Measuring the Wealth of Nations: The Political Economy of National Accounts

by Anwar Shaikh, Ertuğrul Ahmet Tonak

This book provides an alternate foundation for the measurement of the production of nations, and applies it to the U.S. economy for the postwar period. The patterns that result are significantly different from those derived within conventional systems of national accounts. Conventional national accounts seriously distort basic economic aggregates, because they classify military, bureaucratic and financial activities as the creation of new wealth, when in fact they should be classified as forms of social consumption that, like personal consumption, actually use up social wealth in the performance of their functions.

Genres: Economics, Nonfiction, Politics

404 pages, Paperback

First published January 1, 1994

About the author

Anwar M. Shaikh (born 1945) is a Pakistani American heterodox economist in the tradition of classical political economy.

He is Professor of Economics in the Graduate Faculty of Social and Political Science at The New School for Social Research in New York City, where he has taught since 1972.

Carrying on teaching and research that originated in the context of New Left social movements, Shaikh's political economy has focused on the laws of motion and empirical patterns of industrialized capitalism, based on a theory of competition independent of neoclassical economics.

In specific fields, he has written on the labor theory of value, production functions, international trade, neoliberalism, the welfare state, economic growth, inflation, the falling rate of profit, long waves, inequality on the world scale, and past and current global economic crises.

He graduated from Stuyvesant High School in New York City in 1961, received a B.S.E from Princeton University in 1965, worked for two years in Kuwait, and then returned to the United States to study at Columbia University, from which he received his Ph.D. in Economics in 1973. In 1972 he joined the Economics Department at the Graduate Faculty of the New School for Social Research.

Reviews

[Michael · Rating 5 out of 5]

In terms of empirical applications of Marxist political economy, this book is on another level. I couldn’t recommend this more.

INTRO
The issue motivating this book is that conventional national accounts distort the levels and trends of aggregate economic variables like the national product, the surplus product, productivity, etc. Shaikh and Tonak seek an alternative accounting by mapping Marxian value categories on to conventional input-output tables. Central to this is taking seriously the classical distinction between productive and unproductive labor (PUPL) which must be considered in calculating aggregate flows and stocks in the economy.

Shaikh and Tonak describe the basic national accounts which record productive flows (I-O Tables), unproductive flows (financial accounts), and stocks (national balance sheets); move on to mention some technical issues like the imputation of non-money economic activities and the status of the service sector in the PUPL debate; and offer a few comments on prior work, especially of Eisner (1988).

One point on Eisner worth mentioning is that, in agreement with Kuznets, services traditionally conceived of as "unproductive" like national defense or guard duty should instead be conceived as intermediate inputs since they represent a "precondition" as opposed to direct contribution to production. If I interpret this correctly, this means such services would transfer past value instead of subtracting from surplus-value (as is the view of most Marxists who treat unproductive labor as a form of luxury or social consumption). This seems to cohere with Murray Smith's argument that unproductive labor, being socially necessary, should be considered a "systemic overhead cost" and therefore as a part of total constant capital.

CH 2
This chapter provides a cogent defense of the productive/unproductive labor (PUPL) distinction on the basis of a clear separation between spheres of production and circulation. Shaikh and Tonak (ST from now on) divide human activity into 4 types: consumption, production, distribution, and social maintenance. The first does not involve labor, the latter three do. Out of those three, only production involves, well, productive labor in that they use objects of social utility to create new objects of social utility. Distribution uses such objects to transfer ownership of existing use-values. Social maintenance uses objects for private and non-market public purposes.

Crucially, use-values are not limited to corporeal objects but include any "material thing or effect" and therefore include services. ST propose an insightful distinction between usefulness for consumption vs usefulness for ownership that helped clarify which "material things or effects" are "internal to production" or "internal to distribution". So transportation of final goods to consumers is productive in the classical sense according to ST, whereas warehousing goods in one location over another according to cost considerations is unproductive since this has no bearing on the usefulness to consumption....it is internal to the sphere of circulation.

After explicating a concept of PUPL in general, ST concretize it with respect to specifically capitalistic forms of labor. The upshot being productive labor under capitalism is productive of surplus-value. They then criticize the orthodox view which treats all labor expended as productive and end the chapter rooting certain transfers of value/profits on alienation in the distinction between productive and unproductive labor.

CH 3
Now that the theoretical foundations have been laid out, ST begin the task of mapping Marxian categories on to national accounts. First, they introduce the main orthodox accounts: National Income and Product Accounts (NIPA) and Input-Output (IO) tables. These accounts record a set of economic measures including intermediate inputs, wages, and profit types (M, W, P). These conventional measures are to be related to two sets of Marxian measures--the (labor) value expression of constant capital, variable capital, and surplus-value (C, V, S); and the money expression of the same (C*, V*, S*).

Furthermore, these three sets of measurements need to be related at two levels of abstraction. The primary level reflects the flow of these categories in the "real" economy which determines the mass of commodities in the social product (eg. Production and wholesale/retail). The secondary level follows these flows through their motion within the sphere of circulation and includes sectors like finance and government.

Using IO tables with numerical examples for illustration, ST show in what ways the disbursements/revenues recorded as money flows in conventional accounts do or do not correspond to movements of use-value/exchange-value in the capital circuit. One of the main conclusions drawn from this exercise is that gross output as reflected in conventional reports tend to be greater than their Marxian counterparts because the former include the royalties earned by the secondary sector as genuine contributions to the social product whereas the latter recognizes them to be payments for rearranging the ownership of that product created by the primary industries.

ST then drop the assumption of a closed economy and devote a section to transfers in international trade through three mechanisms: producer price vs. purchaser price deviations, price vs value deviations, and direct cross-border payments for wages, interest, rent, dividends, etc. The first is based on the trading margin charged by merchant capital on the outputs of producers before being sold at their final (purchaser's) price to consumers. The second involves deviations of prices of production from labor values due to variations in sectoral capital/labor ratios about the average organic composition of capital. The third is just the money flows representing factor payments and will be treated in a future section.


Finally, the chapter comments on non-market activities usually imputed with estimated values into conventional accounts followed by a summary account of the story so far--complete with a very useful input-output figure where transactions are recorded as uses of commodities along the rows and as exchange-values down the columns:





CH 4
What chapter 3 does for Marxian categories in their money-form of value, Ch 4 does for their labor-form. That is, ST demonstrate how the flows of abstract labor within the economy can be represented consistently using the input-output tools of national accounts. The immediate complication here is that while money measures of value are directly comparable across the economy, homogenous labor is determined at the sector level due to the "heterogeneity of the production process of different categories of commodities" (Saad-Filho). Therefore, ST conceive the production row elements as matrices of input-output coefficients...to which they then apply some matrix algebra operations I don’t really understand (subtract from the identity matrix and invert the difference), multiplied by a row vector of labor hours per output (sourced from BLS data) and wind up with a vector of labor values for units of output which, obviously, isn't given by national statistics.

λ=hp*(I-app)^(-1)


Where:
λ = the row vector of unit labor-values
hp = hours of productive labor per unit output
app = coefficients matrix for productive inputs

ST show how these labor-value measures correspond to the money measures of the same Marxian aggregates (C, V, S) with helpful numerical examples. An important contribution of this book is that it incorporates the distinction between producers' price and purchasers' price into its calculations of aggregates since Input Output tables report flows of commodities at the former and not the latter. This is an advance over earlier attempts (Wolff) which resulted in overestimated money measures of surplus.

CH5
Chapter 5 is the empirical meat of the book and applies the theoretical framework already outlined to the actual I-O and NNIP data for the United States from 1947 to the 1980s. I’ll just list some of their most important findings:

-There is a dramatic increase in unproductive labor as a share of total labor over this period

-Both the mass of surplus and profits grow, both at declining rates, with the rate of surplus declining somewhat faster (due to growth in the unproductive employment)

-The ratio of profits to wages (the profit/wage share of income) is a very poor proxy for measuring the rate of surplus value. This is important since the former measure is often used as a shortcut for estimating the latter in many studies

-Likewise, the ratio of intermediate inputs to wages is a poor proxy measurement for the value composition of capital C/V

-The value composition of capital increased enormously over time

-The rate of surplus value increased quite modestly

-The rate of profit declined markedly as a result of the previous two trends (exactly in accordance with Marx’s theory of the TRPF)

-During the 80s the rise of the VCC slowed while the rate of surplus-value sharply increased leading to a slight recovery in the rate of profit (this supports the idea that the neoliberal assault on labor was an effective response to the crisis in profitability caused by technical progress over the preceding decades which, itself, started to slow due to decreased profit incentives to reinvest surplus in production)

-ST observe a decreasing rate of increase of labor productivity

-Government spending as a share of surplus value remains relatively stable at 35%

-US workers receive a “negative social wage” by paying in taxes more than they receive in benefits on average

-Price-value deviations are empirically negligible. That is, the tendency for prices of production to diverge from embodied labor-values due to the equalization of profit rates across industries with different organic compositions of capital only amounts to about 6-9%

CH 6
This chapter is an extensive assessment of the literature which I won’t go into detail about. I’ll just say it’s an invaluable resource for familiarizing yourself with the different approaches and results Marxist economists have observed over the years this has been studied. This book is essentially a culmination of all that research and it shows.

CH 7
The last chapter just summarizes the theoretical framework, empirical method, notable findings, and advances over earlier studies that this book represents. I’m totally blown away by it and will probably be returning to it as a resource many times in the future.

[Eugene Kernes · Rating 3 out of 5]

What we define as wealth has implication on how we measure it. One of the major issues the Shaikh and Tonak describe is the implication between the difference in productive and unproductive labor, or as they prefer to state it, production and nonproduction labor. All labor produces an outcome that can be extremely valuable to society, but not all labor has an output. Production labor add to the total wealth of society, while nonproduction labor either transfers wealth of society or is just a form of social consumption. Under this framework, changes in employment have different effects depending on the type of labor.

The current Neoclassical framework of measuring production is to include many forms of labor that are actually meant to be social consumption. Making anything that someone is willing to pay for a production activity. The problem with this view, that the authors point out, is the implication on investments change. Increasing social consumption, in the authors framework would reduce the available investment.

The authors pose an alternative way of measuring production. They use a variety of models and charts to show the way their alternative measure would look like. Their methodology is based on the Classical perspective, more of from the view of Marx. Their measures differ from Marx, but are based on Marx. The authors use an input-output account framework, rather than the national-income-and=product account, as it enables greater coverage by tracking the whole product.

The problem with the book is that the general discussion is relegated to a minority position. The impact of particular measures of wealth and the politics of measuring wealth is not the dominant theme. The book is not an easy read and deeply mathematical. Most of the book is trying to explain the math and outputs of the various alternative measures of wealth. The equations become harder to read as the explanation to them is reduced. Although the authors meticulously explain their equation and charts, there appears to be a huge disconnect between the math’s and their meanings.

[Matt · Rating 2 out of 5]

I was recommended this book by a friend, and the best I can say is that Profs. Shaikh and Tonak make some interesting points about how the national accounts and gross national product had been devised, specifically why it shouldn't treat certain activities as output producing (e.g. financial services, ground rent, certain parts of government expenditure like administration and defense spending) and instead as a form of social consumption, thereby getting perhaps a clearly picture of available national savings and real investment. Others in the mainstream neoclassical tradition have put forth similar arguments (see: Kelvin J. Lancaster), but Shaikh's and Tomak's come at it from a classical political economic perspective a la Ricardo and Marx.

Where I get lost, however, is when they delve into the nebulous realm of Marxian economics - labor values, money values, surplus value, the whole transformation problem, etc. To be frank, I think it undermines their whole argument by spending so much time trying to use this novel perspective to try and solve problems most economists had largley moved on from for good reason, including many Marxists from Joan Robinson to John Roemer.

Much of the analytical structure the authors make rests on one critical assumption that allows them to solve the Marxian transformation problem through national input-output tables, and they are clear about it: labor values and producer prices are proportional by some assumed constant. This is a rather bold hypothesis, and without it, sadly, the problem devolves back into the same one Marx had in Capital Vol. III of how 'labor values' map into market prices that is one-to-one (in other words, from a given set of prices, there is an inverse mapping from prices to labor values to allow value-based calculations).

Overall, this book is a good thought provoker for macroeconomists if one just reads the introduction and conclusion, and skip most of the body. For most of everyone else, you probably will find it to be a long and pointless read.