Two things immediately spring to mind:
1. If it's valued in euro then the face value has to be in euro. You can't sell all 100k at same BTC price as if they don't sell quickly and the exchange-rate moves it would be impossible for you to honor the contract to investors who bought at different exchange-rates.
You misunderstand; the face value will always be 0.01 BTC. Originally it was my intention to cap any value increase at 50% but having modeled the situation it became apparent that since we intend to invest it in SHA256 mining hardware we will be generating more anyway; the bond will be repaid by the yield from those machines.
If difficulty is rising too fast then we will take the option to repay the bonds early. Even in my worst case scenarios we can repay the bond early. However, paying interest linked to might BTC would become too risky, hence the fixed-fiat yield.
However, you also assume that BTC will rise. Personally I think there is a fair chance it will crash again soon, in which case the bond holders would be in a better position in terms of yield than they would be if it were BTC-linked. Personally I see this feature as quite a nice bonus for the investor - their returns are stable regardless of BTC's price.
2. You can't issue a bond which is only secured against the assets purchased with capital raised from the bonds:
a) That would necessitate you keeping seperate accounts in respect of the bonds - with all purchases recorded as being purchased with bond-capital or purchased with other capital.
I fail to see how this is any different to a hire purchase agreement or other secured loan? At Memset we regularly take out loans against rapidly-depreciating hardware? We are more than capable of the necessary accounts administration to record which assets belong to the bondholders. Again, we do this at Memset and I shall with Wood Tech (the legal entity which will own the tangible assets).
b) It's grossly unfair to purchasers of the bonds. They take all losses associated with the bonds but have profits capped at 22% growth expressed in euros.
This is not how the principal investor we have lined up views it. There should be no losses associated with the bond. CipherMine will take all necessary measures to ensure that it is repaid, as any company would in such a situation. The ultimate back-stop would be to sell the assets themselves, though by then even in a dire scenario we should have been able to repay most if not all the face value from the mining yields. That they are secured by the assets purchased should only ever come into play if something went disastrously wrong (ie. CipherMine had completely exhausted its funds attempting to honor the loan).
Further, the first tranche of this bond will actually go towards paying me back for eight HashFast BabyJets I've purchased with my personal overdraft, so the bond holders and CipherMine shareholders have the additional surety that we'll be pumping out the cash necessary for repayments within two months (20th-30th Oct).
Bonds are issued by the company and secured against ALL its assets. Bond-holders have senior claim over equity holders.
Not necessarily. As mentioned I would compare this to a hire purchase agreement, but such vehicles do not fit neatly into BTC TC's categories. Burnside specified that the only difference between a BOND and a LOAN is that the latter is issued by an individual. This could therefore be considered secured loan, and one one at an extremely high interest rate at that!
EDIT: It occurs to me that when you say "based on EUR/BTC value at time of issue" you may mean at time dividend is paid rather than at time the bonds were issued.
No, I meant at time of issue. In other words, if I issue the first 25,000 bonds next Monday the price dividend price would be set at that point. I cannot keep track of exactly when they get purchased, but I've realised my mistake now which you describe here:
- An interest rate of ~22%/year (0.38%/week) based on EUR/BTC value at time of issue (eg. ~€0.0032 / bond / week).
- Bonds to be issued in tranches at discretion of CipherMine management; they shall only be issued if the management is satisfied that they can be repaid and that there is a useful way to spend them.
To clarify my point 2 above, the clash is between the two items quoted above.
If you sell the bonds in tranches then there'll be a different exchange-rate when each tranche is sold. You then can't honour the contract in respect of all tranches as there's no way to pay different amounts of interest to different tranches of bonds. Unless you wanted to pay 22% of of face value at the most beneficial (to bondholders) exchange-rate at which you'd ever sold bonds. That pretty much negates the whole point (for you) if exchange-rate moves significantly in EITHER direction between the placement of tranches (as, depending on direction of the move, you'd either have to pay new bondholders a far higher rate than you wanted or upgrade heavily the rate paid to existing ones).
I now see that I cannot differentiate between bonds in terms of issue either, a silly oversight. I could adjust the dividend yield as more bonds are issued, making it the average, but that would be complex and confusing for the investors. Alternatively it might be more sensible to issue the lot all at once and then just keep the BTC that we didn't need and take the hit on a slightly higher-than-necessary interest payment while we hold the funds. We can certainly afford to do this with minimal risk (ie. if we don't want to spend it all at once, just hold the BTC and pay the EUR interest).
This second approach would also support keeping the interest at a fiat-fixed rate. Editing the security now...
In that case my point 1 doesn't apply - though it then makes no sense as the interest would be based on BTC face value anyway (just with two currency conversions to get back to where you started from).
See above re. both points.
Kate.
EDIT: I've updated the security listing to address some of the issues raised.