By Jim Harper

It has been suggested in some quarters that the government —
instead of profit-motivated private enterprise — should provide financial
services. In past friendly debate with a colleague who supports the idea, I
concluded mockingly, “Yes, we’ve got to get the money out of banking!”

Last week, the Federal Reserve Bank of Boston (Boston Fed) and the MIT Digital Currency Initiative (MIT DCI) released their initial research on the technical infrastructure that might underlie a US central bank digital currency (CBDC). They call their project “Hamilton.” Reading it in contrast with the Federal Reserve’s (the Fed’s) outing on the topic from late January, one might conclude that Project Hamilton wants to get central bankers back into banking. Their technical choices contrast favorably to what we know of the Fed’s plans so far.

In a post discussing the Fed’s CBDC paper two weeks ago, I sought to highlight a tension. The Fed paper was agnostic on whether the Fed should produce a US CBDC, but it appeared committed to an account-based system, should one emerge.

One way of assessing payment systems is to look at them as
account-based or token-based. An account-based system requires collecting
information about people, their holdings, and their value-transfer activities,
which is costly in terms of privacy. A token-based system is mostly about
controlling counterfeiting. It requires little or no personal information to
function.

The Boston Fed and MIT DCI paper is lengthy and complex, and it says a simple “account-based” versus “token-based” lens may be insufficient given the “complexity of choices in access, intermediation, institutional roles, and data retention in CBDC design.” But the paper makes a choice between whether Hamilton will track balances (i.e., accounts) or funds (i.e., tokens).

Open cryptocurrency networks are token-based. They work by serially
logging transactions and thus the tokens associated with them. An “output” in
one transaction is an “input” into the next. When a token is at rest with a
given holder, it is referred to in the data system as an “unspent transaction
output” or “UTXO.” That is what Project Hamilton has opted for. “We chose to
build Hamilton in the UTXO model.”

That is not the only choice that leans in favor of privacy.
To lighten the load on a system built to process 100,000 or more transactions
per second, Hamilton would store cryptographic hashes of transactions rather
than the data-heavier transactions themselves. That means information such as
public keys, which are readily used as identifiers, would not be in a single
database the way they are — and are widely replicated — in a public blockchain.

How does all this restore bankers to the role of banking? In the ill-fated “BitLicense” proceeding, New York state’s superintendent of financial services sought what I called “intolerable financial surveillance, forcing firms to collect detailed information on every last transaction in which their customers engage.” Here was a consumer protection agency not helping consumers protect their privacy but sacrificing it wholesale to presumed law enforcement needs. There are other agencies of government whose role is to seek after those things, and in doing their work, agencies should strike balances against each other rather than unite against the people.

The Fed’s commitment to an account-based system has hints of similar muddiness as to role. It commits to a certain data model and the financial surveillance status quo without an evident purpose in payments or banking. The Fed paper says:

A
general-purpose CBDC would generate data about users’ financial transactions in
the same ways that commercial bank and nonbank money generates such data today.
In the intermediated CBDC model that the Federal Reserve would consider,
intermediaries would address privacy concerns by leveraging existing tools.

The Boston Fed and MIT DCI paper shows that the data
generated by a general-purpose CBDC is highly configurable. A CBDC doesn’t have
to be designed in such a way as to lock in the status quo. In its proper role,
the Fed should be somewhat antagonistic to financial surveillance, given its
costs to banking and payment systems. Let law enforcement interests make the
case that extending the current regime would be wise.

As I wrote late last year, a rule of thumb that I think makes sense is to have a CBDC system that can “support all information policies, from entirely private transactions equivalent to cash, to closely monitored transaction flows, to public broadcast of, say, government spending.” I have not been able to digest all of the Boston Fed/MIT DCI paper, much less its significance, but it is hewing much closer to that principle than the Fed.

The post Project Hamilton compares well to the Fed’s central bank digital currency approach appeared first on American Enterprise Institute – AEI.