> Nobody can know how this will pan out, but I think a reasonable guess is that as QE is allowed to unwind we experience recessionary forces that we are no longer willing to accept or able to withstand - given the increased fragility and interconnectedness of our economies. And so we will see QE3 and QE4 and so on, with related ZIRP[2], until we finally see a deflationary collapse.
It seems like the flaw is in the assumption that QE be required to "unwind" at all.
When the fed creates/destroys money, there are at least two things it can do with it. One is to buy/sell treasury bills, the other is to lend more/less to banks (who use it to make loans to consumers).
We currently have two problems: 1) too many loans to consumers, 2) too many outstanding treasury bills. So why doesn't the Fed set a near-zero interest rate for treasury bills but a higher interest rate for banks? That keeps them from having to destroy the money entirely and cause deflation to get consumer debt back to a rational level, while at the same time removing treasury bills from the market and reducing the interest rate the taxpayer has to pay on the outstanding debt -- solving both problems at the same time.
The result would be for the Fed to hold trillions in treasury bills indefinitely, but so what? The "harm" is that it allows Congress to borrow money more cheaply. That seems like a benefit provided they don't abuse it to the point of causing too much inflation -- which certainly doesn't appear to be happening in practice. The debt numbers would continue to grow on paper, but if it's all just going to be held by the Fed using created money, and the amount of money being created isn't causing hyperinflation, are we not good?