Robert Waldmann (and others) on Dean Baker (and hence Paul Krugman) and the burden of the debt on future generations

January 22, 2012 · by wciecon · in Fiscal policy, Macro, Nick Rowe · 159 Comments

Robert Waldmann here.

Bob Murphy here.

Richard Williamson here.

I've done my bit, and have failed. Robert Waldmann may succeed. If Robert Waldmann fails too, that would be very bad for the credibility of economics blogging.

159 comments

Patrick · January 24, 2012 - 12:48 pm · Reply→

I’m with Noah Smith on this. Sure, Nick is obviously right as far it goes, but it seems to me we miss something big by just thinking in terms of the oldsters borrowing to fund a giant kegger and sticking the kids with the bill. Along with getting the bill, the youngsters get a functional civilization (roads, bridges, health care system, parliament building, etc). There’s value in that.
Nick Rowe · January 24, 2012 - 1:15 pm · Reply→

Patrick: I’m fine with that way of thinking about it. It all depends what the government spends it on, and how much that benefits the kids. Kegger or roads? Is the kids’ value of the road bigger than the value of their bill? If so, no worries.
“Sure, Nick is obviously right as far it goes,…”
If I could just convince everyone of that, I’d be done. But so many are still saying I’m obviously wrong!
anon · January 24, 2012 - 1:22 pm · Reply→

Nick, http://delong.typepad.com/sdj/2012/01/is-there-any-reason-to-think-that-the-interest-rate-on-us-government-debt-in-the-future-will-be-greater-than-the-growth-rat.html
This is what I’m asking, why would the governement ever borrow at rates above the growth rate except to combat inflation?
Your assumption is that they ALWAYS do so. You call borrowing at BELOW the growth rate a ponzi scheme witch is always sustainable, but if you always borrow at ABOVE the growth rate isn’t it always unsustainable?
Wouldn’t borrowing at a level consistent over time with the growth rate still require taxes to pay back bondholders? There would just be no net (monetary) burden, what matters is how we spend the money.
(In your first post you write that the government borrows money from cohort 1, then gives the borrowed money right back to cohort 1? This seems obviously corrupt and a bad way to spend money. Wouldn’t a non-corrupt governement buy something? Or have a multiplier-type rationale for making a transfer payment?)
primedprimate · January 24, 2012 - 1:28 pm · Reply→

Nick, I know from your earlier responses to my comments that you agree with me that owners of government bonds may very well bequeath their bonds to their children even in a non-RE world. For instance, I think that it is quite possible that the wealthy own a majority of the bonds and most of the wealthy bequeath a significant portion of their wealth (including bonds). I also think RE is not true, because many people (predominantly those less wealthy) are credit constrained. All the repeated assertions by you and others that RE needs to be literally true for the debt to not be a burden, create a false dichotomy, “either debt is a burden or bond financed lump-sum tax cuts/transfers have no effect.” The conditions required for debt to not be a burden are much much weaker than the conditions for full RE.
I think only empirical work can show to what extent government bonds are bequeathed rather than sold to the younger generation.
Maybe the issue is that typical economic modeling assumes homogeneous agents and it is hard to informally analyze an economy where some people (bondholders) leave bequests, while others are credit constrained.
I share your sense of frustration – it bothers me too that Krugman has not responded to your posts and I understand it completely if you are choosing to invoke RE to strengthen your rhetoric in order to provoke a response.
Ralph Musgrave · January 24, 2012 - 2:36 pm · Reply→

Min, This whole argument is based on the assumption that government does NOT default. That’s a reasonable assumption because small changes in the extent to which oldies sell rather than bequeath bonds to the next generation won’t have any effect on the liklihood of government defaulting.
Nick Rowe · January 24, 2012 - 2:57 pm · Reply→

anon: why should the government borrow? because it doesn’t want to cut spending or increase taxes. It wants to postpone doing that to the future, for good reasons or bad.
“(In your first post you write that the government borrows money from cohort 1, then gives the borrowed money right back to cohort 1? This seems obviously corrupt and a bad way to spend money. Wouldn’t a non-corrupt governement buy something? Or have a multiplier-type rationale for making a transfer payment?)”
It’s equivalent to a bond-financed tax cut. Any time the government buys things and borrows to do so, it’s equivalent to: an increase in spending matched by an increase in taxes; plus a cut in taxes matched by borrowing.
primed: back a couple of weeks back I thought I had figured out some subtle way that all bonds could be bequeathed while RE could still be false. Now I’ve forgotten what I figured out. Meanwhile, I am desperately trying to explain the whole thing to a bunch of people who haven’t even gotten to square one. They desperately want to believe that the debt is a 0% burden, and they desperately want to believe that RE is 100% false, and they can’t figure out any relation between the two.
anon · January 24, 2012 - 4:29 pm · Reply→

Nick, I’m totally confused.
1. I keep asking about borrowing on interest rates not LESS THAN or MORE THAN but EQUAL TO the growth rate. And you keep on responding with no explicit references to interest rates. Specifically, do you think EQUAL TO interest rates are still a burden? And what rationale does a government have for keeping interests rates PERMANENTLY above the growth rate? (Does your apple model require the permanently above interest rate to get its result?) (I posted a link to a Delong post where (I think) he shows the interest rate staying even with the growth rate over time. Which might be why Krugman and Baker might make that assumption.)
2. I was confused by the rationality/realism of your apple model. You had no taxes (plus no trade) to begin with and the government borrowed then gave what it borrowed right back to the people it borrowed from at no cost. (I still don’t understand how this isn’t corrupt, stupid, and/or unrealistic.) A bond-financed tax cut is presumably done to stimulate trade. Would a government borrow from some people and turn around and cut only the taxes of the people they borrowed from?
Bob Murphy · January 24, 2012 - 4:33 pm · Reply→

Nick Rowe is apparently the muse for my outbursts of fiction. I dedicate this one to the people who are now telling Nick, “Duh! Nobody is arguing with that Nick. But we’re talking about debt in the real world we are actually facing.”
Nick Rowe · January 24, 2012 - 4:44 pm · Reply→

Bob: I think I now have someone saying that no MMTer ever said anything like that too. Read the comment thread here, especially between me and Jos.
Does this mean you, me and Robert Waldmann have won? I expect it does. But God it’s been an exhausting slog!
Min · January 24, 2012 - 5:07 pm · Reply→

Ralph Musgrave: “This whole argument is based on the assumption that government does NOT default.”
Both Nick Rowe and Bob Murphy present scenarios where eventually the gov’t is not able to roll over the debt and so must either raise taxes or default. They are not assuming no default. It is that crisis that makes the burden of debt real. And that burden falls on the younger generation, who have bought gov’t bonds but do not get the expected payout.
It seems to me that we have a test of their scenarios in real world gov’t defaults. Upon whom does the burden of default fall?
Nick Rowe · January 24, 2012 - 5:50 pm · Reply→

And from Stag Hunt.
Richard Williamson · January 24, 2012 - 6:06 pm · Reply→

Nick: I’ll believe you’ve really “won” when everyone then goes on to accept Steven Landsburg’s point that if the debt isn’t a burden due to bequeathing (i.e. older cohorts not changing their total consumption in response to the debt), then it doesn’t matter whether the debt is held domestically or internationally. The national/international distinction only makes sense when making the (invalid) identity argument. And I’ll be satisfied if people accept that marginal increases in public debt must cause marginal increases in bequests in order for there not to be a burden (which somehow seems less plausible when you put it that way, at least to my mind) (excluding sustainable ponzi financing, of course).
I’d also really like Paul Krugman to admit that his argument was straightforwardly invalid, and requires an assumption he has criticized in the past as being a ‘dubious doctrine, even when done right’ in order to get to the same result. But I don’t have high hopes for that.
Kevin Donoghue · January 24, 2012 - 6:10 pm · Reply→

Nick,
Yes, you’ve won but before you get too carried away please take a look at what you’ve won. As I said upthread, Dean Baker acknowledges that there are possible worlds in which what you and Robert Waldmann say is true.
This might be a bit less than you were hoping for. But Robert Waldmann will settle for it.
Ramanan · January 24, 2012 - 6:20 pm · Reply→

Nick,
“Ramanan: OK. I would say it is roughly right to say they are born with a debt. The twist is that they might be able to pass on that debt, plus interest, to the next set of newborns.”
Reminds me of the book “A Brief History of Time” by Stephen Hawking which begins like this:
“A well-known scientist (some say it was Bertrand Russell) once gave a public lecture on astronomy. He described how the earth orbits around the sun and how the sun, in turn, orbits around the center of a vast collection of stars called our galaxy. At the end of the lecture, a little old lady at the back of the room got up and said: “What you have told us is rubbish. The world is really a flat plate supported on the back of a giant tortoise.” The scientist gave a superior smile before replying, “What is the tortoise standing on?” “You’re very clever, young man, very clever,” said the old lady. “But it’s turtles all the way down!” ”
🙂
Steve Roth · January 24, 2012 - 6:40 pm · Reply→

Here’s why I’m still unsure and/or confused:
The model used to demonstrate future burden assumes fixed or arbitarily increasing productivity and zero real growth.
But isn’t that the whole crux? Doesn’t a “burden” ultimately mean less apples eaten/produced, supplied/demanded per person in the future — lower real GDP (growth)?
Isn’t the question whether economic growth — change in production/consumption per capita — will be less in toto over x years if the government borrows today? Isn’t that what determines whether future periods/generations will have to consume less — be “burdened”?
It doesn’t seem like accounting can answer that, absent behavioral assumptions.
(And yes I’m not counting the potential benefits of spending the borrowed money; call it net zero.)
Nick, if I’m totally not getting this don’t spend a lot of time. Just tell me and I’ll go do more mental pushups…
Nick Rowe · January 24, 2012 - 6:59 pm · Reply→

Steve: you are not getting it. But I would like you to get it.
Hold GDP=Consumption constant. Future generations can still consume less. Sounds totally impossible, right? Sounds like I am just doing the adding up wrong, right?
The government gives each young person in cohort A a bond worth 100 apples, paying 10% interest per generation, which he keeps till old. The A grow old. Each young person in cohort B gives 110 apples to each old person in cohort A (in exchange for the bond+interest), who eats them then dies. B grow old. Each young person in cohort C gives 121 apples to each old person in cohort B (in exchange for the bond), who eats them then dies. Then the government decides the debt is too big, taxes each C enough apples to buy back the bonds. So the young C eat 121 fewer apples when young. Then everything goes back to normal.
Nick Rowe · January 24, 2012 - 7:25 pm · Reply→

Richard: it is very rare for people (let alone economists!) to do what Bob Murphy did, and say “Hey look at what I wrote in my textbook! Man was I wrong!”
The best most of us mortals manage is to say “Hmmm, OK, maybe there’s another angle on this” (usually the best I can manage), or “I didn’t mean that!”, or just stop saying what we used to say.
Let’s see if Paul Krugman ever says again: “What I was actually saying, of course, is that debt is a liability that we pass to the next generation — but it’s also an asset that we pass to the next generation.”
You have a good point on the debt to foreigners point. But I confess that I have difficulty keeping my head straight on that one too.
Everyone: Check Richard’s ,latest by the way.
(I am so proud of myself: I have finally learned how to embed links! Applaud everyone! It’s taken me 3 years! Bet I forget tomorrow).
anon · January 24, 2012 - 7:27 pm · Reply→

Nick, I now see you replied to me twice earlier. (For some reason I could only see one.)
Yet in that post you still have me saying, “why can’t the interest rate permanently below the growth rate.” When I’m saying, “why can’t the interest rate be equal to the growth rate.”
I’m saying to have the interest rate permanently above the growth rate is just as unfair an assumption as having it permanently below.
I’m saying Krugman and Baker might be assuming it’s equal to the growth rate. And I posted a link to Delong showing that in reality it has been (over time) just that.
Also while money illusion could occur, inflation could also occur. But this does not mean inflation will shift the burden onto the old. There could be equal and offsetting inflation and money illusion.
Nick Rowe · January 24, 2012 - 7:36 pm · Reply→

anon: Exact equality between r and g is something that could happen with probability p=0.000000000000000001. I don’t have the energy to describe what happens there accurately or correctly. It’s a borderline case. But no, that’s not what Paul Krugman is talking about. Because r relative to g only matters for whether or not taxes need to increase. He talks about taxes needing to increase. once he does that he’s in my world, and my argument takes over.
Bob Murphy · January 24, 2012 - 7:58 pm · Reply→

@Richard Williamson: Yes you’re totally right, the only way for people to really “get” all this if when they see that foreigners holding the debt isn’t really the issue. So in one sense, we have Landsburg to thank for that aspect of it, but on the other hand, it’s infuriating that Landsburg (to this day if I’m not mistaken) thinks Krugman never made the mistake Nick and I are attributing to him. I truly believe it’s because Landsburg is a mathematician first, economist second, and so he correctly thought about this stuff in terms of intertemporal budget constraints or something, rather than in terms of faulty intuition. So that’s why Landsburg doesn’t see how obvious it is that the foreigner non sequitur (or Krugman’s thought experiment involving the Woody Allen reference, where some dictator out of the blue creates new debt but hands the bonds out to everybody) show Krugman was committing the fallacy that Nick and I both suffered from in our past. (Nick’s distant past, my past from November.)
@Nick’s remaining critics: I understand why it seems weird to you that Nick makes some ridiculously strong assumptions, and keeps walking around thinking he has proved a general result. But the reason is that Dean Baker and Paul Krugman based their arguments on an “insight” that was totally wrong. They said things like “it’s especially nonsensical” to worry about burdening our children with government debt, and their reasoning is wrong. The easiest way to show that it’s wrong, is to isolate away all the extraneous things that could mask either the insight or the fallacy, depending on who’s right. So we don’t have foreign trade in Nick’s model; nobody got mad at him for that. We don’t have income taxes in Nick’s model that might cause people to harvest fewer apples in period 3; nobody got mad at Nick for that.
But by the same token, we can’t have the government building bridges in period 1 either, because that might mask the gross harm of the debt to future generations. On net there could be a benefit, and Nick wanted to clear that away.
Remember, Dean Baker and Krugman were saying that if we assume away foreigners, distortionary taxes, etc., then IN PRINCIPLE the debt can’t burden people in the future. Go read Krugman’s post with the Woody Allen reference. It’s a very compelling perspective, IF our grandkids just wake up one day and find themselves holding bonds–as they would in Krugman’s eccentric dictator scenario.
But if even one of our grandkids had to reduce his earlier consumption in order to be holding that Treasury bond when the tax man goes around to collect its interest payment, then the people alive at that point on net suffer a burden from the debt. It’s true, maybe the burden is offset by whatever the gov’t originally spent the borrowed money on. But the burden is there. Hence, Dean Baker and Krugman were using a bogus argument to tell people worrying about the debt burden was nonsense.
anon · January 24, 2012 - 8:09 pm · Reply→

Nick, I’m not saying exactly equal, but roughly equal over time.
If growth goes up (roughly, over time) taxes would need to (roughly, over time) increase to pay off bond holders.
Steve Roth · January 24, 2012 - 8:24 pm · Reply→

Nick: ‘kay: so in some future period(s), the older people eat some of the younger people’s apples — while the younger people are standing there watching them do it! (Fools.) That’s the OLG.
Consumption in that period (and the next) is unchanged, but since it’s shifted from young to old in that period, consumption per generation is changed. The young fools eat less apples in their lifetime.
So we’re not necessarily being selfish at our children’s expense (damn, that’s a relief!), but if growth is less than interest rates so increased taxes are eventually necessary, we’re forcing some future cohort(s) to be selfish at their children’s expense. (Greedy bastards.)
I think I got it. That right?
Nick Rowe · January 24, 2012 - 8:53 pm · Reply→

Steve: you are nearly there. But not quite.
The government gives us a bond-financed transfer payment. We eat more when we sell the bonds to our children. The government then rolls over the debt, borrowing to pay the interest. Our children eat less when young but more when old, but are not worse off over their whole lifetimes (the interest is enough to compensate them for having to postpone part of their consumption until they are old. Then, after several generations, the government decides to increase taxes on our great-great-great grandkids. Who get screwed. They eat less over their lifetimes.
Nick Rowe · January 24, 2012 - 8:54 pm · Reply→

We are being selfish at our great great great grandkids expense.
Bob Murphy · January 24, 2012 - 9:34 pm · Reply→

The reason I am continuing to post here, is that (in contrast to many of the naysayers at my blog) the people here seem genuinely interested in figuring out what Nick is saying. So let me try offering more clarification:
Strictly speaking, Nick and I are being sloppy when we say “government debt can burden our great-grandchildren.” Krugman and Baker are correct that debt per se doesn’t hurt society; every person’s debt corresponds to some other person’s asset (or even the same person simultaneously, if it’s a taxpayer who has Treasury bonds in his portfolio).
Really what the Rowean (?) position is, is that future taxation necessary to service/retire the debt we create today imposes a gross burden on our grandkids. When the government spends money today, it can finance it by taxing the people alive today. In that case, clearly our grandkids aren’t paying for it. However, another mechanism is for the government to borrow the money today. If that debt gets rolled over continuously–with no taxation to service it–and then there is a one-time big tax in the year 2100 to pay the whole thing off at that time, then the taxpayers who are alive at that moment bear the full brunt of our spending. The bond market simply allows the government today, to bring that discounted future tax revenue forward in time.
@Nick Rowe, this is where Landsburg’s point about the foreigners comes into play: Once we realize that it’s not the debt per se that is the issue, but the taxation necessary to finance it, then we should see that the identity of the people who hold the Treasury bonds at that time (and hence get paid the interest/principal) is irrelevant. Look, Nick, you have been hammering home the point that even if the debt is held domestically, then our grandkids are still worse off collectively, because the ones holding the bonds earlier in their lives reduced some of their consumption to be in the position of holding those bonds. So, you know that the mere fact that the interest/principal payments go into their pockets in the year 2100 (or whatever), doesn’t mean that the people alive in 2100 have a wash.
But Landsburg can be taken to say: By the same token, the Americans alive in 2100 aren’t any worse off, if you now inform me that their interest/principal payments get funneled into the pockets of Chinese investors. Why? In this scenario, it means the Americans alive in 2100 never had to earlier reduce their consumption. No, the Chinese investors did.
So Landsburg is really saying, “Given that Americans living in the year 2100 will be taxed to service government debt, those people are not affected by whether the debt has been rolled over via American or Chinese investors.”
However, where Landsburg went terribly awry (in my opinion) is when he concluded, “Therefore, Krugman was basically right on all this stuff, it’s just that ‘we owe it to ourselves’ even if we owe it to the Chinese.” That characteristically Landsburgian conclusion was wrong (I claim) both in letter and in spirit.
Nick, please tell me you wholeheartedly agree with the above, otherwise we are in trouble.
Steve Roth · January 24, 2012 - 9:36 pm · Reply→

“our great-great-great grandkids…eat less over their lifetimes.”
Do all the apples that they don’t eat lie on the ground and rot? Or are they eaten by their parents, while the kids stand there watching (making consumption in the period unchanged)?
Bob Murphy · January 24, 2012 - 9:37 pm · Reply→

Oops one little footnote to my discussion of the foreigners vs. domestic financing thing: To make that argument work perfectly, you have to assume that the investors don’t benefit from the possibility of investing in government debt. I think that’s what Landsburg implicitly assumed. I.e. the government doesn’t do investors favors by providing a debt market.
If investors do benefit from that possibility, then it gets a little trickier. But if you make the weaker assumption that the Treasury auctions don’t discriminate, and that Chinese investors have just as much freedom to bid on Treasury debt as American investors, then I think we’re back at Landsburg’s result.
Nick Rowe · January 24, 2012 - 9:45 pm · Reply→

Bob: I read it twice. I wholeheartedly agree. You explain the foreign-held debt issue clearly.
Yes, we have very good commenters here. Like Steve (a good blogger in his own right).
Steve: “Do all the apples that they don’t eat lie on the ground and rot? Or are they eaten by their parents, while the kids stand there watching (making consumption in the period unchanged)?”
No apples ever rot. Consumption in each period is always unchanged. The kids’ apples are eaten by their parents. But the parents are just being compensated for their apples that were eaten by the grandparents, when the parents were young.
Nick Rowe · January 24, 2012 - 9:48 pm · Reply→

Landsburg’s “mistake” is that he automatically assumed Ricardian Equivalence, and assumed Paul Krugman did too.
anon · January 24, 2012 - 11:03 pm · Reply→

Nick, did you go to the Delong link I posted? Doesn’t that show an interest rate roughly equal to the growth rate for a fourty year period?
Wouldn’t taxes have to rise with the growth rate equal to the interest rate? Because as the economy grows so do interest payments.
So Krugman could be assuming they’re (roughly) equal (over time) and talk about raising taxes? Because the growth and interest cancel out.
Bob Murphy · January 24, 2012 - 11:35 pm · Reply→

Anon, for me that Brad DeLong post means (a) DeLong knew that Krugman/Baker had stepped in it, so (b) he was trying to give them an out, by showing that in practice their conclusion was OK.
Bob Murphy · January 24, 2012 - 11:39 pm · Reply→

Anon: Another thing, more directly answering your question (and I think Nick would agree): No, Krugman was NOT assuming that the debt wouldn’t be a burden, because the economy would grow faster than the principal. (I.e. g higher than r.) It’s true, Krugman was aware of that possibility, and perhaps in his mind that made him comfortable saying the debt was no problem.
But clearly his main point–echoing Baker–his “ace in the hole” if you will, was the fact that Americans largely will owe the debt to themselves in the year 2100 (or whatever). Krugman even talked about the taxes necessary to service the debt, and how those tax payments would just go right into the pockets of other Americans.
Once the government starts taxing to service (not even to repay) the debt, then the people alive at that time have a gross burden imposed on them. (It could be counteracted by the benefits of the space program, interstate highway system, freedom from Nazi rule, or whatever the deficits had paid for, back when their grandparents’ government spent the money. But the taxation is a burden on them.)
So no, it is crystal clear that Krugman’s case was NOT resting on the assumption that g > r. In that scenario, the government never has to impose a tax to deal with the growing debt. That’s why the debt isn’t a burden in such a scenario, because nobody is ever taxed to service it.
anon · January 24, 2012 - 11:59 pm · Reply→

Bob, I’m not saying Krugman was assuming growth higher than interest. I’m saying Krugman and/or Baker (maybe) were assuming growth (roughly) equal to interest.
Taxes would still rise, but only as growth and interest rates rise.
If Delong showed that in practice this assumption is true, why wouldn’t Krugman assume it?
JKH · January 25, 2012 - 12:57 am · Reply→

“The government gives each young person in cohort A a bond worth 100 apples, paying 10% interest per generation, which he keeps till old. The A grow old. Each young person in cohort B gives 110 apples to each old person in cohort A (in exchange for the bond+interest), who eats them then dies. B grow old. Each young person in cohort C gives 121 apples to each old person in cohort B (in exchange for the bond), who eats them then dies. Then the government decides the debt is too big, taxes each C enough apples to buy back the bonds. So the young C eat 121 fewer apples when young. Then everything goes back to normal.”
As you may recall, I’ve agreed with your particular model from the start.
But the following occurred to me (rather suddenly):
I’m assuming cohort C is taxed 133 apples in your model.
(I’ve assumed a final additional year of compounding, but the actual number doesn’t matter for my illustration here. I suppose they could be taxed 121 at the beginning, or 133 at the end. They’re down 121 in actual start of period apples, because they bought the bonds. They’re down 133 in opportunity cost end of period apples, because they didn’t end up selling their bonds to the next cohort. Either way of looking at it, they’re down. I’ll use 133, because I want to compare the actual case in which they are taxed, with a counterfactual case that continues the previous pattern of selling bonds at the end of the period.)
So, under that assumption, as per your model, C pays the tax with the bonds, both valued at 133 at the end of the period. And C is now down 133 apples in end of period opportunity cost (121 apples in actual start of period cost), because it didn’t get to sell the bonds to a subsequent generation D.
That’s the argument in a nutshell for the future generation burden, I believe.
But at the same time, cohort D, the one that follows C, no longer pays C the 133 apples it would have paid had C not been taxed and sold the bonds to D.
That means cohorts D and beyond in total are up 133 apples in the scenario where C is taxed, compared to the scenario where C is not taxed.
When C is taxed, C is worse off by 133 apples, but cohorts D+ in total are better off by 133 apples.
The net result for cohorts C, D, and beyond is that in total they are no better or worse off, whether or not C is taxed.
So there’s no apparent burden for future generations, net.
(But there is a distribution issue among future generations.)
What’s the response to that?
Is your response still bullet-proof against Krugman et al?
Would they accept a “proof” that it’s sufficient to demonstrate that a single generation C loses due to the tax, even though all subsequent generations in total gain because of it, to the degree that they perfectly offset C’s loss with their gain?
Do you agree that all generations beyond C, in total, gain because of the tax to C?
And that all generations beyond B, in total, are unaffected in total because of the tax?
If not, why not?
And if you agree with that, and that therefore it’s only a distribution issue, isn’t that the kind of distribution issue that’s covered by Krugman’s general caveat regarding distribution (whether or not he thought of this aspect in particular)?
Bob Murphy · January 25, 2012 - 2:12 am · Reply→

Anon:
You agree that both Dean Baker and Krugman talked about there being distributional issues within a given (future) generation, right? It’s what they called “a different kettle of fish” or something. So they are clearly intending their argument to cover a scenario where the government is engaged in taxing, in order to at least partially service (perhaps not pay down) the debt. Their argument was wrong, for the reasons Nick gave.
Now you seem to be asking a related question: If g roughly equaled r over a long period, would there ever be any need for the government to tax people ever to service the debt? No, there wouldn’t. E.g. let’s say the gov’t has a debt-to-GDP of 50% in the year 2012. For a few decades g < r, so that debt-to-GDP rises to (say) 110%. But the government doesn’t tax anybody to service the debt, it just keeps letting it roll over and grow exponentially. This eventually is OK, because during the next few decades, g > r, and the ratio falls back to 50%. Yeah, that could happen cyclically, and so long as g = r in the long run, there would be no reason for the system to blow up.
However, part of the problem (and this is what I think Waldmann was getting at in the first comment at DeLong’s post) with this sort of thinking, is that r isn’t independent of the debt size. If the debt gets to be 110% of GDP, investors might start getting nervous and insist on higher interest rates to keep rolling it over.
Richard Williamson · January 25, 2012 - 2:18 am · Reply→

@Nick
If Krugman ever says something like that again, tag me in…
@Everyone
This is the simplest way I’ve come up with to make the key point: bonds are an asset that derive their value from the ability to tax. Therefore, you are going to start getting distribution effects when you start changing the tax base (people being born) without changing the bondholder base. Work from there.
Bob Murphy · January 25, 2012 - 2:21 am · Reply→

JKH wrote (to Nick):
And if you agree with that, and that therefore it’s only a distribution issue, isn’t that the kind of distribution issue that’s covered by Krugman’s general caveat regarding distribution (whether or not he thought of this aspect in particular)?
No, absolutely not! No quarter for Krugman!
I don’t know why this is so hard for people to accept. Dean Baker and Krugman screwed up, period. I am 99.99% confident of what led them to write the false things they wrote, because I wrote the exact same falsehoods in my textbook treatment of this issue.
They weren’t thinking of “future generations” in terms of people living for several periods down the road. Instead, they were thinking in terms of single snapshots of a given future period. They were looking at, say, the year 2105, and thinking that whatever the level of the federal debt in that year, it wouldn’t make the people living in 2105 richer or poorer collectively. Sure, some people might get money taken from them in taxes to service the debt, but other people (who would also be Americans, to the extent that the debt were held domestically) would get those payments put right back into their pockets. So yes, Baker and Krugman concede there is a “distributional” issue, but they are talking about the moment-to-moment redistribution occurring when Peter is taxed to make an interest payment to Paul. They weren’t at all thinking that, say, the people who live from 2100 – 2160 could all be be made poorer, while the people from 2161 onward could be richer.
Anyway, we can modify Nick’s example to show that even on your terms, Krugman is wrong. Have an infinite number of generations. The debt is continually passed down, with every young person being taxed enough to cover the interest payment. I.e. the principal doesn’t grow.
So in this revised scenario, the first generation gains, everybody else for the rest of time loses. Since this is theoretically possible in a world satisfying Krugman’s assumptions (no distortionary taxes, no foreigners holding the debt, etc.), he is wrong when he concludes that “it’s especially nonsensical to talk about burdening our children.”
Oliver · January 25, 2012 - 5:46 am · Reply→

seems the spam filter ate my first attempt at this comment:
So in this revised scenario, the first generation gains, everybody else for the rest of time loses.
In my opinion, this clearly violates your own condition that there are always 100 apples consumed per period (assuming people don’t live forever).
But to quickly recap:
Krugman reads Bob Murphy’s table horizontally and concludes ‘there is no burden, period’.
Rowe & Murphy read it vertically across overlapping generations and say ‘there can be, bahhhh’.
Some thoughts:
Both cohorts (pardon the pun) assume Ricardian Equivalence is bunk. With RE, there can be no burden, nor any benefit, ever, by definition. I.e. there are no options to choose from. Without RE, on the other hand, the options and variations become infinite. That includes an infinite number of possibilities that makes one or the other cohort worse off over their lives. Finding one such example out of an infinite pool does not seem particularly ingenious.
Do you deny that for every such particular case, there exists a fiscal operation that could nullify the incurred ‘injustice’? E.g. the overendowed age group could be taxed to subsidise the underendowed one?
And in what meaningful sense is the inequality in consumption that results from your initial act of borrowing different, say, from inequality arising through any other course of capitalist exchanges without government? I don not hear you screaming for a 100% inheritance tax, free schooling from crib to grave or any other egalitarian scheme that would guarantee that every generation consumes 50+50 apples, no matter what. Why the outrage? Why the ‘Eureka’ moment?
Krugman is right in the sense that there is nothing that the initial borrowing sets in motion that cannot be rectified – assuming time doesn’t end suddenly.
You are right in the sense that an act of borrowing sets in motion a chain of events, which, if left unattended, may have adverse effects for some and positive effects for others over time. Considering that logic applies to absolutely everything we do or don’t do in life, the attention these posts have garnered seems somewhat exaggerated.
Oliver · January 25, 2012 - 5:55 am · Reply→

Nick, seems the spam filter ate my comment multiple times. either that or I’m on your black list.
Nick Rowe · January 25, 2012 - 6:05 am · Reply→

Oliver: sorry. Our spam filter disagrees with you. I retrieved all 3 versions, then unpublished the first 2 versions. Let me know if it’s still not right.
Nick Rowe · January 25, 2012 - 6:24 am · Reply→

Oliver: “Krugman is right in the sense that there is nothing that the initial borrowing sets in motion that cannot be rectified – assuming time doesn’t end suddenly.”
I disagree. Once the first cohort has eaten extra apples and died, it’s too late to turn back the clock and revert to the status quo. All you can do now is decide which of the current and future cohorts will pay for those apples with higher taxes.
“So in this revised scenario, the first generation gains, everybody else for the rest of time loses.
In my opinion, this clearly violates your own condition that there are always 100 apples consumed per period (assuming people don’t live forever).”
I disagree. It’s still C(young)+C(old)=100 for all time. There are two ways in which future cohorts could be worse off:
1. Same lifetime consumption, but lower lifetime utility, because they are consuming (say) 40 when young and 60 when old when they would prefer a {50,50} package.
2. Lower lifetime consumption, but lifetime consumption asymptotically rising towards 100 as we go further and further forward in time. E.g. cohort C consumes {20,70}, cohort D consumes {30,65}, cohort E consumes {35,63) etc. (Something like that.)
Nick Rowe · January 25, 2012 - 6:31 am · Reply→

Bob: “However, part of the problem (and this is what I think Waldmann was getting at in the first comment at DeLong’s post) with this sort of thinking, is that r isn’t independent of the debt size. If the debt gets to be 110% of GDP, investors might start getting nervous and insist on higher interest rates to keep rolling it over.”
It’s not just the increased risk of default and nervous investors. Assume zero default risk. People won’t want to consume 10 apples when young and 90 when old (which is what is implied by a high debt/GDP ratio) unless you offer them a very high interest rate to persuade them to defer consumption.
JKH · January 25, 2012 - 6:39 am · Reply→

Bob Murphy wrote:
“I don’t know why this is so hard for people to accept”
The record of entries on this blog will show that I agreed with Nick’s model before you did. And I’m just suggesting some testing of it around the edges that I haven’t seen yet. It would be nice to tighten up the argument where possible, given the amount of time you guys are still spending defending it.
“Anyway, we can modify Nick’s example to show that even on your terms, Krugman is wrong.”
I don’t think your suggestion does that. Running cash interest payments would have to be inter-cohort for your revision to work. There’s every reason to observe they’re intra-cohort. So they’re a wash in terms of the specific cohort burden model. Besides, Nick effectively capitalized the interest payments in his model by including them in the debt accumulation that’s sold cyclically from old cohorts to young cohorts. And that’s consistent with bond pricing mid-coupon in the real world. Any way you cut it, it’s the capital value of the bond that threatens in terms of a potential generational burden, not the cash interest payments on it.
I’m just trying to help here to tighten up your arguments. Cut me some slack. You have to be quite specific about your definitions and assumptions in these models and exactly what it is you think you have proven.
Kevin Donoghue · January 25, 2012 - 6:43 am · Reply→

Oliver: “the attention these posts have garnered seems somewhat exaggerated.”
To be fair, a very small number of people have got rather excited. That’s all really. Certainly I agree with Dean Baker that, if we’re worried about the burdens we are passing on to future generations, there are lots of things which should have higher priority. The youth of 2112 will presumably be able to vote for default on the national debt. They probably won’t find a way to put the oil back under the ground or re-freeze the Arctic Sea.
Having said that, I still think it would be nice to get the theory right.
JKH · January 25, 2012 - 6:47 am · Reply→

Here’s another test around the edges.
You all should be a bit more careful before you bash MMT full bore on this issue.
The MMT literature is rampant with the idea that budget surpluses are almost always evil.
That means that paying down debt is almost always evil.
And that’s inherently sympathetic to the view that taxes to pay down debt are a burden.
If you pay close attention to MMT, I think you’ll find that the assumption of no future burden is consistent with an assumption of not taxing to pay down debt. Again, much of the MMT story leans heavily along such a theme.
And I doubt you’ll find Mosler contradicting this.
This is from memory; I’m not about to look for the “proof” of it; but I’ll stand to be corrected on it, even immediately if the counter punch is available.
Nick Rowe · January 25, 2012 - 6:48 am · Reply→

Kevin: “The youth of 2112 will presumably be able to vote for default on the national debt.”
Too late. We’ll all be dead by then. They won’t be able to get back the apples we’ve eaten. But yes, it’s not the only intergenerational issue.
Gotta go teach the next generation.
K · January 25, 2012 - 8:19 am · Reply→

Nick: “Too late. We’ll all be dead by then. They won’t be able to get back the apples we’ve eaten.”
No. But they can choose to consume their own production. It’s their aging parents (or the creditors) who’ll be left holding the bag having given their own production to their parents in the hope of doing the same thing to their kids. In both cases it’s their own fault: the parents should have defaulted themselves, rather than attempting to perpetuate the fraud onto the next generation; the creditors bought the bonds knowing full well the next generation had no moral obligation to make good on them given that they never consented to nor benefited from the initial issue.
Oliver · January 25, 2012 - 8:26 am · Reply→

I disagree. Once the first cohort has eaten extra apples and died, it’s too late to turn back the clock and revert to the status quo. All you can do now is decide which of the current and future cohorts will pay for those apples with higher taxes.
You mean which current or future generation will consume less apples? That’s the same in your model, and yet it’s a condition that consumption is alway 100.
Armed with my background as an architect (which includes some knowledge of simple engineering principles and blissful igonrance of any higher math), I conclude that your model is over determined. Only in a world where apples are not money, paying taxes and consuming less apples is not necessarily the same thing. I say this distinction is necessary to make your model work. Under your 100 apple/period condition, as soon as one generation is taxed to pay for the apples some past generation has consumed, without those taxes flowing back into current consumption, one of four things must happen: 1. The price of apples declines = deflation. 2. The overall money supply increases exogenously. 3. Someone dies early to repay for Nr.1’s sins (Christ, that sounds familiar) 4. Less apples are consumed, which violates your initial condition and is the same as 3 in effect.
You need more moving parts or more martyrs.
And the utility argument, which is quite clearly right, is a separate issue, and perfectly compatible with the ‘distributional’ argument by Krugman, I’d say.
Oliver · January 25, 2012 - 8:28 am · Reply→

Oliver: sorry. Our spam filter disagrees with you.
Oh dear. 🙂
anon · January 25, 2012 - 10:04 am · Reply→

Bob, there would be taxes in any borrowing scenario regardless of the interest rate. (Even if the interest rate were permanently below the growth rate.) Because the principle still has to be repaid.
I think Baker talks about Bill Gate’s kids owning most of the debt. I assume this would be a distributional issue. But there would still be no necessary burden as long as the interest rate equals the growth rate over time.
The last part of your post seems to be invoking the invisible bond vigilantes. I believe Japan has debt to gdp of over 200%.