Neo-Ricardian factor shares

Note:   I want you all to be highly critical of my posts on factor shares – and where you can throw literature at me.  I wrote a bunch of posts in a single day based on one book (and some prior knowledge), I have no appeal to authority here and would love to have your ideas thrown in there 🙂

Last time I said I thought Neo-Ricardian was Post-Keynesian – it seems like they have similar authors, but the essays focus on different “important elements”.  Furthermore, the direct issues with “neo-classical factor models” are indeed clearer here!  However, the focus seems much more on short-term, and in some ways monetary policy, issues rather than long-term trends in factor shares.  That is ok, but it is important to note that some of the Post-Keynesian models in these chapters are discussing different questions than the questions the earlier part of the book was based on.

Note:  The author cites Roncaglia (1978) as discussion on why research should move from marginalist to a classical surplus approach again – haven’t read it, but hope to once I have proper internet access.

As a starting point, interest rate and the rate of profit will be used interchangably.  It starts off by noting Sraffa discussing capital, and downward sloping factor demand.  Marginalists tend to presume that a higher price (interest rate) leads to higher capital utilisation.  However, capital as an input is related to other inputs AND is a composite good (made up of many types of capital) – implying there is no clear correspondence.  Eg, a lower real wage for labour may lead to more capital use due to higher profits (income effect dominating substitution effect), or the relationship between forms of production may shift discontinuously due to the complicated relationship between many inputs.  This is all very true.

Note:  Mas Colell discusses it post this essay here.  It also bears similarities to the discussion of “demand” in general equilibrium models.

The author then goes on to state that this undermines the traditional idea of an automatic mechanism for full employment, namely the falling capital-labour ratio.  I accept the criticisms – but lets be a bit careful, they are an attack on the problem of aggregation, not the mechanism of adjustment per se.  We can’t really criticise aggregates as unrealistic and misleading, and then use an argument on the basis of an aggregate, unless neo-classical economists actually believe this is the way the ‘market clears’.  Note:  This moves into a discussion of multiple eqm.  Yes, multiple eqm, sun-spots, etc, are important things to study and to try and understand – but they generalise neo-classical theory rather than invalidate it from what I can tell.

The author then says that marginalists and classicalists have fundamentally different “visions” of the economy:  a production function vs a circular flow.  I have to be honest that this baffles me, they are both representations of the same thing, one in a static stock basis and one in a dynamic flow basis.  We really need a consideration of both to build a representative model (either implicit or explicit), and they both have the same spectacular flaw (which in actual work both sides will likely clear up), in the oversimplified way they do not have individuals making choices.  I feel that neo-classicals are being a bit strawmaned here.

Note: Another point, Sraffa’s response to “the problems of aggregation” is too … use a different type of aggregation based on idealised types of classes which don’t formally exist and are practically immesurable.  Don’t get me wrong, the characterisation may well be useful, but it is just a bit interesting to do that after criticising other economists on the basis of aggregation issues.

Sraffa defines a wage-profit curve f(w,r) where w is wages, and is bounded at W (the wage with zero profits), and r is profits and is bounded at R (the profit with zero wages).  The author then spends the next page attacking simultaneous determination of output-price relationships and supply and demand, which supposedly only function outside of normal circumstances – I think we need to be a bit more careful here, this attack is more relevant for transition paths than it is for steady states and much of the work being quoted is steady state type work.

Given this wage-profit curve, which is downward sloping, the author says we have once again discovered class exploitation and conflict.  I find this problematic.  What sort of outcome holds with zero wages?  What sort of outcome holds with zero profit?  These are functional parts of the curve, and illustrates to me that the curves relationship seems circumspect.

Still taking that curve, the author suggests we might be able to close the model by picking the subsistence wage.  While this was common for classical economists, our ideas of wage determination have improved significantly since then, so an exogenous wage set at a subsistence level doesn’t seem like an appropriate closure.  The author then states we need to think of other ways to determine this and get our income distribution.

Technological-Normative

Taking last times discussion on Pasinetti, gives the Kaldor-Pasinetti theorem, which was that a given rate of accumulation in the long term and the savings rate of workers and capitalists we will have a level of profit (and growing output).

We discussed this last week as an accounting identity, namely the Pasinetti result where P/K was a positive function of the savings rate of capitalists and the flow of investment relative to capital.  But like a lot of this, accounting relationships are not behavioural, there is a lot of talk about animal spirits and natural forces as “shifting” variables – that is fine – but these relationships are not choices of firms or workers.

The author states a similar view to mine above with regards to Pasinetti’s derivation of “natural” profit levels:

But again nothing is said about the existence of forces that automatically insure the attainment of such an “equilibrium” value in an actual market economy: we are still confronted with a normative notion, not with a positive explanation of income distribution in actual economic systems. (Let us note that a normative analysis is by itself insufficient to specify the direction of cause-and-effect relationships).

As a result, this section appears to imply that the prior Post-Keynesian discussion was on issues of interest, defining stylized facts and relationships that will hold.  However, the ability to discuss the true impact on the distribution of income is “more complicated” – a position I can appreciate.

The monetary rate of interest and the rate of profits

This section I am leaving to the side – it is interesting, but it is an issue that may lead to more heat than light if I describe the short section written in the book!  I will quote the first paragraph as a note of the direcion it is going:

According to classical economists, it is the size of the surplus accruing to capitalists, and hence rate of profits, which determines the pace of capital accumulation and hence the rate of growth in the economy [Note:  For Post-Keynesians it is the rate of capital accumulation that determines profits].  However, the development of a full-fledged structure of financial intermediaries in the capitalist economies may cast doubt on the relevance of the classical causal chain to contemporary conditions.  Recourse to finance may provide entrepreneurs with command over resources in a way that does not automatically correspond to ownership deriving from past income, and hence influenced from past income distribution.

Conclusion

The piece concludes by stating that a sole broad model of income distribution and its “optimality” is unlikely.  As a result, it is better to investigage and discuss individual policies on their own merit.  I certainly have to agree with this.  They then say we should obviously reject marginalism and use a surplus approach in the Neo-Ricardian fashion – an argument I certainly haven’t been persuaded on 😉

I get the impression that there is an insurmountable gap here between Neo-Ricardian and Mainstream Economics – or at least was in the 1980s.  We both accept issues with heterogeniety, aggregation, and the importance of looking at small conditional questions – this is excellent.  Furthermore, we both accept that data needs to be the arbitrer of competing theories – superb!

However, can the data we have actually close the gap between the two ways of conceptualising macroeconomic outcomes? No – theories drive the way we interpret data, and some questions will not get a definitive answer one way or the other.

My solution?  Keep having people do both, create arguments, share and critique them.  Insure that analysis gives a transparent account of assumptions around human choice, and that the trade-offs involved are clear.  Not much about “solutions for factor shares” in this post – and I think that is a positive thing, ultimately the key point from Neo-Ricardian’s is to say “things be complicated”.