Hacker News new | comments | ask | show | jobs | submit login
The ICO paradox and how to fix it (sia.tech)
112 points by zherbert 11 months ago | hide | past | web | favorite | 101 comments

The way to fix it is through enforcement of the SEC laws. We aren't going to combat fraud without proper enforcement. The industry needs more regulation because right now most investors are getting fleeced and getting left with nothing because there is a large pool of unsophisticated investors.

This solution would give SIA the ability to regulate these ICOs and that has even more issues (what makes SIA authoritative)? SEC regulation even though seems like a bitter pill for people who are in the cryptocurrency space is needed for things to actually get done since every fraud basically turns away investors that actually want to fund a real project.

Note that the TCO structure proposed in the article is compliant with SEC regulations and is under SEC oversight; they're not proposing self-regulation.

Ultimately, SEC regulations don't allow people to do what they want to do (and I don't think the limitations of SAFTs/TCOs have become obvious yet), so you're really just calling for ICO-like fundraising to be banned. That would be fine with me, but don't expect people who raised millions to just give that money back with no resistance.

"Ultimately, SEC regulations don't allow people to do what they want to do"

Because people did do the exact same thing before, and it turned out to be rife with fraud. That's why the regulations were put in place.

"so you're really just calling for ICO-like fundraising to be banned."

No, they're just calling for it to be called what it is: A Security.

Agreed. Many ICOs have been securities offerings.

Not all of them have been, however. Ethereum was arguably not a security. It's not necessarily a security offering if you are putting something up for sale that has a value of its own, or will have a value when the transaction occurs, such that people buying the thing are not expecting to get anything like a "return on investment", and are not expecting to somehow profit from the work of the company who is making the sale.

Consider Kickstarter-style presales: the people who buy these offerings are buying a product, not investing in the company. If you were, for example, just selling a token where the token has some intrinsic value (such as enabling you to pay for distributed compute or storage), and where you're clear to buyers that there's no expectation of any kind of return, just the token's intrinsic value, then that's probably not a security.

I believe I also read that there are requirements that the thing-you're-buying have value at the time the transaction takes place, though I'm less sure about this. If you take presales for a token that doesn't even exist yet, with a plan to fund actual development of the token off the presale funds, then that may be a security. If the token that you sell has value at the time you sell it, or will have value at the time the transaction takes place, then that's probably not a security.

If you read the Ethereum project's presale materials, they were very careful to stay within all of these boundaries, from what I can discern. This is my layman's analysis of securities law - IANAL.

One thing that confuses me about securities is how vague the description is. Why isn't a concert ticket a security if the majority of purchasers are reselling them?

Presumably because first-party ticket sellers don't encourage resale. In the U.S., there's a legal test to determine whether something is a security: https://en.wikipedia.org/wiki/SEC_v._W._J._Howey_Co.

A concert ticket isn't an ownership interest, it's a license/permit to attend an event.

> SEC regulations don't allow people to do what they want to do

That is a feature, not a flaw.

Opinions vary.

It is a flaw for those banned from being recognized as "accredited investors".

It is a feature for those lucky to have competition banned.

Can you be banned from being an accredited investor? AFAIK, the only requirement in the US is that you have a net worth of $1m or an income of $200k a year. It's not perfect, but it's the government's best way of determining whether you have the ability to suffer significant financial losses.

Again, there are 2 equally valid explanations.

One is to say that the government want to protect people against themselves, to make sure they can withstand significant financial losses.

Another explanation is that those who match these conditions (I'd guess about 2% of the population) have successfully lobbied the government to maintain their advantage against the remaining 98%.

I do not understand why people downvote when I just say out loud the less palatable alternatives views. It seems to me we are not considering all the possibilities there.

Their advantage isn't that other people aren't allowed to make certain types of investments, though. It's simply that they have more money. Even if "accredited investor" weren't a thing, they'd still have access to better investment opportunities, because they can offer better terms. Compared to someone who has 1000x more money than you, all you can offer is the willingness to take on more risk — which is exactly what accredited investor rules are meant to protect people against.

> The industry needs more regulation because right now most investors are getting fleeced and getting left with nothing because there is a large pool of unsophisticated investors.

I agree with the premise

> The industry needs more regulation

but not the reason,

> unsophisticated investors

If you are an unsophisticated investor trying to invest in an ICO and you lose out big, I think that is on you. I don't try my hand in algorithmic day trading, or long term value-based investing either because I know I am not informed enough to be able to make money, so I stay out.

With the social and financial security nets we have, a person’s bankruptcy is on all taxpayers (as soon as they get food stamps, Medicaid, etc.). There’s an incentive to protect people from devastating financial mistakes, regardless of political slant.

That is one of many arguments that we should cease to be so socialistic.

So let people starve so that a few others can get richer?

The question, as always, is what did those people do to earn the food with which to not starve?

(Many people do just fine by producing things of value and engaging in voluntary acts of trade.)

I mean, I'm coming from a place where I don't think we should let people die because they are currently not economically productive.

If you think that people should die if they can't produce a surplus, you're free to have that opinion. Just don't hide it behind phrases like "we should cease to be so socialistic".

I also hope youre the first to stand up and refuse help when you find yourself in a situation where your skills are no longer valued

> I also hope youre the first to stand up and refuse help when you find yourself in a situation where your skills are no longer valued

I absolutely would -- because if my skills are no longer valued, it's time to pick up some new skills.

Many will say “cut food stamps and Medicaid”

>If you are an unsophisticated investor trying to invest in an ICO and you lose out big, I think that is on you. I don't try my hand in algorithmic day trading, or long term value-based investing either because I know I am not informed enough to be able to make money, so I stay out.

Like it or not, but there are plenty of "unsophisticated investors" that don't know to stay out.

There's an old saying about fools and their money...

It's like saying that anyone losing something in free market had only himself to blame and there should be zero regulation of markets of any kind. Yeah, that works out perfectly in the long run. /s

This is doubly a strawman. "Don't invest in things you don't understand" is pretty much rule 1 of investing, which is quite different from the general case of participating in a market. Anyway, GP already said they agree that regulation is needed.

Your words are true at face value.

SEC enforcement does not fix the problems that we discussed in the post. Many projects have raised money through Reg D exemption filings (such as Blockstack and Filecoin), but they are still ultimately selling utility tokens.

We are arguing that the incentives behind ICOs of utility coins are bad. I could easily devote more blog posts to the legal/regulatory side, but that is not our intention for this post.

My biggest concern for the whole crypto industry right now actually is strong regulatory action. The problem is that a lot of these things look, smell, and act like securities, and a lot of people think that all tokens should be classified as securities (including regulators).

I can understand that position, because many of the tokens and ICOs today are structured like securities and as speculative investments, and are primarily used as a way to make fast money instead of as a way to move technology forward. And people are using them that way, and I agree that we should be regulating these types of structures.

The problem when you start to believe that all tokens should be classified as securities. Tokens like bitcoin and siacoin have important uses and advantages that go beyond speculation and cannot exist absent a blockchain token. If the everyday consumer needs to comply with securities law in order to use bitcoin, you won't get to the future where YouTube videos are being streamed over a decentralized CDN, where WSJ paywalls use decentralized payment platforms, where various ecosystems can interact seemlessly and trustlessly thanks to the power of decentralization.

I worry because more and more I am seeing posts that are skeptical of the actual utility of "the blockchain" (much like during the dotcom bubble we started seeing people who were more and more skeptical of the utility of "The Internet") when the utility does exist and with the blockchain legitimately is a world changing technology. But taking full advantage of this technology means allowing consumers to interact with it (buy, sell, trade, etc.) with as few barriers as possible, and that means that when we bring forward appropriate regulation, we need to make sure we carve out places for tokens to exist other than centralized, regulated exchanges. (though, perhaps only tokens following XYZ criteria)

Part of the problems, as I understand it, is that ICO's are different enough from traditional securities that present regulations in some important ways don't fit. I have read the SEC recognizes this and is working on new regulations.

That said, the article says that if ICO's were designed the way they are recommending, a lot of the present problems would go away on their own.

In parallel though, many entrepreneurs are staying on the sidelines because of fear of the regulatory bodies. I think the SEC should adopt something the like the proposed Arizona crypto laws.

That's definitely true. Going through this process has been extremely complicated and uncertain.

> right now most investors are getting fleeced and getting left with nothing because there is a large pool of unsophisticated investors

Fret not, this situation cannot last.

Investors are acting out of greed. They are perfectly willing victims.

Please no more regulation, or American influence/special position ("I can get my money back, because you sold me a security, and the price didn't went to the moon").

Fraudulent ICOs are written and warned about wide and far. It is easy to do due diligence. People who don't research any of their investments, need no more protection, than those who put everything on black, buy art that devaluates in value, or buy a second-hand Zune for their grand children.

Participants in any investment fraud at doing so out of greed. The whole bloody point of investment is greed. If I wasn't greedily chasing returns, I'd keep my money in my mattress.

That does not mean that fraud should be legal. Stealing from stupid people is still stealing. Stealing from greedy people is still stealing. I don't understand how this is controversial.

No I would say much investment is not about "greed". It is wise and prudent to invest in good things. A farmer invests in the seeds and time it takes to grow the crops.

But yes I agree fraud should be illegal and misleading advertisement is a bit of a fraud already, mis-representation of facts.

But informed people are hard to mislead, the question is who can you trust to give you correct information? I say that is best done by a democratically elected government.

The job of government is to protect us, not only from hostile foreign powers, but also from fraudulent information. Sometimes the two go together. Not all governments will protect us against fraudulent information, only the representative, democratically elected ones will.

The entire point of investment is that you believe that you will have returned to you more than you put in. Otherwise they'd just give the money to the organization as a donation.

People only have themselves to blame if they invest in something that is not vetted by credible parties. What you're suggesting is to make the entire market a safe space, where people are forcibly infantilized and have their contract freedom limited, to protect a small minority who don't use their freedom responsibly.

Regulation just means banning an entire category of voluntary interaction and creating a centralized gatekeeper with the power to lift that ban on a case-by-case basis if a project meets its approval.

Such an approach to dealing with complex social issues like fraud is lazy and flies in the face of a free society.

Treating tokens like securities would mean treating token sales like IPOs. Right now it costs $6 million to do an Initial Public Offering of a stock. If you want to eliminate 99% of market activity, and exclude 99% of the population from having market access, that's how you do it.

Anyway your regulations are unenforceable. Decentralized exchanges and the multi-jurisdictional nature of the market assure that. For the first time in history, you can list your financial asset on a globally accessible exchange without needing anyone's permission.

If regulations are unenforceable, what reason does anyone have to believe that anything about these tokens can be enforced?

Personal rights are enough to protect token rights, because all that's required for someone to be secure in their possession of a token is for criminals not to be able to torture them into giving up their decryption key, and practically every country in the world enforces personal rights.

The blockchain secures everything else in a censorship resistant way. Some governments may be able to go to some extreme lengths to control information flows on the internet and stop people from having unrestricted access to the blockchain, but I do not think this will be commonplace, both because there would be significant push back from most populations, and because the government would not want to create an illiberal environment that's unattractive for investment.

As for what keeps token issuers honest: the value of reputation, and the fact that their reputation is at stake. The reason high-ranked eBay sellers deliver the goods they promise is not to avoid being charged with fraud. It's to maintain their 99% positive rating. And eBay has a strong incentive to maintain an honest rating system because they want to maintain their reputation as a trusted market. The world is filled with relationships that consist of iterative interactions, and iteration creates incentives for cooperative behaviour.

ICOs are mostly useless - they are great for the founders but useless for investors. For now there is the hype that delivers new and new greater fools - but that will eventually stop and ICOs will face the hard reality of not being a very good at aligning the incentives of founders and investors. The problems are numerous - starting from the general problems of crowd-funding which can work for big companies (i.e. the traditional stock exchange way) - but is dismall for startups, to the specifics of ICOs which don't actually offer any formal claim to investors. TSOs seem to stay at the 'general problems of crowd-funding' level. There is also the theory that ICOs can enable funding of development of open protocols (think SMTP) so that the claim is on the network not on the company. I find this theory very vague.

>Utility tokens exist to provide access to a good or service on a decentralized, blockchain-based network. [...] Siacoin is specifically designed as a utility token and has never been used for fundraising.

And yet it looks and behaves exactly like a speculative vehicle, just like all the other crypto coins.

That's true, but that is not the point of the post whatsoever.

>>Counterintuitively, developing a working Y-Coin network may actually decrease YCN’s value, because it would increase YCN’s usage. This is because, in order to access network utility, Y-Coin users need to spend their YCN. This, in turn, increases YCN’s token velocity, which puts negative pressure on YCN’s price.

This is clearly wrong. Value comes from demand for the currency, which increases when there are non-speculative sources of demand, like having an amount of that currency on hand for use in payments.

Demand for the currency can also increase absent non-speculative sources of demand.

That's the problem we're highlighting. We feel that many of the dev teams today are more focused on increasing the demand by chasing speculation instead of chasing utility. By changing how the dev team makes money, you focus the dev team on chasing actual utility instead of chasing the more short term goals that can pump coin price.

I'm the author, if anyone wants to discuss!

Interesting article.

I'm not convinced transactions are so bad for utility token value. Token price is the token GDP (total value of goods exchanged for it) divided by velocity. If you double the number of payment transactions, you double the velocity but also the GDP, for a net zero effect on price.

But it's certainly true that a fee-earning token gives a stronger incentive to developers. Sia is its own network, right? Would it be fair to say that on a project implemented on top of a blockchain like Ethereum, you would dispense with the utility token and just use ETH, and apportion the fees to the "fund" holders?

I've seen a fair number of projects do this. Then people started thinking it'd be less likely to pass muster with the SEC, compared to a utility token. (But now it seems you can't avoid the SEC in any case.)

Disclaimer: Not too familiar with ICOs, and this might be ignorance on my part but genuinely interested in learning.

My understanding was that ICOs were basically a way to invest in companies/ideas smiliar to startups, and therefore reap the benefits if the project succeeds, ie upside.

Were the Siafunds given to investors as a way to reward them for investing in the company and therefore accruing in value as it grows and succeeds?

If so, is that a way to back out valuations for the company/product? (If that is the case, is it true that as more people use siacoins more revenue generated -> more profits -> value of Siafunds goes up)

I think the better way to think about it is that you are investing in the network itself (at least in Sia's case). Siafunds could be valued with an NPV of the future revenue generated by users buying storage on the Sia network.

It's much harder to figure out how to value utility tokens. Here's an interesting starting point: https://medium.com/@cburniske/cryptoasset-valuations-ac83479...

author of a decentralized cryptographic system myself (we're ranked #2 in Blockchain on GitHub: https://github.com/topics/blockchain), and I have some concerns:

How do you prevent siacoin storage owners from forking the chain to reject the siafund storage fee? They have an incentive to do so, and technologically nothing stops them.

The fee improves the health of the network, because it incentivizes siafund holders to do whatever they can to increase the amount of data being stored. So Nebulous Inc. (developer of Sia and the largest holder of siafunds) wants to make Sia scalable and easy to use; other siafund holders might develop third-party apps or simply buy ads for Sia.

Disclosure: I co-founded Nebulous.

I love what sia is doing, however I was never clear why sia wouldn't have made siacoins inflate proportional to MB*seconds available?

It seems like producing some public dataset (the blockchain so far for instance) and some proof of work that is reasonable to do once and store, but unreasonable to produce for each proof of storage, make the proof of work tied to a private key and then give each user the ability to mint new coins at some discount rate for storing the blockchain.

There are a couple constants there (the cost of the proof of work, the discount rate for storing the public info as opposed to a file contract) that need to be set correctly, but for some values the cost of the token should be pushed down to the cost of storing a certain amount of data for a given time.

This would give a coin that is pegged to roughly what you want the contracts denominated in and would encourage network utilization.

Are the constants too hard to choose (you'd probably need to modify the proof of work and discount rate)?

What are y'alls thoughts?

EDIT: sorry this is not the best description, I can try and find some of my older better write ups of the idea if it's not clear what I am trying to say.

Do you think with the first TSO's starting to come out, will we ever see attempts to remove the flaws of a TSO (accredited investors) vs crowd funding positive, at-least at some sort of happy medium, were more people can have access to vc level returns but also can understand the risk? should there be a licence process for the average person to become an accredited early stage investor.

also how do you feel about the path to TSO if you wish not to ICO, build your MVP blockchain project, grow your project attempt to become profitable/ acquire further investors then go for the TSO ?

thanks for the great article though !

How does one execute an ICO that is compliant? Can you please point to specific details or methodology?

While I cannot give legal advice, generally the only acceptable path is for a Reg-D filing with the SEC. This means that your token is considered a security, but the filing is technically an "exemption" filing.

So far I believe the only other projects that have pursued Reg-D filings did so under the SAFT framework, which is currently under scrutiny by the SEC. We may be the first project to raise money for a token that itself is a transferrable security, and pursue filing under Reg-D.

honestly it's a bit of a mess right now

Be careful regarding its transferability.

Under rule 144 you can have your original investors trade it after a lock-up period. But even with that, Reg D securities are not transferred very easily.

I had to learn all this when raising money for Qbix the last 7 years.

You MAY be able to have some sort of derivatives which are traded, like taxi medallion leases, and maybe they won't be considered securities since there is no expectation of profit (but then why do people buy them?)


Under current rules, after 1-yr lockup period investors are able to sell to other accredited investors.

Hopefully by then we'll have some new, regulated exchanges that can automate the process.

Hopefully the investors realized that their lockup may extend longer than they thought.

I am not a lawyer, but my understanding is that accredited investors do not need an exchange or a regulated entity to sell their tokens to other accredited investors. They have some compliance burden, but can, under the right circumstances, comply with that burden absent a regulated exchange or broker dealer.

I am not a lawyer, and the above paragraph may be incorrect.

Register with the SEC.

Edit: Why did this get downvoted? That is the right answer to the question!

That's an insufficient answer considering how many different flavors of registrations the SEC has; e.g. Reg A+ vs. Reg D.

Even so, that's what SAFTs tried to do, but it looks like the SEC may be concerned about their structure.

It's extremely simplistic. In some circumstances, the coin offered in an ICO might not be a security, for instance, and then it doesn't come under the SEC's jurisdiction.

They have said that pretty much every coin offered in an ICO does qualify as a security.

Every coin the chair has looked at.

The TLDR is that you just make sure that you are only selling your unreleased token to qualified investors.

A qualified investor is someone who has a million dollars in assets or something like that.

This is also what we did when designing Intercoin (intercoin.org)

In my opinion, it has just the right structure. The more communities install Intercoin's open source platform, the more ITC tokens would be worth. Thus our interest is to build the best open source software we can and literally give it away to as many organizations (colleges, cities etc.) that we can. We can even have Intercoin power other startups' business models, who would otherwise be competing with us - like Colu or Moocho.

A community might use ITC tokens to back their economy in all sorts of ways:


Raising money for an actual project by selling 100% of the tokens

And most of all - an internal currency and democratically controlling the monetary policy (to implement eg Unconditional Basic Income or loans).

In short - you need one main token network as a store of value, and sidechains for actual spending.

One coin can't be both.

With securities you’re limited to 2500 investors before you need to do a public registration.

That's true, but there's a ton of uncertainty with regards to who the "issuer" is for token sales.

And I believe it's 2000, not 2500.

I can't imagine why, unless someone is trying to be willfully ignorant.

Massive correction like we see now https://www.coingecko.com/en is a good way for the market to readjust itself and wake up to all the ico greed is good

Security token offerings (STOs) are steadily gaining popularity these days, as was predicted for 2018.

A few relevant projects if you are interested in this subject are:

* Securitize.io who assists companies in raising capital through STOs

* GBX.gi who is creating a security token exchange for vetted projects and an ecosystem to provide such vetting (based in Gibraltar, where they have regulatory support from authorities)

* Kairos, a company out of Miami who successfully did a security + utility token raise earlier this year.

That being said, valuations for these security tokens are still crypto-exuberant. I was on the phone today with a company with $1M in revenue raising $20M at $40M pre-money valuation (i.e. a 40x multiple for a company that has little to no growth rate and an unproven business model). In all honesty, even though the deal is bananas by conventional VC standards, I think they will probably raise that money very quickly.

US citizens can't register for an account on GBX.gi.

> liquid capital in the form of Bitcoin, Ether

Not sure that you're using the word "liquid" properly. Show me how you can convert $50M USD's worth of BTC or ETH into USD in three business days or less.

It's a great question. Most exchanges like Gemini have OTC desks that are highly liquid. We have a direct point of contact at Gemini, and can convert as much as we want.

Gemini also has daily auctions where you can move a ton of crypto.

There's also other well-known OTC desks like Cumberland Mining

Umm, at any major exchange?

The initial setup, of linking your identity to an account would take a long time I INITIALLY, but after coinbase knows what your identity is, has informed the IRS, and the government knows that you are not engaging in money laundering, it is quite quick.

50 million is easy, and markets trade that amount every day. Current volume is something like 400 million dollars of Bitcoin in the last 24 hours.

You can do that pretty easily on any of the major exchanges.

Can you? Someone's selling the MtGox bitcoins, $400m worth, and this halved the market price: https://www.coindesk.com/mt-gox-trustee-sells-400-million-bi...

Most volume on most exchanges is fake "wash trading". https://medium.com/@sylvainartplayribes/chasing-fake-volume-...

That's because they sold it with large market sells, stupidly. People also knew they were selling. I can't speak to OkEx, but at least two of the major top ten exchanges i'm reasonably confident have volume at least in the range of what they claim.

Yeah, but...then it isn't liquid. CEOs of big companies pre-announce stock sales all the time, and prices don't often move >5%. An announced sale of bitcoins shouldn't move the prices in a liquid market. It suggests a buyer/seller mismatch.

Announced sales of company stock absolutely move (and absolutely should move) the price of a company's stock.

>halved the market price


where are you getting halve?

From the market peak, not a 2h graph.

look at the x axis. it goes back to feb21.

This is a classic "after it, therefore because of it" fallacy. $400M over a few months is a drop in the ocean in the bitcoin world.

Bitcoin daily (!) trading volume is $6.1 billion today.

Is that money in and out of exchanges? Or day trading volume? And how can we trust that figure?

Wash trading probably

oh? Last i checked they all had $10k/day limits or less. Can you cite one that can give me $50M REAL USD in a REAL bank account in USA for that much BTC, today?

The $10k limit is generally for ACH withdrawals. Wire transfers tend to have higher limits.

Gemini has no limit on wire transfers: https://gemini24.zendesk.com/hc/en-us/articles/209113906-Wha...

Straw man.

Can you cite self service online stock trading platform that can give me $50M REAL USD in a REAL bank account in USA for that much stock, today? (No, even if one wanted to, the USD doesn't move fast enough)

Coinbase daily limits can go up to $100k or $250k I think. Most financial transactions in excess of those amounts (in Bitcoin or any other asset class) probably involve an amount of personal service and a dialog between the two parties.

Morgan Stanley. Charles Schwab. Chase.

You can't create an account with them, transfer in $50M worth of stock, and transfer out the USD all in the same day.

Frankly, I'd be fairly surprised if you could even transfer $50M into a self-serve account the same day your account is created - likely some compliance and human-intervention guardrails would slow this further than transferring several thousands of dollars worth of stock. But even so, USD transfer is going to take at least a day.

Transfers between accounts != liquidity. Liquidity is selling an asset without moving the price. If you sell $50m of AAPL, a highly liquid stock, the price may move a bit, but not a ton. If you actually sold $50m of bitcoin, you would probably have trouble getting the buyers AND the price would fall.

According to the current orderbook on GDAX, you can market sell $50M of BTC, and the price would drop to a little above $7000 (on GDAX). So, no, you wouldn't have trouble getting the buyers since they are right there on the order books, but yes, the price would fall. You'd still get your $50M.

Bitcoin is certainly much less liquid than AAPL, but thats somewhat off topic.

Person I was replying to was talking about daily transaction limits, not liquidity. My point was that the standard he proposed is not even met by the biggest online stock exchange platforms.

Very easy, just buy USDT and then Bitfinex will gladly allow you, institutional investor, a cash out ;) . /s

Yeah, definitely wrong terminology. ETH isn't liquid at all, it's gas.

ETH and BTC are highly liquid these days. You may be saying that ETH's "intent" is to serve as "gas" but irregardless the markets are highly liquid and the order books are deep.

I think they were just making a pun.

This might come across as a bit snarky but it is not intended as such at all...

Why did you post this link?

I would assume that the HN crowd would be aware of what an ICO is at this stage. Am I wrong?

I had no idea what ICO stood for, and couldn't find an explanation in the article, so I posted this to save others a click.

Applications are open for YC Summer 2019

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