Hacker News new | past | comments | ask | show | jobs | submit login

It will be interesting to see if interest rates in government controlled currencies and crypto currencies will stay diverged in the long run.

Without the distorting action of governments printing money, interest rates might be set by market forces in "crypto land".

This might lead to a long term situation where artificially low interest rates are paid in government controlled currencies, but market prices are paid in crypto currencies.

Or will cheap interest rates in government controlled currencies somehow bring down the interest rates in crypto currencies?

The 10 year yield of bonds in Europe is at 0%. While US bonds are currently at 1.5%. This might be an indicator, that rates in one currency will not completely control rates in other currencies.




I think the primary reason interest rates are high in crypto lending right now (also on USD stablecoins) is leveraged speculation - people using crypto as collateral to leverage their crypto positions higher (and/or yield farm) and are prepared to pay a relatively high interest rate for it (~10% on stablecoins) because the expected gains are a fair bit higher.

I don’t expect this will last forever


Anecdotally many crypto holders prefer to not liquidate their crypto positions but instead take out loans via BlockFi to purchase hard assets like real estate. With cash in hand and now a property, they can get a cash out refi and the crypto loan is not a taxable event since the crypto was just collateral.


Like this?

0: Holder has ETH

1: Holder borrows Tether, provieds ETH as collateral

2: Holder uses Tether to buy a house

3: Holder borrows Dollar, provides house as collateral

4: Holder buys Tether with Dollar

5: Holder pays back Tether, gets back ETH.

6: Holder now has ETH + House + Dollar Dept

If so, why couldn't they lend the dollars to buy the house in the first place? The bank which lends the dollars certainly could make a contract that the dollars only can be used to buy the house?


Real estate financing is primarily done by lenders with very traditional underwriting practices that don't factor in crypto or avoid it entirely. Even HNI/private banking doesn't consider crypto part of a liquid portfolio. You can go to Interactive Brokers and get a 1.x% rate with 2x (margin) or 5x (portfolio margin) but that's off non-crypto. To get that same leverage with crypto, you can put in a down payment for real estate (20% down) in 5 properties for 5x leverage. BlockFi/Tether etc. make that liquidity possible without liquidation. Feel free to DM me for more conversation on this topic.


Rates are lower on margin loan, and they’re easier to get compared to a proper mortgage (assuming, you know, you have value assets to borrow against)


Rates might be lower but they're not usually fixed. They're easier to get assuming your liquidity is in equities versus crypto or even a stable income. There's a reason mortgage financing is still more prevalent than margin loans despite what you've stated which are inherent benefits.


You say a bank loans money to someone who owns a certain house, but not to someone who uses the money to buy the exact same house?


Yep this is the other use case for crypto lending


You say that one can create leverage by borrowing? How does that work?

Is it a contract like "You borrow me 1 ETH and I will pay back 1.1 ETH in a year. Except when X happens, then I will pay back 2 ETH"?

If so, what is X?


The leverage is speculation using borrowed money. See: https://www.blueleaf.com/articles/how-leverage-works-in-inve....


That does not answer my question: How a smart contract can implement lending in a way that both parties are satisfied:

- The lender gets their money back plus interest.

- The borrower makes a profit if the price of the borrowed currency goes up.


Majority of the lending platforms to date are centralized and custodial solutions


Check out Aave or Compound. You basically use your crypto as collateral, and do whatever you want with the borrowed money. If you want to take the borrowed money and get even more (leveraged) exposure to crypto, you can. You don’t pay back more than the borrowed capital plus interest.

https://docs.aave.com/faq/borrowing


Since ETH staking returns do represent a (sort of) risk free rate for ETH denominated loans, low fiat rates would pull down on those rewards, if the exchange rate volatility comes enough for people to stomach the currency risk.

The same sort of flows balance exchange rates and interest rates among national currencies, though as you point out, sufficient risks and controls abound to prevent complete parity between the rates.

I’d challenge the notion that one side is paying market rates, and one isn’t, at least in nominal terms. Bond purchases and ETH stakes are both market transactions at the prevailing rate, just with different structural forces in play.


I agree completely. I think the greatest value proposition of non-government money (e.g. gold, bitcoin) — at least in the long run — is the ability to escape from zero/negative interest rates.

I also agree that the different between EUR and USD bond yields seems to indicate that there exists no reliable way to arbitrage the spread in bond yields between different currencies (or, at least, that there exists a certain spread beneath which it's not profitable).


You're comparing a risk-free rate (government bonds) with a rate on risky investments. Not really comparable at all. You can also find high-yield bonds (aka junk bonds) denominated in USD.

Also real interest rates are set by the market not governments. Monetary policy has, at best, only small effects on real rates (in theory it should have none).


> You're comparing a risk-free rate (government bonds) > with a rate on risky investments.

I don't think so. What is the "risky investment" here?

I compared lending of different currencies. In the case of Euros or Dollars, the lender is a government. In the case of crypto, the lender is a smart contract. Both are assumed to be reliable.


If your smart contact is truly reliable and anyone can lend on it at a rate of their choosing, and all loans are fully and perfectly collateralized, the natural interest rate is exactly the risk free rate. As above, for ETH I think that ends being the return on staking, minus the costs of actually running a staking node - so pretty close to zero?


The last resort of the US, UK, etc. governments is to print money if they really want to avoid defaulting on their debt. Can smart contracts print money? Surely there must be some way for the counterparty in the contract to default.


"Defaulting" in crypto means the collateral goes from the borrower to the lender. The collateral is always more valuable than the borrowed asset. Otherwise, all borrowers would "default" all the time. As there is no other downside to it than to lose your collateral.


So, the borrower starts with X BTC. They post X BTC as collateral and borrow X-y BTC (where y > 0). Once the loan is paid off, they get the collateral back. This means they end up with X BTC minus the interest paid on X-y. Why would anyone do that?


I am not sure if lending makes sense if the collateral is the same asset that is borrowed.

A more typical example I can envision:

Someone owns land in Decentraland. The land is an NFT on the Ethereum blockchain. To make profits from the land they need to put a hotel on top of it. But they don't have the means to buy/build the hotel. So they lend Decentracoins (some other asset on the Ethereum blockchain) and provide the land as collateral.

If all goes well, the borrower buys/builds a hotel with the decentracoins. Makes more decentracoins from visitors. Pays back their debt.

If it does not work out, the land goes to the lender.


> I am not sure if lending makes sense if the collateral is the same asset that is borrowed.

It doesn't. And if the collateral is a different asset, there is no certainty that the value of the collateral exceeds the value of the principal. So there's a risk involved and that explains the premium over the risk-free rate. It's got nothing to do with the fact that the loan is denominated in some cryptocurrency.


> It doesn't

Proof needed. I can in fact think of counter examples:

Say there is a DAO that gives more voting power if you own more ETH. In that situation, it might make sense to borrow 900 in ETH with 1000 ETH collateral. Then you make your vote on the DAO with a power of 1900. And pay back 910 in ETH to the lender. The vote on the DAO might trigger an action that is worth more than the 10 ETH you paid in interest. For example if you run a company and the vote on the DAO was to buy a service from your company.


> it might make sense to borrow 900 in ETH with 1000 ETH collateral. Then you make your vote on the DAO with a power of 1900

No, you would vote on the DAO with a power of 900, because your original 1000 ETH collateral is being held by the lending protocol


Not necessarily. There are mulitple ways it could work.

Two examples:

The DAO could support the lending protocol. Counting your assets in the lending protocol towards your voting power.

The lending protocol could support the DAO. Not allowing you to withdraw you collateral but allowing you to signal something to the DAO.


> What is the "risky investment" here?

Well, you tell me. Where do the profits come from?

> In the case of crypto, the lender is a smart contract.

That doesn't make sense. The smart contract is a contract, that is, an agreement between two or more parties. An agreement is not a lender. The lender is one of the parties.


> The lender is one of the parties

The "parties" do not know each other. And it does not matter who put the contract up, who put assets in and who borrowed assets from the contract.

Because everything is in the smart contract. Even the assets. Are you aware that Ethereum contracts hold assets?


Let's pretend the lender is "smart contract" (even though it's not). You lend X to the smart contract, and after a while the smart contract pays you X+y. Where does the y come from?


> even though it's not

We should sort this out first. It does not make sense to discuss higher level concepts if we disagree on lower level concepts.


the risk-free bonds have failed numerous times. Often they wiped out investors entirely.

Full list: https://en.wikipedia.org/wiki/List_of_sovereign_debt_crises

They cannot print value, only paper.


And you'll find these failed governments bonds were paying a considerable risk premium, i.e. a spread over the risk-free rate.


Not necessarily. Since the government can print the money they lend, they can dictate the risk premium. They can keep it as low as they want.

Italy and some other European countries are basically broke. But they pay less interest than the USA.


The risk premium is the interest rate spread that investors demand. Not clear how the government can "dictate it" by printing money.


The government prints money and buys government bonds with it. And other assets. When you have infinite amounts of money, you can dictate the risk premium.


> When you have infinite amounts of money, you can dictate the risk premium.

But how? If you're an investor who is considering buying government bonds, how can the government dictate the interest rate that you are willing to accept in return for buying the bonds?


read up on yield curve control. https://www.stlouisfed.org/on-the-economy/2020/august/what-y...

Basically: central bank has infinite money and buys all the debt at 0%, and outbids everyone else.

Simple.


The government is an investor. If the government buys enough bonds at X% interest rate, that is the market rate.


Sorry, that's not how it works. The governments runs an auction where the bonds are sold to investors. That sets the market rate. The government would not buy its own bonds in the auction (it wouldn't make sense) and cannot force investors to buy the bonds at a particular rate.


In Europe, the government does buy its own bonds. Via the central bank. They print the money. Hold it for a few days. Then buy government bonds with it.

Read what the european central bank (ECB) writes about it:

https://www.ecb.europa.eu/pub/pdf/scpwps/ecbwp1956.en.pdf

https://www.ecb.europa.eu/mopo/implement/pepp/html/index.en....

Search for "government bonds" in these papers.


The central bank buys bonds in the secondary market, and any effect on the risk premium is only indirect (investors perceive the bonds as being safer). This is completely different from the government "dictating the risk premium". The risk premium is the spread deemed necessary by the market to compensate for credit risk, and as such it can't be dictated by anyone.


Then there is also the downstream effects of buying these "safe" bonds on the secondary market: strip mining the "highest" quality collateral out of markets, and making markets ever more reliant upon fewer on tradeable cusips…


Yet, i've seen defi protocols that tried to fix the rate + maturity of said "bonds" they were offering and be complete baffled why demand for them changed when market conditions changed… wish there were more voices like yours in the sea of fecal matter floating around in the ecosystem…


Thank you for wading in and trying to speak to them sensibly




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: