Nonstandard Preferences and Monetary Policy
Published on 24. October 2014
Book
VIII, 103 pages
978-3-86624-610-2 (ISBN)
Description
Not only since Daniel Kahneman won the Nobel Prize in Economics, have nonstandard preferences and the discipline of behavioral economics become popular subjects of discourse both in the academic arena and the public forum. University classes in nonstandard preferences are quickly booked out, and its concepts have wandered into popular culture from the printed page to the wide screen of Hollywood. More importantly, however, the perspective of nonstandard preferences is increasingly influencing the analysis and design of public policy.
For instance, consider the case of corporate retirement saving plans. There are, essentially, two possible systems. Either an individual is not enrolled in a pension plan unless he or she actively elects to participate in one or the individual is automatically enrolled unless he or she choses to opt out of the plan. For a fully rational individual with the kind of preferences that standard economic theory assumes, these two systems are equivalent. But as Madrian and Shea (2000) show, making enrollment the default option can significantly increase retirement savings. Possible reasons could be inertia in decision making or procrastination. Whatever the actual reason, results like these have prompted policymakers to react. In the UK, legislation is currently coming into effect that requires all employers to automatically enroll their workers in corporate pension plans.
In this thesis, we review a subset of the near infinite set of nonstandard preferences and use two specific preference formulations to analyze their effects on monetary policy. The monetary framework we use, falls into the class of search theoretic models in the vein of Lagos and Wright (2005). The Lagos and Wright framework is based on explicit frictions that make money essential in the sense that money expands the set of possible allocations and can thereby be welfare increasing. Trade between agents in this framework is typically decentralized and happens in pairwise meetings, thereby creating a double coincidence of wants problem. In the absence of a record-keeping technology and public memory, these frictions imply that in any trade there must
be immediate settlement with (fiat) money. Money hence is essential. In contrast to previous monetary models that make money intrinsically valuable or impose cashin-advance constraints, the introduction of these frictions explicitly gives money a medium of exchange role in these models. Such models are used in Chapters II and III.
Chapter I delves into the vast literature on nonstandard preferences and their use in macroeconomics. It is by no means meant to be an all encompassing review of the entire literature, but rather an inquiry into the evidence for and the use of three particular types of preferences. The first subset of preferences that we analyze offer scenarios for explaining people's desire to commit. These preferences, including hyperbolic discounting - which we shall meet again in Chapter III - can shed light on questions such as: Why many people chose to role over credit card debt at relatively high interest rates, while at the same time holding liquid assets that pay relatively low interest? Chapter I then continues to look at preferences that include relative comparisons. These preferences evolve around the idea that humans care less about absolutes and more about relatives. These preferences can, for example, play a role in explaining the equity premium puzzle. These preferences play a large role in Chapter II. The closely related topic of reference-dependent preferences then forms the last building block of Chapter I.
In Chapter II, we introduce a concept from the literature on relative preferences into a monetary model. In a search theoretic model of money, 'keeping up with the Joneses' preferences create a consumption externality that leads to overconsumption. The monetary authority can discourage such overconsumption by raising interest rates. We show that the Friedman rule need no longer be optimal and that the welfare costs of inflation are significantly lower than the standard model suggests. Hyperbolic discounting - the idea that an individual's discount rate is not constant - can have similar effects on optimal monetary policy, as discussed in Chapter III. The third and final chapter again uses a search theoretic model and extends it to include hyperbolic time preferences. The introduction of such preferences again leads to a higher optimal interest rate and lower welfare costs of inflation.
For instance, consider the case of corporate retirement saving plans. There are, essentially, two possible systems. Either an individual is not enrolled in a pension plan unless he or she actively elects to participate in one or the individual is automatically enrolled unless he or she choses to opt out of the plan. For a fully rational individual with the kind of preferences that standard economic theory assumes, these two systems are equivalent. But as Madrian and Shea (2000) show, making enrollment the default option can significantly increase retirement savings. Possible reasons could be inertia in decision making or procrastination. Whatever the actual reason, results like these have prompted policymakers to react. In the UK, legislation is currently coming into effect that requires all employers to automatically enroll their workers in corporate pension plans.
In this thesis, we review a subset of the near infinite set of nonstandard preferences and use two specific preference formulations to analyze their effects on monetary policy. The monetary framework we use, falls into the class of search theoretic models in the vein of Lagos and Wright (2005). The Lagos and Wright framework is based on explicit frictions that make money essential in the sense that money expands the set of possible allocations and can thereby be welfare increasing. Trade between agents in this framework is typically decentralized and happens in pairwise meetings, thereby creating a double coincidence of wants problem. In the absence of a record-keeping technology and public memory, these frictions imply that in any trade there must
be immediate settlement with (fiat) money. Money hence is essential. In contrast to previous monetary models that make money intrinsically valuable or impose cashin-advance constraints, the introduction of these frictions explicitly gives money a medium of exchange role in these models. Such models are used in Chapters II and III.
Chapter I delves into the vast literature on nonstandard preferences and their use in macroeconomics. It is by no means meant to be an all encompassing review of the entire literature, but rather an inquiry into the evidence for and the use of three particular types of preferences. The first subset of preferences that we analyze offer scenarios for explaining people's desire to commit. These preferences, including hyperbolic discounting - which we shall meet again in Chapter III - can shed light on questions such as: Why many people chose to role over credit card debt at relatively high interest rates, while at the same time holding liquid assets that pay relatively low interest? Chapter I then continues to look at preferences that include relative comparisons. These preferences evolve around the idea that humans care less about absolutes and more about relatives. These preferences can, for example, play a role in explaining the equity premium puzzle. These preferences play a large role in Chapter II. The closely related topic of reference-dependent preferences then forms the last building block of Chapter I.
In Chapter II, we introduce a concept from the literature on relative preferences into a monetary model. In a search theoretic model of money, 'keeping up with the Joneses' preferences create a consumption externality that leads to overconsumption. The monetary authority can discourage such overconsumption by raising interest rates. We show that the Friedman rule need no longer be optimal and that the welfare costs of inflation are significantly lower than the standard model suggests. Hyperbolic discounting - the idea that an individual's discount rate is not constant - can have similar effects on optimal monetary policy, as discussed in Chapter III. The third and final chapter again uses a search theoretic model and extends it to include hyperbolic time preferences. The introduction of such preferences again leads to a higher optimal interest rate and lower welfare costs of inflation.