« The equivalence between national debt and transferable monopoly rights | Main | Labour market flows revisited »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

If you add debt into the model, and assume the government is in deficit, the net effect wont be the same. If the central bank buys bridges, and the government buys less bridges in response, you will still have 10 bridges, but this time the corresponding rise in public debt will be less - which means people don't have to expect as strong fiscal consolidation in the future, so the building of 10 bridges will be more stimulative.

This doesn't mean I support the convoluted 'people's QE' however.

Britonomist: debt is already in there. It won't make any difference. The amount of government debt *in public hands* will be exactly the same, because the government will be issuing $3 billion less bonds, but the Bank of Canada will be buying $3 billion less government bonds. And the government owns the Bank of Canada's balance sheet.

Not sure I follow, are you assuming that in the first scenario the central bank simply buys 3 billion in bonds instead? This is only the same if the central bank promises not to sell the debt back to the public or the central bank simply retires the debt, and that the public don't suffer from 'debt illusion', whereby they assume this central bank held debt is still a public liability.

I agree with Britonomist: Nick’s argument does not put PQE into check mate. But this argument does.

There are two possibilities. Either more bridges are needed PLUS stimulus is needed. Or more bridges are needed, but stimulus is not needed.

In the second case, the logical course is for government to borrow so as to fund bridges and let interest rates rise as a result of that borrowing, or fund the bridges out of tax. The anti-stimulatory effect of the rate rise or tax hopefully negates the stimulatory effect of spending money on bridges.

In the first case, borrowing or extra taxes are pointless in that extra stimulus is needed: the simplest course of action is simply to print money and spend on bridges. But (and this is a big but), it is not easy to start infrastructure spending quickly, come a recession, nor does it make sense to stop building bridges when they’re half constructed if the recession suddenly ends.

Ergo infrastructure should not be used to deal with recessions: other forms of spending should.

Peoples’ QE (where it consists of printing money and spending on infrastructure) is in check mate. In contrast, printing money and spending on other stuff DOES make sense. But Keynes said that in the early 1930s and most MMTers subscribe to that view, far as I can see, so that’s not an original idea.

Britonomist:

In the benchmark scenario, with regular QE, the govt buys 10 bridges and issues $10B bonds, and the CB buys $3B bonds.

In the experimental scenario, with silly QE, the govt buys 7 bridges and issues $7B bonds, and the CB buys 3 bridges

Since the govt owns the CB's assets, the govt indirectly owns $3B of its own bonds in the first scenario (which means they wash out), and indirectly owns 3 bridges in the second scenario.

And it makes no difference if the 10 bridges were originally tax-financed (as long as the CB is not at a corner solution where it runs out of bonds to buy, or the govt is at a corner solution, where it wants to buy negative bridges).

Nick, as I said you still have two issues. As I said, if the central bank pledges that the $3B is only going to be held by them temporarily, then it doesn't wash out, it is still ultimately a public liability. If you're suggesting that the central bank should just retire much of the debt it holds, I fully agree!

Good post!

Ralph: "Ergo infrastructure should not be used to deal with recessions: other forms of spending should."

Did you say the bridge was needed? Are you arguing recession is a bad time to start building public infrastructure project? What is a better timing for long project - good times? Note that even "other forms of spending" will target to encourage private sector to commit long projects.

I think infrastructure tends to require a lot of planning and can last several years, infrastructure spending probably should be invariant to the business cycle.

Britonomist: I'm still not following. The CB could sell the bridge at some future time, just like it could sell the bonds at some future time. (Sure, one's a bit more marketable than the other, but the government would presumably buy it, if it thinks that X bridges should be publicly rather than privately owned, since this is just fiscal offset going in reverse gear.)

Would the central bank selling the bridge imply some increase in liabilities to the public, like selling bonds to the public would?

If the government buys the bridge from the central bank, in exchange for $1B in bonds, it's a wash. And that is exactly what would happen. (If the government wanted the bridge to be privatised, it would presumably have sold the bridge to the private sector in any case, if it had originally been bought by the government and not by the central bank).

Nick,

"2. Suppose the government of Canada normally buys 10 new Canadian bridges at $1 billion each. Then the Bank of Canada prints $3 billion, and instead of lending it to the government (buying government bonds) like it normally does, it buys 3 new Canadian bridges."

As a bridge builder, why would I settle for $1 billion for each bridge I sell? Why not $1 trillion, or a $ Gazillion. It's not like I am going to put the Bank of Canada out of business by raising my asking price. If I am one of many bridge builders (in competition) and I know the central bank is going to buy 3 new Canadian bridges, I get together with my (former) competition and explain it simply - we can beat our heads against each other or we can merge and charge whatever price we like.

I don't get it, why would the central bank need to sell the bridge back to government? Why can't it just transfer it for free (and yes the same can apply to bonds, the current problem is central banks have pledged they wont do that)?

Agree. This is a political argument, which really annoys us Paleo Keynesians. The assumption is that monetary policy = good, and fiscal policy = bad, so let's call fiscal policy monetary policy. But what can you say, we live in a world in which Roger Farmer thinks its OK for the CB to buy stocks but not for the government to buy bridges.

I think I get what you're trying to say though you don't really spell out what this has to do with Brits. Corbyn and "People's QE"?

Your point here seems not to be about a silly type of QE, but about a silly way of delegating fiscal authority. In both examples A and B, there is $3b of QE that's used to finance bridge-building. But in example A, fiscal authority is with government, whereas in example B, fiscal authority is with the central bank. That's the only difference. And I agree, giving fiscal authority to the central bank is silly. It's just not the right institution for the job.

It seems to me your point is to the Sumnerites and their idea that in a ZLB low NGDP growth situation the central bank should inflate by buying up non-traditional assets, eg corporate bonds. Which effectively turns the central bank into a fiscal authority that does fiscal policy through lending directly to the real economy on non-commercial terms. How that's supposed to be different from and better than the fiscal stimulus so dreaded by Sumnerites I don't quite follow.

But what Wren-Lewis is promoting is helicopter money, ie a regular fiscal expansion explicitly funded by monetary expansion. The relevant comparison would be, in example A the government builds 10 bridges in year 1 and 2, and the central bank does no QE, and in example B the government builds 10 bridges in year 1 and 13 bridges in year 2, and the central bank does $3b of QE.

The crucial difference from previous QE programs is the direct explicit link between the QE program and a fiscal expansion. That makes it a much more powerful, but also much more dangerous proposal, especially if the central bank lacks the independence to reject government pressure to continually increase monetary expansion to fund continually increasing fiscal spending. There is actually a record of that turning into repeated adding of 0s on the bills.

PS I meant to say, in example B the central bank does no QE in year 1 and $3b in year 2.

"I don't get it".

I think it is easy to get it if you exchange "the central bank" and "the government" to mean "the consolidated government". You can consolidate them as a thought experiment because the government is a sole owner and beneficiary of the central bank and its profits. Of course the nature of their asset ownership might be different as they have different mandates but with (silly) QE those mandates are now overlapping.

"Why can't it just transfer it for free". Yes it could but given above why would it change anything. It would be just a part of its profit remittance. You cannot create anything, apart of politics, by cooking the books between the government and the central bank.

Britonomist: If the central bank needs to reduce the money supply, to reduce inflation (or NGDP growth) it will need to sell an asset in exchange for that money it buys back.

rsj: "The assumption is that monetary policy = good, and fiscal policy = bad, so let's call fiscal policy monetary policy."

Maybe. Or maybe if the central bank does it it's good, and if the government does it it's bad?? Or maybe they are just conceptually confused, and don't understand the full implications of the fact that the government owns the central bank. My gut says the last.

I don't think Roger is saying the government shouldn't buy bridges. I think he's saying build bridges if you need more bridges, not if you need more AD.

Tom: yes, Corbyn's QE. Funny thing is, a couple of weeks back I wrote a post about some very obscure Canadian monetary cranks, without naming them, and the Brits thought I was talking about Corbyn's QE (to which my post applied equally well, but I hadn't seen the connection). So I never know whether to spell out who/what the post is about, or whether to leave it blank, so the reader can see applications that I have missed.

"Your point here seems not to be about a silly type of QE, but about a silly way of delegating fiscal authority....And I agree, giving fiscal authority to the central bank is silly. It's just not the right institution for the job."

Yep. Imagine the Bank of Canada trying to decide all the concrete (heh!) details about what sort of bridge etc. But my main point was even if the Bank of Canada got it right, it would have no effect, because the government would do the opposite. Even if all works perfectly, adding a second person to make the decisions already being made by one person won't have any effect.

The Summerite point is that with a sensible and credible NGDP level path target, a lot less QE would be needed. And you only need the central bank to (threaten to) buy other stuff if it runs out of government bonds to buy. Buying bridges would be for when the CB has already bought everything else.

*All* (permanent) increases in (central bank) money supply *must*, sooner or later, be spent by the government. Because the government owns the central bank. So all money is helicopter money, in that sense. The only questions are: when? and on what?

"PS I meant to say, in example B the central bank does no QE in year 1 and $3b in year 2."

Well, I was really comparing:a) $3B regular QE (buying bonds) : b) $3B silly QE (buying bridges instead of bonds)

Jussi,

Infrastructure investments last a very long time. They should be made if they look like they will pay for themselves over that “long time”. Thus the fact that we are currently in a recession (or not) is near irrelevant. I.e. there’s no harm in boosting infrastructure spending by a SMALL AMOUNT in a recession, but DOUBLING infrastructure spending in a recession would be crazy. Apart from anything else, in the case of the UK, the requisite skills aren’t there. See:

http://www.theguardian.com/business/2015/feb/10/uk-plumbers-builders-engineers-skill-crisis-economy

I am not a great fan of Jeremy Corbyn. However, you seem to have misunderstood the proposal. Let’s backtrack.

Jeremy Corbyn was elected leader of the Labour party in what was a major popular left-wing revolt against the previous 20 years of Labour party history. One of the policies put forward by Corbyn was ‘People’s QE’ which was an idea that originated from a left-wing economist called Richard Murphy.

After the election, a number of UK academic economists and financial journalists began to critique People’s QE and also to critique each other’s critiques of People’s QE. Unfortunately, few if any of them did any research on what the People’s QE proposal actually was. The result – they created even more confusion than normal!

Why is that important here? Well, you now seem to be doing the same thing.

Here is a quote from Richard Murphy in a reply to an article by a BBC journalist called Robert Peston:

“So let me be clear, at a rather boring technical level I agree with him (Peston). Which is why I have been, and Jeremy Corbyn has also been, very keen to make clear that the Bank of England will not be investing in roads, or houses, or green energy. It would in fact be buying bonds from a National Investment Bank which would, under government direction and subject to government guarantees, engage in such investment issues. All that the Bank of England would do would buy some new forms of bond: it would not manage a single project or decide upon any investments.”

http://www.taxresearch.org.uk/Blog/2015/08/12/robert-peston-on-peoples-qe/

And here is a link to a search for posts on People’s QE on Murphy’s site:

http://www.taxresearch.org.uk/Blog/?s=people%27s+qe&searchsubmit=

I would be interested in your thoughts on the ACTUAL proposal. For example, does the idea of having separate bonds for the NIB and other government activities make any difference? Is the ‘QE’ aspect of the proposal just a variant of helicopter money where the money goes to the NIB?

Jamie: But that's just the same. If the National Investment Bank is financed by the Bank of England, it's just like adding another department to the BoE. The UK government can already do investments. And when the BoE buys government bonds, it's already lending to the government's investment bank. The whole thing is a shell game.

" For example, does the idea of having separate bonds for the NIB and other government activities make any difference?"

No, not if both are guaranteed by the same entity. (It might make both a little less liquid, which would be a bad thing.)

" Is the ‘QE’ aspect of the proposal just a variant of helicopter money where the money goes to the NIB?"

Yes. But all (permanent) increases in money are always ultimately helicopter money anyway, if the govt owns the CB.

At best (and the analogy doesn't really work, because demographics are relatively predictable) it's like the Canada Pension Plan's assets being kept on a different set of books than the rest of the government's assets.

Or, if it's exactly the same people deciding how many bridges/sewers to buy, with some of the financing coming from the BoE and the rest of the financing coming from the govt, it is *exactly* the same as my first example, where *I* decide how many apples/bananas to buy, except the govt gives me $3 instead of 3 apples. Where that $3 came out of my own pocket.

What worries me is that many Brits won't see through the shell game.

"Infrastructure investments last a very long time. They should be made if they look like they will pay for themselves over that “long time”."

Yes, one should only start Net Present Value > 0 projects. But the timing of fiscal stimulus depends on how fast money gets spend - that being typically a span of few years on infrastructure projects.

@Nick Rowe
"it is *exactly* the same as my first example, where *I* decide how many apples/bananas to buy, except the govt gives me $3 instead of 3 apples. "

Clearly the difference is that I might have to borrow $3 to buy 3 apples (if I am short of cash), and so does the gov. if it pays for my apples through deficit financing.

The cost of borrowing is interest. On 3 apples it is not much, but on the UK's total debt, accumulated over time it comes to close to 3% of GDP.

If PQE buys apples, the financing cost of buying the apples is 0. The money is spent into the economy. So, it is cheaper for the government, as it saves interest over time.

(The saving of interest payments for the gov. is not why PQE is proposed in the UK, btw, it is proposed to provide a stimulus if times are bad - it could be used for providing insulation for badly insulated houses, for example, which could be done much quicker than a new bridge.

However, if all government expenditure was financed by PQE - again not a current proposal - it would in time allow the UK to save all its interest payments, which since 1950 have accumulated, so that they now make up 2/3 of UK's gov. debt)

https://radicaleconomicthought.wordpress.com/2015/09/24/pqe-how-to-cut-the-uks-debt-by-two-thirds/

Nick: You DID make a reference to Becker's Rotten Kid Theorem! (-;

@Usselmann:

> The money is spent into the economy. So, it is cheaper for the government, as it saves interest over time.

No, no it does not.

Remember that between these examples, we hold constant the size of the central bank's balance sheet, both in the moment and over time. Interest "saved" in one version of this story is interest paid to the central bank and remitted back to the government in the other. It's a wash.

For what you propose to have any effect, the "bank buys bridges" world would necessarily require that the central bank hold a larger asset position than in the status quo world. Achieving that necessarily requires a change in the central bank's mandate, but changing the mandate would achieve this difference of books without changing the type of assets purchased.

Matt U: "If PQE buys apples, the financing cost of buying the apples is 0. The money is spent into the economy. So, it is cheaper for the government, as it saves interest over time."

No. There is no effect on the government's interest cost of investment spending. Because you get exactly the same $7B of bonds *in public hands* in both cases. True, without PQE there is the interest on the extra $3B of bonds that the government has to pay the Bank of Canada. But since the government owns the Bank of Canada, the Bank of Canada gives all its profits to the government, so the government pays that interest to the Bank of Canada out of one pocket, and gets it back in the other pocket.

See my old post on this. (That post was supposed to be about a group of Canadian monetary cranks, but it applies equally well to "PQE".

"(The saving of interest payments for the gov. is not why PQE is proposed in the UK, btw, it is proposed to provide a stimulus if times are bad - it could be used for providing insulation for badly insulated houses, for example, which could be done much quicker than a new bridge."

Leaving aside the question of whether fiscal policy does or does not provide "stimulus", and just assuming it does, PQE would provide zero stimulus, because (as I argue above) it would be 100% offset by regular fiscal policy.

kevin: Ooops! What was that Holy Grail sketch?

Here's how I see it. People's QE originally came up as an idea as a policy alternative to QE during a recession at the Zero Lower Bound.

Bankers' QE was unpopular and not that effective. Instead of giving printed money to the banks to invest, give it to the government to invest.

You would sidestep the austerians and deficit fetishists who obsess about the budget. There wouldn't be more debt. All opponents could do is warn of inflation - which is what they did with Corbyn - but that's part of the point. During a recession, you're trying to close the output gap and get back to full employment and the inflation target whereby the central bank would raise rates to fight inflation. People's QE is money-financed government spending and a way to fight recessions without worsening the debt position.

Simon Wren-Lewis and others say you could just do a helicopter drop and give people newly printed money.

David Beckworth has been making a point about the importance of the *permanent" increase in the monetary base when doing QE. QE wasn't that effecive because of the inflation target and promise of price stability. Same thing would hold for People's QE. A more effective form would involve linking it to the NGDP path trend level. Inflation would go above the old target in order to get catch-up growth.

As it was QE didn't provide catch up growth, it just helped closed the output gap, which is partly why it was disappointing.

"If you think the government should build more bridges, then just say so. And if you think you are going to be the next PM, then just say you are going to do it when you are PM. It's really silly to say that when you are PM you will order your central bank to do your shopping for bridges for you."

If the PM orders the central bank provides newly printed money to finance the building of bridges during a recession when short rates are at the zero lower bound, then the budget deficit won't blow up.

What happened in the U.S. was that after the President ordered Congress to build more bridges, the government deficit went way up. Conservatives harped on it day in and day out. Then before the recovery was complete; before we were at full employment and the inflation target, the government turned to austerity. It ordered less bridges than it otherwise would have because of concerns over the budget deficit.
This hurt the recovery. A "Silly QE" would have sidestepped deficit concerns. Bridges would have continued to be built until the recovery was complete.

No what was the central bank doing during all of this? Well maybe your point is that none of this matters because the central bank can compensate for fiscal austerity. But in reality it doesn't because of politics. Bernanke constantly said fiscal austerity was hurting the recovery probably because there was only so much stimulus the central bank was willing to do for fear of a backlash. The central bank was already under pressure for its unconventional policies. And it was under pressure and subject to criticism because its Bankers' QE didn't seem to work very well.

Peter K: " Instead of giving printed money to the banks to invest, give it to the government to invest."

That is how it is designed to look. And, to be charitable, maybe the proponents of "PQE" really did think that is what it meant. But it's just an accounting shell-game. The government's budget constraint is exactly the same either way.

Peter K: "If the PM orders the central bank provides newly printed money to finance the building of bridges during a recession when short rates are at the zero lower bound, then the budget deficit won't blow up."

Nope. It's exactly the same budget deficit either way, except for creative accounting. If you consolidate the government's and central bank's balance sheets (since the government owns the central bank), you can see right through it.

"Britonomist: If the central bank needs to reduce the money supply, to reduce inflation (or NGDP growth) it will need to sell an asset in exchange for that money it buys back."

Yes and the central bank already has plenty of assets to do this. There is a difference between saying "we might sell back these bonds and add to the public debt IF inflation starts to get too high", which is a conditional statement and "we absolutely definitely will give these bonds back to the public no matter what, regardless of inflation, this is definitely only temporary, you have more public debt whether you like it or not, deal with it!" which is what central banks are currently actually doing. Why can you not see that!?

There is absolutely no reason right now for an investor to think bonds held by the central bank are not still a liability for the public, none at all, unless the central bank explicitly states it will retire them or not sell them back to the public.

"Nope. It's exactly the same budget deficit either way, except for creative accounting."

No it's not the same. In this case, spending increases, borrowing remains the same. If on the other hand, the government borrows to spend, and the central bank buys those bonds, everyone in the public will consider it both a spending increase AND a borrowing increase, it doesn't matter if a few economists and accountants consider those bonds bought by the central bank 'a wash' and the deficit to have not increased; NOBODY IN THE PUBLIC WILL THINK THAT, the press will certainly not report it that way - as far as the public is concerned, the debt has increased no matter who holds it, any ricardian equivalence effects will still hold. Signalling matters, expectations matter. Unless the central bank comes out and says "we're not necessarily going to sell these back to the public", the public will consider it more debt, the government will consider it more debt, period.

@nick rowe

"PQE would provide zero stimulus, because (as I argue above) it would be 100% offset by regular fiscal policy."

This fiscal offset idea: I think you mean that the interest provided by the bridges with bond finance would not be available by the private sector, if bridges were financed by PQE. (PQE is supposed to work by a National Investment Bank selling bonds to the Bank of England. The bonds pay interest to the Bank of England, but as they are both government agencies, money form one governemnt pocket goes into the other)

So the private sector loses the income it would have had, had it lent the money to the government. That interest income is the now lost by the private sector.

That is exactly the same as taxing bondholders who bought bonds to finance bridge construction, the exact amount of interest the bonds pay. Here the interest income from bonds is also lost by the purchasers of the bonds. (Which of course is never done)

So you would argue there is fiscal offset, as not giving the private sector bond income (through PQE) is the same as taxing the bond income.

That is true, however, the whole of the population (tax base) is of course normally taxed to pay interest income on government bonds. So it boils down to that PQE is really a redistributive policy. It redistributes income from the rich (that is the section of the population which normally hold bonds) and makes it cheaper for the poor to have bridges.

I agree it is a a fiscal offset, (if that is what is meant here) but the effect of the poorer section of the population having more money, is of course stimulative to the economy. As that is the part of the population with a higher propensity to consume.

Matt: "This fiscal offset idea: I think you mean that the interest provided by the bridges with bond finance would not be available by the private sector, if bridges were financed by PQE."

No. The private sector would earn the same amount of interest, whether the bridges were financed by issuing bonds or by PQE. Because with regular QE, the BoE would have bought $3B worth of bonds, leaving only $7B bonds in public hands in either case.

Britonomist: why should (expected) future money supply growth be in any way affected by the choice between PQE and regular QE? The central bank is lending the government the same $3B in either case.

Unless there's some purely symbolic/signalling effect of buying bridges instead of regular bonds? "Hey", says the BoE "I wouldn't be buying bridges if I planned to sell them again soon, because bridges are so damned illiquid!" (like burning your boats?)

Just a disclaimer I consider actual PQE as proposed by Murphy et al a strange and probably unnecessary policy, so I can't comment on that specific policy. But regarding this more abstract case:

I'm not considering future money supply growth expectations, I don't regard those as particularly relevant, since they probably will be the same across both scenarios. However, I do consider that the expected path of fiscal policy will be different - if the public perceive there to be more debt, the public will expect more fiscal consolidation. I believe the public & government will not lower their perception of debt much just because the central bank is holding government liabilities temporarily. I do believe they would lower their perception of debt if the central bank buys bridges - I can't see them considering those bridges a liability in the same way bonds ae.

"I don't think Roger is saying the government shouldn't buy bridges. I think he's saying build bridges if you need more bridges, not if you need more AD.
"

I could just as well say -- invest more if you need to invest more, but not if you need more AD.

Needing more AD pre-supposes the private sector's decision making process about what investments should be made isn't working. The patient is sick, not healthy, and so the cures aren't the same medicine we would give to a healthy patient. Lowering interest rates channels resources to those who have collateral -- primarily land -- but not those without. Building bridges channel resources to those who are in the bridge-building business. I don't think that these are less deserving than landholders. In fact, I'm not sure why one would be worse than the other, or why the economy would be more distorted by systemically subsidizing real estate versus infrastructure investment.

Thanks for the reply, Nick.

Re: the shell game nature of many QE-related proposals, there’s an important reason why that intellectual path is so often taken. The traditional “Chinese wall” between fiscal and monetary policy is a powerful taboo, established after harsh experiences. There are many movements that want by some route or another to break that taboo and go through the Chinese wall, but few that are willing to plainly admit that that’s what they’re trying to do.

NGDP targeting is all about that. Its sole attraction is that it provides ideological support for breaking through the Chinese wall separating fiscal and monetary policy when growth is still too low at the ZLB.

And my point about Sumnerites is that they don’t admit that’s what they’re doing: they purport to loathe fiscal stimulus in any situation. But the only important proposal they’re making is that when growth is still too law at the ZLB, the central bank should provide fiscal stimulus through a kind of monetary financing of industrial policy, by communicating to the market the prices and conditions at which it’s willing to buy new issues of corporate bonds. Such signaling can’t work without genuine, specific commitment.

Re: the fiscal authority reducing spending to the extent the central bank takes on quasi-fiscal authority, that depends on the political system. In the US with the Congress controlling the budget and a more presidentially influenced, bipartisan/independent central bank, not necessarily so. The taboo against central bank-run fiscal policy is strong in the US but if the polarized D/R presidential/Congress dichotomy persists, it’s possible the Fed could consider some kind of quasi-fiscal stimulus in the next recession, and if that happened I don’t see why to expect Congress to react by cutting its budget, except perhaps over a long run. These sorts of policies are after all meant to be short-term stimuli not permanent. Arguably buying mortgage bonds was a kind of Fed step into industrial policy, though the Fed only slightly nudged prices (mortgage rates) while leaving it to government to set lending conditions. In a hypothetical Corbyn-led UK with a Corbynist BoE, it seems to me not at all hard to imagine a “people’s QE” funding increased overall transfer payments to households.

Anyway, “people’s QE” aka helicopter money does not require any transfer of fiscal authority to the central bank. If the government boosts spending by 3% of GDP and the central bank boosts QE by 3% of GDP at the same time, that’s helicopter money. QE aka monetary financing is just one of the ways fiscal expansion can be financed.

Re: “*All* (permanent) increases in (central bank) money supply *must*, sooner or later, be spent by the government. Because the government owns the central bank.” Certainly currency is a public liability and not really very different from a short-term government bond. And issuance is usually by central bank purchase of financial assets, which is a kind of spending by a public authority. But not all spending is final spending or spending in the real economy, and not all public authority is appropriately called government in all contexts. Also central banks issue currency by making loans, usually to commercial banks.

Re: “So all money is helicopter money, in that sense. The only questions are: when? and on what?” No, I disagree. What makes helicopter money different and powerful is the fiscal expansion. When monetary financing merely substitutes for debt financing to maintain a stable rate of fiscal spending, the only thing that changes is that the private sector accumulates currency instead of accumulating sovereign bonds. That private accumulation of currency rather than bonds does not imply any greater later fiscal spending.

Britonomist: I'm not sure I understand you. But if you are saying "Sure, it's an accounting shell game, but that accounting shell game might fool the public, and government, into doing more (total) fiscal expansion than they otherwise would" then I get your point.

Tom: I'm not sure you are getting my point. And I'm certainly not getting yours.

Let me try one more time. The advocates of PQE say that if the CB gives the government an interest-free loan that will lead to a bigger increase in G than the same $ amount of regular QE. And I'm saying no it won't, because regular QE is also giving the government an interest-free loan. The only way PQE could work is by fooling the government, or the people, if they can't see through the accounting shell game.

Where's JKH, when I need him?!

Well I definitely understand and agree with this new point you've just made. There is clearly no difference whatsoever between a zero-interest CB loan to government and a 100% interest CB loan to government, so long as the CB pays its profits to government. Anyone who suggests there's a difference isn't thinking it through.

I thought you were arguing that helicopter money spent directly by a central bank wouldn't provide any stimulus because the government would cut back on its spending by an equal amount. Which I think generally would not happen, at least not in the short run. And anyway there's no reason why "people's QE" can't be spent by government.

Let me try to clarify using a similar method to yours.

In year 1, the government raises $4 trillion of revenues, spends $5 trillion, issues $1 trillion of bonds and the central bank buys none.

In year 2:

A: $4t revs, $5t spend, $1t bond issues, zero QE.
B: $4t revs, $5t spend, $1t bond issues, $1t QE.
C: $4t revs, $6t spend, $2t bond issues, zero QE.
D: $4t revs, $6t spend, $2t bond issues, $1t QE.

A is no change. B is $1t of QE. C is $1t of traditional fiscal stimulus. D is $1t of helicopter money.

Say that in year 1 market rates on overnight interbank loans and short-dated sovereign bonds were 5bs and IOR was 5bps, and latter isn’t changed in year 2.

By far the most important difference among these is that A and B provide no fiscal stimulus while C and D both provide $1t of fiscal stimulus. Fiscal stimulus directly boosts spending and nominal output, and depending on the extent of slack capacity also boosts inflation and interest rates.

A and B are nearly the same, and C and D are nearly the same except:

- B and D give a shorter-dated mix of public liabilities in private hands (more currency, less long bonds). This avoids the crowding-out effect of sovereign long bond issuance. Or put another way, monetary financing when substituting for long-dated debt financing creates “reach for yield.”

- B and D create to some extent a hot potato effect on currency, suppressing its value relative to other currencies and financial assets. But this seems to vary by context and is sometimes hard to distinguish. (EG Japan QE generated lots of obvious FX pressure, US QE much less; QE effects on bond and equity prices are debatable).

- D (helicopter money) is not restrained by market discipline, and the habitual practice of it strongly correlates with chronic high inflation.

D and B both feature monetary financing and deficit spending, but B merely substitutes monetary financing to maintain a stable level of deficit spending, whereas D adds monetary financing and fiscal spending. That's why helicopter money is much more powerful than QE alone.

Whether or not Britain needs fiscal stimulus, I would be uncomfortable with a PM candidate arguing that the differences between a standard fiscal stimulus (C) and "people's QE" (D) are important.

"Britonomist: I'm not sure I understand you. But if you are saying "Sure, it's an accounting shell game, but that accounting shell game might fool the public, and government, into doing more (total) fiscal expansion than they otherwise would" then I get your point."

That's kind of my point, but I actually think it's QE which is more of the shell game, whereas a direct helicopter drop - as in literally just printing money and buying new stuff or financing tax breaks - is much less ambiguous and much easier for the public to understand. PQE seems shell-game-ish however, which is why I'm skeptical of it.

Tom: "I thought you were arguing that helicopter money spent directly by a central bank wouldn't provide any stimulus because the government would cut back on its spending by an equal amount."

I am arguing that. And I am arguing that precisely because "There is clearly no difference whatsoever between a zero-interest CB loan to government and a 100% interest CB loan to government, so long as the CB pays its profits to government."

It's just like my first example, where I buy 3 apples less because the government taxes me $3 and gives me 3 apples.

The central bank takes away the government's interest-free loan, and invests it in 3 bridges, so the government invests in 3 fewer bridges.

The problem that "people's QE seems to be trying to address is when the government department in charge of buying bridges, is being told not to do so because during recessions we need to reduce deficits ("austerity". If government was doing its job of investing in activities with present costs and future benefits according to an NPV criterion, there there would be no reason for the central bank to do "people's QE." [This does not mean that "people's QE" is the best or even a good solution to misguided fiscal "austerity."] This is sort of the reverse of the problem that one might need more ST government paper than would be produced by the NPV rule if the central bank were constrained to use no instrument except purchase of ST government paper.

Both fiscal policy and monetary policy must reckon with the other not being optimal and that alters what is optimal in its case.

Ok, I got you Nick.

I think your argument makes some sense for a parliamentary system government with a captive central bank. If there's one united group of people who control both the government and the central bank, and that group's deciding how much money to create and how much to spend. But if some group of people are wacky enough to give the central bank fiscal tasks when they control the fiscal authority anyway, I wouldn't dare predict what they would do with their control over the fiscal authority.

And in other systems I think a central bank decision to directly lend or spend into the real economy likely wouldn't provoke any compensating fiscal tightening, at least not in the short term. For example in the US it's currently nearly impossible to pass a sequester mid-year. And the short term seems to me the only point anyway. There's more reason for central banks to do such things when power is fragmented, eg the FOMC might see itself as rescuing the economy from a Congress of nutters.

I think this is all far-fetched for now, but if rates are still very low going into the next recession, these sorts of ideas might muster political weight.

Cheers and all that.

ThomasH: [For some reason your comment got caught by our spam filter, and I had to fish it out. This happens sometimes.]

Monetary policy nowadays takes it for granted that fiscal policy may or may not be optimal. The job of the Bank of Canada is to target 2% inflation, regardless of what fiscal policy does, so it makes adjustments accordingly. And we hold it accountable for doing so. We expect it to know it lives in a second-best world. And the fiscal authority is supposed to be accountable for the right level and composition of spending and taxes (much harder to define) but not accountable for keeping inflation on target. A clear division of accountability.

Tom: glad you got it! Even if we still disagree. (I just can't have written my post as clearly as I thought I did.)

[Off-topic: just re-reading your very thorough Malaysian Air post. Time for an update?]

Peter K: " Instead of giving printed money to the banks to invest, give it to the government to invest."

That is how it is designed to look. And, to be charitable, maybe the proponents of "PQE" really did think that is what it meant. But it's just an accounting shell-game. The government's budget constraint is exactly the same either way.

Posted by: Nick Rowe | October 14, 2015 at 10:48 AM

Peter K: "If the PM orders the central bank provides newly printed money to finance the building of bridges during a recession when short rates are at the zero lower bound, then the budget deficit won't blow up."

Nope. It's exactly the same budget deficit either way, except for creative accounting. If you consolidate the government's and central bank's balance sheets (since the government owns the central bank), you can see right through it.

----------------------------------

So with quantitative easing ("Bankers' QE") it's also creative accounting and the budget is consolidated? So the central bank doesn't "create money" out of thin air by printing it? What am I missing?

Are you hung up on the inflation target and a permanent increase in the monetary base? That's seems to be the basis of David Beckworth's oppostion. Again I don't see the difference between QE and People's QE except that in the latter the government directs investment.

Peter K: "If the PM orders the central bank provides newly printed money to finance the building of bridges during a recession when short rates are at the zero lower bound, then the budget deficit won't blow up."

Nope. It's exactly the same budget deficit either way, except for creative accounting.

----------

When the Fed did QE1-3, it didn't blow up the deficit. It added to the Fed's balance sheet. But the balance sheet really doesn't matter. They can just hold on those securities forever and never sell them.

Do you not agree? It is all just creative accounting? If not, how is People's QE any different. And don't say People's QE is different just because of the way *you* alone characterize it in a straw man argument fashion. It should be seen as analogous to QE.

"Monetary policy nowadays takes it for granted that fiscal policy may or may not be optimal. The job of the Bank of Canada is to target 2% inflation, regardless of what fiscal policy does, so it makes adjustments accordingly. And we hold it accountable for doing so. We expect it to know it lives in a second-best world. And the fiscal authority is supposed to be accountable for the right level and composition of spending and taxes (much harder to define) but not accountable for keeping inflation on target. A clear division of accountability."

In the U.S. and U.K. that's not how it has been working. The central banks have been forced to compensate for fiscal austerity and used QE to target 2 percent inflation. But the recoveries were very disappointing.

So next time their economies have a recession at the Zero Lower Bound, we could repeat the experience with QE and disappointing recoveries. Or possibly the could do more, better QE to bring about a swifter recovery. Or they could try something new like People's QE.

In the U.S., People's QE would create stimulus and help bring inflation back to the 2 percent target more quickly. It would go on the Fed's balance sheet and not add to the budget deficit.

And so when conservatives call for a reduction in the deficit, there would be less of a deficit to reduce and the economy would be hit by less fiscal austerity as the Fed is trying to hit its 2 percent inflation target.

It does matter to the Fed how much fiscal austerity Congress is throwing at it. Bernanke regularly complained about it. Because of politics, the Fed wants to do least amount of QE possible. They want out of the uncoventional policy and ZIRP businesses for they create controversy for the Fed and threaten its independence.

I do believe "Silly QE" would be more stimulative and bring about a swifter economy recovery than Bankers' QE. It is more direct at creating AD and jobs. Once the recovery is over and the Fed has hit its 2 percent inflation target, the fiscal authorities can cross off the investment projects that "Silly QE" has taken care of with the central banks' newly printed money.

PeterK: "So with quantitative easing ("Bankers' QE") it's also creative accounting and the budget is consolidated? So the central bank doesn't "create money" out of thin air by printing it?"

I never said that. I would say the opposite. The central bank can and does create money by printing it. (And if we consolidate the balance sheets, then the government/central bank does create money by printing it.) And that is equally true in both regular QE and silly QE.

"What am I missing?"

I have no idea. I do not understand you; and I'm pretty sure you do not understand me.

I see PQE as (potentially) a way around constraints on the central bank and fiscal policy. Whether this makes sense in the UK or Canadian political system I do not know. I'm looking at a system constrained by political limits on the amount of QE that can be done* and political constraints on the amount of investment in NPV greater than zero projects that can get done because politicians think (or at least pretend to think when the other party holds the Presidency) that debt, or increases in debt, or rates of increase in debt are always and everywhere bad.

In that kind of system letting the Fed buy the bonds of a "National Investment Bank" (especially if the Bank is instructed to cost inputs at marginal social cost not market prices) MIGHT be an end run around the constraints on monetary and fiscal policy I see. But if I had one wish it would probably be for ngdp targeting rather than PQE + NIB.

* It is not clear how independent this constraint is from the constraint on having the inflation rate be high enough above 2% for long enough to get the price level back on its pre-crisis trend.

PeterK:

"as analogous to QE".. "a swifter economy recovery"

Bankers QE and silly QE both have similar monetary effect - and PQE might be better in other ways, no silliness there. So PQE might have a swiftier effect and all. But that is not the point the post made, PQE might still be silly, do you see why?

It is surely silly to send wife to buy a car just because she is able to fund it, wouldn't it be better to do it yourself (assuming you know more about cars) and ask your wife just to fund it? Wouldn't it in similar fashion be silly to make the Central bank "to hammer the screw" if we can do the very same PQE without it by using a screwdriver? Nick argues that yes we can do PQE without using (directly) the central bank.

So, in other words, "bankers QE", buying treasuries is what the Central Bank is setup for. PQE not so much as buying bridges is usually done by an other agency. Yes, it still not silly if the Central Bank buying would create something special which couldn't be pursued by other means. But as Nick explained that is not the case. The Central Bank doesn't create anything special by buying directly the bridges comparing the case it buys the bonds funding the bridges and the buying is done by the other agency.

Nick Rowe:

"PeterK: "So with quantitative easing ("Bankers' QE") it's also creative accounting and the budget is consolidated? So the central bank doesn't "create money" out of thin air by printing it?"

I never said that. I would say the opposite. The central bank can and does create money by printing it. (And if we consolidate the balance sheets, then the government/central bank does create money by printing it.) And that is equally true in both regular QE and silly QE."

Okay I think I know what you're saying. For both Bankers' QE and Silly QE, the consolidated government/central bank balance sheets are constrained.

But by what are they constrained? Inflation? Too much debt? What would help illustrate the case is to compare two scenarios: one with QE and one with People's QE and describe how the consolidated government/central bank balance sheets differ and how the balance sheet differ separately.

Jussi, proponents of People's QE have said that the Central Bank would help finance a National investment bank which would know what to do.

"The Central Bank doesn't create anything special by buying directly the bridges comparing the case it buys the bonds funding the bridges and the buying is done by the other agency."

In one case, the governments' budget deficit increases. It will need to raise taxes after the recession is over to pay for the bridges. In the other case the central bank funded the buying of the bridges so the budget deficit wouldn't go up. Nick insists the consolidated central bank/government budget is still constrained so it doesn't matter that the government budget deficit doesn't go up.

But it does matter because of politics. Maybe Nick is saying they could just get the same recovery by doing a larger QE without PQE and no one is disputing that. But with a PQE, projects and jobs and income are created directly instead of indirectly by the banks.

If the U.S. had done "Silly QE" instead of Bankers' QE, I bet the recovery would have been much quicker and it would have cost the "consolidated budget" less. There would have been less fiscal austerity forced on the economy, because the budget deficit would have been smaller.

I'm pretty smart and if Rowe can't explain why PQ is silly in simple terms then I believe the problem is with his explanation and his conception of the issue.

Peter K: "Okay I think I know what you're saying. For both Bankers' QE and Silly QE, the consolidated government/central bank balance sheets are constrained."

Not quite. For both Regular QE and Silly QE, the consolidated government/central bank balance sheets are *exactly the same*.

"Maybe Nick is saying they could just get the same recovery by doing a larger QE without PQE ...."

No, that is not what I'm saying. Since regular QE and silly QE are exactly the same, and will lead to exactly the same change in G (if any), you will get exactly the same recovery in both cases.

Nick,

I haven't read all of the above comments, so maybe this is old news. But, I think here is what is confusing people.

1) By calling it "Silly QE," you are making it sound like QE has no effect on real output. And since your readers know plenty of models where QE has real effects, they are saying you must be doing something fishy.

2) But in reality, you are saying that having the central bank directly buy publicly needed goods has no effect on real output *relative to a standard QE program* where the central bank buys bonds and the government runs a deficit to invest in the public goods.

Is that right?

Supposing my first comment is correct, Nick, then the way someone might justify "People's QE" is like this:

"Sure Rowe, I agree with you that there is no significant difference between normal QE + bigger fiscal deficit versus People's QE + smaller deficit. However, the public is stupid. They get worried when the government issues another $500 billion in bonds, even if they are absorbed by the central bank. So that's why we can push through People's QE more easily than what is, I grant you, an equivalent program."

Bob: you are right. But I am saying more than that. I am saying that silly QE, relative to regular QE, will not cause G to be higher.

And someone *might* try that counterargument, but if so, silly QE only works because it fools people into doing something they would not want to do. Simply lying about the deficit numbers would have the same effect.

Calling silly QE QE is silly. The CB is actually doing the same thing in both cases. What's different is that with an NIB, fiscal policy decisions are being partly outsourced to another (possibly more competent) entity, with the side effect that they also don't show up on government books.

But, as rsj rightly points out, seeing as fiscal policy is considered bad while monetary policy is good, maybe keeping the public eye on a non-existent monetary distinction to obfuscate from the actual accounting gimmick might not be so silly after all. Also an NIB could be partly equity financed and more easily dissolved if its investments go belly up. An NIB would make the distinction between government investment and transfer payments clearer.

And whether G is higher vs. the current arrangement seems to me less of a function of whose books the investments show up on as of whether or not the investments are made or not. Presumably (I haven't read the proposal) the NIB would be mandated to invest some percentage of GDP, whereas with ordinary QE, commercial banks are not mandated to lend out the reserves they obtain.

In all, an NIB arrangement seems a sensible way to marry classical banking with non-cyclical / long term government investment goals while shielding it a: from day to day politics but b: also from the commercial banking system. And with that, the functions of the CB as government (investment) banker and of manager of the private banking system become easier to distinguish, too.

PQE may not actually be distinguishable from ordinary QE - from the CB's point of view, it's rearranging the deck chairs. But it is a sensible rearrangement nonetheless, at least on paper (and assuming you're not idiologically wedded to free markets).

Nick, let's put aside economic logic for a moment. Cameron is trying to run away from austerity but the Brits now believe the lies of austerity (because he kept repeating them.)
Therefore Cameron is looking for a way to have a more expansive government while keeping a fig leaf that he is still all about austerity.

The mess is surely all about Cameron not having the guts to say 'Gordon Brown was right all along.'

Chris: If we want to draw a line in the sand, to protect us against our own profligacy, then a debt/GDP target would be a sensible line to draw, even though any particular line would be arbitrary, to some extent. "Flexible debt/GDP targeting" (budget to bring the debt/GDP to x% over h years) would be the counterpart to flexible price level path targeting. If Gordon Brown had brought the debt/GDP ratio down below target in the good years, it would have been much easier to defend looser fiscal policy in the bad years. Drunks daren't touch a drop, for fear they will fall off the wagon again.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad