[Economic Model Suggestion] Adaptive on-boarding price for stage 1


The problem
An issue which has risen is the transition from Stage 1 to Stage 2.
The issue with the current model is as follows:
During stage 1, the only way to buy XRD is thought the AMs for 1:1 XRI:XRD trade. However, while the trade is 1:1 ratio, the reserves (Pmin) are actually much lower. This gap between the buy price and the value of Pmin at the end of stage 1 results in a large risk. Investors are asked to buy at a value much higher than what the reserves can ensure them. There is a possibility that this may lead to a panic sell in which the price may drop to Pmin, the moment we transition to stage 2 and people can freely trade on the DEX.

The suggested solution
My suggestion is to implement an adaptive XRI:XRD ratio during stage 1. For instance, this can be set to be max(X, 1.1*Pmin) where X is some minimal value as Pmin starts at 0. This ensures (1) higher incentive to buy early as you may gain more from an increase of value, and (2) when we transition to stage 2, the gap between the buy value and the floor, e.g., the potential risk, is small.

How it would work
User buys FIAT tokens from the AM and converts them to XRI.
The user converts XRI to XRD based on the price of 1.1 * Pmin. The surplus is not distributed during stage 1.
As the buy price is higher than Pmin, the reserves relative to XRD grow and thus Pmin rises.
When we transition to stage 2 the maximal value any user has bought is at most 1.1 * Pmin and in case of a price drop that user is ensured a loss of no more than 10%.
The incentive to buy during stage 1 is based on an increase in value and not redistribution. Stage 2 and 3 will enjoy redistribution effects when the DEX is active.


@tesslerc, this could be a significant improvement over the current proposal. Not only does it provide much less risk for new buyers of XRD, but from what I can tell it also causes Pmin to grow faster, and the buyer’s investment to grow faster too, with fewer required new reserves each week. I’m putting together a spreadsheet to test it out. I think that this, combined with your other suggestion for matching in between Pmin and Pmax, could improve how quickly XRD becomes stable.


@steve, can you share the spreadsheet once you have it ready?


I think this proposal goes a long way to solving current incentive problems. It allows for more free market action. ATM, stage 2 might begin with the risk of a 90%+ drawdown, while the upside is 10% maximum. More frightful than that is the fact that redistribution will not occur unless pmax is hit. Since we have defined a top, we must reach a demand that is constant enough to hit that, although most all supporters will have bought in stage one. Asking people to buy a top is counter intuitive. The stage (pun intended) is set, then literally to sell off stage two. Without the incentive for upside price growth, and lack of buyers, redistribution might stop completely for several months (or years). Price will not likely hold up and stage one holders will be asked to hold with paper loses through this period of time. We should not expect them to do so, creating more downward pressure over time. The only way to profit will be seen as daytrading, or buying right above pmin and selling somewhere higher.

  • How do you calculate rM in this scenario? I believe this will lead to a much lower Pmin for a long time, when rM is calculated accordingly.

  • Also I wouldn‘t calculate 1.1xPmin but more like 1.1xMax(Pmin;y), where y could be 0.5 for example, to not go crazy with new supply without corresponding reserves, especially in the beginning.


@mario my proposal is to remove rM during Stage 1. Instead of having incentives through redistribution, Stage 1 incentives are to buy low while Pmin is rising. Stage 2 and 3 have the DEX and thus we can have better mechanisms that are based on redistribution of funds.

Stage 1 becomes a little similar to regular ICOs in which the gain is through increase of value.
Stage 2 and 3 add our economic model when the DEX is functional.

I agree with your second point. Since Pmin starts at 0 you will want a minimal starting value which isn’t 0. This helps increase Pmin faster and thus a faster path to stability.


Then a strong dillution could occur during stage 1 and incentives/ROI are lowered again. I‘m against this idea.


What do you mean by strong dillution? Can you elaborate and explain why?

Why are incentives / ROI lowered?
Let’s assume Pmin reaches 0.5 by the end of Stage 1.
You buy in at 0.01, you have an x50 ROI by the end of Stage 1. The current model will give you much much less. Redistribution has decaying gains as the price grows, this has linear gains. The ROI is actually expected to be higher as far as I can tell.


If you buy your stake during stage 1 and suddenly big money comes in, you would only get dilluted by the new investors and NOT participate at this new demand. A huge amount of Rads would be created and you would be left out with your bought Rads. Big dillution forever, and the gain? Only a slightly better Pmin grow in the beginning. You think Way-Less-Rads x HigherPmin is better than MoreRads x LowerPmin. But that could be shortsighted and only benefit the people who want to cash out early. One incentive to stay in the system would be gone from stage 1(rM), and another incentive would grow (bigger Pmin). I believe the first is better.


I think the point you missed is the ROI from value increase. Actually this model (for good or bad) results in higher ROI for early investors than the redistribution model.

Assume you buy when Pmin is 0.1. your buy price is 1.1 * Pmin = 0.11. Meaning for $1 you get ~9.1 XRD, meaning your ROI is x9.1. How much ROI is expected from the current model, depends on the rate at which you want to get to Pmin=0.9, but I can hardly believe during Stage 1 people will get more than x9 ROI.

BTW, at the expense of Pmin rising at a slower rate, you can take some of the difference between the buy price and Pmin and turn it into rM.


@tesslerc, I’m still finishing up the spreadsheet. I had one thing that I ran into while creating the spreadsheet that wasn’t clear to me from your original description. Since we are not distributing the reserves during Stage 1, how much of the XRD that gets created goes into the reserves, and how much is given to the buyer?


If nothing goes to rM and you buy at Pbuy = max(1.1 * Pmin, 0.2) then buyer gets 1 / Pbuy tokens and the reserves grow by Pbuy - Pmin.

Where 0.2 and 1.1 are parameters you can change to control the ROI and rate which Pmin grows.

On second thought, Pmin is going to grow slower during Stage 1 with this approach. But it reduces risk due to the transition to stage 2.
Also, since this should calculate on a per-coin basis it’s hard to simulate with a spreadsheet. I’ll try to cook up a Python script to calculate it properly tomorrow.


So, as an example:

Pmin = .3
Pbuy = .33

Spend $1 to buy 3.03 tokens. Increase reserves by (.03 * 3.03) = 0.09. Is that right?

I agree that this is probably not sufficient to grow the reserves quickly. I was thinking that for each token purchased, another could be created that goes into the reserves. I’ll do some experimentation.


Ah sorry, reserves grow by $1 in this case. My mistake.
If we buy for $1 then the delta, i.e., (Pbuy - Pmin) * $1 = 0.03 is the surplus which increases Pmin. This is the “extra” above the base “value” of the token.


I’ve created a spreadsheet to try to reproduce the adaptive stage 1. Unfortunately, it didn’t do very well at increasing the Pmin during stage 1. I think that it’s just not able to give enough to the reserves relative to how quickly the M(t) grows.

Perhaps I didn’t implement the spreadsheet correctly. Feel free to play around with the numbers to see what you come up with.

I’ve also experimented with using a “Pmin + 0.2” method, rather than multiplying by 1.1. This gives better results at increasing Pmin, but does carry a higher risk for investors. I’ll post that spreadsheet once I’ve completed it.


Yes, that makes sense.
Since buying at Pmax introduces high risk, it also increases Pmin faster as the difference Pbuy - Pmin is much higher. Even if a fraction goes to increased supply.