Analysis on treasury spending, bonding curve + impacts on DXD secondary market


Here I analyzed some basic economics of DxDAO, its treasury, spending, and bonding curve.

DxDAO is an organisation that should be viewed similar to a start-up and hence reflect this sentiment. The current bonding curve promotes too much inflation, which disincentives earlier adopters and contributors to DAO, and even holding of the asset. I propose some potential solutions along with my analysis on treasury spending and the interaction between the bonding curve and DXD secondary markets.

Overview / TDLR

  • dxDAO’s current treasury is healthy, holding ~22,000 Ether

  • dxDAO has spent 412 Ether in the last 4.5 months

  • At current holdings, the treasury is projected to have 20 years of runway at a static spending rate, and 7-8 years at 50% year-on-year growth (parabolic growth scenario). Urgent funding is not required.

  • The bonding curve raised over 10,000 Ether in August, minting approximately 20,000 DXD. It allows buyers to enter but unable to exit due to enar non-existent buyback curve.

  • DXD is strongly correlated to ETH on the downside.


Currently, these are DxDAO’s assets, of which, ETH is the main spending power as we received DMG to participate in the DMG governance / ecosystem.

The DAO currently holds nearly 22K in Ether, so what are the treasury’s ETH spendings? I grabbed some etherscan data and this is what I was able to come up with.

If we project the treasury spendings so far into a table:

Month ETH USD
May 132.534 $27,559.62
June 36.318 $10,798.09
July 196.269 $74,851.85
August 27.040 $11,475.89
September (so far) 20.800 $8,060.00
ETH USD
Total spending 412.960 113548.356

And a chart:

In total, dxDAO has expensed 412 ETH (US$ 113,500 at time of ETH spent) over the past 4.5 months which is an annualized rate of 1,033 ETH. At the current rate of spending, dxdAO has a 20 year runway. Assuming dxDAO’s spending grows by 50% per year, which is a parabolic growth in funding, we would have enough treasury to last the next 7-8 years. From what I could find, a normal tech start-up generally raises enough for a 2-3 year runway before the next funding round. Below is a chart that shows the average distribution period in which a company has raised for, from a sample data of 13,916 start ups.

By modeling a parabolic growth in spending, increasing year-on-year spending by 50%, current treasury can last us into 2028, meaning at current rate of spending, we do not have without need to worry about the lack of funding.

The calculation is strictly made in ETH and compares ETH spending with ETH treasury holdings. It also does not take into account spending other assets in our treasury as this has been negligible thus far, and also does not account for DXD holdings and vesting from the treasury. Finally, the assumption has been made that the DAO’s treasury will not grow, though obviously it can grow via revenue capture, DXD vesting, and asset price appreciation

Currently, the bonding curve alone has raised over 10,000 ETH in the month of August, and nearly 22k Ether in last 4 month, which so far has shown that raising some capital is not that difficult.

In August we also saw significantly inflation the total circulating supply of DXD by nearly 20,000, which in my opinion is way too much inflation and too much DXD shareholder dilution. Is this significant amount of inflation worth the 10000 Ether to be added to the treasury?

DXD is meant to give holders the economic output from the DAO, however if this share is inflated significantly, then early buyers will benefit less than the DAO’s actual growth, inflation not only mean suppression of DXD price, but more so dilution of revenue share in the future that will be generated. The growth of DxDAO should not be captured by the inflation in DXD.

Furthermore, I think anything that inflates significantly in supply is unattractive to hold, but DXD is meant to be an equity that is held by active participants like us.

Given that the treasury is relatively in a comfortable spot, I think proposing a potential new curve that induces less inflation, or pausing the curve, make sense.






On bonding curve and impact on DXD secondary markets

Currently, the linear bonding curve trades in ETH into the treasury, and mints (and inflates) DXD, the DXD goes to the investor.

This week, Bitcoin fell from 12 to approx 10k, and Ether fell from mid 400s to mid 300s. Meanwhile, DXD suffered a correction of ~40% against ETH, which in USD value was a high of 450$ to 150$ (data from uniswap), and currently trade at 240$.

Sure, this is a volatile market and I am not here to say we need to do things to prevent DXD from ever dropping ever ! and only go up - but if currently the back curve is very weak, which means if you invested into DXD, you can only sell on the secondary market, and because the secondary market is very thin, it is easy to drop the price. Whenever secondary markets move slightly above the price of DXD, arbitrage comes in to mint new DXD to sell on the secondary markets. (more in depth below)

Currently, the Uniswap pair has less than 200,000$ of DXDETH liquidity, while bigger purchases on the bonding curve often involve multiple hundreds or thousands of Ether. This mechanism allows DXD secondary markets to drop very easily because buyers can easily get in on the bonding curve but will suffer

For example, if a buyer buys 197 DXD on the bonding curve, his slippage would be less than 1%, however if he instantly regret it or needs to de-risk from volatile markets, or need cash or whatever other reasons. he will end up pushing down Uniswap price by 30%







Some of my thoughts:

From previous discussions, it was mentioned that:

Adjusting the bonding curve > but what kind of modifications ?
Pausing the bonding curve > but when to unpause ?

I think this post answers that the treasury is wealthy enough that DxDAO should be perfectly fine if treasury grew at a slower pace, or if it is paused. Personally I think I am in favor of making the bonding curve significantly steeper, which likely results in less treasury growth, and less DXD inflation.

Discussion

If we pause the curve completely, when do we unpause? I think if projected treasury spending can not sustain for longer than 3 or even 5 years, it would make sense to unpause. Considering how much DxDAO raised in the last 4-5 month, it seems raising is not a big problem, and pausing seems feasible.

If we make change the curve, how steep do we want it to be? and how exponential?
Just as an example,

Here is an example bonding curve but with a exponential modifier of to the power^1.001 for each new DXD minted. Again, just an example, i think a better curve can be produced, this is just for sake of visualizing a exponential bonding curve.

Ending this topic with a poll here, curious as to the results, personally, I am open to both, but if its an adjustment to the bonding curve, then a follow up discussion is needed for a better modling, in my opinion.

  • Pausing
  • Adding exponential factor
  • Fine with both
  • Against both

0 voters

References:

Treasury spendings:

DXdao Budgeting

https://alchemy.daostack.io/dao/0x519b70055af55a007110b4ff99b0ea33071c720a/history/

https://etherscan.io/address/0x519b70055af55a007110b4ff99b0ea33071c720a#internaltx

Bonding curve trade history:

https://etherscan.io/token/0xa1d65e8fb6e87b60feccbc582f7f97804b725521?a=0x0000000000000000000000000000000000000000

Treasury holdings address

https://etherscan.io/tokenholdings?a=0x519b70055af55a007110b4ff99b0ea33071c720a

Uniswap pair information:

https://uniswap.info/pair/0x1c9052e823b5f4611ef7d5fb4153995b040ccbf5

Others:

https://medium.com/journal-of-empirical-entrepreneurship/how-much-runway-should-you-target-between-financing-rounds-478b1616cfb5

https://www.svb.com/blogs/lewis-hower/startup-burn-rate-cash-flow

https://www.inc.com/magazine/201602/helaine-olen/startup-burn-rates-spend-money-to-make-money.html

https://www.investopedia.com/terms/b/burnrate.asp

https://corporatefinanceinstitute.com/resources/knowledge/modeling/burn-rate/

https://en.wikipedia.org/wiki/Burn_rate

https://www.crunchbase.com/organization/ethlend

https://www.crunchbase.com/organization/makerdao

12 Likes

This is excellent and clearly outlines the very legitimate economic headwinds for DXD. The balance of supply and demand must be adjusted given the thin secondary market and advent of Dex arbitrage. Ignoring this is naive at best and does not have the DAOs best interest in mind.

Like all problems, this doesn’t need to be solved all at once with a silver bullet. The curve can be paused until April 1, 2021 with a planned revote to pause for another 6 months if needed.

In the meantime, more permanent solutions can be debated and if a more permanent solution is agreed upon then it would be implemented and replace the pause.

Imo there isn’t much need for debate. The case for action is clear and we can decide to take incremental steps toward a more robust system.

9 Likes

I do like the idea of having some predefined period of pause happening to the curve and then move to a vote as to what modifications should be made to the slope of the curve.

+1 @CLCL on taking the time to do this analysis

6 Likes

I support making the bonding curve exponential (the “adding exponential factor” option).

Some reasonings:

a). DXD token has a continuous emission, but at a lower rate due to the higher marginal cost of obtaining a DXD from bonding curve

-> (1) investors valuing DXD can still get in through the bonding curve (especially when the price on uniswap is too high). This brings more stakeholders and potentially more contributors to the community. And I believe this is net good.

-> (2) the net benefit of having a marginal ETH in the treasury is always positive, as “you never have too much money, or a large enough war-chest” . I think the cash hoarding strategy by (high-monthly-revenue, high cash flow) tech giants provides a strong evidence.

b). Having a well-designed bonding curve, in the long-term, prevents price overshoots in market peak and trough.

This is what I understood what @CLCL meant in this graph:

at current bonding curve slope, when ETH price drops, investors wishing to exist DXD tanks the price at uniswap -> when the ETH price drops significantly in terms of USD, such that the USD DXD price from the bonding curve is lower than the USD price of DXD on uniswap -> arbs buy DXD with ETH on bonding curve and further sell DXD on uniswap, until marginal profits = 0.

I agree with him that having an exponential bonding curve solves this specific issue.

Apart from preventing the DXD USD price from going too low too fast, it is also healthy to precent the DXD USD price from going too high too fast. This could be achieved, also, with a well-designed bonding curve and the help of arbs.

3).More on the point @CLCL made about lacking exit liquidity for (large) DXD holders:

Now that the DAO treasury has a large stack of ETH, it makes sense for the DAO to provide more exit liquidity. e.g. instead of having 10% of new raised ETH going to the buy back curve reserve, we can vote to raise to ratio to 50% or even 80% temporarily so that there is enough exit liquidity for big holders who wish to adjust their DXD positions.

Even better, we could make the ratio dynamic, a function of how many ETH there are in the treasury and how many ETH there are in the buy-back reserve. e.g. R = f(ETH_{in treasury}, ETH_{in buy back}, Other relevant factors).

4 Likes

Great work @CL

Further to this, it’s important to understand inflation (by way of Linear minting of DXD) negatively impacts not only early investors, but the DAO’s underlying cashflow too.

Due to crypto’s volatility each Work Request Proposal is to be paid in a $USD amount examples here:

So with every passing month as more DXD is minted, the DXD to ETH purchasing power is technically offset by the rate of DXD being minted, which translates to work being undertaken in the future will cost more than now it does now.

What this means is a linear Bonding Curve may feel like a ‘safe’ means of fund raising, however unless the DxDao intends on a linear progression of work being undertaken to match the linear rate of Minting, a linear Curve will negatively impact the amount of future work it can afford, with respect to what it could have otherwise done with a Pause & Start means of Fund Raising.

So how do we determine when to pause and when to start fund raising?

What the DxDao needs is a Road Map of Projects its likely to undertake in the next 2-3 years with their estimated costs, this coupled with the estimated budget of running costs for business as usual will give the information required each year to allow the DxDao to make a decision on when to restart a paused Minting process again, if the DxDao doesn’t have a foreseeable need to fund raise over a given time frame, Minting for the sake of fund raising doesn’t really make sense.

5 Likes

Great analysis. The data shown by OP clearly shows that the fundraising goal of the DAO has been achieved for now. I am in favor of pausing the bonding curve now and take the time to think about how we want to address this issue long term.

3 Likes

I see how you came to this conclusion, however both the curve and uniswap are permission-less and non-kyc. What I mean to say is that the curve does not offer a cheaper way to get in, but rather an arbitrage opportunity for someone only seeking profit and who isn’t concerned with the DAO.

If you think coming in through the curve helps price discovery, well then that’s a different argument. However any rational purchaser would go to the cheapest marketplace and in theory we should expect uniswap and the curve to be priced the same (unless of
course the curve is putting artificial pressure on price).

2 Likes

I absolutely agree with you on this.

I should have explained in more detail in the original reply.
My whole line of reasoning should be:

Premise 1: when DXD price is higher on uniswap than that on the bonding curve, arbitrageurs will come in and make the two prices converge (eventually).

Premise 2: when DXD price is lower on uniswap than that on the bonding curve, arbitrages cannot take place as the buy-back price is too low.

Premise 3: when DXD price is lower on uniswap than that on the bonding curve, those who buy DXD are more likely to be holders.

Conclusion 1: arbitrages happens only when DXD uniswap price is higher than that on the bonding cuve.

Premise 4: When a large/very-large DXD buy takes place: the price increases on the bonding curve is smoother, predictable, model-able and are public information. In comparison, at least currently, the price increase on uniswap will be significant.
https://uniswap.info/pair/0x1c9052e823b5f4611ef7d5fb4153995b040ccbf5

Conclusion 2: if an uninformed whale market bought DXD on uniswap, arbitrages will make DXD on both platforms converge. (most likely to a point in between the high uniswap price and low bonding curve price).

Premise 5: Without a bonding curve, where uniswap is the main ex for DXD, a whale large buy might keep the DXD price high for a long time, if a small number of long-term DXD holders are willing to sell.

Conclusion 3 : Bonding curve price provides a more predictable and modle-able higher bound for DXD prices because of arbitrages won’t drive the bonding curve price higher than that on uniswap.

Premise 6: the splippage on the bonding curve is much lower.

Final conclusion: investors valuing DXD can still get in through the bonding curve (especially when the price on uniswap is too high).

I am aware that there are many assumptions in my reasoning. If you have very different assumptions, implicit or explicit, you can very well conclude very differnetly.

Nice analysis.
We all know the bonding curve should be modified

1 Like

A minor quibble: You seem to treat “quadratic” and “exponential”

Sorry about being slightly off professional terms, im pretty poor with professional nomenclature of mathematics, i actually hate math, haha

3 Likes

Further to this, it’s important to understand inflation (by way of Linear minting of DXD) negatively impacts not only early investors, but the DAO’s underlying cashflow too.

Thanks for pointing this out - current design of the bonding curve and DXD tokenomics is somewhat, analogous to - having a guaranteed linear price where you can purchase Microsoft stocks from, and you can simply wait until Microsoft becomes successful before you finally purchase it, and you can comfortably earn the revenue Microsoft is making, even though its the early investors time and capital that funded and build the company that is ultimately making the revenue that you are obtaining - which is quite flawed.


Also, may I start a side discussion on DAO generating revenue ?

With dHedge and confirmed to use Mesa and many other interested projects coming up, I don’t see why we can’t be obtaining revenue from it ? it seems clear to me at least, that Mesa is a very attractive product for initial listings.

I also dont think we need to ask for a significant amount of fees, like Uniswap charging 0.3% trading fees, maybe even 0.1% is a good start, that would make us charge less than even Coinbase, I don’t think the traders would mind it, at least I won’t, but I am open to discussions and suggestions.

5 Likes

I almost voted “against both” but, in the end, even though I’d rather have a trade optimizer implemented (which will in effect pause the minting of new DXD without having to officially pause the curve, see here and here) I would not be against pausing if the majority decides for it.

Very good post @CLCL with some good points. I agree that pausing or adjusting the curve is needed at this point in time. Im fine with both.

Maybe we should create a separate thread for this topic. It is a great topic to discuss.
New thread here: DXdao generating revenue discussion

2 Likes

So i asked aavegotchi / aragon about their bonding curves on how they derived it

I was told they used Bancor’s model.

Here are some relevant links.

https://fundraising.aragon.black/
https://blog.bancor.network/how-liquid-tokens-work-a4ba30f2482b
https://blog.relevant.community/bonding-curves-in-depth-intuition-parametrization-d3905a681e0a


4 Likes