«Abstract While payment card usage has increased dramatically, the stock of outstanding currency has not declined as rapidly. We analyze changes in ...»
For these regressions, we divide cash outstanding/GDP into three categories— large, medium, and small denominations. The medium note category was chosen by surveying the central banks of the respective countries to determine which denomination was prevalently distributed by ATMs. Denominations above this threshold were categorized as “large,” while those below this threshold (including coins) were categorized as “small.” In some countries, only a single currency note is commonly distributed by ATMs (e.g., the $20 note in the United States). In other countries (e.g. Germany), ATMs routinely provide a set of different denominations. In such cases, we categorize all of the ATM-distributed denominations as medium. Table 4 summarizes the data on currency denominations and ATM conventions in each of the countries in our sample. If the main denomination for ATM withdrawals is also the highest denomination (as is the case in Japan with the ¥10,000 note), we categorize it as large.
While the ATM-dispensed banknotes form the bulk of transactional currency demand by volume, they represent a potentially noisy data source for identifying links between the growth of alternative payment instruments and cash demand. We expect smaller denomination notes and coins to be a good gauge of the changing needs of merchants and other receivers of cash to make change for purchases. Since change is given only in the case of cash transactions, changes in the stock of small notes and coins likely provide the most accurate reflection of fluctuations in transactional demand for currency.14 There are several additional benefits from partitioning the currency by denomination.
First, by isolating ATM-distributed notes we attempt to provide a more precise estimate on what we term the ATM demand factor. Given that consumers are primary ATM users and that, in our sample countries, ATMs are the main source of cash for consumers, the primary demand for small denomination notes and coins is by receivers of cash to make change. By analyzing nonATM-distributed notes, we are able to focus on merchant demand for currency. Second, if the share of self-employed affects currency demand by facilitating black market activity, we would expect to identify that relationship in the large denomination category.15 If, on the contrary, the self-employed stay away from electronic payments for cost reasons, we would expect to find an effect on demand for small denominations. Finally, since the bulk of foreign currency demand is limited to large denominations, we expect the small- and medium-notes demand estimates to be quite robust to restricting the data sample to countries without substantial foreign demand for their currency.
Table 5 shows the results of fitting a country fixed effects model with cluster-adjusted standard errors to each denomination category. We chose the FE cluster-adjusted specification since its non-parametric formulation gives us a chance to get standard errors right without making strong assumptions. It is also the model in which debit terminals had the weakest statistically identifiable effect on aggregate currency demand (see Table 3). Thus, it sets a somewhat higher hurdle for the denomination-specific exercise.
Panel A of Table 5 contains full sample estimates. As hypothesized, debit card terminals have a very strong negative effect on demand for small denominations, but not for medium or large denominations. This underscores the idea that by eliminating the need to make change, electronic payment instruments have the most measurable effect on the subset of currency used for this purpose. Also, the ATM infrastructure is found to affect only the ATM-dispensed medium category. This effect is fairly precisely identified (p-value of 0.04) and is negative, indicating that, on net, ATM proliferation lowers the stock of medium denomination notes. The bank branch infrastructure has a significant (and positive) effect only on bank-distributed denominations – i.e. those that cannot typically be obtained from ATMs. The ratio of selfemployed is influential only for small denomination demand, consistent with the hypothesis of high fixed costs of installing electronic payments terminals for small merchants.16 Interestingly, we fail to find a statistically measurable effect of short-term interest rates on disaggregated currency demand. Even the large denomination notes, which in their role as store of value should reflect the opportunity costs of currency holdings, have at best a marginally significant effect (p-value of 0.11). However, to the extent that the full sample does not account fully for foreign currency holdings, the estimated response to fluctuations in domestic short-term rates may be muted. Thus, we repeat the exercise on the truncated sample that excludes Germany, Switzerland, and the United States.
Panel B of Table 5 presents the results for the truncated sample. This time, there is a very strong and precisely identified effect of changes in short-term interest rates on changes in largedenomination currency holdings. Importantly, all other results are very similar to those obtained for the full sample, consistent with the expectation of more stable coefficient estimates across samples. In addition, the full and truncated samples have similar results for small and medium denomination classes, consistent with the assumption that these denominations are less likely to circulate in foreign countries.17 The specifications in Table 5 do not include time fixed effects in addition to country fixed effects to avoid overfitting a fairly small data set. However, this may raise concerns that the negative relationship between the stock of small denomination currency and the extent of debit card infrastructure is due to secular time trends, which are spuriously reflected in these variables. We believe this concern to be misplaced. Conceptually, the observation that only cash transactions require making change establishes a strong prior of a direct link between the decline in change-making currency and the increase in non-cash payment infrastructure. For this relationship to be due to passage of time (or some other omitted strongly trended variable), one needs to argue that fewer transactions involved small notes and coins, whether for execution or change-making, at the end of our sample than in the beginning. With respect to the former, fairly stable inflation rates over our sample period make it unlikely that the average size of cash transactions changed enough to cross the threshold into the medium-denomination notes. With respect to the latter, price rounding has been rare in countries in our sample. In few places where it occurred, the magnitude of rounding has been far from levels necessary to raise the likelihood of making change for cash transactions in anything other than small denomination currency.18
5. Conclusion Despite the strong growth in the adoption of electronic payments throughout the 1990’s and the early part of the 21st century, cash usage remains significant in most OECD countries.
Although, generally, both consumers and merchants may prefer debit cards to other payment alternatives for certain types of transactions, the resilience of cash demand suggests that electronic alternatives have not succeeded in mimicking all the benefits of cash.
A key feature of our approach is the ability to disentangle cash’s dual roles—store of value and payment medium. We are able to isolate the transactional role of cash by focusing on the small denomination class which we define as currency and coin that are lower in value than that commonly dispensed by ATMs. We find that the effects of substitution to electronic payments are largely confined to demand for small-denomination currency, typically used to make change for purchases. Since change is given only in the case of cash transactions, changes in the stock of small notes and coins provide a better gauge of fluctuations in transactional demand for currency. We believe that our paper is the first to attempt identification of cash demand through its “trickling down” to change-making small denomination notes and coins.
We also find that the demand for small denomination currency is not affected by changes in the interest rate, a key feature of the Baumol-Tobin money demand model. Our result is consistent with Alvarez and Lippi (2007) who also find that financial innovations reduce the interest rate elasticity of certain denominations. Their study showcases a stochastic cash inventory model motivated by proliferation of ATMs, and thus the empirical findings focus on ATM-dispensed currency (our medium denomination category). Our finding of substantial interest rate sensitivity of demand for high denomination notes (especially in countries that do not have significant proportions of their currency stock circulate outside of their borders) suggests a persistent role for cash as a store of wealth.
There are several broader long-term implications that follow from our results. First, as ongoing financial innovation lowers the cost of participating in financial markets and raises the returns of doing so, cash holdings may lose some appeal as store of wealth. In this case, currency stocks that are increasingly comprised of large denomination notes, will primarily reflect demand for underground transactions in which anonymity is highly valued. This type of demand for currency may be less sensitive to interest rate movements, potentially resulting in more stable money demand functions. Second, from a public policy perspective, currency may be the preferred medium for the government to collect taxes in the form of seigniorage particularly from otherwise untaxed underground activity. Third, greater conversion to debit cards where change is not required may also decrease the price stickiness of certain types of goods such as vending machine and newspaper purchases that have historically adjusted very infrequently. Indeed, recent research has found evidence of faster price increases for traditionally cash goods (e.g. tollway fees) as payments for them became predominantly electronic (Finkelstein 2007). Moreover, some industry experts argue that consumers are likely to spend more on goods and services when paying electronically because they are not restricted by cash-in-hand and potential psychological constraints.
In sum, our empirical results suggest that electronic alternatives to cash will indeed reduce the demand for cash of certain denominations. However, the general demand for cash will continue to be strong in the future because of cash’s anonymity and store of value features.
In addition, greater usage of OECD currencies in many emerging economies may result in a shift from domestic money demand in the country of currency issuance to money demand by residents in other countries.
Alvarez, Fernando, Andrew Atkeson, and Chris Edmond. (2003) “On the Sluggish Response of Prices to Money in an Inventory-Theoretic Model of Money Demand.” NBER Working Paper No. 10016.
Alvarez, Fernando, and Francesco Lippi. (2007) “Financial Innovation and the Transactions Demand for Cash.” NBER Working Paper No. 13416.
Amromin, Gene, and Sujit Chakravorti. (2007) “Debit Card and Cash Usage: A Cross-Country Analysis.” Federal Reserve Bank of Chicago Working Paper, WP 2007-04.
Baumol, William J. (1952) “Transactions Demand for Cash: An Inventory Theoretic Approach.” Quarterly Journal of Economics, 66:4, 545-56.
Boeschoten, Willem C. (1992) Currency Use and Payment Patterns, Financial and Monetary Policy Studies, 23, Dordrecht/Boston/London: Kluwer Academic Publishers.
Borzekowski, Ron, and Elizabeth K. Kiser. (2006) “The Choice at the Checkout: Quantifying Demand across Payment Instruments.” International Journal of Industrial Organization, 26:4, 889-902.
Bounie, David, and Abel François. (2006) “Cash, Check or Bank Card?: The Effects of Transaction Characteristics on the Use of Payment Instruments.” Telecom Paris Working Papers in Economics and Social Sciences ESS-06-05.
Carbo Valverde, Santiago, David B. Humphrey, and Rafael Lopez del Paso. (2003) “The Falling Share of Cash Payments in Spain.” Moneda y Credito, 217, 167-90.
Chakravorti, Sujit. (2006) “The Evolving Payments Landscape and its Implications for Monetary Policy.” In Institutional Change in the Payments System and Monetary Policy, edited by Stefan W. Schmitz and Geoffrey Wood, pp. 81-92. London: Routledge.
Chakravorti, Sujit, and Timothy McHugh (2002), “Why Do We Use so Many Checks?” Federal Reserve Bank of Chicago Economic Perspectives, 26:3, 44-59.
Committee on Payment and Settlement Systems (various years), Statistics on Payment and Settlement Systems in Selected Countries. Basle, Switzerland: Bank for International Settlements.
De Grauwe, Paul, Erik Buyst, and Laura Rinaldi. (2000) “The Costs of Cash and Cards Compared: The Cases of Iceland and Belgium.” mimeo, University of Leuven.
Drehmann, Mathias, Charles Goodhart, and Malte Krueger. (2002) “The Challenges Facing Currency Usage: Will the Traditional Transaction Medium be able to Resist Competition from new Technologies?” Economic Policy, April, 195-227.
Duca, John V., and David D. VanHoose. (2004) “Recent Developments in Understanding the Demand for Money.” Journal of Economics and Business, 56:4, 247-72.
The Economist. (2007) “The End of the Cash Era.” February 17-23, cover page and 71-73.
European Central Bank (various years), Payment and Securities and Settlement Systems in the European Union. Frankfurt, Germany: European Central Bank.
Finkelstein, Amy. (2007) “EZ-Tax: Tax Salience and Tax Rates.” mimeo, MIT.
Fischer, Bjorn, Petra Kohler, and Franz Seitz. (2004) “The Demand for Euro Area Currencies:
Past, Present, and Future.” European Central Bank Working Paper Series, N. 330.
Flannery, Mark J. (1996) “Technology and Payments: Déjà Vu All Over Again?” Journal of Money Credit and Banking, 28:4, Part 2, 965-70.
Flannery, Mark J., and Dwight M. Jaffee. (1973) The Economic Implications of an Electronic Monetary Transfer System. Lexington: Lexington Books.
Freedman, Charles. (2000) “Monetary Policy Implementation: Past Present and Future—Will the Advent of Electronic Money Lead to the Demise of Central Banking?” International Finance, 3:2, 211-27.
Friedman, Benjamin M. (1999) “The Future of Monetary Policy: The Central Bank as an Army with Only a Signal Corps?” International Finance, 2:3, 321-38.
Forbes. (1967) “Electronic Money.” April 1, 42-46.