CEPR - Center for Economic and Policy Research

Multimedia

En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Educating Steven Rattner on Government Debt

Educating Steven Rattner on Government Debt

Print
Sunday, 22 January 2012 05:14

Steven Rattner remains convinced that handing future generations trillions of dollars of government bonds imposes a burden on them and is very unhappy that I don't see things that way. Let's try this one more time.

Let's say that we add $10 trillion to the national debt over the next decade. We'll assume that it is owned domestically. That is of course not true, but the foreign ownership of debt is determined by our trade deficit, which in turn depends on the value of the dollar, not the budget deficit. Furthermore, no one is disputing that foreign ownership of U.S. assets (either government debt or private assets) will be a drain on the economy.

At some future point, everyone who owns this debt today will be dead. They will have no choice but to hand this debt on to members of the next generation, either their own heirs or someone else's. (Note, contrary to Mr. Rattner's assertion in his original NYT piece, the debt does not disappear, the ownership is transferred. [Sorry Mr. Rattner, you don't get the $1 million prize.]) This means that future generations will be both paying and receiving debt service on $10 trillion of debt. How is this a burden on future generations as a whole?

Again, the taxes needed to pay the debt service do imply distortions, but the distortions will not be anywhere near the size of the debt. Furthermore, there are all sorts of distortions in the economy, many of which could be much larger, that we never think of as imposing a burden on future generations.

The most obvious are patent and copyright protections. By virtue of these government-granted monopolies, we force people in the future to spend hundreds of billions more per year to buy protected products like prescription drugs and computer software than they would pay in a free market. The additional costs associated with these protections have the same impact on the economy as a tax of the same size. Why are deficit hawks like Rattner completely unconcerned about the implicit tax burden of patents and copyrights that we are imposing on our children?

There is one other logical point where Mr. Rattner needs some education. I had suggested that the Fed could just hold the $3 trillion in assets currently on its books. In this case the interest on the debt is paid to the Fed and is refunded right back to the Treasury. Where is the burden on our kids?

Rattner responds by favorably quoting a blog commentator:

"You don’t know what the Fed will do or be able to do in the next generation."

Of course I don't know what the Fed will do, but this is a matter of public policy. Congress could mandate that the Fed will hold $3 trillion in assets and refund the interest to the Treasury. The Fed can raise reserve requirements (the favored tool of China's central bank) to stem any inflationary impact from this decision. The point of raising this issue is that this is one way that we can issue debt today and impose no debt service burden on future generations. The debt service is paid from the government to the government. That's pretty straightforward, isn't it?

Comments (26)Add Comment
...
written by Robert Waldmann, January 22, 2012 7:53
It is. Not true that the current generation has no choice but to hand over the debt. We can exchange the bonds for goods andbservices. It is perfectly possible to write down a model in which debt makes future generations poorer even if the government can tax without distortions. Peter Diamond wrote that model down decades ago see "National Debt in a Neoclassical Growth Model" AER LV December 1965 pp 1126-1150.

I understand that it is hard to keep up with the literature and one can easily fall 46 years behind. AlsoDiamond is an obscure figure and the AE R is an obscure journal.

This post is a howler on the order of Cochrane on Ricardian Equivalence.

The completely different totally valid point is that most members of the current generation *Choose* to give assets to the next generation as bequests. If their motive is altruistic (a safe bet) *and* people are rational utility maximizers, then debt does not reduce the wealth of future generations. Also tax cuts are completely ineffective as stimulus.

You are assuming that Ricardian equivalence *must* hold. I am sure you don't believe this and do not understand what you have written.
'We Don't Need No Education' is Not a Double Negative for the 1%
written by izzatzo, January 22, 2012 8:27
There is one other logical point where Mr. Rattner needs some education.


Put Another Brick in the Wall for the 1%.

GO PINK FLOYD!

Stupid liberals.
Quantity vs. Quality
written by LSTB, January 22, 2012 8:34
Rattner writes:

Consider the sole comment that followed Mr. Baker’s [so much for Baker's PhD] blog post on Business Insider: … "Budget deficits aren’t being used for investment. They’re being used to fund wars and pay for today’s consumption (healthcare, welfare, etc.)" … I couldn’t have said it better.


These examples are bad investments irrespective of the budget balance. Rattner is pulling a bait and switch: he starts by saying that the quantity of debt is the issue (how much debt plus interest) and now he tells us it's the quality that matters (what it's spent on and who's taxed to repay it).

Baker (and Krugman) are saying Rattner is making a composition fallacy: on net there is no burden and no benefit, though tax collection isn’t frictionless. Rattner then insists some people/institutions benefit more from wasteful government deficit spending than others as recipients of the initial disbursements (e.g. Lockheed Martin), interest on the bonds (probably Lockheed Martin's board of directors and their heirs), and low taxes (Lockheed Martin again).

But Rattner didn't write that essay, and in his original article (at the end) he said that he was in favor of increased short-term deficit spending. Meanwhile, neither myself nor my heirs will suffer a reduced standard of living if we raise taxes on land values, reform IP laws, and rent out EM spectrum rights and geosynchronous orbits.
...
written by Troy, January 22, 2012 8:44
Waldmann is apparently missing the obvious point that for every taxpaying debtor their is a creditor.

Borrowing today is both enriching and penurizing the next generation, in equal measure.

Also, good luck with "We can exchange the bonds for goods and services". Next time I go shopping I'll take my Series EE bonds in and get a quote.
...
written by Robert Waldmann, January 22, 2012 9:21
@ Troy I note that your quarrel is not with me but with Peter Diamond. I suggest you read the paper before asserting that it is obviously wrong.

My claim is that the current generation can sell the bonds to the next generation. The claim is that if the current generation is selfish or myopic, then public debt can crowd out private investment. This makes future generations poorer.

My claim is just that if one owns something, one can sell it. One does not have no choice but to leave it to one's heirs. It is entirely possible to write down a model in which Baker's claim is false. The claim that it conceivably could be false is not an error.

This is not a point disputed in the peer reviewed economics literature. My claim is a brief summary of a paper published in the AER. It is not an elementary error. You are confidently asserting that you understand inter-generational accounting vastly better than Peter Diamond does.

Baker's claim is that we know that there is Ricardian equivalence. This would imply that tax cuts do not work at all as a stimulus.

Only fresh water lunatics (like Lucas) make the argument that Baker made. In fact, the only other example which comes to my mind is Lucas's diatribe against C Romer in which he asserted (not his worst howler) that we know that there is Ricardian equivalence. In fact it is a not very interesting clearly false hypothesis which certainly can be false.

I am writing about a claim that something must be true "have no choice" so I don't need to discuss data. However, the data show that unprovoked tax cuts (1981 and 2001) are followed by high consumption/GDP and low investment, just as Diamond would guess.

Right now we are in a liquidity trap so, only if Baker is wrong, higher deficits would be good for the economy. Only if he is wrong is there any sense in temporary tax cuts designed to stimulate demand. Given the liquidity trap, higher consumption wouldn't crowd out investment but rather cause higher investment.

But if Baker is right, then there is no point in say a Payroll tax holiday (a permanent reduction with replacement by a VAT or higher capital gains taxes would make sense also higher government spending would stimulate demand even if the Baker/Lucas/Barro hypothesis were correct). He is arguing that Keynesian economists are totally wrong about the effects of deficit spending.
Robert Waldmann is right
written by Nick Rowe, January 22, 2012 9:25
And Robert Waldmann's point is well-understood by most macroeconomists, and has been for the last 30 or so years.

Dean: "They will have no choice but to hand this debt on to members of the next generation, either their own heirs or someone else's."

They do have a choice in how they hand on this debt to members of the next generation.

If they sell the bonds to the next generation, as in an OLG model like Diamond's, then there is a burden on the next generation, which pays for the bonds but is not paid for taking on the liability.

If they give the bonds to the next generation, as in Barro's paper on Ricardian Equivalence, then they must have saved the whole of the bond-financed tax cut.

Steve Rattner is right. Unless you want to argue Ricardian Equivalence. Which I assume you don't.
The way government debt usually works
written by Ben, January 22, 2012 10:11
Since the federal government doesn't usually pay down debt but rather rolls it over (we tried to pay it all off but the Bush Admin a d Congress intervened). The holders of 30 year debt today will be paid off and others will buy the new bonds issued to rollover the debt. Dean's model holds in reality. Since the US government controls its currency and its debt is denominated in dollars; we can't be Greece. While morally it would be better if we funded more education or health care and less war; the impact on the debt of its quality is irrelevant.
...
written by ellis, January 22, 2012 10:44
What about the impact of financial speculation on the debt, things like credit default swaps. No one ever talks about that. Is it important? It seems to be in the case of Europe. And perhaps the low U.S. interest rates are due to speculators seeking a refuge from Europe. If that is the case, then couldn't that flow of speculative money reverse, causing much higher interest rates in this country also? Is there anything in those models that take this into account?
If the Economy Is Fully Employed then the Moon is Made of Green Cheese
written by Dean, January 22, 2012 10:54
There are several people on this thread who apparently believe the economy is at full employment. If this were true then additional debt would create a burden on the economy (although it still would not be one to one, much of the impact would come out of reduced current consumption, which means that the debt is a problem for the present generation, not future generations.)
But, since we are not at full employment or anything close, then the government borrowing does not come at the expense of private saving. It generates NEW saving through increasing output.

Come on folks, the problem is a lack of demand. More demand, leads to more output, and (drumroll please) more saving. No Ricardian equivalence her, just the good old Keynesian economics that we all got to know and love many years ago.
...
written by skeptonomist, January 22, 2012 10:59
Here are some important questions. Rattner moves between the worlds of banking, government and journalism. How many revolving doors are involved here, two or three? Which door did Rattner come through to deliver this op-ed?
...
written by AndrewDover, January 22, 2012 11:01
Just a reminder that $4.7 trillion of the U.S. debt is held by other countries.

http://www.treasury.gov/resource-center/data-chart-center/tic/Documents/mfh.txt

Essentially, we chose to lower taxes rather than repay that debt. Roughly speaking, the U.S. built houses with more square feet, and bought larger TVs with the lower taxes.
...
written by skeptonomist, January 22, 2012 11:10
The situation of large increases to debt under full employment is hardly something that we need to guess or theorize in purely hypothetical ways about; it has happened repeatedly during major wars. The most recent and important example was WW II. What were the long-term results? Could we have some effort to connect these debates with reality?
Thank you for taking on Steve Rattner, Dean
written by jhand, January 22, 2012 1:26
When I first read the column in the NYT, and saw how he began with the straw man of progressives holding dangerous ideas that debt didn't matter, I looked for the comments icon to respond. Finding none, I concluded that Rattner preferred to exchange smugs with Joe on Morning Joe, rather than allow us great unwashed to challenge him. I am no economist, but I do read you and Krugman regularly, and knew right away that Rattner's column was nothing more than an attempt to look Serious. Thanks for speaking up on the column. Serious People don't like little people to disagree with them in the comments section. Your doing it here provides a valuable service.
...
written by Nick Rowe, January 22, 2012 4:05
Dean: "If this were true then additional debt would create a burden on the economy (although it still would not be one to one, much of the impact would come out of reduced current consumption, which means that the debt is a problem for the present generation, not future generations.)"

Nope. Take the simplest possible OLG model with people living two periods and no investment. Consider a gift of bonds from the government to the young members of the current cohort (A). So far no change in their consumption. When they are old they sell the bonds to the young members of the next cohort (B). So A's consumption is higher when old, and B's consumption is lower when young. Suppose the government rolls over the debt, and when B are old they in turn sell the bonds (plus accumulated interest) to the young members of C. So B have lower consumption when young, and higher consumption when old, and are compensated for postponing their consumption. But if the government increases taxes to pay off the debt, or even pay the interest (which it must do if the debt is growing faster than the economy), then C will have lower consumption when young, and also lower consumption when old.

From long experience I have learned that the only way economists who have been brought up (like I was ) in the "we owe it to ourselves" school, is if they actually sit down and work through a numerical example for themselves.

Does the result change if there's deficient-demand unemployment? "Not really" is the short answer. It's just that the benefits to cohort A (if there's a multiplier effect) may be bigger than the losses to cohort C (if taxes don't increase until the economy is back at full employment).
...
written by shooter242, January 22, 2012 4:17
Basically this is just a redistribution scheme.
* Interest comes from taxpayer payments to Treasury
* Treasury pays interest to the Fed
* Fed pays the interest back to Treasury
* Treasury spends the interest on Govt programs.

Hmmm. Maybe it's money laundering instead?
Where Does the Money Go When The Old Sell Their Bonds?
written by Dean, January 22, 2012 8:23
Nick,

You are constructing a scenario in which there is a fix amount to be consumed across generations and you have required that the older generation consumes the full amount of any government debt issued during their lifetime. If those conditions hold, sure we all know arithmetic.

However, I see zero evidence for either condition. People in the world do leave inheritances, which means that they do not consume 100 percent of their assets.

More importantly, especially in the current context, output is absolutely not fixed. If the government deficit is used for investment (public or private) that increases output in the future, then there is no reason whatsoever that it need depress future living standards, even the older generation ate 100 percent of it. All you would need is that the return on the investment is greater than the interest rate. With the interest rate on long-term debt negative, this is a pretty low bar at present.
The old use the money to buy goods produced by the young
written by Nick Rowe, January 23, 2012 4:42
Dean: Yes, I constructed a very simple model with fixed endowment and no investment and no bequests to use as a counterexample. I need one counterexample to show that there is something very wrong with the view that it is impossible to put a burden on future generations (except through distorting taxes or lower future capital stock or debt to foreigners). My counterexample shows that these arguments are invalid:

"we can't make apples travel back in time so that today's generation eat apples produced by people living 100 years in the future so we can't impose a burden on future generations"

"we owe it to ourselves so there is no burden on future generations"

My one counterexample knocks out those "impossibility theorems".

OK. Now let's introduce bequests. Imagine a model in which each person has one heir in the next generation (one kid per person, yes, no sex in my model!)

Assume the government gives a transfer payment of one $100 bond to each person in cohort A. That person reasons: "Hmm, my heir, or his heir, will have to pay higher taxes to pay the interest and principal on this bond. Those taxes will impose a burden on my heir or his heir, unless I do something to offset that burden. The Present Value of those taxes is $100. I need to save an extra $100 today to offset that future tax burden. Therefore I cannot sell this bond; I must save it all, including the accumulated interest, and bequeath it to my heir, so that he in turn can bequeath it to his heir."

And we are now in the world of Barro/Ricardian Equivalence. The current cohort save all of the bond-financed transfer payment.

Now introduce demand-deficient unemployment, in addition to bequests. The current cohort save all of the bond-financed transfer payment, and so there is zero Keynesian multiplier. (There may still be the "balanced budget multiplier" from bond-financed increases in government spending, but that's another story).

Bottom line: the "man in the street" who sees the debt as a burden on his kids and grandkids is right. It's the extra future taxes that constitute the burden. Only if he takes offsetting action, by saving all of any deficit, and passing it on as a bequest, is there no burden. But then there's no keynesian multiplier either (except the BBM).

The only ways (I can think of) to escape this are:
1. The government invests in e.g. schools, which benefit future generations and offset the tax burden. The schools are the bequest.
2. No bequests, so the current generation spends the bond-financed transfer, and the keynesian multiplier is big enough to kick-start the economy back to full employment, and it stays at full employment even if taxes are increased later. The next generation pays the burden of higher taxes, but it also gets jobs, so may come out ahead on balance.
3. Samuelson 1958. Assume the real interest rate on bonds is permanently less than the growth rate of the economy, so the debt/GDP ratio falls even if the government rolls over the debt+interest forever, and so taxes need never be increased, so there never is a tax burden on future generations.
It matters what the government buys
written by Nick Rowe, January 23, 2012 5:05
Dean: "If the government deficit is used for investment (public or private) that increases output in the future, then there is no reason whatsoever that it need depress future living standards, even the older generation ate 100 percent of it. All you would need is that the return on the investment is greater than the interest rate. With the interest rate on long-term debt negative, this is a pretty low bar at present."

Yes. Absolutely. Government investments that pass the standard Cost/Benefit test will create benefits to present and/or future generations that exceed the costs.

Plus (and not many Keynesians understand this I believe) such investments will increase permanent disposable income, and will cause a "crowding in" of increased current consumption even in a Ricardian world, and so give a bigger multiplier than just spending on pyramids.

We can't just talk about "increased government spending". It really matters what the government buys with its increased spending. It matters not just for all the standard micro reasons, but for macro reasons too.
They Can't Take It with Them
written by Dan Kervick, January 24, 2012 7:12
There are three scenarios:

1. The older generation passes on all of the the government bonds, totaling $X, as a bequest to the younger generation. The younger generation finds themselves in the possession of financial assets totaling $X and also a government liability totaling $X dollars. Net aggregate impact zero, although there will be various distributional effects depending on the method the next generation uses for retiring the liabilities.

2. The older generation, in possession of government bonds totaling $X, tears up and destroys some proportion of those bonds and passes on the remainder as a bequest to the younger generation. The value of the bonds destroyed is $Y. Thus the younger generation finds themselves in the possession of assets totaling $(X-Y) and government liabilities totaling $(X-Y). Still nets to zero.

3. The older generation sells all of its government bonds, with a total value of $X, to the members of the younger generation for a total price of $Y. That means that in the present, $Y is transferred from the younger generation to the older generation. As a result the younger generation finds themselves in possession of assets totaling $(X-Y) and government liabilities totaling $(X-Y). And they have also relinquished $Y dollars to the older generation. But what happens to those $Y dollars. Those dollars are now passed on as a bequest to the younger generation. So at some later time the younger generation ends up with assets totaling $(X-Y), government liabilities totaling $(X-Y) and $Y returning to them that were relinquished at the earlier time. It still nets to zero.

The older generation can and will indeed sell some proportion of its bonds to the younger generation. But I don't think there is a coherent model in which there is a significant net loss of dollars to the older generation. The older generation cannot take those dollars with them in their coffins.

I also don't see how the taxes make a difference. If the younger generation decides to charge themselves a tax on the redemption of government bonds. The tax, when it is imposed, is just a redistribution of income among the members of the younger generation. The private sector's assets are initially reduced via taxation by some amount $Z. But those additional revenues offset the government's debt liabilities, and will add to the government's net spending back into the private sector.

All the above said, these operations could end up being burdensome on the younger generation if in carrying them out they make stupid decisions about their levels of consumption and investment of real goods in the time periods in which they make the decisions.
...
written by Nick Rowe, January 24, 2012 1:37
Dan:

Your 1 is Ricardian equivalence, which means all of a bond-financed tax cut is saved, and doesn't increase demand. Dean doesn't believe in Ricardian Equivalence (nor do you, if you are an MMTer).

Your 2 is the same as 1.

For your 3, you need to read my original post. The young buy the bonds with apples, the old eat the apples, then die, then the young find that they themselves have to pay the taxes to pay for the bonds they own.

Try my post: http://worthwhile.typepad.com/...lence.html
...
written by Dan Kervick, January 24, 2012 3:39
Nick,

Isn't it only Ricardian equivalence if the older generation reduces its consumption by $X in anticipation of the fact that the the younger generation will be taxed by $X to make the bond payments? But why should they think that way? Leaving aside all MMT-related considerations, and assuming all government debt payments are matched dollar-for-dollar by a tax to fund the payments, the taxes and debt payments only represent a shift of financial resources among the members of the same cohort, and no change in their overall income.

Suppose some group of bonds valuing $X are issued in 2010 and come due in 2020. At that time, those bonds are all owned by people alive in 2020. People alive in 2020 own the liabilities and the assets If the government in 2020 taxes $X to pay off the bonds, then some people alive in 2020 lose $X total in taxes. But some other people alive in 2020 receive $X when their bonds are received. So it's just an income transfer among people alive at that time. This is true for $X, $2X, $3X or any amount less than the total income I.

What happened back in 2010, let's say, when the bonds were first issued? Let's assume, in parallel with the assumption that the government always taxes in order to make debt payments, that the government always spends everything it borrows. So in 2010 some group of people alive in 2010 bought bonds at some total price $Y, which transferred $Y from the private sector to public sector, which is then spent back into the private sector. So the whole operation is just a shift of $Y in income from one group of people alive in 2010 to another group of people alive in 2010.

Is the transfer of some of their income good or bad for the people in 2010? It all depends on what kind of transfer is involved, and whether that transfer results in a net overall increase in demand, production and valuable material activity or not. The usual defense of these kinds of public sector spending operations funded by taxes and borrowing is that they channel financial resources out of places in which they have a relatively high propensity to be held into the hands of people who have a higher propensity to apply them directly or indirectly to increased production, either by undertaking production oneself or by demanding additional consumption goods that will in turn cause additional production.

There is never any literally any "inter-generational" movement of funds. Money flows are always contemporaneous, from people alive at a given time to other people alive at that time. The functional effect of the above would be more-or-less the same if there were a straightforward tax shifting $Y around in 2010, followed by a completely independent tax in 2020 shifting $X in income around. The mediation of bonds, rather than independent, successive taxes, in a way just amounts to a kind of pre-commitment in 2010 to execute the tax-funded income shifting in 2020, and it places some prior constraints on who will be on the receiving end of the shifting: primarily either the people who purchased the original bonds, or the descendants of those people who inherited their assets.

Where some MMT reasoning comes is in relaxing the requirement that the total bond sales in 2010 and the total bond redemptions in 2020 represent equivalents shifts in income. Some of the bonds are in effect purchased by the central bank (at least following open market operations). So some of the money that is added to the treasury in 2010 is not shifted out of private sector accounts, but is simply a credit to the government from its own central bank. And when those bonds are redeemed in 2012, some of the redemptions will not be matched by additional income-shifting taxes, but will instead be redeemed by just another credit from the central bank. The only constraint of this ongoing process of net money creation is the constraint of price stability, which isn't a problem so long as affairs are managed so that the added spending generated by each injection of money creates new goods and services
...
written by Dan Kervick, January 24, 2012 5:16
But some other people alive in 2020 receive $X when their bonds are received.

I meant to write:

But some other people alive in 2020 receive $X when their bonds are redeemed.
Interest Rates and Growth Rates
written by anon, January 24, 2012 8:30
Dan, Nick is assuming the interest rate is held permanently above the growth rate. This is unsustainable.

He says keeping the interest rate permanently below the growth rate is a ponzi scheme. Which is possibly sustainable but may cause hyperinflation.
Dan, IR and GR, continued...
written by anon, January 24, 2012 8:39
But Krugman and Baker may be assuming the interest rate is roughly equal to the growth rate over time. (Which is neither unsustainable nor a ponzi scheme.)

This means taxes will only have to rise (roughly, over time) and the growth rate rises (roughly, over time).

Brad Delong had a blog post showing the interest rate roughly equal to the growth rate for a period of fourty years.

Brad Delong: Is There Any Reason to Think That the Interest Rate on U.S. Government Debt in the Future Will Be Greater than the Growth Rate of the Economy?
IR and GR, edit...
written by anon, January 24, 2012 8:48
Taxes will only have to rise *as* the growth rate rises.
Ponzi Scheme?
written by Dan Kervick, January 24, 2012 11:02
I don't see any way that a sovereign currency issuer can run a Ponzi scheme.

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.

busy
 

CEPR.net
Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.

Archives