Warren Weber joined the Federal Reserve Bank of Minneapolis in 1981 and served as senior research officer in the Research Department from 1989 to 2012, when he retired. Before joining the Bank, he taught economics at Virginia Polytechnic Institute and State University Tulane University and Duke University. He also has been an adjunct economics professor at the University of Minnesota.
Warren’s M.S. and Ph.D. degrees are from Carnegie-Mellon University. His research agenda focuses on monetary and banking theory and history, with particular emphasis on banking in the United States before 1861.
We explore the long-run demand for M1 based on a dataset comprising 38 countries and relatively long sample periods, extending in some cases to over a century. Overall, we find very strong evidence of a long-run relationship between the ratio of M1 to GDP and a short-term interest rate, in spite of a few failures. The standard log-log specification provides a very good characterization of the data, with the exception of periods featuring very low interest rate values. This is because such a specification implies that, as the short rate tends to zero, real money balances become arbitrarily large, which is rejected by the data. A simple extension imposing limits on the amount that households can borrow results in a truncated log-log specification, which is in line with what we observe in the data. We estimate the interest rate elasticity to be between 0.3 and 0.6, which encompasses the well-known squared-root specification of Baumol and Tobin.
We explore the long-run demand for M1 based on a data set that has comprised 32 countries since 1851. In many cases, cointegration tests identify a long-run equilibrium relationship between either velocity and the short rate or M1, GDP, and the short rate. Evidence is especially strong for the United States and the United Kingdom over the entire period since World War I and for moderate and high-inflation countries.
With the exception of high-inflation countries–for which a “log-log” specification is preferred–the data often prefer the specification in the levels of velocity and the short rate originally estimated by Selden (1956) and Latané (1960). This is especially clear for the United States and other low-inflation countries.
Contemporaries and economic historians have noted several features of medieval and early modern European monetary systems that are hard to analyze using models of centralized exchange. For example, contemporaries complained of recurrent shortages of small change and argued that an abundance/dearth of money had real effects on exchange, especially for the poor. To confront these facts, we build a random-matching monetary model with two indivisible coins with different intrinsic values. The model shows that small change shortages can exist, in the sense that adding small coins to an economy with only large coins is welfare-improving. This effect is amplified by increases in trading opportunities. Further, changes in the quantity of monetary metals affect the real economy and the amount of exchange as well as the optimal denomination size. Finally, the model shows that replacing full-bodied small coins with tokens is not necessarily welfare-improving.
Prior to 1863, state-chartered banks in the United States issued notes—dollar-denominated promises to pay specie to the bearer on demand. Although these notes circulated at par locally, they usually were quoted at a discount outside the local area. These discounts varied by both the location of the bank and the location where the discount was being quoted. Further, these discounts were asymmetric across locations, meaning that the discounts quoted in location A on the notes of banks in location B generally differed from the discounts quoted in location B on the notes of banks in location A. Also, discounts generally increased when banks suspended payments on their notes. In this paper we construct a random matching model to qualitatively match these facts about banknote discounts. To attempt to account for locational differences, the model has agents that come from two distinct locations. Each location also has bankers that can issue notes. Banknotes are accepted in exchange because banks are required to produce when a banknote is presented for redemption and their past actions are public information. Overall, the model delivers predictions consistent with the behavior of discounts.
This article describes a newly constructed data set of all U.S. state banks from 1782 to 1861. It contains the names and locations of all banks and branches that went into business and an estimate of when each operated. The compilation is based on reported balance sheets, listings in banknote reporters, and secondary sources. Based on these data, the article presents a count of the number of banks and branches in business by state. I argue that my series are superior to previously existing ones for reasons of consistency, accuracy, and timing. The article contains examples to support this argument.
Before 1789, the individual colonies that would ultimately make up the United States were free to issue their own currencies, and all of them did. The U.S. Constitution, adopted in 1789, took this power away from the individual states. Thus, it might appear that the Constitution left the federal government, which minted gold and silver coins, as the sole creator of currency in the new country. However, this did not turn out to be the case. Although the Constitution took away the power of states to issue money, it left them with the power to charter and regulate note-issuing banks. All of the states ultimately utilized this power, and some went as far as wholly or partially owning banks. In addition, the federal government chartered the (First) Bank of the United States from 1791 to 1811 and the (Second) Bank of the United States from 1816 to 1836. Virtually all of these banks issued notes, and these notes circulated as currency. Thus, by the early 1800s, there were far more entities issuing currency in the United States than there had ever been before 1789. The regulation of these currency issuers varied from place to place and from time to time. We have argued elsewhere (Rolnick, Smith, and Weber 1994) that the intention of the framers of the Constitution was to make the United States a monetary union or a uniform currency area.
This paper investigates antebellum interbank relationships using previously unknown data for Pennsylvania banks from 1851 to 1859 that disaggregate the amounts due from other banks by debtor bank. It finds that country banks, banks outside of Philadelphia and Pittsburgh, dealt almost exclusively with financial center banks. Most had a large, highly stable relationship with a single correspondent bank. The location of a country bank’s correspondent was consistent with trade patterns, particularly railroad and canal linkages. Philadelphia banks, in contrast, did not establish correspondent-type banking relationships. Further, Philadelphia’s correspondent banking market was not highly concentrated, and entry was easy.
Bimetallism has been the subject of considerable debate: Was it a viable monetary system? Was it desirable? In our model, the amounts of each metal are split between coined metal, satisfying a cash-in-advance constraint, and uncoined metal, yielding utility. The ratio of the monies in the cash-in-advance constraint is endogenous. Bimetallism is feasible: we find a continuum of steady states indexed by the constant exchange rate of the monies. Bimetallism is not desirable: among steady states, welfare under monometallism is higher than under any bimetallic equilibrium. Long-run trends in gold and silver production placed limits on the maintenance of bimetallism at any given ratio, but its sudden collapse in 1873 remains a puzzle.
Interest rates under the U.S. National Banking System (1863-1914) appear to imply that banks failed to exploit an arbitrage opportunity for two reasons: yields on government bonds exceeded the tax rate on note issue by approximately 150 basis points, and short-term interest rates varied seasonally. This paper examines whether note redemption costs can explain observed interest rates. We present a model in which redemption costs create a spread between the tax rate on note issue and bond yields and in which temporary seasonal fluctuations in currency demand generate seasonal movements in short-term interest rates. Calibration of the model to actual data lends support to the model’s implications. Further, interest rates are shown not to vary seasonally when banks do not incur the costs of note redemption.
Many have argued that private provision of close currency substitutes may lead to large scale indeterminacies and excessive economic fluctuations. Others argue that money creation can be “left to the market.” Adherents of this viewpoint often point to the Suffolk Banking System as an example of a well-functioning system of private money creation. We provide a framework for analyzing these notions, and for modeling the monetary consequences of the Suffolk system. This system resolves some, but not all indeterminacies. It also can raise steady-state welfare, but may substantially enhance volatility. The model’s predictions are consistent with historical evidence.
What are the conditions under which Gresham’s Law holds? And what are the mechanics of a debasement? To analyze these questions, we develop a model of commodity money with light and heavy coins, imperfect information, and prices determined via bilateral bargaining. There are equilibria with neither, both, or only one type of coin in circulation. When both circulate, coins may trade by weight or by tale. We discuss the extent to which Gresham’s Law holds in the various cases. Following a debasement, depending on the incentives offered, equilibria exist with positive seigniorage and a mixture of old and new coins in circulation.
We examine the behavior of money, inflation, and output under fiat and commodity standards to better understand how changes in monetary policy affect economic activity. Using long-term historical data for 15 countries, we find that, under fiat standards, the growth rates of various monetary aggregates are more highly correlated with inflation and with each other than under commodity standards. Money growth, inflation, and output growth are also higher. In contrast, we do not find that money growth is more highly correlated with output growth under one standard than under the other.
We establish several facts about medieval monetary debasements: they were followed by unusually large minting volumes and by increased seigniorage; old and new coins circulated concurrently; and, at least some of the time, coins were valued by weight. These facts constitute a puzzle because debasements provide no additional inducements to bring coins to the mint. On theoretical and empirical grounds, we reject explanations based on by-tale circulation, nominal contracts, and sluggish price adjustment. We conclude that debasements pose a challenge to monetary economics.
According to previous studies, the demand-liability feature of national bank notes did not present a problem for note-issuing banks because the nonbank public treated notes and other currency as perfect substitutes. However, that view, when combined with nonbindingness of the collateral restriction against note issue, itself an implication of the fact that some eligible collateral was not used for that purpose, implies that the safe short-term interest rate is pegged at the tax rate on note circulation. Since evidence on short-term interest rates is inconsistent with such a peg, that view must be rejected.
In this paper we present a consistent estimator for a linear filter (distributed lag) when the independent variable is subject to observational error. Unlike the standard errors-in-variables estimator which uses instrumental variables, our estimator works directly with observed data. It is based on the Hilbert transform relationship between the phase and the log gain of a minimum phase-lag linear filter. The results of using our method to estimate a known filter and to estimate the relationship between consumption and income demonstrate that the method performs quite well even then the noise-to-signal ratio for the observed independent variable is large. We also develop a criterion for determining whether an estimated phase function is minimum phase-lag.
This paper shows that there can be equilibria in which exchange rates display randomness unrelated to fundamentals. This is demonstrated in the context of a two-currency, one-good model, with three agent types and cash-in-advance constraints. A crucial feature is that the type i agents, for i = 1, 2, must satisfy a cash-in-advance constraint by holding currency i, while type 3 agents can satisfy it by holding either currency. It is shown that real allocations vary across the multiple equilibria if markets for hedging exchange risk do not exist and that the randomness is innocuous if complete markets exist.
This paper explains why the risky notes of banks established during the Free Banking Era (1837–1863) were demanded even when relatively safe specie (gold and silver coin) was an alternative. Free bank notes were demanded because they were priced to reflect the expected value of their backing. The empirical evidence supports this explanation. Specifically, in New York, Wisconsin, and Indiana the expected value of backing was sufficient for free bank notes to circulate at par, which they did. In Minnesota the backing for notes was very poor: they exchanged well below par, being treated as small-denomination securities.
Barnett introduced the use of neoclassical demand-side aggregation theory into monetary economics. More recently he has introduced supply-side aggregation theory into monetary economics. We show that the demand-side and supply-side exact monetary aggregates need not be equal, even if aggregation is over the same component assets on both sides of the market and if all component asset markets are cleared. The non-payment of interest on required reserves produces a classical regulatory wedge between the two sides of the aggregate market. We use time-series methods, including a new Hilbert transform method, to investigate the empirical importance of this aggregate gap.
Gresham’s law often takes two forms: the rule that bod money drives out good money and a qualified version of that rule that requires a fixed exchange rate between the two monies. Yet history contradicts both of these forms. In fact, the exchange rate has never been fixed, and we doubt it ever could be. We propose a new version of the law that is more feasible and more consistent with the evidence. It requires a fixed transaction cost of using currencies at nonpar prices for the rule to apply. Then denomination determines the fate of good money.
In this paper we propose and test a new explanation of bank behavior during the Free Banking Era, 1837–1863. Arguing against the conventional view that free bank failures were due to wildcat banking, we claim they were caused by falling asset prices. Confronting both explanations with our new and detailed data set developed from state auditor reports, we find that the falling asset price explanation of free bank failures explains far more failures than does the wildcatting hypothesis.
This paper presents a direct search approach to the optimization of multiresponse simulation models. The paper develops an approach within the framework of goal programming and uses a modified pattern search routine developed for this purpose. The algorithm and a graphical example are presented. The advantages and disadvantages of the approach determined from computational experiences with the solution procedure are discussed.
This paper examines the controversy concerning the relative desirability of fixed versus flexible exchange rates by examining whether a country can achieve a smaller variance of domestic output around its full employment path operating under an exchange rate rule or under a money supply rule. The paper finds that neither policy will always dominate the other. However, it does find that if shocks in one market of the economy are large relative to shocks to other markets, then one type of rule can be shown to dominate the other.
By examining empirically the growth of government in 34 nations during the post-World War II period , this paper explores whether it is appropriate to make references to such a phenomenon as “Wagner’s Law of Expanding State Activity.” The variation in patterns of growth in governmental spending is so substantial that one can only conclude that “Wagner’s Law” cannot be considered a “law.” Possible reformulations of the research agenda pertaining to the size and growth of government are then sketched. In short, it is suggested that attention should be turned to alternative paradigms for interpreting the disparate empirical evidence, especially focusing on the impact of alternative social institutions on patterns of government growth.
Price level variations cause changes in the real rate structure of any tax system which, as is customary, defines its rate brackets in nominal dollars. This paper traces out the effects of inflation on the real tax rates imposed on U.S. personal income between 1954 and 1970. Despite the substantial 1964-65 reduction in nominal rates, real tax rates have actually risen for many taxpaying units, especially those with low incomes or large families. The analysis suggests, therefore, that inflation has a surprisingly strong and almost certainly unintended effect on the distribution of personal income tax burdens in the U.S.
In a previous paper in this Review, I used the assumption that a representative consumer maximizes utility over a multiperiod horizon to obtain a consumption function which includes the rate of interest as an independent variable. In that study, interest rates were a statistically significant determinant of annual aggregate consumption. In this note, I examine the question whether interest rate changes are also an important determinant of quarterly consumption movements. I will also determine whether the evidence on consumer behavior obtained with the short-run data is consistent with that obtained from the long-run data.
Here you can define the content that will be placed within the current tab.
Some of the Excel files below use Pre-1900 dates that Excel does not natively handle. Download the Pre-1900 Date Function Add-In to overcome this limitation. Once downloaded, copy it to C:\Program Files\Microsoft Office\Office10\Library (for Microsoft Office XP). Then open Excel, go to Add-Ins – Tools and check the corresponding box.
- This data set consists of individual bank balance sheets for the antebellum period in the United States compiled from reports of state banking authorities.
- This file contains a listing of all banks that existed in the United States between 1784 and 1860 along with their opening and closing dates. Further, if a bank went out of existence, its disposition – whether it closed, failed, or other – is given. For the methodology to obtain beginning and ending dates see Early State Banks in the United States: How Many Were There and When Did They Exist? Journal of Economic History, June 2006: 433-455.
- This spreadsheet contains the disaggregated national bank call reports by state and reserve city for each call report date. These data appear as compiled by the Comptroller of the Currency. These data are a “cleaned” version of the data published in the Annual Reports of the Comptroller of the Currency. Where assets and liabilities were not equal for a state or reserve city in the original, they have been corrected to be equal in this data set. This was done by comparing for each asset and liability category differences between totals as reported by the Comptroller and totals category obtained by aggregating the individual state and reserve city data. It should also be noted that aggregates for the entire National Banking System should be based on the individual data in this data set and not those reported by the Comptroller. After 1900 the dates for the data for Alaska and Hawaii that the Comptroller used in his totals do not match the dates given in the individual state reports.
- This spreadsheet contains data for Argentina, Brazil, Canada, Chile, France, Germany, Italy, Japan, Netherlands, Norway, Portugal, Spain, Sweden, UK, and US for the period 1810 – 1995. The data reported are for specie, M0, M2, prices, and output. The results in Rolnick-Weber, Journal of Political Economy (1997) are based on the data in this spreadsheet. For a description of how the data are constructed, see Rolnick and Weber, Staff Report 175 (1995).
- Dividends of Massachusetts Banks, 1807-1861
- Quoted Discounts on State Bank Notes in New York, Cincinnati, and Cleveland, selected dates, 1817-1858
- Quoted Discounts on State Bank Notes in Philadelphia, 1832-1858
- Railroad stock prices, quarterly indices