Saturday, November 30, 2019


  • David Graeber's review of Robert Skidelsky's Money and Government: The past and Future of Government.
  • A TED talk, by Nick Hanauer, on how complexity economics is replacing "neoliberal" economics. He is especially interested in reciprocity.
  • A 2014 interview by Bill Moyers, of Paul Krugman, on Piketty's book.
  • Mariana Mazzucato, with a talk on the value of everything. She also has a 2013 TED talk.
  • Heinz Kurz on the Cambridge capital controversy.
  • Bertram Schefold on the CCC and his recent research.
  • John Eatwell on Joan Robinson, including on the "disgrace" of mainstream economists not taking on board the results of the CCC.

Update (6 Dec. 2019): Added link for John Eatwell.

Saturday, November 23, 2019

Literature Distinguishing Large Corporations And Finance From Competitive Firms

A considerable body of literature has been published, during the last century, arguing that a movement away from competitive markets must be recognized in trying to describe and understanding contemporary capitalism. The literature I am thinking of emphasizes big business, corporations, and finance. Here are some selections, not all of which I have read:

I find that I have provided a similar list before. If I wanted to include journal articles, I would say something about Paulo Sylos-Labini. There was a nearby literature arguing the convergence of different systems across, say, the first and second worlds. Another related literature develops theories of imperialism, especially in the context of north-south relations.

As I understand it, many of the above writers were influenced by Marx. But I think even those who accepted the labor theory of value for competitive conditions, argued that the developments they were writing about implied that it no longer applies. For instance, Baran and Sweezy replaced surplus value by the (non-quantitative?) concept of the economic surplus. I was surprised to find myself arguing against this conclusion in recent work.

Friday, November 15, 2019

The Rate of Profits is Not the Scale Factor

Figure 1: Rate of Profits Unequal to Scale Factor for Rate of Profits

This post continues the example in the previous post. I modify the prices equations so that the rate of profits in producing corn is (s1 r̂), and the rate of profits in producing ale is (s2 r̂). The solution to the price equations are:

pcorn = 16 [16 + (s1 - s2) r̂]/[204 + (3 s1 + 9 s2) r̂]

pale = 32 [10 - (s1 - 3 s2) r̂]/[204 + (3 s1 + 9 s2) r̂]

w = 4 [51 - (9 s1 + 5 s2) r̂ - s1 s22]/[204 + (3 s1 + 9 s2) r̂]

where r̂ is what I have been calling the scale factor for the rate of profits.

I want to show that, only in exceptional cases can the the markups s1 and s2 be rescaled such that the scale factor is equal to the economy-wide rate of profits, whatever the distribution of income. For concreteness, assume that the rate of profits is four times as high in the corn industry, as compared to the ale industry. That is, introduce a new parameter α such that:

s1 = α
s2 = (1/4) α

Suppose the wage is unity and workers receive the entire standard commodity. Then both the scale factor for the rate of profits and the rate of profits are identically zero percent.

Next, consider the other extreme case, where the wage is zero. The maximum rate of profits is 300 percent. For the scale factor to also be three, the numerator for the wage, in the third equation above must be zero, when a rate of profits of 300 percent and the above scale factors are substituted in. One thereby obtains a quadratic equation:

3 α2 + 41 α - 68 = 0

The positive solution is:

α = (1/6)( -41 + 2,4971/2)

For any scale factor for the rate of profits, one can find the wage with these markups. For any wage, one can find the economy-wide rate of profits:

r = 3 (1 - w)

The simplicity of the above equation results from taking the standard commodity as the numeraire. The graph at the top of this post shows the difference between the rate of profits and the scale factor, as a function of the proportion of the standard commodity paid out in wages. The rate of profits and the scale factor are equal only at the two extremes. I guess that in a model with more commodities, the difference does not come out looking like a quadratic function.

This counterexample demonstrates that, in general, one cannot rescale markups such that the scale factor for the rate of profits is the economy-wide rate of profits, however distribution stands. Part of Marx's point in Capital is that observers who focus on surface phenomena will not perceive underlying value relations.

"The relations connecting the labour of one individual with that of the rest appear, not as direct social relations between individuals at work, but as what they really are, material relations between persons and social relations between things."

I guess I have shown that the existence of persistent ratios between the rate of profits in various industries, as with a corporate sector and a sector of small firms and proprietorships, provides another layer of confusion in economic analysis. Here is a source of another illusion created by competition.

Aside: I have stumbled across the International Symposium on Marxian Theory (ISMT). They have a series of books out, with a focus on reacting to the complete works of Marx and Engels (MEGA2). MEGA2 is published under the auspices of the Internationale Marx-Engels-Stiftung (IMES) in Amsterdam and of other groups in other countries. Apparently, volume 1 of Capital varied among editions, and Marx had several drafts. Furthermore, MEGA2 apparently has editions of volumes 2 and 3 that show exactly how Engels edited them. These facts re-enforce the point of this post, not that I want to read all of these variants.

Wednesday, November 06, 2019

An Example Of The Labor Theory Of Value

Figure 1: Variation of Prices of Production with Wages and Markups
1.0 Introduction

This post documents an example in my working paper, The Labor Theory of Value and Sraffa's Standard Commodity with Markup Pricing.

2.0 Technology

Consider a simple economy in which corn and ale are each produced from inputs of labor, corn, and ale. Inputs for unit outputs are shown in the columns in Table 1. Obviously, the units of measure should not be taken serious. Inputs are totally used up in the production of outputs. I abstract from the existence of fixed capital, land, and joint production.

Table 1: The Technology
Labor1 Person Year1 Person-Year
Corn(1/8) Bushels(3/8) Bushels
Ale(1/16) Pints(1/16) Pints

The standard net product consists of (9/16) bushels corn and (3/16) pints ale. The Perron-Frobenius root of the Leontief input-output matrix is 1/4. (The other eigenvalue is (-1/16). The maximum rate of profits is 300 per cent. Labor values are (64/51) person years per bushel corn and (80/51) person-years per pint ale.

3.0 Price Equations

Equations for prices of production are:

[(1/8) pcorn + (1/16) pale]( 1 + r̂) + w = pcorn

[(3/8) pcorn + (1/16) pale]( 1 + s2 r̂) + w = pale

(9/16) pcorn + (3/16) pale = 1

I have taken the standard commodity as the numeraire. This allows one to freely move back and forth, when evaluating aggregates, from labor values to monetary units.

The rate of profits in producing corn is 100 r̂ percent, while it is s2 r̂ percent in producing ale. I am assuming there are persistent barriers to entry or some reason why the rate of profits persistently varies between industries. Some economists talk about dual markets. I can also point to John Kenneth Galbraith's The New Industrial State for a contrast of corporations in the planning system and more traditional firms. Anyways, the solution of these equations is:

pcorn = 16 [16 + (1 - s2) r̂]/[204 + (3 + 9 s2) r̂]

pale = 32 [10 - (1 - 3 s2) r̂]/[204 + (3 + 9 s2) r̂]

w = 4 [51 - (9 + 5 s2) r̂ - s22]/[204 + (3 + 9 s2) r̂]

These equations show that prices of production vary from labor values when the rate of profits is positive. Furthermore, these are not straight lines, although the curvature is not visually impressive in the figure at the top of this post.

Anyways, here is a question. Suppose labor coefficients happen to be a left-hand eigenvector of the Leontief input-output matrix, a very special case. When prices of production are defined with equal rates of profits across all industries, prices of production are labor values in this special case. (The specification of the numeraire does not matter.) Does this property still hold under the sort of markup pricing which I am assuming?

Update (8 Nov 2019): A supporter in email points out a special case. Let s1 = 5/13 and let s2 = 5/29. Then prices of production are labor values. The scale factor for the rate of profits is: r̂ = 3 (1 - w). That is, the scale factor is the rate of profits. Presumably, with these relative markups, relative prices are relative labor values, whatever the numeraire.

Friday, November 01, 2019

Keen's Debunking Economics Most Popular Among Popular Critiques

Table 1: Selected Critiques
Moshe AdlerEconomics for the Rest of Us214
Rod Hill & Tony MyattThe Economics Anti-Textbook134
Steve KeenDebunking Economics, 1st edition253 to 4
Debunking Economics, 2nd edition564 to 5
Paul OrmerodThe Death of Economics103 to 4
John QuigginEconomics in Two Lessons24
John WeeksEconomics of the 1%134 to 5

Steve Keen seems to be the most popular of those writing internal critiques of economics directed towards the common reader. I selected the above books and looked at rankings on Amazon's United States website. You can spend lots of time reading the comments.

I am not sure about how to characterize this genre. I am more focused on theory than offering political programs. Would Robert Reich's Saving Capitalism be excluded? But what about memoirs, such as John Perkins' Confessions of an Economic Hit Man, Stiglitz' Globalism and its Discontents, Thaler's Misbehaving, or Kahneman's Thinking Fast and Slow? These books seem to have much more ratings than the ones I list in the above table.

Why is Keen's book more popular than the other ones in the table? Keen often overstates his case. One reviewer said he confuses necessary conditions with sufficient ones. I'm covered here; I suggested to him, before publication of the first edition, that well-behaved aggregate excess demand curves might exist in special, numeric, cases even if all consumers did not have identical and homothetic preferences. But those who know of Alan Kirman's work, with others, in the 1970s know Keen has a point. You cannot find any other condition than Gorman form that is sufficient to have well-behaved aggregate excess demand curves. And this is true of many other of Keen's points. I had not realized before reading Keen that the standard textbook presentation of perfect competition assumes managers of firms are systematically mistaken in their understanding of the demand curves they face.

Anyways, neoclassical economics is mostly wrong or useless for internal, logical reasons.