The 19 Percent Solution

How to balance the budget without increasing taxes

Here's an appalling snapshot: During fiscal year 2010, which ended on September 30, the government spent around $3.6 trillion, or 25 percent of gross domestic product (GDP), while collecting $2.1 trillion in tax revenue, or 14.5 percent of GDP. The resulting deficit was $1.5 trillion. The total debt held by the public—the sum of all accumulated annual deficits and interest payments—reached 63 percent of GDP.

You have to go back to 1946, in the immediate aftermath of World War II, to find spending that was as large a percentage of GDP. You need to return to 1945 to find a deficit that big on a percentage basis as well. Just a few years ago, in 2007, the debt was 36.2 percent of GDP. If current trends continue, the Congressional Budget Office (CBO) projects, the number will reach 87 percent in 2020. Most economists talk about a debt equal to 60 percent of GDP as a trigger point where investors become very nervous about a country's ability to pay its obligations.

Given all this, it's good that all sorts of people—even the politicians responsible for the problem—are beginning to take spending seriously and even tentatively propose some action. The animating message of the unexpectedly successful Tea Party movement has been to stop the spending. In 2010 Rep. Paul Ryan (R-Wis.) unveiled his "Roadmap for America's Future," which seeks to eventually balance the budget by reducing entitlement spending. Weeks before the November midterm elections, the presumptive speaker of the House, Rep. John Boehner (R-Ohio) released the GOP's "Pledge to America," which declared that "the need for urgent action to repair our economy and reclaim our government for the people cannot be overstated." President Barack Obama convened a National Commission on Fiscal Responsibility and Reform that in December recommended ways to eliminate the primary spending gap—the difference between revenues and total government outlays minus interest payments on the federal debt—by 2015 and actually balance the full budget by 2020. Esquire magazine pulled former Sens. Bill Bradley (D-N.J.), John Danforth (R-Mo.), and Gary Hart (D-Colo.) out of retirement to gin up a package that would balance the budget by 2020. The liberal Center for American Progress, chaired by Clinton administration Chief of Staff John Podesta, has weighed in with a proposal of its own.

[article continues below video]

Unfortunately, the plans produced to date are severely lacking. The Tea Partiers' hearts may be in the right place, but the decentralized movement has yet to deliver any specifics. The GOP pledge is a laughable and already forgotten document, explicitly ruling out cuts in military spending (one of the biggest budget items) and implicitly taking other big-ticket items such as Medicare and Social Security cuts off the table as well. Under the best of circumstances, Paul Ryan's plan, pilloried by detractors as "radical," does not balance the budget until 2063. Radical, that ain't.

John Podesta believes cutting $255 billion—the primary spending gap—from a projected 2015 budget over $4 trillion would "eviscerate the middle class." So his group's plan features major tax hikes instead, coupled with relatively small spending cuts. The Esquire program opts for tax hikes too. The presidential debt commission's final report called for serious cuts to defense and other programs but increased gas taxes and raises the amount of salary subject to payroll taxes. More problematically, its plan depends on hiking federal revenue as a percentage of GDP to at least 21 percent, a level that has never been sustained at least since World War II. A majority of commission members endorsed the plan but not the requisite number needed to present its recommendations to Congress for action. You can almost feel the conventional wisdom hardening like an old man's arteries. "Conservatives must accept that tax increases need to be on the table to achieve fiscal stability," wrote Washington Post columnist Ruth Marcus in November. "Liberals must accept that promised entitlement benefits are not affordable and must be pared." This formulation may sound sensible, but any plan that relies on significantly raising average annual federal revenue is based on fantasy. Since 1950 annual federal revenue has averaged 17.8 percent of GDP, fluctuating within a relatively narrow range (see Figure 1). Despite endlessly creative attempts to squeeze more dollars out of taxpayers, the feds haven't been able to pull in much more than that on a regular basis. 

There's a better way to balance the budget, one that wouldn't raise revenue levels above historic averages or "eviscerate" anything other than bloated spending. To get there, though, we'll need to lay out the basic facts without flinching, especially when it comes to realistic projections of revenue.

Budget Basics

Giving Washington yet more money will grant it more power over the economy, but it won't guarantee anything resembling a balanced budget. That's because federal spending has little connection to revenue levels. The feds can spend more than the available revenue by literally printing money or by taking on massive levels of debt that future generations of Americans will have to pay off. 

The federal budget (see Figure 2) has two basic parts, each accounting for about half of total outlays. "Mandatory" spending includes entitlement programs, such as Medicare and student loans, whose funding is automatic under current law. Some major entitlement programs—most notably Social Security—are "off budget," meaning they are not included in the official spending numbers, making it that much more difficult for taxpayers to understand what's going on. Politicians like to call such outlays "mandatory" because it implies they have no control over that part of the budget, but they can always change the laws that govern the entitlement programs.

The second part of the budget is "discretionary" spending, which includes things such as homeland security, most military spending, and programs such as farm subsidies and aid to schools. Discretionary spending is what the president and Congress decide to spend each year through appropriations bills. Because items come up for annual votes, they are theoretically easier to increase or decrease than programs in the mandatory part of the budget.

In fiscal year 2010, the federal government spent 25 percent of GDP, well above the historical average. Since 1950 annual expenditures have averaged just under 20 percent of GDP. According to the CBO's "alternative scenario" budget projection—which assumes the most likely policy changes, including legislators' concessions to interest groups such as doctors and senior citizens—spending at its current trajectory will increase to 26 percent of GDP in 2020 and to 32 percent in 2030.

The costs of Medicare, Medicaid, and Social Security are an important force behind this growth (see Figure 3). According to the CBO's alternative scenario, the combined cost of these three programs was roughly 10 percent of GDP in 2010 but will reach 12.4 percent in 2020 and 15.7 percent in 2030.

The main driver of the growth, however, is interest spending—the bar tab for our binge. The CBO projects that in the next 70 years, public money spent on interest will grow from 1.4 percent of GDP (or $204 billion in 2010 dollars) to almost 41.4 percent of GDP (or $27.2 trillion in 2010 dollars). In the short term, the cost of our debt will reach 3.8 percent of GDP by 2020 and 7 percent of GDP by 2030. Today spending on interest represents about a third of the cost of Social Security; in 20 years it is expected to exceed the cost of that program.

The 19 Percent Solution

As noted, annual federal revenue since 1950 has averaged just under 18 percent of GDP. Sometimes the percentage is a bit lower, and sometimes it is a bit higher. In 2010, for instance, due to the weak economy, revenue was only 14.5 percent of GDP. Under Bill Clinton, revenues reached 19.9 percent, 19.8 percent, and 20.6 percent between 1998–2000, helping to produce surpluses. Such upticks are typically short-lived, and to the extent that they produce surpluses, political pressure inevitably builds to reduce revenue levels or return the money to taxpayers in one form or another—through rebates, tax cuts, or increased spending. It's worth remembering that both major-party candidates campaigned on tax cuts during the 2000 election.

Under the CBO's basic projection, which amounts to little more than running numbers supplied by politicians, federal revenue will rise to 20 percent of GDP in 2015 and then climb even higher, reaching 21 percent in 2020 and 22.3 percent in 2030. If history is any guide, such estimates are pure fantasy. Under the CBO's more realistic "alternative scenario," federal revenue will equal about 19 percent of GDP within a few years and then stay around that level. Even that projection is a bit of a stretch, to judge from the last 60 years, but it is far more plausible than the basic projection. For the purposes of this budgeting exercise, we are willing to give the alternative forecast the benefit of the doubt.

So as a starting point, we assume that federal revenue will be about 19 percent of GDP in any given year. To put that into plain numbers, the GDP in 2010 was about $14.6 trillion. If the government spent 19 percent of GDP, it would have spent no more than $2.8 trillion, as opposed to the $3.6 trillion it actually shelled out. (We use fiscal year 2010 figures because, as of this writing, no federal budget has been passed for fiscal year 2011.) Federal spending will vary from year to year, especially in response to such unpredictable events as natural disasters, recessions, and acts of war. But like a family that dips into savings or lives on credit cards during short downturns, the government can survive such swings from time to time.

In absolute dollars, getting to 19 percent immediately would mean cutting $829 billion out of the budget, which isn't a politically realistic target at the moment. Getting to 19 percent within a few years, though, would require a series of far smaller cuts because of the expected economic recovery. One of the tasks assigned to Obama's debt commission was finding ways to close the primary spending gap (the deficit excluding interest payments) in 2015. That would mean cutting about $243 billion from a budget projected to be $4.1 trillion.

That's a 5.9 percent reduction, the sort of cut that many businesses and households have managed relatively easily during the recession. In its "A Thousand Cuts" study, the Center for American Progress lays out (with reservations) a series of spending cuts that total $255 billion in current dollars. The trims include a 75 percent reduction in farm subsidies (saving $11 billion), a 12.5 percent cut in military spending (saving $96 billion), and $53 billion in terminated "tax expenditures"—spending done through the tax code such as the child tax credit or the home mortgage interest deduction.

The president's debt commission suggested $200 billion in discretionary cuts for 2015, including closing a third of America's overseas bases ($8.5 billion), freezing noncombat pay for three years at 2011 levels ($9.2 billion), freezing federal pay at nonmilitary agencies for three years at 2011 levels ($15 billion), and eliminating 250,000 nonmilitary contractors ($18 billion). They also suggested reductions in mandatory spending, including the use of a lower-cost "chained Consumer Price Index" to adjust entitlement spending and a plan to fix cost-of-living adjustments for military retirees. All told, they came up with about $56 billion in cuts to mandatory spending for 2015. It should be noted, though, that these cuts are only cuts in name, as the final Deficit Commission plan increases spending by $1.6 trillion in the next 10 years. (The CBO assumes $2 trillion in increases.)

In a related exercise, the Committee for a Responsible Federal Budget, a bipartisan nonprofit housed within the left-leaning New America Foundation, has called for balancing the budget in 2020, when revenue is expected to equal 19.3 percent of GDP under the CBO's alternative scenario. GDP that year is projected to be $19.5 trillion, which means federal outlays should not exceed $3.8 trillion in a balanced budget. To reach that goal would require cutting about $1.3 trillion from projected spending increases during the next decade, or about $129 billion annually out of budgets projected to average around $4.1 trillion. Due to variations in individual budgets and various effects of compounding, that works out to reductions of about 3.6 percent a year. To give you a clearer sense of what getting to 19 percent of GDP means, Figure 4 shows how much needs to be cut each year from six areas of the budget.

On its face, freezing spending at 2010 levels in nominal dollars would also create a balanced budget in 2020. But it would not prepare the country for the explosion in Medicare spending expected after 2020. Furthermore, using 2010 spending levels as a starting point means locking in the spending increases of the last decade. That's true too of using 2008 spending levels, which the GOP House leadership pushes for in its Pledge to America.

The spending increases since Bill Clinton left office have been gigantic. Under George W. Bush and Barack Obama, in constant 2010 dollars, military spending during the last decade increased by 75 percent, Medicare by 75 percent, and Social Security by 37 percent. Some of the increases reflect demographic changes: As the country ages, the number of retirees will increase, and the payouts for old-age entitlements will rise. But the last decade saw no demographic shifts that justify anything like the spikes in spending shown in Figures 5 and 6. More to the point: Precisely because the country is aging, we need to seriously reform entitlement spending lest it swamp every other aspect of government. In 2030, with the baby boom fully in its dotage, the U.S. Census Bureau projects that nearly 20 percent of the country will be 65 or older.

If we were going to peg future outlays to a budgetary baseline from the past, we would do far better to use the 2000 federal budget as a guide. The final year of the last century was a pre-recession year, and it is recent enough to dismiss any complaints that everyday life has changed dramatically since then. Few people were complaining at the turn of the century that our federal spending was at a dangerously low level that did not support the needs of the nation.

Figure 7 compares projected spending in 2020 under the CBO's alternative scenario to what 2020 levels would be if they were based on 2000 spending as a percentage of GDP. In 2000, for instance, military spending was $294 billion, 3 percent of that year's GDP. In 2020 GDP is expected to be $19.5 trillion, so military spending based on 2000 levels would equal 3 percent of $19.5 trillion, or $585 billion. 

The table illustrates the long-term price tag of the last decade's unrestrained government spending spree. In 2001, for instance, Medicaid spending came to $157 billion (in constant 2010 dollars). By 2010 it had increased to $275 billion. If nothing is done, it will cost $546 billion in 2020. If spending stayed the same as a percentage of GDP as it was in 2000, Medicaid would instead cost $234 billion.

Total spending in 2020 is projected to be more than $5 trillion under the CBO's alternative scenario, or 26 percent of GDP. There is simply no realistic way the federal government will be able to raise that kind of revenue. Since 1950 the government has collected revenue above 20 percent of GDP exactly once. That was in 2000 and the percentage was 20.6. Good luck getting to 26.

Debt and Entitlements

The twin specters of exponentially rising debt payments and exploding entitlement programs, especially Medicare, are intricately connected. Given the government's inability to generate revenue sufficient to cover current levels of discretionary spending, pay interest on the debt, and pay for the growing sums devoted to Medicare, Medicaid, and Social Security, it will have to borrow more and more, creating a seemingly hopeless debt spiral.

This is true even if interest rates stay at projected long-term levels. The CBO alternative scenario assumes that interest rates will stay at an average of 4.6 percent from now until 2084. The CBO estimates the rates for government debt will climb from about 2.3 percent now to just under 5 percent in 2030, then stay there for another 54 years. That's a low figure historically, and even though the CBO alternative scenario is understood to be a more realistic projection, this particular estimate is almost certainly optimistic. As our debt level increases, investors could start seeing the United States as a riskier investment and ask for an increase in interest rates to make up for the uncertainty involved in lending us even more money. Any increase in interest rates above the projected levels could make debt uncontrollable and push the country into bankruptcy.

When it comes to entitlements, we need some basic reality checks. Social Security is not a retirement plan as any of us know the term. Taxpayers do not control any aspect of their contributions, from the amounts they pay to the instruments in which their money is invested. The Supreme Court has ruled that Social Security and other entitlements do not impose contractual obligations on the government, meaning that benefits can be cut or abolished altogether. Social Security and Medicare are instead transfer programs that move money from one group of taxpayers to another, irrespective of demonstrated need. (Medicaid is at least means-tested.) Reform discussions should begin with the understanding that such programs function as a form of guaranteed income, not some sort of sacrosanct compact between generations.

Our unsustainable entitlements are a product of a very different America. Created in 1935, Social Security was a response to the Great Depression and widespread poverty among older Americans. Created three decades later, Medicare, which provides medical coverage for those 65 years and older, was likewise motivated by a rate of poverty among seniors that was nearly double that of the overall adult population. Today, by contrast, seniors comprise one of the wealthier segments of the U.S. population. They are much less likely than average to live in poverty, and they are far more likely to own a house and other assets. This makes sense: Seniors, even those who have never earned big salaries, have spent a lifetime working and amassing wealth. According to its administrators, Social Security payments represent about 40 percent of "the income of the elderly" and in 2010, the average payment was $1,170 a month. That sort of payout can be generated at a lower cost through workers investing their wages over their careers.

More important, as the population ages, entitlements will need to be sharply curtailed if the government expects to spend money on anyone with greater needs. It would make more sense to have a system in which individuals who are too poor or sick to take care of themselves would receive financial assistance, but everyone else would be expected to provide for their retirement and health care. Instead, we have a system largely unaffected by changes in income, wealth, and life span.

Despite big increases in life spans and later entry into the work force, full Social Security benefits still start at 65 and partial benefits can be tapped at 62. The only significant adjustment currently on the books is that beneficiaries born after 1960 will have to wait until they turn 67 to get full payouts. It's no wonder that the program will start running a permanent deficit in 2014.

Congress should cut benefits today for people who are 55 and younger. Those individuals still have plenty of time to adjust their expectations about future benefits and plan for retirement. We should gradually raise the initial age of eligibility to at least 70 and progressively increase it to track life expectancy (currently almost 79). We should also means-test these programs so that only those who really need the help get it. Such changes are relatively easy to implement and would allow lawmakers to pass reforms that won't kick in until years down the road.

According to the CBO, the changes to the eligibility age suggested above would gradually decrease Social Security spending by 6 percent through 2040. Means testing would reduce Social Security spending by an additional 6 percent. By 2040 the program would be spending $200 billion less annually than if the status quo remains unchanged.

Similar reforms are needed for Medicaid and Medicare. Funds for Medicaid, which provides health care for low-income Americans, could be transferred to state and local governments in the form of annual fixed block grants. Because they are closer to the actual beneficiaries, more knowledgeable about regional differences in the target population, and more responsive to local needs, state and local governments are better suited to take care of poor people.

When it comes to Medicare, Rep. Paul Ryan's roadmap is a useful place to start. Ryan suggests making Medicare a consumer-directed system by replacing the current program with tax credits and vouchers. In this scenario, instead of reimbursing doctors and hospitals, the government would provide payments directly to individuals, who would then purchase health insurance in private markets. The idea is to encourage competition among providers, increase consumer choice, and encourage cost-saving innovations.

More broadly, we need to rethink "mandatory" outlays altogether. Budgeting would be better served by creating multi-year spending commitments that sunset every five or 10 years. That would allow for planning—indeed, it would force it—on the part of the government and beneficiaries, but it would also allow for meaningful reform and even zeroing out on a regular basis.

Discretionary Spending

When it comes to discretionary spending, there are obvious cuts that any serious spending audit would demand. The Department of Education, created in 1980 under Jimmy Carter, has demonstrably failed to improve education results. Ronald Reagan talked about abolishing the department but did nothing to dismantle it. Under George W. Bush, federal funding for education increased by 50 percent in real dollars without anything to show for it other than higher debt levels. The department should be dumped altogether, along with farm subsidies and virtually all spending by the energy and commerce departments.

Then there is military spending, which at around $700 billion for fiscal year 2010 is one of the largest items in the federal budget. Pentagon spending has become sacrosanct among many conservatives who otherwise attack the government as incompetent and spendthrift. In an October 2010 op-ed in The Wall Street Journal, American Enterprise Institute President Arthur Brooks, Heritage Foundation President Ed Feulner, and Weekly Standard Editor Bill Kristol fretted that "our active-duty military is two-thirds its size in the 1980s," as if the ideal number of troops was set in stone during the last decade of the Cold War. Noting that growth in mandatory spending has outpaced spending on the Pentagon, the authors argued that we should opt for guns rather than butter, writing: "Congress can make a difference…by insisting that the Obama administration endorse responsible defense budgets instead of throwing our money down the well of entitlement expansion."

Brooks, Kristol, and Feulner are right to imply that tradeoffs exist in the budget, but they are wrong about how to address them. The question isn't whether entitlements or military spending should be cut. It's how much both can and should be cut. If the United States cannot afford an open-ended commitment to retirees regardless of need and income, it certainly cannot afford an open-ended commitment to Pentagon budgets hashed out when the Soviet Union controlled half of Europe.

[article continues below video]

Last July, Reps. Barney Frank (D-Mass.) and Ron Paul (R-Texas) wrote a letter signed by 55 other legislators that called for a thorough reappraisal of military spending, noting that the defense budget accounts for 56 percent of all discretionary spending. Within the next few years, defense spending will rise to almost $900 billion annually before tapering down again. Obama's debt commission lays out $100 billion in military cuts, a number that could easily be increased and even doubled through steps such as scrapping obsolete weapon programs and shifting the military to the lighter, smaller strategic model envisioned by Secretary of Defense Donald Rumsfeld before 9/11. While silent on the ongoing wars in Iraq and Afghanistan, the commission suggests that basic war spending be outlined by the president each year, a procedure very much at odds with the way George W. Bush used emergency supplemental spending bills to pay for conflicts that had been underway for years. 

Specific cuts can and should be debated, as the political class belatedly grapples with a crisis of its own making. But one idea that should be rejected early on is the notion that revenue shortfalls got us into this mess and that new taxes are needed to pull the country out of it. As calls for revenue increases come fast and furious, keep the number 19 in mind. Until federal spending is brought down to 19 percent of the economy or less—something that was accomplished with little trouble for the years 1997 through 2002, not to mention most of the period between 1950 and 1970—no serious solution will be possible. 

Nick Gillespie ([e-mail]) is editor in chief of reason.com and reason.tv. Contributing Editor Veronique de Rugy ([e-mail]) is a senior research fellow at the Mercatus Center at George Mason University.