Saturday, 18 October 2014

More on time travel.

My article yesterday on the above topic was not well thought out. But I still don’t think that funding public investment via government borrowing rather than via tax has any significant implications for inter-generational fairness. Let’s run through the arguments.
The obvious and superficial attraction of funding via borrowing is that future generations (who benefit from public investments) have to pay interest on the debt and/or have to repay that debt.
One answer the latter point is the “time travel” argument set out in my article yesterday.
And an answer to the latter time travel argument, set out by Nick Rowe is that time travel is in fact possible in the following sense (assuming I’ve got Nick right). Assume a public investment is made during the working life of a particular generation (let’s call it generation 1). Assume also that government funds the investment via borrowing. Generation 1 undoubtedly suffers a standard of living hit as a result. However, that hit can be nullified if during the retirement years of generation 1, generation 2 has to buy government bonds off generation 1. That obviously helps fund the retirement of generation 1.
Likewise, generation 3 can be made to buy the bonds off generation 2 in the retirement years of generation 2. And clearly that works: that is, it transfers the burden of making a public investment to future generations.
However the big problem with the latter “generation” argument is that it is highly artificial. That is, when governments make public investments funded by borrowing, they don’t in practice set up the above nice clean and simple “succeeding generation” system.
That is, the real world is in practice far more messy. For example, government debt tends to be held by the rich who far from SELLING government bonds to their descendants, simply GIVE it to those descendants. Second, government debt is in practice bought by and held by a variety of actors apart from pension funds, e.g. foreigners, banks and so on. And the latter have a variety of reasons for buying or selling government debt that have nothing to do with inter-generational fairness. Third, the assumption in the above “generation” scenario that government makes public investments in a particular year and none at all for the next ten or twenty years is unrealistic.
The reality is that governments spend very roughly the same amount on public investments every year. And that point alone means that it doesn’t make much difference as far as inter-generational fairness goes, whether public sector investments are funded via borrowing or tax.
And a fourth and final complicating factor is that the population ALREADY HAS approximately the pension set up it wants. That is, in as far as pensions are investment based rather than pay as you go, the above generation 1, 2, 3 etc won’t be very interested in a new facility that enables younger generations to buy bonds or other assets off preceding generations. Put that another way, if the above “clean and simple” system WERE SET UP, it would to a large extent just displace EXISTING SYSTEMS for “making the next generation pay”.


  1. Ralph how can the Gen 1 borrow from Gen 2 at the initiation of this sequence if Gen 2 has no "money"? iow Gen 2 are children when the bridge is built so they have no balances of "money" to borrow.

    iow at the initiation of this sequence, the "money" has to already exist to be exchanged for the bonds... it is in the possession of Gen 1.

    If govt issues bonds before they will fund equal bridge expenditures, they are requiring Gen 1 to defer equivalent consumption (ie "save") to at least until the maturity date when the govt will redeem...

    Govt refuses to let Gen 1 build and use the bridge until somebody in Gen 1 agrees to save.... ("inflation" control reasonings?)

    There is no "borrowing" involved, there is "forced savings" or "coerced savings"... remember the "deficit" is savings... as Warren says "to the penny".... rsp,

    1. Assuming I’ve got Nick right, the assumption is that Gen2 are young (i.e. in their 20s and 30s) but not children. So they ARE EARNERS.

      Re your “coercion”, there is no need to force Gen1 to buy bonds: it can be induced to do so by raising the reward for buying bonds (i.e. raising interest rates) far enough. As to Gen2 buying the bonds off Gen1, there may need to be coercion there. But there is nothing unusual about that: making a contribution to the UK state pension scheme is compulsory for those in work.

  2. I think an important aspect of public borrowing has been overlooked in ALL these posting: Borrowing by the Public is accompanied by a System of Repayment.

    A System of Repayment has two main components:

    1) A designated target to receive the payments. This target is the entity who will not need to work to receive money. This target entity has already worked in the past; worked or someway acquired the stream of benefits that follows ownership of debt.

    2) A method of acquiring the money to be devoted to repayment. This method (for public debt) is usually in the form of a tax on something. The existence of any tax becomes an important component of the future economic environment.

    "Time travel" may be a very misleading description of the future-setting aspects of public debt. The establishment of new public debt is synonymous with the establishment of components of the future economic environment.

  3. Ralph: I think you have got my argument right.

    "For example, government debt tends to be held by the rich who far from SELLING government bonds to their descendants, simply GIVE it to those descendants."

    That is an important point. And it means the burden is on the first generation, and not their kids, or kids' kids (if their kids do the same thing). But if we follow that line of reasoning through to its logical conclusion (which Barro did) it leads us towards Ricardian Equivalence, where a bond-financed tax cut has no effect, because people save all of their tax cut, to give it to their kids.

  4. Rowe's argument is illegitimate.
    1. The debate is usually about the effects of bond sales on different generations.
    The debate is NOT usually about bond sales PLUS taxing or otherwise forcing young people to pay for higher spending by older folks. By adding this new measure into the debate Rowe Implicitly admits that debt increases ALONE are not a burden on future generations.

    2. The redistributive effects of Rowe's extra measure only occur at the time the measure is in effect.
    Extra taxes paid by young people TODAY enable higher pensions TODAY.
    The same measure NEXT YEAR affects disposible income distribution only NEXT YEAR.
    There is absolutely ZERO redistribution over time.

    1. Your point that “Extra taxes paid by young people TODAY enable higher pensions TODAY” is compatible with Nick’s point. For example during the retirement years of generation 1, generation 2 is forced to pay “taxes” (aka buy bonds off generation 1). In that sense “redistribution over time” is possible. That is “higher pension TODAY” (for generation 1) is the result.

      But (to repeat), I think Nick’s idea has problems. One of them, which I alluded to above, is that assuming everyone ALREADY HAS the pension arrangements they want prior to a chunk of public investment being made which exploits the “Rowe theory”, then “everyone” won’t be keen to have Rowe’s theory imposed on them. Reason is that Rowe’s theory (far as I can see) involves increasing everyone’s income in retirement relative to their income when in work.

  5. Ralph: I think my previous comment got lost. Trying again.

    You understand me right.

    "For example, government debt tends to be held by the rich who far from SELLING government bonds to their descendants, simply GIVE it to those descendants. "

    That is an important point. You get no burden on future generations in that case. But if you follow that point through to its logical conclusion, like Barro did, you end up with Ricardian Equivalence. Bond-financed tax cuts are saved.


Post a comment.