The eternal dream of automatic money

The writer is an FT contributing editor

The white paper that launched the cryptocurrencies terra and luna, like all such papers, is composed in the elegant font of a peer-reviewed journal. It cites The Review of Financial Studies and the Journal of Financial Economics. It is a very serious work.

“Terra Money: Stability and Adoption”, released in 2019, lays out two axioms of money. First, a stable currency is elastic — it can expand and contract to counter swings in demand. Second, stability alone doesn’t guarantee that anyone will use a currency. For that, a currency needs a fiscal policy that will spend it on efficient projects that create growth.

Well, yeah. Keeping a currency stable by managing supply is the central challenge of monetary policy. Efficient spending that creates growth is the central challenge of fiscal policy. The two axioms in the terra white paper are inarguably true descriptions of ancient problems.

The paper offers a neat answer: two currencies, paired by design. The value of terra would be pegged to the dollar. What would hold that peg was a guaranteed swap with a dollar’s worth of luna. And luna would have value, because each luna would offer a vote on how to spend some of the seigniorage — profit — from minting terra.

The project failed. Terra dropped to 8 per cent of its value in early May, and luna is worth essentially nothing. There was nothing inherently hare-brained about luna and terra, though. Together, they shared a governance structure that would decide which financial applications could use terra, and which to encourage with seigniorage revenue.

The criteria would be “robust economic activity” and “efficient use of funding”. This should sound familiar, because it’s what legislators do in a democracy. One of these approved applications was Anchor, with its own nominally independent governance structure that offered an interest rate for terra deposits. This is just the basic work of any central bank.

A sharp reader may have noticed the repetition of the word “governance”. It is often difficult to describe how cryptocurrencies work, because the people who create them have a habit of using new words to describe things that already exist.

Words such as “protocol” and “stakeholder” and “governance” are just different ways to say words like “institution” and “investor” and “decisions”. The terra white paper describes a fiscal and monetary system with a remarkable resemblance to the one we already have. The only difference is that in the terra protocol, governance was supposed to ensure that everyone would make good decisions.

We know now that it didn’t work. Governance is tricky. It fails all the time. In this case, the people with a stake in Anchor decided together to offer a 20 per cent interest on deposits of terra, using their share of seigniorage revenue from the currency to fund the operation. This is an astonishing return in any environment, and it encouraged the creation of more terra than could ever be useful for any actual economic activity. Such unsustainable growth is a classical reason for a currency to collapse. In this case, it undermined confidence in terra.

As a species we’ve spent the past 4,000 years trying out a lot of different governance protocols for money, and none of them has been particularly effective at preventing people from making the wrong choice.

The protocol that came out of that original white paper referred to terra as a “algorithmic stablecoin” because of the way it and luna balanced each other under a set programme. It suffered under an old hope: the dream of automatic money that cannot be corrupted by bad human decisions.

The philosopher David Hume, for example, proposed that prices should adjust to the supply of precious metals, which would flow naturally across borders to where they were needed. This was the logic that underpinned the classic gold standard, but as the economic historian Barry Eichengreen has pointed out, governments suspended the standard to accommodate wars and recessions. The gold standard was a protocol with plenty of human agency for decisions.

Milton Friedman, too, wanted human fallibility as far away from money creation as possible. In the 1960s he proposed that the total supply of money grow at a constant rate. Three to 5 per cent was his preference, but consistency and inflexibility were more important to him than the rate. He acknowledged that this would create mild inflation and deflation, but thought it best to take humans out of money entirely. More recently, in the early 1990s, John Taylor suggested that rules for policymakers could prevent them from exercising discretion, with better outcomes.

Without contesting the merits of these arguments, they just don’t seem to be what people want out of money. Liberal democracies have downvoted Friedman and Taylor-like protocols again and again. There are no chains with which we can bind ourselves. We inevitably snap them, and ask for the people in charge of money to make decisions again. Better to recognise that, than to hand-wave it all away with words like “governance”.

This article has been amended to change the photo illustration.