Ideal Banking

This article is featured in Bitcoin Magazine’s “The Withdrawal Issue”. Click here to subscribe now.

A PDF pamphlet of this article is available for download.

“The special commodity or medium that we call money has a long and interesting history. And since we are so dependent on our use of it and so much controlled and motivated by the wish to have more of it or not to lose what we have we may become irrational in thinking about it and fail to be able to reason about it like about a technology, such as radio, to be used more or less efficiently.” – John Nash

Money is a technological tool that humans developed organically out of the necessity of bargaining axioms such as time and space. Many of the financial services that exist today have risen to meet the need of an evolving market, and yet at its most reductive, the modern banking system still represents supply and demand via sellers and buyers. This remains true even when looking into the complicated circuit of the U.S. banking system, including the regional banks providing mortgages for first-time buyers, to corporate debt obligations from large private American banks, to the issuance of government bonds by the Treasury. Only by examining the monetary flow in a logical manner within our current system can we begin to present coherent alternatives to the status quo of a select few holding the special privilege as a world reserve currency debt pardoner. At the center of the circuit of the U.S. banking system sits the Federal Reserve and the Treasury — a proprietary black box chip that controls both the current (short-term and overnight interest rates) and voltage (the issuances of U.S. Treasuries, “USTs”).

“The root problem with conventional currency is all the trust that’s required to make it work. The central bank must be trusted not to debase the currency, but the history of fiat currencies is full of breaches of that trust. Banks must be trusted to hold our money and transfer it electronically, but they lend it out in waves of credit bubbles with barely a fraction in reserve. We have to trust them with our privacy, trust them not to let identity thieves drain our accounts.” – Satoshi Nakamoto

Tracing The Circuit

The reserve asset at the bottom of the stack of the U.S. economy is not the U.S. dollar, but rather U.S. Treasuries. Offshore dollar markets such as the eurodollar have long operated under the illusion of dollar creation by these European banks without hardly touching U.S.-issued government debt. The Treasury issues debt in the form of USTs to be sold to private banks, who later create credit via dollars in their customer accounts in order to finance the budget of the U.S. government, as well as service any outstanding national debt. The idea of issuing new debt to service old debt would seem illogical, and in many ways it is, yet becomes far more conceivable with the proper understanding that not all debt is created equal. Debt, at least in the Treasury issuance example above, is demarcated by both the percentage of profit generated as yield, and the duration until said bond reaches maturity. Historically, and perhaps logically, the longer the duration (twenty years vs one year), the higher the yield (2.4% vs 1.2%, using real rates from March 2022). The most liquid denomination of government debt are short-term Treasury bills, referred to as T-bills, which are any bonds with a maturity date less than one year; generally, the yields on those bonds are most directly influenced by short-term federal funding rates. When the government wants to sell more debt, it can increase the yield on these T-bills by increasing the short-term interest rate on offer, driving yield-seeking capital back into the U.S. banking system in search of profit. When rates rise, the cost to borrow increases and these new debt instruments soak up excess dollar liquidity.

Conversely, when rates fall, the cost to borrow decreases, and thus the demand for personal debt increases. To put it simply, if rates are at or near zero, more people will take on debt due to the negligible additional economic cost of eventually paying it back. When rates are higher, and there is market-high yield to be made on simply loaning dollars to the government by purchasing government-issued securities, there is little available supply to be loaned out, and even less demand due to the high costs of borrowing. The issue with this credit-debt boom-bust cycle is that it is levered by trusted third parties, culminating with a buyer and lender of last resort at the modern Federal Reserve — who are in fact actually limited in their ability to manipulate the short end of the yield curve. The yield curve demonstrates the different yields offered by the bond market, denoted by their duration. When there is unexpected and excessive relative volatility within short-term interest rates, the yield curve can invert, meaning short-term debt now pays a higher yield than long-term bonds. If simply held to maturity, sometimes as long as 30 years, Treasury bonds will never yield a material loss, but if short-term liquidity needs strike a bank in the form of depositors withdrawing, banks are forced to sell and realize a loss.

The health and efficiency of the U.S. banking system can be measured in how volatile short-term interest rates are, the state of the yield curve, foreign and domestic interest in government-issued bonds, and the discrepancy between outstanding liabilities and reserves — be it securities or cash.

The New Dollar: FedNow, Not Retail CBDCs

The dollar has been digitized for a long time; be it the Zelle or Venmo credits in your retail account, or the dollar balance in your checking account at Bank of America. But generally speaking, the mechanisms behind the transfer of Treasuries and other reserve assets backing these numbers on a screen have remained at the technical agility of a fax machine. The dollar may be the world reserve currency, and can be transacted via intermediaries on obvious centralized banker rails, or less obviously on Ethereum rails via ERC-20 tokens in the form of popular retail stablecoins, but the U.S. Treasuries held by these novel credit creators remain the world reserve asset. The public has generally feared the direct issuance of some form of retail CBDC (central bank digital currency) due to surveillance concerns and currency seizure from a centralized issuer, but fewer realize both the level of financial surveillance already imposed by banks, never mind the ability for these trusted third parties to censor, blacklist and even expose retail to their counter-party risk. All of these actions are made increasingly possible via the digitization of the currency with an encroaching reliance on centralized payment rails, but up until this July, the communication network for interbank asset trades has remained lossy and slow.

FedNow, slated to launch next month, serves multiple purposes, but perhaps none as important as creating a much more efficient lever for the Fed to have 365/24/7 influence on overnight banking rates, such as SOFR, effectively setting the cost of borrowing short-term liquidity between fractionalized private banks attempting to meet their depositors’ withdrawals. You have probably heard the phrase “reverse repo” once or twice, but the underlying mechanic is often misunderstood. The “repo” stands for a repurchasing agreement; essentially a contract between two entities in which Bank A, with excess dollar liquidity, agrees to lend cash to Bank B, with overnight liquidity needs, via a short-term loan collateralized by Bank B’s assets such as USTs, with the conditions that Bank B will repurchase their securities, usually the next morning (“overnight”), plus a percentage-based fee that Bank A gets to keep. A reverse repo is essentially the same behavior, except that Bank A is bond-rich, cash-poor and thus asking Bank B for dollar-denominated liquidity. This exact scenario came to fruition within the recent regional bank failures in the U.S., and the Fed created new mechanisms to backstop the liquidity needs of the depositors. In the case of the ever-growing reverse repo market, Bank B is routinely the largest American banks, and sometimes even the Fed directly. FedNow is a digital lever, made possible via the internet, for complete centralized control on the overnight rate of borrowing dollars, the necessary transferring of Treasuries between banks, and thus the reshoring of dollar-denominated activity away from the Eurodollar market, and back to the United States within the scope of the Fed and the Treasury.

It’s not all about payments. We will have exchanges forever. We will have banks forever.” – Calle

Banking Is More Than Payments

Did you notice that at no point above were payments even mentioned? Bitcoin in its current state is not necessarily ready to replace the dollar as a global medium of exchange, which takes advantage of financial services to scale over time and space, but it is potentially poised to replace USTs as a world reserve asset and an interbanking settlement network. For Bitcoin to service the many functions of a banking system, there needs to be further tooling beyond the peer-to-peer payment networks innate to the base layer and the Lightning Network, the most discussed second layer. Paper money represents dollars as cash, a physical bearer asset for settling debt obligations, yet the majority of U.S. dollars today exist solely as credit in a user’s account balance at a trusted third party such as a bank. In stark contrast, Bitcoin itself contains zero account balances, and instead relies on a UTXO model: Non-fungible unspent transaction outputs that when signed and spent can transfer fungible satoshis, the atomic unit of bitcoin, between wallet addresses. The address balance of your wallet is an aggregation of the multiple UTXOs associated with your private key. By sharing a UTXO between two or more parties, typically in the form of Lightning channels, Layer 2 payment solutions create near-instant, probabilistically trustless settlements allowing for account balances. By taking a UTXO and creating a shared channel with a peer, you create the functions of credit and debt within the Bitcoin network. Some instances of LN even allow sub-satoshi denominations such as “msats” — a literally unrecognizable unit on the baselayer, and thus only existing as a form of credit or debt. Due to the nature of Layer 2 solutions having the ability to simulate credit and debt, these services enable a trustless iteration of yield via routing fees, and trust-minimized financial services akin to the traditional banking system. Tooling built on top of Bitcoin can create analogs to legacy loan, yield, and liquidity-sharing services. Unfortunately, a large aspect of the trustlessness of Layer 2s being able to finalize and settle back to the mainchain is an open topological network and an ever-surveilled open ledger, significantly reducing the capacity for private financial exchanges.

“Actually there is a very good reason for Bitcoin-backed banks to exist, issuing their own digital cash currency, redeemable for Bitcoins. Bitcoin itself cannot scale to have every single financial transaction in the world be broadcast to everyone and included in the block chain. There needs to be a secondary level of payment systems which is lighter weight and more efficient.” – Hal Finney

Enter ecash

Chaumian mints were invented by cryptographer and mathematician David Chaum in a 1982 paper titled “Blind Signatures For Untraceable Payments”. Chaumian mints utilize blind signatures to represent ecash in mint-specific denominations to create near-perfect privacy within the federation. This newly found privacy is at the expense of reserve asset custody and potential economic debasement depending on both the coding of the mint instance as well as malicious actions from mint authority signatures; this is a situation nearly identical to the downsides of using a legacy financial institution. Ecash uses a similar token mechanic to bitcoin in that while a single wallet can appear to contain an aggregate account balance, in reality the ecash wallet balance is actually distributed among many iterations of common denominations of ecash tokens issued by the mint. The mint itself is completely unaware of the account which funded the initial issuance of ecash, and at redemption merely sees that it had previously validated this token via a blind signature. When using any privacy-preserving payment protocol, there are always two anonymity sets: inside and outside the protocol. While a Chaumian mint can offer near-perfect privacy when transacting within the federated mint itself, an external settlement from the mint can be noticed with a low number of user withdrawals, unassuming metadata collection, and a multitude of poor operational security choices by users. A user could generate ecash from a Chaumian mint instance via a relatively private sender-side LN payment, take the newly generated tokens and fund another outbound sender-side LN payment with zero ability for the mint to generate user account balance information, nor associated metadata with proper external privacy technique. With cheap, near instant, and perfectly private payments, if authored correctly, Chaumian mints can bridge the gaps between Layer 2 balances and even base layer UTXOs.

The New Mint

Chaumian mint construction types differ mainly in two ways: the federation construction itself and the ecash token denominations it issues. A federation can contain a single signature with administrative access to issuing its ecash, as well as having the ability to sign for the mint’s reserve asset when processing withdrawals. A federation can also enable multisignature capabilities to similar mint duties, distributing responsibilities away from a single point of failure to a quorum of trusted third parties. Ecash token denominations are unique to the mint, but theoretically decided at launch of the instance. In lossy parallel to Bitcoin’s UTXO model, there are no account balances, but rather aggregates of ecash tokens that were issued as common denominations (think $5, $10, and $20 notes). These common denominations allow for greater fungibility and far greater anonymity sets within the mint, especially when combined with issuance validation via blind signatures. All of these decisions, including the relative issuance per reserve asset — say ecash token per satoshi — are to be made by the founders of the Chaumian mint, generally upon its genesis. Cashu is a popular, open-source, single-signature instance (created by open-source developer Calle) that is capable of being spun up quickly, leaning on tooling such as LNBits to create fast and easy operability with users already on the Lightning Network. Fedimint, a multisignature instance, allows for a more decentralized mint consensus among federation members, creating more administrative checks within the mint when minting ecash tokens, and when eventually redeemed, signing transactions to withdraw from the bitcoin reserve.

Coincidentally, the main user concerns when using ecash come from its privacy-preserving qualities. Due to there being no account balances, successfully auditing a mint to check its supposed reserves against its liabilities is rather difficult. And since there are no accounts, a trusted custodian must be responsible for holding enough of the reserve asset against the total supply of ecash held by unknown users of the mint. The mint itself is a trusted third party responsible for both appropriate monetary issuance and being able to make depositors whole at time of redemption. This is another prudent parallel to our current banking system, similarly true in both a regional bank and the Federal Reserve itself, of course, with none-to-little of the privacy benefits. These concerns can be theoretically met with clever proof-of-liability schemes such as the one proposed for Cashu by Calle, which publicly generates a monthly token burn list and a monthly token issuance list, rotating issuance keys after every monthly epoch. Both of these lists simply consist of the blind signatures representing their specific ecash denominations from their issuance, and users can check that their own transactions are present in their respective monthly list. The liabilities of the mint is the difference between the mint and the burn list, and thus should be similarly demonstrated within the reserve asset wallet. Proof of reserves is simple with a bitcoin-backed financial service (a public bitcoin wallet), but proof-of-liabilities is significantly more difficult. Concerns of economic debasement and associated custodial risk are nonnegotiable on the base layer of Bitcoin, and yet these real risks are easily mitigated depending on how you use the mint. If a Chaumian mint instance such as Cashu or Fedimint sees user volume at significant scale mostly for extremely short-term payment needs, proper usage of ecash — funding and withdrawing from a busy mint nearly instantaneously — leaves little time for monetary debasement nor reserve asset theft.

“I believe this will be the ultimate fate of Bitcoin, to be the ‘high-powered money’ that serves as a reserve currency for banks that issue their own digital cash. Most Bitcoin transactions will occur between banks, to settle net transfers. Bitcoin transactions by private individuals will be as rare as… well, as Bitcoin based purchases are today.” – Hal Finney

Minting Your Own Bank

Trust is a necessary component of many of the beneficial financial services employed by the U.S. banking system. This remains true now as well as during the gold window. Loans, fractional reserve banking, and counterparty risk is all possible on a bitcoin standard, much like it was on previous hard money standards. By decentralizing the manipulation of monetary issuance away from central pardoners, bitcoin has supplanted USTs as the ideal reserve asset for a new banking system. While it is perhaps seen as a failure to simply replace the instrument banks use to settle their reserves with bitcoin, the elimination of these special privileges from the Fed as reserve asset issuers — and the replacement being a disinflationary, censorship-resistant asset — will have profound effects on the current status quo of monetary manipulation. Bitcoin’s base layer simply cannot service 8 billion people, but proper tooling in layers can allow this scarce, neutral asset unfettered access to a stable monetary policy; a revolution in banking, financial, and economic reality as we know it. Layer 2s are delegated as such due to their trustless ability to settle back to the mainchain without any third party. But ecash enables an entirely new interoperability between Layer 2s and traditional financial services, with an innate ability to be created specifically and timely in accordance to customer demands and needs. Behind every online community that warrants certain privacy needs for their users could be another unique interaction of Cashu. In order to distribute mining rewards privately, mining pool operators can use tools such as FediPools to maximize anonymity sets derived from mining reward payments.

The future of banking is not stablecoin issuers providing opportunities for the Global South to buy U.S. debt; the future is every website, every digital community, threatening to run their own ecash instance, backed by bitcoin — the only neutral reserve asset — when their current financial counterparties are eventually cut off. David Chaum built the tooling and constructed the ideas needed for everyone to be their own bank in the 1980s, and yet those were the days of double-digit interest rates, and the largest onshoring of dollar demand in the modern economic era. Now, as the U.S. banking system is showing serious fundamental cracks — from UST markets marking unrealized duration risk losses, to increasing depositor centralization in the Big Four American banks, to literal government seizure of some of the largest regional banks in the country — it is no surprise that a second wave to the ecash revolution has begun.

This article is featured in Bitcoin Magazine’s “The Withdrawal Issue”. Click here to subscribe now.

A PDF pamphlet of this article is available for download.