August 1, 2011
Debt and the Deficit
Immediate and Medium Term Debt and Deficit Problems-Becker
As Posner indicates, the US is facing an immediate debt problem, a medium term deficit problem, and long-term debt and deficit problems. We have discussed several times the long-term fiscal problems (see, for example, my discussions on 11/07/10 and 7/17/11), so I will concentrate here on the shorter-term deficit-debt issues.
With the sole exception of the mid-1990s, the federal debt ceiling has always been raised on time during the past 60 years to accommodate the growing level of the debt. Although little time remains to raise the ceiling I do not believe either party will risk the political fallout from not adjusting the ceiling before the government effectively defaults on its obligations. Certainly Republicans remember the political cost to them after the Gingrich-inspired shutdown of the federal government in 1995. So I would be quite surprised if a deal is not reached this time before the US government is forced to delay payment on some of its implicit obligations, such as Medicaid and Medicare payments.
Of course, an absolute ceiling on the debt that is just raised continually makes little economic sense. Clearly, the relevant size of government is not measured by the absolute level of its debt since large economies tend to have much more debt than do small economies. Better metrics of the debt burden would be the ratio of debt to GDP, or the ratio of interest payments on the debt to GDP, or perhaps the share of interest payments in total government spending.
Even the ratio of government debt to GDP does not measure the size of the government's impact on the economy. The present and future impact is better measured by the ratio of government spending to GDP, including any trend in this ratio, plus various harder to calculate measures of the magnitude and scope of government regulations in product, labor, and capital markets.
The major legitimate concern over the growing federal government impact on the economy comes from the rapid growth in the share of government spending to GDP during the past several years. Moreover, this share is projected to continue to grow in the future unless GDP grows by at least 3% per annum over the next 20 years and effective controls are introduced over entitlements for the elderly.
The share of federal government spending to GDP hovered between 18% and 21% from 1995 to 2007, when it was 19.5%. During the next three years, government spending took off and grew rapidly, GDP stagnated, and the ratio of federal spending to GDP reached 25% in 2009. This share is projected at about 25% in 2011. Some of the growth in federal spending was needed to fight the financial crisis, but much of the increase also reflected the desire of many in control of the federal government to increase its role in the economy.
In a Wall Street Journal op-ed piece by George Shultz, John Taylor, and myself on April 4th (Time for a Budget Game-Changer") we lay out a general strategy to gradually lower this share once again to 19.5% by 2021. This strategy reduces government spending by a mere $19 billion in 2011, or only by about 0.1 percentage point of total federal spending. It also allows nominal federal spending to grow on average by 2.7% per year from 2010 until 2021.
Our approach also assumes that nominal GDP will grow by 4.6% during this eleven-year period. Since GDP is assumed to grow considerably faster than government spending, the share of government in the economy is reduced by over 4 percentage points by 2021. If we assume (optimistically) an average inflation rate of 1.7% per year during the next decade, our analysis allows government spending in real terms to grow from its already high level in 2010 by 1% per year until 2021.
The assumed growth in real GDP (after adjusting for the assumed 1.7% annual inflation rate) is a little less than 3% per annum over this period, with about 1% growth per year due to increases in the American population. Three percent per year in real GDP is the long-term growth rate of the American economy since 1880. This is considerably below the usual growth rate of GDP in a decade after a major recession. This is why I believe three percent growth is achievable with tax reform (that would include some tax increases), and with major controls over the growth of spending.
Keeping real federal spending growth to 1% per year will be a huge challenge since entitlements, like Medicare, will grow much faster than that over the next decade without sizable entitlement reforms that greatly cut their cost to the government. Still, the spending growth rate is doable. A major start would be to cut back some of the sizable expansion of federal spending since 2007. That combined with serious reforms that cut government spending on Medicare, especially for the elderly with decent financial means, would go a long way toward keeping the real growth of federal spending in the range of 1% per year for a decade.
Added Monday: first reactions after news surfaced of the possible budget compromise.
The details of the compromise are not yet clear, but apparently Medicare and other entitlements will not be affected, with one exception. Payments to hospitals and insurance companies for the services they provide or finance will be cut, which will either lower the quality of services offered or induce greater waiting times for medical services.
The compromise seems at this point better than doing nothing. But much more will have to be done, rather than simply promised for the future, in order to limit the growth in real federal spending to not much more than 1% per year.
The Depression, the Deficit Debacle, and the Debt-Ceiling Crisis—Posner
The economy faces a short-term problem, a medium-term problem, and a long-term problem. The short-term problem is the debt ceiling, the medium-term problem is the depression that the economy is still wallowing in (the orthodox description of it as a mere "recession" that ended in 2009 is misleading), and the long-term problem is entitlements for old people—Medicare, social security, and (to a lesser extent) Medicaid, which to a significant degree operates as medigap insurance for many old people. The three problems are intertwined. The third, the long-term one, may well be the most serious.
The debt ceiling—a dollar limit on debt owed by the federal government, which can be increased only by Congress's enacting a statute raising the limit—is a dysfunctional method of legislative control over government expenditures. Without the ceiling Congress could still limit borrowing by the Treasury, but it would have to do so by passing a statute. The existence of the ceiling means that enacting a statute is necessary to permit borrowing above whatever amount was specified as the ceiling the last time it was raised by statute. It is harder to pass a new statute than to defeat a proposed statute, and so a determined legislative faction may be able to extort unreasonable concessions by threatening to block the enactment of a statute raising the ceiling. If the intended victims of the extortion balk, and a game of chicken ensues, the statute may not pass; and if as a result the debt ceiling is not raised, when as at present the government must borrow to have enough money to fulfill its expenditure commitments, very serious consequences can ensue. At present the federal government is spending about $300 billion a month, of which about $83 billion is borrowed. If it cannot borrow that amount any more, because every time it borrows (unless it's just rolling over a loan) its debt rises, it will be unable to fulfill its spending commitments. It will not default in the technical sense because its tax revenues are sufficient to service the debt, but contractual debt (bonds and the like) is not the only unavoidable expense of the federal government: there are both the normal civilian and military costs of running the government and the huge entitlements on which a significant fraction of the population is dependent. Inability to pay these costs would be de facto insolvency.
The drastic curtailment of federal expenditures if the debt ceiling were not raised would have a devastating effect on the current very weak economy, because $83 billion in monthly spending cannot quickly be replaced by private spending; it's not as if $83 billion were being shifted from the government to the private sector. It would mean taking a trillion dollars a year out of the economy. Maybe private foundations would take up some of the slack, but if so they would be diverting money from other recipients of the foundations' largesse rather than increasing the net amount of cash available for consumption, unless they diverted money from overseas recipients.
There would be some long-run benefits from eliminating further federal borrowing. The benefits would lie mainly in forcing governmental economies and reducing the annual interest expense of the government. Rational-expectations economists would argue that foreseeing lower taxes in the future would stimulate consumers and businessmen to spend more than now. But consumers, investors, and businesses might be held back by uncertainty; and certainly the short- and medium-run dislocation of an already weak economy could be catastrophic. The reduction in government expenditures could not be matched immediately by an increase in private spending, so overall economic activity would plunge.
The current weakness of the economy cannot be overemphasized. Adjusted for inflation, GDP has fallen since 2007 by 0.4 percent. That means that per capita GDP has fallen significantly because of population growth and that current GDP is almost 10 percent below the GDP trend line of 3 percent a year. There is a better argument for the Fed's stimulating inflation to reduce mortgage and other consumer debt in real terms than for Congress's cutting government expenditures.
An irony in the present situation (and a powerful argument against Republican proposals to lift the debt ceiling for only six months in order to force reductions in government spending that would be necessary to induce House Republicans to support a further increase in the ceiling at the end of that period) is that the actual proposals for debt reduction are meager and probably illusory. Under the deal tentatively agreed upon yesterday the initial cut will be just $900 billion spread over the next ten years, which would amount to only $90 billion a year, or less than 10 percent of the annual deficit in the federal budget. And this is on the assumption that the deficit won't grow. But it is quite likely to grow. True, if the economy recovers, tax collections will increase because incomes will be rising, and even without increased tax collections the deficit as a fraction of GDP (which is what's important—not the absolute size of the deficit) will fall. But against this is the likely rapid increase in entitlements, primarily Medicare and social security, as the over-65 population continues its relentless growth. The only measure to slow this increase that seems politically feasible at present is to change the formula for calculating annual social security cost-of-living increases. A sure sign of the phoniness of the rival Democratic and Republican plans to cut government expenditures was that both rely heavily on a crackdown in Medicare "waste and fraud."
The deal tentatively agreed to calls for further cuts (or tax increases) of about $1.5 billion, also spread over the next ten years, to be proposed by a bipartisan commission and take automatic effect unless Congress acts—as it could do: it could decide to rescind the cuts. Alterations in social security have been taken off the table; and the only cuts in Medicare that the negotiators are permitted to order are reductions in reimbursement of hospitals and physicians—that will have only an indirect effect on the expense of Medicare, by lengthening waiting time for medical services, and that will create pressure to rescind the reductions.
As with adjusting the social security cost of living formula, and the successful effort in the 1980s to raise the eligibility age for social security gradually, large-scale reductions in entitlements are feasible only if phased in gradually—which would have to be done anyway in order to avoid a serious jolt to the current weak economy. The problem is that Congress cannot make credible long-term commitments to reduce spending because it cannot bind subsequent Congresses. The problem is exacerbated by the fact that both political parties are much fonder of increased spending than of increased taxes, so there is built-in momentum for increased government debt. The Republican radicals in the House of Representatives recognize this problem and their frustration is understandable, but national insolvency is not an intelligent solution.