I see a lot of discussion of Fed “debt monetization”, and yet it’s not exactly clear what that term means. Here I’ll discuss two types of debt monetization, and a third policy that is often wrongly viewed as debt monetization.
The best example of debt monetization is a large, one-time increase in high-powered money, which the central bank uses to purchase interest-bearing government bonds. High-powered money is base money that pays no interest, such as currency and zero interest bank reserves.
This policy produces a one-time increase in the price level, but no persistent increase in the rate of inflation.
A second type of debt monetization involves a persistent increase in the rate of inflation, such as we saw during the 1960s and 1970s. This is only effective if a substantial amount of government debt is in the form of long-term bonds. The inflation reduces the value of these bonds in real terms. In contrast, if all government debt is in the form of short-term T-bills, then the interest cost on T-bills rises with the higher inflation, producing no real gain to the government.
This second type of “debt monetization” doesn’t actually involve very much conversion of debt into money. That’s because at high rates of inflation the real demand for high-powered money falls sharply. If the Fed switched to an 8% inflation target tomorrow, then the Fed would likely have to reduce the amount of high-powered money in circulation. (Interest-bearing bank reserves are not high-powered money, as there is no long run fiscal gain is swapping interest-bearing reserves for interest-bearing T-bills.)
Some people regard a “low interest rate policy” as another form of debt monetization. This is false. Indeed low interest rates are not even a policy; they are the outcome of various other monetary policies, plus other non-monetary factors. The Fed has only a very transitory effect on real interest rates, which are determined by underlying economic fundamentals. As a result, a policy of persistently low nominal interest rates requires persistently low inflation, i.e. a tight money policy. That does not provide a fiscal gain to the government, as one can see from the example of Japan.
David Beckworth recently retweeted a couple of Adams making a similar point:
PS. The recent surge in the CPI is not relevant to this post, as it’s likely just a transitory supply-side shock. While I follow convention in focusing on price inflation, NGDP growth is the more relevant variable for this sort of analysis.
READER COMMENTS
iskander
Jul 13 2021 at 4:14pm
Would we not expect the end of the covid shock to be a be deflationary (as it is the end of a negative supply shock)? why is it inflationary or is there (the end of) a simultaneous money demand shock creating net transitory inflation?
Scott Sumner
Jul 13 2021 at 7:22pm
Yes, it should be deflationary. Which suggests to me that we have not yet reached the end of Covid. There are still lots of supply bottlenecks. But this might also reflect policies triggered by Covid, such as enhanced unemployment insurance.
John Hall
Jul 13 2021 at 4:23pm
I think the big misconception is on the relationship between QE and debt monetization. You could argue that excess reserves, by paying an interest rate, aren’t considered high powered base money, but I think you need to spell it out a bit more for the people who don’t understand why that’s important.
Scott Sumner
Jul 13 2021 at 7:24pm
If reserves pay interest, then the gain to the government in buying back its debt is offset by the added cost to the government in paying IOR.
bill
Jul 14 2021 at 8:25am
Actually Loss.
The rate paid on IOR exceeds the rates on 1, 3 and 6 month Treasuries.
Andrew_FL
Jul 13 2021 at 4:54pm
Was debt monetization even effective in the 70s? Long term nominal interest rates got pretty high, too.
Scott Sumner
Jul 13 2021 at 7:23pm
It was only briefly effective, reducing the value of long-term T-bonds issued before the Great Inflation.
Brian
Jul 13 2021 at 7:40pm
Is there a synopsis on how debt monetization happens in high inflation economies? One economics blog comment some weeks ago said that new money is not used to purchase government bonds in Argentina. Which are the big or mid-sized economies where that is the practice?
Scott Sumner
Jul 14 2021 at 1:56pm
You get the same effect if the new money is directly spent on goods and services or transfers.
MarkLouis
Jul 13 2021 at 9:28pm
What would NGDP-targeting call for in terms of current monetary policy?
Scott Sumner
Jul 14 2021 at 1:57pm
Hard to say. Policy seems reasonably on target to me, but the economy’s supply problems are so great that it’s hard to interpret the economic data.
marcus nunes
Jul 14 2021 at 3:04pm
In part that depends on the “end-point” the Fed desires…
https://marcusnunes.substack.com/p/state-of-play-may-21
Michael Sandifer
Jul 15 2021 at 1:22am
Scott, you wrote:
“The Fed has only a very transitory effect on real interest rates, which are determined by underlying economic fundamentals.”
My model predicts that wage-adjustment alone should have allowed for full recovery from the Great Recession within about 6 years. But, that assumes the Fed doesn’t tighten policy at times during the recovery. The Fed ran PCE core inflation at about .4% below it’s target for 8 years of the recovery, which can translate into up to 3-4 times that much in terms of real GDP growth. Obviously, real GDP growth and real interest rates are correlated. Indeed, I think they’re probably equal in monetary equilibrium as they were close to being during the Great Moderation.
And notice in the graph linked to below that the real rate starts to rise in 2012. Were real factors in the economy causing this improvement? It’s conceivable that there were some real factors at work, but then what about claims about demographic factors, shifts to the service sector, etc.?
https://fred.stlouisfed.org/graph/?g=Fl1U
And why was the real rate nearly 3% before the Great Recession, but crashed, and then was negative for years after the Great recession? Would you really attribute most of that to real factors? Did the world suddenly wake up and realize the US had long-term supply-side problems?
And, as mentioned before, real rates were mostly flat during the Great Moderation, in which nearly everyone agrees that policy was pretty close to equilibrium for most of those years. I just don’t see what appears to be mostly a policy of the Fed just following real rates down over 40 years.
Michael Sandifer
Jul 15 2021 at 1:31am
I guess I should address what some refer to as the global savings glut. Perhaps it’s just my stubborn ignorance, but I still fail to imagine how a glut of savings could ever be a bad thing for the US, as it’s so often been characterized. I’m thinking the model economists use for savings and investment is simply wrong. I don’t think that’s an outrageous statement, given how perplexed economists seem to be by what we’ve seen with real rates in the developed world, and related observations in recent decades.
I can’t get passed thinking that a glut of foreign savings should just fuel more investment in the US, if we’re willing to supply enough money to allow the economy to grow at potential. Where am I going wrong, and is there actually an empirical case?
Michael Sandifer
Jul 15 2021 at 6:58am
Is there any doubt that, ceteris paribus, an overseas increase in demand for dollars/dollar assets will lower AD, if not offset by monetary expansion?
So, I’d think that one possible effect would be to allow for some combination of increased investment and consumption, with sufficient accompanying monetary expansion.
Why wouldn’t this be a positive supply shock, if unexpected? Shouldn’t real interest rates rise in response, while inflation falls?
Comments are closed.