TL:DR
Crypto token burn/buyback models work in theory, but not been proven in crypto economics. ICONOMI had a good reason to convert from dividends to a burn model. Their model has failed. AUGUR has no good reason to convert from dividends to a burn model. So don't do it!

Introduction

I wrote this to create dialogue about the upcoming proposed changes to upcoming AUGUR V2. Currently there is a ETH dividend model in place using participation tokens which will be replaced with a buyback / burn system:

A TL;DR of v2 — The use of DAI is planned, implementation of use it or lose it for REP (in a fork), using an auction system for the REP price oracle,rep lacing participation tokens with a buyback/burn system , faster disputing, API fixes, bug fixes and more. Please keep in mind that these are all tentative changes
source

The following topics are covered in this article:

1. Previous attempts at burn model have failed
2. REP as a medium of exchange
3. A crypto token isn't a stock
4. Crypto markets don't behave rationally; a burn model will not achieved it's desired intent

The Story of ICONOMI

ICONOMI was an exciting crypto mutual fund and fiat on-ramp that had a bright future. The project made great promises, however broke several of them very early, with the most controversial feature of creating a token burn mechanism rather than a dividend model to re-distribute profits. This burn model was massively unpopular within ICONOMI, created division in the community, and has just recently failed as a business model, requiring ICONOMI to rescructure.

Iconomi originally had a dividend model, that was detailed in their whitepaper as follows:

Source

The claim was that the ICN token would represent ownership in the ICONOMI platform and would receive dividends. It is important to note, that ICONOMI is a centralized company that exists legally. By creating this model, they clearly created a 'security' token, as this model would probably not pass the 'Howey' test (see further in article for description). After the DAO project was hacked and subsequently ruled a security by the SEC, ICONOMI understandably panicked and came out with what was marketed as a 'new and better' model, based on not giving dividends, but rather burning the ICN token:

Source

While this diagram looks complete, it is important to note comments from one reddit investor who took 'disagreement with this diagram, and any positive outlook for the ICN token.'

1. Statement that Company Assets / Treasury 'can be sold / invested / distributed to the community at any time'. Sorry this is just BS. If it were true we would have kept the dividend model for ICN profits rather than the burn model.

2. Statement that Iconomi activity buys and sells the ICN token themselves when they feel it's under or overvalued. Basically they are creating a sell wall in this case and will prevent any major upside to speculation on future value of the coin. And therefore limit your investment upside. No thanks.
   I dumped my ICN yesterday, sadly. Just can't see this working out.

So the question at this point was, what was the utility of the ICN token? Things got very messy with investors. The best summary from one reddit investor best sums up the situation:

What?
You guys can't keep doing this
You can't completely change how the entire company/system is going to work, and then present it to the community as a fait accompli.
People are saying here in the comments that this is "neat" and a "cool solution" but it's not. Not at all. One reason why this is bad is that you have determined that a core component of the system which you pitched during the ICO, is unworkable. Not only have you decided that payment of dividends to token holders is unworkable, but by extension you're decided that sharing of management fees with tokenholders can't be done either. (Serious talk: If you can't pay out dividends, you can't pay out management fees either).
For an asset to have real and guaranteed market value, it needs to have a use. GNT aren't appreciating simply because people speculate that GNT will keep appreciating. Same for Ethereum. These tokens can be used or invested to return more of the same, or something of value. Bitcoin, Monero, Dash, can be used to buy goods.
Previously, the expectation was that ICN can be used to receive dividends. Nobody thinks that they will ever be able to purchase goods and services directly with ICN.
Now that you've taken dividends and management fees off the table, ICN signifies only an ownership interest, the same as any non dividend-paying stock. but worse.
In the traditional market, if I own stock that doesn't pay dividends, it doesn't have value simply because people will continue to want to buy it. First, it signifies a right to receive proceeds if the company is bought out. Or if the company is liquidated. That's not the case here with Iconomi though, because both of those events require a distribution be made to me as a shareholder, and Iconomi has decided that's unworkable.
Second, it signifies a right to vote in the governance of the company. Putting aside the fact that Iconomi has now made sweeping changes to its structure (change in Service Operator, elimination of dividends) without any votes... your announcement today also implicitly acknowledges another problem. If Iconomi is unable to distribute dividends, it is also unable to distribute voting rights. If there were a way to hold a shareholder vote, there would also be a way to make distributions to token holders.
What we're left with is a situation where ICN's only reason for appreciating in value is that Iconomi may buy and burn more of it in the future. That stinks. We need dividends.
Second reason why this is bad is that in several jurisdictions, capital gains are taxed at a higher rate than dividends. Don't get me wrong; in most places the tax rates on dividends and capital gains are the same. In a very few dividends are taxed more. But generally, the change from dividend payments to buybacks hurts people from a tax perspective (take France as an example).
edit: The more I think about this... I am pretty damned livid right now. What the actual fuck are you guys thinking? You went from a reasonable "Hold ICN in your own wallet or on platform to receive dividends" to "We're doing buybacks instead" with a one-sentence justification:
[Paying dividends] is practically impossible to realise in the distributed economy because it needs 100% collaboration with exchanges.
No. You're lying. Because "hold in wallet or on platform if your exchange won't cooperate" solves that problem. Completely. What are the true motivations behind this change? What the fuck are you guys thinking??? This is horrible.

Generally all crypto markets were in a downtrend, but may investors:

• didn't like the burn model.
• questioned the utility of the ICN token; and
• Despite tokens being burned, ICN token remained under book value

About a year later, the Iconomi burn model failed.

It's not clear if, ultimately the reason to convert the tokens to official shares stemmed from legal reasons, or a failure of the burn model bringing price parity of the ICN tokens market price to at least book value.

REP As a Medium of Exchange

Investors don't want REP tokens in order to get more REP. This is not an ideal model. Vitalik Buterin (Eth Founder) writes on his blog about Medium-of-Exchange Token Valuations. He makes the following important points:

One kind of token model that has become popular among many recent token sale projects is the "network medium of exchange token". The general pitch for this kind of token goes as follows. We, the developers, build a network, and this network allows you to do new cool stuff. This network is a sharing-economy-style system: it consists purely of a set of sellers, that provide resources within some protocol, and buyers that purchase the services, where both buyers and sellers come from the community. But the purchase and sale of things within this network must be done with the new token that we're selling, and this is why the token will have value.

He states:

Let us now look once again at the economic effect on the users. What do users lose by using an application with a built-in appcoin rather than plain old ether (or bitcoin, or USD)? The simplest way to express this is as follows: the "implicit cost" imposed by such a system on users the cost to the user of holding those coins for that period of time, instead of holding that value in the currency that they would otherwise have preferred to hold.

And Concludes:

'appcoins are very much a multi-equilibrium game.'

and

'the market cap of an appcoin depends crucially on the holding time H'

AUGUR created a white paper with an idea making decentralized oracles possible with the use of their token REP and a combination of game theory that ensures the right incentives. The AUGUR V1 White Paper states:

By owning REP, and participating in the accurate reporting on the outcomes of events, token holders are entitled to a portion of the fees on the platform. Each staked REP token entitles its holder to an equal portion of Augur's market fees. The more REP a reporter owns, and reports correctly with, the more fees they will earn for their work in keeping the platform secure.

This structure allowed investors to hold, and deal in their preferred currency ETH in terms of 'fees they will earn'. Or put another way, using Vitaliks analogy, investors will minimize their implicit cost because they are using their preferred currency (ETH).

Decentralized Tokens Versus Stocks

A decentralized token shouldn't be viewed in same context as a legal share because it has no book value. This article is must a must read on token economics.

The more utility REP can provide, the better it will be:

When evaluating a given token-based organization, the more boxes that can be ticked pertaining to the role of the token, the better it would be. The role of tokens is like nails that encroach on your business model. You want more than a single one to hold it firmly in place, and keep it defensible and sustainable.

The reasons for a token to exist in the first place is critical. Without a good reason, in theory, with open source software some other team could fork the project, remove the utility/security token from the code and simply run the project using basic ETH and smart contracts. This would remove economic 'friction'. And mathematically is inevitable unless the project really needed the token.

REP is designed to be a 'toll' for users, in the process of securing the network. It is critical that REPs market cap remains sufficient as to create the right incentives to defend against attacks on the system (detailed in the whitepaper).

Users must be able to see sufficient value in REP if they want to have the market cap appropriately priced. This means that the REP tokens utility is critical. Lets look at a traditional stock:

Common Shareholders' Six Main Rights

1. Voting Power on Major Issues.
2. Ownership in a Portion of the Company.
3. The Right to Transfer Ownership.
4. An Entitlement to Dividends.
5. Opportunity to Inspect Corporate Books and Records.
6. The Right to Sue for Wrongful Acts.

Let's look at AUGUR. AUGUR ISN'T A COMPANY

The Forecast Foundation has no role in the operation of markets, trades or actions created or performed on the Augur protocol, nor does it have the ability to censor, restrict, control, modify, change, revoke, terminate or make any changes to markets created on the Augur protocol. The Forecast Foundation has no more control over the Augur protocol than anyone else using Ethereum.
Source

If I am an investor (or user) in a capitalist system (assuming I am a rational actor), I will be seeking to maximize my profit. I will be looking to get PAID , in a currency I prefer, with the least time, risk and volatility. If I am not getting paid directly, I will want to know what is giving REP value. How can I cash out for fair market value?

Let's look at traditional Stock Rights again, but in the context of holding REP with the new BURN model:

1. Voting Power
   None. I cannot vote to liquidate my REP to ETH or USD with a guaranteed conversion or valuation to book value.

2. Ownership in a Portion of the Company
   None. There is no assets or company.

3. An Entitlement to Dividends
   None.
   As REP generates profits, and is burned, shareholders see a return in the form of increased share value as stock prices rise. REP holders get no additional ETH or REP, and rely on the irrational market to value REP at the new correct price.

Essentially the heart of my argument is that there is no company with a book value to liquidate, and there is no dividends paying cash (or ETH). Instead there is a token being burned in order to become more scarce, and be 'valued' appropriately by new buyers.

As Vitalik put it in his article:

Notice that here, the cycle is not complete, and in fact it never will be; there needs to be an ongoing stream of buyers and sellers for the token to continue having its value. The stream does not strictly speaking have to be endless; if in every round there is a chance of at least v / w that there will be a next round, then the model still works, as even though someone will eventually be cheated, the risk of any individual participant becoming that person is lower than the benefit that they get from participating. It's also totally possible that the token would depreciate in each round, with its value multiplying by some factor f where v / w < f < 1 , until it eventually reaches a price of zero, and it would still be on net in everyone's interest to participate. Hence, the model is theoretically feasible, but you can see how this model is more complex and more tenuous than the simple "developers as seller" model.

The best tokens in my opinion generate some form of return in ETH as a function of ownership. Investors want a dividend, not more of a utility coin designed to get more of same utility coin.. The best model would be to payout dividends with DAI, USDT or in whatever native fees that are collected on the system.

The REP burn model would work if there was an escape hatch button to convert REP to baseline book value using a smart contract. This would set a price floor. Investors need to be able to cash out! You can't have a circular system with a utility token that tries to build itself up over time without offering inherent utility built into the coin that is worth the desired market price.

Crypto Markets Don't Behave Rationally

While in 'theory' a burn model may work in economics for traditional securities, it is far from proven using crtpo. Joey Krug himelf stated the following on the Augur Discord:

If the supply of REP doubled today, I'd bet a lot of money that the price doesn't actually halve due to the market being incredibly inefficient / stupid.

A better argument in favor of direct fee distributions is that the crypto market is irrational and thus something with a decreasing supply it'll value less.

For instance, if the supply of rep doubled today, I'd bet a lot of money that the price doesn't actually halve due to the market being incredibly inefficient / stupid.

And if that doubling was "non circulating supply" I'd bet even more that it'd be the case.

Likewise if the supply of rep halved, I'd also bet a lot of money that the price of rep would not double. That's a far better argument than claiming buybacks don't work.

A burn model may or may not achieve it's desired intent - do we realy want to risk it?

Conclusion

At the very least I would like the Augur team to take a hard look at the proposed burn model, and not repeat mistakes from other projects. Scaling solutions will cheapen the transaction costs associated with the current model - participation tokens. A fee payout should be as simple as possible for REP investors. In addition, REP holders should not universally benefit if they do not actively participate in the system.
Having a utility token that relies on circular transactions will not inspire investors to price REP at the market cap necessary to secure the Augur network.