Consider two possible worlds:
World A: Us old people bequeath our government bonds to the next generation (as a freebie).
World B: Us old people sell our government bonds to the next generation (they pay us for the bonds).
See the difference?
In world A the next generation inherits the tax liabilities inherent in the bonds, and really does inherit the bonds too. It gets given a liability, and gets given the corresponding asset too.
In world B the next generation inherits the tax liabilities inherent in the bonds, but does not inherit the bonds. It gets given a liability, and does not get given the corresponding asset. It pays for the corresponding asset.
World A is the world of Barro-Ricardian Equivalence, where government debt has no intergenerational distributional implications.
World B is the world of Buchanan and Overlapping Generations models, where government debt does have intergenerational distributional implications.
World A is not Paul Krugman's world. Paul is not a fan of Barro-Ricardian Equivalence. So why does Paul suddenly switch to talking like someone who believes in Barro-Ricardian Equivalence when he says:
"Antonio Fatas, commenting on recent work on deleveraging or the lack thereof, emphasizes one of my favorite points: one of my favorite points: no, debt does not mean that we’re stealing from future generations.
….
Debt has distributional implications, and it may have macroeconomic effects because of those distributional issues. But again, all this is within the current generation; it’s not about the present versus the future."
and:
"And the problems with public debt are also mainly about possible instability rather than “borrowing from our children”. The rhetoric of fiscal debates has been, for the most part, nonsense."
Even in: a closed economy; with no investment of any kind; and purely lump-sum non-distorting taxes, we can take apples out of the mouths of yet unborn future generations, and eat them ourselves. Time travel is possible.
Dammit! I thought I had this whole point about debt burden on future generations sorted out back in the great debt blog war of 2011/12.
Or, ignore me. Listen to Simon Wren-Lewis get it right.
Ours the task eternal.
(Antonio Fatas, by the way, is correct in saying that if the government issues debt to buy real assets, and if future generations inherit (as a freebie) those real assets as well as the tax liabilities inherent in the debt, it can all cancel out.)
Worthwhile Canadian Initiative 
@JKH: “the counterfactual is that Treasury never…buys back its debt”
Maybe we should say that’s the factual–at least if hundreds of years counts as “never.”
http://www.asymptosis.com/the-meme-that-refuses-to-die-government-debt-must-be-paid-back.html
Steve,
No that’s not the factual
I said never buys back its debt – not never buys back all of its debt
Clinton surpluses
Canadian surpluses
… and so on
Ken: “Now I realize how silly that worry was, because even if the government taxed all the children and paid back the bond holders, whomever held the $18T in bonds at that point can increase their consumption by $18 trillion”
Even if that’s true, the necessary corollary would be that deficit spending cannot increase current consumption, because the bondholders must decrease their consumption now. So why would anyone want deficit spending?
Since 1947, we ran a surplus in 1947 ($2.4 B), 1948 ($900 M), 1950 ($1.9B), 1951 ($4.5B), 1956 ($1.4B), 1960 ($200 M), 1998 ($2.7B), 1999 ($66.7B), 2000, ($156.B), 2001 ($14B) (10 Years). We ran deficits the other 57 years.
http://research.stlouisfed.org/fred2/graph/?g=10uu
We averaged a deficit of 2.5% of GDP over that period.
In 2014, the U.S, paid an interest rate of 2.4% on its debt, or 1.7% in real terms. Real GDP growth was about 2.5%. This is the typical situation — real interest costs of 1.7% is the average for the postwar period. Obviously GDP growth is higher. This does not mean that we never run even a small surplus, but that we don’t need to. Many people believe that there is an inherent virtue to balancing the budget, but striving for virtue isn’t the same as a fiscal requirement.
Running a surplus doesn’t mean running a primary surplus. Generally interest payments to the public go from a high of 1% of GDP to .05% of GDP. The numbers are a bit tricky — OMB counts “debt service” as principle repayment + interest costs, and debt service appears rather than interest payments in most budget data (which is prepared by OMB). The treasury keeps pure interest payment data and this is available in FRED. But this interest payment data includes payments to the social security trust fund and for debt held by the federal reserve, etc. To get consolidated data, I took total interest payments * Federal debt held by public/total Federal debt, and then divided by GDP to get the interest payments as a percentage of GDP (actually a ratio). This assumes that the rates paid on debt held by the consolidated government is the same as the rates paid to the public.
http://research.stlouisfed.org/fred2/graph/?g=10uW
We can combine interest payments as a % of GDP to estimate the size of the historical primary surplus/deficit as a percent of GDP:
http://research.stlouisfed.org/fred2/graph/?g=10uZ (percent, not ratio)
Point being that running primary deficits is the norm, too. Clinton surplus period is an atypical case, as can be seen in the FRED data, in which small primary surpluses caused much panic about running out of Treasuries.
Going back to the larger point, Krugman was discussing an op-ed about the real economy. Nick counters with an OLG model in which the signs are all wrong. Is the OLG model even appropriate here — I don’t think so. You can’t have it both ways — either this is part of a “rebuttal” of Krugman or this post is it’s own description of whether or not debt is a burden in a looking glass economy in which OLG is appropriate and r > g. In one case, he is engaging Krugman in discussing whether debt is a burden in our economy. In the other case, the discussion is whether it is possible for debt to be a burden in an opposite economy.
rjs: if Paul Krugman said: “r < g, and will be forever, therefore a ponzi scheme is sustainable, and therefore increasing the debt would make both current and future generations better off” that would be very different. None of this “we pay the taxes to ourselves so it can’t be a burden” nonsense. There wouldn’t be any extra taxes.
Of course, less-informed people would really go ape if he said that, so I do have some sympathy. But invalid arguments are invalid arguments. And economics education matters to me, even if it doesn’t matter to those who want a particular policy.
Nick,
When I read his essay, I assumed that Krugman’s mental model was one of wealth concentration (with bequests), not OLG. That is, IMO, why he said “there are distributional consequences”. Again, that is something that can be reversed with a deficit-preserving transfer. This also goes in line with some of his other op-eds. To me, there was nothing controversial about it.
And no, r < g does NOT mean that we can continue a Ponzi scheme forever. Stop saying that. The two are related but distinct concepts. It may well be that if Debt/GDP reaches X, then r will begin to increase for larger values of X (or there will be some inflation). And it could be that wealth inequality plays a big role in determining X. And of course the debt affects wealth inequality. Therefore I can cook up a model in which r is always less than g because the distributional consequences increase wealth inequality which increase X to be always higher than the current debt load. Here, Japan would be my exemplar.
The above just means you can cook up a model that says pretty much anything. I don’t think it’s fair to say “Krugman is wrong” just because you have a very unrealistic model. You can say ‘I have a model in which Krugman’s “distributional effects” are realized as “distribution among generations”, and these effects are so important that we should view government debt as a serious negative’.
The response to that would be a nice statistical debate about whether there is a fiscal multiplier and what the effects of prolonged unemployment are — I don’t think you would win such a debate.
Maybe this small map will help:
rjs: r < g (forever in time) DOES mean you can run a Ponzi scheme forever. But if r is a positive function of debt/GDP, there will be a limit to the SIZE of the Ponzi scheme you can run (measured in terms of debt/GDP.
Yes, that’s true. Sorry!
That would be the intersection of ponzi and r < g. The difference would be r = f(D/G) or something more complicated.
The prospect of significant primary surpluses and a non-trivial reduction in outstanding government debt was a major issue for the US in the late Clinton era. Greenspan made various presentations to Congress, quite anxious about the implications of the anticipated trend.
That concern was part of the real world at the time. The US Fed Chair certainly thought it was.
Of course, it ended up not happening – at least to such a degree. But that’s not the point.
Krugman might have at least made reference to the concept when writing about “we owe it to ourselves”.
JKH,
Regardless of what action generation N takes, there is a possibility that generation N+L will run a primary surplus. That is not enough to say that generation N has burdened generation N+L. You need to argue (as Barro does), that it is inevitable that there will be some generation that will run a primary surplus of some kind (although as you point out, any primary surplus is a harm). That will give you a gross burden, although the combination of deficit spending + debt financing may lead to a net benefit.
IIRC there was a lot of handwringing about deficit doom which lead to a Clinton tax hike, which lead to huge surpluses which led to a tax cut and the (cyclically adjusted) budget went back to it’s (natural) state of primary deficits.
A more nuanced view would be that some levels of debt might be burden, if the increase in debt is not accompanied by offsetting transfers that leave total debt unchanged. And we would look at the interest rates to give us a clue as to whether the offsetting transfers should occur. To see why this is true in the OLG model, it’s enough to tax the older generation by the amount of the increased interest payments and give that money to the younger generation, and to continue this transfer every period. Now you have debt but no inter-generational transfers. The deficit-neutral fiscal policy can undo any such shifts regardless of whether r > g or r < g.
As Krugman said, it really is all about distribution.
rjs,
“A more nuanced view would be that some levels of debt might be burden, if the increase in debt is not …”
That sort of conditionality is fine by me. I guess I interpreted the Rowe and Wren Lewis posts (and Bob Murphy’s coming into the light) all as being not so normative but more expository of a possibility. Analytical completeness … that sort of thing.
I’m suddenly reminded of Krugman’s big “deficits don’t matter” fight with MMT a few years ago.
Plus ca change …
JKH,
Well, if we are going to talk about “big picture” stuff, then the reason why a transfer from old to young in the OLG model undoes the burden on the young is because the old are the ones who have the bonds! But in our world, it’s more of the 1% having the bonds (both old and young). So the appropriate transfer here would be progressive taxation to undo any regressive effects of the government debt. But what are the regressive effects? They arise from the difference between r and g. I.e. if r < g, then wealthy are paying a premium (to the government) to insure themselves, in which case the government is making money off the rich and using it to stimulate the economy. In that case, the government debt is not a burden on those who don’t hold the bonds but is a benefit. If r > g, then rich become richer as a result of their increased bond holdings and wealth inequality increases — it is a burden on those who don’t hold the bonds.
So instead of thinking about a very carefully constructed OLG model, think about rich versus poor. The OLG model is a special case of this. I’m certain that this was what Krugman was thinking about when he mentioned that there were distributional consequences. The problem I have with the OLG model when used for this type of reasoning is that it hides the real issues and masks them with talk about “the future” — i.e. the young, who have no bonds, are punished and the old, who hold the bonds benefit. The issue is now about time but about present wealth holdings, although within the parameters of the OLG model (perfect wealth equality among the old, no wealth among the young) you can frame it as a generational thing. I maintain that, in the context of an op-ed analysis about our world, this is misleading.
This is also why, if the old give their bonds to the young, then the generational issues go away — they went away because the wealth inequality issues went away.
rjs said: “So instead of thinking about a very carefully constructed OLG model, think about rich versus poor.”
Agreed.
rjs,
I’m going to circle back after starting out with barest bones:
Suppose Treasury (TR) issues a 150 year bond.
A2 buys the bond.
TR makes a transfer payment to A1 who uses the proceeds to pay for consumption.
150 years later, Z2 owns the maturing bond.
TR taxes Z1 to pay off the bond.
Z2 uses the proceeds for consumption.
In this model, for the 150 years, I don’t care whether intervening generations sell their bonds or bequeath their bonds. I don’t care whether the sellers are old or young. I don’t care whether they are rich or poor. I don’t care whether the degree of generational overlap approaches zero. I only care that the bonds are outstanding for 150 years and are held by somebody at all times during those 150 years.
And I only claim:
The transfer payment is a benefit to A1.
The tax is a cost to Z1.
That’s all.
And then I say that Z1 is a member of a future generation.
And somebody in a future generation has paid the cost of a benefit for somebody in an earlier generation.
And that constitutes intergenerational transfer of the cost of the debt.
That’s all.
Everything else is interesting but extraneous to that central point.
And I think that’s the essential point that both Rowe and Wren Lewis make.
They are a more specific about overlapping mechanics and relative ages and OGM construction, but as I said earlier I think all that is moot to the only real requirement, which is that the bonds are passed on in some fashion and don’t disappear – until the government makes them disappear by taxation.
For example, I could go to the Wren Lewis post and edit it by cutting out all the extraneous detailed OLG construction and get to this basic point, but I won’t do that right now.
Now, here is Krugman from his last post on this:
“Clearly, the overall level of debt in the U.S. economy has suddenly increased (actually by about $1.6 trillion). But has the nation become any poorer? Is that $1.6 trillion of additional debt money taken from the next generation? No and no: the additional debt represents a claim by one set of Americans on another set of Americans — and we’re talking about people here now, not future generations… Anyway, how we got here should not have any direct bearing on what the debt means now — which is that it’s money we owe to ourselves… Debt has distributional implications, and it may have macroeconomic effects because of those distributional issues. But again, all this is within the current generation; it’s not about the present versus the future.”
And so that final point is just wrong. Clearly it is wrong in the scenario laid out by Rowe and Wren Lewis and above. And because it is wrong, his earlier points while arguably correct in some way, just constitute a lot of hand waving, which is not very good for a guy who is constantly criticizing non-ISLM believers for hand waving.
“We owe it to ourselves” merely amounts to the observation that at any point in time, the world has a balance sheet. Duh. And while that alone may be intended to be enlightening to those who believe government debt needs to be repaid into some right wing black hole, it doesn’t support a larger unambiguous claim that “it’s not about the present versus the future”.
JKH,
Let me add the following observation:
“Debt has distributional implications, and it may have macroeconomic effects because of those distributional issues. But again, all this is within the current generation; it’s not about the present versus the future.”
Failed debt (aka bankruptcy) can have some pretty severe psychological implications for both borrowers and lenders going forward into the future. The future success of a functioning credit market relies upon present success. Trust is a fragile thing. And so I don’t even think it is enough to say “it’s not about the present versus the future” is a false statement because debt bought today can represent a claim on future generations. I would say that it is about present (credit money system) versus future (possibly something else), but not because of intergenerational transfers.
Paul simplifies “the additional debt represents a claim by one set of Americans on another set of Americans”. That is fine when a legal system is sufficient to resolve those claims. But imagine a closed loop debt system – I borrow from Paul who borrows from Jane who borrows from Peter who borrows from me. Suddenly I stop repaying Paul who then can’t repay Jane who can’t repay Peter who can’t repay me. In a bankruptcy proceeding, how is a closed loop of lenders resolved? This is
JKH,
Sure, I get it. But, I hope that you see that this is a benefit/cost only because one group (Z1) was taxed and another group (Z2) had their bond redeemed. Z1 and Z2 are both alive at the same time. So while you can think of this as someone in the past imposing a burden on someone in the future, you can also think of this as a distributional problem between Z1 and Z2 — something that the government can undo by taxing Z2 and giving the proceeds back to Z1.
I don’t want to get into the semantics here. Barro’s theorem is correct, as far as it goes. However the statement that “we owe the money to ourselves” is also correct — we are ignoring bonds held by the foreign sector here, which is important.
So — not arguing with your characterization — we can ask, in our economy, on whom will the burden of taxation to pay for the bond redemption fall? I hope you agree that is really the key question. If the bond holdings of each individual are held in proportion to their overall financial assets (e.g. everyone holds a share of the market portfolio, including the government bonds), then a general tax on financial assets to retire the bonds will be perfectly neutral and not a burden to anyone at all. At the other extreme, a tax purely on labor income to pay for the bonds will fall on those who work and exempt retirees. Barro’s proposal assumes the latter (along with other assumptions). Barro also assumes that those who hold bonds never work, so the entire debt is a pure transfer from future workers to future retirees. You don’t need to view it as a transfer from our generation to future retirees. You can view it as transfer from future workers to future retirees. That distinction is important. The “we owe the debt to ourselves” view is just as valid, if not more valid.
Why more valid? Because we can ask, why would the future generation enact such a transfer? Why wouldn’t the future generation tax financial assets in order to redeem the bond? Why must they tax labor and labor only? Nothing in the terms of the debt requires that labor and not capital be taxed to redeem the bonds. How the future generation decides to retire the bonds is up to them, and not up to us. So we (the generation that incurs the debt) are not burdening anyone. It really is a distributional issue to be decided by the generation that redeems the bond.
JKH said: “TR taxes Z1 to pay off the bond.”
Is Z1 getting taxed all at once to pay off the bond (I assume that means taxed to pay principal)?
rjs: “Why wouldn’t the future generation tax financial assets in order to redeem the bond? Why must they tax labor and labor only?”
If it were known in advance they would do that, nobody would buy the bonds.
rjs,
There is nothing wrong with “we owe the debt to ourselves”.
As I said, one way of illustrating this is to understand that the world has a balance sheet and that one sector’s liability (the government) is another sector’s asset (the ‘non-government’). MMT has made a career out of this. And its fine as far as it goes. Accounting is a good thing. And as several commenters have noted, perhaps this is intended to be Krugman’s core point – disposing of the false meme that debt is net negative as a whole, merely because it is owed. It is owed, but it is also owned.
But Krugman goes further than this, venturing into dangerous denials of distribution effects:
“Debt has distributional implications, and it may have macroeconomic effects because of those distributional issues. But again, all this is within the current generation; it’s not about the present versus the future.”
This is a/the point that Nick Rowe objected to.
“Sure, I get it. But, I hope that you see that this is a benefit/cost only because one group (Z1) was taxed and another group (Z2) had their bond redeemed. Z1 and Z2 are both alive at the same time. So while you can think of this as someone in the past imposing a burden on someone in the future, you can also think of this as a distributional problem between Z1 and Z2 — something that the government can undo by taxing Z2 and giving the proceeds back to Z1.”
These are complementary interpretations, each saying something different about the full puzzle. They are not alternative interpretations. When seen accurately, they are consistent with each other as two different pieces of the full effect. Ideally, for completeness, you should think of both of them at the same time:
First, regarding your undoing example:
Reversing the process by which X causes Y undoes Y. That is logical.
And it is equally logical that not reversing it leaves Y as caused.
This is universally true, I suppose.
So yes, I see that.
It doesn’t seem too relevant to the proposition that X causes Y, though.
Second, yes for sure there are distribution issues among co-existing generation(s) that are currently alive. Again, as I suggested in a previous comment, one of the things in play here is that distribution issues work both cross-sectionally and across time. These are interesting simultaneous effects. The cross-sectional issue is that there is a flow of funds from the tax liability/payment to the bond asset/redemption. There are 2 transaction nodes connected by this flow of funds. One is symmetric – the bond holder exchanges one asset for another. There is no net change in his wealth position, other than in the liquidity attribute of it. The other is asymmetric – the tax payer has lost net wealth. As I noted before, this shows up at the macro level as a reduction in the sector financial balance – the private sector loses net wealth because tax is paid from private sector equity. (That sector effect is manifested in the disappearing bond asset.) So that asymmetry is left dangling out there, which tells you that maybe something else is going on here. And that something else is that a corresponding asymmetry was created in an earlier generation – which was the addition to the private sector financial balance that was created by the infusion of equity in the form of a transfer payment.
So there are distribution effects both intra-generational (co-existing generations) and inter-generational (generations 150 years apart in my example).
But Krugman summarily dismissed the latter.
rjs
“It really is a distributional issue to be decided by the generation that redeems the bond.
I don’t disagree with that.
The future generation makes the policy decision on whether or not to tax.
And that decision is about imposing a cost on the then current generation for the benefit that accrued originally to today’s generation. Or they may carry on and continue to run deficits.
A very loose analogy for the type of decision that needs to be made is that of a loan write-down. Acknowledging and recognizing a loss is a decision to account for something that originated in the past – in a particular way now. But that is a judgment call, analogous to that relating to deficit sustainability. And neither borrower nor lender was intending that there by a loan write-down in the future. It’s just a contingency – in the way that a future primary surplus is a contingency.
“I hope you agree that is really the key question.”
It’s an interesting question, but “key” pertains to the original question of inter-generational transfer IMO, which is a pretty simple idea. There are lots of policy issues that are tangential to that core discussion.
I have two questions/observations (inspired by the communication with JKH with whom I mostly agree):
1. Why everybody (except Nick E) is associating “time travel” to government debt only? It seems incomplete because it works similarly with money or other assets?
2. Would it be more general and complete to say that apples eaten by A creates intergenerational distributional deficit that needs to be paid back in the future? This is motivated by the fact that any asset (question 1) can create “burden” and the fact that the similar constraint can be created without any asset by tax-and-transfer.
JKH,
Ahh, I see. Yes, I agree — with you and the rest of the left econ world that all things equal, deficit spending is expansionary and austerity is contractionary. In the original Barro framework, total output in every generation was fixed — all effects were distributional in nature within a generation, and only these distributional effects were studied by Barro’s original work. This even holds in Nick’s examples. I was continuing in that assumption.
Yes, if Generation N runs a deficit and Generation N+L runs a surplus, then all things equal, Generation N + L may suffer a recession or be forced to throw output away due to sticky prices; As you point out, you’ve zapped away some financial assets from the private sector — but no real assets — so unless you assume perfect price flexibility so that prices fall a bit, some output will go unsold in generation N + L.
But, as Nick will be quick to point out, away from the zero bound, the contractionary effect of the deficit may be undone by the CB in generation N+L with a rate cut, which effectively substitutes private debt for public debt. Just as a rate hike can undo the expansionary effects of the deficit spending in generation N. Therefore the CB can keep output constant and you are back to only looking at distributional issues: young versus old, and how a deficit spending in period N can, through a chain of transfers, hurt the young in period N+L (but also help the old by the same amount, as total output is fixed).
The only time this wouldn’t happen would be if the CB tries to counteract austerity at the zero bound — (even though Nick might argue that QE would help, and I would argue that this is equivalent to deficit spending). So generation N+L only needs to kick the can down the road if it is at the zero bound, in which case servicing the debt is not difficult and it should kick the can down the road.
As a side note, this is similar to the MMT argument, as far as I understand it:
when you run the surplus, you are public assets for private assets, causing an increase in the “leverage” of the private sector and generating financial instability. That would be how MMT would describe the Clinton surplus period. Deficit spending decreases financial instability by undoing this swap. Nick is thinking in terms of interest rates, but the MMT are thinking in terms of encouraging an expansion of household debt to replace the lost government debt.
Jussi,
I think there may be a rough analogy with either a corporate balance sheet or a household balance sheet – but VERY rough and imperfect for various reasons.
Running through analogous pieces:
The bare bones consolidated government balance sheet (ignoring real asset claims and the present value of future taxes) is a net liability position (reserves, currency, bills, and bonds), which is equivalent to a negative equity position from an accounting perspective.
A corporate balance sheet typically has a positive equity position with some debt.
A bond financed government transfer payment increases the negative equity position of the government – makes it more negative.
A bond financed corporate expense (NOT an investment) other things equal reduces the positive equity position of the firm – makes it less positive (because the expense is charged to equity). Note that a bond financed expense is ethically questionable in a corporate context.
But these are directionally consistent negative equity effects in the comparison – they make the equity position worse in each case.
As an outlandish example, consider a firm that issues bonds to make executive compensation bonus payments. That would reduce equity, other things equal, analogous to a government transfer payment (slightly different moral modalities!).
Now consider the case where those executives have gone to jail and new management takes over. They judge that they have too much debt and want to pay it down. One way of doing that is to earn their way out of the situation. Make profits and use retained earnings to pay down debt.
Earnings to the firm in this context are analogous to an equity infusion to government via taxation (improving the positive equity position of the firm; making the negative equity position of government less negative).
And the ‘current generation’ (current shareholders) are paying for the escapades of prior ones (previous shareholders who tolerated reckless management) by seeing their profits used to pay down debt instead of being invested.
This is very rough and dirty. I’m sure there are lots of considerations I’ve left out. Of course, a big one is that taxation is imposed and profits are the result of exchange. That said, deflecting profits into the repayment of ‘odious’ debt is a kind of taxation on the shareholders.
And, of course, this means that no future generation is required to be burdened by current deficit spending because only those generations experiencing positive rates will care about the debt, and in those generations the CB will offset whatever primary surpluses they elect to run. This fixes total output of each generation, and then fiscal transfers can undo the inter-generational issues at any point in time. So debt is only a burden when a future generation makes bad policy choices, succumbing to austerity at the zero bound, or failing to address distributional issues arising from the debt. But any generation, when making bad policy choices, will throw some output away, irrespective of what previous generations did.
Then the next order of question is — should we “tempt” future generations by creating the possibility of bad policy choices. Maybe if no future generation ever inherited debt, there would be no debt hysteria and Alan Simpson would be out of a job. Possibly, but that is not a reason to throw output away in the current generation, if it is in a recession at the zero bound.
rjs,
I should have used Godley/Lavoies’s sector balance language rather than the jargon of M…T.
The latter are a rude representation of ideas developed more stylishly by others; no need to knock them further, but sorry I even referenced them here.
Again, I think the main event for the posts of Rowe and Wren-Lewis is the nature of the SCENARIO of a primary surplus as it relates to this question of intergenerational transfer – not to explore antidotes to the risk of having to invoke such a primary surplus.
Wren-Lewis actually notes about keeping to this core point:
“If you are thinking that these redistributions need not occur if the debt is never repaid or the pension scheme never wound up, then we need to get a bit more realistic and bring in interest rates and growth (and the famous r<>g relationship), which these posts of mine (and these at least as good posts from Nick Rowe) discuss. But the idea with this post is to get across in a very simple way how redistribution between generations can work because generations overlap.” (and then he proceeds to link to the posts that deal with those issue)
Right. Let’s focus.
“you and the rest of the left econ world”
WTF?
If I read that correctly, it’s not a very good insight into intellectual motivation in this case.
“If I read that correctly, it’s not a very good insight into intellectual motivation in this case.”
Woah, 🙂 It’s not an insult!
I was saying that Barro + Lucas and the RBC gang don’t believe that deficits are expansionary or that surpluses are contractionary. That is a controversial statement in some quarters. The notion that total output is fixed in each generation is the background of Barro’s original paper as well as of Nick’s series of posts on this. I just assumed that the CB was managing AD except at the zero bound, so that each generation’s total output is fixed, and the only issue up for grabs would be the distribution between old and young within a generation. If you did not care about this distinction and were only concerned about total output in period n, you wouldn’t use an OLG model.
It was not meant as a slur on left econ, as I agree that surpluses are a burden, cet. par. I was pointing out that your interpretation of the model — that total output will be less if a primary surplus is run — is one that was not shared by the other proponents of the “debt is a burden” view, and is not what Nick’s model is claiming. Both the left and right are going to assume total output is fixed except at the zero bound, and at the zero bound, no one cares about rolling over the debt, so in this particular OLG debate, just assume that total output is fixed in every generation, and all you care about are distributional consequences.
If you want to argue that certain surpluses cannot be offset by the CB, my response is that it doesn’t matter. Say the biggest primary budget balance that the CB can offset is B. B << Total Debt. Nevertheless by offsetting B, the CB has set the policy rate to zero, at which point Total Debt is a non-issue. We only care about Total Debt needing to be repaid when interest rates are high because the savings demands are too low. In the other extreme — interest rates too low and savings demands too high — we are happy to keep rolling over the debt.
None of the above is meant to offend/mock anyone. Please excuse my typos also.
“I was pointing out that your interpretation of the model — that total output will be less if a primary surplus is run — is one that was not shared by the other proponents of the “debt is a burden” view”
I’m a bit confused – can you point me to where I said that? I thought I said we could assume it was unchanged. thx..
P.S. I don’t associate myself with the left – hence my reaction
JKH,
I totally agree with your last posts (esp. WTF? :). And comment at 07:14 AM is spot on. And I think I got your corporate example but isn’t it basically just reinforcing, applied on other domain, what Nick/others said in the first place, which I agree. I’m fine with idea that the debt might indicate (but not necessarily) excess savings taking in place which might mean implications in the future (like here: http://monetaryreflections.blogspot.co.uk/2015/02/more-on-inter-generational-implications.html).
But again why always use debt as an example? Let’s say that money is a “burden”.
So I agree but I think it is not the whole picture: it would be more general and complete to say that first generation, A, creates intergenerational distributional implications by its consumption (apples). How how this “deficit” is accounted is an other interesting question but not a primary driver of the implications.
Also I think Nick stretched it later on too far – my interpretation is that he is saying that amount (apples) of future consumption is reduced by debt incurred today:
“we can take apples out of the mouths of yet unborn future generations, and eat them ourselves.”
There I disagree.
Jussi,
Somewhere in these comments I’ve said (twice I think) that money serves just as well as bonds.
The scenario is just changed from the government funding a transfer payment by issuing bonds to one where it makes the payment by printing money (or crediting bank accounts). And it taxes money directly, extinguishing it, without a bond being part of the equation at that end.
So no problem. But that doesn’t change the intergenerational issue. It’s still there.
Bonds are used in the standard scenario because bonds are standard government financing.
Also, the terminology “debt” is slightly problematic. In the case of money finance, the government has the option of redeeming it by imposing a tax liability. In that sense, the government has the option of converting money to “debt” – at the point of exercising the tax option – because it has decided to pay back the “debt” by exchanging it in return for eliminating the money holder’s tax liability. In the case of bond finance, the government normally has the option of rolling the debt over, so that it doesn’t necessarily have to pay back the debt. This also gets into semantics regarding reserves, currency, bills, and bonds and their connection to the idea of liability or debt. The fact is that the government can redeem any of these things (although currency redemption takes some extra willingness on the part of the holder).
For reference:
Farmer Feb 9:
http://rogerfarmerblog.blogspot.co.uk/2015/02/sam-and-janet-learn-about-debt.html
Wren-Lewis Feb 11:
http://mainlymacro.blogspot.co.uk/2015/02/the-burden-of-government-debt-again.html
Both agreeing with the inter-generational effect
JKH: I fished a couple of your comments out of spam, then unpublished the duplicates. Typepad is playing up again!
JKH,
I’m sorry, my money rant wasn’t targeted to you at all. I can see I should have been more carefully.
I only hoped that the others would also acknowledge this and maybe use money as an example in the future. Just for completeness. But I doubt we can read Rowe’s post how money is a burden 🙂
“But that doesn’t change the intergenerational issue”
No it doesn’t but a more general approach, IMO, can and should be made.
Jussi: Money is a whole different kettle of fish, because: money (at least currency) pays 0% interest, so r < g; if you double the stock of money, by giving all the current generation money, the only thing that happens (approx, flexible prices, etc.) is that each $1 becomes worth half as much, so you haven’t given them anything, in real terms.
But I’m not going there in this post. So don’t come back with a lot of questions.
“Even in: a closed economy; with no investment of any kind; and purely lump-sum non-distorting taxes, we can take apples out of the mouths of yet unborn future generations, and eat them ourselves. Time travel is possible.” (with link)
Is there an assumption that there are 100 apples (or the same amount) produced each year for 100 years?
Jussi,
Read through your equity example. With government debt the onus of providing the returns on investment is up to tax payers as a group. Holding a bond gives you a legally protected claim on tax revenue, whatever the source. One has a claim on many.
With equity, that may not be the case. A equity holder may not have the same legal protections. In fact, the onus of obtaining the returns on investment may be totally up to the holder of the equity (for instance someone who is self employed).
And so your question:
“Can we now say that the equity was burden of unborn children?”
If making decisions that increase the realized value of that equity is a burden, then yes equity can be a burden. Some people struggle with monetary / financial decisions. But it is a burden on the individual equity holder, not tax payers as a group.
Nick,
Have you implicitly assumed the FTPL? What I mean by that is; there is either no money in your model(or returns are indexed to inflation) ie an IOU for 10 apples is ALWAYS worth 10 apples upon redemption. I see no reason, if this isn’t the case, how you can determine on who the burden will fall. What if the young could, theoretically anyways, bargain to be paid 2 apple dollars instead of the initial price of 1 apple dollar. Then the old would get claims worth less in real terms than when they bought them. While I don’t disagree that your story is a possible one it is a very specific and maybe FTPL dependent one; up until the point of debt retirement?
Or have I lost it trying to brand monetarists as implicit believers in the FTPL (when confined to operational realities of cb accounting and government finance);)
Please don’t take me too seriously. Plus, I think you and Krugman have the same OLG model, I doubt he would disagree with you since that wasn’t his point. And neither of you think real apples are travelling through time, literally. All apples eaten in the past were grown in the past. Otherwise we should bring better stuff than apples back from the future. Like hoverboards;)
Just to clarify what I mean by FTPL is the case where you assume the government holds the real value of government debt constant.
Miami: not really FTPL. There’s no money in the model, to keep it simple. Bonds are promises to pay a fixed number of apples. Or you can think of them as indexed bonds.
This debate about debt isn’t really a monetarist vs Keynesian thing. But monetarists are definitely not FTPL people.
You could say it’s about two ways of aggregating:
1. Aggregating across people’s consumption in a given year.
2. Aggregating across year’s consumption for the lifetime of a given cohort.
And being able to see the difference. Even if debt has no effect on 1, it can have big ripple effects on 2.
JKH,
OK, then. You agree that total output if each generation is fixed. As a consequence, the operations of taxation and coupon redemption do not increase or decrease the total amount of goods and services available to the private economy — only government consumption of goods and services can do that. Therefore the taxation and redemption of coupons can at best transfer goods and services from one person to another within the same time period. And that type of transfer can be reversed by a second set of taxation and distribution policies. In which case we can combine the two taxation and distribution policies to achieve a single tax/distribute combination which both redeems the coupons and leaves the total consumption available to everyone in the economy unchanged.
Therefore it does not follow that debt is necessarily a burden on anyone. How the government chooses to redeem the debt, if it all, can be a burden on some people alive in the future period in the economy and an offsetting benefit on others alive in the same period. That is why Krugman said “we owe the debt to ourselves”. Now it’s curious why you would focus on the first cohort and not the second — e.g. you could just as well say that “debt is a benefit on future generations (the ones who hold bonds)” — remember total output is fixed so reduced real income to one group necessarily means increased real income to another. I would call that misleading. More importantly, this transfer of real income is not an inevitable consequence of redeeming the debt. It depends on how the government in the future chooses to conduct its taxation policy.
Which of the above statements do you disagree with, and why?
Nick,
It seems to me that if you view bonds as a promise to pay a fixed number of dollars, and not apples, and view that taxation as the confiscation of dollars, and not apples, from the private economy, then the only way you can not subscribe to FTPL is to assume Ricardian Equivalence always holds. You can’t do that and also argue that debt is a burden to future generations. This is fundamentally tied to FTPL, because if the government taxes $1 from one person and stores the dollar in a warehouse, then that may be a transfer from that person to the rest of the private sector (not to the Government), but if it taxes $1 from everyone, then prices will fall and no one is worse off (assuming flexible prices and identical agents).
@Nick
I certainly agree that money might come with other considerations but not within this model.
@Frank Restly
Why A selling bonds instead selling equity to B makes a difference? It is not about nature of ownership.
Nick,
I don’t understand this argument. Everybody dies sometimes. If one generation sells bonds to next generation it gets money in exchange, which then gets bequeathed, or the government confisticates part of the assets via death/inheritance taxes and so retires the debt. And if the younger generation are buying the assets then they are getting the assets in return. Sure this has generational transfer aspects among the LIVING, but not between generations.
@ Nick
I understand the point you’re making and agree that the sequence of events described in your model leads to a cohort getting the short end of the stick. The metaphor I like is one of airplanes overpacking and weight capacity limits. Subsequent baggage is delayed until the route is cancelled or there’s a flight with spare capacity.
I think that metaphor works.
Delayed baggage still arrives (unless the route is cancelled) if that’s not clear from reading my previous post.
Ugh sorry I don’t think what I wrote is written clearly. Should read:
airplanes, people over packing and weight limits.
I’ll quit now.
I was just teasing about the FTPL thing. Sorry if I opened up a can of worms 🙂
Nick,
Out of interest, do you agree you model works with equities?
Miami: Ah! I sort of get that metaphor. Which brings me to:
rjs: “…remember total output is fixed so reduced real income to one group necessarily means increased real income to another. I would call that misleading. More importantly, this transfer of real income is not an inevitable consequence of redeeming the debt. It depends on how the government in the future chooses to conduct its taxation policy.”
OK. Sit down with pen and paper, and try to come up with a simple numerical example where aggregate consumption in any year is fixed, where the government pays off the debt, and no individual has reduced lifetime consumption as a result.
The only way I understood this stuff myself was by doing something like that. Bob Murphy was the same, except he used an excel spreadsheet.
reason: suppose the old sell their bonds to the young, in exchange for the young’s consumption goods, then consume those consumption goods, then die, and bequeath nothing.