Is Demand for Investment Goods a Derived Demand in Marx?

It is well known that for Keynes the demand for investment goods, as for labor services, is a derived demand. The demand for investment goods ultimately depends on the extent to which they are needed to produce items of consumption. Certainly, one investment good may be required in the production of a second investment good which, in turn, is needed to produce the first investment good. Iron gets sold to steelmakers who sell some steel to iron makers in a circle that never makes direct contact with the production of consumption goods. But this occurs because such maintenance of iron and steel works enables, indirectly, the supply of investment goods for the production of items of consumption.

In Marx’s framework, the need for investment goods ultimately to have a connection to consumption-goods production is not immediately obvious. In his analysis, the motive of capitalist production is surplus value. If surplus value could be created by producing iron to produce steel to produce iron without any connection, even indirectly, to items of consumption, the uselessness of the activity would have no bearing on whether capitalists deemed the activity worth pursuing. This raises the possibility that the demand for investment goods need not always be a derived demand in Marx’s framework.

Against this, though, is Marx’s proviso that a commodity cannot have a value (including surplus value) without a use value. This raises the question of whether iron-steel-iron production, to the extent it is utterly superfluous to consumption, could have a use value.

It might be tempting to say that, from the perspective of capitalists, the quality of creating surplus value is what ensures the use value of such production. But this would presuppose that such production could create surplus value, and that is the question to be resolved.

There seems to be an ambiguity. It would be helpful if Marx somewhere along the line indicated his own view on the matter. He appears to have done so in a passage in the third volume of Capital, chapter 18:

[A]s we have seen (Book 2, Part III), continuous circulation takes place between constant capital and constant capital (even regardless of accelerated accumulation). It is at first independent of individual consumption because it never enters the latter. But this consumption definitely limits it nevertheless, since constant capital is never produced for its own sake but solely because more of it is needed in spheres of production whose products go into individual consumption.

It seems, at least on the basis of the above, that Marx considered the use value of investment goods (constant capital) to be contingent on the contribution made, either directly or indirectly, to the production of items of consumption. If this were not the case, it would be possible for the production of investment goods to occur profitably even when it had no connection whatsoever, even indirectly, to consumption. And, if so, there would be no need for any such connection from the perspective of capitalists.

If the above passage is an accurate reflection of Marx’s position, it seems compatible with Keynes and Kalecki on demand deficiency.

For Keynes, a reduction in the average propensity to consume results in negative income adjustments unless private investment, government expenditure and export demand happen to increase sufficiently to maintain overall demand and income at the same level. He maintained, of course, that there was no automatic tendency for this to occur.

For Kalecki, in the normal situation of excess capacity and unemployment, weak capitalist consumption relative to investment can contribute to a fall in the rate of profit. One way of viewing his ‘tragedy of investment’ is that investment adds not only to profit but to the stock of fixed capital whenever net investment is positive. As a result, profit as a proportion of fixed capital (the realized rate of profit) can fall due to a reduction in capacity utilization. The effect can be offset by an increase in capitalist consumption, the budget deficit or net exports, or by a reduction in worker saving.

Related Posts

A discussion of demand deficiency in relation to Marx is provided in:

Taking Demand Seriously

On the lack of an automatic tendency to full employment in Keynes’ framework, see:

Planned Investment/Saving and Keynesian Causation

Kalecki’s profit equation is discussed further in:

Thinking in a Macro Way


4 thoughts on “Is Demand for Investment Goods a Derived Demand in Marx?

  1. If steel is used to make tools for the iron industry these tools ARE use values and they HAVE a use for the iron industry. Use-value just refers to the specific use of a commodity. If someone is buying it, regardless if it is for personal consumption or productive consumption, than we can assume it has a use. But use value isn’t a transitive property that chains its way down to a final consumption good. The use value of a hammer is that it drives in nails. It is not the use-value of the home that the carpenter has built with these nails.

    In my understanding of the argument in Marx about investment demand, the argument is not that, say, steel production, as a whole, can be completely unrelated to any final consumer goods. The argument is, rather, that in order for the quantity means of production to increase, and thus for the economy to grow, the production of means of production must grow at the expense of the production of consumptions goods. This means that the growth of investment demand will rise faster than the growth in consumption demand. The empirical fact that investment demand can grow faster than consumer demand shows that not all investment demand resolves to consumer demand. If the proportion of trade between iron and steel industries grows faster than the proportion going to consumption goods then part of the social product is not resolving itself to human consumption. This phenomenon does not require or imply that total iron and total steel production happen with no regard to human consumption. It just says that a portion of this production happens without regard to final consumption.

  2. Thanks for your thoughts. I’m not entirely sure what I make of the matter, and continue to think about it, but I find it hard to square your interpretation of Marx’s argument with the content of the quoted passage (“constant capital is never produced for its own sake but solely because more of it is needed in spheres of production whose products go into individual consumption”). That seems crystal clear to me. It says production of investment goods is “never” for its own sake but “solely” due to a connection (direct or indirect) to consumption. I don’t see how it could be stated any more clearly.

    Certainly investment demand can grow faster than consumption demand. That will mean that relatively more resources are being employed to increase future productive capacity. In my reading of the quoted passage, the addition to future productive capacity will be occurring because it directly or indirectly contributes to the production of future consumption items. Some iron-steel-iron production will never directly resolve itself into consumption, and remains in its own closed loop, but it occurs (it seems from the quoted passage) in order to maintain iron and steel production that does ultimately enable production of consumption items.

    As an aside, an increase in investment-goods production does not require a reduction in consumption goods whenever there is unemployment. So, such a tradeoff is typically not operative under capitalist conditions. This would be more applicable to the old Soviet-style economies which were supply but not demand constrained.

    In any case, it is only one passage. If another passage can be cited from Marx’s work that directly and clearly contradicts the passage quoted above, that would carry equal weight in deciphering Marx’s position.

    Maybe, as you suggest, it is incorrect to link the requirement of a use value to the statement that constant capital is “never produced for its own sake” but “solely” due to a connection to the production of consumption items. Perhaps there is another reason for the suggested connection, if in fact it is suggested or intended to be suggested. But in the sense you are using the term ‘use value’, pretty much anything could be said to have a use value. And if that’s the case, why would Marx bother to make the distinction that to have value a commodity must have a use value?

    Note that when you write of a commodity, “If someone is buying it … we can assume it has a use”, this does not resolve whether the commodity was bought despite having no direct or indirect connection to consumption or because it did have such a connection. This is what I had in mind when I wrote of a commodity with no direct or indirect connection to consumption that “It might be tempting to say that, from the perspective of capitalists, the quality of creating surplus value is what ensures the use value of such production. But this would presuppose that such production could create surplus value, and that is the question to be resolved.”

  3. I don’t think the use-value discussion is relevant to the the question of closed-loops in Department One. Again, I think use-value is very simple: if someone has a use for something, whether for productive consumption or personal consumption, than that something has a use. I don’t think there is anything problematic about this. When you say “why would Marx bother to make the distinction that to have value a commodity must have a use value?” I would answer that Marx makes the distinction because things that aren’t useful aren’t sold on the market and thus do not become commodities. I don’t follow you here: “It might be tempting to say that, from the perspective of capitalists, the quality of creating surplus value is what ensures the use value of such production. But this would presuppose that such production could create surplus value, and that is the question to be resolved.” Constant capital doesn’t create surplus value, only labor power does. If workers are being exploited and the products of their labor are being sold then the capitalist gets surplus value regardless of the final destination of the product. The only matter that “ensures the use value” of a commodity is whether or not the purchaser deems the commodity useful for some purpose. The use-value of steel is not the creation of SV but the creation of objects made from steel.

    In regards to the quote from Marx: I am intrigued by it. I hadn’t noticed it before. I’m not particularly good at philological debates about Marx so I can’t assault you with counter quotes and contextualizations. It seems to me that the issue comes down to what we are trying to prove, specifically.

    in Vol. 2 Marx is arguing against Adam Smith’s trickle-down theory of distribution in which all production reduces itself to wages paid to workers. Buy showing the closed loop in D1 Marx is able to show that in fact part of the social product never reduces to wages.

    This argument is then wielded by those who critique underconsumption theory to argue that investment demand is not entirely dependent on consumption demand and that, therefore, a theory of crisis cannot be based on the gap between wages and the total social product.

    Neither of these arguments claim that department one has no relation to department two. I think we would make more progress in this discussion if we were more precise about what we meant by “direct” and “indirect”. If part of the value produced in department one stays in D1 then this production is production purely for D1. But not all of D1 production is for D1 and D1 could not exist without D2. Does this make production within the closed loop “indirect”? I don’t know, it depends on how we define the term. What seems more important is to be clear about what phenomenon we are trying to describe, what questions we are trying to answer.

    Can you expand on your point about unemployment effecting the tradeoff between investment demand and consumption? I don’t understand.

  4. Yes, I can see there is some misunderstanding going both ways. Talking across paradigms can be confusing because of the different terminology, but I think it is worth it in the case of Marx and Kalecki/Keynes. I don’t think their perspectives are the same, but I do think they are compatible in important respects, and possibly more so than is often thought.

    To clarify what I have in mind with the direct/indirect aspect, perhaps it could be spelt out like this. Consider two broad types of production chains, A and B:

    A. Dept 1 -> … -> Dept 1. These production chains form closed loops (e.g. iron-steel-iron) and never leave department 1. The dots indicate that there could be numerous steps in the chain.
    B. Dept 1 -> … -> Dept 2. These production chains can have numerous steps in Dept 1 but ultimately end in Dept 2.

    Marx’s passage seems to suggest that production chains A will exist to the extent that they are socially necessary to meet the requirements of production chains B. Additionally, production chains B will not contain steps that are superfluous to meeting the requirements of Dept 2.

    So Marx’s passage seems to imply that all production is justified by its connection (in some way) to individual consumption. However, in my reading of Marx’s passage – but also of Kalecki and Keynes – this does not mean that an increase in the gap between wages and income necessarily leads to demand deficiency. This is because:

    1. Worker consumption might be maintained through private credit creation.
    2. Capitalist consumption, which comes out of profits, could conceivably rise to offset the reduced consumption of workers.
    3. Even if consumption falls, this might be offset or more than offset by an increase in investment (or the budget deficit or net exports).

    1, in all likelihood, only postpones a collapse in worker consumption until workers find themselves too deeply in debt.

    2 is unlikely, as I think Marx, Kalecki and Keynes would all agree. Capitalists don’t consume as high a proportion of their income as workers. In Keynes’ terms, the average propensity to consume for the society as a whole is likely to fall.

    3 is possible, but there is no automatic tendency for it to occur on the basis of private demand (Say’s Law is not operative in the view of Marx, Kalecki or Keynes). The government could choose to maintain demand, but typically doesn’t do so.

    So, even though it is not a necessary outcome of a fall in wages, in practice relative impoverishment of workers might lead in some cases to demand deficiency (Kalecki, Keynes) or falling profitability (Marx, Kalecki).

    In Kalecki, for example, weak consumption demand by capitalists or increased saving by workers both impact negatively on realized profit and result in lower capacity utilization. This may deter investment. There is less need for capacity-enhancing investment when utilization of existing capacity is low. The likely result is a contraction in output and employment or, in Marx’s terms, a reduced level of value production.

    In Marx’s framework, when the level of value production is taken as given, a fall in wages will clearly boost surplus value. More surplus value can be produced, given the rate of exploitation. But the level of value production is unlikely to remain unaffected by a change in wages. Assuming only surplus value that can be realized will be produced, lower wages and worker consumption imply a lower level of value production unless capitalist consumption or investment offset the effect. If, as Marx’s passage seems to suggest, investment is ultimately connected to individual consumption, the investment might not be forthcoming unless capitalists decide to step up their consumption. More circuitous production chains A cannot arise simply for the purpose of filling the gap. They need to be socially necessary to meet the requirements of individual consumption (if I am interpreting Marx’s passage correctly and this meaning was actually intended by him).

    Some clarifications:
    — “I can’t assault you with counter quotes and contextualizations.” No worries. That makes two of us. (I just happened to notice the passage while reading about something else.)
    — Agreed that constant capital is not productive of surplus value. I just meant that the production processes involved in a closed loop (in Dept 1) will be productive of surplus value to the extent they employ workers provided the closed loop is socially necessary to meet the requirements of individual consumption (assuming, again, that my reading of Marx’s passage is appropriate).
    — My intended point on use value was that although a commodity being bought does indicate a use value, this doesn’t tell us whether the commodity has a use value in spite of having no connection to individual consumption or because it does have such a connection. But I am happy to leave use value out of the discussion as you may be correct about its lack of relevance.
    — When there is unemployment and idle resources, it is possible to increase the production of both investment goods and consumption goods by employing extra workers and other resources in both types of production.

    This last point comes back to the question of what level of value production, in Marx, is to be considered. I don’t think there is any tendency in Marx toward full employment of workers and other resources because he rejected Say’s Law. But if we take the level of value production to be below full employment, the cause of this can be seen (equivalently, in my opinion) as demand deficiency or lack of (realizable) profitability. In Kalecki’s framework, they are one and the same thing, since realized profit is the sum of capitalist expenditure, the budget deficit and net exports, minus worker saving.

    Although Kalecki’s profit equation is in terms of realized profit and Marx’s falling rate of profit is in terms of produced surplus value, if we make the assumption that all produced surplus value is realized, Kalecki’s profit expression corresponds to the amount of surplus value created in production. The tendency of a falling profit rate shows up in Kalecki’s equation in the effect of investment on both profit and the capital stock. Investment adds to both and therefore tends to reduce the rate of profit when net investment is positive unless other components of demand also increase to maintain the degree of capacity utilization.

    If we drop the assumption that all produced surplus value is realized, this only makes Kalecki’s equation more pessimistic from the perspective of capitalists. The realized profit rate will be lower than the produced profit rate. There is still a falling rate of profit tendency in Kalecki. But it is really the other side of the coin from demand deficiency.

Comments are closed.