The objective of the Institute for Policy Reform is to enhance the foundation for broad based economic growth in developing countries.Through its research, education and training activities the Institute encourages active participation in the dialogue on policy reform, focusing on changes that stimulate and sustain economic development.At the core of these activities is the search for creative ideas that can be used to design constitutional, institutional and policy reforms.Research fellows and policy practitioners are engaged by IPR to expand the analytical core of the reform process.This includes all elements of comprehensive and customized reforms packages, recognizing cultural, political, economic and environmental elements as crucial dimensions of societies.

In environments with private information and spatial separation, the ability of agents to establish mutually beneficial arrangements can be limited by their ability to communicate contemporary dealings and histories of past dealings. Indeed, with the extension of some recent work in contract theory and mechanism design, this paper argues that location or person—specific assignment systems, portable object record—keeping systems, written message systems, and telecommunication systems can be viewed as communication systems which are successively more complete in this sense. An attempt is made also to match these various communication systems with systems in use in historical primitive, and/or contemporary societies and to interpret these communication systems as financial structures. Economic agents are naturally separated in time and space. In an economy at large agents reside typically in various locations. work at alternative distinct

This paper presents formal, stylized representation of communication-accounting systems: oral assignment, portable object, written message, and telecommunication systems are considered. The environments that allows this formalization are characterized by spatial separation, private information, and a need to keep track of past actions, transfers, and shocks.

A recent development in economic science is the attempt to integrate monetary theory with the theory of general economic equilibrium. This work takes as its starting point the idea that money cannot have value in standard, general equilibrium models. In these, too much trade can be accomplished in centralized markets (see Robert Clower, 1969, 1971; Frank Hahn, 1973; or Neil Wallace, 1980). Thus, to decentralize or break up the structure, either exchange must be made costly or there must be restrictions on who can trade with whom, and thus such choicetheoretic models offer the intriguing possibility that real and monetary phenomena can be understood as intimately related. This paper continues in the relatively brief, choice-theoretic tradition, motivated by real and monetary phenomena associated economic development and growth: 1) To be noted first is Simon Kuznets' seminal work on national income (1971). In a cross-section study of fifty-seven countries in 1958, Kuznets shows that the share of the agricultural sector, including forestry, fishing, and hunting, in Gross Domestic Product is inversely correlated with Gross Domestic Product per capita. The share of the industrial sector, including transportation and communication, is closely and positively associated with per capita product. The share of the service sector tends to be positively but weakly associated with per capita product, but the share of banking, insurance, and real estate shows a striking rise as one shifts from lowto higher-income countries. Moreover, the evidence suggests that the ratio of industrial prices to agricultural prices is perhaps lower the higher is per capita income, though the evidence on relative prices for the service sector is inconsistent. Turning to long time-series for thirteen developed and four less developed countries, Kuznets finds dramatic evidence for a decline of the agricultural sector and a rise in the industrial sector with per capita income, at least in developed countries. Again, results for the service sector are mixed, but Canada, France, and the United States are positive exceptions. The share of a transport-communication subsector rises quite consistently. Turning next to shares of sectors in the labor force, Kuznets finds, both on a cross-sectional and secular basis, that all the above movements are at least mirrored and in many cases amplified. In particular, both components of the share of the service sector, services and commerce, rise substantially with Gross Domestic Product per capita. 2) To be noted second is the extensive work of Raymond Goldsmith on financial structure and financial intermediation. For the United States, Goldsmith (1958) finds that the activity of intermediaries, as measured by their share in national assets, in tangible assets, and in all claims, has shown a substantial rise from 1860 to 1952. Similarly, Goldsmith (1969) finds that the ratio of financial institutions' assets to Gross National Product rises substantially from 1860 to 1963 in both developed and less developed countries, including Switzerland, Great Britain, the United States, Japan, Argentina, and India. Related, the number of households with savings accounts, the number with life insurance policies, and the number with stock ownership expressed as percents of the population are all low for less developed countries relative to developed countries, and *Professor of Economics, Carnegie-Mellon University, Graduate School of Industrial Administration, Schenley Park, Pittsburgh, PA 15213. This paper was motivated by a conversation with Thomas Sargent and has been aided by helpful comments from Robert Barro, Robert E. Lucas, Jr., Dan Peled, Kenneth Singleton, and Neil Wallace. Financial support from the National Science Foundation, the Alfred P. Sloan Foundation, and the Peterkin Symposium on Foundations of Monetary Policy and Government Finance at Rice University, and research assistance from Pramerudee Townsend are all gratefully acknowledged. I alone assume full responsibility for any errors and for the views expressed here.

In an environment with private information, spatial separation, and limited communication, a currency-like object and more standard named credits can be distinguished. The credit objects can be used among agents in an enduring relationship, that is, among agents with known trading histories, whereas the currency-like object must be used among relative strangers. In this environment, collectively determined Pareto-optimal rules make the level of the currency-like object and the mix of currency to named credits responsive to individual needs and to economywide states. Total indebtedness is determined by the number of lenders, that is, by preference or demand shock, and the mix of currency to credits is determined by transaction patterns among the agents.

In an environment with private information, spatial separation, and limited communication, a currency-like object and more standard named credits can be distinguished. The credit objects can be used among agents in an enduring relationship, that is, among agents with known trading histories, whereas the currency-like object must be used among relative strangers. In this environment, collectively determined Pareto-optimal rules make the level of the currency-like object and the mix of currency to named credits responsive to individual needs and to economywide states. Total indebtedness is determined by the number of lenders, that is, by preference or demand shock, and the mix of currency to credits is determined by transaction patterns among the agents.

This paper explores the formulation and analysis of linear equilibrium models of investment in which learning is perpetual and informationally decentralized firms need never share the same beliefs concerning time series relevant to their decisions. Recursive, Kalman filtering techniques are shown to be applicable in an illustrative, hierarchical information structure, and a nonlinear technique of undetermined coefficients is shown to be applicable in an illustrative, symmetric information structure in which there is a confounding of laws of motion with forecasting problems. The equilibrium time series of these models can display interesting movement in response to shocks and measurement errors, including persistence, certain cross-correlation properties, and damped oscillations. That is, forecasts errors are serially correlated over decision makers and serially correlated over time in a certain crucial sense. More generally, these models do place restrictions on observed time series and can be fitted to data.