CHARTERING THE FINTECH FUTURE.

• Author: Calomiris, Charles W.

What we use as our medium of exchange is subject to dramatic change over time, and sometimes bank regulation has accelerated such changes. The national banking system, founded in 1863, envisioned the creation of a uniform medium of exchange in the form of national hank notes, which replaced the preexisting system of state bank note issuance. But the creation of the national banking system soon resulted in the diminished importance of bank notes as a medium of exchange. Under the new system, state banks faced a prohibitive tax of 10 percent per year on any notes they issued, and national banks had to maintain collateral at the Treasury for their outstanding national bank notes equal to 111 percent of their outstanding notes, and also had to maintain an additional 5 percent in required government-currency ("greenback") cash reserves on hand. That meant that if a bank wanted to make loans, it had to find an alternative to bank notes as a funding source for those loans. Deposits had been growing in importance leading up to the National Banking Act of 1863, but the act accelerated the growth of deposits markedly, and they became the primary funding vehicle for loans. As Comptroller Eckels remarked in 1896: "And thus it has come about that deposit taking is now the feature, and the issuing of circulating notes but the incident, in national banking, instead of, as in the early history of the system, the note-issuing function being the feature and deposit banking but the incident" (Eckels 1896: 565; emphasis added).

Furthermore, bank notes were not issued by all banks prior to the 19th century. Bank notes were a 17th-century innovation, and they were not the primary medium of exchange or the main liability for many important banks in the 18th and early 19th centuries. For many transactions, bankers' acceptances, and bills of exchange were both the primary vehicle of credit and the medium of exchange, and banks like Amsterdam's famous Wisselbank functioned primarily as a clearinghouse for such bills.

Clearly, the history of successful bank chartering informs us that banking has always been defined by the core functions that banks engage in--lending funds or clearing payments, or both. In feet, the word "bank" has been linked to both of those functions, and scholars debate whether payments transfers (initially accomplished on a "bench") or the creation of a portfolio of loans (a "mound," or bank, of loans) has the greater claim to the origins of the word. The particular means banks use to lend or transfer payments changes over time as a function of technological and regulatory changes. In particular, transfers can be made via bills of exchange, bank notes, deposits, credit cards, electronic balance transfers, or exchanges of cryptocurrency tokens via blockchain. History also teaches us that banks don't always provide both lending and payments services. Some banks specialize in one or the other. Indeed, I will show that it requires some rather complicated and specialized economic modeling assumptions to explain why banks sometimes choose to bundle lending and payments services within one intermediary. Those assumptions do not always hold, which explains why bundling is not always a good idea.

Sometimes changes in banks' structures and functions are predictable. The rise of deposit banking in the mid-19th century United States was predictable as a matter of arithmetic if one recognized that banks would continue to act as lenders (given that notes could no longer serve as a funding source for loans after the passage of the National Banking Act). The rise of nationwide universal banking in the United States after 1980 was also predictable, given the evident inefficiencies of the preexisting U.S. banking system (Calomiris 2000; Calomiris and Haber 2014).

Similarly, the demise of traditional models of banking today (including nationwide universal banking provided by today's too-big-to-fail banks) has similar elements of predictability based on clear trends that are driving change. In this article, I consider why current changes are occurring and consider what the new structure of chartered banks likely will be in the future. I don't offer a single forecast of that future, but rather a conditional set of forecasts. If special interests, many of which already are currently struggling hard to preserve the status quo, fail to halt the path of progress, then I believe that technology wall lead us down a path of substantially increased efficiency and stability and the expansion of chartering to encompass novel banks. But the evolution of banking has never been entirely determined by technology or economic logic. Politics is at least equally important in shaping the chartering of banks. If special interests are successful in blocking progress (as our history shows they often have been), then a very different path--one of persistent inefficiency and instability designed to preserve the status quo--is also possible, at least for the foreseeable future.

The article is organized as follows. First, I consider the post-1980 emergence of a nationwide universal banking system and explain how and why technological changes now favor "unbundling" and the ascendance of new fintech banks capable of providing services that threaten that status quo. (1) A detailed analysis of how fintech banks are improving financial inclusion, not just improving efficiency, for existing bank customers is provided. Second, I describe how the chartered banking system could and would evolve over the next decades if special interests fail in their attempt to preserve the status quo. In the near term, this evolution could see substantial numbers of fintech shadow banks becoming chartered national banks, including many that do not rely on deposits as a source of funding. As part of that analysis, I show that there may be substantial advantages from the standpoint of efficiency, convenience, and stability to encouraging the creation of a chartered national bank network of stable-value coin banks issuing nondepository liabilities. Finally, I identify the powerful special interests that are attacking, or may oppose, the chartering of fintech banks.

From Bundling to Unbundling

In the 1980s and 1990s, the United States moved from a system in which banks were fragmented by location, and in which financial services were provided by specialist firms (bank lenders, insurance companies, broker/dealers, and asset managers) to a system dominated by nationwide universal banks. By 2000, a handful of large banks operating throughout the country provided an unprecedentedly wide range of services. Based on the evident historical shortcomings of the U.S.'s fragmented financial system (see Calomiris 2000; Calomiris and Haber 2014), the new banking structure made sense as a means of achieving greater portfolio diversification through geographic integration across bank locations, reusing customer relationship information, and taking advantage of advertising and marketing economies of scale. After two centuries of regulation-induced geographic and service fragmentation, by 2000, it seemed that we finally had arrived at what some of us imagined would be a new nirvana of stable and efficient nationwide universal banking.

But only 20 years (and one major financial crisis) later, the bloom of efficiency and stability is off the rose of nationwide universal banking. We experienced one of the worst financial crises in history in 2007-2009. Since then, the traditional chartered banking system wallows in a state of unprofitability and inefficiency. For the first time in history, new entry into chartered banking has been virtually nonexistent for over a decade. Banks' services remain expensive (and some have become more expensive since 2009), and more than 60 million Americans are still described an "unbanked" or "underbanked."

As has always been the case in banking history, the drivers of these facts are regulation and technological change, which are themselves interdependent. With respect to regulation, the merger wave of 1980 to 2005, which produced the integrated nationwide banking system, occurred as part of a political bargain that drove merging banks to increase their real estate risk exposures, thereby also increasing systemic risk (Calomiris and Haber 2014: chaps. 6-8). The Card Act of 2009 and the Dodd-Frank Act of 2010 did little to remedy those incentives (banks' exposures to real estate risk remain very high), but instead added to the already heavy compliance burdens and other costs banks bear (Calomiris 2017; Calomiris and Campello 2018).

With respect to technological changes, new methods for providing loans and payment services by "shadow banks," especially by fintech banks over the past several years, are accelerating the long-term trend of financial disintermediation from chartered banking by providing more attractive alternatives to customers (Jagtiani and John 2018; Thakor 2020). According to Statista, the chartered banks' share of personal loans granted fell from 40 percent in 2013 to 28 percent in 2018 while fintech banks' personal loans rose from a 5 percent market share in 2013 to 38 percent in 2018. Interestingly, these new competitors are structured very differently from traditional banks. They tend to focus on one or two lines of business, and typically provide either loan services or payments services, but not both. In sharp contrast to the pre-2000 trend toward universal banking, fintecli providers are demonstrating a new model of financial intermediation "unbundling." The new wave of innovative, low-cost, unbundled fintech providers are making behemoth universal banks look as necessary as buggy whips. Such providers are gaining market share in both the payments and lending side dramatically over the past several years, are out-competing traditional banks for talent, and are attracting huge amounts of new investor capital owing to their extremely high...