The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Note: This report is available in its entirety in the Portable Document Format (PDF).
Almost all U.S. federal budget experts predict a financial crisis in the long run if entitlements are not reformed or taxes increased far beyond what Americans have known. The financial community's lack of concern is more than a little puzzling. There is no doubting today's arithmetic. Almost half of civilian spending outside interest on the national debt goes to people
65 and over, with Social Security and Medicare the dominant programs. Spending on the elderly will explode as health costs continue to outpace income and the number of elderly soars as life expectancy increases and the baby boomers age. Meanwhile, the number of workers available to pay this staggering bill will stagnate because of low birth rates since the mid-1960s.
Although a financial crisis is widely forecast absent reform, few analysts speculate on its timing or evolution. Their reluctance is understandable. Financial markets are notoriously fickle, especially in their attitude toward government budgets. In January 1980, markets threatened to
become unstable when President Carter recommended a $16 billion deficit in his 1981 budget. He was forced to withdraw that budget and propose a new one that showed balance.1 Three years later, markets accepted President Reagan's actual 1983 deficit of $208 billion with relative equanimity.
As this brief is written in June 2004, financial markets are relatively calm, even though the 2004 deficit will near 4 percent of the gross domestic product (GDP) and Congress has recently added a prescription drug program to Medicare that significantly worsens the long-run problem. The lack of worry on Wall Street is buttressed by Standard and Poors (S&P) and other rating services that give the U.S. debt the highest possible credit rating. In discussing the United States, S&P notes the need to reform Social Security and Medicare, but their analysis suggests that flexible financial and labor markets and stable, transparent political institutions provide the U.S. economy with significant advantages.2
A key question is whether our political system promotes reform. Winston Churchill once remarked that democracy is the worst form of government except for all the others. Democracy is clearly at its weakest in dealing with powerful interest groups. By some measures, the elderly are our nation's most powerful interest group.3
The experience of other Western democracies is not reassuring. Many European countries that face worse demographic pressures than ours have undertaken only the mildest reforms or none at all. In exceptional cases, such as Italy, Sweden, and the United Kingdom, severe budget and economic pressures have provoked significant reforms. A similar crisis may be necessary to provoke reform in the United States. Even without a crisis, Congress did undertake a long series of painful, deficit-reducing incremental reforms in the 1980s and early 1990s. In theory, Social Security and Medicare could also be reformed incrementally. But no one is pursuing this course and soon it will be too late for incrementalism.
The Nature of the Crisis
The severity of crises varies significantly. A financial meltdown can cause a recession, but even the Great Depression did not greatly increase the risk that the United States would out-and-out default on its sovereign debt. Indeed, explicit governmental defaults are quite unusual. A greater danger is that governments, trying to avoid recession, will respond to financial crises by inserting excessive amounts of liquidity into the economy. Accelerating inflation will then cause the real value of the debt to erode. There would, in other words, be an implicit default. This type of default can happen completely by accident.
We saw inflation accelerate in the late 1980s when the Federal Reserve injected large amounts of liquidity into the economy after the stock market fell about one-third in October 1987. Many argue that the Federal Reserve overdid it and inadvertently caused the acceleration of inflation that lasted until 1990. The Federal Reserve also responded aggressively to the bursting of the stock market bubble in 2001-02 and the associated recession by lowering nominal, short-term interest rates to levels not seen for almost 50 years. It is yet to be seen whether the Federal Reserve is overdoing it again.
Even if an acceleration of inflation occurs by accident, rising prices can be a highly convenient response to a soaring debt-GDP ratio. By increasing the nominal dollar value of the denominator more rapidly, inflation reduces the growth in that ratio and acts as a tax on assets denominated in dollar terms. The implied reduction in purchasing power releases resources for government use. Creating money is also highly profitable for the government because the costs are trivial.
Because inflation can be so convenient, a strong temptation exists to inflate purposely as the public debt rises rapidly. However, any hint that a government is about to do this would cause a severe negative reaction on bond markets.
In any case, inflation provides only a temporary respite. Nominal interest rates quickly rise as inflation rises. If inflation becomes severe, long-term rates are likely to rise so much that the government would find it unattractive to sell long-term bonds. Even investors in inflation-indexed bonds would demand high real rates because of uncertainties caused by lags in indexing. Thus, inflation's beneficial impact in lowering the debt-GDP ratio is soon offset by a soaring interest bill on new debt issues.
Although past unanticipated inflations were equivalent to a default of some U.S. debt, these inflationary periods over the last 20 years never became severe. Perhaps financial markets take a relatively sanguine view of future budget pressures because they expect a crisis severe enough to provoke reform, but not severe enough to greatly damage the real value of the debt.
When politicians in democratic countries lack the will to take politically difficult yet necessary steps, they often seize on a crisis as an excuse for doing the right thing, even when the crisis and the policy problem are unrelated. The last major U.S. Social Security system reform occurred in 1983 when the Social Security trust funds emptied. Apolitical rather than an economic crisis ensued. The elderly feared losing all benefits, even though trust fund outlays only slightly exceeded trust fund revenues.
When Congress acted, they did far more than resolve the short-run crisis. The most important change in the system—a gradual increase in the normal retirement age—did not affect anyone over 45 in 1983 and had no immediate effect on the trust fund. Swedish Social Security reform was provoked by a serious budget and economic crisis. However, the reform took a long time to design and implement—longer than the duration of the crisis. Nevertheless, the Swedes persevered with a reform other countries are now emulating.
The problem with waiting for a mild crisis is that a serious one might intervene. This possibility becomes more likely every day that reform is delayed. If current policy is not changed, the amount of debt created over the next 40 years will make the debt issued to finance World War II seem relatively small. The worst of all possible worlds will occur if a crisis does not provoke reform until the economy suffers lasting harm from runaway inflation or a very severe recession. When reform eventually comes—as it must—there will be a major social cost as well because Social Security and Medicare benefits will have to change abruptly without giving current beneficiaries and those near retirement much time to adjust.
The Timing of the Crisis
A crisis could occur tomorrow because of a plummeting dollar or for some totally unexpected reason. Or a crisis could be delayed for 30 years. Long-run budget projections suggest, however, that without reform, the budget situation will deteriorate very rapidly in the 2020s. A crisis could not be delayed much beyond the end of that decade.
The Congressional Budget Office (CBO) recently produced six long-run budget scenarios and the General Accounting Office (GAO) produced two. This brief focuses on the three most pessimistic projections, although these scenarios are not especially pessimistic. It is easy to imagine much worse outcomes. The other five scenarios assume either that health care will be reformed or that tax burdens will increase at very high levels compared with those of the past.4
The worst deficit scenario comes from GAO. In it, all expiring tax cuts from 2001, 2002, and 2003 legislation are continued through 2014, after which revenues are held constant at 17.7 percent of GDP. Discretionary spending is held constant relative to GDP and long-run Social Security and Medicare outlays are based on the intermediate assumptions outlined in the 2004 trustees report. The Medicare trustees assume that the rate of growth of outlays per beneficiary decelerates in the
long run until it is 1 percent higher than the rate of per capita GDP growth.
The implications of this scenario are truly frightening. The national debt surpasses 100 percent of the GDP in 2022, and by then, the debt-GDP ratio is growing by 7 to 8 percentage points per year. The ratio passes 150 percent in 2028 and 200 percent in 2032 (figure 1).
The worst CBO scenario is not quite as bad, but still alarming. It is more pessimistic than GAO in its per beneficiary health spending, which is assumed to grow 2.5 percent faster a year than per capita GDP growth. However, CBO is more optimistic in its defense outlays, which are assumed to decline relative to GDP in the long run. Also, CBO's revenue projections are higher in the long run at 18.4 percent of GDP—the average of the previous 30 years.
In this scenario, the debt-GDP ratio does not exceed 100 percent until 2027, but, as in the GAO scenario, the ratio is rising very rapidly at that time. The deficit exceeds 10 percent of GDP in 2027 and interest costs near 20 percent of total spending.
In the second most pessimistic CBO scenario, the rate of health expenditure growth is similar to that in GAO's scenario and defense outlays fall even more rapidly relative to GDP. The debt-GDP ratio does not pass 100 percent until 2037 but is rising rapidly in the 2020s. It passes 50 percent in 2024 and is 60 percent only four years later.
If the world unfolds in line with any of these scenarios, an interesting question arises about exactly when financial markets become alarmed. Do they have to observe the debt-GDP ratio rising at unsustainable rates? Or will they be impressed when the CBO's 10-year baseline projections begin to
show an exploding debt? If the latter, financial instability could emerge in the next decade.
But to believe the latter scenario, it is necessary to ask why financial markets should put so much more confidence in 10-year baseline projections than in the 30-year projections discussed above. The 30-year projections are, in fact, more reliable because the analysts projecting 10-year baselines are forced to work with unrealistic policy assumptions. These analysts must assume that discretionary spending is constant in real terms and that Congress does not correct for surprises on the revenue or spending side of the budget. The policy assumptions used in the three scenarios
discussed above seem much more realistic; Congress adjusts to budget surprises, which is more in line with historical experience.
So it remains a puzzle why financial market players are not pushing reform more vigorously. It may be because life is too comfortable given unusually low inflation and interest rates. Even unemployment is low compared with the average of the previous 30 years. But life could become very uncomfortable very quickly. That is the trouble with financial markets. They never tell you what they will do next.
Notes
1. A weakening economy helped cause the actual 1981 deficit to grow to $79 billion.
2. As reported by Philippe Sachs and John Chambers, "United States of America," http://www.standardandpoors.com, October 10, 2003.
3. The largest organization representing the elderly is the AARP with 35 million members. Compare this number to the 4.2 million members of the National Rifle Association, another politically potent group.
4. For more information, see Congressional Budget Office, The Long-Term Budget Outlook, December 2003 and the General Accounting Office web site, http://www.gao.gov/cghome.htm.
About the Author
Rudolph G. Penner is a senior fellow at the Urban Institute, where he holds the Arjay and Frances Miller chair in public policy.
Eight Budget Scenarios
Two General Accounting Office (GAO) simulations are discussed here, along with six scenarios produced by the Congressional Budget Office (CBO).
The most pessimistic GAO simulation assumes that discretionary spending remains constant relative to GDP and that Social Security and Medicare outlays follow paths consistent with the intermediate projections of the Social Security and Medicare Trustees. Medicaid outlays follow CBO's "intermediate" assumptions. All expiring tax cuts are assumed to continue and revenues level off after 2014 at 17.7 percent of GDP. The results of this scenario are discussed in the main text.
It may be overly pessimistic to assume that discretionary spending remains constant relative to GDP, because defense has been on a long downward trend relative to the size of the economy. On the other hand, the trustees' intermediate assumption regarding Medicare growth may be overly optimistic and more important. The amount by which the rate of growth of spending per beneficiary exceeds per capita GDP growth is assumed to decline to 1 percent a year in the long run—an excess growth rate lower than Medicaid has experienced through most of its history.
GAO's "baseline extended" scenario is much more optimistic. Discretionary spending is held constant in real terms for 10 years and then grows with the economy. All temporary tax cuts are allowed to expire and revenues are then held constant relative to GDP at their implied 2014 level of 20.1 percent, or almost 10 percent above the average of the past 30 years (18.4%). Social Security, Medicare, and Medicaid spending paths are based on the same assumptions as in the previous scenario. In this scenario, the debt does not exceed 100 percent of GDP until 2038. However, the scenario seems overly optimistic in that neither political party wants to see all the temporary tax cuts expire and it would be extremely difficult politically to keep real discretionary spending constant for the next 10 years.
The most pessimistic CBO scenario has Medicare and Medicaid costs per beneficiary rising at a rate of 2.5 percent a year more than GDP per capita. This matches the historical average, although there have been signs of a deceleration in recent years. Defense costs follow the administration's 2004 Future Years Defense Program and then remain constant in real terms. As a result, defense outlays fall from 3.7 percent of GDP in 2010 to 2.8 percent in 2030. All temporary tax cuts are continued,
implying under CBO assumptions that revenues reach 18.4 percent in 2012. They are assumed to remain at that level. The results of this scenario are discussed in the main text. Although its assumptions regarding medical costs may be overly pessimistic, the assumption regarding defense does not seem to allow for the possibility that the United States may face new adversaries, such as China, in the long run.
In the second CBO scenario, the same revenue path is combined with a lower spending path. Health cost growth slows to the point that per-beneficiary costs exceed GDP per capita growth by only 1 percent. Defense spending slows even more rapidly and amounts to 3.1 percent of GDP in 2010 and 2.0 percent in 2030. The results of this scenario are discussed in the main text.
The third CBO scenario combines the same revenue and defense assumptions with the assumption that per-beneficiary health costs will grow at the same pace as per capita GDP. No serious budget problems emerge in this scenario, but the health cost assumption seems totally implausible unless there are major reforms in Medicare and Medicaid.
In CBO scenarios four, five, and six, the different spending assumptions of the previous three scenarios are combined with a "high" revenue path. These scenarios assume that all temporary tax cuts expire and that real growth is allowed to push taxpayers into higher tax brackets. The number affected by the alternative minimum tax soars until the total tax burden reaches 24.7 percent in 2050. This rate is much higher than GAO's high revenue path and assumes a 2050 tax burden one-third higher than the average of the last 30 years.
The results of scenario four are similar to those of scenario two. Scenarios five and six imply no budget problem in the long run, but seem hopelessly optimistic.
About the Series
National Budget Issues analyzes wide-ranging budgetary matters that cut across all Urban Institute research areas. This series reveals the tough choices that the nation must grapple with through the budget, ideally balancing the costs and benefits of government, while also considering society's needs and taxpayers' burdens. Urban Institute's budget researchers—senior fellows and experts—include former high-ranking officials at the Congressional Budget Office and the U.S. Department of the Treasury.
The author would like to thank the Smith Richardson Foundation for financial support and Leonard Burman, Paul Hewitt, James McTigue, and Paul Posner for useful comments.
Latest Research
Browse by Topic:
PDF
Printer-friendly summary
