By Chandrashekar (Chandra) Tamirisa, (On Twitter) @c_tamirisa

The problem of aggregation in macroeconomics is a complicated one. Dislocated from microfoundations, John Maynard Keynes had articulated in abstraction a new framework of looking at the wealth of nations as a whole during the Great Depression (besides The General Theory of Employment, Interest and Money of 1936 by Keynes, for a seminal textbook see Gregory Mankiw’s Macroeconomics).

Macroeconomics is not derivable from microfoundations. There is no theory of profit maximization of nations, or utility maximization by entire societies. The fundamentals of demand and supply are derived from the framework of the Keynesian Cross. The marginal propensity to consume (MPC) has no counterpart in investment. There is no concept such as a marginal propensity to invest (MPI).

The only efforts to root macroeconomics in microeconomics, similar to public economics, were started by Robert Solow when he used the production function to relate real output to real capital in a hybrid theoretical model and later by Milton Friedman for modeling consumption.

Keynesian macroeconomics relies on statistical data aggregation methods for the product markets and money markets to apply the analytical framework of Keynes and Robert Mundell. And from this the rest of open economy macroeconomics is derived. The variables in the output equation, Y (output), C (consumption), G (government spending), I (investment) and NX (net exports) are all empirically measured. Methods to empirically measure aggregates were put in place in the United States after World War II (after the New Zealander A.W. Phillips used similar data in the later 19th century to derive the Phillips Curve for U.K data). All economic forecasting is statistical.

It, therefore, behooves to use the econometric framework to prove a point about the definition of society as an aggregation of individuals from sample data. Let us consider for example two variables WE and I. I is a variable in the domain { 1 le I le n}. The following econometric equation:

i = n

we = ∑ I **(1)**

i = 1

If WE is regressed on attributes of I (in a data vector), assuming equality at the beginnings in access to basic education and health, to describe every individual in the sample, it would be clear that the coefficient on each member of the set I ∊{1 le I le n} would be different corresponding to their weight in producing we. The hypothesis that these coefficients change over time (in a panel-data model) relates to other variables exogenous to this model.

In macroeconomic terms, we = Y, the gross domestic product (GDP) of a country. The society is derivable from the basic macroeconomic cycle of gross domestic income (GDI) and the gross domestic product (GDP), where income = expenditures within the margins of statistical error.

A corollary of (1) would be to prove the disintegration of imposed equal weights on individuals in a society in the sample to return to the conclusions of (1) over time as a different approach to critique Marxist economics.

Likewise,

j = n

WE (global) = ∑ we **(2)**

j=1

where we ∊{1 le j le n} is a nation or a society. In macroeconomic terms, WE = global output or simply the sum of GDPs of all countries in the world.

The Solow growth model for the very long run, under its principal assumption of exogenous technical change, is inapplicable for individual countries but is applicable for the world as a whole because it is a closed-economy micro-macro model.

These two above equations, under the assumptions of neoclassical economics, must yield similar results, but only adjusted for scale. Meaning, efforts by governments to raise living standards are the equivalent of profit maximization by a firm and to raise the quality of life, the equivalent of utility maximization.

The corollary of (1) can analogously be written as a corollary of (2).

If the left hand side and the right hand side of the above two equations are reversed, it is self-evident that the individual is not endogenous to the model, we being the exogenous variable in that case, meaning, the individual cannot be Granger-caused by any society, but any society can be Granger-caused by a collection of individuals.

All people and all countries are not equal, but only *created* equal.