Month: July 2021

What if anyone could issue a digital rupee?

The venture capitalist Fred Wilson makes an important distinction between dollar stablecoins like Tether/USDT, Circles’s USDC, Synthetix’s sUSD on the one hand, and Central Bank Digital Currencies like the proposed US government-issued digital dollar (or China’s e-yuan) on the other. He says

“But there is another more important reason to want stablecoins to win over CBDCs – competition.”

“When you have competition, you get innovation, new features, composability, and a host of other important benefits. When you have a monopoly, like the US Government or any government, pushing out the alternatives and forcing us to use their digital dollar, you lose all the value of competition. And that would be a terrible thing.”

– Stablecoins vs CBDCs, AVC.com

Wilson’s post highlights something fascinating that is already happening:

For the first time in history there is a distinction between the imposition of a fiat currency like the dollar – which only governments can do – and coins based on that fiat currency.

Imagine the Indian government creating the rupee standard, and many companies issuing competing rupees with different properties but the same value – this is already happening with the dollar! [1]

Why would people want a coin/token that represents a currency but that isn’t issued by a government? In a word, Programmability.

Several months ago, I had imagined what new things become possible when you can program processes, checks and balances, and automation into money itself (“China and Programmable Money“). Here is an example of loans:

Each… lender implements its own loan accounting system that’s tied to it. This is why transferring a loan between two lenders, or between two borrowers is cumbersome. With programmable money, either a lender or a third-party could create credit tokens that represent aspects of a loan: interest rate, lock-in, transferability, a claim on other (similarly digitised and programmable) securities, and so on. Buying and selling loans, combining them, trading them could become a lot simpler.

Since then, it’s become clear that China’s digital currency will in fact be programmable to some degree (“China’s new digital yuan test shows it can be programed to confine utility“) – the government is programming specific public use cases into the digital yuan.

… instead of issuing free digital yuan to lucky citizens that they can use wherever the e-CNY payment is supported, the latest test comes with a theme called “low carbon summer transportation” that gives away the e-CNY as allowances to encourage citizens to take more public transportations. As such, the free e-CNY will arrive with a pre-programed utility for paying for subway and bus tickets through Chengdu’s Tianfutong mobile app or for shared bikes on the Meituan mobile app…. But it appears users won’t be able to convert the values of the e-CNY coupons for taking rides into the general purpose e-CNY stored in their digital yuan wallets.

So then. The government is one entity that can program money. Private and public companies could do likewise. Individual people will be able to program money – I could lend money to a friend under specific conditions that are relevant to only them and me. Ultimately there’ll be hundreds of thousands – if not millions – of such programs written for digital money.

Instead of having a single currency token then, say the digital rupee, whose roadmap alone determines what can be built on it, India could have multiple rupee tokens issued by different public and private entities. They are each valued at one token = one rupee. But they are branded differently and they’re optimised for different sorts of use cases. The only thing these entities cannot do is unconstrained issuance. Asset-backed stablecoins should be audited to prove they have enough reserves, and synthetic rupee providers will need to reference the right source, or oracle, for day to day (minute to minute?) rupee value.

This already exists in a clumsy way with prepaid wallets. They are essentially branded tokens that are asset-backed and have closed use-cases. But they are completely vertically siloed. Everything I do with my Amazon Pay balance has to be inside Amazon, and likewise for Paytm. There’s no single wallet where I can view my ‘Rupee’ balance, my Amazon Pay balance and my Mobikwik balance. There is no exchange where I can convert between them. The use cases are severly limited, mostly tied to e-commerce. The overhead to create and issue one’s own prepaid currency is extremely high, so there are just a handful of such branded rupees.

Natively digital programmable money can blow this ecosystem open, like the use cases above and in my blog post on programmable money, making it more valuable for everyone – not least incumbents like Paytm and even Fastag.

It’s unlikely that the Chinese government will look kindly on private issuance of different types of e-yuans. But as the Indian government and central bank look to pilot their own digital currency, they should not only build programmability deeply into it, they should also allow for similar issuance of dozens, hundreds of branded rupees by anyone who wants to [2].

—
Footnotes:
(1) It looks like stablecoins will be heavily regulated in the USA rather soon)

(2) The Indian government today also heavily restricts foreign ownership of currency. If such rupee tokens are tradable, it becomes a lot more difficult to restrict their ownership. But whether the government should be more liberal about issues like this, or even the free float of the rupee’s value, is another rabbit-hole with strong pro and cons that’s way out of the scope of this site.

The $100 billion question about stablecoins

About three weeks ago, the billionaire Mark Cuban called for the regulation of so-called stablecoins, or dollar-equivalent crypto tokens. This was after he lost a significant amount of money in TITAN, a token that was used to ensure the stability of the stablecoin IRON. TITAN went from $60 to zero in under a day. The story behind that is interesting in itself, and you should read this excellent Twitter thread from Frances Coppola for more:

In this post, though, we’re going to focus on what Cuban said about stablecoins. Part of his statement to Bloomberg was

there should be regulation to define what a stable coin is and what collateralization is acceptable. Should we require $1 in us currency for every dollar or define acceptable collateralization options, like us treasuries or?

To be able to call itself a stable coin? Where collateralization is not 1 to 1, should the math of the risks have to be clearly defined for all users and approved before release? Probably given stable coins most likely need to get to hundreds of millions or more in value in order to be useful, they should have to register.”

To this, a reader and friend asked me the following:

This makes me ask the more fundamental question. Is there a way to make a reliable stablecoin but not offered by a state?

My view of this is twofold.

One, any private attempt to create an asset-backed or currency-backed stablecoin is futile. 

Two, we don’t need asset backing to create a stablecoin. 

Let’s talk about One: why private stablecoins are near-impossible to pull off:

Asset backed coins are always reactive to demand and supply. A dollar backed stablecoin must bring actual dollars or dollar equivalents onto their books when demand is high, and must take them off the books when demand it low. This is fundamental to maintaining the price at USD 1, but is always post-facto. This is the same whether the underlying is fiat, crypto, commodities or others. Overcapitalisation is one answer, but only in a finite band. If a stablecoin becomes popular, the surplus can quickly be eaten up. In the case of crypto backed stablecoins, a wide rise or fall in the underlying token’s price could do the same.

Fiat has no such problems because they are units of account enforced by governments. A dollar is One Dollar because the US government says so, and the US government ensures it is the only one who can say so. Dollars are not backed by anything, and nor are other fiat currencies.

Of course, there could be excess demand for money, for dollars, and the federal reserve could print more, but (and this is some simplification here) if it prints more than the economy ‘needs’, it leads to a glut, which we call inflation, which is simply that goods begin to cost more because a dollar is ‘worth’ less. But that last fact is always expressed in terms of the price of other things. A stablecoin is one such thing. 

This is why the only practical fiat stablecoin is one that is issued by a government.

But that is only half the picture. 

So then let’s talk about Two: whether asset backing – state-mandated fiat or otherwise – is even necessary for a stablecoin:

USDT is the most popular stablecoin. And that popularity also makes it the most interesting one – because Tether doesn’t really try hard to hide the fact that its asset backing isn’t really solid. The recent case with the state of New York required it to disclose that the vast majority of the backing was commercial paper and secured loans, not actual cash, which was under four percent. And yet there was no dip in volume, no volatility in price. It remained rock steady at USD 1.

This is because Tether serves one critical purpose: people need a crypto token equivalent of One Dollar.

Tether being early to the race set off a virtuous cycle: the more people owned it, the more other people were likely to own it, secure in the belief there were buyers for what they held. Put another way, that they could purchase other crypto tokens – ie other goods – for their USDT, and it would be measured in something familiar, dollars.

Tether had created Fiat. 

As the fintech commentator Patrick McKenzie wrote, although far less admiringly,

At this stage, at this volume, Tether’s price is anchored in the widespread belief that it represents One Dollar. 

That’s it. Not in the authenticity of its auditors’ report, not in the reputation of its founders.

PS Tether isn’t the only dollar stablecoin. Other major ones are Coinbase and Circle’s USDC, Binance’s BUSD and MakerDAO’s DAI, which together have a market cap of well over $100 billion as I write this. But Tether does represent over 60% of that.

So to conclude. 

Yes, an asset-backed stablecoin is only practical when it is issued by the same entity that issues the asset itself, ie a government. 

But being backed by an asset isn’t a prerequisite for a stablecoin when it becomes so popular it behaves like its own unit of account.

Apparently it is, in practice, possible to create a private dollar based on faith. That said, I find it hard to imagine how this party can last much longer for Tether. We shall see.