«Abstract While payment card usage has increased dramatically, the stock of outstanding currency has not declined as rapidly. We analyze changes in ...»
Goodhart, Charles A. E. (2000) “Can Central Banking Survive the IT Revolution?” International Finance, 3:2, 189-209.
Humphrey, David B., Aris Kaloudis, and Grete Øwre. (2004) “The Future of Cash: Falling Legal Use and Implications for Government Policy.” Journal of International Financial Markets, Institutions, and Money, 14:3, 221-33.
Humphrey, David B., Lawrence B. Pulley, Jukka M. Vesala. (1996) “Cash, Paper, and Electronic Payments: A Cross-Country Analysis.” Journal of Money, Credit, and Banking, 28:4, 914-39.
International Monetary Fund (various years), International Financial Statistics. Washington, DC: International Monetary Fund.
Klee, Elizabeth. (2006) “Families’ Use of Payment Instruments During a Decade of Change in the U.S. Payment System.” Federal Reserve Board, FEDS working paper 2006-02.
Kroszner, Randall S. (2003) “Currency Competition in the Digital Age.” In Evolution and Procedures in Central Banking, edited by Dave Altig and Bruce D. Smith, pp. 275-99. New York: Cambridge University Press.
Lambert, Michael J., and Kristin D. Stanton. (2001) “Opportunities and Challenges for the U.S.
Dollar as an Increasingly Global Currency: A Federal Reserve Perspective.” Federal Reserve Bulletin, September, 567-75.
Markose, Sheri M., and Yiing Jia Loke. (2002) “Can Cash Hold Its Own? International Comparisons: Theory and Evidence.” mimeo, University of Essex.
Petersen, Mitchell. (2006) “Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches,” Northwestern University, working paper.
Porter, Richard D., and Ruth A. Judson. (1996) “The Location of US Currency: How Much is Abroad?” Federal Reserve Bulletin, October, 883-903.
Rogers, William.H. (1993) “Regression Standard Errors in Clustered Samples.” Stata Technical Bulletin, 13, 19-23.
Rogoff, Kenneth, (1998) “Blessing or Curse? Foreign and Underground Demand for Euro Notes.” Economic Policy, 13:26, 261-303.
Schneider, Friedrich. (2006) “Shadow Economies and Corruption All Over the World: What Do We Really Know?” IZA Discussion Paper No. 2315.
Seitz, Franz. (1995) “The Circulation of the Deutsche Mark Abroad.” mimeo, Deutsche Bundesbank Shy, Oz, and Juha Tarkka. (2002) “The Market for Electronic Cash Cards.” Journal of Money, Credit and Banking, 34:2, 299-314.
Stix, Helmut. (2003) “How Do Debit Cards Affect Cash Demand? Survey Data Evidence.” Empirica, 31: 2-3, 93-115.
Tobin, James. (1956) “The Interest Elasticity of Transactions Demand for Cash.” Review of Economics and Statistics, 38:3, 241-47.
U.S. Department of Treasury. (2003) The Use and Counterfeiting of United States Currency Abroad, Part 2. Washington, DC: Department of the Treasury.
Van Hove, Leo. (2004) “Cost-Based Pricing of Payment Instruments: The State of the Debate.” De Economist, 152:1, 79-100.
Whitesell, William. (1992) “Deposit Banks and the Market for Payment Media” Journal of Money, Credit, and Banking, 24:4, 483-98.
Woodford, Michael. (2000) “Monetary Policy in a World Without Money.” International Finance, 3:2, 229-60.
Data Appendix We employ three different data sources. For payment data, currency outstanding, population, and banking infrastructure, we use the Bank for International Settlements’ Committee of Payment Settlement Systems’ Statistics on Payment and Settlement Systems in Selected Countries (Red Book) and the European Central Bank’s Payment and Securities Systems in the European Union (Blue Book).19 We also contacted several central banks for clarification and to fill in some of the missing data. For short-term interest rates, foreign exchange rates, gross domestic product, and the ratio of self-employed to total employment, we use the International Monetary Fund’s International Financial Statistics.
The dependent variable in our regressions is the ratio of outstanding currency to GDP.
The Red and Blue books provide currency outstanding for each note denominations and coin type. Cash outstanding includes holdings by businesses, individuals and financial institutions but excludes cash held at central banks. For euro countries, we only consider the period from 1988 to 1999, which eliminates the last two pre-Euro years in which currency stocks (particularly those of large-denomination notes) started to decline in anticipation of the regime switch. For the other countries, the sample period is from 1988 to 2003.
In several countries higher denomination coins were introduced during our sample period, replacing, in some cases, paper notes of similar value. In these cases, we kept the coin and note within the same denomination category. In addition, certain currency denominations were replaced by new denominations. In these cases, we kept both new note and the old note that it was replacing in the same category.
As mentioned before, Germany, Switzerland, and the United States have a sizable fraction of their currencies held outside their countries’ borders. Since foreign holdings are much less likely to be influenced by domestic cash and debit card infrastructure, we need to adjust our cash measures. In the case of the United States, we use the Federal Reserve Board’s Flow of Funds Accounts series on currency holdings abroad. For Germany, we utilize the time series estimates of Fisher, Kohler, and Seitz (2004).20 Data limitations prevent us from making any adjustments for Swiss currency stocks.
The following variables were taken from the Red and Blue books: number of ATMs, number of POS debit card terminals, number of debit cards, and number of bank branches. Each of these variables was normalized by per capita terms by dividing each variable by the population of each country during that year with the exception of POS debit terminals. POS debit terminals were normalized by 1,000 individuals.
As in any cross-country study, certain variables may not be defined consistently. For example, definitions of debit cards vary somewhat across countries. Although “debit” transactions are usually associated with nearly immediate withdrawal of funds, in some countries (e.g. France) “debit” cards include those in which the funds may be withdrawn up to one month following the transaction. Because in most countries in our sample, PIN-based debit cards are much more widely used than signature-based cards and data for PIN-based transactions are more consistent, we focus on the usage of PIN-based debit cards. In most countries in our sample, all debit cards are PIN-based.
Some recent studies find evidence linking the gradual reduction in check transactions to growing usage of debit cards in the Unites States (Borzekowski and Kiser 2006; Klee 2006).
Duca and VanHoose (2004) review this literature.
For more recent models using an inventory approach, see Alvarez, Atkeson, and Edmond (2003).
Lambert and Stanton (2001) and Porter and Judson (1996) discuss foreign demand of U.S.
For theoretical models that address the link between size of transaction and payment instrument, see Shy and Tarkka (2002) and Whitesell (1992).
For robustness, we repeated the analysis with currency holdings normalized by total household expenditures. None of the reported results was affected by this modification.
As a consequence, we have an unbalanced panel in which non-euro countries are overrepresented.
Using transactions data directly to explain fluctuations in cash demand presents an econometric problem. A flow measure of debit card use is an endogenous variable, in the sense that it is influenced by the same factors that affect transactional cash demand. Some of these factors may be unobserved or unavailable, making coefficient estimates on debit card transactions both biased and inconsistent. One example of such an omitted factor is the cost of debit card transactions, which clearly influences both cash and debit use in transacting payments.
In Amromin and Chakravorti (2007), we regress debit card transactions on the number of debit cards in circulation and debit card terminals. We find that the installation of debit card terminals is able to best explain the increase in debit card transactions.
For more discussion of our data set, see the data appendix.
We discuss direct inclusion of tax burden measures into the model in section 4.E.
The results for a full sample of 13 OECD countries are qualitatively similar, although they are somewhat less stable across different econometric models. They are available in the older version of the paper (Amromin and Chakravorti 2007). In the subsequent analysis of denomination-specific demand, we report the results for both sets of countries, as foreign demand is limited to large-denomination currency.
We also estimated the model under the assumption of first-order autocorrelated error terms, as well as the random-effects model with clustered standard errors. Both of those methods of accounting for within-country serial correlation resulted in somewhat smaller changes to the pooled OLS estimates.
In the United States, some merchants, such as grocery stores, are willing to give cash back to their debit card customers. This service is analogous to cash back received for checks, a common means to acquire cash in the 1980s. Some analysts have suggested that these merchants provide cash back services to reduce their cash handling costs. However, this practice is not widespread for most countries in our sample.
Arguably, even the lemonade stand operator wishing to avoid tax payments would at some point have to convert coins and small denomination notes it collects for larger notes suitable for storing wealth or transacting.
We experimented with adding various measures of the tax burden to the regression. Among the variables we evaluated were: the statutory point-of-sale tax rates (VAT or the weighted sales tax rate), average marginal tax on earned income, and both the contemporaneous and the lagged ratio of tax revenues to GDP. In each case, we failed to detect a significant direct effect of tax burden measure on demand for any denomination class. This is consistent with at least some studies of cash demand in OECD countries (Boeschoten 1992; Doyle 1999). More importantly, none of the results reported in Table 5, was materially affected by any of the tax variables.
In Amromin and Chakravorti (2007), we also report regressions using DGK breakdown.
While we find similar qualitative results, the precision of the results improves by considering three categories instead of two.
For instance, Finland and the Netherlands do not issue eurocent coins and encourage merchants to use 5-cent increments in setting prices. Although this development took place outside our sample window, it is indicative of the type of denomination-specific change associated with rounding.
We used the Blue Book for Austrian and Finnish data.
Since holdings of domestic currency abroad are heavily concentrated in large denomination notes (Porter and Judson 1996), we subtract the estimated foreign holdings from large denomination notes series for both Germany and the United States.
Sources: Bank for International Settlements’ Committee of Payment Settlement Systems’ Statistics on Payment and Settlement Systems in Selected Countries (Red Book) and the European Central Bank’s Payment and Securities Systems in the European Union (Blue Book) Note: Data on currency/GDP ratio are adjusted for currency circulating abroad (see Appendix for details). The diamonds depict the relationship between the aggregate currency/GDP ratio and the debit card infrastructure in 1991, while the hollow dots depict the same realtionship at the end of the decade. The corresponding dashed lines represent cubic splines fitted to the data points in a given year.
Table 1. Number of Per Capita Non-Cash Payments by Type of Instrument, 2005
Table 3. Various Econometric Models of Aggregate Currency Demand (1) (2) (3)