Capital Theory: An Austrian-Marxian Synthesis

Readers of this blog will know that I am not generally very sympathetic to Austrian economics. There is one point on which the early Austrians did contribute an interesting idea to the world of economics: namely, their theory of capital. This does not mean that the Austrian theory of capital is valid — as I shall show in a moment it is deeply flawed — but there is an idea that can be salvaged from the wreckage that is Austrian capital theory.

To my mind there are only two coherent theories of capital — and the mainstream possess neither. The first is what might be called the Ricardian or the Marxian. This is what might be called the ‘labour theory of capital’. The idea is that capital is effectively embodied labour or, to use Marx’s colourful phrase, ‘dead labour’. Marx unfortunately contaminated this concept with moral judgements, as he so typically did. In Das Kapital he wrote:

Capital is dead labour, that, vampire-like, only lives by sucking living labour, and lives the more, the more labour it sucks. The time during which the labourer works, is the time during which the capitalist consumes the labour-power he has purchased of him.

But it is perfectly possible to extract the interesting point being made here: capital is accumulated labour. That is, when a machine is built it is built using human labour and in that regard it ‘stores up’ this labour. To the extent that it is built using previously accumulated capital, it is also effectively using previously accumulated labour time.

The second approach to capital is the Austrian one. In the Austrian theory capital is effectively embodied time, in the sense that it is time spent on the production of one good or service rather than another. G.L.S. Shackle provides an interesting gloss on this in his book Economics for Pleasure. He writes:

[Böhm-Bawerk] gathered ‘produced means of production’ of every kind under the heading of capital. And capital, he said in effect, is the visible symptom of the part played in the productive process by the lapse of time between the putting-in of services of labour and land, the ‘original means of production’, and the enjoyment of the fruits of that process at a later date. How can we say that ‘time’ is productive? Because given quantities of human effort and of ‘land’ can yield a larger quantity or better quality of product if we are willing to wait longer for it. (pp212-213 — Emphasis Original)

Do you see the slight inversion taking place here? What Böhm-Bawerk did was to change the emphasis. Rather than saying that capital was embodied labour, he said that it was embodied labour time that could have been spent on something else. The emphasis was laid less on the ‘labour’ and more on the ‘time’.

The idea here is that we as a society could work to produce consumption goods in lesser quantity/quality now or we could use our efforts to produce investment goods that will in turn produce consumption goods in greater quantity/quality in the future. While this is not a bad way of looking at the problem it quickly runs into problems when Böhm-Bawerk and the Austrians try to turn it into a theory of the interest rate.

This was because they were pre-Keynesian and the Austrians did not understand that real capital — machines etc. — must be firmly distinguished from financial capital and that the market for the latter operated in an unusual way. They assumed, implicitly, some sort of perfect knowledge on the part of the market. So, left to itself the interest rate would tend to equality with the profit rate on investment goods. Thus, the rate of profit would come to represent the ‘reward for waiting’, as Alfred Marshall would put it. Of course, after Keynes we came to know that what determined the interest rate was actually the liquidity preference of the market.

This is because financial actors — that is, savers and their investment managers — are faced with an uncertain future. They thus evaluate various investments not so much by their real return but rather by their prospective liquidity. When people feel pessimistic about the future they increase their holding of liquid assets, causing interest rates to rise, while if they feel optimistic about the future they increase their holding of non-liquid assets, causing interest rates to fall.

Another way to think about this was that the Austrians and the other marginalists confused stocks with flows. They assumed that the interest rate was reflective of savings flows. So, savings would flow into a pool of investment at a given rate. This rate of savings would determine the interest rate which would in turn determine the profit rate and thus the investment rate. But the way this really works is that there is already in existence a big pool of accumulated savings. There is also the ability for financial investors to borrow and thus create additional financial savings. When sentiments change this pool of investments may suddenly shift. For example, if people become pessimistic more people will hold more liquid assets and interest rates on everything else will rise. The financial markets are thus a bit like that scene in Alice in Wonderland where the Mad Hatter tells everyone at the tea party to change places.

There is also a disconnect between the interest rate and investment. The Austrians and the other marginalists assume that there is some sort of linear relationship here. But there is not. The rate of investment in real capital is dependent on the state of confidence. This, in turn, is mostly dependent on the level of effective demand that is present in the economy at any given moment in time. You can see that there is no relationship in the graphs below which plot the rate of investment against the real interest rate.

Lagging the investment rate by a quarter to allow for time for the interest rate to transmit to increased investment makes basically no difference. The R-squared remains basically the same. This can be seen intuitively from the chart anyway.

If we detach Austrian capital theory from the Austrian theory of the interest rate, however, we get something far more usable. We can also easily couple Austrian capital theory with Ricardian/Marxian capital theory. Once an economy reaches full employment — which we need not assume as some sort of ridiculous teleological end-point — the main constraint on production is labour. We then recognise that the question becomes for society where they want to allocate its labour time. Here we can emphasise either the Austrian time component or the Ricardian/Marxian labour component. It really makes no difference. If the amount of labour time is constrained then we have to decide how much is channeled into consumption and how much is channeled into investment for future consumption.

There is one more point, however, that needs to be stressed and which is not contained in either theory: namely, that of technology. How do we conceive of technology in this framework? We cannot, after all, consider it like land; an ‘original’ or ‘given’ means of production. Rather we must see it as a product of a type of labour; namely, intellectual labour.

Now this where another approach might be interesting to supplement the above mentioned ones. This is actually a neo-Marxian approach that has recently been developed in philosophy by the French philosopher Bernard Stiegler in his three volume work Technics and Time. Stiegler draws on the work of the phenomenological philosophers to argue that technology is actually a physical embodiment of human memory. Think of it this way: all technology is the embodiment of human intellectual labour. In other words: all technology is the embodiment of human thought. Technology is then a sort of ‘deposit’ of human thought. In this sense it is very similar to memory.

Thought of in this sense capital becomes a number of things. It becomes, first and foremost, embodied labour time — whether we want to emphasise the time aspect or the labour aspect is completely arbitrary. But it also becomes embodied memory or embodied human knowledge.

Ultimately though these are all just metaphors. And really capital theory is a very secondary, perhaps even tertiary, part of macroeconomic theory. We really cannot say a great deal about it. We can tell stories and create metaphors but beyond that we can say little else. Trying to understand the interest rate, for example, in terms of the theory of capital is a total dead-end. But also trying to come up with some ‘objective’ theory of distribution as Marx did is also a dead-end.

This shows us something interesting: any economic theory that gets overly bogged down in the theory of capital is probably not a very interesting or useful economic theory. Thankfully, since the 1970s Post-Keynesians have moved away from capital theory. Indeed, Joan Robinson said it was a dead-end after the capital debates in the 1970s — something that Nicholas Kaldor recognised in the 1960s while everyone else was obsessed.

The Austrians and the Marxians are still obsessed. But all they are doing is weaving ideological narratives out of metaphors and in doing so thinking they are doing something scientific. What they are really doing is engaging in storytelling that justifies their a priori political belief systems. As we have seen above we can tell very nice stories and deploy very nice metaphors. But they are just stories and metaphors.

Interesting work is being done on capital accumulation, however, by the Schumpeterian school. The most popular of these is Mariana Mazzucato. They eschew the model-oriented approach and instead go out and look at how technology and knowledge become embodied empirically. Their results are very interesting. In a later volume of Review of Keynesian Economics (ROKE) I will be publishing a review of Mazzucato’s book (see early draft here) where I tie what they are doing back to Post-Keynesian economics using the old, and still very flexible, Harrod-Domar growth model.