theoretical analysis by Arrow and Debreu (1954) demonstrates that general equilibrium for a competitive economy can be achieved under the assumptions of complete markets and perfect information, and there is no role for monetary factors or transactions costs. Stiglitz (1994) recognized that Arrow-Debreu's analytical insight was to identify the singular set of assumptions under which Adam Smith's invisible hand proposition would be valid.5 Under this set of assumptions, the pursuit of self-interest is conducive to competitive equilibrium. But it is argued also that the relevance of this neoclassical model of general equilibrium for welfare economics would be rather limited in the absence of perfect information and in the absence of important markets for risk allocation. The assumption of perfect information implies that the set of information available is fixed and invariable to the behavior of individual economic agents, independent from the pricing system, insensitive to changes in other economic variables. It is research about the implications of imperfect information for welfare economics that gave birth to the information paradigm, which addresses the information-theoretic concerns about the distortive effects of imperfect and asymmetric information on optimal resource allocation. The efficiency of competitive economies, which is considered as the first fundamental theorem of welfare economics, is deemed, according to Stiglitz (1994), to be fundamentally flawed. This assertion is based on theoretical evidence from Greenwald and Stiglitz (1986, and 1988) that markets are not constrained Pareto efficient under imperfect or asymmetric information and an incomplete set of markets for risk allocation.
This result follows from the existence of externalities in the decisions of some economic agents that are not taken into consideration by others. For instance, the purchase of insurance reduces the incentive to avoid the occurrence of a risk event, leading to moral-hazard problems. Also, Stiglitz (1994) argues that competitive market equilibrium with imperfect information is not necessarily described by market conditions where demand equals supply. The assumption that the pricing system ensuring market-clearing conditions is linear may not be tenable in light of price discounts relative to purchased quantities. The incomplete set of markets can also be explained by prohibitive information costs and transactions costs, which render difficult the inception of markets for risk allocation under all contingencies and all future delivery dates. Furthermore, Stiglitz (2011) notes that the recent literature on general equilibrium indicates that even under rational expectations, markets are not necessarily (constrained) Pareto efficient. This degree of market efficiency is never achieved under imperfect and asymmetric information and incomplete markets for risk allocation. It is the failure of modern macroeconomic models to account for market inefficiencies that limits their relevance for prediction, policy, or explanation purposes.
Rethinking Conventional Economics and Finance
Thus, the scope and limits of Arrow-Debreu equilibrium analysis are subject to continuous scrutiny. But this general equilibrium model provides a theoretical framework for optimal risk sharing. In an ideal conventional financial system, financial markets and financial intermediaries provide opportunities for intertemporal consumption smoothing by households and capital expenditure smoothing by firms. Savings represent a trade-off between current and future consumption, and real investments represent present expenditures with expectations of future economic output. It is natural that attitudes toward risk, including income risk and consumption risk, differ across market participants. The Arrow-Debreu framework allows for the distribution of risk in the economy among economic agents according to their respective degrees of risk tolerance, as noted by Hellwig (1998). In addition, the significant advances in general equilibrium models and finance theory have, nevertheless, provided a rigorous analytical framework for the examination of an ideal financial system for optimal allocation of risks in the society. They provide also clear evidence about the existence of a trade-off between risk and return and about the concept of no-arbitrage asset pricing, which underlie the capital asset pricing model and the arbitrage pricing theory.
However, the recurrence of financial crises has exposed the inherent instability of the conventional financial system. Reinhart and Rogoff (2009) provide evidence from the history of debt crises about the universality of serial defaults. The procyclicality of the financial system reflects the propensity of the banking system to expand credit during economic booms and restrict it in response to economic downturns. This procyclicality is, as noted by Rochet (2008), intrinsic to the financial system, but it is associated with financial fragility, which as defined by Allen and Gale (2009) reflects the potential for small shocks to generate significant effects on the financial system. As argued by Stiglitz (2011), there is a general recognition of the failure of standard macroeconomic models to predict the U.S. financial crisis or to understand the extent of its implications. It is further argued that the pursuit of self-interest “did not lead, as if by an invisible hand, to the well-being of all.” Indeed, Mirakhor and Krichene (2009) argue that the Arrow-Debreu conceptualization of an exchange economy based on risk sharing was transformed in steps into an economy based on risk transfer and eventually on risk shifting to taxpayers through government bailouts. As argued by Reinhart (2012), elevated levels of government indebtedness are conducive to a resurgence of financial repression, as reflected by tightly regulated financial environment. Since financial repression involves a distortion of resources allocation, as noted by Cottarelli (2012), this process is not consistent with the ideal conventional financial system and the Arrow-Debreu competitive economy with optimal allocation through risk sharing.
In light of the properties of the conventional financial architecture, there is an ongoing debate about rethinking the foundations of macroeconomics and financial economics, and reconsidering the implications of behavioral economics and behavioral finance for policymaking, regulatory, and academic purposes. In this regard, Stiglitz argues that “New Macroeconomics will need to incorporate an analysis of risk, information, and institutions set in a context of inequality, globalization, and structural transformation, with greater sensitivity to assumptions (including mathematical assumptions) that effectively assume what was to be proved (for example, with respect of risk diversification, effects of redistributions). Agency problems and macroeconomic externalities will be central” (2011, 636–637). The misalignment of incentives, moral hazards, information asymmetry problems, and regulator's capture stemming from risk transfer activities indeed contribute to the complexity of an interconnected financial system and to the complexity of prudential regulation. It is for these reasons that Bean (2009) also argues for the need to reconsider the role of financial intermediation in the development of macroeconomic models, in consideration of the peculiar properties of the balance sheets of financial intermediaries.
The renewed argument is thus made also for abolishing fractional reserve banking with the aim of dissociating the credit and monetary functions of commercial banks. As argued earlier by Fisher (1936), the merits of the Chicago Plan for monetary reform, created by some Chicago economists during the Great Depression based on the requirement for hundred percent reserves against demand deposits, include the attenuation of business cycle fluctuations, elimination of bank runs, and reduction of the levels of public and private debt. The revisit of the Chicago Plan by Benes and Kumhof (2012) using a dynamic stochastic general equilibrium model provide analytical evidence that the implications of the monetary reform program are strongly validated. It was also found that altering the banks' attitudes toward credit risk resulted in additional benefits, including significant steady-output gains due to the reduction or elimination of distortions such as interest-rate risk spreads, and costs of monitoring credit risks. There is also a potential for steady-state inflation as the focus of banks is directed towards the financing of investment projects as the government's ability to control broad monetary aggregates is increased. These analytical results are also consistent with the proposal for limited-purpose banking advanced by Kotlikoff (2010), which argues for confining banks to their core and legitimate function of channeling savings toward real investment. It is further argued that financial intermediation through limited purpose banking is less prone to breakdowns and that trust in the financial system would be restored.
In the aftermath of the U.S. financial crisis, there has also been a renewed focus on the role of morality and the relation between finance and good society, which is examined by Robert Shiller (2012),
As noted by Stiglitz (1994), it is also possible to identify other singular conditions for markets to be constrained Pareto efficient. The existence of risk markets, or lack thereof, would be irrelevant to Arrow-Debreu analysis if all economic agents were identical and faced with identical shocks. Such conditions would preclude the development of markets for risk allocation since the rationale for securities trading is to allocate risks between different economic agents holding different pricing expectations.
