The Next Generation of Monetary Models
Economists ponder the mysteries of money at Minneapolis Fed conference.
Published September 1, 2004 | September 2004 issue
"What is the thing called money?" asked University of Minnesota economists John Kareken and Neil Wallace over 25 years ago. The question seems elementary, but it links intimately with others far more complex: How does the value of money relative to goods and services get determined? Why do people use money as opposed to a number of other possible trading arrangements? Can government policy improve economic welfare by changing the forms of money that a nation uses? Economists have wrestled with these issues for decades—for centuries, in fact—grappling with subtle interrelationships among time, inflation, interest, banks and the thing called money.
To a substantial degree, these efforts have aimed at spanning the gap between what economists know about the microeconomic behavior of firms and households and their understanding of the broader macroeconomy: national economies, aggregate money supplies and the operations of central banks. Key to this effort is developing a solid theory of how money works in an economy.
In December 1978, Kareken and Wallace organized a conference at the Minneapolis Fed to focus on such issues. The 37 participants—including many of economics' leading scholars—struggled to develop economic models that would capture money's elusive character. The conference proceedings, published in a volume titled Models of Monetary Economies, became a classic in the field, providing the bedrock for future theoretical work. But many questions remained.
The second generation
Over a quarter century later, the Minneapolis Fed held a second conference, Models of Monetary Economies II, co-sponsored by the University of Minnesota, to review what has been learned, to focus on current debates and, implicitly, to set forth the agenda of monetary economics for coming decades. The planners of this "next generation" conference were Fed adviser Narayana Kocherlakota of Stanford University and Randall Wright of the University of Pennsylvania.
Over 130 economists attended the two-day conference, held May 21 and 22, 2004, at the University of Minnesota. Both days consisted of two sessions, with three papers delivered and discussed in each. Underlying this discussion were several shared understandings:
- Money plays a number of roles: It is a medium of exchange, a unit of account, a store of value. People hold onto money because it makes buying things easy—economists call this "the transaction motive." They also hold it because it gives them a sense of security that they can buy things in the future—the "precautionary motive." But holding money also entails risk: Inflation decreases the value of money day by day, and thieves can steal it.
- Money has several forms: Gold coins are a form of commodity money—money with intrinsic value. Bills and token coins are considered representative money when they are claims to things that have intrinsic value. Otherwise, bills and token coins, which have no intrinsic value, are considered fiat money.
Day 1: Money models and monetary policy
The three papers presented in the first session went well beyond these common understandings. The first paper, presented by Ruilin Zhou of the Chicago Fed and Pennsylvania State University, analyzed the efficiency of an expansionary monetary policy relative to laissez-faire or contractionary policies and suggested that monetary policy can play a role in redistribution of wealth and welfare. William Roberds of the Atlanta Fed delivered the session's second paper, which pointed out that money provides privacy. It allows anonymous exchanges, which can be of social value "in economic situations where the parties in the transaction cannot trust each other." In the third paper, Brown University's Peter Howitt developed a model in which fiat money is used as a medium of exchange in an economy organized into "shops"—that is, specific locations where buyers and sellers meet to exchange money for products (a model resembling the real world)—and showed that fiat money is essential to the efficient operation of the economy because it saves in transaction costs.
The second session dealt with monetary policy explicitly. The first of these papers, delivered by Minneapolis Fed visiting scholar Ricardo Lagos of New York University, looked at the impact of inflation on economic output. Inflation motivates people to not hold money since the value of their (noninterest-bearing) cash is eroded by inflation, and the model developed in this paper demonstrated that efforts to avoid this "inflation tax" always reduce welfare and often reduce output.
Antoine Martin of the Kansas City Fed presented the second paper of the afternoon, analyzing the redistributive impact of monetary policies and indicating that if people in an economy differ in their preference for holding money, a monetary policy that sets nominal interest rates at zero may not maximize social welfare, contrary to conventional wisdom. In the final paper of the day, Nobuhiro Kiyotaki of the London School of Economics presented a model with a liquid asset—money—and an illiquid asset—capital—and then used it to explore the effect of a binding liquidity constraint on asset prices, savings behavior and resource allocation.
Day 2: Imperfections and banking
The conference's third session focused on economies with imperfect information and participation. Notre Dame's Christopher Waller delivered the day's first paper, looking at the issue of monetary neutrality—that is, does increasing the money supply increase economic output? This model showed that an unexpected increase in money supply can have "beneficial effects in the short run ... even though it is neutral in the long run" and that under some conditions, money injections can permanently increase output by altering the distribution of money balances.
In the day's second paper, Shouyong Shi of the University of Toronto addressed two paradoxes in monetary theory: Why do government bonds pay interest, though they're virtually risk-free? And why don't people use them as a medium of exchange?
Shi analyzed these questions by integrating nominal bonds into existing models of the microeconomic foundations of monetary theory.
Allen Head of Queen's University gave the third paper, describing a model to explain why consumers face a greater range of prices for goods when inflation rates are high. The model suggests that price dispersion results from buyers having incomplete information about prices offered by different sellers; increases in money supply will increase inflation and price dispersion and lower welfare unless buyers are induced to search more widely for lower prices, a costly effort in itself.
The final session, focusing on money and banking, was chaired by John Kareken, so it was appropriate that his former colleague at the University of Minnesota, Neil Wallace, was the next presenter. Wallace described a model that explored the implications of "inside" and "outside" money—that is, money issued by private banks and money issued by the government. The distinction has historical relevance, noted Wallace, but also contemporary application: Discount window operations, intraday credit and regulation of stored value cards all represent current challenges on the border between inside and outside money.
The next paper, given by Ricardo Cavalcanti of the Getulio Vargas Foundation, also developed a model with privately issued money to study the connection between inside money, the velocity (that is, the speed of circulation) of bank liabilities and the regulatory environment that affects banking profitability, concluding that "monetary stability depends on central-bank policies and liquidity conditions in general."
And the final paper of the conference was presented by Randall Wright of the University of Pennsylvania. The paper, said Wright, built a new theory of money and banking based on the 17th century origins of banking: goldsmiths who issued notes to those who deposited gold with them for safekeeping. The model uses a "search" framework in which buyers and sellers exchange in centralized and decentralized markets; in those markets, checks drawn on bank deposits may be useful as a means of payment because cash is subject to theft, just as gold was in the 17th century.
Twelve papers and much discussion later, the mysteries of money remained intact. Still, the economists had shared and debated new paths for linking monetary models to the real world of money, and they had reached a fuller—if still incomplete—understanding of how money works in an economy.
In addition, they'd reinforced the partnership between the Minneapolis Fed and the University of Minnesota that sparked the 1978 conference. And they'd bridged the first generation of monetary economies, led by Kareken and Wallace, to the second generation, for whom the intellectual challenge of money retains an undeniable allure.