A Random Coefficient Model of the Demand for Liquid Assets

Abstract

THE PAST DECADE has produced an abundance of empirical studies which estimated the demand functions for liquid assets based on temporal cross-section data. The focus of these studies has been the estimation of the degree of substitution between the liabilities of bank and nonbank financial intermediaries. These substitution relationships are recognized as being highly relevant to such diverse issues in the monetary literature as the appropriate definition of moneyS the macroeconomic effects of monetary policy, and the allocative consequences of the growth in financial intermediation. The richness of the temporal cross-section data base allows a varied menu of alternative model specifications and, consequently, a cornucopia of estimates of the elasticities of substitution between different liquid assets. Several studies have used specifying assumptions which constrain behavior to be stable, both over time and over geographic units (see, for example, Cohen [1], Hartley [5], Kichline [6], and Lee [7]). Other studies have assumed geographic behavioral stability while testing temporal behavioral stability (see Feige [3]). A recent survey of studies of the demand function for demand deposits utilizing temporal cross-section data revealed a remarkable robustness of the estimated elasticities to significant departures from the more conventional model specifications; however, none of the prior studies allowed for behavioral differences across geographic areas (see Feige [4]). To the extent that geographic boundaries reflect differences in both lnstitutional arrangements and individual decision-making, the

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 37 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.