Reconstruction of Macroeconomics: Methods of Statistical Physics, and Keynes’ Principle of Effective Demand

Over the last 50 years, macroeconomics has drastically changed. Born in the midst of the Great Depression, it was once synonymous with Keynesian economics. However, under the flag of micro-foundations, it has turned to neoclassical macroeconomics.
 
Modern micro-founded macroeconomics is proud of its solid micro-foundations by which the profession understands that optimization of a micro-economic agent such as the household or firm is explicitly analyzed. In this book, I maintain that standard micro-foundations are on the wrong track. Ironically, modern macroeconomics, which emphasizes the significance of micro-optimization so much, takes it too carelessly.
 
The fundamental problem is that modern micro-founded macroeconomics rests on representative agent assumptions. In the case of real business cycle (RBC) theory and endogenous growth theory, this is obvious, for a literally single representative household/firm is assumed. In contrast, in the case of Lucas’s rational expectations model, labor search theory, and more recent heterogeneous agent new Keynesian (HANK) model and mean field games, the problem is much subtler because in these models, apparently heterogenous agents are assumed. However, economic agents differ only to the extent that the realization of relevant stochastic variables is different. They are assumed to face a common stochastic distribution with which they optimize. I contend that the assumption that all the economic agents share a common stochastic environment, namely, the assumption of representative constraints, is wrong. It is essentially nothing but a stochastic version of representative agent assumptions, not worthy of micro-foundations.
 
In modern micro-founded macroeconomics, aggregate demand has been purged as a determinant of the level of macroeconomic activity. This makes macroeconomics more or less Walrasian if on the surface it is different from Walras’s original framework. The tragedy is that the macroeconomy in which we happen to live is not Walrasian. In contrast to financial markets, prices play only a limited role in the real economy.
 
I provide Keynes’s principle of effective demand with a solid micro-foundation based on methods of statistical physics. Aggregate demand is the fundamental factor that governs the matching of workers and firms in labor markets.
 
Demand plays a central role not only in the short run but also in the process of economic growth because in advanced economies, saturation of demand for the existing goods and services constrains economic growth. Growth is generated by product innovations. This is where Keynes and Schumpeter meet.
 
The role of the financial market must be reconsidered. No doubt a well-developed financial market contributes to economic development through efficient resource allocation. However, it occasionally disturbs the macroeconomy. Efficient market hypotheses must be discarded as universal theory; “bubble” is a meaningful concept. I argue that the notion of “efficiency” differs fundamentally in the real economy and the financial market.
 
Macroeconomics requires different methods than microeconomics, that is, methods of statistical physics. Macroeconomics must change.
 

(Written by YOSHIKAWA Hiroshi, Professor Emeritus, Graduate School of Economics / 2024)