Since the pioneering work by Levhari and Patinkin (1968) and Sinai and Stokes (1972) several empirical studies have attempted to identify the role of real money balances on aggregate production function. However, this bunch of empirical work followed different tracks. Some researchers based on the initial contention made by Sinai and Stokes (1972, p. 290) that “real money balances increase the economic efficiency of a monetary compared with a barter economy” (e.g., Prais, 1975; Ben- Zion and Ruttan, 1975; Short, 1979; Subrahmanyam, 1980; Khan and Ahmad, 1985; Betancourt and Robles, 1989) have argued that real money balances should be included as a separate input in the production process together with conventional inputs (i.e., physical capital and labour).
However, this production-based approach has been heavily criticized by several other authors (e.g., Fischer, 1974; Claassen, 1975; Davidson, 1979; Nguyen, 1986) arguing that viewing real money balances as a conventional factor of production is conceptually flawed as the technical relationships of the production process are confounded with the underlying exchange relationships that real money balances actually reflect. In an attempt to overcome this criticism recently Delorme et al., (1995) and Nouzrad (2002) argued that real money balances do affect the overall productivity in the economy indirectly through their impact in the efficient utilization of the conventional inputs. Specifically, they suggest the indirect use of real money balances as a factor affecting technical efficiency rather than as a separate input in the aggregate production process.
The main objective of this paper is to enrich our knowledge on the true effect of real money balances on the production process, by providing empirical evidence of their impact on TFP changes in a sample of 79 developing and industrialized countries during the 1965-1992 period. For doing so we integrate both production-based and technical inefficiency approaches into a single framework using the notion of non-neutral stochastic production frontier proposed by Huang and Liu (1994). This general specification allows to statistically examine each of the aforementioned approaches concerning the treatment of real money balances in country’s’ overall productivity as a special case within the proposed formulation. That is, formal statistical testing is used to check whether real money balances should be included only in the production frontier or in the inefficiency effect model, or in both. In the latter case, an empirical evaluation of the true effects of real money balances on TFP changes can be obtained using the integrated primal approach developed by Bauer (1990), Lovell (1996) and Kumbhakar (2000).