The Biden Administration’s Strange Stance on Crypto and the Climate
How did a debate over crypto tax reporting requirements veer so wildly off-track?
The debate on Capitol Hill over a cryptocurrency tax reporting provision in the bipartisan infrastructure bill got really weird Thursday night when Senator Mark Warner (D-Virginia) proposed an amendment that would favor one energy-inefficient mechanism of blockchain validation over a more environmentally friendly alternative … and then the White House backed Warner’s amendment (with Treasury Secretary Janet Yellen personally lobbying senators to support it).
What on earth is going on?
I don’t have a theory as to why Senator Warner—or Secretary Yellen—would do this. But the Warner amendment would impose potentially significant environmental costs for approximately zero tax-enforcement gains.
Getting up to speed
First, a quick recap of events: On Sunday night, Senators Kyrsten Sinema (D-Arizona) and Rob Portman (R-Ohio) introduced a 2,702-page amendment to the infrastructure bill moving through Congress that included, starting on page 2433, a provision on information reporting requirements for brokers of digital assets (cryptocurrencies).
The Portman-Sinema amendment, as it has come to be called, would have clarified the definition of “broker” in section 6045 of the Internal Revenue Code to include “any person who (for consideration) is responsible for regularly providing any service effectuating transfers of digital assets on behalf of another person.” The upshot of being a broker under section 6045 is that Treasury and the IRS can then require you to track your clients’ adjusted basis in their assets and report that information, along with your clients’ gross proceeds, when they sell or exchange those assets.
The Portman-Sinema language could have been interpreted in a number of ways—for example, as applying only to centralized cryptocurrency exchanges like Coinbase and Kraken, or as applying to a very wide range of cryptocurrency market actors. Portman’s office sought to clarify the scope of the amendment Sunday night. “This legislative language does not … force nonbrokers, such as software developers and crypto miners, to comply with I.R.S. reporting obligations,” a Portman spokesperson said in statements to the New York Times, Washington Post, and other outlets. “It simply clarifies that any person or entity acting as a broker by facilitating trades for clients and receiving cash must comply with a standard information-reporting obligation.”
As I argued in my last post, applying basis-tracking and information-reporting requirements to centralized exchanges is eminently reasonable. These exchanges already are subject to know-your-customer requirements, and they should be able to file information returns (on a Form 1099-B) with relative ease. Indeed, Treasury almost certainly had authority already to impose information reporting requirements on centralized exchanges under the preexisting section 6045, which defines a “broker” as “any … person who (for a consideration) regularly acts as a middleman with respect to property or services” and which allows Treasury to require basis-tracking for any “financial asset.”
The problem with the Portman-Sinema amendment, as narrowly construed by Senator Portman, is that it still would leave lots of room for tax evasion. For example, an individual could buy digital currency (“coins”) on Coinbase, using U.S. dollars from a bank account or credit card. She then could transfer those coins to a wallet, which is a device or program that stores the “public key” needed for receiving coins and the “private key” needed for sending coins. A wallet can be a piece of software downloaded onto a personal computer or smartphone, or it can be an actual piece of hardware that stores private-key information offline.
The individual could then transfer coins from her wallet to someone else’s wallet without returning to the centralized exchange. She also could swap her coins for a different type of coin on a decentralized exchange, which is a platform that facilitates peer-to-peer transactions. A robust information-reporting regime for cryptocurrency will likely require some way to deal with wallets and decentralized exchanges. But simply saying that they have to file Form 1099-Bs won’t be enough: there are millions of wallets already out there on U.S. smartphones that don’t have know-your-customer capabilities, as well as dozens of decentralized exchanges all around the world operating on open-source protocols (created—in some cases—by anonymous developers). The cat is out of the bag already, and it will take a lot of work to get the cat back in.
This is why I said of the Portman-Sinema proposal: “If the provision that ultimately passes is as narrow as Portman’s office suggests, then it will probably have a modest but positive effect on cryptocurrency tax compliance …, but the hard problem of cryptocurrency tax compliance won’t have been solved.”
Proof of work vs. proof of stake
Separate from the problem of how to actually regulate wallets and decentralized exchanges, the Portman-Sinema text raised questions about the status of validators. Validators are the computers (and the people running those computers) that maintain the blockchain. One might plausibly describe validation as a “service effectuating transfers of digital assets,” but validators typically don’t know the identities of the individuals whose transactions they are validating—much less the other information (address, Social Security number, adjusted basis) that would be needed to generate a Form 1099-B.
The primary methods of validation are proof of work and proof of stake. Proof of work—which is the mechanism used by Bitcoin—involves computers all around the world racing to solve extremely hard math problems and then “proving” their answer to other computers in the network. One problem with proof-of-work validation is that it is extraordinarily energy-intensive: according to the Cambridge Bitcoin Electricity Consumption Index, Bitcoin currently uses more terawatt-hours per year than all of Bangladesh (population 163 million). There is a vibrant debate about just how bad for the environment Bitcoin really is, given that some of the energy comes from carbon-neutral sources or from byproducts of oil extraction that otherwise wouldn’t make it to the grid. But the energy intensity of Bitcoin is clearly a real problem—and it’s the reason why, for example, Elon Musk has said that Tesla no longer will accept Bitcoin as payment for its cars.
Proof of stake—used by some smaller cryptocurrencies—involves significantly less energy than proof of work. There are different versions of proof of stake, but to oversimplify (and probably to reveal my own very surface-level understanding): In a proof-of-stake system, validation responsibilities are randomly assigned to one validator, and then another randomly chosen group of “attestors” double-checks and signs off on the first validator’s solution. The term “stake” comes from the fact that in order to be chosen as a validator, one must ante up (“stake”) a certain number of coins. If the validator proposes a block to which others attest, then that validator receives a reward. If the validator fails to solve the problem in the requisite amount of time or proposes a block that others reject, then part or all or her “stake” will be lost.
There are questions about whether proof-of-stake mechanisms will turn out to be secure from malicious attacks on a large scale and over the long run, and I don’t know enough to even have an opinion on the subject. But from an energy-use perspective, the benefits of proof of stake are overwhelming. Ethereum, the second largest cryptocurrency by market capitalization, currently operates on proof of work but plans to migrate to proof of stake as early as December of this year. If cryptocurrency is here to stay, we should all really hope—from a climate perspective—that this migration succeeds.
Wyden-Lummis-Toomey
On Wednesday, Senators Ron Wyden (D-Oregon), Cynthia Lummis (R-Wyoming) and Patrick Toomey (R-Pennsylvania) introduced an amendment that, if enacted, would have more or less aligned the Portman-Sinema text with Senator Portman’s description of that text. Specifically, it would have added:
Nothing in this section … shall be construed to create any inference that a [“broker”] includes any person solely engaged in the business of—
(A) validating distributed ledger transactions,
(B) selling hardware or software for which the sole function is to permit a person to control private keys which are used for accessing digital assets on a distributed ledger, or
(C) developing digital assets or their corresponding protocols for use by other persons, provided that such persons are not customers of the person developing such assets or protocols.
The first carveout—for “validating distributed ledger transactions”—would seem to exempt proof-of-work and proof-of-stake validators from filing Form 1099-Bs that they couldn’t possibly have filed anyway.
The second carveout (“selling hardware or software …”) is a bit less clear. It’s designed for wallets, but it wouldn’t seem to apply to most wallet providers in the United States—for two reasons. First, most firms with a U.S. presence that provide popular wallets also do something else. For example, Ledger—which makes a popular hardware wallet—also makes an app that allows users of the hardware wallet to buy, sell, and exchange coins. Second, most popular software wallets in the United States do more than just store private keys. They also make it possible to send coins to other wallets.
The third carveout (“developing digital assets and their corresponding protocols …”) would exempt coin creators, like Bitcoin creator Satoshi Nakamoto. No one knows who or where Satoshi Nakamoto is, so it wouldn’t mean much to require him/her/them to file Form 1099-Bs. Less clear is whether the exemption would apply to the sometimes-anonymous developers of decentralized exchanges, which are sometimes described as “liquidity protocols.” In any event, applying information reporting requirements to decentralized exchanges will require much, much more than changing the definition of “broker” in section 6045. Developers of decentralized exchanges can avoid information reporting requirements by leaving the United States. Treasury and the IRS then would need to order U.S. internet service providers to block access to websites like uniswap.org and pancakeswap.finance. Maybe Treasury and the IRS ought to have this authority, but the original Portman-Sinema text didn’t provide it.
Warner-Portman-Sinema
On Thursday night, Senators Warner, Portman, and Sinema filed a competing cryptocurrency amendment. Theirs reads, in relevant part, as follows:
Nothing in this section … shall be construed to create any inference that a [“broker”] includes any person solely engaged in the business of--
(A) validating distributed ledger transactions through proof of work (mining), or
(B) selling hardware or software the sole function of which is to permit persons to control a private key (used for accessing digital assets on a distributed ledger).
The Warner-Portman-Sinema amendment differs from Wyden-Lummis-Toomey in two important ways. First, and most destructively, Warner-Portman-Sinema exempts only proof-of-work validators—not proof-of-stake validators—from information reporting. This makes zero sense. Why would the United States favor the more energy-intensive validation mechanism? And why would a climate-conscious administration lobby senators to support the environmentally-unfriendly amendment?
Neither Senator Warner nor the co-sponsors nor the administration have provided any explanation. Now, one might say: “Don’t worry too much because imposing information reporting requirements on proof-of-stake validators would be so crazy that Treasury wouldn’t ever do it.” But if Warner-Portman-Sinema is the law on the books, it’s likely to dampen enthusiasm for proof-of-stake coins in the United States (since U.S. users will know that Treasury has a statutory loaded gun that would effectively prevent U.S. users from “staking”). In particular, this could create problems for Ethereum’s migration to proof of stake, since after Ethereum makes the switch, it won’t be easy to switch back. The Warner-Portman-Sinema amendment throws a wrench into Ethereum’s plans for … what reason? The most charitable interpretation, I think, is that it was a drafting error by someone who didn’t understand that “proof of work” is not the only mechanism for validation.
The second difference between Wyden-Lummis-Toomey and Warner-Portman-Sinema relates to “developing digital assets and their corresponding protocols.” The Wyden-Lummis-Toomey amendment would exempt these folks; the Warner-Portman-Sinema amendment wouldn’t.
Supporters of the Warner-Portman-Sinema amendment are right that a robust information reporting regime will need to include decentralized exchanges … somehow. But again, none of the proposals on the table give Treasury the broad authority it will need to shut U.S. users off from those exchanges. And if we’re going to wall off a portion of the Internet to U.S. users—which may turn out to be necessary—we probably ought to have a debate lasting more than a few hours about whether the tax-compliance benefits outweigh the Internet-freedom costs.
Speaking of proof: All this is further proof that the Senate should not have rushed to write complicated cryptocurrency tax compliance rules as part of a must-pass infrastructure package without a committee markup. There are a ton of other ways to improve tax compliance that don’t have climatic consequences (e.g., fund the IRS!). We certainly need more robust information reporting requirements for cryptocurrency. But Warner-Portman-Sinema doesn’t solve the tax-compliance problem—and it exacerbates a planetary problem along the way.