Wednesday, July 21, 2010

Very Simple Debt Dynamics

The starting point is the budget identity: the sum of all budget positions in a country, public and private combined, is equal to the current account surplus or deficit. Separate out public and private and you get

Private budgets + Public budgets ≡ Current account

As a mental exercise, suppose the CA is fixed. In that case, if the public budget position increases by $1B (either a smaller deficit or larger surplus), the private budget position must fall by the same amount. Of course, no single component is fixed; they all change simultaneously in order to maintain the identity.

“Identity” is an important word here. What you see above is not an equation (with an equal sign), but an identity with an extra little bar. This means it is not a behavioral relationship, one which may or may not be true depending on how agents behave. It is an identity: it’s always true at every moment of every day, brought to you by the gods of accounting.

Since it has no behavioral content, it tells you nothing about what adjusts to what, or how. Nevertheless, it is a useful starting point—the most useful starting point.

Consider a first problem, adjustment. This arises when a deficit country (this refers to a CA deficit) is no longer sustainable. A crisis could take the form of a run on foreign exchange, or a sudden stop in external lending. A deficit country, by virtue of the budget identity, is a net borrower: its households, firms and government borrow more than domestic savers can finance. This exposes the country to the risk that external capital markets will shut down or become too expensive to access. Adjustment means that a country must do two things, which according to the identity are actually one thing: rapidly reduce the sum of public and private borrowing and reduce the current account deficit. Surplus countries don’t have to adjust; they are net lenders. They face default risk, but that can’t be remedied by changes in their own policies, at least not to a first approximation. (At a detailed level it depends on how large and connected they are.)

Keynes didn’t like this asymmetry of adjustment; he thought it lent a deflationary bias to the global economy. He was right, but there wasn’t anything he could do about it. This asymmetry is the dominant fact about the international financial system today as it was in his time.

Adjustment is what Greece and Hungary are going through right now, and what the other peripheral European countries are staring at.

Additional governments who are not threatened by adjustment are nevertheless planning to cut public (fiscal) deficits in the near term. Some of these are deficit countries like Britain, which is not completely crazy to worry about abandonment by international markets. Some are surplus countries like Germany, which is completely crazy (in this respect).

That brings us to the second problem. Recall that, if the CA is unchanged, any reduction in fiscal deficits must be offset by an increase in private deficits. There is no plausible mechanism for this, however. In fact, the most probable adjustment to preserve the identity will be the CA itself. If private borrowers do not change their behavior, or if they actually continue to deleverage, the logical sequence is that fiscal tightening will lead to a decline in national income and therefore a surplus or reduced deficit on the CA. In other words, the more elastic component of this identity is the external position, due to contractionary fiscal policy. But reductions in national income will also reduce public revenue, which means that the achieved reduction in the fiscal deficit will turn out to be less than the intended. To summarize, unless you can tell me why fiscal tightening (government spending cuts and tax increases) will cause households and businesses to become greater net borrowers, the tightening is offset instantaneously and unavoidably (this is an accounting identity after all) by some combination of automatic stabilizers and reduced imports, both due to a shrinking economy.

What does this mean for intelligent policy? First, Keynes was right: if the private sector has stopped borrowing, the public sector must leap in and take its place, and this must continue as long as the private sector remains skittish. If that imperative leads countries into the maw of adjustment—well, we need international institutions that spread adjustment across surplus and deficit countries alike, so that the contractionary impact on the latter is offset by the stimulative impact of the former.

Second, if lots of bad debts have been incurred, and if the amount of time it will take to wind them down and return to healthier levels of consumption and investment is too long, it is better that there be an orderly writeoff of a large chunk of the debt overhang. Alas, this was not central to the bailouts of the last two years as it should have been, even though the financial system had accumulated trillions of dollars in bad debt. Either we do this in a rational, civilized way, or the economy will sputter until default breaks out chaotically.

Unfortunately, the creditors are in command across the world economy and can think only of squeezing out every last cent of their assets.


Don Levit said...

Would you consider the U.S. a deficit country?
How would you measure when the debt becomes "too high?"
Would intragovernmental debt increase the total debt, reduce the total debt, or have no effect on the total debt?
Don Levit

Peter Dorman said...

The US is most definitely a deficit country: our CA deficit is getting up into the 5% range, which is dangerous territory.

It's important to distinguish between different deficits. The external deficit is the current account deficit, mostly governed by trade. The fiscal deficit is the government's budget deficit. Then there are private budget deficits: financial firms, nonfinancial firms, households.

When you ask whether "the debt" is too high, I assume you mean the fiscal deficit. Short answer: (1) At the moment it is too low, and this is why we have an unacceptably high level of unemployment and a real risk of deflation. (2) It is manageable in the medium term because the rest of the world is willing to finance it, and even if they weren't, the Fed can do it and take advantage of the dollar's position as the world's reserve currency. (3) In the long run it is necessary to reduce the average ratio of the fiscal deficit to GDP to the level of economic growth, or a bit below. Very doable.

If by intragovernmental debt you mean debt between different levels of government in the US, the answer is that "the fiscal deficit" is the composite of local, state and federal government. If you mean international sovereign credit arrangements, well, government borrowing is borrowing, and government lending is lending, so in that sense it doesn't matter who the other side of the transaction is. The US fiscal deficit is the same size whether it is financed by German pensioners or a Kuwaiti sovereign investment fund. Am I missing something?

Don Levit said...

By intragovernmental accounts, I am referring to the government borrowing from itself, primarily through the Medicare and Social Security trust funds.
Up until 1983, Social Security and Medicare had little surplus, basically one year's of payments.
After 1983, when a new financing plan was designed, the surpluses increased exponentially.
Now, when the outgo exceeds the annual income of the trust, the public debt is increased to make up the difference (unless there is a budget surplus).
This happened for the last 3 years with the Medicare HI trust fund, and occurred this year with the Social Security trust fund.
Many economists downplay intragovernmental debt, and place primary importance on the debt to the public. This is understandable in that the principal is rolled over, as with the public debt, and the interest is paid for by issuing Treasury securities. So, up until now, for the most part increasing intragovernmental debt has had little or no effect on the present economy. The thinking may change in the near future as various trust funds start to dwindle.
Don Levit

Peter Dorman said...


Now I understand where you're coming from. One can slice the accounting identity up into as many pieces as one likes. You could subdivide the public sector into various separate budgets, but what I offered -- at the simplest possible level -- was a sort of unified public budget including all levels and funds. This means that the current SS surplus represents a source of government saving, to be deducted from deficits elsewhere. When SS starts to draw down its accumulation, this will be a source of dissaving. No argument there.

Quantitatively speaking, the non-Medicare portion of SS will have only a small influence on the unified public budget post 2020. This is separate from the question of how large the unified fiscal deficit should be in the first place. The implication of my post is that this is inseparable from the challenge of closing the CA deficit. As long as the US runs massive trade deficits, we will either be a big private borrower (as in the mortgage fiasco), a big public borrower, or a bit of both. And shutting down public borrowing at a time of private deleveraging simply hammers national income and employment.

Don Levit said...

It seems like you prefer increasing public debt to increasing private debt: sort of the lesser of 2 evils.
Is that because public debt has a more deferred effect on the economy than private debt?
I agree with you about the effect that Medicare will have on the budget compared to Social Security.
While we are looking at the total economy, I found a startling bit of information yesterday.
In a paper entitled "Analytical Perspectives, Budget of the U.S. Government, Fiscal Year 2009" it states on page 345, "In 1984 a new system was set up to finance military retirement benefits on a full accrual basis. In 1986, full accrual funding of retirement benefits was mandated for federal civilian employees hired after Dec. 31, 1983. Since many years will pass between the time when benefits are earned and when they are paid, the trust funds will accumulate substantial balances over time. THESE BALANCES ARE AVAILABLE TO FINANCE FUTURE BENEFITS, BUT ONLY IN A BOOKKEEPING SENSE. THESE FUNDS ARE NOT SET UP TO BE PENSION FUNDS, LIKE THE FUNDS OF PRIVATE PENSION PLANS. WHEN TRUST HOLDINGS ARE REDEEMED TO AUTHORIZE THE PAYMENT OF BENEFITS,THE DEPARTMENT OF THE TREASURY WILL HAVE TO FINANCE THE EXPENDITURE IN THE SAME WAY AS ANY OTHER EXPENDITURE: BY USING THEN CURRENT RECEIPTS, BY BORROWING FROM THE PUBLIC, OR BY REDUCING BENEFITS OR OTHER EXPENDITUIURES.
I was amazed to find out that the federal employees' benefits were financed like those for Social Security. Since state and local governments have pension plans, with mostly private assets and a trust that is devoted to accumulating those assets, I thought the federal plan for employees worked the same way.
Don Levit

Peter Dorman said...

The problem with private borrowing is that it is in the hands of the public, largely beyond the reach of policy, certainly in the short run. Public borrowing is a policy variable. It also happens, under the current conditions, that nearly all non-sovereign assets are regarded with terror by investors, but the government can borrow cheaply at will. My biases, such as they are, reflect these circumstances. What I really want, however, is a speedy and non-austere end to the current account deficit.

As for your discovery, this holds equally for SS. It should not be viewed as shocking (as seems to be implied by ALL CAPS). First, a dedicated fund that holds treasuries can always cash them in as long as Uncle Sam doesn't default. This puts public retirement trust funds in the same position as you or me with our personal retirement funds invested in the same assets (if we have them). And yes, retirement benefits will have to be paid when they have to be paid; there's nothing we can do about that, is there? That's an argument for foresight, but not for or against the benefits themselves.

Don Levit said...

The Social Security system has been based on the philosophy that benefits are financed by payroll taxes, that the system was to be self-sustaining, that no funds would be used from general revenues.
Being that the payroll taxes go into the Treasury, like all other revenues, and the "excess" was spent on current government expenses, what is left other than general revenues to pay for benefits?
It is indeed shocking that the federal employees retirement is paid by 100% Treasury securities, which have already been spent for current expenses - for they are merely bookkeeping entries.
How is that full accrual funding?
There is no reason, financially, or legally, in my opinion, why the employees of the federal government cannot have a pension plan like the employees of a private company.
When benefits are paid, public debt will have to be increased (unless we have a surplus or benefits are reduced).
This seems to be set up by design to increase the public debt!
Don Levit

Unknown said...

Really its a great post! All the informations are in your post are making knowledgeable us.Thanks a lot for this nice post.

debt advice