an argument can be made that if the housing market goes busts, rates will actually be muuuuuch lower. look at japan. so in a "relative" sense, declining rates act like rising rates if their decline is not speedy enough to buffer even faster declining cashflow of debtors. "rates have to eventually go up"...i hear this every day, and it's true that rates 100 years from now will likely be higher, but i don't see the evidence to support the argument that long term rates will be higher in 3, 4, even 8 years. The fed needs to keep debtors afloat by liquifying our economy with negative real rates. That means keeping rates on the short end as low as possible for long as possible; to zero for a century if need be. economic softness on the long end causes investors to clamor for bonds, and pushes rates down. in 2008, a 2% return might be great, if housing and the stock market are losing 10% annually from here on.