Hacker News Reader

Devil's advocate: 2008 was bad because of regulatory failures (specifically deregulation). Lack of transparency was only part of the problem. So if you want to avoid 2008 in cryptocurrency, the answer would be more regulation, not less.

> So if you want to avoid 2008 in cryptocurrency, the answer would be more regulation, not less.

To expand on this point. Deregulation with a federal reserve fallback is especially dangerous because it's essentially fake deregulation. When you know that you're too big to fail and the government is going to bail you then your risk to allocate capital effectively becomes zero.

I think the Boeing CEO (or some other airline) was famously quoted for this that they do not need any cash buffers, as the government will always bail them out, no matter what

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3571441

Thus, it’s ok to spend extra cash to stock buybacks and maintain no cash reserves.

I mean, he isn't wrong, is he? If he's wrong, then maybe the next time Boeing needs a bailout, we should make an example of them and not bail them out. If we don't have the balls to do that, why should they have a cash buffer?

Was 2008 an issue of lack of regulation or lack of enforcement?

I thought it was more about "too big to fail" attitudes preventing enforcement from being fairly applied to the largest participants.

It's a little unclear, because (like many things in public policy) there's several putative explanations for what caused the crises, peoples' preconceived notions color which ones they favor, and it's manifestly unclear who is right. And it should be remembered that "everyone" is a perfectly plausible answer.

At a broader level, there is most consensus on the idea that the key failure in 2008 was that no one--neither investors nor creditors nor credit-rating agencies nor regulators--correctly understood the risk profiles of the new financial products, and things that were thought to be very safe or at least combined uncorrelated risk profiles suddenly started all taking in unison at the same time.

> It's a little unclear, because (like many things in public policy) there's several putative explanations for what caused the crises, peoples' preconceived notions color which ones they favor, and it's manifestly unclear who is right.

I thought 2008's cause was perfectly clear: the banks all sold insurance policies on each other (in the form of credit default swaps) that they couldn't fulfill. When one bank went under due to bad bets in the market, this triggered policies that, if fulfilled, would bankrupt the next bank. That in turn bankrupted the next, and so on.

People focus on the conditions that bankrupted the first bank, but that was just the final snowflake that happened to trigger the avalanche.

But why was there bad debt on the market ? Yes the defense mechanisms couldnt work since they were interdependent, but I saw a hairdresser on tv talking of how she used to flip 600k houses before the crisis, makes me think she s also part of the problem by failing to obey general accounting principles (dont borrow what you cant repay just because market goes up).

And no, saying natural idiocy in the citizenry is to be embraced, does not simplify the issue: it was caused by natural idiocy which was enabled by lax enforcement. There are countries where your credit score matters not at all and your current ability to actually repay is analyzed.

You can fix the banks all you want but if people take stupid loans you ll never run out of ways to crash.

> But why was there bad debt on the market ?

This is my point, the focus on bad debt is like obsessing over the last snowflake that fell before the avalanche. The bad debt didn't cause the avalanche, the underlying structure of credit default swap contracts between the banks did. They were all writing each other checks that they couldn't cash. If the underlying problem of unaccounted-for credit default swaps didn't exist, the bad debt would've just been another snowflake falling on the ground, essentially.

Doesn't all insurance work like this though? Insurance companies can pay for individual losses, but are bankrupt if everybody loses at once.

Yes, but these weren't insurance companies, they were retail banks. That was the problem.

That doesn't square with Goldman's bankers chopping up and retranching known low quality derivatives, submitting them to the rater, and not being surprised when they were rated far less risky than the same mortgages when rated in previous tranches. To some degree there is blame to be shared, but there is also fraud for which nobody was truly held accountable.

Not sure if it’s not both.

In America, the SEC had been systematic defunded and weakened - over successive administrations.

The argument, in the years before 2008, was that more regulation only holds wealth creator back. Especially when the market self corrects itself.

To suggest otherwise was even nearly un-American.

The SEC policy had adapted to pursue slam dunk cases, ensuring that it’s resources were spent in achieving some form of guaranteed deterrence.

However things like CDOs and the rest were out of its reach.

This leads to the low oversight.

The self correcting aspect of the market was also more mythical than practical.

As a financial analyst we were encouraged to ensure that model predictions were within range of other firms.

Multiples out of range meant that you were likely wrong.

Faith in the market became a shortcut meaning “someone else has done the homework”.

With all the rating agencies beholden to the purchaser of the rating, you start setting up the dominos for 2008.

Too big to fair came after 2008. Before then it was assumed that if a large institution did fail it would be orderly and the system could take the shock. That turned out to be wrong when it did happen...

It was entirely a result of deregulation.

https://en.m.wikipedia.org/wiki/Glass%E2%80%93Steagall_legis...

Right, OP, CDOs were also very innovative when they first got rolled out!

(OR) these sort of crisis cannot be fixed by regulations. Having more of the same may not necessarily fix the issue.

Hot take: 2008 was caused by over regulation, resulting in broken incentives for rating agencies and preventing any real bad outcomes from them just rating everything as great.

Finding a trustworthy oracle is always a problem.

Either you can let (systemically important) people buy whatever junk they want, or you can mandate a minimum bar.

But if you mandate a minimum, who judges what does or doesn't meet it?

The issues weren't with risk per se, nor any minimum bar, but with risky assets dressed up to deceive buyers (and those that audit collateral) into thinking they're not risky.

If something is too complex to interpret: maybe that's not immediately a problem, but it likely shouldn't serve as collateral for a loan (or similar), especially not systemically.

So this conceptual oracle doesn't need to be perfect: it just needs to accept a few false negatives (unnecessarily rejected collateral) in order to ensure there are few false positives (unwisely accepted collateral). Philosophically it certainly feels to me like an easier problem than finding a more generally trustworthy oracle; but these kind of things certainly are tricky.

I don't think this is an oracle problem. I watch reviews online for products I want to buy. Seems to work fine. But let's say now legal action is threatened for any negative reviews. What happens to the trust now of online reviews?

It's more than that. Do you read reviews paid by you or by the seller of the product they are reviewing?

I've kinda lost track of the analogy at this point. I'm surprised I was downvoted so much. I didn't have a strong case to begin with. To try and carry on, if I was paying for product reviews like companies do for the rating agencies, I'd expect them to be even more accurate and impartial! If the companies (reviewees) are paying the reviewers (bribes?), I'd expect them to be crap. But companies are not paying Fitch to rate them are they. Other investors are paying Fitch for the supposed impartial review. But since high ratings makes markets go up, and monetary policy is greasing the wheels so that nothing can fail, it caused all ratings to creep up. Because after all the rating are based on empirical data and nothing is allowed to fail (bail out).

> But companies are not paying Fitch to rate them are they.

They often are.

A lot of times a large investor pays them to look at a possible investment, and this case is quite ok. But the official ratings they are listed with are set by a third party mandating the company "go pay Fitch to rank you". It's not plain bribery like companies paying reviewers, but it's quite shady.

I understand what you're saying but ultimately it's still the investors paying Fitch. You can clearly see they have no quid pro quo with the company or product they are evaluating. All your saying is who's invoice the bill appears on but the company is not paying that bill. The investor pays through the acquisition.

There is no investor at this point. The mandate is made by a 3rd party that is expected to be neutral. The investor only appears after the entire thing is settled.

Yes, it's not a case of fraud. It's a case of a process structured in a way that facilitates fraud. (Who pays for the invoice has the power to negotiate its value.) It is a very different thing, but it's still a problem to be solved.

I think the rating system he referred to was supposed to be independen.

Why are stable coins interesting?

It's basically a thing guaranteed by single entity that promises that if the price drops they will buy tether at personal loss.

It's as safe as any pre-crypto, single company e-currencies. All of which died.

The only thing they have going for them is that they managed to place thrmselves as currency of the crypto market.

Because not all depend on a company buying at personal loss. DAI has a complex mechanism (at least to me). To keep itself oscillating around 1usd. https://en.m.wikipedia.org/wiki/Dai_(cryptocurrency)

DAI looks interesting. What about Binance USD, USD Coin, Terra USD, TrueUSD, Paxos Standard, Fei Protocol, Liquidity USD, Frax, sUSD, USDX [Kava], mStable USD?

How many of those try to do something interesting and how many are just single entity backed coins like Tether?

Stable coins like Tether are not interesting, for precisely the reason you stated. DAI is decentralized, not like Tether.

Stable coins with centralized generation are not interesting.

> Blockchains like Ethereum are currently far more transparent than the state-sponsor US monetary system. I hope to live in a society where economic experiments can be executed, new lessons learned, and transparency maintained. If stablecoins are persecuted, the need for cryptographic zero-knowledge proofs will allow these systems to continue with far less transparency.

This is the part that excites me about Ethereum, and also scares me when I hear talk of "regulation" for the space. By definition, a lot of the projects on Ethereum are a LOT more transparent than the traditional finance world. I can interact with something like DAI, and see exactly what backs it at every second of every day. I have an idea of what would happen if certain assets that backed it dropped too quickly.

With most of the traditional finance world, it's all very grey and hidden. I'm told that there is regulation "protecting me", but in a lot of cases that regulation doesn't really accomplish its goal and instead acts as a moat.

> By definition, a lot of the projects on Ethereum are a LOT more transparent than the traditional finance world

The important stuff is not transparent. For example, how do you know that Tether is backed by real USD? That Binance isn’t manipulating prices via wash trades? That the promises made about various coins are being upheld? etc.

Crypto is reliving the era of https://en.wikipedia.org/wiki/Wildcat_banking

> With most of the traditional finance world, it's all very grey and hidden

What specifically are you referring to with “traditional finance”?

> The important stuff is not transparent. For example, how do you know that Tether is backed by real USD? That Binance isn’t manipulating prices via wash trades? That the promises made about various coins are being upheld? etc.

You are describing so called centralised finance (CeFi) whereas the most high quality Ethereum projects focus on decentralised finance (DeFi). DeFi tries to break any chains to opaque companies and have all actions happening transparently and auditable on-chain. However ones like Tether murky the waters a lot here.

> DeFi tries to break any chains to opaque companies and have all actions happening transparently and auditable on-chain

How can you create anything of value¹ by limiting yourself to on-chain transactions?

If value is created, the transaction would cover something happening in parallel, and your transparency does not extend to that.

¹ I do not consider lotteries or similar zero-sum games as value creation, but those can be 100% on-chain.

Any economic system works because people choose to believe in it. Specifically, people with guns. For instance, I have US Dollars, and I can exchange US Dollars for goods and services because everyone agrees to let me do so. I don't have them on me though--my bank has a record of how many US Dollars I have, and if they suddenly decide I don't have US Dollars for a sufficiently bad reason, that's some type of embezzlement, and I can go through a process involving highly trained professionals arguing on my behalf in front of someone wearing a black robe, and eventually men with guns will force them to give me my US Dollars back.

Likewise, we also all choose to agree that the owner of a car or a piece of land is whoever holds legal title to that car or piece of land according to some records maintained by some government office. If you think you own my car, and you make a duplicate set of keys and drive away in my car, I can get men with guns to go try to stop you from doing that because the official record says it's my car. But if you're the repo man, and I miss enough car payments, you can get the men with guns to take your side instead of mine, according to a similar process involving trained professionals arguing in front of someone wearing a black robe.

So yeah, the actual value isn't on-chain. What the chain replaces is the system of implicitly trusted institutions keeping records and adjudicating disagreements. Sure, out there in the real world, you have my car, but it's the record keeping that helps the men with guns decide whether to let you keep it or throw you in a cage for the next few years.

You can create value by for example connectind lenders and borrowers. The money could be used in real economy, albeit does not happen too much yet today. For example, developing nation export companies can get a dollar loan with better terms they would get from their local corrupted bank.

Developing nation export companies can already borrow money through international markets. If they are going through the corrupt local bank, it's either because they are forced to by legislature (In which case crypto does nothing for them), or because they have corrupt executives who are getting kickbacks (In which case crypto also does nothing for them.)

Is there an actual problem that this solves?

How would you ensure that the companies repay their loans?

> As a different matter, how can society say "2008 bad", which resulted in little economic progress, and then threaten problem solvers from making new solutions? Innovation is a hedge against regulator lethargy.

What do you believe caused regulators of financial instruments to be created?

If you think DAI is exciting, check out Liquity : it's entirely managed by smart contracts and economic mechanisms, with no governance managing interest rates like with Maker.

One Explainer video with the founder: https://www.youtube.com/watch?v=bXLTE-5BkhA

I think DeFi is really getting interesting.

A bulletproof algorithmic stable coin like IRON (dropped to $0.70) and SafeDollar (dropped to zero)?

https://rekt.news/safedollar-rekt/

https://rekt.news/iron-finance-rekt/

Algorithmic stable coins have a terrible history of failing. I wouldn't discredit tether, a simple redeemable token, then promote DAI. Also Ethereum contracts have a bad history. Their first launch they got hacked for $150 million USD and then reversed the pseudo decentralised network to try and get that money back

> I hope to live in a society where economic experiments can be executed

Totally read experiments as experts

Isn't DAI backed by USDC and Tether in part?

The debt ceiling for Tether is currently set to 0 [0] for dai. And USDC + other stablecoins do partially back it [1].

[0] https://daistats.com/#/collateral

[1] https://dune.xyz/SebVentures/maker---accounting_1

You are correct. Last time I checked, more than 50% backed by USDC. Makes no sense as the whole point of censorship resistance is defeated.

from https://daistats.com/#/overview it's currently ~26% USDC

It's also 200% collateralised so USDC could go to zero (or freeze all of the DAI locked USDC) without breaking the peg on DAI

Where/how did you check this? >50% of USDC sounds unreasonably high.

In part yes, along with many other assets. They've stated they'd like to lesson the use of stables as collateral and are still brainstorming solutions. The ability to add real estate as collateral should be interesting.

> Tether is rat poison

It would be very useful if we could kill vermin with it.

Stablecoins reinforce dependence on the State, as they further entrench the fiat system that Bitcoin was meant to replace. It seems to me that this, and the rest of the "Decentralized Finance" movement is not intending to replace the current financial system, but rather reinforce it.

I don't think financial innovation is the answer to preventing another financial crisis considering that's exactly what caused the 2008 crisis.