What happens is the amount of BTC one receives when burning renBTC slowly decreases over time, but at the same time the amount of renBTC one receives when minting BTC also increases.
For example, right now if you have 1 BTC and you want to mint renBTC, you will receive 1 renBTC minus the .1% fee. If you have 1 renBTC and you want to burn for BTC, you will receive 1 BTC minus fees. If a 5% continuous fee is activated, then every 24 hours the peg is reduced by 5%/365. So in Exactly one year, burning 1 renBTC would get you .95 BTC minus the fee. But, if you want to mint, 1 BTC would give you 1.05263 renBTC (so if you burned it within the current 24 hour period you would get back 1.05263 - 5% = 1 BTC, minus fees). Every day leading up to that one year mark, the % difference increases slightly so once you hit the 365th day you are at 5 %.
This is a smart way of doing it, but I agree with others who are against it as knowing 1 renBTC = 1 BTC is very psychologically comforting and I feel also lends a certain power and credibility to renBTC if only superficially. Having anything less than a 1:1 peg tarnishes the image of it being a perfect representation of BTC on other chains. I would only use continuous fees as a last resort. It is much too early to consider this path with so much more coming to affect the economics.
As far as raising fees now, I am for the most part against it echoing other sentiments that we do not have sufficient data to test their affects. I would rather wait for further integrations before adjusting the fees. We are not requiring a certain TVL/TVB ratio at the moment and we have enough darknodes for the amount of volume being processed. However, if fees must be raised or testing needs to happen now, I like Loong’s idea of an algorithmic model that adjusts to volume. Raising fees and then backtracking if it appears we are hurting volume or adoption isn’t a good idea.