yeah, I mean you’re a bit of a strange philosopher. I know you were saying it was multi-disciplinary but I never expected somebody to mention a proportional controller in a philosophy thesis. Because that’s my background, electrical engineering. And I don’t know many philosophers you can use lp_solve.

[PD] I started off studying physics and philosophy in parallel, so that’s why I have the mathematical background. And originally my interest in philosophy was more in philosophy of science. I wrote my Master’s thesis in philosophy on the philosophy of quantum physics, so something completely unrelated.

But I was always a communist and politically interested. And I thought for my PhD I wanted to do something more relevant, but where I could also use my particular skills. I knew I wanted to go into philosophy and one of the reasons is that I just have such broad interests and philosophy is one discipline where you can you can do anything really. You can do the philosophy of economics, or political philosophy, you can also do philosophy of science.

And so I knew this would be, by the nature of it, an interdisciplinary project. And I got Paul [Cockshott] involved, he’s a computer scientist, of course. And he was the primary intellectual inspiration for the project as well. But then because I knew I wanted to do it in philosophy, we got a philosopher to be the primary supervisor. Paul was the co-supervisor. And then, I don’t know how much of the thesis you read, I mean a lot of it is standard political philosophy as well. I start off with very basic philosophical questions in terms of how to distribute things and the nature of rational choice, and these kinds of things. So, I try to ground it in in philosophy, but then I also get to these more technical things.

[ATO] No I like that. I think it’s a really nice combination. Because I probably have very similar interests to you. I’m very interested in the philosophy of science in particular. And you know there’s that saying of Richard Feynman that ‘the philosophy of science is as useful to scientists as ornithology is to birds.’ Well, I don’t agree with him on that. I don’t agree with him, particularly not now, where … anyway we can maybe talk about that again.

[PD]A lot a lot of physicists think like that. So, when I when I talk to other people in physics about philosophy of science they’re always like ‘we don’t need philosophers to tell us how to do science’. But, I mean, clearly there are ways of doing it wrong and there are ways of doing it right. Theorising about that is leaving the realm of science, in a sense, because you’re not doing science anymore, you’re talking on a meta level about science. And then you’re basically doing philosophy at that point.

[ATO] Yes, definitely. But I think there could be a lot more of that general discussion, which is necessary and important, and then going to ‘okay, but also we live in a real world, and we want to do things.’ I mean, in terms of philosophy I’m a pragmatist. As in John Dewey, and so on. So, ultimately, everything for me comes back to ‘what are we going to do?’. And so I thought that was great, you know, ‘yeah this is this is what Robert Nozick said, and so forth, and here are some critiques, but anyway so I was using lp_solve basically I figured out how to optimize the production plan.’ And I think there should be more of that.

All right, look, I’ll let you go. It’s great to talk to you, and we’ll talk again next week.

[PD]All right, I’ll see you next week then, bye.

[ATO] Thank you for watching.

If you got anything from this video, then please press the Like button, consider Subscribing, and, if you really enjoyed it, then comment your favourite ice cream, pizza topping, or whatever other generic, boring, unrelated, nonsense will appease the Algorithm.

There’s a lot more to come. We’ll keep exploring better futures for humanity until we get there.

And as always, I want to read your thoughts in the Comments section below. This channel has a wonderful audience, and there are usually some very interesting comments under the video, so let’s continue that.

That’s all for now. Our democratic future lies After The Oligarchy.


Dapprich, Jan Philipp (2021). Optimal Planning with Consumer Feedback: A Simulation of a Socialist Economy.

Dapprich, Jan Philipp (2020). Rationality and Distribution in the Socialist Economy. PhD thesis.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s