Hacker News Reader

Comment by rglullis | Hacker News Reader

> Sure that's useful, but that' assuming that these coins should be worth something and are worth holding in their own right

Take the fiat-backed version of these tokens (not Tether, but ones that are backed by somewhat reputable companies) and you can have a distributed Foreign Exchange who can be reasonably profitable just by being a liquidity provider. Early last year I was getting 2% per month on USDC/DAI and about 1% per month on USDC/EURS. Even today one can get a lot more than you'd get on a savings account.

View on HN

> Early last year I was getting 2% per month on USDC/DAI and about 1% per month on USDC/EURS. Even today one can get a lot more than you'd get on a savings account.

Ask yourself how risky this investment is, and why existing financial services aren't arbitraging it away. 1% a month is ~12% per year, which is comparable to CCC+ corporate bonds.

Risky? Yes, substantially.

Question for you, though: how would "existing financial services" arbitrage it away? By entering the crypto market and copying Uniswap's contracts and just use smaller fees?

> Risky? Yes, substantially.

Then why not park the money in existing financial instruments which have a larger risk profile than savings accounts? CCC bonds offer >14.7% yearly returns. Argentina's 10 year bonds have ~50% yearly return.

> Question for you, though: how would "existing financial services" arbitrage it away?

If you think the annual returns are priced appropriately due to risk of default, then they can't. I only brought this up because you compared it the returns on a savings account.

Otherwise, they could just offer a 5% savings account, deposit the money into the crypto market you're talking about, and make an easy 6% return.

> Otherwise, they could just offer a 5% savings account, deposit the money into the crypto market you're talking about, and make an easy 6% return.

No, you can not. If you go to learn how Uniswap (the AMM system being discussed) works, you'll see that liquidity providers revenue comes from the swap fees. There is no compound interest, or any compounding effect for that matter.

If more people went on to become liquidity providers, the profitability of the liquidity pool would go down unless there was corresponding increase in the swap activity. There is no arbitrage there.

> If more people went on to become liquidity providers, the profitability of the liquidity pool would go down unless there was corresponding increase in the swap activity. There is no arbitrage there.

That's the definition of arbitrage.

Ok, let me rephrase... there is no "sustained" arbitrage opportunity there.

My point is that anyone that went on to offer 5% on the hopes of making more than that for any prolonged time would be making the exact same mistakes as BlockFi/Celsius/all others did.

There's no such thing as a sustained arbitrage opportunity. The whole point of arbitrage opportunities is that they are self-limiting.

For example, let's say there's a delta in terms of Bitcoin pricing in Korea vs the US - due to banking requirements. Let's call this the "Kimchi premium."

Arbitrageurs find a way to buy at the US price, sell at the Korean price, and keep the difference. Their buying in a low-cost market and selling into a high-cost market ("capturing the premium") brings down the price discrepancy. You can only keep doing this as long as the structural inefficiency exists and also the number of participants remains low. The more participants, the more volume, the lower the premium.

This btw is how SBF made all his money.

I understand what arbitrage is. I was arbitraging Bitcoin between Brazilian and American exchanges in 2012 already.

My point in saying that there is no "arbitrage" is that you won't make any money if you offer 5% to people and expect to make more than that. Could you perhaps do with a little less pedantry and understand the overall message, or are you going to continue with needless pontification? It's late where I am and I would like to go to sleep.

You’re using words that you don’t understand the meaning of.

Anyways, the person you’re responding to isn’t being pedantic. They’re describing a complete mechanism for how a bank could arbitrage the money.

If they could get 11% returns from AMMs with the same risk profile as a savings account, then they could essentially make free money by setting up savings accounts for other people, where they could offer them 5% returns, and then reinvest that money their clients deposited into the AMMs and receive 11% back.

Of which they would only need to pass on 5% to the clients who deposited the money into the savings account, keeping 6% of the returns without having to commit a single dollar of their own.

So why hasn’t any bank done this yet?

And you are trying to argue claims that I never made, such as:

> get 11% returns from AMMs with the same risk profile as a savings account

Where did I say that risk of being an LP is the same as putting the money on a savings account? I didn't.

You are making all this exposition and trying to lecture me based on an assumption that I never made and that I know to be false. IOW, this is at best a strawman and at worst it's disingenuous. In either case it's fucking annoying.

Read again. All I said was that even today being an LP in a stabletoken is beating a savings account.

- Did I say it was without risks? No.

- Would I say that this is relatively low risk, compared with other "investments" in crypto? Yes.

- Compared with money in a bank? No.

- Do I think it is worth it? As part of your strategies, yes.

- Would I tell people to take money from their savings and do this? No, of course not.

- Why not? Because not only it has risks, but also because if more people did it the (and if the transaction volume at the exchanges stayed the same) the ROI would go down.

- Is that the same as "arbitrage"? Not exactly. Liquidity providers make money even if all they are doing is to buy and sell tokens that keep a peg to the USD. The issue here is that flooding the pool with more cash is diluting the profitability.

- Could some whale come, put a bunch of liquidity in a pool and keep it only while the profitability is higher than the savings account? Shouldn't that count as a "arbitrage opportunity"? No, because there are costs to on-and-off ramp, there are risks associated with crypto and if you want to do that kind of arb, you'd be better off my looking into crypto money markets - which also have its own risk/reward profile.

What else do you need to be spelled out so that you can stop with the stupid strawmen?

You said:

> Early last year I was getting 2% per month on USDC/DAI and about 1% per month on USDC/EURS. Even today one can get a lot more than you'd get on a savings account.

I don't understand why you compared USDC's return to a savings account's if you didn't mean that USDC is a better investment than a savings account.

> - Compared with money in a bank? No.

Yes you did, above.

> - Did I say it was without risks? No.

> - Would I tell people to take money from their savings and do this? No, of course not.

Then it makes no sense to compare USDC to a savings account. The risks are much higher.

> - Could some whale come, put a bunch of liquidity in a pool and keep it only while the profitability is higher than the savings account? Shouldn't that count as a "arbitrage opportunity"? No, because there are costs to on-and-off ramp, there are risks associated with crypto and if you want to do that kind of arb, you'd be better off my looking into crypto money markets - which also have its own risk/reward profile.

But all arbitrage is risky, that's not against the definition and neither are on/off-ramp fees. Even if you're "better off" doing something else by your criteria, the opportunity is still there.

Edit: not to be too pedantic but:

> - Is that the same as "arbitrage"? Not exactly. Liquidity providers make money even if all they are doing is to buy and sell tokens that keep a peg to the USD. The issue here is that flooding the pool with more cash is diluting the profitability.

You're getting dollars somewhere and selling them where you will get more money for it. The opportunity goes lower as more people use it. Still sounds like an arbitrage to me, or something that's functionally the same.

> I don't understand why you compared USDC's return a savings account

First, not just "USDC". I was describing the process of being a liquidity provider, which involves a token pair.

Second, the fact that I compared with a savings account does not mean that I am saying that the risks are the same. I used savings account because I wanted to illustrate that this is something that can be done holding only "stabletokens", i.e, without speculating on the value of the token itself. I could've used money markets or fx funds, the idea would be the same.

> Not to be pedantic (...) The opportunity goes lower as more people use it. Still sounds like arbitrage to me.

Imagine you have an ice cream stand and you start making good money because you find a nice spot by the beach. Some days you are selling close to 100% of your stock, some days you even need to go get more at the supplier. Some of your competitors find out and move their trucks next to you. Nobody changed their prices, and the ice cream being sold by all of you are all of equivalent quality. Now, the street as a whole is still a "good" spot, the customers are still spending the same amount of money as before, but you are not making as much money because now you are splitting the customer base with the other guys.

Is that "arbitrage"? I certainly wouldn't call it that. There was at first some information asymmetry, but calling every asymmetry an "arbitrage opportunity" seems like a stretch to me.

The arbitrage is the ability to buy ice cream where it's cheap (the store) and sell it where it's expensive (the beach). The margin compress as more folks arrive to sell because your fixed costs (labor, the van, cogs) remain fixed while your sales volume drops.

Definition is: "the simultaneous buying and selling of securities, currency, or commodities in different markets or in derivative forms in order to take advantage of differing prices for the same asset."

So in this case you're buying a commodity (ice cream) at the store and selling it at the beach. You are taking advantage of differing prices for the same asset in different markets.

So yeah, still arb, imo.

If buying at X and selling at X + ε is "arbitrage", every retailer is merely "arb-ing".

Even looking at the definition you gave I don't think it pass the smell test, the scenario does not require the buying and selling to be simultaneous... not in the ice cream analogy, and much less in the real uniswap mechanism.

All arbitrage leads to the elimination of the arbitrage. That’s the whole point of arbitrage. It’s never sustained, because if it was, it would be the financial equivalent of a perpetual motion machine, simply generating value out of nothing.

Argentina quite reliably defaults on its bonds much more often than once a decade.

It's not so much that it's risky, it's that they'd have to spend a lot of money on the compliance position to tell them it's not and ... wait yeah, the bonds are a better idea in every way.

1%-2% per month doesn't pass the smell test for something stable. You're taking on some risk, it's just hidden from you. And thats the worst kind of risk to take on. Just look at the extremes. The big risk here is impermanent loss which is inevitable unless the relative price remains stable. Or they could both go down and relative to fiat you'd be down as well

Did I say it was stable?

> The big risk here is impermanent loss.

How would you get impermanent loss on a pair of tokens that are both pegged to the dollar?

Pegs exist until they break.

Mightier institutions than crypto coins have been brought to their knees trying to defend a peg.

These mightier institutions were trying to defend a peg while dealing with fractional reserve and/or a deficit in the trading sheet.

A stabletoken like USDC is not any of that. It is (or should be) backed 1:1 by dollars, and the institution holding USDs should be profiting only for the service provided, not by re-investing the money.

> It is (or should be) backed 1:1 by dollars, and the institution holding USDs should be profiting only for the service provided, not by re-investing the money.

Should be, perhaps. But it does not appear to be. Tether's statements indicate that they are very much in the fractional reserve game, and while Circle has removed the asset breakdown from their attestations, the last one I found that had it indicated that they were also playing the fractional reserve game.

Given that the two largest stablecoins have admitted to playing the fractional reserve game, are you willing to admit that the peg turn out to only be as solid as, say, the GBP/ESM peg was?

If you have any attestation from USDC saying that they are not holding enough USD or commercial paper to back their reserves, you can make a lot of money by a combination of (a) legal action against them (which is easier to do than compared with Tether) and (b) shorting it, as there are a lot of institutional investors who would be interested in breaking them.

I said USDC has said that is using fractional reserves. I did not say that it is insolvent.

(That said, I do believe that USDC and USDT are likely to be insolvent, primarily on the basis that assets equaling 100% of liabilities to three significant figures is suspiciously precise. More likely for USDT, as their attestation also notes that they're using non-standard accounting practices, although USDC gets a ding for not providing any breakdowns.)

> (a) legal action against them (which is easier to do than compared with Tether)

What standing has anyone to sue USDC right now? You need to have suffered damages, and so long as the peg has maintained, damages have not yet occurred.

> (b) shorting it, as there are a lot of institutional investors who would be interested in breaking them.

How do you short them? You'd need to acquire likely >10 billion of USDC to sell without providing 10 billion of liquid USD to Circle or any of its confederates that they could use against you.

> I said USDC has said that is using fractional reserves. I did not say that it is insolvent.

It doesn't matter. USDC promises 1:1 backing of dollars for cash or "cash-like instruments". If they are not able to not attest this and they minted more USDC than they should, they are breaching this promise.

The asset classes they have previously admitted to investing in are similar asset classes to those that money market funds have invested in, and historically, that has not proven sufficient to always maintain a 1:1 backing. And also, I believe, not inconsistent with a description of "cash-like instruments" (at least insofar as any legal liability could result).

Of course, the more you compare stablecoins to money market funds, the more you wonder why anyone should prefer the former over the latter: stablecoins appear to be a less regulated, more opaque version of money market funds, with more meager rewards and a richer repertoire of liars and scam artists.

Are you sure you are talking about Circle (USDC) and not Tether (USDT)? Let's not confuse one for the other, because I already said right at the top that Tether is not to be trusted.

The peg could be broken as we've seen with a number of stable coins so far this year. I've explained in previous comments why algo-stablcoins are doomed to fail. The traditional stable coins require a lot of trust and I'm not sure how they fall in the capital structure of the company. Even if it were backed 1:1 with fiat, if someone like Coinbase goes into bankruptcy, is the pool of money for their stable coins protected?

> if someone like Coinbase goes into bankruptcy, is the pool of money for their stable coins protected?

Theoretically, yes. Circle's reserves are separated from Coinbase. So one could be worried about their assets if they were being held by Coinbase, but if you are holding USDC on your own Circle should still be able to redeem USDC for dollars.

> How would you get impermanent loss on a pair of tokens that are both pegged to the dollar?

A "peg" for a cryptocurrency or token is a promise, not a guarantee.

Then your concern is not about impermanent loss, but simply of the token's value going to zero.

Real (traditional) successful market makers are rather more sophisticated than anything that could possibly run on the Ethereum blockchain. The fact that a market maker is a decentralized automatic magical thing that is supported by cryptography does not mean that it has a chance of competing with a good algorithm that runs on an actual computer (or FPGA or ASIC!) if the cryptocurrency market ever becomes legitimate to the point where modern market makers fully participate.

You are missing the point. Those providing liquidity are not making money because of their sophisticated trading strategies. They make money by being the exchange.

What is “the exchange”? They are making money be being an exchange and the major market maker on it. But the fees are insane by traditional finance standards, and the barrier to entry of creating a new exchange is pretty low.

The barrier to create a new exchange is almost zero. The code from Uniswap is open source, you can clone and be running a competitor in a day.

The challenge would be in providing liquidity at levels that could compete with Uniswap's. You can go buy your userbase like Sushiswap did, but you'll see that LPs already learned their lesson that what matters is the transaction volume and these governance tokens being given as an incentive are worthless, so you need to buy both makers and traders.

If you are willing to keep a lot of capital around for a long time and taking the risk of impermanent loss, eventually you will be able to take users from Uniswap. But who would do that?

You're aiming for the 2% per month, they are aiming for your 100%.

The only question is who gets there first.

Who is "they"?

Previous investors and potential grab the bag and run ponzi scheme owner.

There are no "previous" investors at a Uniswap Liquidity Pool.

If you are going to attempt taking a dump at me, at least have some idea of what we are talking about.

the people who took the 2% for a while before you did. And got out before it died.

Sorry if I was unclear, you might be in this camp too.

There was nothing to be "taken". Liquidity providers on Uniswap are market makers. They are (automatically) buying and selling tokens, and receiving a fee for their service.

Anyone can create a liquidity pool of any pair, as they see fit. It does not require any "prior" input. There is no staking, no dividend or interest of any type.

The profit comes solely from the fees, and the profitability of the pool does not depend on how many people "get in" or "get out", it depends solely on how many people are willing to swap token A for token B. The more people transacting, the more fees are generated.

You started this thread with some completely wise-sounding quip, but each of your responses are showing that you have absolutely no idea of how things actually work on decentralized exchanges, and that all you have are blank statements and preconceived notions about crypto.

Where is that return of 2% per month coming from, and why are they giving it to you?

Users of a token swap pay a fee to have their tokens swapped. This fee gets distributed among the people who are providing liquidity proportional to how much liquidity they provide compared to everyone else.

They are paying a transaction fee because they've deemed that token swap as being the most convenient or cheapest way to make the exchange.

Reply on news.ycombinator.com