Public debt: don’t target the quantity, target the price

by John Quiggin on April 2, 2021

As I’ve mentioned previously, when I started work on Economic Consequences of the Pandemic, I assumed I’d be writing a polemic against austerity, as I did in Zombie Economics. Based on the last crisis, it seemed likely that any stimulus measures would be wound back rapidly, leading to a sluggish and limited recovery. That’s pretty much what is happening in Australia, where I live, but not in the US, where the book will be published. On the contrary, Biden’s policies are pretty much what I would have advocated (certainly if you take into account the razor-thin majorities he is working with). And, with luck, the main elements will be in place by mid-year, long before my book can appear.

So I’m refocusing on the issue of debt and how it can be managed. This was the central issue after the Treaty of Versailles, and also in the return to the gold standard, which prompted The Economic Consequences of Mr Churchill. io o

My central conclusion is a simple one. Rather than aiming for a fixed ratio of public debt to GDP, governments should aim to control the long-term rate of interest on inflation-protected bonds, and set it at a rate of around 1 per cent, about equal to the long-term rate of productivity growth. Since rates are well below that now, there is plenty of room for more public investment.

More over the fold

Instead of targeting the quantity of public debt it is better to focus on the price, which is best measured by the real (inflation-adjusted) rate of interest on long-term government bonds. For the US, this is represented by the interest rates for Treasury Inflation-Protected Securities (TIPS), the principal of which is adjusted in line with inflation. Currently, the rate of return on 10-year TIPS is negative (about -0.5) while the rate on 30-year TIPS is just positive (about 0.1 per cent).

These rates have been declining slowly over recent years. However, the passage of the American Recovery Plan and the announcement of the Biden Infrastructure package, involving around $5 trillion in new expenditure led to an increase of around 0.5 percentage points.

get the right strategy on public debt, it’s worth considering why any limit might be imposed. The answer is that lenders might refuse to buy more bonds and demand the repayment of the existing debt as it falls due. If governments cannot raise the money required, as has happened on many occasions, a crisis will ensue. There are three main concerns here

  • If debt is denonominated in a foreign currency, and the domestic currency depreciates, the burden of repayment can increase rapidly
  • If bond buyers fear future inflation, they will demand higher rates of interest to compensate for this
  • If bond buyers fear that the government will default on its obligations, they will be unwilling to buy bonds and will demand an interest rate premium

The US does not face the first problem, since its debt is denominated in US dollars. The second is not as big a problem as it seems, since government revenue will rise broadly in line with inflation. Nevertheless, to clarify the issue it is best to focus on TIPS

If investors fear a default, the real rate of interest on inflation-protected securities will increase. Unlike exchange rates and expectations about inflation, which can change rapidly, real interest rates typically move very slowly.

As can be seen below, the rate of interest on 10-year TIPS has declined gradually since the turn of the century, falling from a little over 2 per cent to a range between 0 and 1 per cent in the years before the pandemic. Rates spiked briefly by 2 percentage points in the worst of the financial crisis, before returning to the previous trend.

There was also a brief uptick during 2012 and 2013. The rate fell to negative levels between 2011 and 2013, following large scale purchases of government bonds (quantitative easing), followed by a return to just under 1 percent. The reversal, occurring when investors expected a rapid reversal of ‘quantitative easing’ , is sometimes referred to as the ‘taper tantrum’, but what is more striking is how modest these fluctuations have been.

The key implication here is that, if long-term fiscal policy is focused on maintaining a low and stable real rate of interest on government debt, there is little likelihood that it can be derailed by a sudden change in investor sentiment. rate
The stability of rates can be enhanced by a shift to longer term financing. There is a strong case for relying more on 30-year bonds and encouraging retirement income systems that invest in these bonds to provide secure incomes.

The ultimate long term security is a perpetual bond, like the ‘consols’ on which the British government relied in the 19th century. The advantages of perpetual securities have been discussed by both conservative and liberal writers.

Where should the target rate be set? In a world of technological progress, we expect that there should be investment opportunities that yield a positive rate of return, roughly measured by the rate of multifactor productivity growth (the growth in output from a given input of labour and capital). This is about 1 per cent.
Even before the pandemic crisis, the TIPS rate was consistently below 1 per cent. This implies that, with a 1 per cent target, there is substantial room for public investment financed by long-term debt, even after the passage of the infrastructure bill.

On the other hand, there are good arguments for gradually unwinding the expansionary measures adopted specifically in relation to the pandemic emergency. This would provide more room to move in the event of a similar emergency arising as unpredictably as the pandemic and, before that, the GFC [these events weren’t, in fact, unpredictable and were in fact predicted, but even those who feared such events couldn’t say when they would hit]



Peter Tulip 04.02.21 at 9:48 am

Governments have very little control over the real long-term interest rate.
Better would be real net interest payments.
Furman and Summers suggest this should be 2% of GDP.


Tim Worstall 04.02.21 at 11:01 am

This might be a little difficult. We currently use the size of the public debt – through QE – to try to target nominal interest rates and inflation. Using the same mechanism to target real interest rates at the same time might be a little more complex than the single mechanism can manage.


MisterMr 04.02.21 at 1:31 pm

This is slightly OT but, seriously, I don’t understand why people believe that the government eimitting bonds is less inflationary than the government printing money.

Honestly the government of the USA emitting bonds for 100$ is the same thing of the government of the USA printing a 100$ bill, since it can always repay the bond by printing the bill.

So I don’t really understand the problem of investors being able to demand an higer rate of interest, really this just happens because the government chooses to not print the money.

So what is really the point of not printing the money in the first place, if the government is still emitting money in the form of bonds?
Is the reason that bonds pay interest so this shifts up the average interest rate? I honestly believe this is the real reason but it is the opposite of what everybody says, meh.


John Quiggin 04.02.21 at 8:00 pm

@1 I’ll discuss Furman and Summers. They are moving part of the way to what I suggest

@2 Monetary policy should be targeted at nominal growth.

@3 The point about money is that the holders can spend it whenever they want. If governments want resources they can use, they need to reduce private spending power accordingly.


MisterMr 04.02.21 at 10:25 pm

@John Quiggin 4

Yes but people can ALSO spend bonds whenever they want: they just have to sell them for money, something that they can do istantly through a bank (though perhaps with some loss). The point is the total stock of banknotes represents a stock of savings, not a flow of income, and so do bonds.
The government emitting bonds doesn’t limit private spending power more than the government emitting banknotes, the only difference is that the government has to pay interest, IMO.


John Quiggin 04.03.21 at 5:20 am

@5 People can only sell bonds if other people buy them. No change in private sector spending capacity


nastywoman 04.03.21 at 6:48 am

As long as ”The Casino” treats ”public debt” in the sam way as private debt –
”All depends on the credit rating of the debtor (aka: ”Country-Government”)
and if a country has as excellent of a credit rating as for example Japan –
the public debt actually does matter as little as for example in the US – where somebody once said:

We can print as much money as we want.
(while Greece nearly went belly UP – if the Euro and the other credit worth EU countries wouldn’t have saved it)


MisterMr 04.03.21 at 8:26 am


I don’t want to hijack the thread, so tell me if I’m becoming boring, but:

A owns 100$ of bonds. A buys 100$ of icecream from B, thus is in debt to the bank for 100$, while Bis creditor to the bank of 100$. Then A sells the bonds to the bank to repay its debt, and the bank uses the bond as capital to stay solvent relative to its debt to B: no real banknote changes hands.

In the real world, a great amount of government debt is owned by banks, who use it as capital to balance out bank accounts, so those bonds have already been turned into money, its not like they are really illiquid because the bank system turns them into liquidity.

Now if the government of the USA couldn’t print money, USA bonds would have a certain chance of not being repaid and thus would not be as good as money, and thus banks that use such bonds as capital would be at risk from a government bakruptcy (as happens with italian banks with italian government bonds, because Italy can’t print euroes) but as the USA can print dollars, this turns USA bonds into zero risk assets, therefore they are already as good as banknotes from the point of view of banks (actually better since they pay some interest), and therefore as the banks use them as capital they automatically turn them into liquidity as bank accounts.

So in pratice the fact that the government of the USA could print banknotes automatically turns USA into money base (m4), without the need for the government to actually print the banknotes.


Tim Worstall 04.03.21 at 11:33 am

“@2 Monetary policy should be targeted at nominal growth.”

Quite so. Which will make it a little difficult to use the same policy lever to target the real interest rate…..


Rapier 04.03.21 at 11:34 am

It’s pretty funny to see the word “investors” strewn about when discussing long term Treasury rates since there are almost no investors left in the market. Using the term market ever so loosely. The Fed is now buying Treasury Notes at a rate of around 90% of new issues on a monthly basis. The Fed is most definitely not an investor. In its decision making return on investment never enters the picture.

“Investors” in this case is a euphemism. A polite fiction allowing the pretense of a market by totally ignoring the mechanisms by which the system functions. Admittedly the market for Treasury paper is nominally a market with bids and asks and trading but the simple fact is that the Treasury coupon ‘market’ is a centrally planned phenomena.


eg 04.03.21 at 3:34 pm

@J0hn Quiggan #4

“If governments want resources they can use, they need to reduce private spending power accordingly.” While true, this statement is incomplete: it only holds once all real resources are already employed — I think it’s been a very, very long time indeed since such a condition was true for the USA.


John Quiggin 04.03.21 at 8:22 pm

@8 Let’s leave it there

@9 Quite so, as you say. That’s why long-term public debt should be the instrument used to target the long-term real rate, as I said in the OP

@10 Incorrect. Here’s the data


nastywoman 04.04.21 at 9:25 am

and as –
with this Pandemic –
nobody actually – really cares about debt anymore –
(as WE HAVE to ”Keynes” the world into the future)
why not writing about ”the real problem”?
Like… like – see – I’m now flying next week to THE homeland to get vaccinated BE-cause I can’t get it here in Germany – while here in Germany – the country where supposedly everybody cared about ”public debt” – EVERYBODY cares soooo much more about getting vaccinated.

And just yesterday a very, VERY angry ”FGB” (Full Bloodied German) – neighbour – told me:
”We should have done Germany FIRST! – as it was US the TurkishGermans who invented AND HAD the FIRST Vaccine – and then this Americans and Brits come along and BUY, BUY, BUY it ALL away from US without ANY consideration for their national debt –
and NOW – WE the Germans who wanted to be ”the good not nationalistic Germans anymore” – get laughed at by the Brits and Americans – who died and dying three times faster as we did and do?

AND as that is the argument of some Nationalistic German Right Wing Idiots – and I really don’t like Nationalistic German Right Wing Idiots – can’t one of you guys on CT – who sometimes writes for Anglo papers – write about THAT –
instead about ”debt” –
nobody really cares about anymore in this Pandemic?



Rapier 04.04.21 at 11:30 am

I should have said that the Fed with it’s purchases of Treasury coupons and MBS is adding to it’s balance sheet at a rate that is about 90% of Treasury issuance every month.

The Treasury currently is sitting on $1.1TN but has to reduce that to $125bn by the end of the fiscal year because that was stipulated last year. Why? Got me. The point is some of the money for the stimulus bill is already in the bank and that will be spent down which will reduce Treasury issuance, for awhile. By summer however the Treasury is going to have to goose issuance again and when it does the Fed is going to have to increase purchases of Treasuries and MBS. NotQE II we will call it. It has to because long Treasury rates have been rising for 6 months despite the fact that the Fed is ‘printing’ 90% of the money needed to fund that borrowing. The rise in T Note rates and fall in their price is becoming an existential problem for the Primary Dealers who are holding huge amounts of the stuff at a loss. Funded by short term borrowing. Houston we have a problem.

So let’s dispense with this “investors” stuff when talking about the Treasury market. The Fed like the BOJ, ECB and BOE will now have to fund a huge portion of their respective deficits for several years. Of course central banks funding government debt to a significant degree violates Central Bank Rule #1, which is, don’t do it. The thing is everyone wants them to do it but bless their little hearts they can’t bring themselves to say they are going to do it. When they do they still won’t say it but it is coming soon enough. Like the day 10yr Treasuries hit 2% yeild.


John Quiggin 04.06.21 at 7:29 pm

Rapier, you’ve added MBS to the numerator, but not the denominator.


John Quiggin 04.06.21 at 7:34 pm

More importantly, the trends I’m talking about began even before the first round of QE during the GFC.


Rapier 04.06.21 at 10:45 pm

RE 15.

Ah, what? Well no. Any asset purchased by the Fed are in the numerator, aka money printed VS the denominator, the total of Treasury auctions which I suppose I have to stipulate goes to cover the deficit.

Or have it your way whatever that is. I suppose one could think that MBS purchased by the Fed from the Primary Dealers is actually, well I’m not sure in regard to Treasury borrowing, well fine. Why bother in this day and age with the mechanisms of the monetary/credit system and all that old fashioned double entry bookkeeping. In the Age of Q the world of Economics isn’t far behind in eagerness to bury the Enlightenment. Let me assure you that double entry bookkeeping was a foundation of the Age of Reason. The books don’t balance? No sweat.

Comments on this entry are closed.