[BTC-TC] Deprived Mining Speculation (DMS)

[Deprived]

Following text added to the end of the contract (the unplanned closure section) to address the issue identified earlier of a scenario where a majority of remaining shares are inactive and so no vote can be passed.

--- New text to be added to end of contract below this line ---

In the event of a vote with a proposal to split funds being left up for 7 days and (in either vote) the total of (Yes votes + No votes) being less than 50% of all outstanding shares then the Manager will determine a fair (in his best judgement) split of funds and execute final payments in accordance with that decision.  This clause is added specifically to address the scenario where most shares have been redeemed with a large part of those remaining outstanding being unable or unwilling to participate in reaching closure.

[Deprived]

Following text inserted into the section of the contract relating to investment:

--- Inserted text starts below this line ---

Entities may be removed from the fund's white-lists by a majority vote of DMS.SELLING.

Votes may only be initiated by the Manager (at his discretion).  Investors may propose candidates for white-listing - but have no right or ability to compel a vote to be held.

--- Inserted text ends above this line --

These minor additions were made for the following reasons:

No means existed by which entities (issuers, counter-parties or assets) could be removed from our white-lists.  Whilst that was hardly serious (I can always just refuse to invest in them, expose capital to them or have exposure to them via loan collateral) it is preferable to be able to remove them completely so they gain no credibility from being on our white-list.

No means existed by which investors could compel a vote - but that left an ambiguity over whether a vote could be initiated by investors.  They now explicitly can NOT.  That is necessary to avoid the situation where someone unsuitable gains significant holdings of DMS.SELLING and then initiates a vote to have themself white-listed (though they could still not compel me to expose capital to them).

[odolvlobo]

Nice work. I feel that it is a bit complicated, but of course you have to work with the tools that you are given. I think it demonstrates the need for a futures contract based on difficulty.

How did you arrive at the initial price of 405 times dividend and the buyback price of the mining shares at 365 times daily dividend? These imply a discount rate (assuming dividends are paid forever) of 3.5% and 3.8% per difficulty period, respectively. The rates seem too low in the short term (given recent history, they should be 15%) and too high in the long term (assuming Moore's Law, they should be 1.8%).

You didn't specify when the first dividend will be paid. It affects the price but guess it doesn't really matter.

Edit: Warning -- my math could be wrong.

[Deprived]

Nice work. I feel that it is a bit complicated, but of course you have to work with the tools that you are given. I think it demonstrates the need for a futures contract based on difficulty.

How did you arrive at the initial price of 405 times dividend and the buyback price of the mining shares at 365 times daily dividend? These imply a discount rate (assuming coupons are paid forever) of 3.8% and 3.5% per difficulty period, respectively. The rates seem too low in the short term (given recent history) and too high in the long term (assuming Moore's Law).

You didn't specify when the first dividend will be paid. It affects the price but guess it doesn't really matter.

I agree with your estimate that the discount rate is too long in the short-term and too high in the long term.  That actually was one of the factors constraining the range in which buy-back had to be set.  The value was determined by trying to find a value where there could legitimately be speculation on the value of MINING.  To pick extreme examples - if buy-back was at 10 days dividend (and PURCHASE was sold at a small amount over the capital needed to guarantee Buyback initially) then the price of one PURCHASE would never cover the amount a MINING could expect if dividends were paid perpetually.  At the other extreme if buy-back was at 1000 days then there's no way it would ever be enforced - and the need to always have sufficient capital to cover a buy-back would mean that a LOT of money would be charged for PURCHASE with no realistic chance of the majority of it ever being needed.

Beyond that there's no precise formula I used to define it.   With current difficulty trends it's above what MINING could expect to earn in any reasonable period.  With a more flat difficulty progression it becomes LESS than anticipated long-term earnings.  For more informed investors/sepculators on this fund it's predicting the point at which difficulty will (relatively) stabilise that's key - and I've tried to set a buy-back that practically allows betting on that prediction (not always - but at some stages).

365 is, of course, the number of days in a year.  Having determined that in my estimation the optimal value (by which I mean one where BOTH MINING and SELLING have a change of profit) lay somewhere in th 300-500 range it seemed a nice number to pick.

The 405 is just the 365 plus a small bit extra to allow for short-term variance (spikes) - so as to ensure that in most scenarios capital can remain above the amount needed to execute a buy-back and still leave a bit left over for SELLING to get.

There are possible models for difficulty where this breaks down at some point - either ending with forced closure and all funds going to MINING (where difficulty becomes almost static for a long period - or even decreases) or with MINING dividends reducing to a trivial amount with the vast majority of capital having been returned to SELLING (where difficulty rises astronomically).  Those who believe in either of those scenarios obviously have a chance to try to make a very significant profit by trading appropriately.

Neither of those endings would be a failure for the fund.  In many ways the fund CAN'T fail (other than if I manage to lose its capital in investments).

• If one set of MINING or SELLING make a killing at the expense of the other then the fund has facilitated the mass transfer of capital from those who were wrong to those who were right.  I don't see that as a bad thing.
• If some years down the road neither side has lost or gained a lot then this may not have been great as an investment - but it will have been absolutely brilliant at demonstrating the exact true value of PMBs over a lengthy period of time.

I believe the actual outcome will lie somewhere between the two options above - probably nearer the first than the second.  At times the market price MAY represent a true value - but I believe the market isn't informed enough AND too many elements are unpredictable anyway for any sustained period during which the trading price of MNING is consistently an accurate representation of real value.

[Deprived]

You didn't specify when the first dividend will be paid. It affects the price but guess it doesn't really matter.

Missed responding to this.

First dividend will be paid at the first occurence of 16:00 GMT after the first units of MINING are distributed.  So if you hold MINING then any time 16:00 GMT arrives you'll be due a dividend.  If you hold PURCHASE then the same is true so long as SOMEONE holds a MINING.  The minor distinction is in case trading gets approved shortly before a dividend-due time (like within a few hours).  In that scenario I'd hold off doign any swaps (including my own) until after the dividend time.

Reason for that is that from first sale until first dividend, PURCHASE will be sold at a fixed price.  At the time of first dividend that price will change - and from then on all new sales will be at a slight premium to the actual value of existing units.  I therefore want to ensure some reasonable window for everyone who wants to market-make initially to get in at the same price (the very first change in price will probaly be very slightly upwards due to the 2% markup exceeding the likely amount of the dividend paid).

As soon as approval is received (and I notice) I will, of course, post indicating the amount and time of the next dividend for MINING and the initial sales price for PURCHASE.

[Deprived]

Here's some numbers to consider :

Apologies for poor formatting - I've copied and pasted from my spreadsheet here.  I'll try to remember to paste this after every difficulty change (along with other parts of it showing the calculation of NAV/U, dividends of SELLING etc - SELLING will tend to only receive a dividend after a difficulty change where difficulty rises a lot.).

Block reward   25
Difficulty   15605633
Hashes   5000000
   
Daily Dividend    0.00016113
50 days (Min Liquid)    0.00805649
100 days (Forced Close)    0.01611298
365 days (Buyback)    0.05881238
405 days (IPO)    0.06525757
400 days (Post SELLING div)    0.06445192
410 days (Pre SELLING div)    0.06606322

If we received moderator approval before the next dividend changes then:

The very first batch of PURCHASE would be sold at : 0.06525757 BTC each (with BTC-TC limiting us to 6 decimal places that would be 0.065258).  If the securities are approved for trading whilst I'm not around (e.g. in the next 8 hours or so whilst I sleep) then anyone intending to buy in at the start can place Bids at that price.  There's no need (or point) in over-bidding one another - ALL bids at that price will be filled.  And the price will remain at that until the first dividend is paid.

MINING would receive a dividend of 0.00016113 per day until difficulty next changed.

The 50 days value represents the amount of capital per share (of MINING + PURCHASE) that has to be kept liquid at all times.

The 100 days value is the capital per share (MINING + PURCHASE) which, if fallen below, triggers automatic closure of the fund with payout of all cash to MINING.

The 365 days value is the amount that would paid to MINING if the fund were closed now (after a vote from SELLING).  Paying more than that for MINING is totally stupid (paying anywhere near it isn't too bright either for that matter - unless you have good reason to believe difficulty won't rise much for quite a long time).

The 405 days value is the IPO price for PURCHASE.

The 400 and 410 days values represent the range in which capital will be targetted to stay.  If it rises above the 410 day limit then a dividend will be paid to SELLING to reduce it down to the 400 day level.  The only time this will usually happen if after a difficulty change where difficulty significantly rose (it CAN on rare occasions also occur at other points - but those will be exceedingly uncommon).

[carnitastaco]

Approved!

[Deprived]

Approved!

Yep - just got back and all 3 are approved.

DMS.PURCHASE will be placed for sale now at the price shown 2 post above:

405 days (IPO)    0.06525757

That price will be maintained until 16:00 GMT tomorrow when the first dividend to mining will be paid.

Shortly after 16:00 GMT today (just over 2 hour's time) I will send out the first units of MINING and SELLING in return for any PURCHASE That have been returned by then.  The delay before first exchanges are processed (and first dividend not being until tomorrow) is so that :

1.  There is a decent period (over a day) in which everyone can buy at the same price.
2.  The few hours of no exchange avoids one person (even me) being the only one selling MINING or SELLING (my own exchanges for my first batch will be processed along with everyone else's).

Expect the first units of MINING and SELLING to appear for sale shortly after 16:00 GMT today (in 2 hours time).  They will be at whatever price investors who converted PURCHASE choose to sell them.

[Deprived]

A FEW PREDICTIONS

As well as allowing the market to determine a price for fixed-rate hashing, the performance of this fund will also allow testing and examination of market participants in a few other ways.

Here's a few predictions I'll make now - I'm not going to explain the reasoning in detail behind these predictions (as doing so would make them less likely to be correct).  In general terms these predictions are based on my view of how many market participants assess the value and desirability of investments.

1.  I expect there to be far more trading volume in MINING than in SELLING despite an equal number of both being issued to investors.
2.  I expect MINING to trade at a significantly lower price per hash than PMBs on BTC-TC (by which I mean at a 25% or more discount - not just some small amount that could be accounted for by contract differences).  Initially this is inevitable (as PURCHASE is being sold below the price of PMBs) but I expect it to remain true in at least the medium term (months).
3.  I expect there to be a significant quantity of PURCHASE that never gets exchanged for MINING or SELLING and which is held long-term.

This post will be not be edited - so in some months time we'll see if any/all of the above were correct.

[alexius89]

The Whole project  is for me a bit confusing....perhaps because of my english knowledge...
anyway  could you tell me the dividends per day  with 1 (one)  DMS  Mining share?

Whats the difference between DMS PURCHASE,  DMS MINING & DMS SELLING......please make use of easy and basic english..;-)

[Deprived]

The Whole project  is for me a bit confusing....perhaps because of my english knowledge...
anyway  could you tell me the dividends per day  with 1 (one)  DMS  Mining share?

Whats the difference between DMS PURCHASE,  DMS MINING & DMS SELLING......please make use of easy and basic english..;-)

I can give a one line description of each - but unless you understand most of the detail it's unlikely you'll make a profit from investing/trading in them.

DMS.MINING - Behaves like a 5 Mega-Hash/Second PMB (The things people call perpetual mining bonds).
DMS.SELLING - Is a bet that people will pay too much for DMS.MINING.
DMS/PURCHASE - Is a way to get one each of MINING and SELLING at a price that will probably deliver a small profit.

The real profit comes from correctly pricing MINING and SELLING and trading with people who get it wrong.

At present difficulty MINING will pay a dividend of 0.00016113 per day.  That will change (almost cetainly downwards) when difficulty next changes.

[carnitastaco]

A FEW PREDICTIONS

As well as allowing the market to determine a price for fixed-rate hashing, the performance of this fund will also allow testing and examination of market participants in a few other ways.

Here's a few predictions I'll make now - I'm not going to explain the reasoning in detail behind these predictions (as doing so would make them less likely to be correct).  In general terms these predictions are based on my view of how many market participants assess the value and desirability of investments.

1.  I expect there to be far more trading volume in MINING than in SELLING despite an equal number of both being issued to investors.
2.  I expect MINING to trade at a significantly lower price per hash than PMBs on BTC-TC (by which I mean at a 25% or more discount - not just some small amount that could be accounted for by contract differences).  Initially this is inevitable (as PURCHASE is being sold below the price of PMBs) but I expect it to remain true in at least the medium term (months).
3.  I expect there to be a significant quantity of PURCHASE that never gets exchanged for MINING or SELLING and which is held long-term.

This post will be not be edited - so in some months time we'll see if any/all of the above were correct.

agree with #1 and #2, not sure about #3, seems wrong to me  but I'm sure you have a reason....in other news TAT VMs entire offering just sold out instantly...lets do this!!

[Deprived]

Have sent MINING and SELLING to everyone who sent PURCHASE back to me so far.

Will be checking for more transfers regularly whenever i'm online.

[Deprived]

Sorry if this sounds stupid but:
How exactly to I send the PURCHASE to you? With the Internal Asset Transfer in the Portfolio menu?

Also what is the spelling of your account? In the text it says DeprivedMining but in the Issuer row it says DEPRIVEDMINING.

Thanks

DeprivedMining is the correct spelling - I'll nudge burnside about the all-caps version as it's wrong and potentially dangerous.

Internal transfer from Portfolio menu is the way to do it, yes.

On a seperate issue do NOT expect much volume of trading until after the first dividends are paid tomorrow.  A lot of investors compare securities on dividends - and until they see one won't trade.

[Deprived]

I've caught up on transfers again.

The inconsistency in the capitalisation of issuer names has now been corrected  - thanks burnside for fixing that promptly.  I don't believe there was any actual risk (you can't create accounts with same name but different capitalisation) but it removes some confusion.

It's interesting to watch the spreads on SELLING and MINING tighten as, even with relatively few traders, competition forces it.  We're already past the point where you can BUY MINING 'hash-power' for LESS than you can SELL TAT 'hash-power' - suggesting that the market's definition of PMB value (in DMS) will be even lower than TAT had already reduced it to.

[radiumsoup]

I'm curious, as I'm not really sure how BTC-TC works behind the scenes regarding dividend scheduling - are the currently queued dividends static or based on the number of units sold as of when the dividend was queued? (There have been 1139 units sold as of this post and there are a little less than 17 hours remaining until the first dividend, if I'm not mistaken.) In other words, if you sell more units in the interim, will the currently queued dividend amount change to correlate with the number of outstanding units at the ex-dividend time?

This really is a fascinating set of securities, I'm watching very closely to see how the market acts on them.

[Deprived]

I'm curious, as I'm not really sure how BTC-TC works behind the scenes regarding dividend scheduling - are the currently queued dividends static or based on the number of units sold as of when the dividend was queued? (There have been 1139 units sold as of this post and there are a little less than 17 hours remaining until the first dividend, if I'm not mistaken.) In other words, if you sell more units in the interim, will the currently queued dividend amount change to correlate with the number of outstanding units at the ex-dividend time?

This really is a fascinating set of securities, I'm watching very closely to see how the market acts on them.

I'm also puzzled on how it works.

The dividends were scheduled (by me) based on a fixed payment per share (NOT a total amount).  It seems that, at a minimum, the amount reported in the system as going to be paid is not correctly updated when shares are sold.

From past experience I believe that it WILL pay the correct amount when the dividend is actually processed.

Unfortunately the reason I scheduled the dividends was NOT to schedule the dividends (sounds crazy I know) but mainly to try to allow people to see what the next dividend would be - and it fails at that.  So, as I actually need to be around when dividends are paid (to catch on exchanges and to adjust PURCHASE price after) I'll be cancelling the scheduled dividends and doing them manually.

The procedure for paying dividends will be as follows:

1.  Calculate what the dividend will be per MINING share (this can be done at any time after midnight - earlier if I'm confident difficulty won't change).
2.  Shortly before I pay dividends, halt trading on PURCHASE and remove any orders placed by issuer.
3.  Make sure I'm up to date on all transfers.
4.  Pay dividends to MINING and to PURCHASE.
5.  Calculate current NAV, NAV/U and whether a dividend is due to SELLING (usually will only be the case on the day after a difficulty change - and definitely won't be the case until the next difficulty change initially).
6.  Calculate new selling and buyback prices for PURCHASE.
7.  When dividend has left my account reenable trading on PURCHASE and place new Asks and Bids (Bids will only be placed if doing so would not reduce liquid capital too low - until we have significant investments I'll always be able to place Bids and probably will be able to nearly all the time.

It sounds pretty complicated but will only take a few minutes of my time (plus a wait of a few minutes between steps 6 and 7) - other than the very first time (tomorrow) when there'll be a longer delay due to me recalculating everything by hand to ensure the spreadsheet is correctly set up.

There's also a step 8 - which will occur tomorrow and probably most days - of posting some extracts from the spreadsheet here so everyone can see how the dividend, new NAV/U and new selling price were calculated.  When I do that tomorrow I'll also explain every line in the data dump.

[Deprived]

Further to my above post it appears it IS updating the amount to be paid correctly - just with a fairly significant delay.  Guess it only gets updated when the dividend-payment cron job runs.

[Deprived]

Just a quick note.

Transactions on BTC-TC are cached - meaning it can take up to about 5 minutes for transfers to show up at my end.

RIght now I KNOW someone bought 1000 PURCHASE and then transferred them to me (as they've gone from outstanding) - but until the transaction shows up for me I can't send the MINING and SELLING (as I don't know which account to send them to).

[ThickAsThieves]

I've caught up on transfers again.

The inconsistency in the capitalisation of issuer names has now been corrected  - thanks burnside for fixing that promptly.  I don't believe there was any actual risk (you can't create accounts with same name but different capitalisation) but it removes some confusion.

It's interesting to watch the spreads on SELLING and MINING tighten as, even with relatively few traders, competition forces it.  We're already past the point where you can BUY MINING 'hash-power' for LESS than you can SELL TAT 'hash-power' - suggesting that the market's definition of PMB value (in DMS) will be even lower than TAT had already reduced it to.

I'd argue that selling less than I do is playing with fire. If you compete with the costs of actual hardware you will find people buying your product rather than real mining, which will distort the potential for increased difficulty over time.

At roughly .03-.035 and below, DMS mining is a "deal" in this sense.