The "Strategic Bitcoin Reserve" is an Absurd Idea

It is understandable why BTC owners would want the U.S. government to bid up the price of their assets. That does not make it good policy.

Governments are in the business of transferring spending power from some people to other people.¹ The hope is that these transfers make society as a whole better off, set a reasonable floor for everyone’s living standards, and are established through a process that is accountable and transparent to the citizenry.

But there is no single right way to do this, so politics often involves clashes of narrow interests at least as much as disagreements over principles.² There is always a danger that a small but organized group will try to push through changes that would benefit them at the expense of everyone else.

This is the most useful perspective for understanding a proposal from Senator Cynthia Lummis of Wyoming³ to “implement a 1-million-unit Bitcoin purchase program over a set period of time to acquire a total stake of approximately 5% of total Bitcoin supply, mirroring the size and scope of gold reserves held by the United States”.⁴ Purchases of that size—by a buyer that is indifferent to value—would almost certainly push up the price, currently around $61,000, and generate a windfall for current holders. Little wonder that this idea is extremely popular among the small number of people who happen to own bitcoin.⁵

According to section 9 of the bill’s text, these purchases would be financed by seizing remittances and surplus funds from the Federal Reserve, but mostly by having the Fed credit the Treasury’s General Account (TGA) with hundreds of billions of U.S. dollars in exchange for revaluing its claim on the Treasury’s gold holdings. This gold is still booked at a statutory price of just $42.22/ounce, compared to a current market price of around $2,400/ounce.

The plan is therefore similar to “mint the coin”, but with gold instead of platinum. I am also reminded of the way Argentina’s central bank used to book profits on its holdings dollar-denominated assets every time the peso depreciated, and then monetize those profits by printing more pesos to cover the budget deficit. Either way, once the Treasury spends the new money credited to the TGA on bitcoin, the dollars will end up on private balance sheets, most likely as interest-bearing claims on the Fed such as bank reserves or reverse repos.⁶ The net effect would be to create new interest-paying liabilities without any additional interest-bearing assets, limiting future remittance payments to the Treasury and commensurately increasing the federal budget deficit.

It is not clear to me how any of this would “strengthen the financial condition of the United States, providing a hedge against economic uncertainty and monetary instability”, as the bill claims, although it does seem like a great way to inflate the dollar value of bitcoin and other cryptos.

Suppose we give Lummis and her supporters the benefit of the doubt. Could there be other, legitimate, justifications for this move? This is what Lummis said in the press release announcing her proposal:

Establishing a strategic Bitcoin reserve would firmly secure the dollar’s position as the world’s reserve currency into the 21st century and ensure we remain the world leader in financial innovation…The United States currently maintains strategic reserves in certain hard assets critical to American national security and independence such as gold and petroleum. Establishing a strategic Bitcoin reserve to bolster the U.S. dollar with a digital hard asset will secure our nation’s standing as the global financial leader for decades to come.

In her speech at the Nashville Bitcoin conference, she also said that “this is our Louisiana Purchase moment,” and implied that the rising value of the government’s bitcoin holdings would eventually allow the Treasury to buy back all of its (dollar-denominated) debt.⁷

None of this makes much sense. But it is worth thinking through why it does not make sense, because strategic stockpiles and reserve currencies are important concepts that are often misunderstood.

Trying to Understand the Bitcoiners’ Argument

Shortly before the proposed legislation was published, Sam Lyman, the director of public policy for Riot Platforms, which mines bitcoin and manufactures electrical equipment, published an opinion column in Forbes outlining why the government should have a “strategic bitcoin reserve”. Lyman has a background in Republican politics and he was cited in a FOXBusiness article that anticipated Lummis’s announcement, which suggests that he may be among those who helped design the proposal. His thesis is that “the United States can use bitcoin’s unique properties as a form of digital gold to deter China, Russia, and other competitors from leveraging their physical gold reserves to undermine US dollar dominance”.

I have no idea what that means, and Lyman’s column does not explain how this would actually work—or why it would matter either way—but he does cite a paper (from 2023) by Matthew Pines of the Bitcoin Policy Institute that has a lot more nuance. This is the key passage:

Allowing Bitcoin to monetize alongside (or outpacing gold) would disproportionately benefit the U.S. (whose citizens and firms hold potentially a majority of all Bitcoin, and whose companies and capital markets would grow in tandem). That is, while China and Russia double-down on analog gold, the U.S. can countermove to digital gold.

As China faces intermittent power crises and a mammoth energy import appetite, North America’s (and our close ally Australia’s) prodigious energy abundance gives our states and locales a natural advantage to compete in the global, zero-sum proof-of-work hash race for the block reward…A gradual shift to a global monetary system centered on a neutral reserve asset that we hold a large share of would allow the U.S. to reverse the structural trade flows which have eroded our manufacturing capacity and made us acutely vulnerable to Chinese supply chains…Any dramatic increase in the value of Bitcoin will result in new power centers concentrated in those states and nations early to adopt.

In other words, if bitcoin becomes substantially more useful for buying goods and services than it is now, then whoever accumulated the most bitcoin first would potentially benefit. A world where everyone uses bitcoin for trade is a world where lower-cost electricity directly translates into cheaper imports.⁸ Lummis—and Pines and Lyman—seem to be proposing that the U.S. government accelerate this process by bidding up the price so much that the rest of the world feels compelled to buy out of fear of missing out. This does not seem particularly likely, but even if it were, it is not clear why a “strategic bitcoin reserve” would be a particularly compelling use of taxpayer resources.⁹

The Point of Reserves

Governments maintain portfolios of real and financial reserve assets for the same reasons that people do: you never know when you might have unexpected expenses or unexpected shortfalls in your income. Accumulating these assets (and buying insurance) comes at the cost of being able to enjoy fewer goods and services, but it also limits your downside risk when things go wrong.¹⁰

Determining which assets to buy, and in what size, should be a function of the liabilities and risks you are trying to hedge. For example, I live in the United States, which means that my expenses are denominated in U.S. dollars. It would not make sense for me to hold my emergency cash in Australian dollars or Canadian dollars, unless I regularly traveled to or lived part-time in those places. Nor would it make sense for my emergency cash portfolio to be so large that it could cover decades of expenses, because that would only be possible due to some unwelcome combination of living far below my means and under-investing in risky assets that should appreciate over time.¹¹

Emergency cash and insurance are helpful, but they are not sufficient for a comprehensive household reserve portfolio. So, like most people, we keep inventories of toiletries and non-perishable food in our house, both for convenience and to hedge against the risk of shortages.¹² We also have a car, even though we do not need it for our daily activities, because the times when we really would want to have a car are also the times when it would be a huge hassle to have to rent one on short notice, or rely on alternative methods of transportation.

This logic applies just as well to governments that are trying to minimize downside risks for their citizens.

A country that is a net importer of commodities might want to accumulate physical stockpiles of oil, gas, and grains, but also should avoid holding too much because excessive inventories can be costly and unhelpful.

Investing in real assets that reduce dependency on imports altogether—such as domestic oil and gas, non-fossil energy sources, and technology to improve agricultural productivtity—would be even better than stockpiles, because those investments would reduce the country’s liabilities, rather than simply hedge part of them.¹³

Similarly, governments worried about shortages of manufactured goods such as face masks, microprocessors, or munitions should not just think about building up their inventories, but about developing the domestic capacity to make whatever is needed. Building and maintaining this capacity is also costly—almost by definition, the world would end up with more than enough to meet global demand under normal circumstances—but it has the advantage of flexibility. A diversified and sophisticated manufacturing base, which includes skilled labor as well as factories and machinery, can adapt to make different types of critical goods faster than any stockpile.

Financial assets can also be helpful for governments that want to hedge their risks. Emergency cash is not as helpful as real assets for managing acute shortages of physical goods, but it can be very helpful for societies that are vulnerable to balance of payments crises because they borrow in foreign currency.¹⁴ There are several reasons why so many non-Americans like holding USD-denominated assets, but an important one is that many non-Americans have USD-denominated liabilities. The dollar is a “reserve currency” precisely because it useful for repaying those liabilities. The Fed has tried to reduce the appeal of holding dollars as reserve assets by promising to lend cash in unlimited amounts when needed to a select group of countries, but most societies are outside of this circle of trust.

Bitcoin Would Be a Poor Reserve Asset

Bitcoin has several unusual properties:

• It creates a (more or less) immutable record of all transactions

• Anyone can make bitcoins with enough electricity, computing power, and a permissive legal regime

• Its production is decreasing asymptotically to zero, which puts a hard ceiling on supply at 21 million BTC

• It is not backed by any government, nor by any claim on any other asset or income stream

• Its price is extraordinarily volatile, but overall has gone up by an incredible amount since inception

While these properties are very appealing to some people, they should not be appealing to anyone trying to construct a sovereign reserve portfolio.

Crucially, the price volatility of bitcoin amplifies risks, rather than hedging them. The price fell by half in the first weeks of the pandemic, and by almost 80% between mid-November 2021 and November 2022. At a time when poorer countries might have wanted an asset to protect them from currency depreciation against a surging U.S. dollar, holding bitcoin would have been a disaster. This volatility is only going to get worse as the price of new supply rises and the total outstanding amount—already almost 20 million BTC—gets closer to the cap.

More generally, there are no liabilities denominated in bitcoin, which means that there is no risk in the world for which holding bitcoin is a better hedge than something else. Even if the proponents of the “strategic bitcoin reserve” got their way, and bitcoin became “monetized”, it would still be less useful as a reserve asset than stockpiles of critical goods, domestic productive capacity, or even foreign currencies. Bitcoin might become more useful than gold, but the price of gold is much less volatile because its supply is more flexible and because it has real uses. Gold is also harder to steal (or misplace).

On top of which, there is no guarantee that bitcoin can even be converted into needed goods in a pinch.¹⁵ Before the most recent Russian invasion of Ukraine, Putin was convinced that the Russian government’s large holdings of gold and foreign currency—as well as its position as a large net exporter of essential commodities—would help insulate his country from any sanctions imposed by the U.S. and Europeans. But despite considerable creativity on the part of the Russians, export controls and secondary sanctions have demonstrated their ability to limit what Russians can import irrespective of their apparent ability to pay.

Transactions in bitcoin are harder to supervise than transaction in dollars or euro, but not impossible. Besides which, there will always be some point at which the flows of bitcoin are translated into the currencies that people use to pay their bills. There is a reason why yuan-denominated trade with Russia is sensitive to U.S. government policy.

But there are even more problems with the idea of a “strategic bitcoin reserve”.

Holding Bitcoin Would Not Fix the Problems with the Global Dollar System

The basic problem with the U.S. dollar’s persistent popularity outside the United States is that it makes the dollar more expensive than it otherwise would be. That has some benefits for many Americans—most notably, cheaper imports and cheaper foreign vacations—but also costs, particularly an underdeveloped manufacturing sector and higher borrowing. The current system also creates costs for many people outside the U.S., with the result that the world as a whole is probably worse off. I have plenty of ideas on what could be done to fix this, but it is worth noting that a greater international role for bitcoin would not help.

Right now, U.S. residents in aggregate spend more than they earn, covering the difference by selling financial assets to the rest of the world. Or, equivalently, people outside the U.S. produce more than they use domestically, selling the surplus to Americans while recycling their proceeds by buying U.S. assets. If foreigners in the aggregate remain determined to spend less than they produce, saving the difference in financial assets, but buy bitcoin instead of USD-denominated assets, then the immediate next question should be: from whom?

From the perspective of balance of payments flows, the currency in which transactions are denominated is not important. What matters is the location of the counterparties. Borrowers outside the U.S. issue USD-denominated debt and attract inflows to their countries all the time. If there were a country of bitcoin, and the supply of BTC-denominated assets were flexible, then a surge in global demand for BTC might correspond to a shift in the global balance of payments such that bitcoinland ran a larger deficit and the U.S. deficit shrank.

But there is no bitcoinland, so in practice what would happen is that some people would buy bitcoin and others would sell it for foreign currency—and then either reinvest that currency in assets or use it to buy goods and services. There would be no straightforward impact on global imbalances, just an extra step in the process by which purchasing power is transmitted across borders.

And if most bitcoin ends up being held by Americans, whether because of the lower cost of electricity or government subsidies, then any international “monetization” of bitcoin would definitely replicate the current system in which foreigners transfer spending power to the U.S. at the expense of foreign consumers and U.S. producers. The Lummis proposal would not fix this because it would be printing dollars to buy BTC mostly from U.S. residents (or foreign criminals, terrorists, and rogue states), rather than buying assets from foreigners to increase their purchasing power.

There are good reasons to worry about the dollar’s use in the international financial system. There are also good reasons for the U.S. government to think more strategically about hedging risks via portfolios of reserve assets. But there is no good reason for a “strategic bitcoin reserve”.

1. From a certain perspective, the government also transfers spending power across time for the same people. Think of public school students who grow up to be taxpayers or workers who retire and get Social Security and Medicare. Despite these conceptual links, there is little direct connection between what is paid in and what is taken out for any given individual. Government-backed student loans would be the closest example, except that few are ever repaid in full.

2. The “vulgar Marxist” view that politics is only about the clash of economic interests is uninformed, because many people are animated by genuine beliefs about what is good and bad that are independent of their narrow material interests. Nevertheless, there are some cases, such as this one, where the simple-minded materialistic perspective is analytically useful. Note that this is not a feature of democratic politics specifically.

3. She identifies as “The Bitcoin Senator”.

4. The government already holds some BTC and other crypto assets thanks to asset seizures from criminals, and it seems that she wants to move those bitcoin into the proposed reserve, which might limit the total amount spent on purchases.

5. Just 7% of U.S. adults used any form of cryptocurrency for any reason in 2023, down from 10% in 2022 and 12% in 2021, according to the Federal Reserve’s latest comprehensive survey of households’ “economic well-being”. Not all of those Americans hold bitcoin, presumably, and most bitcoin is held by relative few entities. The North Korean government also has a lot of crypto, although only some of it is bitcoin.

6. In theory the money could all transform into physical currency, which is not interest-bearing, but that is not the likeliest outcome.

7. That last argument also seems to have had some appeal for Donald Trump. He also wants law enforcement agencies to hoard all crypto assets seized from criminals, instead of selling them to compensate victims.

8. Of course, having lower-cost electricity already generates many benefits for consumers and businesses without any need for bitcoin. Consuming exorbitant amounts electricity to support the bitcoin network sacrifices those tangible benefits for, at best, hypothetical future gains.

9. For the sake of argument I am assuming that there is more here than a desire to extract subsidies from taxpayers.

10. This is also the argument for a tax-financed welfare state, which is essentially a form of forced savings+insurance plus redistribution from higher-income to lower-income households. As with voluntary forms of savings and insurance, whether the welfare state is a good deal for any individual is a function of the price paid relative to the risk-adjusted expected return.

11. There could be circumstances when the sort of person who bets on exchange rates might want to own foreign currencies, but those positions would be separate from reserve holdings. Similarly, if someone had a strong view on the attractiveness of risky assets relative to cash, she might want to temporarily rotate out of stocks etc, making it appear as if she were hoarding more emergency funds than needed. But that would be different from a target cash allocation.

12. It has been years since the pandemic basically ended, but it can still be difficult to find everything I might need in stock in the stores I usually rely on.

13. Employ America has done fantastic work showing how government can encourage the private sector to do this, both for crude oil via the Strategic Petroleum Reserve and for other commodities.

14. For most individual euro area countries, the euro is effectively a foreign currency. In theory you could have a balance of payments problem without any FX-denominated debt if foreign demand for local assets fell off a cliff and/or local demand for foreign assets spiked, making the currency depreciate enough to make imports prohibitively expensive. In that case it would be good to have FX reserves in USD, EUR, etc. The International Monetary Fund’s “assessing reserve adequacy” page is a good resource for those looking for an orthodox treatment of the subject.

15. Leaving aside the fact that the Lummis proposal specifically limits sales of bitcoin from the “reserve”.