One of the largest issues associated with the issuance of Federal Reserve digital currency — a new form of digital currency that can be owned by the public — is its impact on the financial system. Fed digital currency may offer the Fed an attractive means for soaking up excess bank reserves at lower cost than the current approach of paying interest on reserve balances and reverse repurchase transitions. But if Fed digital currency becomes too popular, it could negatively impact the cost and availability of business and household credit.

 

The Fed has yet to decide whether it will issue a digital currency that can be held by the public. Yet today most money is already in digital form. The largest cache of digital currency is the $18.1 trillion in deposits at federally insured depository institutions. These deposits are electronic bank money. If an insured institution fails, a depositor is certain to recover balances up to the $250K federal deposit-insurance limit, but excess balances are at risk of loss.

 

Shares in money market mutual funds (MMFs) are digital and a close substitute to bank deposits. MMF shares must first be converted into bank deposits before they can be used to pay for goods and services, but under normal market conditions, MMF shares convert one-to-one in value to bank deposits. Today, $1.5 trillion is invested in retail MMFs while institution-only MMF shares total $3.2 trillion.

And then there is central bank money, much of which is digital in form. Central bank deposits are a liability of the central bank (the Fed), and digital deposit balances are never at risk of loss because central banks can always meet withdrawal demands by converting deposit balances into physical currency.

 

The Fed currently issues two forms of central bank money: physical Federal Reserve notes — the dollar bills in your wallet — and digital dollars held as reserve balances at Federal Reserve banks. At present, Fed digital dollars can only be held by banks and other specialized financial institutions.

 

In times past, banks economized the amount of reserves held at the Fed because, like Federal Reserve notes, reserves paid no interest. Minimum reserve requirements forced all banks to hold some reserves, and the overall system included excess reserve balances that were needed to settle daily interbank payment system transactions. Today, reserve requirements have been abolished and the Fed pays interest on all bank reserve balances.

 

The few non-bank institutions allowed to hold reserve balances at the Fed do not earn interest. Institutions like Fannie Mae, Freddie Mac, and the Federal Home Loan banks need Federal Reserve accounts to settle transactions in the interbank payments system. These institutions try to earn money on the reserve balances they hold by lending them out in the federal funds market.

 

In early 2008, before the Fed began paying interest on bank reserves, there were $817 billion in Federal Reserve notes outstanding and banks held only $9 billion in reserve balances at Federal Reserve banks. Today, banks hold about $4 trillion in interest-earning reserve balances and Federal Reserve notes outstanding have grown to $2.2 trillion.

 

Unlike bank deposits, public Federal Reserve digital currency would be a direct liability of the Fed and free of default risk, meaning that a digital currency owner would always be able to redeem a digital Federal Reserve dollar one-for-one for a Federal Reserve note.

 

The Fed creates new reserve dollars, Federal Reserve notes, or public digital currency (should the Fed decide to issue it) by purchasing a limited set of eligible securities. As a result of the its purchases to date, the Fed owns $5.7 trillion of marketable US Treasury securities, $2.3 billion of debt securities issued by federal agencies, and $2.7 trillion of mortgage-backed securities issued by Fannie Mae, Freddie Mac, and Ginnie Mae.

 

To reduce reserve balances, the Fed enters into reverse repurchase agreements with eligible counterparties. In a reverse repurchase agreement, the Fed pays a counterparty to trade its Fed reserve account balances for an equivalent amount of securities from the Fed's portfolio holdings. This transaction locks up reserve balances that otherwise might be loaned out in the federal funds market which would cause the funds rate to dip below the Fed's policy rate target. The Fed currently drains about $2 trillion in reserve balances using reverse repurchase agreements, with most of the participation coming from MMFs.

 

If the Fed makes central bank digital currency available to the public, the public must pay for it by converting bank deposits, Federal Reserve notes, or MMF shares into digital currency. The volume of bank deposits and MMF shares converted into Fed digital currency will have important ramifications for the availability of credit.

 

On average, banks raise 80 percent of their funding from deposits. Deposits are the cheapest source of bank funds. If there is a large demand for Fed digital currency paid for using bank deposits, banks could experience a reduction in funding and a substantial increase in their cost of funds. Banks would likely react by reducing lending and passing the additional costs on to customers in the form of higher loan interest rates and increased deposit fees. Bank credit for businesses and households could become harder to acquire and more expensive.

 

MMFs are a key purchaser of commercial paper, an important $1 trillion market where high quality corporations and businesses borrow short term. With interest rates at historical lows, slack loan demand, and the cost of deposit insurance premiums on the influx of COVID relief deposits, banks actively encouraged large corporate depositors to withdraw their funds. Many of these funds flowed into MMFs.

 

In view of the prior discussion, it is clear that, under normal market conditions, the public's demand for Fed digital currency will depend importantly on four factors: (i) whether or not the digital currency pays interest; (ii) the level of short-term interest rates; (iii) the magnitude of uninsured deposits and MMF balances; and, (iv) the cost of Fed digital currency transactions. In contrast, under financial crisis conditions, large volumes of bank and MMF deposits would likely run to the safety of central bank digital currency — an issue I discuss elsewhere. In the remainder of this post, I consider the demand for Fed digital currency under normal market conditions.

 

If Fed digital currency pays no interest, and the costs of using the new currency are similar to the costs of using bank deposits, the demand for the currency will likely come from businesses with large, uninsured bank deposit balances and institutional MMF funds. This scenario becomes more likely if interest rates stay depressed and the Securities and Exchange Commission adopts its "swing pricing" requirement for institutional MMFs. Swing pricing requires MMFs to discount the value of redeemed shares when the fund experiences net outflows. With near-zero MMF yields and swing-pricing risk, MMF institutional investors would almost certainly find Fed digital currency an attractive alternative. If short-term interest rates rise appreciably, MMF returns could become sufficiently attractive to retain institutional investors. Regardless, large MMF outflows would raise the cost and reduce liquidity in the commercial paper market.

 

If Fed digital currency were to pay interest, depending on the rate paid, it could pull significant balances from banks and MMFs. History demonstrates that Fed reserve balances were an unattractive alternative when short-term market interest rates were positive and reserve balances paid no interest.

 

The appeal of Fed digital currency could be enhanced if it is linked to a new payment system architecture that reduces transactions costs relative to bank deposit transactions. However, with present technology, it is unclear whether a blockchain-like payment system would yield cost savings. Moreover, should a cost-reducing technology be realized, it could also be applied to reduce transactions costs in the traditional bank deposit payment system. On balance, it seems unlikely that transactions costs will be a durable competitive advantage for a Fed digital dollar.

 

At present, there are a substantial amount of excess reserves in the financial system as a result of Federal Reserve and federal government COVID relief programs. The Federal Reserve might find Fed digital currency an attractive way to soak up reserves at lower interest costs than the Fed pays on reserve balances and reverse repurchase transitions as long as the new digital currency does not negatively impact the cost and availability of credit. But if the design of Fed digital currency proves to be more attractive than anticipated, the new currency could reduce availability and increase the cost of business and household credit.