MICROBUSINESS ACCOUNTING AND MACROECONOMIC ACCOUNTING: THE LIMITS TO CONSISTENCY*

Abstract

The paper is concerned with analyzing the consistency problem that arises when the macroenterprise sector of a nation's accounting system is put on a microdata foundation. This foundation is composed of sets of microbusiness accounts, after some appropriate rearrangements and reclassifications. We pose the question: can the macroenterprise sector accounts be regarded as a consolidation of (observed) microbusiness accounts? The answer is positive from a purely conceptual viewpoint, but negative from a statistical viewpoint which preserves the decision‐making records of microbusiness units. The latter phenomenon is referred to as the limits to (statistical) consistency while attempting to maintain the viability of a national accounting system.The analysis proceeds by exploiting the structural properties of market transactions matrices for a nation's economy. The results are sufficiently general to encompass the case where the transaction matrices are initially characterized by both sectoral discrepancies and transaction flow category discrepancies. In this general context it is shown that the statistical inconsistency potentially resulting from the replacement of the macroenterprise sector by an aggregation of microbusiness units has certain properties with economic meaning. This leads to a discussion that explains the ultimate rationale of statistical inconsistency: the fact that different microeconomic decision units may have different views and knowledge of common market transactions. The paper concludes with some implications for future research that appear to follow from the historical development of the subject matter.