Greg Mankiw’s exam question

Via Ron Paul and Dean Baker (really):

  1. According to Congressman Paul, to deal with the debt-ceiling impasse, we should tell the Federal Reserve to destroy its vast holding of government bonds.
  2. Because the Fed might have planned on selling those bonds in open-market operations to drain the banking system of the currently high level of excess reserves, the Fed should (according to Baker) substantially increase reserve requirements.

This would be a great exam question:  What are the effects of this policy? Who wins and who loses if this proposal is adopted?

My answer is a little different from Greg’s, though without denying he has outlined a possible and indeed plausible scenario (he argues that destroying the bonds is irrelevant and #2 would be contractionary).

I don’t quite view the Fed as a pure Ricardian shell of the government as a whole, at least not for every marginal choice.  So if the Fed has its stock of bonds destroyed, it wishes to recapitalize itself, in part because it doesn’t fully trust the indirect option on recapitalization through Congress and it wishes to restore its financial power base.  It can do this in a number of ways, but the simplest is to print up more money and buy something valuable.  We get QEIII, though on which assets is unclear.  The shadow banking system expands, while the higher reserve requirements (assuming a penalty rate of interest) contract the non-shadow banking system.  Overall that’s a bad mix, though we have temporarily solved the debt ceiling problem.

In the longer run the Fed is more subject to the will of Congress (it required some kind of implicit permission to recapitalize, and thus some kind of deal was struck) and that is probably contractionary after the new QEIII wears off.  Overall, it can be said that when there are agency relationships it matters in whose pocket the assets are stored and thus full Modigliani-Miller neutrality will not hold.

If you’re from Minnesota (I am not), you will start by asking how the destruction of the bonds is done in a “balanced budget” manner or not.  Is an offsetting change in fiscal policy required to maintain a balanced budget constraint, or can we think of the bond destruction as itself an act of fiscal policy?

Returning to the previous track, you might alternatively think that Congress and the Fed cannot cut a deal and, after this policy change, the Fed cannot recapitalize.  In that case the Fed’s “sopping up the expanded monetary base” option is weaker than before, this is viewed as implicit monetary expansion if only probabilistically.  Instead of QEIII we get a probabilistic QEIII.  A boom is boomier because M2 is rising and there is less mopping up, but there is probably less expansion in the more pessimistic scenarios because until there is strong positive pressure on M2 the absence of the mopping up option doesn’t mean that much.

Those are not the only scenarios.


The Fed cannot reduce the asset side of its balance sheet without reducing the liability side. Burning the bonds would leave the Federal Reserve bust. Ron Paul and even you seem to forget this. One way to think of this is that the bonds are bought with the excess reserves generated by the banking system at the Federal Reserve. These reserves are assets of the banking system and liabilities of the Fed. By burning the 1 trn $ of bonds, the Federal Reserves will be in the red at a about 1trn this year. And if it doesnt recapitalise (with Treasury reissuing the 1trn) then the banking system is bust.

"And if it doesnt recapitalise (with Treasury reissuing the 1trn) then the banking system is bust."

- The plan would be to have the Treasury reissue the bonds

This debate raises all kinds of interesting questions.

Would this simply be an accounting gimmick? That's what Prof Mankiw claims, that this is just debt that the government owes itself. But that's only true at a certain definition of government. Other definitions, such as central government which excludes the central bank, lead to different results. In that case the US Federal Govt (the central government) just saw the debt it needs to finance or refinance drop by $1.6 trillion. That's a huge amount. In that sense it is no different than if the Chinese chose to burn their own Treasury holdings.

Of course, such a definition begs the question to what happens to the other side, the side that chose to forgive the debt. If the Chinese did that it would reduce their purchasing power. But that's not true of the Fed which can, after all, just print money. Carsten Valgreen claims that if the Fed is in the red the banking system would go bust. But how? There are plenty of current or historical examples of "bust" central banks with very solid banking systems (Costa Rica and Chile, to name but two).

Prof Mankiw further claims that if the Fed used reserve requirements as a policy tool that would amount to financial repression, denying loans to some market players. But isn't that the whole purpose of a more restrictive monetary policy? Now, maybe it turns out that selling Treasuries is a more efficient way of carrying out monetary policy than reserve requirements, but if that's the case is there any empirical evidence? Again, there are plenty of countries across the world that use reserve requirements and they seem to be doing OK.

A possible risk is more political than economic in nature. If this were carried out, would the federal government get used to this, and expect such Fed bailouts in the future? Given human nature the most likely answer is yes. And in the future such actions could happen under circumstances that had inflationary results.

I was quite surprised by the suggestion from Ron Paul. It is surprising how rapidly he is closing the holes in what started out as very naive economic theory.

My question about part I is the signalling effects such a move would have on the bond markets. Would it signal an end to bond issuance in a stronger way than the current recalcitrance Congress?

Tyler, I cannot ignore your sloppy scholarship on monetary issues. It's too much.
I'm from Minnesota where I got my Ph.D. But I'm Argentine and I have seen all kinds of gimmicks to deal with fiscal crises. Contrary to what Greg M. says and to your nonsense comment, the Treasury can collect the seignorage from the issue of currency and therefore Paul's proposal should be modified to limit the reduction in the amount of Treasury debt accounted as a FRS's asset to the amount of currency in circulation, that is, to around $1 trillion at the end of June 2011. Now you should explain to your readers why my proposal makes sense and they should ignore what Greg and you wrote.

"But I’m Argentine"

That explains it.

I'm not sure, but isn't this pretty much straight Modern Monetary Theory? Government bonds themselves are essentially an accounting gimmick (and a vehicle for manipulating interest rates), rooted in the mistaken notion that government must borrow to spend. This is complicated and obscured by the rather byzantine statutory relationship between the treasury and the fed, whereby the fed can't buy bonds directly from treasury -- they have to be bought by banks, then the fed buys from them.

An alternative: Obama instructs the treasury to just keep spending, disregarding the debt limit -- not selling any bonds to replenish its account at the fed. Is the fed gonna bounce treasury's checks? Maybe yes, but, might it not instead just issue the necessary reserves, without the circuitous kabuki bond dance?

As for the Mankiw/Reinhart/Gross hogwash about "financial repression," long government bonds are essentially a gift to banks, paying interest on risk-free holdings. Is removing that gift (say, by only issuing interest-free bills, a.k.a. dollar bills) "repression"? And really: who's being repressed financially in the current economy? My heart really bleeds for Bill Gross.

While the Fed is paying IOR, isn't this even simpler than Mankiw described? It seems to amount to the Federal Reserve selling the bonds to the banks and renaming them "reserves".


If the Fed sells the bonds then the obligation to repay them back remains, so I think it's different. What Paul and Baker are arguing, as I read it, is to eliminate this obligation altogether.

You guys are thinking too hard. Rep. Paul's proposal would make explicit exactly how much of the current debt the US does not intend to repay. That might be good, or bad, depending on your view; but that's all it is.

This is so bizarre for a layman to read. All I know is that in addition to watching bond spreads, the one thing that kind of makes me okay with the printing of money is that I pretty well understand the explicit lever we can pull to sop it back up when things heat back up. I don't understand what this means.

I understand your frustration. It's the same frustration that serious economists should feel about the failure to explain clearly monetary issues. Unfortunately most economists are still addicted to the theoretical approach of monetary aggregates that combine means of payments with financial assets. Today in all economies there is a government-issued currency and for accounting purposes it is a liability of a central bank, but since this currency is no longer backed by gold or any commodity, for accounting purposes the double-entry principle requires to enter an asset (in the Fed they call it U.S. Treasury bill but I prefer to call it confetti). The modified proposal I suggested in a previous comment amounts to replace the confetti with toilet paper for accounting purposes. It doesn't have any economic effect because the seignorage was already collected by government (in the past 100 years). The only effect is to circumvent the debt ceiling because someone made the mistake long ago of considering the confetti as debt when it was not government debt --of course, let us hope that accountants and lawyers agree that toilet paper is not government debt, otherwise it would be stupid to replace the confetti.
BTW, the rent that the issuer of currency earns at the time of issuing it. You cannot earn rent twice on currency already issued (in other words, if you need additional rent, you have to issue more currency).
Also, don't get excited by Paul's proposal because most likely there will be a simple agreement to increase the debt ceiling by a small amount, enough to finance the government for one or two months.
It's unfortunate that Tyler has failed consistently in explaining monetary issues.

Sorry, it should have said: BTW, seignorage is the rent that the issuer of currency earns at the time of issuing it.

Telling "the Federal Reserve to destroy its vast holding of government bonds," may violate the Fourteenth Amendment to the U.S. Constitution, Section 4 of which states:

The validity of the public debt of the United State... shall not be questioned.

Read the whole thing.

I agree with Steve Roth.

Any inflationary effect took place when the Treasury spent the money ir got from the Fed. That's already done. Forget it.

Yes, the Treasury did not directly sell the bonds to the Fed. Big deal! The Fed might as well have trucked money to the Treasury building and unloaded it in the lobby.

Now, to the question itself. What happens if the Fed shreds the bonds?

Answer: Not much of anything. The Treasury learns it will escape paying interest or redeeming the bonds. Everyone may get a bonus after such good news.

But back at the Fed some people are worried because they can't someday sell bonds that have been shredded. They feel their power is diminished. Then they get over it.

Their accountant now says the Fed is bankrupt. So an asset is created "Expected recovery of $1 trillion from someone, someday". Problem solved.

First, this is a very bad exam question because nobody knows the correct answer since the experiment hasn't been done. To those who think the bonds being destroyed are an accounting gimmick or confetti, why are they valued so much by the market? Hard to imagine how destroying an asset of some value won't negatively effect the dollar (a matching liability) thru either inflation and/or additional steep falls vs. other currencies. Raising reserve requirements may draw those dollars out of circulation but velocity is likely to increase to compensate for the Fed destroying their balance sheet. Furthermore this will be a clear signal to markets that the Fed is giving up even more of their independence also a negative for the dollar. Lastly, Ron Paul of all people should be more skeptical of this type of action for it is nothing more than central planning, his punishment should be to write a 1000 times, "Central planning doesn't work even when 'Austrians' do the planning."

I issue a negotiable instrument to myself that says "I promise to pay $X on Y Date." Someone may like to buy that note from me, therefore it is an asset with value, therefore destroying the note will have negative consequences for me.

Do you see the illogic?

not what's happening, the Fed has already paid for those bonds with something that it wants those on the receiving end to believe has value. It is not just the left hand loaning the right hand money as Mankiw asserts

It is also be worthwhile to do some public choice type analysis on #2, not just #1. It is incredibly likely that whatever reserve caps are put on get "gamed" in some important way.

(Such as using CDSs to allow you to hold AA-rated tranches of mortgages as "reserve" capital.)

This may be what you meant by "shadow banking"

A huge question that is being ignored is, how would capital markets respond to this effective debt forgiveness?

I have a feeling Wall Street and Shanghai might be more than slightly discouraged by the US government forgiving its own debt. What happened to the whole independence of the central bank?

Dean Baker goes on and on about how if US bonds held by the US govt via the Social Security trust fund aren’t honored in full it will be a default that ruins the credit rating of the USA.

Now he says that if US bonds held by the US govt via the Federal Reserve are cancelled, it will a great thing. Let’s do it!

I’ve always found him entertaining.

Irregardless of the immediate effects of Ron Pauls suggestion it seems inevitable that the U.S. is going to default at some point in the future. The only question is whether it will be done with straight printing or "haircuts". Haircuts (even severe ones) to those bonds held by other portions of the government but not to outside holders would seem to be the least troubling to the markets. I don't know if it would be enough.

I suppose there is a third option. Simply have the FED buy more and more bonds in perpetuity, I have a hard time seeing this as significantly different from straight printing. I guess thats why the MMT guys call it an accounting gimmick.

Someone, please confirm/deny whever this part of Mankiw really is nonsense defying Monetary/Fiscal Policy 101, as it seems to me? Is he finally heading into GOP fantasy neverland? Currently, Mankiw sounds to me like a person who does not understand fundamentally how money+CB works, and it's kinda hard to bellieve.
In fact, money that FED gets from those bonds do disappear from economy, and the Treasury does lose those repaid funds, and can't use it for other stuff. And FED does not equal government.
Not to mention all second-order market confidence stuff wrt apprent direct inflation-tax budget funding..

>> 1. According to Congressman Paul, to deal with the debt-ceiling impasse, we should tell the Federal Reserve to destroy its vast holding of government bonds.
>> ...
>> -Okay. Here is my answer:
>> Part 1 is just an accounting gimmick. Since the Fed is really part of the government, the bonds it holds are liabilities the government owes to itself. Destroying the bonds has no direct economic effect. It is just like an increase in the debt ceiling, without any other policy changes attached.

If the mainstream is suddenly surprised that Ron has a good idea, I may need to update my priors.

If General Motors gave away all of its assets, its stock would fall to zero value. If GM then issued a billion new shares and sold them, the price of those new shares would be zero. Recapitalization won't work for GM, and it won't work for the Fed either.

Suppose that GM issued a billion shares of GM stock but did not sell them. Instead it gave them to GMAC, a wholly-owned subsidiary. And GMAC later decided to give them back or destroy them.

GM stockholders might be happy because they would not have to consider the possibility that GMAC might at some time sell them and dilute their stock. Beyond that possibility, the existence or destruction of those shares would have no relevance to them.

Would people worry that GMAC was now in trouble? Yes. No. Maybe. Did GMAC use those shares as collateral for loans they took out? If they were just sitting there, it wouldn't matter. If people thought GMAC could get new shares from GM if needed, why would it be an issue? Would GM let GMAC fail and then argue that GM's losses were limited to its ownership because GM was a separate corporation? It would be a sad day for GM when they just let GMAC fail.

But as it turns out, GM sold majority shares in GMAC in 2006. Then in the crisis GMAC made enough loans to support GM's survival that GMAC itself was threatened. The federal government bailed them both out, and for awhile was the majority owner of both.

So, does the federal government own the Fed? If not, who does own it?

Just to be clear, Mankiw is not arguing that it is contractionary in the sense that tight money is contractionary (less NGDP), rather that it is contractionary in the sense that distortionary taxes are contractionary (less RGDP.)

What a strange question for an exam. In the first place supposedly the Fed is "really" part of the federal government so the bonds have no value. Therefore it makes no difference whether the Fed destroys them or not. But the Fed is a bank with a balance sheet to balance. It simply can't just destroy a financial asset without something happening on the liabilities side. To every action there is an equal and opposite reaction.

Comments for this post are closed