Bill Woolsey opened a long comment in my last post by writing, “Suppose the government decides to increase tank production. It might fund this by printing currency, by borrowing money, or by raising taxes. The tank manufacturers benefit the same amount regarless of this choice.”
Hang on a second, fellas. You are changing the question.
When someone asks, “Does it matter where the government injects new money into the economy?” the natural interpretation is to say that we first inject $x billion into one place, then we hold everything else constant, and instead inject the $x billion somewhere else. If things change, then we say, “Yes, it matters where the government injects new money into the economy.”
In contrast, Bill (following in Sumner’s footsteps) is asking, “Does it matter to the recipient how the government finances a new expenditure?” (I actually think it does matter, but even if it didn’t, the question is still a different one from the one we were all supposed to be answering.)
More generally, my stance isn’t some arbitrary approach. It is exactly this kind of behavior by Scott and many other proponents of “Bernanke needs to do more!!” that has me so upset. When QE3 was announced, Sumner et al. were mildly pleased, they just wished the numbers had been bigger. That’s because Sumner cares about NGDP first, NGDP second, and NGDP third. Whether the Fed bought MBS, Treasuries, or Robert Murphy books, he didn’t care. The important thing was, pump more money into the economy and get the public to think future NGDP would be higher.
So that’s not at all a scenario of, “The government was going to buy a certain amount of MBS, and the question is, should it do it by raising taxes, or by….” If the Treasury had started buying $40 billion of MBS each month, and paid for it by raising income taxes, do you think Scott would have approved? Of course not.
So Woolsey’s question, though perhaps interesting in its own right, has no bearing on this argument over the Austrian usage of Cantillon effects.