Here's the decision that we could have made. At the end of 2014, we had sold assets and we were down to about 3.1. Yes. We're down to about 3.1 billion in assets. So, the main way you protect against spreads, if you're taking spread, we just have to liquidate the assets. So, at that point, we could have chosen to just sit there with a much lower portfolio with the proceeds in cash, we would have earned less money. And as I look at 2015, I scratched my head multiple times looking at it, we would have had a small amount of CMBS, we would have had a larger percentage of hybrid ARMS, we still would have taken the markdown on the short hybrid ARMS, which we don’t think is a long-term issue. And then which is more of a technical issue. And then the CMBS, we would still have on the balance sheet is would have been high quality, we were still taking some lagging on it, I'm not sure that we would have ended up in the exact thing. We had lower earnings to offset a smaller move in book value, but it probably ended up in the same place, some were between zero and negative-2% loans we had. Those asset plans is on the balance sheet. I think what asset class would probably perform better, probably the [30 or 15] [ph] fix from just a spread basis. We've been in a relatively narrow range, heavily supported, obviously from the fed balance sheet. And we've chosen instead of that product, you invest in the CMBS agency which did obviously lead to a larger book value move. Where we surprised by the move? The big surprise is really into evidence, is what happened in the swaps market or swaps, spreads, type and became in the correlations and the relationships changed in 2015. So, that was a surprise. And so, we have said we anticipate surprise and what we don’t want to do is take too much risk that would obviously take us out of the ballgame.