2) BitAssets will now receive a predictable, hardcoded 5% interest return, and short positions will pay a 5% borrowing cost.
There has been some discussion in the BitShares Status Update thread but I think this topic deserves its own thread.
After thinking about this for some time I wanted to share some thoughts. My question is:Why do BitShares need interest payed by the shorts to the longs?
Without it the BitAsset would track the real-life asset much better.
Essentially, what you are creating with this interest/dividend is an bond with a 5 % coupon and infinite maturity. The price of this asset will be highly susceptible to market interest rates changes
. In theory causing the price of the asset to change the same magnitude as the interest rate: for example a rise from 3 % p.a. fo 4 % p.a. (4/3 = a 33 % rise) should lead to a drop in asset value by 25 % (to 3/4). Compare this volatility to a 30 year (the longest maturity available) US treasury bond.
The price of the BitAsset will be (much) higher than the real-life asset
where the coupon (5 % assumed) is much higher than the market interest rate. For BitUSD with a US treasury bond yield of 3.9 % p.a. this effect might be negligible (the NPV calculation quoted in the BitShares white paper states 114 %*, I arrive at 128 %). For assets which normally don't fetch an interest (Gold, JPY, etc.) this will be a huge difference: A 100 year 5 % coupon JPY bond in an environment where you normally get 0.8 % p.a. would trade near 400 %, a 30 year bond just above 200 %.
In BitShares there is no need to charge the shorts a penalty. In real life you have to pay for shorting because you need a counter-party. Being long in any asset is quite a natural thing and you can do that on your own. Being short requires somebody to lend you the asset first before you to then sell it. This loan is open as long as your short position is open. In Bitshares being long and being short seem to be equivalent
as both need a counter-party, so why disincentivise shorts? Would this not choke the supply?
If you want to fix these issues then at least every BitAsset will need its own interest rate. But my gut feeling is, the lower the rate the better, zero best.* Is that calculation available anywhere? The paper only states the result.