Winner of the New Statesman SPERI Prize in Political Economy 2016

Wednesday, 11 February 2015

The burden of government debt, again

Here is an attempt to clear up some of the confusion that still exists on this issue.

1) Government debt can be used to redistribute income to current generations from future generations, even if the aggregate level of consumption in each period remains the same. Proof by example: see here (or many similar proofs from Nick Rowe - his latest post on this is here).

Note: I think people get confused because although the first part of the proposition seems intuitive (if the government cuts taxes and pays for this by borrowing, surely those receiving the tax cut could spend it on themselves and be better off as a result), the ‘even if’ part seems wrong (if we are taking from the future to give to the present, current consumption must rise and future consumption fall). Those who have worked with OLG models, like Roger Farmer, find it easier to work through the logic as to how this is possible.

2) The size of government debt is not a good indicator of any burden. It is possible that government debt is positive, but there has been no attempt at intergenerational transfer. Proof 1: taxes on the young are cut, and the young save all the tax cut by holding the extra debt. Proof 2: borrowing for a capital project that benefits current and all future generations equally.

Note: This is important, as Noah Smith notes in this post. The size of government debt is not equal to the ‘burden’ on future generations. Indeed, positive debt is compatible with there being no burden at all.

3) There are probably much more important mechanisms going on right now that are transferring consumption from the future to the present: in some countries rising house prices, and climate change. No proof, just an opinion.

Note: despite this, if you think your grandchildren will have such a wonderful life compared to yours because of technological progress, you might not be too bothered. It is also worth remarking how potentially inconsistent it is to argue that we have to reduce debt now for the sake of future generations, and at the same time argue that it is too costly to take action now to mitigate climate change. 

4) Even if no intergenerational transfer is involved, high government debt could reduce future consumption for two quite plausible reasons: productive capital may be crowded out, and the tax required to pay the interest on the debt is distortionary (i.e. reduces output below optimal level). Proof: countless papers in the literature.

Note: I think it is wrong to describe both mechanisms as model specific, because you have to make quite extreme assumptions to avoid them. It is for this reason that I worry about high government debt in the long term. I have not heard anything to convince me that either mechanism is unimportant, or of any countervailing mechanism. (The need for safe assets could argue for high gross government debt, but not net debt, where the difference could be a large sovereign wealth fund.)

5) None of these arguments justify austerity at the Zero Lower Bound. Proof: countless posts by various people, including myself.

Note: For example, crowding out happens through high real interest rates, which are hardly a current problem. Nor is scarcity of labour arising from tax distortions.

Final thought. Think about government debt as a way of providing intergenerational insurance against negative demand shocks. When those shocks happen, the state pays out by cutting taxes (say) and increasing its debt. In normal times taxes rise again and the debt is gradually reduced. In this case allowing debt to rise following one of these shocks is no burden.  


  1. "Think about government debt as a way of providing intergenerational insurance against negative demand shocks. When those shocks happen, the state pays out by cutting taxes (say) and increasing its debt. In normal times taxes rise again and the debt is gradually reduced. In this case allowing debt to rise following one of these shocks is no burden."

    That would be a good thing to do, it might even be more than mere insurance since shock effects can destroy things (as an analogy losing 10.000 pounds at once does more damage to most people than losing 100 pounds per month spread out over 100 months).

    But of course it's not as simple as that. There's always the temptation to borrow beyond this insurance, just as there is always the temptation to not pay off the debt during "good" years. You might trust the Germans or the Dutch with such a plan, but not the French or the Greeks.

    1. "You might trust the Germans or the Dutch with such a plan, but not the French or the Greeks."

      Quite so.And that should really be the last word.

    2. I find this increasingly unconvincing with time. Even when there is a cleqr cut, textbook case for spending more, to delay debt payments and deficit reduction, politicians have proved to be extremely eager to cut spending... If they do not use this enormous crisis as an excuse for spending more, what would it take for your little scenario to take place?

      Currently, irresponsibility takes the form of shrink the state program, deficit reduction and high-end tax cuts, not fiscal profligacy.

    3. "politicians have proved to be extremely eager to cut spending"

      You do not live in Southern Europe, do you? Cutting spending (without greater or equal cuts in taxation) can take considerable outside pressure to be implemented. The US in 2002-2007 is also an example: the economy was doing great and still the debt-to-GDP-ratio was creeping upwards.

  2. I don't think I'm understanding the argument, because it seems obvious that if you look beyond money and debt, how well off future generations are will be determined by what they produce, and government debt plays no direct role in that.

    Suppose a future economy produces X items. Now suppose the government borrows lots of money by issuing bonds, then the future economy still produces X items, but the holders on those bonds have a larger claim on those X items.

    So it seems to me that the welfare of the future generation is determined by real resources and how they use it to produce things, while government debt only has distributional effects on the future generation.

    I can foresee 2 possible ways this story can be wrong:
    1. If the bonds are bought by foreigners, basically the current generation decides to give more things to foreigners in the future.
    2. Government debt having second-order effects by influencing how future generations can use their resources for production.

    Are you mainly referring 2nd point?

    1. No. It works because you persuade the young to consume less today, and more in the future. Have a look at the post of mine that I link to under point (1).

    2. Thanks. I understand your argument now, however, I don't see how that counts as making future generations poorer.

      In that simple model, at each time point 100 goods are always produced, so of all people alive at each time point, their total welfare is the same. Sure, it's possible for government policy to force someone to consume less over their lifetime compared to another person in the past (there are examples even simpler than yours: the government taxes the young to give to the old at time T, as a once-off, never to repeated again, so the young at time T are screwed, compared to all other generations). Really, you're example is within generation transfer.

      Politicians talk about debt as intergenerational theft, that we are burdening future generations with massive government debt that they will have to pay off, making them poorer as a result. This is the main assertion that either needs to be proven or debunked. But in your model, the people alive at time T+10 or T+1000 still produce 100 goods regardless of the debt at time T, so the people alive at time T are not poorer, don't make fewer goods, and not burdened by the debt at time T.

    3. While MM's example works, I don't think it actually captures the major reason why debt redistributes intergenerationally. Production is affected by who is in debt and who is a creditor. Put in the worst way possible. Suppose there is one Striver in the population, and the rest are farmers. The Striver grows his own food, but also makes iPhones and sells them to the farmers in return for future production. After ten years the Striver retires, stops growing his own food and stops making iPhones, and lives off the food that the farmers owe him/her.

    4. Which is basically the same as saying there are only distributional effects, like the other Anonymous (not me) said. The money borrowed in the present is still being spent somewhere so today's people as a whole do not consume less. of course it is possible that money is being spent on people who will be dead by the time the debt will be paid off and being borrowed from the people who will still be alive by then. In that case you do get delayed consumption.

    5. Anon at 4:37. Its symmetrical, if the debt is paid off. So the current old get to consume more when debt is created because the young postpone consumption. At some point in the future the debt is paid back, and one generation is worse off, but aggregate consumption stays the same because consumption is no longer postponed.

    6. Original anon here.

      Yes, I think it is possible that debt can affect production, and it can have effects in the future through this particular channel. This was a indirect second-order effect which I was referring to.

      But back to the main question (at least in political discourse), that debt burdens future generations with having to pay it off, making them poorer, I don't see how Simon's arguments can be used to show that it does burden future generations in this regard. I've even made a simple spreadsheet similar to his example to illustrate:

      I suspect that in order to argue that debt makes future generations poorer by burdening them with having to pay it off, one would need to argue based on the the effects of debt on future production or foreign ownership of government bonds (ie foreigners being able to take the outputs of future production).

    7. MM at 05:01

      Why is it assumed the money borrowed now comes from the young and benefits the old? I know that makes sense in general (people tend to only loan money they'll get back before they die and older people consume more state spending and have the large political clout to demand that spending), but it does not always have to be that way. In principle a government could issue bonds to people in their 30s and spend the money on child care for parents in their 30s, as an example.

  3. "Government debt can be used to redistribute income to current generations from future generations"

    No, it can not. Do you know this paper?

    1. Have a look at the proof by example.

    2. "Say that is period T+10, when the young get to keep their 50, but the old who only got 40 when young only get 50 when old. So we have a clear redistribution from the old in period T+10 to the old in period T."

      No, this is a redistribution between young and old in period T+10 not from the old in period T+10 to the old in period T.
      The old are losers because the young change the rule, and give them less, in period T+10.

      I think that Abba Lerner explained very well this point, and it seems to me that your example is quite similar to the one he was criticizing.

    3. Our children's government will have just as much money to spend as our government has today. The same as my parents government had. The limit of spending in all three generations is the capacity of the economy to provide goods and services while controlling inflation at as near full employment as you can get. The 250% Debt to GDP we had after the war, has not stopped my generation from buying train loads of I-Phones has it.

      There is no sovereign currency "saving" by a government that issues its own currency; as there is no government "debt" account. (Why would a government that issues its own currency, want to borrow back what it had spent into existence previously?).

      A government that runs a budget surplus today, is just permanently destroying financial assets that belong in the private sector today. Sovereign currency flows into the economy when the Treasury spends it (out of thin air). It flows back into thin air when the government gets it all back again, 100%, in Taxes.

      So, if you want to reduce the governments, so called, "national debt"; which strangely is exactly equal, Pound for Pound, to the private sector's "savings"; then spend every Pound of your savings in the most tax inefficient way possible. Mr Osborne is going to do his best to make you anyway.

      Not spending money today is a legacy you don't want to leave your children. They will not appreciate motorways with bucket sized pot-holes in them; falling down schools and having to wait four weeks for a Doctors appointment because we are not training enough skilled people and generating the jobs to keep them employed!

    4. New Anon -

      The focus should be on a transfer of WEALTH, and its redistribution upward.

      Consider this scenario. First, a group of high net worth people are given a tax break. The govt stills needs cash (in part because they cut these taxes) so they ask if this group would give them the cash in exchange for an asset instead of this foregone tax bill (they get an asset or wealth, it is a govt bond).

      SWL makes the clear and apt point above: "even if the aggregate level of consumption in each period remains the same" it is definitely a transfer and a redistribution. Many comments on Nick Rowe's blog and Paul Krugman's several columns get hung up on the income flow matter, missing the point that this is a public financing scheme that concentrates wealth upward (i.e; we cut-taxes-and-then-borrow from the same group of people by giving them an asset now). Nick Rowe and SWL are correct it is a transfer and redistribution, it happens to be upward.

      I can see why advocates for the wealthy want to obscure and confuse the discussion hoping economists and people just look at consumption or income-flow matters. Analytically we can recognize that some type of spending, using the transferred income flows, can be wise or not (used by lender or borrower classes or by the govt). Some uses may increase society's welfare in the future, as noted. But the consumption view is basically an exchange between lenders and borrowers, and it is an "identity" or zero net sum equation of cash, principle and interest, if you forget that there is also a bond/asset here too.

      The group that gets the bond instead of a tax bill do not want people to focus on this political engineering (it is as bald a rent-seeking result as I can think of) - it is a scheme of public financing that transfers wealth upward. It was called a supply-side tax cuts by advocates, or trickle-down by skeptics, remember?

      If you feel mechanistically that the cutting of taxes is needed because your economy is in trouble and is way below capacity, then cut it for those of lesser means, they have the higher propensity to spend. Unless your purpose is to concentrate wealth, do not cut taxes for the wealthy group, borrow from them.

      Makes no common sense and I really don't want my govt used for this purpose.

    5. Well, Simon??
      It is OK he has his theory no need for real life.
      Like the "natural" rate of unemployment.
      Economists are a really damaging profession.

  4. Just a note to say, if you don't know already, your London Review of Books 'The Austerity Con' article is up on their website, subscription free.

  5. Professional nit picker than I am, I’ve spotted a nit in Simon’s article.

    He says “Think about government debt as a way of providing intergenerational insurance against negative demand shocks.” That sentence conflates two issues that (as Simon himself points out) are separate. He refers to that seperateness when he says “positive debt is compatible with there being no burden at all”.

    That leads to another very relevant point, namely that the size of intergenerational transfers (far as I can see) is determined SOLELY by the size of a country’s pension provision. Plus the extent to which debt is involved in pension schemes is infinitely variable: that is, some pension schemes involve no debt at all because they are “pay as you go” (e.g. the UK’s state pension scheme). But there’s still plenty of intergenerational transfer there: that is, youngsters sacrifice their standard of living so as to support oldies living at the same time in exactly the same way as they do in Nick Rowe’s intergenerational theory.

    And finally I’d like to thank Oxford professor of economics, Simon Wren-Lewis for writing essays that he allows the plebs to mark. In view of the above nit, I’m knocking off a mark. I.e. I’m awarding 9 out of 10...:-)

    1. Ralph, don't worry about negative demand shocks, remember the MMT motto:-

      Mosler's Law: There is no financial crisis so deep that a sufficiently large tax cut or increase in public spending cannot deal with it.

  6. Thank goodness, a sensible viewpoint on this topic. I can always count on Simon.

    Government debt does not make future generations poorer, because debt payments go to bond holders, both of whom are around in the future. Net result: zero. Sure, there could be second-order effects like crowding out or redistributionary distortion, but you can have those effects without government debt too (say, high taxes and high spending), and you can have high debt without those effects (given sufficiently wise private investors and tax policy).


    Kenneth Duda
    Menlo Park, CA

    1. Ken: I think you missed what Simon is saying. Look at his point 1 again.

    2. Point 1: "surely those receiving the tax cut could spend it on themselves and be better off as a result".

      Spend it on themselves, exactly how would they do that? Let's see, could they possibly spend their tax cut on something that was sold to them by someone who was employed to make something they wanted?

      "In a simple neoclassical model where Ricardian equivalence holds, the net effect of government deficits on savings is zero. In an overlapping generations model [OLG], government deficits will generally decrease net savings". Which has since been proved to be utter nonsense. (Acorn)

    3. Thanks Nick, but I'm pretty sure I read Simon accurately. He says debt can be used as part of a government consumption redistribution scheme from younger to older. I agree. You made this point on your blog as well. The part that I cannot agree with is the naive view, which is so naive that sophisticated economists like you can't even understand what it is. It goes like this:

      "The national debt is $18 trillion! That's $54,000 for every man, woman, and child who will have to pay it back 30 years from now. We're stealing from our children and grandchildren!"

      It makes it sound like some invisible force is going to reach into the pockets of a bunch of people who will be alive in 30 years and help itself to $54k, reducing the lifetime consumption of each person by $54k.

      As you well know, that's not it at all, because for each $54k that the government takes from one person as part of repaying debt, it gives $54k to the bondholder, who can then spend or invest it. So much for reduced aggregate consumption or reduced aggregate investment. Redistribution, yes. Poorer in aggregate, no.

    4. Kenneth Duda11 February 2015 at 11:53

      I certainly hope that isn't macroeconomics. If so,to quote Frances
      Coppola, it's lost its marbles.

    5. Ken: to a first-order approximation (it only works exactly if the debt is small, or if preferences are linear), and if r > g, the cost of the debt is the debt itself.

      Assume r = 5%, and g = 0. Assume for simplicity the debt is never paid off, and the government collects taxes just sufficient to pay the interest on the debt. If the debt is 100 apples, I have to pay people 5 extra apples to be indifferent to postponing their consumption of 100 apples one year later in life. If I tax them 5 apples a year to pay that interest, they are 5 apples per year worse off. The Present Value of 5 apples per year forever, discounted at 5% interest, is 100 apples.

    6. Ken: or try it this way.

      Suppose we give all the old people today 100 apples, and have a debt of 100 apples. The Marginal Benefit of the debt is 100 apples (because the old people today get to consume an extra 100 apples). If the Marginal Cost to future generations was always less than 100 apples, the Marginal Benefit would always exceed the Marginal Cost, and a utilitarian would say we should always make the debt as big as we possibly can. (It's just as daft a policy conclusion as comes from saying the Marginal Cost to future generations is always zero.)

    7. Nick, I really appreciate the attention, that you take the time to respond.

      I feel bad feeling like I'm repeating myself. I think I understand and agree with your points, which I'd summarize as: the government can redistribute, including across generations, and debt can be redistributive. You made that argument in both of your comments above. And I completely agree.

      My point is the government debt does not make us collectively poorer. For example, in your scenario:

      > If I tax them 5 apples a year to pay that interest,
      > they are 5 apples per year worse off.

      Yes, agreed. But you didn't mention that the bond holders are 5 apples per year better off. So there has been redistribution, but no change in aggregate production or consumption of apples. Future generations are not poorer. Some in future generations are poorer due to the government debt, but some are richer, and it cancels out. If you agree with that, then I think we are agreeing on everything, even though it feels like you're trying to convince me of something that I think I already believe.

      Communication == challenging.


    8. Kenneth, "As you well know, that's not it at all, because for each $54k that the government takes from one person as part of repaying debt, it gives $54k to the bondholder, who can then spend or invest it."

      The bondholder could already do that by selling the bond. So his lifetime income is unchanged. The taxpayer's lifetime income is, indeed, reduced.

    9. OK, here is a test question for the above et al:-

      "The government announced in the budget that the Treasury will no longer issue interest bearing bonds. All existing bonds (Gilts) will be redeemed for Reserves as they become due over the next thirty years. Central Bank (CB) open market operations will continue as will CB payment of interest on Reserve accounts as required. Alternative Reserve quantity control mechanisms will be introduced by the CB.

      All those in favour say Aye

    10. Ken: in my example (where the government only pays the interest on the debt, and where all individuals are identical):

      The debt has no effect on any individual's total lifetime consumption of apples. The only difference is debt makes is to postpone his consumption of apples. Relative to the no-debt scenario, he consumes 100 apples less when young, and 100 apples more when old.

      But the discounted Future Value of his lifetime consumption is reduced by 5 apples (assuming a 5% interest rate per period), because apples when old are worth less to him than apples when young.

      His lifetime utility is reduced by exactly the same amount (strictly true only if the debt is small or preferences are linear) as if you destroyed 5 of his apples when old.

      Imagine that bananas were worth 5% less than apples to him, and you took away 100 apples and gave him 100 bananas in return. His total consumption of fruit is the same, but he is poorer, and his utility is lower. ("Bananas" are "apples when old".)

      This stuff is hard. It took me ages to get my head around it. Once you go down the OLG rabbit hole, and come out the other side, you end up right back alongside the unsophisticated people who think that if you tax someone 5 apples tp pay the interest on the debt, that makes them 5 apples worse off. They are right (at the margin).

      (If I relax my assumption of small debt or linear preferences, the only difference it makes is that the rate of interest becomes a function of the debt, which complicates things.)

    11. Nick, I agree again. The government, in your example, for sure is reducing the present value of every individual's consumption of apples by 5%, because deferring consumption into the future reduces its value compared with all consumption at the present time.

      Maybe if I pose my point as a question: you seem to have assumed both that all individuals are identical, and that no one is receiving the 5% interest payments, and that the government is raising taxes in order to make them... So, in your example, where are the interest payments going?

      I think this is Kruman's point. The "Eek! Debt!" people seem to assume interest payments are going to bad guys, or into a giant hole, when in reality, they have to go to someone, and that someone can _increase_ present consumption relative to baseline by the precise same amount as the "identical individuals" in your example have to reduce it.

  7. It might be nice if we looked at historical case studies, rather than looking at theory.

    A country that is heavily indebted to foreigners, would have future generations paying tax to pay the interest on that debt. This may involve a welfare loss in the sense that perhaps that tax might have gone on something else or enabled reduced tax freeing up resources and incomes for other activity.

    But whether it is good or bad will depend on the actual case. Paying war reparations ( a la Germany in the 1920s would be obviously negative. Paying back a cheap loan from a development bank may be a positive net transfer to future generations.

    1. There are certainly cases dating back to classical times of states being seriously weakened by heavy debt burdens - making them very vulnerable economically and more seriously politically and strategically. Do economists study this, or do they only look at theory? More recently we can look at instances like the South American Debt Crisis or Asian Financial Crisis. The latter countries in particular suffered extraordinary humiliation and hardship. Back in those days economists were all saying that relying on foreign currency denominated debt did not matter. It was good - all comparative advantage etc etc. Suggesting capital controls would be seen as very heterodox, even eccentric. (Tobin suggested this and this was another reason he was ostracised from the profession - some people still say controls are bad and people like Mahathir were wrong. Thankfully now the IMF has done a 180 on this and says they are appropriate in some cases.)

      Now it is OK to say these things - this is a subject prone to swings in conventional wisdom.

    2. There is a huge difference between a government that owes its own currency, and one that owes someone else's. If there's an exchange rate shock, you are in for a very rough ride as the size of your debt can expand at the same time interest rates rise, a double whammy for sure. On the other hand, if you owe your own currency, you have many more options.

      I have only been thinking about economics for about three years, so I imagine I arrived on the scene after conventional wisdom shifted to the idea that owing someone else's currency is dangerous. It sure looks dangerous to me. I hate owing dollars. I don't like it when the city of Menlo Park or the state of California owe dollars. But the Federal government owing dollars seems like a different matter --- not without real effects (like the shifting of tax burden from bond-holders to taxpayers) but much less scary, particularly when we're at the ZLB and 30-year bonds are yielding a pocket full of dryer lint.


    3. Nick, in response to another commenter, you wrote:

      > Suppose I have to bribe you with interest to get you to postpone
      > consumption, because you don't want to postpone your consumption. Then
      > when it comes to paying the bribe, I take it out of your pocket and give
      > it back to you. You are worse off.

      This is an interesting addition to your argument. You are arguing that debt makes everyone worse off, because:
      1) the (non-bond-holding) taxpayer is worse off because taxes to pay interest reduce the present value of all future consumption;
      2) the bond-holder is worse off because back when the bond was sold, he had to be bribed in order to agree to shift consumption into the future.

      Now I finally understand. I am arguing that debt does not affect aggregate consumption. You are arguing that debt hurts aggregate utility. We don't actually disagree.

      I'm not sure what I think about whether debt hurts aggregate utility. If we assume that the interest rate is set so that bond purchasers are indifferent to present versus future consumption, then I think bond purchasers utility is identical in either scenario, with or without debt, assuming bond purchasers are fully rationale. It's less clear for the non-bond-purchasing taxpayer. One could make a crazy argument that they voted for whom they voted for because in aggregate, that voting choice maximized their lifetime utility, and in that model, debt is aggregate utility neutral, but that model doesn't seem credible. Now is seems to matter a great deal what utility the non-bond-purchaser gets from the government's fiscal policy, i.e., were public goods created as a result of the bonds. If the answer is no, then I believe you are right, that the debt reduced aggregate utility, i.e., the redistributional effect of the debt was utility reducing.

      But I am still right that the debt is neutral in terms of aggregate consumption, for what it's worth.

      Thanks again for the discussion. I hope we get to meet someday. I am interested in supporting work in NGDP targeting, which is something you have written about quite a bit on your blog.


      Kenneth Duda
      Menlo Park, CA

    4. Ken: finally, I think we are on the same page!

      1. Debt does not reduce aggregate consumption in any future year, by assumption (it might do if it crowds out real investment, or if distorting taxes have disincentive effects so people produce less, but those were never at issue, so we are assuming those away).

      2. If the government pays down the debt at some future period, the lifetime consumption of those paying the higher taxes to pay down the debt will be reduced, even though aggregate consumption in any year is unchanged.

      3. If the government just taxes enough to pay the interest on the debt, but not pay down the debt itself, the lifetime consumption of those paying the taxes is unchanged, but their lifetime utility is reduced.

    5. > I think we are on the same page!

      *Phew* yes indeed. Thanks for your patience. It is such a privilege to get one-on-one coaching from a brilliant macroeconomist.

      > but their lifetime utility is reduced.

      Exactly. That's been my complaint to the Keynesians. They talk as though when at the ZLB, it doesn't matter what the government spends on, because any spending is better than not spending. I agree with the Keynesians that when there is inadequate aggregate demand, government borrowing and spending is utility improving compared with doing nothing. But: with proper monetary policy, there would never be inadequate aggregate demand for long enough for the fiscal stimulus to actually work better. And because individuals are making their own utility maximizing spending decisions, we are all better off than we would be with fiscal stimulus.

      Keynesians suffer from the cognitive illusion of the ZLB. The ZLB is merely the point where the wrong monetary policy (interest rate targeting) breaks down completely. If the CB doesn't use interest rate targeting, then the ZLB is irrelevant to monetary policy. Why is this not obvious to more people? Krugman won't address NGDP targeting directly, calling it "unproven". Simon here is willing to talk about it, but mostly to criticize market monetarists for practicing "faith-based economics". There is a lot more than faith behind the idea that targeting NGDP stabilizes the macroeconomy better than the Taylor rule, especially at the ZLB, where my eight-year-old daughter could probably stabilize the macroeconomy better than the Taylor rule. There may not yet be a fully worked out DSGE representative-agent model of NGDP targeting by pegging NGDP futures, but ... How do we move this ball forward? How do we convince Simon (and DeLong and Krugman and the other good Keynesians) that NGDP targeting would kick the current Fed policy regime's butt? That even if I'm wrong, and fiscal stimulus is still sometimes needed, NGDP targeting would work so much better than the Taylor Rule that Keynesians should line up behind it?

      Most importantly, how do we get a central bank to give it a real try?


      Kenneth Duda
      Menlo Park, CA

    6. I don't really understand the arguments here, but I very much appreciate the various high powered economists engaging and trying to explain to the rest of us. So thank you to Simon, Nick and Ken (and others as I read down the comments).

    7. Luke, if you are implying that I am a high-powered economist, that would truly be a first... :-)

      Regardless, I share your gratitude towards Simon, Nick, and Roger.


  8. I think that a major reason for the conflicting arguments on ths topic is because economists (aware of OLG models) use the term "generation" differently from most non-economists, when considering the present vs. the future. For non-economists, when thinking about 2115 vs. 2015, they are thinking about everybody who will be alive in 2115 vs everybody who is alive now, and this is what they mean by the future generation vs.the current generation. For economists, there will be 2 generations in 2115 and there are 2 generations now. For economists, transfers in 2115 between those who will be old in 2115 and those who will be young in 2115 can be treated as part of the way an "intergenerational" burden of debt can be transmitted. For non-economists this is just an issue of internal redistribution within the the 2115 generation, and is no different than the issue of whether the tax to pay off bond holders in 2115 is a progressive income tax or a regressive sales tax. Almar

    1. Yeah, confusion over definitions definitely plays a role. That and the fact that it almost becomes a philosophical discussion.

  9. Prof. Wren-Lewis,
    You referred to transferring "income" from one generation to another. I don't think this is possible. Income is a flow variable, and I do not see anyway in which flow variables can be transported across time. What does get transported across time are claims on future income. Bonds only allocate income. Income itself can only come from real economic resources, such as land physical capital physical resources, technology and labor...or more precisely, labor power, which is not the same as the number of workers. The generation that "borrows" is also the generation that gives up alternative uses for those resources. So if the government borrows in order to build an aircraft carrier, that borrowing displaces or crowds out other possible ways of employing economic inputs in that current period...assuming that there is no slack in the economy and the economy is not at the ZIRP. In any event, the economic inputs are consumed in the current period, not in some future period. You cannot call upon the flow from future economic resources anymore than you can call upon a from the past. What you can do is defer consumption of stock variables such as oil, minerals, some kinds of physical capital, etc. If consumed today, those stock variables would not be available for future generations. I believe that is the proper way to think about how borrowing to increase current consumption today could in a sense rob future generations. It's because today's consumption makes the stock of those inputs unavailable to future generations. Of course, this is true for all kinds of borrowing, not just government borrowing. OTOH, simply saving more so that future generations will have more doesn't always wash. For example, cutting back on education spending or job training or physical infrastructure spending reduces the stock of future economic inputs that are available to future generations. Even pure consumption items such as unemployment compensation helps future generations because the alternative is the pre-mature death of current period workers. Even a slave requires food in order to be productive. I find this whole approach of using the language of bonds to describe intergenerational equity issues completely confused. Economists should think in terms of how borrowing affects the supply of future economic inputs rather than relying upon this misguided analogy with bonds that they borrow from finance. If you think in terms of how borrowing affects the stock of future real economic inputs, then it's pretty clear that the effects of borrowing are utterly ambiguous. Borrowing cuts both ways.

    Finally, my take is that Nick Rowe and Roger Farmer are guilty of what I call the Doctrine of Immaculate Annihilation. Notice that a future generation is taxed which creates a burden. But also notice that the tax revenue seems to completely disappear. Nowhere does it re-enter the economy. The tax revenue is again treated like a stock variable and I believe this is because their models are built around transferring a bond, which is only a claim to future revenue and not itself future income. Bonds are not income. In the Rowe and Farmer version it's as though an unlucky future generation works overtime to pay future bondholders, who then burn the money. Yes, in that scenario future workers would be made worse off.

    1. Anonymous: "Finally, my take is that Nick Rowe and Roger Farmer are guilty of what I call the Doctrine of Immaculate Annihilation. Notice that a future generation is taxed which creates a burden. But also notice that the tax revenue seems to completely disappear."

      Nope. We are not stupid.

      Suppose I have to bribe you with interest to get you to postpone consumption, because you don't want to postpone your consumption. Then when it comes to paying the bribe, I take it out of your pocket and give it back to you. You are worse off.

    2. Nick Rowe:

      If you bribe me with interest to postpone my immediate consumption then you have to ask what happens to the portion of my income that I loaned to you. There are three possibilities:

      (1) You burn the money. In this case the current generation in the aggregate is worse off because I cut my consumption (making me worse off) but you did not increase your consumption. Your current period income is the same as it would have been had I not lent you the money, so your burning the money lowers my current period consumption without raising yours. In this case the current generation bears all of the burden. Notice that this is equivalent to a recession because money has permanently leaked out of the income flow.

      (2) Suppose you take the money I loaned you and invested it in some productive venture. Then the income of some future generation will be higher. So again, borrowing for productive investment does not add any burden on future generations. All of the burden is on the current generation that has to postpone immediate consumption.

      (3) Finally, if you take the money and choose to go on a partying binge of increased consumption today, then there cannot be any burden on future generations. This is counter-intuitive, so let's take an example. Suppose we both earn $10 each in period 1, so total income is $20. You want to borrow $1 for immediate consumption. I agree. Now I can only consume $9, but you will consume $11. Total expenditures are still $20, which is exactly the same as it would have been if I had not loaned you the money. Borrowing within a generational cohort for immediate consumption makes one person better off and another person exactly worse off. It's a wash. But because it's a wash there is no burden to pass along to a future generation. Also note that you cannot borrow from a future generation; you can only borrow from other members of the current generation.

      Ah, you say. What about the interest that I didn't receive from you because you robbed my pocket and used the proceeds from your ill gotten gain in order to pay me back with my own money. Am I worse off? Yes. But you're better off by exactly the same amount. I lose interest income, but you're off the hook. No one is saying that borrowing doesn't have distributional impacts within each generational cohort. And no one is saying that those distributional impacts might impose a burden on a future generation. But distributional impacts are a second order issue, and those impacts can be either positive or negative. The point is that the borrowing itself does not create a burden for a future generation. For the entire cohort the interest payments between debtor and creditor are a zero sum game.

      If there is any inter-generational burden, then it is not due to bond swapping. This whole financial approach to understanding inter-generational equity issues is completely misguided. It's a case of economists trying to be too cute by half. The economic burden can only occur under conditions in which increasing current period consumption reduces the economic inputs available to future generations. Every gallon of gasoline that I use today is a gallon that is not available for some future generation. But we have Harold Hotelling to help us figure out this problem. To put it another way, you can pull forward of consumption of stock variables like oil and iron ore, but you cannot pull forward consumption of flow variables like income.

    3. Anonymous:


  11. You are, of course, completely right Simon on every point. I am amazed that there is still debate about any of these issues.

    It is, however, worth separating the effects of a debt financed transfer in a general equilibrium model where the equilibrium (or equilibria) is (are) Walrasian, from the experiment we are discussing in the real world where there is clearly involuntary unemployment in the sense of Keynes. It is conceivable, in a non Walrasian model, of whatever variety you favor, that expansionary debt financed policy can provide a free lunch.

    That may not be the best way to restore full employment: but that's a diffentent discussion.

    1. Roger: yep. But you've spent most of your life thinking about OLG models (remembering when you told me about Samuelson 58 back in grad skool). It's hard.

      But yes, you would think that Paul Krugman would understand the basics.

      BTW, a small aside regarding your excellent Twitter argument: if r=g=0% both with and without debt, then preferences must be linear. So the other guy's assumption (r=g=0% both with and without debt) contradicts your assumption (concave preferences), unless the debt is very small.

    2. Nick, R=g does not imply linear preferences. For example, log preferences, no population growth, equal endowment in both periods, equal utility weight on both periods. This is the knife edge case where the monetary equilibrium and the autarkic equilibrium coincide.

      I applaud you for taking on this issue in the first place.

    3. Thanks Roger. It's been a draining issue, especially since OLG models aren't really my forte 9unlike you).

      I didn't explain myself very clearly on the R=0% thing.

      If R=0% initially, and then debt increases by a non-trivial amount, so consumption gets postponed to later in life, but we still have R=0%, then preferences would have to be linear (otherwise R would increase above 0%, with concave preferences). I only mention this because you were arguing with someone on twitter who made the assumption that an increase in debt did not cause R to increase above 0%, so he was implicitly assuming linear preferences.

  12. "The size of government debt is not equal to the ‘burden’ on future generations. Indeed, positive debt is compatible with there being no burden at all."

    Why then do you and Nick Rowe write about this topic using the misleading title The Burden of Debt? Why not call it The Burden of Crumblies or the like?

    You complain about confusion but you add to the confusion yourselves.

    1. Kevin: do you want to call it "the burden of the future tax liabilities that may or may not be inherent in the debt"?

      Simon's case 1 is where the current generation gets the transfer payment, saves it, and also pays 100% of the taxes (it's Ricardian Equivalence). Case 2 (where the government borrows to build schools) is where I would prefer to say "the kids get their education which is an asset, and they also get the burden of the debt, and if it's a good school the asset is worth more than the liability, so the kids come out ahead on net".

    2. Most of these posts seem to be intended as ripostes to Paul Krugman's statement that debt is money that we owe to ourselves, so I'd say a better title would be The Perils of Aggregation. What your various cases are showing is that there is precious little that can usefully be said about the burden of debt in general.

    3. Yes, Kevin is right, this is the exact issue. Krugman says, "the debt scolds are completely wrong because debt payments don't disappear, they go to bondholders, so aggregate income and consumption is unaffected." Nick and Roger and (to a lesser extend) Simon say, "Oh no, Krugman is completely wrong, it is at least possible for debt to be part of a larger dynamic in which aggregate utility is somewhat reduced."

      Excuse me?

      How about, "Krugman is completely right, of course. Debt payments do not reduce overall consumption. However, there are some second-order effects we should be concerned about if the government is taking on too much debt... (insert stuff about intergenerational redistribution and aggregate utility reduction through deferred consumption here)."


  13. I like this wording:

    " 4) Even if no intergenerational transfer is involved, high government debt could reduce future consumption for two quite plausible reasons: productive capital may be crowded out, and the tax required to pay the interest on the debt is distortionary (i.e. reduces output below optimal level). Proof: countless papers in the literature."

    where the word "may" is used.

    It seems in discussions of crowding out the statements are more along the lines of "Any government borrowing at full employment necessarily crowds out private projects, which is of course a bad thing".

    Yet, even at full employment (whatever that is these days), if a government project has a higher ROI than the debt cost, the project increases wealth. And if the public ROI exceeds private ROI (increasingly likely the more enthusiastic the privates become), then the public borrowing must be done, no matter what.

    So I like the "may".

    It would be interesting if among the many papers discussion crowding out some discuss the history of such instances.

  14. are you saying that the government debt incurred by my parent's generation (for WW2, for instance) reduced the output of goods and services in my generation?

    1. RJS: No, they're not saying that aggregate output now is lower because of WW2 debt. I finally figured these guys out (thanks to Nick Rowe's patience). They're saying that a government that carries debt reduces the aggregate utility of future generations compared with having no debt but things otherwise being the same. The intuition is that because people are able to spend in ways that maximize their utility better than the government can, the distortion effects of the transfer payments from taxpayers to bondholders (interest and/or repayment) has no effect on the utility of bondholders, but reduces utility of the taxpayers. This is because buying the bonds was voluntary (hence utility neutral) but the higher tax payments to fund them are compulsory (and thus utility reducing). Aggregate output and consumption are completely unaffected by the government transfers.

      You can offset this argument if you assume that the government spent the borrowed money in ways that increased aggregate utility, for example, by winning a war, the loss of which would have plunged us into decades of fascist rule.


      Kenneth Duda
      Menlo Park, CA

    2. @rjs, John, Kenneth and Kevin There are qualifications in Simon's post because the economic model he is using allows for different cases and the question of which case applies is an empirical one. The point that Nick made back in 2011, is that the notion that 'debt is money we owe ourselves' is seriously misleading.

      Debt does not have to reduce aggregate consumption in order to pose a burden on future generations. The distribution of income between current and future generations, and I would add, between the 1% and the 99% is not a small matter.

      When central banks around the world bailed out the banks, that money went to the owners of the banks, and predominantly to the 1%. The debt that was incurred imposes a tax liability on all future generations, including the 99%.

      Don't be fooled by the fact that Simon, Nick and I use simple examples to make a point or that we use generational models to discuss the broader question of income distribution.

      I do not know why PK chose not to respond the consensus that developed amongst academic bloggers when this issue was first aired back in 2011.

      And your 'excuse me' point Ken; is missing the main point. If you want to wade in on behalf of your idol, there is a more effective argument. It is not that the distributional burden of debt is second order, (and I would dispute even that) it's that the model we are all discussing assumes full employment. If increased government spending increases employment; there are first order benefits from funding expenditure with debt.

    3. kenneth, that assumes my taxes were or will be raised to pay interest on the bonds...i haven't seen evidence of that, nor any indication that it will ever be necessary...

    4. "The intuition is that because people are able to spend in ways that maximize their utility better than the government can ..."

      Which is the rub. Because that may or may not be true, in so many different ways.

    5. Roger, I'm not sure whom you're referring to as "my idol". I can only assume you're referring to Krugman. I appreciate a lot of what he has to say, but I am a much bigger fan of Scott Sumner, and in fact I personally funded the Hawtrey Chair in Monetary Policy at Mercatus to help Scott develop his ideas in this space. As you know, Sumner believes that fiscal stimulus is almost always a mistake, as any fiscal expansion would be offset by monetary contraction assuming ideal monetary policy. i.e., under ideal monetary policy, "crowding out" would become true. My views align much more closely with Sumner's than with Krugman's. I think Scott is probably right that the right monetary policy regime (NGDPLT or some other targeting of a nominal aggregate) would eliminate all "free lunch" cases of fiscal expansion, e.g., would eliminate the case where resources would have otherwise been wasted had the fiscal authority not consumed them.


      Kenneth Duda
      Menlo Park, CA

  15. Sorry to say this, but the whole debate is something of a distraction. It is always true that there is no distributional effect of fiscal deficits as a mode of financing, separate from what it finances, across time periods. No one disputes this, although this illusion is at the heart of public misconception. The dispute is over distribution across generations.

    In order to demonstrate an initial generational transfer, you need first of all to make the assumption that the object of spending produces benefits disproportionately for those who are (a) alive and (b) older at the time of expenditure. Then you have to assume that no or an insufficiently small proportion of bonds are bequeathed from the old to the young. Those two assumptions give you an initial situation in which two generations overlap the same time period, and the older has acquired a net benefit at the expense of the younger. This transfer can then be transmitted over future time periods provided the new generations always purchase the outstanding bonds from the older (or retire the bonds and issue new ones). There is no doubt that this logic is correct; it can be shown formally.

    The reason there is still a dispute is that many of us, including yours truly, doubt that this set of assumptions accurately characterizes the world we live in, where, among other things, bequests are substantial. Indeed, you can also construct a model in which a future generation benefits at the expense of the present on the basis of sufficient wealth transfers from old to young.

    Simon, I’m rather surprised by the position you’re taking. Much of what’s been wrong with macroeconomics (and also micro) in our experience is the construction of models, internally consistent, on the basis of counterfactual assumptions, which are then applied to the real world. No one is saying, for instance, that the RBC model is “wrong” in the sense of logically inconsistent, just that the assumptions (and conclusions) are empirically disconfirmed. The same goes, I think, for the claim that fiscal deficits transfer wealth from future generations to our own. This is why I wasn’t part of the “consensus of academic bloggers” Roger refers to back in 2011—yes, I posted vociferously on this—and why I’m still not persuaded.

    On a secondary note, in response to Roger’s point about the 1% and the 99%, there is no dispute here. Financing spending by borrowing from the rich rather than taxing them is transparently distributional, and this effect is transmitted over time through differential ownership, and especially differential inheritance (Piketty) of bonds. But this is an intragenerational effect, since it doesn’t alter the factors that govern net transfers across generations.

    1. I'm not persuaded that OLG models are based on counterfactual assumptions that should be ditched in favour of ones where bequests are complete in a Ramsey sense. But I'm happy to be persuaded if you can. The real world of course involves bequests, and the baseline OLG assumes none, but unless bequests are complete something of the OLG character survives.

    2. Are you sure that complete bequests would not result in transfers from the present generation to future ones, and that an intergenerationally neutral regime would not have partial bequests? If this has all been worked out somewhere, please point me to it. Intuitively it’s not at all obvious.

    3. Peter and Simon
      It is still true, with bequests, that a debt financed transer to the present generation leads to a welfare loss for all future generations. There is fully worked out example here . This is a little more complicated than two period OG, but arguably more realistic. It is composed of patient and impatient agents, like Krugman's Sam and Janet, but one can think of these agents as dynasties of infinitely lived families who pass on wealth to their children. Sometimes parents have no children and dynasties end. Sometimes children quarrel with their parents and form new dynasties.

      People consume a fixed fraction of their wealth in every period. That is a consequence of the assumption that preferences are defined, each period, over the logarithm of consumption. Arguably, that is a realistic assumption. The model is stochastic, there are random shocks to beliefs; but that is not essential. The lesson from this model, that is relevant to this debate, is that a newly created dynasty must rely on the endowment of its members. In the model this is one 'apple' every period into the infinite future. When government creates debt to finance a transfer to current generations, it does not alter the endowment of a new born dynasty. But is does alter their wealth; this is defined as the discounted net present value of all future endowments. The creation of debt causes the consumption of every new born dynasty to be lower in every period. This is for two reasons. First, each new dynasty pays taxes to finance the transfer to pervious dynasties. Second, the discounted value of future income is lower because the interest rate is higher when government creates debt.

      There is a, very real, sense in which debt is a burden on future generations. It is NOT a second order effect, even in a model with full employment and perfect markets. It is a transfer from our children and our children's children to ourselves.

      Does this mean that a debt financed transfer to the current generation is always welfare decreasing for our children. NO. The entire debate that we are having here assumes full employment. That is NOT the world we live in and there are potentially separate benefits from increasing debt that arise from its potential effects on aggregate demand.

      The, much more interesting debate, is: should we stimulate demand by increasing government debt and using it to purchase goods? Or can we stimulate private demand by open market operations in the asset markets? I wish we could refocus the considerable intellectual energy that we are all expending on a game of intellectual gotcha, to the much more important question of how best to "end this depression now".

    4. The here in my previous post should have linked to the paper "Global Sunspots and Aasset Prices in a Monetary Economy". You can find an open link as the first paper at the top of my website. My skills at embedding http links in the comment section are clearly not up to snuff.

    5. Roger, I looked at the Sunspot piece, and, as I said on your blog, I’m not familiar enough with the literature to evaluate the modeling choices you made. Nevertheless, it was not clear to me how you arrived at your conclusion about intergenerational transfers, and your summary here doesn’t explain it either.

      The money quote: “ The creation of debt causes the consumption of every new born dynasty to be lower in every period. This is for two reasons. First, each new dynasty pays taxes to finance the transfer to pervious dynasties. Second, the discounted value of future income is lower because the interest rate is higher when government creates debt.”

      First reason. It isn’t clear why the recipients of new dynasty tax payments aren’t other dynasties from the same cohorts or even from successor cohorts, if they have inherited a portion of the bonds. With bequests, can’t the recipients be from any cohort, earlier, the same or later? Or is there a restriction on the structure of bequests that prevents this?

      Second reason. First, you are assuming crowding out, which, as you say, implies a supply-constrained world. Second, are you assuming a wedge between the market interest rate and individuals’ discount rates? This would be required in order to demonstrate that higher interest rates were welfare-reducing, yes? More generally, this argument renders your conclusion conditional on the utility framework: it is not about changes in real consumption but the changing present value of unchanged future consumption. In lay language, fiscal deficits may not reduce the consumption of future generations, but they will value their future consumption less because you raised their interest/discount rate. What you think of that depends on how much credence you place on the utility apparatus. In any case, this is hardly what the Great Debate was about. I am happy to have a new Great Debate about the welfare-reducing effects on future generations of raising their discount rates.

    6. Peter, it's easier than this. The "Eek! Debt!" people make ridiculous claims about how $18T in government debt will mean our children are $18T poorer. Krugman points out that this is silly, because debt is "money we owe ourselves". In typical Krugman form, he's clever and essentially right but is also stretching and oversimplifying. That annoys a bunch of sensible economists, including Roger and Nick and Simon here, who point out that debt is not entirely benign for several reasons:

      1. the future payments mostly go to the rich

      2. the future taxes reduce lifetime utility of the non-bond-holders (i.e. most people)

      3. the additional government spending (at the time of the borrowing) can reduce utility because how likely is it that the government spends money as effectively as private parties, or because of "crowding out" (which is kind of the same thing), and

      4. by the way, the debt can be part of a larger government program that does shift consumption away from future generations and towards people alive today through a chain of intergenerational transfers, so it's possible we are effectively "borrowing" from our children, in some sense, assuming social security collapses at some point etc.

      Everything the economists are saying here is right too. They are annoyed by Krugman's oversimplification. I am discouraged (and I expect you are too) by the way Krugman's simple-and-true point (that the "Eek! Debt!" people are idiots) gets lost in side issues.

      The better idea is to drop this sad topic (as you suggested at the top of this thread) and focus instead on what can we really do about output gaps at the ZLB? Is fiscal stimulus better than nothing? Is it possible to organize a political consensus around fiscal stimulus? Should we also try to do better in monetary policy, e.g. NGDP level targeting? (I personally think NGDPLT would be the biggest single improvement we could make in the way we as a society handle output gaps and the ZLB.)


      Kenneth Duda
      Menlo Park, CA

  16. It appears that you and Farmer and Rowe are the only economists in the world that understand this point. Although Nick managed to convert an Austrian a couple of years ago - sort of like Cosmo Kramer working his magic on a lesbian.

    There should be a club for this.

    (Krugman just needs to work on his accounting I think - for this and other reasons.)

  17. on Monday, zero hedge posted a chart from morgan stanley showing the net global issuance of government debt in 2015 will be negative, because the central banks will buy it all…
    so, if all the notes and bonds are being bought by central banks, who then collect the interest on those bonds and in turn remit that interest back to their Treasury, how can that debt be a burden on anyone, much less to a future generation?

  18. I'll just leave the same question here that I did at Nick Rowe's blog:

    There is one thing (at least) I know I don’t understand about how you and SWL and RF are interpreting the economics of this.

    It has to do with the emphasis you all place on the distinction between bequeathal and selling of bonds.

    Consider the final time period where the tax is imposed.

    Two scenarios:

    One where young Z1 has bought the bonds from old Y1.
    Young Z1 then gets taxed.
    So old Z1 will have no bonds to sell and therefore no related consumption.

    The second where young Z1 has inherited the bonds from old Y1.
    Young Z1 gets taxed.
    And old Z1 will have no bonds to sell and therefore no related consumption.

    It seems to me that these two scenarios are identical in terms of the effect of a marginal tax (actually, a marginal primary surplus) on the marginal inter-generational distribution of consumption.

    In both cases, old Z1 has no future consumption flowing from his ownership of a bond today.

    The fact that in the first case he paid for the bond and in the second case he inherited the bond should not be an issue, IMO. It is the marginal effect on inter-generational distribution of consumption that is the issue. The fact that Z1 ends up with more net wealth in the case where he didn’t have to buy the bond is moot to that marginal intergenerational distribution effect.

    That’s why I said earlier that in my stripped down version of the most bare bones model, the issue of the bequeathal/sale distinction is secondary and in that way not really relevant. The only relevant thing is the marginal primary surplus at the end of the game, which is also what I said several years ago.

    Do I have this wrong? I know that the bequeathal/sale distinction seems central to the way you illustrate this. Roger Farmer seems to use it a lot also, but for additional reasons it seems that I don’t fully understand. Am I missing something essential here? If so, I don’t see it.

    Is it possible that this distinction is more an inherent feature of OLG models, and that while OLG models have been used by economists in this debate, it is actually not essential in illustrating the specific intergenerational distribution issue that has been the subject of this debate?

  19. I'll also leave the same question here that I did at Nick Rowe's blog:

    Why the debt gets passed forward if g=0 and r>0 (e.g. Rowe's model). It is a ponzi scheme and everybody should know that.

    This is quite obvious from the second cohort point of view. Letting A consume more means that B have to trust that they can pass the debt forward. This is not a certainty and thus B will not buy the debt.

    Sketch of Proof by induction: Let's say we know that at time T+n this unsustainable debt have to be paid back or will be defaulted. Then the cohort T+n (n doesn't have to be known, it is enough that this will happen eventually) will pay the price and everyone knows that. Then T+n will not buy the debt T+n-1 owns. But T+n-1 knows that and it is not going to buy it from T+n-2. So by induction T+1 will not buy the debt. The second cohort shouldn't voluntarily buy A cohort's debt because it will be strictly worse off.

    So isn't the model logically inconsistent? The additional assumption that cohorts are not logical can be made but is the model then useful anymore?

    1. the "debt" does not have to be paid back...

      & that's the problem with all discussions on US or any sovereign debt; the use of that word debt, which carries the confusing connotation that it’s something that has to be paid back…that leads some to compute how much each taxpayer would owe, something i even did myself when Reagan was president and probably even did as recently as 2008…it’s a nonsense exercise, because it isnt something we owe; the way that US notes and bonds function in the economy is more akin to the way a ten dollar bill in my wallet functions as money for my personal use…

      all money is loaned into existence, so expanding debt by either the government or the private sector is absolutely necessary for economic growth…any attempt by the government to pay down the debt will result in a depression, just as it did the previous six times in US history that the government ran large surpluses...

      the idea that we are borrowing from China also has its origin in the thinking that our script is somehow debt, leaving us with this picture of Uncle Sam going hat in hand to the Commies to get the money to run the government…the trade deficit with China is the only reason that China ends up holding a large portion of our script…every time Walmart orders another boatload of trinkets from a Chinese supplier, they pay for it with a bond or note (or an electronic transfer thereof) …to get rid of our notes and bonds, the Chinese would have to stop trading with us..

      the idea of the Chinese “selling their US holdings” is equally ludicrous; what would they want in its place, a stack of greenbacks? or would they prefer Euros, or Swiss or German notes with negative interest rates? yen, when the “debt” to GDP ratio of Japan is 260%? and if they do buy yen bonds, then Japan buys US Treasuries..

      so i dont see our “debt” as a problem that has to be addressed; if anything we need to issue more of it, to allow our economy to reach its potential…

    2. Debt is highly complex issue, noted also by Wren-Lewis in the previous debt post. I think we need first to get the simplest models working and focus on them.

      If we cannot formulate a simple model with rational agents that shows the intergenerational transfer, I think it is fair to say that Krugman was right to neglect the whole mess.

  20. This comment has been removed by the author.

  21. It is not clear to me why inter-generational transfers are treated differently to intra-generational transfers. I think confusion does originate in the claim that increased public debt places an “increased burden on future generations”, and many do not think about overlapping generations. But even if there are no overlapping generations, current consumption can be at the expense of future generations if current borrowing creates claims to foreigners. If large public sector borrowing (or private sector borrowing) causes significant current account deficits, it is possible that future generations will have to run trade surpluses and therefore will have lower levels of consumption. No?

  22. Here is a concise example showing the burden on the future generation of tax payers.

    The government borrows £ 1 Billion from the central bank and pays that to its employees. 30 years later the government repays that debt by taxing the then current generation of tax payers.

    The result is that the heirs of the government spending either,directly through inheritance or indirectly through commerce, have more money and the tax payers have less money.

    1. BS. Government deficits are the private sector's surplus. They are not a burden on our children. You could leave us young people good infustructure and a healthy enviroment though.
      Central bank "independence" is BS too. It has to be answerable to the treasury and parliament, otherwise we get 1910 constitutional crisis.
      There is no need to "repay" the "debt", the IK govt is a currency issuer.
      The government works by crediting current accounts. The borrowing is voluntary.
      The correct way to look at it is taxes give demand to the currency (you can't pay TV licence or Council Tax in gold or bitcoin!)
      The non-govt sector then needs the currency.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time.