This post triggered a horror-show of ignorance and insults in the comment section. So against my better judgment I’m going to push back one more time at a set of arguments that don’t stand up to close scrutiny. Here’s a typical comment (by Suvy):
This is complete garbage. For every dollar that’s printed, it takes away from my purchasing power. By printing money and stashing them in bank reserves, the banks gain spending power at my expense. Inflation is a tax; printing money is a tax. The first people to get the new money benefit the most. If the banks get the money first; they benefit the most at my expense. It makes a huge difference on how the new money is introduced.
Here’s how things actually work. The government prints up a million $100 bills at virtually zero cost. The bills are used to buy $100 million worth of Treasury securities. The existing stock of currency sees its purchasing power fall by $100 million in the long run. So the government gains a profit of $100 million, which economists call “seignorage.” Existing money holders see a loss of $100 million in purchasing power (in the long run.) That’s because someone has to pay any tax. And it’s a wash for the person or bank selling bonds to the government; they get $100 million in one asset, and give up $100 million in another. Now let’s start going over the myths:
1. “It helps the bond seller because they earn commission on the sale.” That’s actually true, but a trivial gain relative to the size of the monetary injection. Not important in a macro sense, as I think even my opponents would agree.
2. “It helps bond holders because it drives up the price of bonds.” That might be true or it might be false. But even if it is true, the price of bonds rises due to what’s called the “liquidity effect,” which will occur regardless of who gets the new money. That’s because interest rates are the opportunity cost of holding money, and thus if prices are sticky then interest rates may fall in the short run, as a way of inducing people to hold the extra money until prices adjust. But that’s equally true if the money is injected by paying the salary of government workers, rather than OMOs. Or buying gold instead of securities.
3. “It helps the bond holders because the government is buying bonds, rather than something else.” Holding fiscal policy constant that is false, as the amount of debt held by the public would be exactly the same even if the government were paying the salaries of government employees with the new cash, rather than injecting the money via OMOs. The only difference would be that with the cash paid to government workers, the Treasury would issue $100 million in fewer bonds. In contrast, the OMO leaves the gross debt unchanged, but the net debt gets reduced by $100 million as the Fed buys up that much in Treasury debt and takes it out of the public’s hands.
4. “It helps the people or banks that get the new cash because they get the first chance to buy up assets that are about to appreciate in price.” Where does one even start with the series of mistakes in this claim? Obviously for auction style asset markets there’s no advantage in getting the money first, as prices respond instantly to the news. But let’s say I was wrong; let’s say the price of gold rises slowly in response to an injection of new currency. The claim would still be wrong, as holding newly injected money doesn’t give anyone any sort of unfair advantage. If I heard about the Fed’s new OMO, and thought gold prices would rise gradually in response, I’d simply call up my broker and buy some gold. What if I don’t have any currency? I’d charge the purchase on my credit card. What if it was some product you could only buy with cash? Then I’d simply get money from an ATM machine, or a bank, and make the purchase.
This is a good example of the fallacy of composition. In aggregate, the total level of nominal purchases is constrained by the amount of currency in circulation. But not at the individual level. Hence being the first to get the new money doesn’t confer any advantage at all–as the new money has no more purchasing power than the existing money. A dollar is a dollar—and a $100 bill is a $100 bill.
Indeed if you really thought that asset prices would rise with a lag, then holders of the new money would actually see a loss of purchasing power due to inflation. The banks Suvy thinks are making out like bandits see the purchasing power of their massive ERs fall by 1.5% per year via inflation, and only get a measly 1/4% in IOR. I oppose any IOR, but they are hardly making out like bandits. To avoid the loss of purchasing power banks might want to spend the new money right away. But Goldman Sachs isn’t even paid cash for the T-bonds the Fed buys in OMOs. I don’t know how they are paid, but whether it is a paper check or some sort of electronic transfer to a bank account, the point is that G-S would have to actually convert that monetary injection into cash before spending it on the goods that are going to go up in price–if cash was “purchasing power.” But I could do the same, even if I never participated in OMOs! Getting the money first gives G-S absolutely no advantage over me in terms of being able to speculate in assets that might rise in value. And of course if it was goods that one could buy without cash (i.e. most goods) then there would be no basis at all for the view the receivers of new base money have an unfair advantage.
OK, so simple logic tells us that it can’t possible matter who “gets the money first.” But might it matter what the money is spent on? This question is more complex. If the new money is not used to purchase T-bonds, then some other asset holders might benefit—as you’d be combining fiscal and monetary policy. Indeed even a purchase of gold would now be equivalent to fiscal stimulus, as gold is basically just a commodity. However I’d make the following observations:
a. During normal times when interest rates are positive then the size of monetary injections is typically quite small as a share of GDP, so the macroeconomic impact of some unconventional purchase–say equity mutual funds–would be utterly trivial compared to the macro impact of a larger monetary base. On the other hand if they purchased some thinly traded good, say rare stamps or coins, then obviously although the macro impact would be small the impact would be significant in those very small markets. But central banks don’t do that.
b. During periods where the interest rate is zero it is possible that the monetary injections would be quite large. In that case it might matter whether the Fed bought T-bonds or MBSs or foreign exchange. However since MBSs created by the GSEs have been backed by the Treasury, they are now very close substitutes, so even here the effect would be quite small. And if the Fed bought more foreign debt, that would be mostly offset by foreigners buying more US debt. And most importantly, any unconventional purchases by the Fed would matter, if they mattered at all, due to the asset being purchased, not who gets the money first.
To summarize, it makes my brain hurt to try to understand how anyone could think it matters who gets the Fed injections first, assuming the new money is sold at fair market prices. Needless to say I can’t rule out the theoretical possibility that G-S gets some sort of sweetheart deal and earns above fair market commissions on OMOs in Treasury bonds. If so then that’s a scandal that needs to be investigated. But it would hardly be the basis for a serious theory of the macroeconomic effects of monetary policy. It would be like claiming the Federal government doesn’t really control the monetary base because the North Koreans produce some $100 bills as well. True, but nearly irrelevant.
PS. I hope people realize my $100 bill example was a simplification of reality. The modern Fed usually injects money first as reserves, then they get converted to $100 bills when the public’s demand for currency rises. Nothing in my post hinges on that complication. Before 1913 the entire base was cash.
PPS. If people who disagree with me want to be taken seriously they might start by being a bit more polite. I find that’s usually correlated with intelligence. For those who were polite in the previous comment section my apologies for this tirade.
PPPS. I was asked to comment on Noah Smith’s recent critique of David Beckworth’s post on the fiscal multiplier. I basically agree with Noah, and would simply add that his arguments also suggests that the standard arguments in favor of the effectiveness of fiscal stimulus are also mostly flawed, in basically the same way that he claims Beckworth’s arguments are flawed (ignoring expectations channels, etc.) When it comes to fiscal stimulus it’s all about faith—the data tell us almost nothing.
PPPPS. Some of the smarter commenters tried to convince me that money is non-neutral. My entire blog is devoted to the proposition that money is non-neutral–I don’t need convincing. But that has no bearing on the claim that the effect of money policy depends on who “gets” the money first. Non-neutrality in the labor and goods markets comes from sticky wages and prices, and non-neutrality in debt defaults comes from sticky nominal debt.