Related Links

• Why Stablecoins Fail: An Economist’s Post-Mortem on Terra

Transcript

Tim Sablik: Hello, I'm Tim Sablik, a senior economics writer at the Richmond Fed. My guest today is Russell Wong. Russell is a senior economist at the Richmond Fed who previously worked as a senior analyst for the Bank of Canada. His research focuses on payment systems including digital currency, which is the focus of our conversation today.

Russell, welcome to the show.

Russell Wong: Thanks for having me.

Sablik: Today, we're going to be talking about digital currencies which I think listeners are probably familiar with, a few of those probably most notably bitcoin or maybe dogecoin. Some people may have also heard of ether, another commonly used digital currency. I think one thing that I know about those digital currencies — and probably others know as well — is that they can be quite volatile. Especially during the last couple years during the pandemic, we've seen the price of bitcoin skyrocket and then plummet and then bounce all around in between.

Today, I want to focus a little bit on the topic of stablecoins, which you recently wrote an Economic Brief about. We'll put a link up to that. I was wondering if you could explain what stablecoins are and how they're different from these other cryptocurrencies that people may be more familiar with.

Wong: You rightly mention that the price of bitcoin [is] very volatile. You can easily lose five or 10 percent overnight. Buying and selling cryptocurrency like bitcoin is not very liquid. This means you have to pay some extra fee and wait for up to an hour to finalize the transaction.

There's a demand for safe and liquid cryptocurrency in order for people to park their money and for payments. Stablecoin has come to serve this market niche. Stablecoin try to maintain their price at $1.

Sablik: So, that means one unit of stablecoin is equal to one U.S. dollar?

Wong: Right, exactly.

Sablik: So, they're trying to maintain a more stable source of value. That ties into another question about how stablecoins and this decentralized finance system … compare to a traditional banking model.

Wong: Yeah, I think this is a good question to ask. We know a lot about how traditional banking works, so is useful to understand how stablecoin works when we use the lens of traditional banking.

Your checking account is useful for you to buy stuff and pay the bills. One dollar in your checking account is as good as a one-dollar bill in your wallet. It wasn't always the case. We have seen the money issued by the primary banks was trading less than $1 during the free-banking era. In the old days, checking accounts or primary bank notes were pegged at $1 because they were redeemable in other assets, gold or other government assets like money or government bonds.

So, back to the future, what makes a stablecoin stable? It's very similar, but the backing assets and the mechanism are different.

Sablik: That's a good segue. Maybe we can get into what those different methods are. My understanding is that some stablecoins might be backed by an organization that intervenes to maintain the price, while others rely on a computer algorithm. Could you explain the differences between these approaches and why you might choose one over the other when setting up a stablecoin?

Wong: Yeah, sure. I think there are several ways to peg the price of stablecoin. The simplest way is to back it by cash or equivalent. It is the case of some stablecoin — which [are] the biggest ones like Tether and USDC — the issuers make money [by] holding the cash or equivalent on their balance sheet as a backing asset to the stablecoin.

But you cannot make lots of money by sitting on this cash reserve because they have low returns, which is the downside of being safe and liquid. So, here comes something more complicated — a version of stablecoin where it's trying to be backed by an algorithm. It's what we call the "algo stablecoin."

There are two major types of algo stablecoin, but the main idea is trying to use cash as the reserve. The first one [uses] over-collateralization, for example the Maker DAI. In this case, there is an algorithm where you can give the algorithm $1 worth of ether, another popular cryptocurrency which is not stable and floating around. In return, the algorithm will log the ether and give you back 66 cents of the Maker DAI. Every Maker DAI is backed by more than 100 percent of the assets, which is ether. The 50 percent above each [DAI] absorbs any price fluctuation of the reserve, which is ether in this case.

Sablik: Gotcha.

To further the analogy to traditional banking, you mentioned about how the more stable the account is, the less lucrative it can be. I think most people would think their traditional checking account doesn't necessarily pay either any interest or very low interest. It sounds like with these algorithm stablecoins, they're trying to find ways to increase the returns but still maintain that stability.

One algo stablecoin, Terra UST, recently attracted a lot of attention earlier this year because its algorithm actually failed, so its peg to the dollar collapsed. What happened in that case?

Wong: It is a very interesting example.

We just mentioned one algo stablecoin which is the Maker DAI, which [uses] over-collateralization to peg its price at $1. Terra UST is another algo stablecoin but [does] something to try to be smarter, maybe too smart. I wrote an Economic Brief explaining what had happened and the mechanism in detail.

In a nutshell, there's a Terra blockchain and the Terra blockchain is issued through two cryptocurrencies — one is UST you just mentioned, which is stablecoin, and the other is the LUNA, which is the floating one. Every unit of UST is redeemable to $1 worth of the LUNA.

In this case, the price of LUNA doesn't quite matter. If the price of LUNA is $2, then a UST is redeemable to a half LUNA. When the price of LUNA is 50 cents, the UST is redeemable to two LUNA. So, in this case, the quantity of the LUNA will automatically adjust to give you $1 worth of the money for every UST in the background.

The way Terra can do it is because they can create — they call it mint or burn — the LUNA from thin air. Unlike the typical example, you are bounded by the ether for the case of the Maker DAI. You tied your hands by the supply of ether on your balance sheet. In [the case of] LUNA, you can only create it from thin air, so there is no physical constraint for Terra.

So, you can see as long as the price of LUNA is positive, the price of the UST should be automatically pegged at $1. It is only worth $1, in this case. In the banking example I mentioned, LUNA is like the bank equity and the UST is like the bank deposit, and it is only redeemable from the deposit to the bank equity.

This is, so far, in theory. In practice, redeeming UST from Luna or the other way around always involves some extra fee … called a spread. Also, redemption capacity limits how many LUNA can be redeemed or burned every day. There are deeper reasons for introducing these extra frictions, which I explained in the Economic Brief I mentioned earlier.

In a normal time, this spread and the redemption capacity [don't] quite matter. Most of the LUNA and the UST are traded among users on marketplaces like Curve Finance rather than involving any actual algorithm. When people believe the algorithm is working, they will trade UST at $1 automatically. In other words, what is backing the UST is the market rather than the actual algorithm.

But you can see the market approach depends heavily on the liquidity available. So, when the liquidity evaporate[s], people will go back to the algorithm. But then they find that the friction of the spread and redemption kick in and prevent them from redeeming UST. So, it's [people trying] to sell UST at a cheaper price and eventually picking a peg. This is essentially what happened in May of this year which made the UST totally fail.

Some people may be thinking the UST failed because UST is a Ponzi scheme or under some speculative attack. In the Economic Brief, I debunk this explanation. I think the deeper reason is the interaction of the unstable liquidity and the real-time human friction of the algorithm.

Sablik: That's interesting. I think it gets to something you were alluding to, which is this question about whether UST's collapse revealed some inherent flaws with stablecoins. Conversely, does it reveal some strengths of the traditional banking regulation that the digital currency space lacks?

Wong: It seems to me the heart of the Terra problem is the liquidity, which is flighty and fragmented. Flighty liquidity was a concern to traditional banking, too, which is why we have bank runs. That's why we have FDIC to provide deposit insurance — to eliminate bank runs. [Also,] when banks on short notice need liquidity, they can always go through the fed funds market, borrow from each other, or go to the Fed discount window. [Finally,] banks are supervised by the Fed and FDIC or OCC to prevent something bad from happening.

In sum, I think the concern is there is no deposit insurance for stablecoin. The algorithms don't borrow or lend from each other and there's no central bank standing behind stablecoin. The spirit of cryptocurrency is to avoid a central bank, so there seems to be some fundamental tension.

Sablik: Assuming that these issues can be solved, how might stablecoins benefit average households over traditional bank deposits?

Wong: I'm not quite sure. What is the real benefit of having stablecoin? I think the spirit of cryptocurrency is about disintermediation. In other words, you avoid the middleman. If you think about it, most financial activities is about the middlemen making the connection from deposits to lending, or in the case of IPOs, it's raising money for equity. The impact of disintermediation is interesting, whether there is stablecoin or not. Disintermediation is a trend we are seeing right now, even before we are having any crypto.

Sablik: It'll be an interesting space to continue watching, I'm sure.

Thanks very much for coming on today to talk about stablecoins. As this space continues to evolve, I'm sure we'll have to have you on again soon to talk about all the new developments.

As always, I'll remind our listeners who are interested in keeping up with the research that Russell or other economists are doing, you can head over to Richmondfed.org and check out the [Economic] Research section there. And if you enjoyed the show, please consider leaving us a rating and review.