What part of "counter-example" don't you get?

Suppose you claim that it is impossible for a debt to create a burden on future generations unless: it causes a reduction in investment that reduces future output; the higher future taxes cause disincentive effects that reduce future output; it is owed to foreigners so we have to pay part of our future output to foreigners and have less future output for ourselves.

Suppose I then come up with one counter-example, with no investment, no disincentives, and no foreigners. Where the debt can't reduce future output, because I'm holding future output constant. But where the debt does make some future generation worse off.

Doesn't that prove that your claim is wrong?

You don't start complaining that my counter-example is "unrealistic", because I'm assuming output is constant, or making some other assumption you don't think is realistic. It's a counter-example.

Here's my (old) counterexample:

Assume: closed economy; no investment or real capital of any kind; lump-sum non-distorting taxes with zero collection costs; positive real interest rate and zero real growth; exogenous full-employment level of output; apples are the only output good; apples cannot be stored; identical agents; overlapping generations; no funny stuff.

Suppose the government makes a transfer of 100 apples to the current cohort, financed by borrowing. Does that create a burden on future generations?

The government borrows 100 apples from each of cohort A, then gives each person in cohort A a transfer payment of 100 apples. It is exactly as if the government had simply given each person in cohort A an IOU for 100 apples. That IOU is a bond.

So far there is no change in cohort A's consumption of apples.

Cohort A then sells the bonds to the younger members of cohort B. So each person in cohort A gets an extra 110 apples (assume 10% interest per generation), which he eats. Cohort A then dies.

Cohort A is better off. Each member of cohort A eats an extra 110 apples. In present value terms, those extra 110 apples are worth 100 apples at the time the transfer payment is made.

Cohort B eats 110 fewer apples when young, but 121 extra apples when old, and they sell their bonds to cohort C. Although cohort B eats 11 more apples in their lifetimes, the present value of their total consumption of apples is the same. The rate of interest must be high enough to persuade them to eat fewer apples when young and more apples when old, otherwise they wouldn't have bought the bonds from cohort A. So cohort B is not worse off.

But (given my assumption) the debt is rising faster than GDP. The government knows this is unsustainable. It cannot rollover the debt forever, because eventually the next cohort will be unable to buy the bonds from the older cohort. So the government decides to pay off the debt by imposing a tax of 121 apples on each young person in cohort C, which it uses to buy back the bonds from cohort C.

Each member of cohort C eats 121 fewer apples.

Cohort A eats more apples, and cohort C eats fewer apples. It is exactly as if apples travelled back in time, out of the mouths of cohort C into the mouths of cohort A. (With interest subtracted as they travel back in time through the time machine.)

P.S. For those of you who missed last year's argument, here is Bob Murphy's excellent (and very funny) review. HT Daniel Kuehn.

Comments

[Sergei]

"Suppose the government makes a transfer of 100 apples to the current cohort, financed by borrowing"

Zero real growth assumes that government performs a stabilizing function in the economy. I.e. it borrows anti-cyclically. Do you assume cycles away?

[Ryan V]

I can't really understand why people are arguing with you on this.

1. Under certain plausible conditions, debt imposes a burden on future generations.
2. (Useful) investment provides a benefit to future generations.
3. It's possible to offset the burden through other government or private actions.

I'm sure I haven't read (much less understood) every position in this debate, but it seems to me like everyone basically agrees with 1 through 3 above. It's true that 1+2 or 1+3 may not impose a net burden, but describing that as "Debt never imposes a burden on future generations" is clearly false.

[Nick Rowe]

Sergei: It's a C O U N T E R - E X A M P L E.

Ryan: me too. But then I remember the trouble I had getting my head around it 30 years ago, and escaping from the "we owe it to ourselves" gestalt.

[Bill]

Exactly, Nick. Well said. I had a similar argument when I mentioned "what if a magical pony appeared from the sky and stole all of the apples from Cohort B? They would then be worse off than Cohort A! You can't say my idea is unrealistic, it's a counter-example!"

[Ralph Musgrave]

Crucial to Nick’s argument is the claim that “Cohort A then sells the bonds to the younger members of cohort B.” Why would they? What is it about a rise in government debt that actually induces the two cohorts to do that deal?

There are several cause-effect chains or transmission mechanisms here that MIGHT lead to the above deal being done, but this is complicated and I have doubts. For example, assuming extra government debt has no effect on the price of bonds, those extra bonds are extra assets that younger cohorts COULD PURCHASE with a view to boosting their retirement incomes. But if everyone is already happy with their retirement arrangements, why would A and B would do the above deal?

[Andy Harless]

It is reasonable to (1) concede that you have disproven the original assertion but (2) argue that the case for which you have disproven it is not an interesting case and (3) modify the original assertion by adding a condition that makes it more interesting. That's essentially what I did in the comments section of your original post, and you provided a counterexample to the modified assertion. Specifically, I said to add the condition that we are below full employment when the debt is issued, and you showed (to my surprise) that it could still be a burden. What frustrates me now is that you won't make that argument "in public." (I don't think you ever did a blog post on it.) Instead you're arguing the semantics of the full-employment case. I think the slack case is where it gets interesting: even though there are apples sitting on the ground that people can just pick up and eat and would otherwise just rot, it turns out that, by eating those apples via government debt we are taking apples away from a future generation. That's counter-intuitive but it's true (at least for some reasonable cases). In that case it's not just a counterexample; it's a realistic one: in the comments section of Dean Baker's latest post, I used myself as an example and talked about how I would screw over the next generation if the government borrows money to make a transfer to me while the economy is below full employment.

I suppose I could write a blog post about this myself. But that would have to wait until I finish my quarterly economic outlook, otherwise my boss will get mad that it's not done.

[Nick Rowe]

Bill: your magic pony is a foreigner, and reduces the output of apples available for domestic consumption.

Ralph: cohort A is going to die, so wants to sell its bonds and consume the proceeds before that happens. And if the price is right, individual members of cohort B will want to buy those bonds, consume less when young, and consume more when they are old.

[Nick Rowe]

Andy: you are right, of course. I ought to do a post where we start out with deficient-demand unemployment. And I think I could show that in some cases the next generation would be worse off and in other cases the next generation would be better off on net. (Though in both cases monetary policy *ought* to be able to do the same job with no burden at all on future generations, which was the point of my last post.) But my brain is too fried to do it properly. And I keep on having to take the argument back to step one.

[Sergei]

1a. Lets assume debt is a burden. How? We force everybody to pay 10% in real terms for no economic growth. Why? Wait until conclusion
2a. Conclusion: debt is a burden
3a. What part of "debt is a burden" dont you get?

1b. Lets assume parallel lines intersect
2b. Parallel lines intersect
3b. What part of "parallel lines intersect" don't you get?

The latter case of intersecting parallel lines is the layman understanding of Lobachevsky geometry. Funnily enough it is a wrong understanding of Lobachevsky.

[rsj]

But who is claiming that it is *impossible* for *any* government policy to not lay a burden on future generations under any possible condition?

If the government builds a bridge that no one wants, that places a burden on future generations.

If the government lowers interest rates causing an asset bubble, then that places a burden on future generations.

If the government doesn't engage in fiscal policy when it should, causing liquidation of capital, then this places a burden on future generations.

[Min]

"Suppose you claim that it is impossible for a debt to create a burden on future generations unless: it causes a reduction in investment that reduces future output; the higher future taxes cause disincentive effects that reduce future output; it is owed to foreigners so we have to pay part of our future output to foreigners and have less future output for ourselves.

"Suppose I then come up with one counter-example, with no investment, no disincentives, and no foreigners. Where the debt can't reduce future output, because I'm holding future output constant. But where the debt does make some future generation worse off.

"Doesn't that prove that your claim is wrong?

"You don't start complaining that my counter-example is "unrealistic", because I'm assuming output is constant, or making some other assumption you don't think is realistic. It's a counter-example."

Nick, I do not have a dog in this fight. I like all of you guys. :)

However, when you made your first post last year, I had already read Dean Baker about the putative future burden of the national debt. I went back and read him again, and I read Krugman. I did not believe that they were making the claim that you thought that they were. Your counter-example was against something that they did not say, IMO. Instead of proving that the national debt would be a burden on future generations, a claim that Mankiw appears to make in his textbook, you proved that, given what I called a crazy assumption in addition, the gov't might have to default eventually. Whether we would call that a burden is another question. (Mankiw says that a future generation would have a lower standard of living.)

IMO, the reason that your colleagues now object that your counter-example is unrealistic is that Baker and Krugman, in their original posts, were assuming that things would go along pretty much as they always have. They may not have stated that assumption, but it is a common one. They did not, IMO, make the extreme claim that the national debt could under no conceivable circumstance become a burden.

[Determinant]

Nick, your counter-example is fallacious. It begs the question. No, not "begging the question" meaning raises the question, it relies on the question's original assertion for its own answer, it is therefore not falsifiable.

The better answer is that the borrower failed to take into account lack of growth, he knew his assets would not cover his debt liability and so he know the debt would be a burden. Your counter-example proves nothing.

The real answer about debt is that sometimes it is a burden, sometimes it is not, it depends on what happened to the assets behind the debt. Your examples are as fallacious as the naive assertion I referred to that in an elementary macro sense, we borrowed national debt from ourselves and therefore don't owe anything. It's an accounting tautology and tautologies are fallacious and uninformative.

[brad delong]

"with no investment, no disincentives, and no foreigners. Where the debt can't reduce future output, because I'm holding future output constant" why is the interest rate on government debt larger than the growth rate of the economy? Seems to m that if you made your model consistent you have a Diamond-like sustainable Social Security system...

[Frances Coppola]

Interesting counter-example. However, I am puzzled by your assumption of a positive interest rate. Where there is persistent zero growth, surely there would also be a persistent zero interest rate? Unless of course you LIKE having no growth.

[Lord]

Have to agree assuming a constant interest rate is a real weakness of this model. In any realistic scenario, the rate would fall to zero no matter what the government promises, whether through discounting of the ious or inflating away of their value. It is just a promise that can't be kept. Like promising 4% growth.

[Nick Rowe]

Brad: thanks for dropping by!

Brad and Frances Coppola: If the rate of time preference proper is positive (it might be), then the rate of interest could exceed the growth rate.

But yes, if r is less than g, then we are in a totally different world. (I think of that as the Samuelson 1958 world). Provided r is less than g forever, there is no burden on the debt, because you never need to raise taxes on future generations to service the debt. Debt is good, in that world. (Provided you don't create too much debt and force r above g).

[Rob]

You're last post did not read to me like it was saying:

"Don't assume that government debt is harmless to future generations - I have previously provided an example that shows that it can in fact be be harmful"

It read to me more like it was saying:

"Don't use government debt to raise AD because it harmful - use monetary policy instead"

The second statement seems to assume that government debt is always harmful which doesn't follow from your counterexample , which only tells us that under some circumstances debt may be harmful but says nothing about how likely these circumstances are to exist in the real world.

[Lord]

The rate of interest may, but that just means the rate rises over time as the next generation becomes less and less willing to accept certificates and taxes make up the difference until production falls. That is where you flash one of those 110 apple certificates and get handed some rotten apple from last year.

[Nick Rowe]

Rob: OK. I can think of examples where deficit financed government expenditure would be a good thing. Like maybe building schools. But I would want to build those schools and finance them by borrowing even if we didn't need more aggregate demand. Use monetary policy to keep aggregate demand at the right level, and use fiscal policy to do good microeconomic things.

[Rob]

"Use monetary policy to keep aggregate demand at the right level, and use fiscal policy to do good microeconomic things"

I agree with this - I just don't think one can arrive at that conclusion from the OLG models alone.

[Peter N]

This is really a social question, not a purely economic one. After all, you're talking about dividing future society into creditors and debtors. If the debtors come to believe that the creditors are taking an unfair advantage and thus getting to consume more than their share of current production, they will attempt to repudiate, socialize or inflate their way out of debt. This is obviously not in the creditors' interest.

If you have a regime of constantly increasing per capita real debt, sooner or later creditors will begin to believe that they won't receive the real value for their assets that they had planned on (indeed, it may well be mathematically impossible given the ratio of debt to production), either because of inflation or repudiation. They then will lobby the government for policies that favor their interests (like austerity). They don't want the debtors (who are probably the majority) to devalue their assets. They will also try to improve the quality of their assets.

This should have a familiar look, I think.

[brad delong]

I'm still not sure why these effects you have added by building a model in which you can shift period-t consumption from the period-t young to the period-t old are large relative to the Okun Gaps closed by expansionary fiscal policy. Could you enlighten me?

[Nick Rowe]

Brad: they may or may not be. Assume a multiplier of around one. The benefits of higher GDP to today's cohort will be then roughly equal in magnitude to the burden of lost consumption on future cohorts. (Of course, if monetary policy could close the Okun Gap, there wouldn't be that debt burden on future cohorts.)

If r>g, the primary burden of the debt on future cohorts is equal in magnitude to the debt itself. That is certainly not peanuts.

Let me pose a different question: it is common for people who deny the primary burden of the debt to allow that there may be a secondary burden, due to the disincentive effects of distortionary taxation. IIRC, most estimates of the excess burden of taxation are a lot less than 100% of the revenue collected. It doesn't make sense to talk about that secondary burden and ignore the much bigger primary burden.

This is certainly not to deny that there can be benefits too from deficit spending. And those benefits will sometimes exceed the costs. Especially if we are talking about government investment, and real interest rates are low or even negative. But the burden on future generations is a cost, and should not be hidden. (Though, if monetary policy could do the same job of increasing AD, why not do that instead?)

Rob: "I agree with this - I just don't think one can arrive at that conclusion from the OLG models alone."

Agreed.

Peter N: if you are saying that the distribution of wealth between current and future cohorts is a political question -- I agree.

[Phil]

Wouldn't cohort A get only 90.9 extra apples, rather than 110? The present value of the bonds is less than 100 apples, not more, isn't it?

[david]

@Nick

Typing < carelessly makes the blog interpret it as HTML code and the rest of your comment may disappear. Type &lt; instead (& followed by "less than").

[Dan Kervick]

Nick, it seems to me that based on your earlier discussion in the previous post you have changed your argument from the old one you advanced in those earlier posts. Your old claim was:

1. Unless Ricardian equivalence is true, then debt is a burden on our descendants.

But now you are asserting something like:

2. Unless Ricardian equvalence is true, then if our descendants purchase our debts from us, and we don't spend what they pay for the purchase in ways that ultimately benefit our descendants, then the debt is a burden on our descendants.

You've even said that you are including under the category of burdens such a thing as a "negative burden", which is another word for an asset. So I don't think there is a whole lot left of your claim. You're saying that unless Ricardian equivalence is true, debt burdens the next generation - except in those cases where it doesn't.

But beyond that, I would like to argue that even in your bad-case scenarios, it is not the debt that is the burden, but the descendants' decision to purchase the debt that in some cases turns out to be a bad deal for them.

Make it simple: Suppose that The Public borrows 1000 apples from Bob, who is himself a member of The Public and gives Bob in return a bond for 1100 apples to be paid in five years. Suppose Bob's son then buys the bond from Bob for 1050 apples. Suppose The Public eats the 1000 apples borrowed from Bob and the gluttonous Bob then eats the 1050 apples his son gave him.

Now suppose every member of The Public dies suddenly - from apple poisoning perhaps. They are replaced in their social positions by their heirs - The New Public - which includes Bob's son and which reaches the age of majority at the very same same time, and takes over all of the rights and obligations of The Public. Bob's son has a public bond for 1100 apples, which The New Public then pays off. It seems to me that these three things have happened:

3. There was a transfer within the older generation of 1000 apples from Bob to The Public during the era of The Public.

4. There was a transfer within the younger generation of 1100 apples from The New Public to Bob's son, during the era of The New Public.

5. There was a net transfer from the younger generation to the older generation of 50 apples.

So in conclusion we can say that both Bob, and by extension The Public of which he was part, received a net gain of 50 apples. And we can say that Bob's son received a net gain of 50 apples. And we can say The New Public of which Bob's son was part, suffered a net loss of 50 apples. But notice:

6. It is not the debt that burdened the younger generation. The debt itself only resulted in a transfer within that generation of 1100 apples. That might be a bad thing because of its bad distributional consequences, but it is not an intrinsic generational burden (and if we told a different story in which all of the members of The New Public purchased part of Bob's bond, then we don't even get the distributional harm.) The net generational loss resulted from the decision of Bob's son to purchase the bond from his father. He gave 1050 of his generations' apples to his dad, and in return the other member of his generation gave him 1100 apples. So he made out well; but his generation lost out.

7. If Bob's son had not purchased the debt, and had only inherited his dad's bond when Bob died, then there would have been no generational loss.

8. If Bob had died without selling his bond and as a result his asset and The Public's liability had simply been extinguished, then there is no effect on the next generation at all - only an initial transfer within the older generation.

So it's not the debt issuance that was the problem. It was the younger Bob's decision to purchase some of that debt from the original lender that created the problem for his generation.

And again, this bad case scenario all depends on the fact that we are talking about apples and extra consumption by the older generation - instead of extra investment by the older generation. If what The Public had borrowed from Bob had been bricks, and if Bob's son had used bricks to purchase Bob's bond, and if The Public and a more philanthropic Bob had then used the bricks to build schools to educate their heirs, then the next generations would have suffered a net loss of 50 bricks, but received a large benefit in the form of an education.

[david]

If I'm reading the model as outlined by Bob Murphy correctly, as long as the long-term growth rate is expected to exceed the real rate on government debt, then it is unambiguously welfare-improving to keep selling coupons for wealth created by future growth to the present generation, until the real rate equals the growth rate.

It's what private actors would do, if any private institutions were expected to last that long, but of course governments are the only issuers of multiple-decade-long bonds.

Is this right?

[Nick Rowe]

Phil: if cohort A gets the bond worth 100 apples (by assumption) when they are young, and sells it for 110 apples when they are old, they get to eat an extra 110 apples when old.

david: thanks. That has caught me out before.

And now Paul Krugman has responded. And I was planning a quiet relaxing weekend!

[david]

If you can, you could edit your 06:09 PM comment to restore what followed <, I'm curious what you wanted to say to Brad.

[Nick Rowe]

david: Ooops! I didn't notice that. Edited. Thanks.

Dan: "But beyond that, I would like to argue that even in your bad-case scenarios, it is not the debt that is the burden, but the descendants' decision to purchase the debt that in some cases turns out to be a bad deal for them."

It is individually rational, but collectively irrational, for the young in cohort C to buy the debt from the old in cohort B.

[Dan Kervick]

In my hypothetical example, agreed Nick. So maybe they shouldn't do it?

And in the alternative example in which the goods the younger generation uses to pay for the debt are to be invested in ways that benefit posterity, rather than consumed, then the purchase of the debt is not even collectively irrational.

I believe your overall case, going back to those older posts, is all based on the stereotype that public borrowing is all motivated by the borrowing cohort's desire to live high on the hog, rather than its desire to increase public investment.

I haven't even brought out the case of r < g, which I know you have already accounted for.

[david]

Can't help but feel a lot of the objections are due to a sense that government debt is being bashed rather than due to objections to the abstract OLG framework being invoked.

And of course a lot of the older endogenous types don't think g is exogenous to the amount of debt being raised at all, which just confuses the issue.

[Dan Kervick]

But who is claiming that it is *impossible* for *any* government policy to not lay a burden on future generations under any possible condition?

Right rsj. Nick's original claim was that debt is a burden unless Ricardian Equivalence is true. But it seems to me the claim now is just that there are possible cases in which the issuance of debt, combined with certain ways of transmitting that debt to the next generation, ends up imposing a real cost on the next generation. But who would ever have argued against that? Here's an easy way: The public borrows a billion dollars from Slippery Sam at 11 million percent interest. When the debt comes due, the public has to print the funds to pay it, causing astronomical inflation that utterly destroys the purchasing power of every one but Slippery Sam - which doesn't even help him much since the economy ceases to function. That's a counterexample too.

[rsj]

And the odd thing is that we are now living in the aftermath of a collapse of private debt. That is in most places (but not Canada, where house prices continue to go up).

The popping of this bubble is an enormous burden on successive generations, and even in Canada, there are huge transfers from young families purchasing their first home to incumbent owners who take delivery of capital gains due in large part to low interest rates as set by the central bank.

Yet there is zero discussion of how monetary policy creates burdens for future generations. And this is the the big story of our generation, completely missed by standard macro, and certainly missed by Nick's paleo-monetarist macro.

So why even talk about the potential burden of government debt imposed by fiscal policy, when the big story of our era is the actual burden of private debt imposed by monetary policy?

[Too Much Fed]

How about assuming:

debt is denominated in medium of exchange

wealth/income inequality

too many apples gives the person a bellyache or other GI problems

"Cohort A eats more apples, and cohort C eats fewer apples. It is exactly as if apples travelled back in time, out of the mouths of cohort C into the mouths of cohort A. (With interest subtracted as they travel back in time through the time machine.)"

Could there be something else traveling in time?

[Too Much Fed]

rsj said: "Yet there is zero discussion of how monetary policy creates burdens for future generations. And this is the the big story of our generation, completely missed by standard macro, and certainly missed by Nick's paleo-monetarist macro."

Not enough microeconomics. Not enough questioning of assumptions. No mention of the retirement market.

And, "So why even talk about the potential burden of government debt imposed by fiscal policy, when the big story of our era is the actual burden of private debt imposed by monetary policy?"

because both types of debt can impose a burden?

savings of the rich = dissavings of the gov't (preferably with debt denom in med of ex) plus dissavings of the lower and middle class (preferably with debt denom in med of ex)

imo, need to discuss how gov't debt and private debt are similar instead of different.

[Duritobance]

I don't get this analogy...You'll have to pardon me if this sounds stupid, but apples disappear after you use (read: eat) them...Money, however, is spent and ends up in the hands of someone else...It doesn't disappear...Please, tell me why this analogy is supposed to work?...

[JKH]

Nick,

As was the case a year ago, your argument depends totally on the assumption of a future tax.

And as was the case then, your argument is valid based on that assumption.

And that is what your counterexample here assumes.

The purpose of the future tax is to reduce debt.

In effect, the tax “destroys” bonds.

That destruction of bonds prevents some future generation from selling those bonds to the next generation.

And that prevents that same future generation from gaining the benefit of an inter-generational transfer of apples/money that otherwise might have occurred.

And that’s all your argument depends on.

Those who claim the opposite case only assume that there is no such tax.

And that’s all their argument depends on.

[JKH]

i.e.

its actually the assumed destruction of debt that causes the future burden

[JKH]

and, generalizing from your title, why is it so difficult to sort out the dimensions of counter-factual space in economics?

[123]

Nick:"Though in both cases monetary policy *ought* to be able to do the same job with no burden at all on future generations, which was the point of my last post"

Monetary policy does the job and one of the consequences is the increase of the private debt. But the burden logic applies to the private debt too.

[Phil]

>"Phil: if cohort A gets the bond worth 100 apples (by assumption) when they are young, and sells it for 110 apples when they are old, they get to eat an extra 110 apples when old."

Ah! I get it now. Thanks.

[Nick Rowe]

JKH: "its actually the assumed destruction of debt that causes the future burden"

Nearly agreed. It's the tax to *service* the debt. If the present Value of those tax liabilities equals the value of the debt (which it does unless you can Ponzi), it's the same thing.

"Those who claim the opposite case only assume that there is no such tax."

That isn't correct. They say "we are paying the tax to ourselves, so it isn't a burden".

[Shining Raven]

JKH nails it and boils it down to the essence. I completely agree with this.

The point is of course obvious: If you start some kind of scheme where the working population supports the non-working (old, retired) population, and you end it at some point, without giving the "last" generation the chance to reap the benefits, then of course they get shafted.

However, the problem of supporting the retired population does not go away, just because you end one particular scheme to organize this support and to allocate consumption among the population.

Your model presents a false alternative: The only way to allocate consumption between the old and young generation living at the same time in the model is by way of the bond scheme. If there is no bond scheme, then there is no re-distribution of consumption from young to old. Thus there can be a generation that has to carry a net burden, because they don't receive a transfer from a younger generation (since the "game" ends).

But in the real world, old people have to eat, need medical care, need a place to live, anyway, and whatever they consume, members of the younger generation cannot consume. So you always need to distribute consumption to the old generation, and there is no possible scenario under which the old generation does not place a "burden" on the young generation. There is no choice but to re-distribute, and the bond scheme is only one way to do this.

So sure, it is conceivable that at some point a younger generation decides to end a particular re-distribution scheme, to not replace it by anything else, and to let their retired parents starve (although politically this is extremely unlikely, because old people *vote*), and then obviously this is a net burden on the people who starve, and who supported their own parents and don't get anything out of the system.

In reality, if you end the "bond retirement system" and retire the debt, so that there is no way to "save" for retirement, the burden of care would fall back onto family and relatives, and they would once again have to reduce their consumption, in order for the old to consume anything. But debt or no debt, this transfer has to take place in any civilized society.

[K]

Shining Raven,

You could equally well think of it as a "stock scheme." When we are young we build capital assets whose expected output is adequate to support us when we are old. No intergenerational transfer required unless you count revenues from renting out stuff we made when we were young. It's the same thing as the bond scheme *assuming* the bond proceeds were invested in productive assets.

The real difference between the stock and government bond scheme is that government tends to be captured by the old who can use the bond scheme to take consumption from the young who aren't necessarily willing counterparties to the deal. The old are the agents of both parties.

[Frances Woolley]

Shining Raven: "But in the real world, old people have to eat, need medical care, need a place to live, anyway, and whatever they consume, members of the younger generation cannot consume. So you always need to distribute consumption to the old generation, and there is no possible scenario under which the old generation does not place a "burden" on the young generation. There is no choice but to re-distribute, and the bond scheme is only one way to do this."

First point: a bond scheme is only one way to redistribute. For the other ways, see my post "You can't escape demographics: quit whining and deal with it.".

Second, we do not know what level of consumption is sustainable with the type of population structure we will have once the baby boomers retire and we're faced with a seriously inverted population pyramid.. Back in the 1960s, before Canada began its attack on poverty among the elderly, the image of despair was a lonely old lady eating cat food. Now seniors have the lowest rate of poverty of any demographic in the population.

Sure, seniors have to eat. But how much, and what, is another question. It's not going to be steak and caviar for everyone. Maintaining the rapidly swelling older population in the style to which they wish to become accustomed will break the bank.

[JKH]

Nick,

There’s either a tax or there isn’t.

If there isn’t, there’s an increase in the deficit as the counterfactual to the tax.

If there is an increase in the deficit, there’s an increase in the debt.

So the tax avoids the increase in the debt in the counterfactual.

In that sense, counterfactual debt is destroyed - although granted, I appear to be “cheating” slightly in that counterfactual-type interpretation. But there’s a bit more:

Actual debt is destroyed if the tax covers more than the interest, which was my assumption. I recall this issue from last year’s discussion. I think I’m assuming a primary deficit in saying all this, and you’re assuming primary balance in effect. With primary balance, it is the interest compounding of the debt that eventually strains the system at the margin. But to my point, that doesn’t preclude a primary deficit in conjunction with that. I think you are assuming the “minimal case” for a future burden in that sense, whereas I’m being more general. My general case isn’t wrong; it’s just not absolutely necessary to prove your point?

“Those who claim the opposite case only assume that there is no such tax - That isn't correct. They say "we are paying the tax to ourselves, so it isn't a burden".

My point there is that if a correct counterargument to yours exists, it must rely on the assumption that taxes won’t be paid, whereas your argument relies on the assumption that taxes will be paid.

And I think your further point is that an incorrect counterargument could interpret the economic effect of the payment of taxes in the wrong way, as per your explanation.

Subtle difference there – the payment of taxes is a necessary condition for a correct counterargument - but insufficient as a correct argument unless the argument interprets the payment of taxes in the correct economic way. And I think the correct way is as I described above, primary balance or primary deficit.

[JKH]

meant:

"the non-payment of taxes is a necessary condition for a correct counterargument"

[JKH]

shoot

meant:

"the non-payment of taxes is a necessary condition for a correct counterargument - but insufficient as a correct counterargument argument unless the argument interprets the payment of taxes in the correct economic way."

whatever,

But you're right; if they're arguing that paying the taxes to ourselves is the reason why there's no burden, that's incorrect

how about:

payment of taxes is the destruction of counterfactual debt under primary balance and actual debt under primary surplus?

(I'm going to fast and tripping here; sorry)

[JKH]

9:52 should read:

"I think I’m assuming a primary surplus in saying all this, and you’re assuming primary balance in effect."

disaster - sorry

[Nick Rowe]

JKH: Assume the intertemporal government budget constraint holds (that means Ponzi is not possible).

Then Outstanding debt = PV of primary surpluses

[wh10]

Yeah - it all depends on whether or not taxes need to be paid. I thought we settled all of this. So Baker's idea is no more zombie than Nick's - it depends on the assumptions.

So you guys should argue about whether taxes eventually need to be paid. That's the interesting part!

[Nick Rowe]

wh10: "So you guys should argue about whether taxes eventually need to be paid. That's the interesting part!"

Agreed. It's also the very hard part. Because we don't know for sure if r will be less than or greater than g in the future. And I can't get my head around it. The best I could do is this post.

[Dan Kervick]

I don't agree with JKH that the crux of Nick's argument is the future tax to pay off the bonds, and that you get a sound argument if that future tax exists. The future tax itself just transfers money from people in that future generation to other people in the same generation.

What Nick's argument depends on are the claims that (i) bonds are transmitted from older generations to younger generations by the members of younger generations transferring things of value to the older generations in exchange for the bonds, and (ii) the value that is transferred to the older generations from the younger generations remains entirely with the older generations to which it is transferred. As a result, Nick is positing a permanent net flow of value from young to old. Nobody loses so long as r < g, and each generation receives more in transfer from its posterity than it transferred to its ancestors, and the process continues indefinitely. But if there is a last generation, then that last generation gets stuck with the loss from its own transfer of value to buy the last issue of bonds.

But if you sufficiently relax either (i) or (ii) above, you don't get the problem.

I don't really see what Nick's argument has to do with debt and bonds. The bonds are a red herring. Suppose instead we just have a permanent tax on all working people below a certain age which is used to support the lives of all retirees above a certain age. The tax is a continuing drain of value from young to old, and the young who lose value recover it when they are are themselves old. But if there is a last generation, they don't recover anything and are stuck with the loss. The bonds just add an extra inessential wrinkle to this scheme: we imagine instead a process in which the old first borrow value from each other in exchange for bonds promising future return of the value plus interest, and then they tax the younger generation the full sum of that borrowed value plus extra and give them the bonds in return. Same result. It's not the debt that is the burden. It's the tax. Similarly in Nick's cases it's not the bond issuance that is the burden; it is the ongoing decision by the young to buy the bonds from the old that causes the problem.

And so if the sum total of a society's public debt were contracted solely to boost the consumption of the older members of society, and if the real value to boost this consumption were always taken from the younger members of society, then the Rowe mechanism would be operative. But neither if things is uniformly true in the real world. Public borrowing is used to finance a variety of public goods and services, many of which directly or indirectly benefit the younger members of society. And in many cases the older members of society borrow from each other to do this, and the interest is paid in "real time" so to speak via an intra-generational transfer among the older members of society. And bonds that are transmitted to the young are often gifted to them, not sold.

The real problem with public borrowing, when it is a problem, is that it represents an ongoing transfer from the less affluent to the more affluent. The affluent lend the less affluent something off value and are paid back with interest, and as a result the rich get richer.

[wh10]

"I don't agree with JKH that the crux of Nick's argument is the future tax to pay off the bonds, and that you get a sound argument if that future tax exists. The future tax itself just transfers money from people in that future generation to other people in the same generation."

You're assuming that the tax is used for govt spending. If it's not, non-govt NFA are reduced across the board.

[JKH]

Hang on, Nick.

I think I was right all along, notwithstanding the primary balance detour, etc.

Your argument depends on the fact that some future generation doesn’t have bonds to sell to the next generation.

The reason they don’t have bonds is that bonds have been repaid with taxes.

If existing bonds have been repaid, they have been destroyed by taxes.

If existing bonds have been repaid in the current accounting period that means there’s been a primary surplus in the accounting period. Taxes have been sufficient – not only to pay interest on current bonds (preventing an addition to the deficit and debt) but to pay down existing debt and bonds.

In any event, bonds have been destroyed by taxes.

And it is the destruction of bonds that prevents them from being sold to the next generation.

And therefore the current generation incurs a burden relative to prior generations who could sell their bonds to the next generation.

Forget the PV stuff on the budget constraint. Think future accrual accounting – which is what I described above.

Wh10 has the right idea with “NFA” reduction. That means taxes destroy debt.

And:

“So you guys should argue about whether taxes eventually need to be paid. That's the interesting part! ... agreed.”

That’s what I said basically.

[Dan Kervick]

wh10, sure, but I'm trying to meet Nick's argument on its own ground. Isn't his counterexample about cases in which the public pays off its bonds in the final generation by taxing itself?

[wh10]

Dan, in step 1, the govt taxes the nongovt sector, which lowers NFA in the nongovt sector by that much. Step 2, the govt uses their account marked up by the amount of taxes to repay the bonds - this is just a swap of that tax money for bonds, with no counteracting increase in NFA. NFA has been net lowered by the amount of the tax.

[wh10]

But I guess my comment that you must have been assuming the govt spends the tax money didn’t make much sense either. That’s not necessarily a redistribution, but it is a net increase in non govt NFA.

[Dan Kervick]

Wh10, I don't think I follow you. That's what I said: If the government pays off its bonds by taxing the repayment amount and then spending it on repayment, then the result is only an intra-generational transfer, not a net gain or loss among that generation. On Nick's picture, the reason the imagined final generation loses out is because they have paid the departed older generation for the bonds before the final repayment occurs.

But I'm trying to stick within Nick's framework which depends on thinking in terms of real value - things like apples - and not monetary operations. My argument is that even if we do that, his bad-case scenario packs in a few important assumptions that need not, and in many real world cases do not, correspond to reality.

If we bring in the fact that in the real world what is being taxed, borrowed and repayed is not apples, but money, and that money is being created and destroyed in the process of carrying out all these operations, then we bring in a bunch of additional factors about level changes, government employment of resources without corresponding taxes and borrowing, etc. that Nick seems to want to leave out of the picture.

[Dan Kervick]

then we bring in a bunch of additional factors about level changes, government employment of resources without corresponding taxes and borrowing, etc.

should be

then we bring in a bunch of additional factors about price level changes, government employment of resources without corresponding taxes and borrowing, etc.

[wh10]

"Wh10, I don't think I follow you. That's what I said: If the government pays off its bonds by taxing the repayment amount and then spending it on repayment, then the result is only an intra-generational transfer, not a net gain or loss among that generation."

No, it is a net loss - NFA is net lowered. My example is pretty clear. If the govt needs to pay down $100 in bonds, and it taxes $100, and then swaps $100 cash for $100 bonds, NFA is lower by $100. It's not just swapping the $100 around from the within the generation.

[wh10]

Now I agree I am talking in nominal terms there, not real, but the portion of your writing that I quoted seems to be discussing matters in nominal terms as well.

[ws]

You say apples can not be stored.
So you can eat only apples that are growing on the trees NOW.
Who produced the extra 110 apples that cohort A ate?
What happened to the 121 apples that cohort C didn't eat ?

[Too Much Fed]

JKH said: "Actual debt is destroyed if the tax covers more than the interest, which was my assumption."

This is a problem with gov't debt. There are lots of times when the gov't debt is not set up to pay back both interest and principal so there is no rollover risk (the gov't debt/bond is paid off). I believe almost all private debt is set up so that interest and principal payments are made.

[Dan Kervick]

OK, I see what you are saying wh10. It does make a difference between whether we are talking real or nominal here.

[Kaleberg]

That's absolutely correct in a zero growth society. If you allowed for growth, for example, improved apple breeding, tree planting and more effective picking and packing, the system would be sustainable with each cohort eating more apples than the previous.

It's like physics. Near absolute zero, liquids can flow without friction, but don't rely on this in your own home or industrial plant.

[Min]

What is the burden?

Let us assume Nick's basic setup: An island economy with no economic growth, no population growth, and no money, where everyone eats apples. Once a generation, the gov't sells bonds to the younger generation for apples and uses the proceeds to retire the bonds of the older generation. The interest on the bonds is r per generation.

One year a crisis develops. The older generation bought the bonds for X apples, but the younger generation cannot afford to pay X + r*X apples to them. As the date for the new bond auction approaches, it dawns on members of the younger generation that they cannot afford to buy enough bonds to pay off the older generation. They talk among each other, and finally the prime minister calls a caucus of the younger generation to discuss the problem.

After the caucus the prime minister announces to the older generation, "We are sorry, but we cannot afford to pay you what we owe. The bond auction would fail. We have to default on your bonds. What we can afford is to pay you Y apples instead of X + r*X. That is still more than the X apples you paid to your parents' generation. That's the best we can do. Take it or leave it."

The older generation caucuses with the former prime minister. After some grumbling they decide to accept what they can get.

The gov't sells 0% interest bonds to the younger generation for Y apples and pays them to the older generation. The younger generation then throws a week long celebration to commemorate the agreement and entertains the older generation.

One generation later the gov't sells 0% interest bonds to pay Y apples to the older generation. It also declares Sustainability Week as a national festival.

Soon after the original sale of the 0% interest bonds, a reporter interviews the prime minister. Here is a portion of that interview:

Reporter: Several members of the older generation are complaining about the deal. They bought their bonds in good faith and feel that receiving only Y apples places an unfair burden upon them.

Prime Minister: That is their greed speaking. Any burden is in their minds, and is self inflicted.

Reporter: Their greed?

Prime Minister: Yes. When they bought their bonds they were both greedy and irrational. One of the human cognitive skills is the ability to imagine yourself in another's place and to understand their thoughts, feelings, and behavior from their standpoint. That is called empathy. If they had put themselves in our place, they would have realized that we could not afford to pay X + r*X apples. They would have foreseen the eventual default if they asked for so much. They failed to do so because their greed was greater than their empathy. Their emotions overcame their rationality.

Reporter: But they were following tradition.

Prime Minster: Yes. The tradition contained a fatal flaw that was not obvious to our ancestors, who somehow believed that it could go on forever. But it was unsustainable. With our 0% bonds we have implemented a sustainable tradition. It is curious, in a way, that our ancestors, who at some point consumed the same number of apples in old age as in their youth, set in motion events that culminated in our current system, in which we consume more apples when old than when young, something that we much prefer. They were greedy and irrational, but changed our world for the better.

[J.V. Dubois]

Oh my ... Please not this again. Guys, it is CRYSTAL CLEAR that Nick is as right as it is ever possible to be right. Please don't do this. I already had to battle my cognitive dissonance once, when Nick showed me by this example how wrong my thinking is - and now I am starting to lose my faith that there are some fairly smart people doing macro. I beg you - stop putting up these silly counter arguments, or at least use Google search on this page. The only thing you all achieved so far (DeLong and Krugman included) is that I am starting to feel that there is no hope to push forward those slightly more complicated measures like nominal income targeting.

Nick: Sorry for my totally unproductive rant, but this is all so incredibly frustrating. I feel for you ...

[Nick Rowe]

JV: I feel a bit the same way. But then I remember how long it took me to get my own head around this stuff.

[anon]

J.V. Dubois, you can still get "burdens" with "income targeting." What happens if inflation outpaces growth? The old consume more than the young.

(BTW, I think your counterexample is valid, but calling it unrealistic is a valid argument.)

What's missing? We have a debt crisis. What do we do?

1. Try to inflate away the "burden" and/or give the bondholders a haircut.

2. Resign ourselves to cyclical burdens. (Perhaps trying to "correct" them is either too hard or will do more harm than good.)

[Ritwik]

Can we all agree that :

1) No AD policy is distribution-neutral. (Except, perhaps, helicopter drops).

2) All distributively non-neutral policies are burdens on some cohorts, relative to some baseline.

3) To the extent that cohorts defined by age/birth can be assumed to share tax regimes, variations in tax rates are burdens on some 'generations' and transfers on others. This is true for any variation in tax rates.

But why to talk about overlapping generations when one means equi-fiscal cohort. Heterogeneity of lifetimes means that 'all people born today' is not necessarily a better Borgesian category for fiscal outcomes than 'all people alive today'. It could be. But that's not obvious.

[Ritwik]

Sorry, I should have said - 'why to talk about overlapping generations' when one means 'overlapping equi-income fiscal cohorts'.

[Andy]

So let's flesh out the apples analogy a bit. Now, since apples are the only product in the economy, let's assume that taxes are collected in apples.

Scenario 1, no government borrowing.

Generation 1 - Let's say the economy produces 1000 apples, the government collects 200 apples through taxation, and distributes 200 apples to citizens.

Generation 2 - The economy again produces 1000 apples. The government again collects 200 apples through taxation, and distributes 200 apples to citizens.

Scenario 2, government borrowing.

Generation 1 - Let's say the economy produces 1000 apples, the government collects 200 apples through taxation, borrows another 100 apples at a 10% coupon FROM PRODUCERS WITHIN THE COUNTRY, and distributes 300 apples to citizens.

Generation 2 - The government wants to pay off its debt. The economy again produces 1000 apples. The government raises taxes to 310 apples, distributes 200 apples to the broad population, and repays its debt of 110 apples to the producers it borrowed the original apples from.

Where does the impoverishment occur? The population still produces and eats 1000 apples in either scenario. Who within the population changes, but that's all; it's redistribution.

Now of course, if you borrow the apples from a foreign country, it is a different story.

[Nick Rowe]

Andy: see today's post.

[Shining Raven]

Andy: I take the same view as you, i.e. consider the distribution of consumption in a horizontal sense, and there of course you only have a re-distribution of the production at any point of time among everybody then living.

I believe that this is actually the question we are facing, so this view makes the most sense to me.

Nick is taking a longitudinal view, integrating up the lifetime consumption of any cohort of people of the same age.

His point is then trivial: If there is a change in the horizontal re-distribution mechanism during the lifetime of any one cohort, this can end with them having overall a smaller lifetime consumption than other cohorts. I.e. if they forgo consumption in their youth, to support the old people then living (horizontal redistribution), and if the horizontal re-distribution scheme is then ended when they themselves expect to receive benefits, but before these actually arrive, of course they will be able to consume less over their lifetime than other cohorts, who either did not pay in or did receive their benefits.

Sure, if Congress cancels Social Security today, everybody who has paid in and has not yet received his full benefits gets shafted.

However, this is always a decision that the people living at that time make, about a distribution of consumption and resources at that time.

It is not a forgone conclusion following from the existence of the debt.

Overall, I don't find this view very useful, since it does not really tell us anything about how to solve the horizontal distribution problem, which in my view is the real problem.

[Too Much Fed]

Shining Raven said: "Sure, if Congress cancels Social Security today, everybody who has paid in and has not yet received his full benefits gets shafted."

While the people who received benefits at the beginning without paying anything in received the benefits. Right?

Is Medicare the same way?