« Health Spending Numbers: An Update on the Long-Term | Main | Accounting for central bank profits »


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

Who gets the helicopter money? Who decides who gets what?

Your "back-of-the-envelope calculation" completely ignores the property and ownership implications of money.

Money is property, just like your house and your shares of a company are property. We can logically relate ownership of money to ownership of shares in a company. Possession of money becomes equivalent to owning shares in a company, only now the company is the national economy.

In companies, it is common to give shares to the CEO as an incentive to grow the company. Surprisingly, if the CEO was given NEW shares in the ratio of 5% of the outstanding shares every year , after about 14 years the CEO would own over one-half the company.

Now we have an illustration of the danger of helicopter money. When the CB prints new money every year, it is a transfer of ownership from existing owners to whomever government decides to award with the benefits of newly generated money. Obviously, some will benefit but always at the expense of others.

Avon: The (Federal) government decides who gets it (or what it's spent on). Just like normal, when the government decides how to spend the profits it receives from the Bank of Canada (or uses it to keep taxes lower or transfer payments higher than they otherwise would be).

Roger. Take a simple case: the government/central bank prints enough new money to double the money supply, and gives the new money to the tallest half of the population. Assume the price level doubles as a result, but real GDP (etc) stays the same (I'm assuming money neutrality to keep it simple). In the new equilibrium, there has been a transfer of wealth from short people to tall people. We call this "the inflation tax". If the original money supply was $100 per person, it's as if the government took $50 from each short person and gave it to a tall person.

But if money is (short run) non-neutral, real GDP will increase, and both short and tall may be richer.

Just to be clear, Nick, you support the idea that Parliament should direct the central bank to give money to people who the government likes?

Personally I think all monetary policy should be conducted this way. It seems arbitrary to me that mainstream monetary policy is conducted through the banking system, whereby (at least in some countries) this just seems to spike the mortgage market rather than other industries, yet helicopter money is about "the government decid[ing] how to spend the profits it receives from the Bank of Canada". All monetary policy involves this political choice, so let the CB get on with determining the money supply, and let the elected government decide where it should go. Always.

Just to be clear Avon: stop being an ass.

Alex: Groan.

Ours the task eternal.


Nick, While you are thinking about the mare's nest, help me with this paragraph:

"As we have made clear, in the market for bank reserves where interest rates are actually determined, there is no such thing as a well-behaved downward-sloping money demand function. Unless interest is paid on excess reserves, the central bank must supply the specific amount of reserves banks demand – no more and no less. Moreover, cash does not influence the setting of the interest rate. In real life, central banks meet entirely passively the public’s demand for cash (e.g. Grenville 2013). If they did not, either the amount in excess of desired balances would be converted into bank deposits and then switched by banks into excess reserves, or it would fall short of the demand, frustrating the public, which would presumably turn to alternative means of payment."

Isn't the normal assumption that when the central bank doesn't supply enough base money to meet demand, in the short run interest rates rise high enough so that the amount of cash the public wants to hold again equals the supply? Then in the long run prices adjust downward to restore equilibrium. What am I missing? Why can't central banks do an exogenous increase or decrease in the base, when there is no IOR?

> So helicopter money finances a deficit of 0.25% of NGDP every year. It's small beer, but a continuous flow of small beer. It's normal. This is roughly what happens every year on average in Canada, for example.

I don't think this is what happens on average in Canada.

From the BoC year-end report, the Bank's "comprehensive income" was about $1.2bn, against an approximately $2tn NGDP. This gives a money-financed deficit of roughly 0.06% of NGDP, or roughly one-quarter what you back-of-the-envelope estimate. Even including the bank's own expenses as part of this deficit gives $1.6bn of mostly-seigniorage revenue, for money-financing of at most 0.08% NGDP.


You misunderstood me. I'm not trying to be an ass. You are suggesting that the central bank should become a political actor. I cannot understand how some economists miss that if you make the central bank a political actor then the bank cannot be independent. Lobbying at the central bank itself will become a huge industry. Giving money to people you like is an intensely political decision. That power must come with accountability to the voters. That's why we elect the people who set the budget. Our democratic institutions will suffer if unelected central bankers start making political decisions.


"to get my brain working again, while I gather my strength to disentangle this mare's nest"

I posted this the day before the nest:


Similar theme, different style.

I know you're a speed reader, so there may be some points there that will help with synapse refiring. Hopefully that, rather than a severe headache.

if you assume a 5% currency/Nominal GDP ratio then you get 5% NGDP growth if you increase the base by 0.25% no matter if you use OMO or HM to achieve this. And it makes no difference if the CB prints up the money and gives it to the govt (who as a result create less bonds that year) or buys govt bonds directly - you end up with the same level of govt debt held by the public in both cases.

However (still assuming a 5% currency/Nominal GDP ratio) if the govt decides, after getting additional HM , to increase the deficit , then you end up with NGDP the same but govt debt higher than if OMO alone was used.

If you relax the 5% currency/Nominal GDP ratio and assume the ratio goes up when people are given additional bonds (because they feel richer) and down when the CB does OMO (as this increases the money supply but lowers velocity) then things change a bit. In the short-term you have to lower interest rates with OMO compared to HM, especially if the govt increases its deficit after the HM starts.

I think in the long term however these differences go away (more or less). For a given level of govt deficit the amount of bonds held by the CB when it has been using OMO to increase the money supply will save the govt from paying the same amount of interest to the public as it would equally save by not having to issue them in the first place with HM.

Avon, but Central Banks already are political actors. They control monetary policy, and it is a political and moral choice right now to do it via the banking system, which has knock on effects for different industries. The Central Bank could just as easily give money directly to consumers, or fund investment bonds which citizens can use as claims on potential future revenues of companies they like. Both those choices would potentially structure the economy differently, so the current set-up is not apolitical.

As for independence, this need not be compromised any more than it is right now. Elected government sets the target (NGDP level growing at 5/6%, inflation of 2/3%, unemployment of 4/5%, whatever), as usual in many countries. Central Bank determines how much "helicopter money" it thinks the economy will need over the next quarter to sustain that target in the medium horizon. Central Bank "gives" the elected Treasury that money. Treasury sets up a predetermined formula with the approval of Parliament for how much helicopter money will go to what function (20% mailed to households, 10% income tax cut, 20% to business investment bonds, 20% to public sector investment, 10% to science research, 20% into the banking system, for example). The Treasury "sends" the money in those predetermined proportions to the various sectors of the economy. Iterate.

The only lobbying that really comes in here is to Parliament to get them to change that formula. But that's okay, the job of monetary policy is to keep the economy ticking over, the job of politics is to say what sort of economy we want. And that requires political decisions on what areas monetary policy should affect the most. So that lobbying is just the democratic accountability which ought to be there now, but it isn't. Helicopter money puts this out there transparently.

No need for you to do any thinking, Nick. I've done all the thinking here:


Is this a good way to look at it?

Fiscal Policy = Selling people bonds, then giving them back the money back = giving bonds away
Helicopter Policy = giving money away
Monetary Policy = Adjusting the ratio of bonds to money in the economy.

Monetary policy can always be used at any time to convert past fiscal policy into helicopter policy and vice versa.

To get a consistent 5% annual increase in NGDP (assuming on average a 5% currency/Nominal GDP ratio) you will need the monetary base to increase by 0.25% of NGDP a year. This can be done either via direct helicopter money, or by fiscal policy , that the CB then converts into helicopter money via the use of monetary policy.

So to maintain a smooth positive growth path for NGDP sufficient fiscal/helicopter policy will be needed, but it doesn't matter which , since monetary policy can always be used to determine how much of the combined value of both policies is held by the public as bonds v money.

Majro: (JKH should chime in on this one). There are two ways to do the accounting, and I have glossed over this:

1. If the BoC prints $100 in 2016, we call that $100 profit in 2016 (ignoring expenses). That's the way I'm doing the accounting.

2. If the BoC prints $100 in 2016, and buys (say) a perpetual bond that yields (say) 5% interest, and pays $5 per year, we call that $5 per year profit for every year that follows. That's the way the BoC does the accounting.

(I forget the names of those two different accounting methods.)

Since the present value of $5 per year forever at 5% interest equals $100, we can reconcile the two methods. And in a steady state where inflation and real GDP and interest rates are constant we get exactly the same answer either way. But in a growing economy the second method gives smaller profits as a % of NGDP. Plus there are fluctuations as well.

Scott: you are missing nothing (much).

When I read their words "... the central bank must supply the specific amount of reserves banks demand – no more and no less." I reach for my shovel. I fail ECON 1000 students for writing gibberish like that. There's a supply *curve* (or function), and that curve is not the same as *quantity supplied*, and the CB does not make that curve perfectly elastic at an exogenously-chosen interest rate, but adjusts that interest rate as a function of what is happening in the economy, which depends, inter alia, on what commercial banks are demanding.


# 2 is accrual accounting

I would say that # 1 is not a legitimate accounting option anywhere

For example, it's not the same thing as marked to market accounting, which is the general type of accounting that is the alternative to accrual accounting

In my opinion, what economists are doing in the way they treat seigniorage is treating central bank equity as if has market value - like a common stock - and using that as the contribution to the consolidated deficit

again, thats not the same thing as marked to market accounting for central bank profit

e.g. the stock market capitalizes the expected future interest margin effect of any change in a commercial bank balance sheet - but marked to market accounting alone won't capture that effect


What they are doing is treating currency like paper gold. The government printing up money and spending it is like a gold prospector picking up gold nuggets and spending them under a gold standard.


With all the small beers you mentioned in your blog piece, I think you'll be needing a helicopter to get you home.

Bill Woolsey,

I can understand that.

I think what I described is what they do in effect - in the context of the relationships that connect accrual accounting, marked to market accounting, and the market value of equity.

That's the effect in conventional accounting terms, although the intuition may be more basic as you describe.

"So the stock of currency will be permanently higher by 0.25% of NGDP."


The stock of money has increased by 5%. Whether that translates into income depends on how much of it is spent as opposed to retiring debt or sticking it under the mattress. Even if it is spent, how much of it ends up in the domestic income stream (and enhances domestic employment/output/inflation)depends on leakage to foreign goods (which is more than likely going to be a high percentage).

Scott S said:

"Why can't central banks do an exogenous increase or decrease in the base..."

Because this hasn't worked so well since 2008, that's why helicopter money is in contention.


An example:

Set up Carleton Bank with one 5 year fixed rate mortgage for $ 100,000 and one demand deposit for $100,000

The mortgage earns 3 %

The demand deposit pays 0 %

The directors of Carleton Bank gift 1 share of common stock to Nick Rowe

That means the book value of Carleton Bank's equity is $ 0

Suppose everybody thinks that the interest rate on Carleton Bank's demand deposits will be 0 % for a very long time

This situation is directly analogous to central bank seigniorage

Nick may be able to sell his share for a market value of $ 100,000 and realize a profit or he may record an unrealized gain of $ 100,000 on his own financial statements

But that will correspond to neither accrual accounting profit for Carleton bank which is $ 3,000 per year, nor to marked to market accounting profit for Carleton Bank which is accrual adjusted for market value changes on the mortgages over each reporting period

Nick's realized or unrealized capital gain will not be recorded as Carleton bank profit

And the government's unrealized gain in the notional market value of central bank capital will not be part of central bank profit

Although the government may choose to include that notional market value of central bank capital in the consolidated budget - highly weird but it could work that way in theory and consistent with accounting norms otherwise - but they'd better mark to market their bond liabilities too if that's the case

JKH, does the price inflation target apply to the lender of last resort function of the central bank to the commercial banks?


"....nor to marked to market accounting profit for Carleton Bank which is accrual adjusted for market value changes on the mortgages over each reporting period"

Isn't the mark to market valuation a once off (as long as market conditions don't change) valuation increment of the bank's assets (the mortgage) which will appear in the balance sheet and the increment of value over book will appear in the profit statement and will also be shown as an increment in equity? Hence, Nick can sell his share at a profit.

There are two models/worlds:

A) Every one are acting on present values, money is super-neutral and HM doesn't change anything. But I think government bonds still matters because their present value changes counter-cyclically with the interest rates, deficit spending is name of the game.
B) Real world where at least some of the agents are liquidity constraint - money isn't neutral. I _think_ money is not neutral even in the long run.

Banks are closer to A) world, which I _think_ means that's why QE doesn't work very well; common people are very much in B) world, HM will work.


Under accrual accounting:

The market value of the $ 100,000 mortgage at the time the mortgage is booked on the balance sheet is $ 100,000 - and that becomes the book value.

That book value doesn't change.

Similarly, the book value of the demand deposit is $ 100,000 and that doesn't change.

And the net book value is 0, which is the book value of equity, and that doesn't change.

Under marked to market accounting, profit and the book value of equity change by the change in the market value of the mortgage - the change starting from, but not including the starting value of $100,000.


somebody might try and argue - based on what I've described so far - that the market value of the kind of demand deposit I described (analogized with central bank money) would be 0 under marked to market accounting

but that's wrong:

banks can "sell" liabilities as well as assets

when they sell liabilities, they normally pay to do so instead of getting paid as happens when they sell an asset like a mortgage

if the market value of the demand deposit were in fact 0, nobody would be willing to assume it or "buy" it at that price

because the buyer/assumer would be liable to repay the book value of that deposit if/when the depositor wants his money back - and the buyer/assumer of the deposit liability won't have the cash to do that because he hasn't been paid to assume the liability

so that's a contradiction

"Under marked to market accounting, profit and the book value of equity change by the change in the market value of the mortgage - the change starting from, but not including the starting value of $100,000."


I'm not quite sure what you're saying.

Let's say the market value of the debt has gone to $110,000. The assets of the bank are now $110,000. Equity has been increased by $10,000 and deposits are still $100,000. So we have in the balance sheet:

equity = assets - liabilities

10,000 = 110,000 - 100,000.

Yes? No?

(And the profit statement (assuming no other changes) will show a provision for increase in value of assets of $10,000, which will show up as an increase in accumulated profit in the balance sheet - i.e. as an increase in equity.)


yes - perfect

(under marked to market accounting)

I like Bill's analogy with the gold miner.

JKH: "...because the buyer/assumer would be liable to repay the book value of that deposit if/when the depositor wants his money back - and the buyer/assumer of the deposit liability won't have the cash to do that because he hasn't been paid to assume the liability..."

But if central bank money is irredeemable (unlike under the gold standard), it's not like commercial bank money. And the whole point of helicopter money is to create an excess supply of money (at the existing level of NGDP) so that people go out and spend it and increase NGDP, rather than redeeming it at the central bank.

(But I'm not sure whether this discussion of the two methods of doing the accounts really matters, since you get the same Present Value of central bank profits either way you do it. But it's still an interesting (to me) discussion.))


Since comments don't seem to be enabled for your post over at MR, I'll just congratulate you here. Great post!

The HM crowd seem to believe that if you don't think about the balance sheet consequences they'll somehow disappear. They thereby convenienlty miss the equivalence between the self inflicted black hole in the CB balance sheet and the self inflicted, bond financed government deficit in 'normal' finance even without QE.

This is why I like to point out that any type of monetary financing always involves three parties, the bond issuer / seller, the payee and the money issuer / bank. Nothing, except intentional obfuscation of institutional division of labour to further a technocratic agenda, is gained by conflating the first two within the same institution.

This blindness towards the first party is also apparent when Nick says that CB money is not a liability in a meaningful sense. According to this view (to the extent that I have not misunderstood it) a liability in a meaningful sense would only exist if CB money were convertible, i.e. had a fixed promised conversion rate, into another asset (e.g. gold or foreign currency). But this wrongly places the focus on the relation between money and an exogenous asset, whereas what actually determines the quality as a liability in the end is the ('horizontal') relationship between bond issuer and the payee. This holds irrespective of whether any complicating 'vertical' commitment is made by the issuing bank and is equally applicable to other money issuing institution besides the CB.

Just done a new post on the accounting question. Maybe continue comments on that particular question over there? (As you wish.)

thanks Oliver !

Thanks Nick, So what they really meant is that the central banks cannot adjust the supply of base money without disturbing interest rates. OK, but that's not what they said, and if it is what they meant then it's exactly as beside the point as you suggested. They need to communicate more effectively.

Henry, Your comment has no bearing on what I said. Read the post I was commenting on. It has nothing to do with whether QE is effective, they are talking about conventional monetary policy when rates are positive.

JKH, I read some of your post.

"The public receiving the money is interested in the money only, and has no interest whether or not bonds are issued."

I think the "public" is interested in whether a bond is issued or not.

"they are talking about conventional monetary policy when rates are positive."


"Conventional monetary policy" - are you sure about that? They're talking about a monetary expansion funded by printing money - in their words, by a permanent increase in non-interest bearing central bank liabilities. Is this conventional?

"When rates are positive" - are rates positive to any significant level now? In some countries they're negative. This is the context in which HM would be applied. They also argue that HM could result in interest rates permanently at zero.

Massive QE resulted in a massive increase in bank reserves with a feeble/very slow effect on the real economy (particularly in Europe and Japan) and interest rates heading towards zero and beyond, with deflationary pressure alive and well. They argue that HM could also have minimal expansionary impact.

That was the context for my rough and ready comment.

"They argue that HM could also have minimal expansionary impact."

Actually, Borio and co talk about the "models" giving this result.

It's difficult to decipher whether the paper is about what the authors think or what the authors think about what others think. And if it's the latter, whether they are merely reporting what they think others think or they actually endorse what others think.

Nick, is Borio's "mare's nest" that different from Kocherlakota's recent-ish piece on helicopter money?


Kocherlakota's seemed like a reasonable view to me, no?

Kocherlakota seems to not make anything of the fact that Fed money printing finance increases money base while the debt financing doesn't.


Regarding the exchange between you and S. Sumner and his concluding question ...

They are referring to the quantity of excess reserves supplied in order to achieve a supply demand equilibrium, whereby the central bank supplies that quantity of excess reserves that will cause the trading rate for fed funds (the Fed effective as it is known) to hit the target rate - when the target rate is positive and IOR is zero (e.g. pre-2008 Fed), which is the first system they describe.

Otherwise, the sensitivity of the effective rate to an excess reserve supply that that is more or less than the correct supply to hit the target rate is such that rates will plummet or skyrocket respectively toward zero or toward the LLR rate.

I.e. the demand curve for excess reserves in such a system is hyperelastic around the target/effective equilibrium supply. The demand curve is only slightly less elastic than the supply curve, which is conceptually vertical at the point of the desired equilibrium. You can see this in the graphs they provide.

What they describe is operationally correct in terms of that function of trying to keep the fed effective rate in line with the target rate.

But it is not the best writing for explaining this type of thing IMO.

Here is the very best IMO:


(link to PDF there)

Returning to Sumner's question, the authors are referring to a floor system with a positive target rate and IOR = target rate (as with the post-2008 Fed).

They are saying that paying that rate on HM created reserves means there is no effective fiscal difference compared to paying similar interest on bond financing (or at least on short term bill financing)..

Conversely, not paying IOR will drive the Fed effective rate to zero as noted above, no longer allowing for a positive target rate - which is what they mean by losing control.

Once that control is lost, no amount of HM will change the fact that the effective rate will remain at zero.

It a moot point under the assumption of an undesired consequence.

But if the target rate is actually set at intended rate of zero forever, there is no problem. But that's a choice in itself as opposed to becoming an unintended loss of control when there is no intention of setting the target rate permanently at zero.

All that said, where the authors are correct, I think there is a small wrinkle of logical error in how they conclude their piece - quite separate from everthing they have right, which is almost all of it.

Maybe later on that last point.

Sorry -

From above, should read:

"The demand curve for excess reserves in such a system is hyper-inelastic around the target/effective equilibrium point. The demand curve is only slightly less inelastic than the supply curve, which is conceptually vertical at the point of the desired equilibrium."

Elasticity translation is not my favorite task.

JKH, let's say the fed sets the fed funds rate at 3% and wants to keep it there.

Next, let's say entities want to have more currency and less demand deposits. Nothing else changes. This causes a shortage of vault cash, central bank reserves, or both.

What happens next?

Does the fed supply more currency and/or central bank reserves to keep the fed funds rate at 3%, or does the fed allow the fed funds rate to rise even though the fed wants it to be 3%?

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad