Quote from: gog1 on November 16, 2014, 05:11:33 PM

2586, much appreciate people giving thoughts to the issue - but I see this as a "big" change. The (x=0, y=0) => lend at highest available swap demands will kill all the swap demand; I wouldn't be shocked if a lot of people leave it or make it x=0 , y =0.

That would be the same as offering your swaps at a fixed rate of 0%. It seems like very few people if any would choose to do that. When was the last time that the USD swap book had no demands on it?

After thinking about it some more, I like my second idea better, though:

Quote from: 2586 on November 16, 2014, 12:02:09 AM

Place new swap offers at x%
Reduce swap offer rate by y% for each hour(or minute?) that it remains unfilled
Do not reduce swap offer rate below z%

This one also has the advantage that default values can be provided, since it will still create a range of offers even if everyone uses the same parameters.

Offers slowly cascade down until they reach the current supply/demand equilibrium point. As demand for swaps rises, it pushes that equilibrium rate up. As swap supply rises, it gradually pushes the equilibrium rate back down. You could set everyone at a default of x=1%, y=0.1% (per hour, or 0.00167% per minute), z=0.05%, and let people fiddle around with it from there. With those settings, your offers usually won't sit idle for more than 10 hours, which is probably better than the FRR on average. When demand goes crazy and the offer book gets cleared, your new offers start getting filled near %1, which they'd never do if set to the FRR.

Quote

I believe my idea of creating an 'effective FRR' by charging a markup over FRR (do not change the way FRR is calculated) involves less disruption to the way people react to it right now, it somewhat solves the wall issue as lending offers that are taken slightly below FRR will still result in FRR increasing.

This would cause FRR offers to sit idle for even longer than they already do, and doesn't remove the wall. It's less bad than the current FRR setup, but still has the same primary weakness: spikes in demand aren't apparent until the offer book has been entirely cleared out. The wall is allowed to float up slowly, but the price (interest rate) signalling mechanism still gets suppressed. The main benefit of your proposal would be that it'd make the FRR less attractive to lenders. May as well just remove the FRR and its distorting effect entirely.