Yes, absolutely. I mean in a sense, if your asset flow is a long-dated asset flow and you come to a view that you want to actually shorten your duration, that just compounds the difficulty of shortening your duration, so we accept that. And I think we have come to a view that while in general, we would like to keep these assets on our balance sheet, we want to -- for 2 reasons, we want to at least experiment with some sales of them. And the 2 reasons would be, one, that I think it would be a good market test for us to sort of know whether the market likes our product. And so you can believe you've got a great product but it's a nicer, more comfortable feeling for people who are obsessed with risk to say but the market also likes your product. And it does mean that you build yourself the optionality that if, in fact -- particularly managing the negative convexity that's inherent in those products, if you get beyond the scale of where you're saying, "I'm not sure I'm comfortable with the scale of that kind of risk that I can manage it," you have the option and you built the system. So we are building the system so that we can do some periodic mortgages sales there. But yes, to accentuate your natural asset origination as long, then that means that, that puts even more pressure to say, "Well, then you ought to sell." Where we have a securities portfolio, you want to offset those with selling those securities. And you can say on -- collateral damage of that is it's going to kick up security gains for you. It happens to come at a good time because you're also getting squeezed on the margins by the same thing that gave you those security gains.
Gabriel Dechaine - Crédit Suisse AG, Research Division: I appreciate that response. And then, Ed, while I've got your attention here, the dividend, 5% increase it's healthy. But maybe my expectations were too high, that seeing your payout ratio relatively lower than your peers and having a -- bumped up your payout ratio target last quarter, I would have expected -- or at least I have the expectation that you would've been going closer to 10% or something like that. Is there any particular reason that you held back?