The big disadvantage is that you can get a much better yield investing in AGNC instead of AGNCP. But, like people have said, if AGNC's earnings deteriorate, you have less risk of a dividend cut or worse holding AGNCP. However, if things go horribly bad with the business, the preferred investors could still take substantial losses. For example, back in the Financial Crisis of 2008, investors who owned what they thought was safe securities (bank issued preferred stock) took heavy losses. I would expect that AGNCP stock would follow AGNC's stock price down if things went horribly bad so you are getting a small amount of safety in AGNCP and giving up about 6% in yield.
The losses on Preferred stock were because investors sold at a low market price. For instance BAC Preferred PPS dropped to $16. The price recovered as the company recovered. The only way to lose all is thru BK. all Preferred shares are different you must read the prospective to get the details.
Ok. Firstly.. Agnc will only yield 10 percent soon based on NAV. So 8 percent is good for the safety of no cuts. Second.. The agnc common would have a 0 dividend before you can suspend (delay) agnc preferred dividends. Any dividend on common even 5 cents would mean preferred gets paid.
The divs on the preferred are cumulative. I.e., the company can reduce the dividend, but has to make it up at some time in the future.
AGNC has to pay 90% of its profits, so it won't cut the preferred until the common div is wiped out, as you said. But if it also has to cut the preferred div, it means the company is no longer making even (1/.9) * 50 = 55.6 cents per preferred share per quarter, and there are a very small number of such shares. I.e., profits will be very skinny, and heading lower, probably negative.
Probably into a situation where winding-up the fund is the best course of action.
So, if the company is unable to pay the preferred div, it's a good chance they'll never have a profit to make it up with again. The "safety" is only of benefit in a regime where the company is super-close to losing money but not actually losing money. It's a razor-thin range of livelihood.
Meanwhile, 99.9% of the time, you're just giving up the higher yield and higher chance of equity appreciation on the common.