This is an interesting exercise. I did suggest pump and dump type but I did not really think through the whole scenario. To summarize into steps:
- Buy/Have lots of STEEM. Wait for things to settle.
- Start 3.5 day conversion.
- Buy more STEEM. Manipulate the price upwards.
- Sell SBD?
This carries a significant amount of risk as stated, because SBD is not guaranteed to hold its value during this time. To complete the cycle you need to try to exploit the reverse conversion. Let me try...
Let's say P(t) is the STEEM price feed over time.
- Buy/Have X amount of STEEM.
- Start 3.5 day [reverse] conversion at 1 STEEM --> P USD
- Buy more STEEM to pump price.
- Now have P(t_4) * X SBDs.
- Start 3.5 day conversion back. 1 SBD --> 1/P STEEM
- Dump the STEEM to tank the price.
- Now have P(t_4) / P(t_7) * X STEEM.
The amount of STEEM you just gained is now whatever difference you were able to manipulate the price of STEEM by. The scary part here is the SBD price does not factor into this scheme at all! Even if you add a spread, as you say it may not help here.
Hmm.. I can't seem to find a flaw... it's possible that it's not that easy to pump and dump STEEM with STEEM being more mature. The motion of steps 3 and 6 may also end up in a local loss for that amount too? I'm not familiar with how pump and dumps work in general, but my impression is that you need tokens with sufficiently low circulation / trade volumes to carry out the scheme.
The scenarios we are discussing really do illustrate how easy it would be for someone with a lot of money and a little bit of thought to exploit the reverse peg idea that is currently being considered by witnesses.
Regarding your comment about it not being easy to pump and dump a mature cryptocurrency such as STEEM. I agree, and I think it would be nearly impossible to make a profit pumping and dumping STEEM by buying STEEM on the markets and then selling STEEM on the markets. However, the reason I think a pump and dump like that won't work isn't because the pump is difficult because anybody with a huge amount of money could easily pump STEEM. The pumping isn't the difficult part, and the problem with the pump and dump is that once you've pumped the price of STEEM up, then you wouldn't be able to find enough people willing to buy STEEM from you at the pumped up price. You couldn't exit your STEEM position without crashing the price. However, the reverse peg idea currently being proposed makes the dump a very solvable problem because you wouldn't have to convince a single person to buy your STEEM at the pumped up price because you could just dump your STEEM at the pumped up price on the Steem blockchain by converting it into SBD, and the only challenge you would face would be to keep the price of STEEM pumped up for a 3.5 day conversion process which shouldn't be overly difficult since most STEEM is powered up and it would take at least 7 days for 1/13th of it to hit the markets. Also, you wouldn't need to sell a penny worth of the STEEM you bought on the markets. In my crude example I didn't get into how you could cash out of SBD, but if you were able pump STEEM up to 5x or 10x and turn $100 million into 500 million SBD, then I don't think it would be too difficult to get roughly $500 million back out over time especially if the SBD is designed to be pegged at a dollar.
Oh of course... SBD being pegged would certainly make this easy to cash out over time. So that really would be enough then. I am now realizing that the reverse direction conversion really is not symmetric at all to the forward conversion because the price feed is on the steem side. Wow.
About my scenario, it's a fun cycle scenario where you can keep multiplying the steem holdings, and tanking the price is part of the scheme. So if tanking the steem takes even less time/volume, you end up with much more steem.
I feel these scenarios need to be addressed before anyone agrees to this.
I have been collecting our discussion into another post which I hope to augment soon.
Your cycle scenario provides a nice window to see the major problem with the proposed two-way SBD peg which is that conversions (even very large conversions) can be done without directly impacting the market price, and this is the feature of the two-way SBD conversions that can be exploited.
I find it interesting that your step 2 conversion is essentially selling STEEM without the sale directly impacting the markets and your step 3 is buying STEEM with impacting the markets.
I've also written a separate post related to what we've been discussing, and I think it can be helpful for people to see both your example of how to exploit the two-way SBD as well as my example.
I agree this issue needs to be addressed. We're doing our part with trying to get the word out there. It would be nice if someone could make a cartoon video explaining the problem in simple terms that is easy to see visually.