Commenter Negation of Ideology is as perplexed as I am:
Scott is clearly right on this, as I explained on an earlier thread. But I’m still trying to understand the objections. Is it the tiny margin on bonds that primary dealers make? If instead, the Fed simply logged onto Treasury Direct and bought and sold bonds directly from the Treasury would that solve the issue in your eyes?
Then, of course, the monetary base would expand by the Treasury selling less bonds to the public than otherwise.
Maybe that’s the issue we should talk about. Does the Richman complaint about OMOs disappear if the Fed buys directly from the Treasury? This is actually identical to my example where the Fed injects the new money by paying the salaries of government employees that would have been working even without the monetary injection.
Or maybe the issue isn’t commissions; it’s who gets the new money (as Richman claims.) Here’s another way of stating that issue:
Suppose the government buys a billion dollars in goods from two different government contractors. One is paid by the Fed with $100 bills with big pictures of Ben Franklin. The other contractor is paid with $100 bills collected via taxes. Since they are the older bills they have small pictures of Ben Franklin. Is the contractor who receives the “big Ben” bills somehow better off because he “gets the new money from the Fed?” If so why? Do bills with big pictures of Ben have more purchasing power?
Obviously not. So if getting money directly from the Fed doesn’t give you more purchasing power dollar for dollar, then the issue must by the commission, which Negation of Ideology alluded to. So Richman’s claim would be that bond dealers get more purchasing power because they get a sweetheart deal from the Fed. The Fed could inject money more cheaply by buying bonds directly from the Treasury
Fed. Maybe so, but people who know much more about this than me suggests the commissions are extremely low, merely a few basis points. Perhaps there’s a scandal there, but how can that possibly be of macroeconomic significance?
Commenters also raised the issue of how banks are favored because they get the new money. We need to contrast two cases:
1. Normal times. When T-bills earn positive interest rates banks want ERs about as much as a hole in their head. The minute they get ERs they try to get rid of these non-interest earning assets by buying other assets, such as T-bills. All asset prices adjust in the ultra-short run until the public is willing to hold the extra base money as cash. Then NGDP starts to rise. In the long run interest rates and real asset prices return to normal after all wages and prices have adjusted. In the ultra-short run the hot potato effect works through asset prices adjusting, and then over time it works through higher nominal spending on goods and services.
2. The zero bound. Now banks do want to hold those ERs. But guess what, the cash will mostly flood into the banks even if first injected to the public. The public is not allowed to hold deposits at the Fed (much less earn interest on cash). The public can only hold base money as cash. If the Fed dumps a $100 billion in base money into the public (say by buying bonds from the public) they will immediately deposit most of that money in the bank. Now it’s the public that views truck loads of cash as being about as useful as a hole in its head. If you think it’s unfair that banks end up with most of the new base money when rates are zero, you need to remember that no one else wants it! OK, the 0.25% IOR program is unfair, and favors banks. But surely that’s not what all this Cantillon effect stuff is based on. The program only began in 2008.
PS. dtoh has also had some good comments on how bond purchases actually work.
PPS. All those commenters trying to convince me that QE affects stock prices, bond prices, redistributes income, etc., please stop beating that dead horse. He isn’t going to get any deader. I know all that.