Although the debate on the effects of QE on the economy and whether QE is deflationary or inflationary has (at least in my mind) been settled, there seems to exist a rather going "concern" on whether the buying of government bonds from the Central Bank equals money printing. Whether it does or it does not have separated people into two camps: those who believe that too much QE can lead to hyperinflation and those who believe that it will cause nothing of this kind.
But first things first: QE, although usually defined as the purchase (by the Central Bank) of government bonds in the possession of commercial banks. Yet, there is also an additional point: the one where the Fed purchases new bond issues. To see why this points holds see the following breakdown:
As it is obvious from the above, the only major change in the categories is the increase in foreign demand for US debt and the increase in the Fed's share of debt. Essentially, as many have argued before, when QE is initiated, collateral in the form of government bonds becomes more scarce in the economy. This is supposed to make the banks focus their funds elsewhere, meaning an increase in lending. These operations are usually conducted by the Fed either at the expiration and the re-introduction of a bond or by direct "investment" in the markets.
The point to be made here is that if the bonds are purchased from the pile of existing bonds then QE is nothing but an asset swap: cash is exchanged with bonds (both at zero risk for the bank). This does not mean an increase of the money supply whatsoever. Yet, if the government decides to issue additional bonds (remember the whole "raising the debt ceiling" debate?) then the Fed is essentially creating new money by purchasing some of them.
Remember that in order for newly printed money to enter the market it has to either be channeled through government spending or by throwing these amount off a helicopter. (If we choose the former then Monetarism and Keynesianism are essentially saying the same thing.) Thus, if the Fed purchases bonds from new issues, then it is essentially creating new money to enter the market via the government spending channel. Here, we are talking about money which did not exist before. Again, if it was a bond rollover then we would be talking about an accounting increase in cash and a decrease in government bonds (both on the asset side of the balance sheet) which have no effect on the money supply. If the money is lent, then we have an indirect increase in the outstanding amount of money in the economy, via the money multiplier, yet, this is not money printing. It is printing only if the Fed purchases new bond issues.
Now suppose that the Fed buys some new bond issues. Should we experience hyperinflation? The answer is no and not because increasing the money supply does not mean an increase in inflation. It is simply because of timing. Since the money base (M0) is just about 1/3 of the broad money in the economy (MZM) the effects of a rise in M0 just offset the decrease in MZM due to deleveraging. This is the major reason why the money supply in the US has been increasing over the last year despite the decrease in loans. It is just now, that the increase in bank lending has returned to its "normal" growth rate, that QE has began tapering.
Is QE a panacea? Obviously not. As already said, it may cause short-term asset bubbles and disinflation as a result of increased investment in the stock market. Still, those are just short-term effects and compared to the contrary (in the case of the US, a huge depression). In addition, just like fiscal stimulus, it can only be implemented when times are bad. In booms, QE and fiscal stimuli can cause private investment "crowding out" thus forestalling growth and creating additional inflation. In booms, the latter two tend to cause more damage than good. The two camps referred to at the beginning of this article can both be right but at different times: when times are good, QE can cause high inflations. When times are bad, it does not.
Overall, QE remains a good idea, despite its short-term side effects. The big question of whether the ECB will be able to apply something like it in the Eurozone remains to be answered.