Overview

The advent of new payments technologies including new applications for rapid payments, stablecoins and other innovations raises important questions about the long-run outlook for the “currency franchise” of major central banks.  Perhaps the most basic question is whether a physical “token” like central bank currency will go the way of vinyl records and rotary phones over time and, if so, what that might imply for central bank and government finances.  This note offers some general observations on this issue along with a few simple calculations aimed at providing rough estimates of the potential financial implications that a sizable shift away from central bank physical currency might have for central bank and government finances.

To preview the main results, under plausible alternative scenarios for the path of currency in circulation over time, and assuming that central banks do not compete with stablecoins through their own issuance of digital currencies, central bank net incomes could be significantly diminished relative to a baseline scenario in which the ratio of currency to nominal GDP remains roughly constant.  For example, based on two alternative scenarios, rough calculations suggest that a significant shift away from physical currency holdings over time could reduce cumulative Federal Reserve net income over the period 2025 to 2055 by $1.5 trillion to $2.5 trillion relative to the baseline scenario.

The remainder of this note proceeds as follows.  Section 1 provides some background information on the balance sheets of major central banks.  Section 2 discusses the value of the currency franchise for central banks with a particular focus on the Federal Reserve.  Section 3 works through some back-of-the envelope calculations for long-run projections of the value of the currency franchise in a baseline scenario along with two alternative scenarios in which the demand for physical currency declines relative to the baseline scenario.  Section 4 concludes.

Background on Central Bank Balance Sheets

As shown in figure 1, one of the most notable features of central bank balance sheets over the last 25 years has been the dramatic expansion in the size of central bank balance sheets relative to nominal GDP.  The surge in central bank balance sheets began with the onset of the global financial crisis (GFC) in 2008 and was amplified over time by the European crisis in 2012 and most recently the response of central banks to the pandemic.  The expansion of central bank balance sheets reflected increased lending through various emergency credit programs and the purchase of large volumes of securities and other financial assets, the latter intended to help provide policy accommodation with short-term interest rates pinned at the zero lower bound.  The bulk of the expansion in central bank balance sheets was financed by increases in reserves held by depository institutions along with other short-term liabilities.  As the global economy recovered in the post pandemic period, most central banks have been in the process of gradually paring the size of their balance sheet.

While a lot of the “action” on central bank balance sheets over recent years has centered on the asset side of the balance sheet and the associated increases in reserves, there have been some interesting developments with a less attention-grabbing component of central bank balance sheets—physical currency.  Physical currency represents a very significant portion of total central bank liabilities in many countries, and as shown in figure 2, currency outstanding as a share of nominal GDP is significant in many countries as well.  In a number of countries—the UK, Australia, Canada, and Japan—the currency to GDP ratio has been fairly stable over recent years at levels ranging from 2% to 4%.  The currency to GDP ratios for the Euro area, Switzerland and the United States are higher than for other countries.  Demand for these physical currencies may have been boosted by a number of factors including strong safe haven demands during the GFC, the European financial crisis, and the pandemic.  In addition, the opportunity cost of holding non-interest bearing currency was quite low for many years; indeed, in Europe and Switzerland, negative interest rates provided a financial incentive for investors to hold physical currency rather than holding other financial assets bearing negative returns.

In contrast to the other countries shown, the currency-GDP ratio for Sweden declined sharply over the last 15 years.  In the case of Sweden, the public has enthusiastically adopted electronic platforms such as Swish for retail payments.  Indeed, according to the Riksbank Payments Report for 2025, many small businesses no longer accept physical currency and the government has now initiated efforts to encourage cash usage (see box below).^[2]  The trend in currency outstanding in Norway over recent years is similar to that for Sweden, and the currency-GDP ratio is even lower in Norway than in Sweden.  Currency is still used in some transactions in Norway, but for practical purposes Norway is often cited as an example of a cashless economy.

In addition to its traditional role in facilitating retail payments, currency is also an important source of net income for central banks.   For the purposes of this note, it is useful to think of the asset portfolio of central banks as comprised of two components—a “currency-financed” component consisting of the share of total assets implicitly financed by physical currency and a second component—the “other-liability-financed” component—representing the share of assets financed by all other central bank liabilities. Of course, currency pays no interest, so the currency-financed component of the balance sheet is assured to earn positive income as long as the central bank’s assets earn a positive return.  One could think of the net income generated by this currency-financed component as the value of the currency franchise for central banks.  The franchise value of currency then depends directly on the public’s demand for physical currency.  The currency to nominal GDP ratios displayed in figure 3 are a useful indicator of currency demand, and the cases of Sweden and Norway suggest that currency demand can decline dramatically as the public adopts alternative payment methods.^[3]

The Currency Franchise and Federal Reserve Net Income

This section takes a closer look at the role of physical currency in generating net income for the Federal Reserve.  Figure 3 provides a very high-level snapshot of the structure of the Federal Reserve’s balance sheet as of July 10, 2025.^[4]  As shown in the left column, almost all of the Federal Reserve’s assets are held in the form of securities—primarily Treasury securities and agency MBS.  On the liability side of the balance sheet, physical currency and balances held by depository institutions are the two largest items.

The Federal Reserve’s net income stems primarily from the excess of its interest earnings on assets relative to the interest expense on its liabilities.  With this balance sheet structure, the Federal Reserve typically generates substantial amounts of net interest income.  Over the past 20 years, the Federal Reserve has generated more than $1 trillion in net interest income.^[5]  The Federal Reserve’s excess earnings above its operating costs are remitted to the U.S. Treasury every week with corresponding credits to the Treasury’s account at the Fed.

A substantial portion of total Federal Reserve net income can be attributed to the currency franchise component of net income.  Figure 4 splits Federal Reserve net income into two pieces attributable to the two balance sheet components discussed above.^[6]  As one would expect, net interest income from the currency-financed component has been steady over time.  Income generated by the other-liability-financed component has been larger at times over recent years with income generated by a much-expanded securities portfolio more than offsetting the increase in the interest cost of reserves and other liabilities that financed the securities holdings.  However, the other-liability-financed component of net income is also more variable and turned negative in late 2022 in light of the rapid tightening of monetary policy that was necessary to address elevated inflation.  Nonetheless, cumulative net interest income over the period from 2005 through 2024 amounted to about $1.2 trillion.  Of this total, net interest income associated with the currency financed component totaled more than $700 billion over the same period.

In addition to providing a stable base of interest income, the currency-financed component of net income provides a modest amount of hedging for total net interest income as well because it tends to be negatively correlated with interest income generated by the other-liability-financed component of the balance sheet.  When short-term rates rise rapidly, net income for the currency-financed component tends to move up gradually as the return on the Federal Reserve’s assets gradually moves up as well.  In contrast, net income associated with the portion of the portfolio financed by other liabilities drops because increases in short-term rates show through immediately and nearly one for one to increased Federal Reserve expenses while the effect of higher short-term rates on the return on assets is much more gradual.

Potential Implications of a Shift Away from Physical Currency for Federal Reserve Net Income

Some back-of-the-envelope calculations based on alternative scenarios can shed light on the possible repercussions of a trend away from physical currency for Federal Reserve net income.  Many projections for physical currency assume that demand for U.S. currency would hold roughly constant as a share of nominal GDP over time.^[7]  As shown in Figure 5a, the baseline scenario adopts the assumption that the currency to GDP ratio holds constant at the level in 2024Q4—about 8.1 percent.  As shown in Figure 5b, using the long-term projections for nominal GDP from the CBO’s Long-Term Budget Outlook, U.S. currency in the baseline scenario would be projected to increase from the level of about $2.5 trillion in 2024Q4 to $7.2 trillion in 2055.^[8]  The first alternative scenario—the slow growth scenario—assumes that the currency to nominal GDP ratio falls gradually to a level of 4 percent in 2055—about equal to the current currency-to-GDP ratio for Canada.  In this scenario, nominal currency increases over time but does not keep pace with the expansion of nominal GDP.  The second alternative scenario—the currency decline scenario—assumes that the currency to nominal GDP ratio falls gradually over time to a level of 0.8 percent in 2055—about equal to the current currency-to-GDP ratio for Sweden.  In this scenario, the level of nominal currency declines substantially over the 2025-2055 window by about $1.7 trillion.

Table 1 provides estimates of net interest income associated with the Federal Reserve’s currency franchise over time.  As a rough approximation, one might assume that the Federal Reserve’s assets would earn a return over time similar to that for the average rate on all federal debt outstanding. Using the projections for this rate from the CBO’s long-range projections, the cumulative nominal value of interest income associated with the Federal Reserve’s currency franchise over the period 2025 to 2055 in the baseline scenario would amount to $4.8 trillion.^[9]  The other rows of the table show the results of similar calculations under the alternative scenarios for the outlook for currency.  In the slow growth scenario, nominal currency still increases over the 30-year projection period from the level of $2.4 trillion in 2024Q4 to $3.5 trillion in 2055.  However, the lower path for nominal currency than in the baseline implies a substantial reduction in Federal Reserve net interest income associated with the currency-financed component of the balance sheet.  Cumulative nominal income associated with the currency franchise totals $3.4 trillion in this scenario, about $1.4 trillion lower than the comparable figure in the baseline scenario.

In the currency decline scenario, the shift away from physical currency is much more pronounced.  The level of currency declines from $2.4 trillion in 2024Q4 to about $800 billion by the end of the projection period.  The much lower path for nominal currency implies a much lower trajectory for income generated by the currency-financed component of the Fed’s balance sheet.  In this alternative, cumulative nominal income associated with the currency franchise drops to $2.3 trillion, about $2.5 trillion lower than the comparable figure for the baseline scenario.

The bottom line from these exercises is that plausible assumptions for a long-run decline in the demand for physical currency could have significant implications for the Federal Reserve’s net income and remittances to the U.S. Treasury.  Federal Reserve remittances under either of the alternative scenarios would be considerably lower than under the baseline.  Lower Federal Reserve remittances would also increase the net interest cost of financing the government in the consolidated balance sheet of the Federal Reserve and the Treasury; in effect, the assumed shift in public preferences reduces the quantity of zero-interest currency outstanding and increases the amount of interest-bearing Treasury debt held by the non-Fed public by the same amount.

Moreover, under the assumptions about the returns on Federal Reserve assets, Federal Reserve remittances would gradually become more variable over time as the relatively stable source of net income from the currency-financed component of the balance sheet diminishes in importance.  Greater variability in Federal Reserve net income could imply more frequent scenarios in which net income turns negative.  In addition, the loss of a sizable stable source of net income would imply that negative net income scenarios could persist for longer than would otherwise be the case.  While periods with negative net income do not affect the Federal Reserve’s ability to implement monetary policy or meet its financial obligations, such episodes may be quite uncomfortable from a public relations perspective and could invite public scrutiny with adverse implications.

Caveats

One important caveat to the analysis in the discussion above is that it is by no means certain that demand for U.S. currency will diminish over time in a way suggested by the alternative scenarios.  It’s quite possible that the unique aspects of the demand for U.S. currency will result in continued steady growth of U.S. currency over time.  In that case, the United States would continue to enjoy the fiscal benefits of large stocks of physical currency as a very low cost source of financing.

Another important caveat in the discussion above is that the analysis does not take account of possible effects of changes in payments technology on Treasury yields.  For example, some observers have suggested that a shift in the public’s demand in favor of a new payments technology such as stablecoins could drive a higher demand for Treasury securities and lower the cost of Treasury debt financing.  This could be an important effect for some types of household portfolio shifts but it seems unlikely to be the case for a shift away from currency in favor of stablecoins.  In that scenario, the shift in public preferences would not generate a net increase in the demand for Treasury securities.  Additional Treasury securities held as reserve assets by stablecoin issuers would be matched by corresponding reductions in Treasury holdings by the Federal Reserve.  Of course, stablecoins may substitute for many other types of assets beyond physical currency. ^[10]  In this case, it’s quite possible that a shift in investor portfolios toward stablecoins could put downward pressure on Treasury yields, particularly at the short-end of the curve.  The net effect on government finances would then depend on the relative magnitudes of various factors; as noted above, a decline in currency outstanding would, all else equal, increase government financing costs while non-currency inflows to stablecoins could drive a decline in Treasury yields that would tend to lower government financing costs.

While the estimates here focus on the fiscal implications for the United States of a shift away from U.S. currency holdings, similar issues would apply in other countries.  The magnitude of the possible effects in various countries is closely related to the current currency to nominal GDP ratio across countries.  As shown in Figure 2, the currency to GDP ratio is even higher in the Euro Area and Switzerland than in the United States, suggesting that overall government finances in these countries could also experience a significant deterioration if new technologies resulted in a substantial shift away from currency usage.

Conclusion

New payments technologies are proliferating and appear likely to become more important over time.  Legislation under consideration in Congress may help to support the growth of the stablecoin industry by establishing clear ground rules for stablecoin issuers.  The development of these new technologies presents many opportunities for improving the efficiency of the payment system, but it also presents a host of new policy challenges.  This note focuses on the perhaps underappreciated implications that new payments technologies including stablecoins could have for central bank and government finances.  The analysis suggests that a significant shift away from physical currency in favor of new forms of payments could result in meaningful reductions in central bank net income and corresponding increases in overall government financing costs.  This conclusion assumes that central banks do not compete with stablecoins by issuing their own central bank digital currencies.  While the alternative scenarios considered here might be viewed as far-fetched relative to current circumstances, the example of a sharp decline in currency usage in Sweden and Norway highlights the potential for new technologies to significantly affect currency usage.  Indeed, in some ways, the scenarios considered above were relatively conservative in assuming that the decline in currency demand played out slowly over a 30-year period.  Based on the experience in many other aspects of modern life, the shift away from legacy technology could occur much more quickly than assumed above.  At a minimum, the discussion above points to potential direct financial risks to central banks and the federal government that appear worthy of consideration in contemplating legislation and other policy steps related to the development of new payments technologies.

[1] ©2025 Andersen Institute for Finance & Economics. All Rights Reserved. James Clouse is a Fellow at the Andersen Institute. This note has benefitted greatly from comments of colleagues including Fabio Natalucci, Robert Wright, Craig Torres, Brian Boeckman, Anya Parikh and Leo Bruk. This material is confidential intellectual property of the Andersen Institute for Finance & Economics. The views expressed in this note are those of the author and do not represent an official position of The Andersen Institute for Finance and Economics or affiliated organizations. By viewing this Andersen Institute Note, you agree that you will not directly or indirectly copy, modify, record, publish, or redistribute this material and the information therein, in whole or in part. No warranty or representation, express or implied, is made by the Andersen Institute or any of its affiliates, nor does Andersen accept any liability with respect to the information and data set forth herein. Distribution hereof does not constitute legal, tax, accounting, investment or other professional advice. The information provided herein is not intended to provide a sufficient basis on which to make an investment decision. Recipients should consult their own advisors, including tax advisors, before making any investment decision.

[2] See Payments Report 2025 | Sveriges Riksbank

[3] The currency to nominal GDP ratio for the United States is likely an imperfect measure of the demand for currency given that many estimates suggest that roughly 50 percent of U.S. currency is held abroad.  In some ways, that unique aspect of demand for U.S. currency arguably might make the U.S. less vulnerable to a dramatic shift away from currency to the extent that foreign demands are strongly related to the anonymity of physical currency.  On the otherhand, foreign holders holding physical currency as a safe haven store of value may find that stablecoins have some attractive features relative to physical currency.

[4] See Table 5 of the H.4.1 Statistical Release for that date, Federal Reserve Balance Sheet: Factors Affecting Reserve Balances – H.4.1 – July 10, 2025.  The figure cited for securities holdings includes unamortized premiums and discounts.

[5] As the Federal Reserve responded to the surge of inflation in 2021 and 2022, its interest expenses increased substantially.   While the net interest income associated with the currency franchise remained positive, net interest income for the assets financed by other liabilities turned negative.  Following its long-standing procedures, the Federal Reserve largely ceased remittances to the U.S. Treasury in the fall of 2022 and has recorded cumulative negative net income as a “deferred asset” that will be paid down when net income turns positive again.

[6] Net interest income attributable to the currency-financed component of the balance sheet is calculated by applying the ratio of currency to total Federal Reserve assets to total gross interest income reported in the Federal Reserve’s annual financial statements.  Net interest income attributable to the other-liability-financed component of the balance sheet is calculated by subtracting this figure from total net interest income reported in the Federal Reserve’s annual financial statements.

[7] See, for example, the projections for U.S. currency in the Annual Report of the System Open Market Account, omo2024-pdf.pdf and the discussion in the CBO document Recent Changes to CBO’s Projections of Remittances From the Federal Reserve.

[8] See The Long-Term Budget Outlook: 2025 to 2055 | Congressional Budget Office

[9] The calculation for interest income of the currency franchise in each year applies the CBO projection for the average interest cost of federal debt in that year to the quantity of currency outstanding as of the end of the previous year.  As noted above, the average cost of federal debt outstanding is used as a rough proxy for the return on Federal Reserve assets over time.

[10] See TBACCharge2Q22025.pdf for an interesting discussion of a range of policy issues associated with stablecoins including the potential for stablecoins to drive a net increase in demand for Treasury securities.