Idea: on the way to "good money" - a crypto coin ETF based on PeerAssets

[size=14pt]Introduction[/size]

What I’m going to propose is neither new, nor revolutionary.
It just tries to recombine some things we already have.
And it might serve a need of people.
Plus it’s a step into the direction of offering “good money”.

Good money could be used for all those uses equally good

[ul][li]unit of account[/li]
[li]cash[/li]
[li]reserve for future payments[/li]
[li]deferred payments[/li][/ul]

But we don’t find good money in the world.
As all major international currencies are issued by central banks.
They serve as unit of account and for cash purchases, but they are not reliable and suffer from inflation, hyperinflation or monetary reforms.
They fail to be used for holding reserves for future payments and for deferred payments, because of the unreliable, uncalculable inflation rate.

To make a token good as reserve for future payments or for deferred payments, you need to base the value of the tokens on commodities, raw materials, food, industrial products (finished and unfinished) and alike; things you or others want to buy with the tokens - now or in the future.
If the tokens keep the value stable measured in that basket of commodities and goods, they are stable and not prone to effects of inflation caused by central banks.
If the token transactions are processed on a blockchain they can have the benefits of crypto currencies while being more treasured than fiat money.
Some might perceive it as drawback that these tokens aren’t stable in relation to fiat money. For others the reason for that is the reason to prefer them to fiat money.

At traditional exchanges there’s a lot of trading regarding commodities, raw materials, food, industrial products including derivatives. It’s easy to find a price index. It would be easy to base a token on a combined index. The fiat money price of that token would be floating. The composition in terms of commodities, raw materials, food, industrial products would change together with the change of the individual relative value.
But the token would at each time allow you to cash out the value of a% of commodityThis, b% of commodityThat, c% of raw materialWhatever, d% of… etc. (keep in mind that a, b, c, etc, can change over time!), which would lead to stable buying power.

[size=14pt]Why making the efforts?[/size]

One reason to make the efforts is to make the world a better place.
This alone is a good reason. But I see a chance to make money from it by making buyers/customers happy with it.

How does the issuer make money, if the tokens are meant to be stable in relation to the index? And how does it make buyers happy?
In short: the tokens are sold for an offset and bought back at face value. They save the buyers from mirroring an index on their own. Nothing new. Just another ETF, but…

Longread: the issuer requests an offset when selling the tokens of quite a few percent. 10% or way more doesn’t look unrealistic, especially not if you account for the real inflation rate of e.g. USD. I’m aware that the Fed proclaims a low inflation rate, but depending on what you look at, USD had a considerable inflation rate during the last 8 years. You save the customers creating a basket of all that commodities and goods, provide them with a token that’s fungible and can be transferred in small units, save them from the exchange fees they have to pay when buying e.g. exchange traded funds (ETF) and offer a token that represents all those commodities and goods.
I’m favoring ETF, because that would save from some trouble dealing with way too many individual products/index funds.
From that offset the operational costs are paid and on top of that revenue can be made.
In difference to a standard ETF the tokens would “owned on a blockchain” and could be transferred just like crypto currency units. You would be on the way making “good money” by creating such an index.

[size=14pt]Would that it were so simple[/size]

The devil is in the details.
As the tokens represent a basket of commodities, you need to buy those ETFs when a customer buys tokens. After all you need to back (ideally 100%) of the tokens by that basket to make the token a reliable and trustworthy product.
That creates two problems:

[ul][li]you have to pay the exchange fees when buying the ETF[/li]
[li]you need to keep the ETF[/li][/ul]

The first problem costs money and might require minimum purchase/sell amounts of tokens (not very convenient).
The second problem can lead to confiscated ETFs in case a government might not want such a business.

[size=14pt]Damn, so close - what now?[/size]

A solution that is less than ideal, but maybe a step into the right direction is right in the field where this token shall be placed: crypto currencies.
Instead of ETFs that stand for commodities, goods, food you make a kind of crypto index. You issue tokens that represent onwership of (just as an example) a% BTC, b% LTC, c% ETH, d% XMR, etc. - you create a crypto coin index.
The tokens in the index need to support multisignature transactions and most have a high trading volume at a tight spread, otherwise you need to limit the amount/value of tokens that can be bought per time to adjust the price.
If a customer buys a token, the proceeds are invested in the crypto coins listed in the index.
The crypto coins are being bought at an exchange and deposited in publicly known multisignature addresses. The sale offset is accounted separately and operational costs are being paid from it. If there’s money in excess, the revenue can be invested in development or distributed as dividend, e.g. as tokens (after the index crypto coins have been bough, of course)
That way customers can always verify that the tokens are backed by the crypto coins they represent.
Just as with creating a basket of commodities/goods, it saves the customers from putting all together on their own.
The easiest way for making it on the own would be to buy a lot of different crypto currencies and manage at least all the private keys for them, maybe even wallets, clients and blockchains. The alternative would be to leave them at exchanges, but that’s not exactly one of the most safe places.
While this crypto coin ETF is not exactly “good money”, it’s a chance for people to participate in the (price) development of crypto currencies without the need to deal with all of them.
They can buy and sell ETF tokens for BTC. If they buy from the issuer, there’s a buy offset. If they buy on the market, the spread will be somewhat smaller.

[size=14pt]How to do it - in a nutshell[/size]

PeerAssets seem to be a platform on which that could be implemented.

I think you’d need (at least) two PeerAssets:

[ul][li]one that represents the issuer (at the beginning distributed across the board of directors, who can use them for voting)[/li]
[li]one that represents the fund tokens (which can be bought/sold for BTC)[/li]
[li]and the third I can think of are tokens (corporate shares) that entitle you to receive dividends, but that have no voting rights; they are optional and can come later[/li][/ul]

I already have spent some time thinking about the implementation, but like to postpone writing more about it than I already did.
Before I want to know whether I’m completely wrong with my idea to base this on PeerAssets.
The PeerAssets white paper says e.g.
“PeerAssets enables easy querying of the blockchain for relevant transactions and offers some extra features like shareholder voting and dividends payouts.”
And the deck, deck spawning, cards and card transfers appear sufficient to do the job for all the two (three) PeerAssets/decks I mentioned above.

What are your thoughts?

Interesting idea, but not sure if it is viable. The cost of buying and selling the underlying coins (exchange fees) and the cost of development (developers) and maintaining the trading (traders) is probably requiring such a high fee that it won’t be interesting to buy the coin in the first place. I doubt whether anyone would pay a premium of say 10% assuming that is adequate.

I’m not sure either :slight_smile:
As any new business this would not have a good chance to be profitable from the start.
Money would need to be spent on development. Operational costs would be there. That’s why there’d be a board of directors, who own the corporation that issues the tokens. These tokens are the compensation for paying the development and operational costs.

To make the tokens attractive for buyers, the sale offset at the beginning would need to be as low as possible. I wouldn’t advocate treating that as advertisement and applying no uplift, but 2% offset should be enough to compensate the exchange fees and the spread.

The 10% is just a guess. As soon as there are tokens and some volume, you can get a feeling for what people are willing to pay by looking at the exchange rate of the tokens. I suppose to see the exchange rate to be soon be above the face value of the tokens. You can orient the sale offset toward the market rate; of couse being low limited at like “face value * 1.02”.
Customers would pay an offset to receive a service.

A sale offset way beyond 2 or 3 percent is of course not attractive for people who want to spend a lot of money on a portfolio of crypto coins. They are better off creating their own portfolio by buying all the crypto coins they want.
But tell me: would you rather spend $200 for a portfolio of 20 different crypto coins, making all the buys, handling 20 different crypto coins (keys/wallets, blockchains), or would buying an ETF representing them for $220 look rather attractive?
And if they do their math right, they are aware, that they have to deal with exchange fees and spread as well. The first few percent offset are merely compensating exchange fees and spreads.

You don’t even want to deal with big amounts of money from the start, because that creates some sort of trouble. So those, who would complain more about 10% offset, because 10% of $50,000 are a lot of money, aren’t your target group. If the amounts are too big, you will run into trouble from having an increased spread at exchanges, if you want to buy in a short period of time and from even moving the market. This is eased to some degree depending on the croypto coins (and their ratios) you have in the index, but still a problem.

I’m not talking about a single ETF here. I’m talking about the concept.
You can of course make different ETFs with different risk and volatility attributes; top dogs, penny stocks, emergent coins etc.

I believe this can be done properly in the form of a DAO, using PeerAssets, that provides a proper incentive structure for maintaining the code base responsible for automating the exchange. You make a valid point when you say,

In terms of managing the trading I believe the only inherent cost is developers as this should remain automated . The exchange fees should be pre-calculated based on which exchanges are being utilized by the DAO. This is an expected cost that should be forwarded to the customer as they would encounter it regardless of using the service provided by the DAO.
In order to make this service profitable the DAO can decide on a markup that is competitive.

Hi, @ConfusedObserver

I had similar idea, though it was more like a hedge fund for cryptocurrencies. PeerAssets is perfect tool to implement something like this.

If you are willing to get started with some sort of paper explaining what is to be done and how should it be done. Things can get moving for this project even with first beta releases of PeerAssets protocol as the thing is not complicated.

The way I see it the most complicated thing about it is selecting the basket. How to filter out the noise of random pumps?
For example Monero now. It was a good buy at 300k satoshis and DAO would profit from including it into the ETF, but buying Monero above 1M satoshis is very risky move as Monero will probably be worth 300k satoshis again in few months.

Thank you for your feedback. It’s good to know that I’m not way off.

If I could filter out random pumps, I were more clever than the market, which I’m unfortunately not.
If you have a basket with a sufficiently big number of assets, you can hope for always buying some low and others high.
As most crypto currencies are more or less tied to BTC, they’ll fluctuate with BTC as well. Even a well-balanced ETF won’t be perfectly stable, but more stable than single crypto currencies if you measure the value in BTC.
And as long as you have BTC in the basket, you’ll even out the effects of BTC surges or drops, because normally most other crypto currencies behave anti-cyclical under that condition whereas they tend to keep their BTC ratio, if BTC makes only small steps (ramdom pump&dumps excluded).

The only requirement that needs to be met for a rather small sale offset is a big trade volume at a tight spread over a given period.
I’d like to target periods of 6 to 12 months. Having that volume spread across different exchanges is also important.
Monero for example fails to meet the “trade volume over period x” requirement, if x is 6 or 12 months. Just as an example, Dash would be a better candidate.
The less (reliable) the trade volume, the bigger the sale offset needs to be to include the risk of having assets with a reduced trade volume in the ETF.

Like I already said: I imagine different ETFs. The ETF designed to be the most stable product that’s being issued, has different requirements than say the one with emergent crypto coins. In that one you need to include coins that can’t have a trading history, because they’re too new.

You need a transparent prospectus for each ETF to allow customers assessing the composition, the strengths and weaknesses, opportunities and threats.
May they pick the right one; some will fluctuate much more than others.

At the moment I still have one challenge and one obstacle that stop me from writing a paper.

The challenge is the trading. I need to mull over this some more.
You will have a hard time getting such an ETF listed at an exchange. Which means you need to start with trading the tokens off exchange and tie that trading with the trading on exchanges.
Automating the buys/selles of the ETF assets should’t be much of an issue for a smart programmer. Having funds at the exchange is required to provide smooth operation. I have a proposal for how to arrange the trading to make it safe (from an economical point of view, not a technical).
When selling tokens off exchange you need to deal with a limit beyond which the price can’t be guaranteed, because you need time to transfer additional BTC to the exchange(s). Just like when exceeding certain limits at shapeshift (which in that case is rather associated with an increased average offset when trading big volumes). But there’s a lot of moving parts and I prefer simple solutions. I haven’t found one so far.

The obstacle is the question: how to provide an incentive for the board of directors to behave?
The board of directors is (in my design) holding keys for multi signature addresses in which the majority of the assets are stored.
Only a fraction that’s meeting the trade volume is on the exchanges (that fraction doesn’t worry me, I have a solution for that).
But the corporation (if successful) will soon deal with assets of a value that dwarfs the value of the corporation.
How will you stop the board members from colluding and stealing the funds?
Requiring collateral from the board members is no solution, because that will only postpone the point of time when the assets are more valuable than the corporate value plus the provided collateral.
At least if the corporation is successful and deals with six-fugure sums or more.

[quote=“ConfusedObserver, post:6, topic:4068”][…]
The obstacle is the question: how to provide an incentive for the board of directors to behave?
[…]
How will you stop the board members from colluding and stealing the funds?
[…][/quote]

In the original concept of “The DAO” the “Curators” were there to prevent such kind of attacks: https://blog.slock.it/on-contractors-and-curators-2fb9238b2553#.2757sg7vg

Tho I myself can not say if it is a proper solution.

So first of all, let’s pretend we’re making a PeerAsset that is pegged to the PPC price. You would form the issuing multisig, then have people deposit PPC into the issuing address. For every PPC deposited, the issuers would issue 1 PseudoPPC. Now, have the multisig run a script where every PseudoPPC sent to x address would be burned and 0.9 PPC sent to the redeemer of the PseudoPPC.

Now, going to 50/50% split PPC and BTC, have the multisig account go buy BTC with half the PPC. Upon redemption (45%PPC/45%BTC/10%profit), the multisig account can just slightly modify the script to send other currencies too. The biggest problem here is pretty straightforward: How does the multisig know which BTC address to send to? The most native solution I can think of is to use the trick where you can generate a BTC address from a PPC private key, and send the coin to that address.

For asset creation, I guess you could just say, “If the same private key has sent both BTC and PPC to the two different issuer addresses, then make assets for them. Otherwise, do nothing.” Then, making a new asset requires you to put in coins for each type in the ETF. However, most people will just buy theirs on the open market from somebody else.

Allow me to explain the exchange part a bit more. I think there’s a misunderstanding.

At the beginning of the business there’s an exchange run on a corporate website, because the ETFs won’t be listed on popular exchanges. I’ll call it “corporate exchange” hereafter.
That corporate exchange won’t deal with the grade of multisig addresses of the cold-storage multisig. 2 of 3 multisig might be considered for security reasons, but not the full charge.

How does that corporate exchange website work?
People deposit BTC, trade them for tokens and withdraw the tokens - or vice versa. On the surface it’s as simple as that.

As soon as the tokens get listed on exchanges, it should be easier to gain traction.
But it won’t be easy to get the tokens listed there. It’s a bit more complicated for exchanges than with native tokens on a blockchain. But I know that tokens based on Omni are traded on exchanges. So it’s not entirely impossible.

I haven’t spent much time on investigating PeerKeeper. How far is PeerKeeper from being a part of that corporate exchange?

[quote=“ConfusedObserver, post:9, topic:4068”]Allow me to explain the exchange part a bit more. I think there’s a misunderstanding.

At the beginning of the business there’s an exchange run on a corporate website, because the ETFs won’t be listed on popular exchanges. I’ll call it “corporate exchange” hereafter.
That corporate exchange won’t deal with the grade of multisig addresses of the cold-storage multisig. 2 of 3 multisig might be considered for security reasons, but not the full charge.

How does that corporate exchange website work?
People deposit BTC, trade them for tokens and withdraw the tokens - or vice versa. On the surface it’s as simple as that.

As soon as the tokens get listed on exchanges, it should be easier to gain traction.
But it won’t be easy to get the tokens listed there. It’s a bit more complicated for exchanges than with native tokens on a blockchain. But I know that tokens based on Omni are traded on exchanges. So it’s not entirely impossible.

I haven’t spent much time on investigating PeerKeeper. How far is PeerKeeper from being a part of that corporate exchange?[/quote]

It will be pretty easy to integrate PeerAssets based tokens on any exchange.

Sorry to come in so late. I have been interested in cryptoETFs for a long time.

My comments specific to the proposal are based on Quite good money ETF · GitHub which might be different from what have been posted above.


First the tone and pitch of the proposal might need a shift. For example

While this crypto coin ETF is not exactly "good money", it's a chance for people to participate in the (price) development of crypto currencies without the need to deal with all of them.

There is no need to be appologetic for not being a “good money” because no one else has made one.

If you try to sell “a chance for people to participate in the (price) development of crypto currencies” nobody will be interested. Instead try to tell its a chance for people to diversify investment and participate in the crypto technology revolution by “Just take my money”. More below.


Second, why is the ETF denominated in BTC? BTC is highly volatile against fiats and most people in the world earns fiat salary. Why not price the fund in USD and accept BTC as payment for purchase? As I show below the fund could be more stable than BTC.


On component assets, my idea is to diversify customers fund into an fund consisting of cryptocoins and their derived assets in three categories: the top N (marketcap on CMC) POW coins, top N POS coins, and top N strong-anonymous coins that have been issued for more than 1 year, and have multisig feature that enables a decentrialized reserve.

The choice of categories is to reduce miner-user interest misalignment risk in POW and authority clamp down risk. The number N is for diverification within a category. The categories can be adjusted as severe risks are identifies.


I think multisig signers shouldn;t be confused with directors. The signers only need to be reliable and could just have minimal technical knowledge to release the fund. The directors need to have knowledge, experience, and judgement in solving business and tech issues.

I think that dividends from profit paid to shareholders awarded to directors should be appropriate incentive.


I have also though about it for some time. The following is how this should be tackled in a two step approach,

  1. Create an index
  2. Make the ETF track the index

My idea is that the fund will track a what I call the Minimal Variance Index. Here after I refer to this index fund as the Fund.


The Minimal Variance Index (MVI)

Suppose there are M assets in the index. The index value (in “point”) is calculated by adding component contribution of every asset. The fraction, or weight, of contribution by the i-th asset in the index is Wi.

Wi = 1/Vi / sum_over_i(1/Vi) ,

where Vi is the price volatility of the price of the i-th asset’s USD price (Pi) in the last e.g. 6 months, normalized by the average of Pi.

Since Vi is a measure of volatility, the above ensures that more volatile assets get less weight in the index.

Vi is chosen to be a function of Pi (the exact mathematical definition needs to be studied) so that the index itself will have minimal fluctuation relative to its average value. Hence the name Minimal Variance.

The index value = sum_over_i (Pi * Wi * some-factor) , for example.

The performance of the index should be tested with historical data, given the set of component assets.

Making and provding index feeds can be a business by itself. Dow Jones Index makes 130,000 indices.

Make the Fund

The Fund can track the MVI by owning all component assets. Synthetic indexing using futures is not discussed here.

The unit of the Fund token is set such that each fund unit is worth one index point.

Buying and selling are made to balance among the assets so that their relative contribution to Net Asset Value (NAV) of the Fund equal to their respective weights (a process called Rebalance).

The weight of every component asset changes. Updating of the weights of all assets in the fund according to the current index can be half yearly or quarterly if the process (e.g. component picking ) is costly, or can be daily if the process is largely automated.

The timing of rebalancing and weight updating can be periodical or with randomness to discourage index front running.

The Fund NAV is calculated daily, if not in realtime. The value of a fund Unit is calculated by dividing the NAV by Fund unit in circulation.

A customer buys the Fund by buying the crypto Fund token from the Fund organization using bitcoin , for example. The sales proceeds, minus fees, are patitioned according to the current weights and used to purchase respective assets immediately. A customer can also buy the Fund with a basket of assets, and paying a better fee. All assets are put into multisig addresses, and become part of the Fund Reserve, save for some liquidity fund.

The Fund token owner can always redeem and get back a basket of coins according to the current weights, or get btc from selling these coins by the Fund. The Fund is backed by a full decentralized reserve.

On the secondary market, the token can be sold/bought according to supply/demand.

The Fund is managed by managers, and directed and paid by Fund shares owners through consensus and voting. The Fund tokens are generated on blockchain from grant voting.

The reserve signers sign to release funds according to the Fund status calculator, which feeds the Fund website to publish current status of the Fund. The signing can be automated. The signers are just holders of keys.

The Fund front desk is an instance of bots that takes buyers fund, converts to the basket of assets, calculates fees (revenue plus e.g. 2%), calculates Fund tokens from the basket of assets yielded (so the buyer pays the exchange cost), issues Fund tokens to the buyer, and sends the assets to the Reserve. In principle electronic money in any form can be accepted as payment as long as it can be converted to cryptocoins, as the payment funds don’t need to be held by the Fund. The bot also takes the Fund tokens from a customer can sells assets according to the currrent weights and pay the customer in bitcoin, or if wanted, in a basket of assets.

The sales fee minus cost of operations is profit. When accumulated profit is more than d% of full Reserve , the excess will be issued to Fund shares holders as dividends. The Fund shares owners have an interest in the health and performance of the Fund.


Value to the customers

  • Exposure to the best cryptocoins and assets without buying all of them.
  • MVI dynmically assigns greater weights to the more stable coins in time, averages volatility out in many assets.
  • Rebalancing sells high and buys low, and is proven to increase customers’ value in volatile market.
  • Index fund saves cost and is well recognized to perform better than hand-picking assets (even by experts).

Value to the investors

The Fund can start small. Most work can be tried by human initially to gain operational knowledge. Automation costs but can be build gradually. Once the in frastructure is made the long-term operation is almost automated. The cost of managers, directors, signers, server renting, and tech support during operations is limited. The fee collected is unlimited.

More products can be added with little cost to cater different customer need.

Brilliant post @mhps.

I am glad that you are interested in doing this as well. I have spent some time trying to figure out how to organize this specific DAC as well.
This looks like an market niche worth filling, but we should be quick. Competition is not sleeping.
I hope plans for this DAC can be completed before PeerAssets is complete so we can beta-test PeerAssets and start a marketing campaign for this in parallel.

Where this project idea is now has quite diverged from what OP proposed, it became more serious I would say. I think you can can help us to accumulate all the different ideas into one sane one.

You can join discussion about this here: https://peercoin.chat/channel/quitegoodmoney-etf

I am glad you find the ideas useful. I wonder why no one else has already done it.

I have problem connecting to the chat. I have also general problems against chatting for serious work as information density is low compared with forums and there ofthen is no archive – even there is it is hard to find specific information. I know there are projects developed mostly using slack.

What problem are you having? Is there an error? What does it say? Are you using the Mobile app or are you opening it using mobile browser? With the app you need to enter in the URL in a specific way or it won’t connect.

Also, I understand what you mean about it being easier to track discussions with the forum. Luckily, discourse is almost ready I think. We’re in the stage of trying to figure out which server to use.

True, but it’s not good money according to Hayek’s prospect.
I’m not the marketing type of guy and I always like to have the chance to under promise and over deliver.

Because it shall show that the QGM ETF is stable or appreciating in its BTC value.
If you price in USD it’s fluctuating more than when priced in BTC, albeit less than BTC/USD - in my theory :wink:
When pricing it in USD, you should always provide charts which show BTC/USD as well.

Of course. During start the organization will need to be slim. Combining the roles of multisig signers and in personal union makes sense.

Fully agree!