Marshall’s 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 🙂

Now here is an area I don’t know much about.  Alfred Marshall is an amazing economist, Keynes said this about him – yes Keynes – so obviously he has a lot of respect.  He even has the typical microeconomics demand curve named after him (Marshallian demand).  However, it appears he had a distinct model of income distribution – one that was classicalist in its view of rent and factor supply (and its interest in secular changes such as population growth), but marginalist when it came to discussing demand for factors of production.  The model discussed in the essay comes solely from Principles, rather than other work by Marshall.

The description of the elements in Marshall’s theory, from the essay, are as follows:

1. Marshall’s description of factor demand was strongly marginalist, and clearly pointed to the idea of an aggregate production function – one that produced something that would be considered as real net national product in modern times (what he terms National Dividend).
2. Rent appears to be treated differently, and in some ways more modernly, than it had been previously.  The heterogeniety of land (such that the average cost is less than the marginal cost/price received) remains an important issue.  However, Marshall also discusses rents accruing to skilled workers – note that prior classical analysis (including Marx’s analysis) tended to directly assume that wage differentials were the result of differences in labour productivity.  Here Marshall assumes that there are genetic differences in skill (more heterogeniety), and that individuals talents are completely unobservable prior to choosing (and being stuck in) an occupation.  In this case, there is a bias towards occupations that reward skill – but given that the payment of a wage is still based on observed productivity of the individual (such that choosing to go for the occupation that reward skill becomes a “risky choice” since those without talent will end up with lower wages than the less variable alternative occupation), the allocation is still efficient.  It would be interesting to read Marshall on this, and compare his discussion with that of Robert Frank on “superstar” jobs (link).
3. The supply of capital is the standard stock, with an investment flow that comes from savings – which is determined by the interest rate (which is where the demand for capital comes in).
4. As was touched on in rent, it is Marshall’s exposition of labour supply that adds the most. Here are the listed assumptions in the essay (Note, many of these are testable nowadays, and they actually touch on some big questions that exist in labour and welfare economics):
   1. The family is the unit for labour supply decisions (not the individual)
   2. The occupation choice is made by parents on behalf of their children, and is largely irreversible.
   3. Entry into an occupation requires “fixed costs” of investment in human capital – this varies by occupation.
   4. Parents (as the occupation choosers and trainers) and employers (as trainers) have no claim on the extra earnings additional human capital would provide the worker.
   5. The occupation choice is a function of wages, working conditions, hours, job security, tenure, and cost of training.  This can be ranked as a money-equivalent “net advantages”.
   6. Expectations are myopic (or extrapolative).  As a result, supply adjustments in the labour market are very slow, taking generations rather than years.
   7. Workers of different efficiency get different time wages, but the same “efficiency wage” (this is like thinking about the wage in quality adjusted terms – not the modern concept of higher wages increasing labour effort).
   8. Worker efficiency is endogenously relative to consumption and hours.
   9. Supplies of managers and businessmen are determined in the same way as skilled labour (wish I had seen this Marshall quote when I was doing my masters – given that it was about skilled labour with managers as my example.  Always good to quote the legends 🙂 ).

Sidenote 1:  On the note of labour markets, much of the talk about “types of consumption” and their impact on efficiency reminds me a lot of economists.  This is all fine, but the problem is when we start imparting this normative significance – as if production has value beyond the revealed consumption value of the person making the choice.

Sidenote 2:  With social mobility, it appears Marshall’s view of parent having difficulty trying to move kids into occupations “above” their own touches on the same mechanics as Bourguignon (1981) regarding social mobility (where Bourguignon is focused on non-concavity, a situation that could be created by credit market imperfection, driving permanent inequality between otherwise similar people).  Very cool!

Even with the greater detail in the labour market than prior theories, the type of assumptions that have been made still give us a situation where, as Marshall says:

Capital in general and labour in general co-operate in the production of the national dividend, and draw from it their earning in the measure of their respective (marginal) efficiencies.

And the description of equilibrium is one of very long periods of time, and “tendencies” for economic variables to shift towards this.  Note:  This is because this specific model was create to think about questions of the long-term – not because Marshall ignored other issues, which were covered in other areas.  Furthermore, secular trends that exist are seen as important for looking forward – but don’t invalidate the usefulness of considering a long-term stationary state.  These assumptions and justifications all sound much like introductory growth theory!

Conclusion

The essay concludes with three points:

1. Marshall’s theory bears a strong relation to classical tradition in terms of factor supply (albeit with more detail on the labour side).  His deviation was in terms of factor demand.  However, rather than solely focusing on the principle of substitution and marginal productivity, he instead focused on labour efficiency.
2. The fact Marshall was unwilling to include important issues on “adding up”, increasing returns, and the representative firm (issues developed elsewhere in his value theory) seems strange.
3.  His criticism of neo-classical theoriests wasn’t on the basis of their view of factor demand, but factor supply. He felt the neo-classical theoriests needed a more general framework for considering supply aspects – hence why he fleshed out modern sounding ideas on labour supply.

My impression of all this is that I think Marshall is an even more incredible economist than I had realised before.  The agency he appears to be giving to labour supply decisions vastly exceeds the agency provided in the prior essays, and can be seen as a precursor to modern labour economics.