It’s clearly a budget. It’s got a lot of numbers in it.
—George W. Bush, 20001
In August 2011, just weeks after the debt ceiling standoff ended with an agreement to cut government spending by more than $2 trillion over ten years, Hurricane Irene took aim at the East Coast of the United States. Tens of millions of people from Miami to Boston nervously watched forecast models that predicted hurricane-force winds sweeping across the North Carolina coastline and up the Atlantic seaboard before dumping torrential rains directly on New York City and New England. New York mayor Michael Bloomberg took the unprecedented steps of evacuating parts of lower Manhattan and shutting down the subway system. Even though the hurricane weakened slightly before striking New York, it was still one of the most expensive storms in U.S. history.2
How did people know where Hurricane Irene was going to strike, how strong it would be, and what kind of damage it could do? Hurricane warnings are issued by the National Weather Service (NWS), which gathers meteorological information across the country and around the world, including radar and satellite data. The National Hurricane Center (NHC) tracks all tropical storms that originate in the North Atlantic Ocean and develops computer models that predict how strong they will be and where they will turn. The NHC’s Storm Surge Unit forecasts how high sea levels will rise; local governments rely on these forecasts when deciding whether to evacuate low-lying regions. The “Hurricane Hunter” pilots who fly into hurricanes to gather data are from the Air Force Reserve and the National Oceanic and Atmospheric Administration. The data generated by the National Weather Service (NWS) are used (sometimes along with privately gathered data) to produce the forecasts trumpeted by private media outlets such as the Weather Channel, AccuWeather, and local news stations around the country. Finally, money to help communities recover from hurricanes and other natural disasters is routinely appropriated by Congress and distributed by the Federal Emergency Management Agency.
This is one of the things people expect the government to do: protect them from risks that are beyond their individual control. In 2005, Senator (and later presidential hopeful) Rick Santorum, an outspoken free market advocate, sponsored a bill that would have prevented the National Weather Service from providing the public with weather forecasts that might compete with private companies; but even he still insisted that the NWS should continue gathering data and providing it to those private companies.3 In other words, the government shouldn’t compete with the private sector, but should still fly planes into hurricanes so the private sector can make money. When House majority leader Eric Cantor said that emergency appropriations in the wake of Hurricane Irene should be offset by spending cuts, he was criticized by Republican governors Bob McDonnell of Virginia and Chris Christie of New Jersey, both members of the party’s conservative wing, for putting spending cuts ahead of emergency aid.4
Flying planes into hurricanes, developing weather forecast models, and everything else the federal government does costs money. But for most of the past half-century, the government has been spending more than it brings in. In 2010, the most recent year for which official figures are available, the federal government deficit was $1.3 trillion—the difference between $3.5 trillion of spending and $2.2 trillion of revenues.5 Spending, at almost 24 percent of GDP, was at the second-highest level (after 2009) since 1946; revenues, at less than 15 percent, were at the lowest level (tied with 2009) since 1950. The national debt—the total amount the government has borrowed to fill budget gaps over the years—was more than $10 trillion at the end of 2011 and growing rapidly. Dealing with the national debt will be one of our country’s major challenges for this decade. But before we can talk about what (if anything) to do about the national debt, we need to understand what those numbers mean.
Before digging into the numbers, there are a couple of important points to remember. The first is the difference between the annual budget deficit and the national debt. The deficit is the gap between the government’s spending and tax revenues in a given year, which it makes up by borrowing money. The debt is the total amount owed by the government at a specific moment, so it represents the accumulation of all previous years’ deficits.
Second, when making comparisons across time, it is important to measure both deficits and the debt against the overall size of the economy, which is typically represented by gross domestic product (GDP): the total value of all goods and services produced in a given year. Paying off the debt, or just making interest payments on the debt, means allocating some of our national income to that purpose. The more income we generate as a society, the easier it is to support a given amount of debt.
Today, most Americans think that the federal budget deficit is primarily due to “spending too much money on federal programs that are not needed or wasteful” and that the deficit should be reduced mainly through spending cuts—in the abstract.6 At the same time, however, majorities oppose spending cuts not just to Social Security and Medicare, but to national defense, homeland security, antipoverty programs, education, aid to farmers, and even the arts and sciences. The only thing that a majority believes should be cut is foreign aid.7 That’s because Americans typically think that foreign aid accounts for 25 percent of the federal budget; in fact, the correct figure is closer to 1 percent.8 So when people consider what the government actually does, it’s not clear that they still think it spends too much.
What does it mean to spend “too much money,” anyway? In the short term, the government does not have a hard budget limit: it can either borrow money (if the bond markets are receptive) or raise taxes (if it has the political will). Whether it spends too much really depends on what it is spending that money on.
In 2010, $196 billion went to pay interest on the national debt, which was $7.5 trillion at the end of 2009. Assuming that we do not want to default on our debt, those interest payments are the price that must be paid for deficits incurred in the past, regardless of current policy decisions. Subtracting off that $196 billion from total spending, primary (noninterest) spending was $3.3 trillion. The primary deficit, which measures the impact of current policies, excluding interest to pay for past deficits, was $1.1 trillion.
That $3.3 trillion in spending is divided into discretionary and mandatory spending.9 Discretionary spending is money allocated by Congress in specific appropriations bills that must be passed each year. This includes spending on defense; on most federal agencies such as the FBI, the Federal Aviation Administration, and the National Weather Service; and on various programs in areas such as veterans’ health care, education, and housing. Theoretically, in any year, Congress could decide to appropriate no money for, say, the Pentagon, in which case it would soon run out of money to pay soldiers, refuel airplanes, or buy new equipment. In practice, however, appropriations in one year are usually adjusted up or down from whatever the appropriations were in the preceding year. Mandatory spending, by contrast, does not require annual action by Congress.10 In this case, an existing law, such as the Social Security Act, requires the federal government to spend money unless that law is specifically amended or repealed. Most mandatory spending goes to programs such as Social Security, Medicare, Medicaid, and food stamps. The laws establishing these programs require that money be spent for all people who are entitled to benefits because they meet certain eligibility criteria (hence the common label of “entitlement programs”). This means that when more people qualify—for example, when a recession makes more people eligible for Medicaid—spending must go up.
Mandatory spending dominates the federal budget (see Figure 4-1). In 2010, it came to $1.9 trillion, or 59 percent of primary spending. The large majority of mandatory spending goes to Social Security ($701 billion), Medicare ($446 billion), and Medicaid ($273 billion), which together account for almost one-half of all primary spending. This is the outcome of the long-term trends discussed in chapter 3: as the population ages and health care costs increase, mandatory spending goes up, even if these programs are not becoming more generous. Since 1962, mandatory spending has grown from 28 percent to 59 percent of total primary spending;11 by 2021, under conservative assumptions, it will be over two-thirds of primary spending.12 This matters because if you want to balance the budget through spending cuts, mandatory spending is hard to cut. Procedurally, Congress has to take action to pass a law that reduces benefits, rather than simply appropriating less money for some program. More importantly, Social Security, Medicare, and Medicaid are popular programs. Most Americans currently benefit from or will someday benefit from Social Security and Medicare. A majority of Americans say that Medicaid is important to their families, which may seem surprising for a program dedicated to the poor—until you realize that it served 68 million people in 2010.13 This is in part because Medicaid, unlike Medicare, pays for long-term care (although beneficiaries must first exhaust all of their assets)—relieving many middle-class, working-age people from having to pay for their parents’ care. In other words, most federal spending is spending that people like, and that will grow year after year under current law.
Figure 4-1: Federal Spending, 2010 Billions of Dollars
While mandatory spending largely goes to pay for individual benefits such as retirement income and health care, most of what we typically think of as government activity—national defense, border security, aviation safety, the FBI, hurricane planes, disaster relief, highways and bridges, regulation, research and education, national parks, and so on—is paid for by discretionary spending. In 2010, more than half of all discretionary spending ($689 billion) went to national defense. Everything else, conventionally called nondefense discretionary spending, amounted to $658 billion. While mandatory spending has been steadily growing over the past half-century, discretionary spending has been falling from over 12 percent of GDP in the 1960s to just 7 percent in the decade before the financial crisis (see Figure 4-2).14 (It climbed to more than 9 percent of GDP in 2010 because of additional spending under the 2009 stimulus bill, but is falling again now that most stimulus programs have run out.) Most of this long-term decline has been due to major reductions in defense spending since the Cold War—from more than 10 percent of GDP in the 1950s (even after the Korean War) to 3 percent after the collapse of the Soviet Union.15 (Defense spending climbed again, to almost 5 percent of GDP in 2010, because of the Afghanistan and Iraq wars.) Nondefense discretionary spending has been relatively stable at around 4 percent of GDP over the past fifty years, growing slightly in the 1970s before falling back thanks to the cutbacks of the Reagan and Clinton eras. (Because it includes neither popular entitlement programs nor national security, nondefense discretionary spending is also the category that is most often singled out for budget cuts; the month before Hurricane Irene, the House Appropriations Committee voted to cut the Hurricane Hunters’ budget by 40 percent—a savings of $17 million.)16
Figure 4-2: Composition of Primary Federal Spending, 1962–2010
In other words, most of the federal government has been getting smaller, not larger, over the past half-century. Not counting Social Security (and Medicare, which didn’t exist yet), primary spending averaged 15 percent of GDP in the 1950s. In 2007, before the recession began, primary spending other than those two programs was just 11 percent of GDP. In other words, except for Social Security and Medicare, the government has been steadily shrinking. The government’s civilian workforce, which fluctuated between 0.9 percent and 1.1 percent of the population from 1954 to 1991, was only 0.7 percent of the population in 2010.17 Seen in this light, President Eisenhower, not President Obama, presided over a large and expensive federal government.
Social Security and Medicare, of course, are large and expensive. They represent a profound transformation in the role of government in American society. While the federal government spends less money and employs fewer people for everything else it does, it has become an important insurance provider to virtually the entire population.18 The government has always played an important role in protecting Americans from risks that they were unable to insure themselves against; Social Security and Medicare marked the expansion of that risk management function into the areas of retirement income security and health care for the elderly.19 Social Security, created in 1935 at the height of the New Deal, protects people from not saving enough for retirement, losing their savings to bad investments, outliving their money, and the premature death of a working spouse through its Old-Age and Survivors Insurance program; through Disability Insurance, it protects people from becoming unable to work. Medicare, created in 1965 at the height of the Great Society, protects people from becoming uninsurable due to poor health, from long-term health care inflation, and from needing to pay for expensive health care in old age.
These programs are the two largest examples of social insurance in the United States. They provide insurance because people pay for protection from outcomes that may or may not occur; we don’t know who among us will live to be one hundred, but Social Security protects all of us from completely running out of money before we die. They are social insurance because they pool risk across most or all of society; although largely funded by individual contributions that resemble insurance premiums, their financing and benefit structure partially reflect the idea that everyone should be able to participate and that the total burden should be shared fairly. With Social Security, you make mandatory contributions while you work; when you retire or become disabled, you receive benefits based on the amount that you contributed. With Medicare, you make mandatory contributions while you work and also pay insurance premiums in old age; in exchange, Medicare pays for most of your health care costs after you turn sixty-five.20 As of 2010, 54 million people received Social Security benefits and 47 million were enrolled in Medicare; 157 million people paid Social Security payroll taxes and (along with their spouses) could expect to receive benefits in the future.21
Unlike traditional government activities, social insurance programs pay concrete, individual benefits to just about everyone. This makes them the opposite of public goods like national defense and environmental protection, which we all benefit from and whose benefits cannot be divided up by individuals for themselves. Social Security and Medicare, by contrast, pay cash directly to you or to your health care providers. Because these programs pay individual benefits that people value—guaranteed retirement income or health care—they substitute for insurance that otherwise most of us would want to buy anyway. Without Social Security, people would want to save more for retirement, which means they would pay money to asset management firms instead of to the government. Without Medicare, people would try to buy health insurance from private insurers; but depending on their age and health status, they might not be able to buy insurance at any price. In other words, if we cut Medicare, our taxes will go down, but most of us will go right out and spend most of that money, or more, on health insurance—assuming that we can.
Over the past half-century, then, federal spending has gone up largely because the government’s insurance operations have grown. Those operations exist because, as our society has become richer, we value insurance more, and we expect to get it from the one entity large and creditworthy enough to provide it—the federal government. In recent decades, spending on social insurance programs has grown mainly because of increasing life expectancy and increasing health care costs, not because the programs themselves have become more generous. The 2003 expansion of Medicare to cover prescription drugs did increase Medicare spending. On the Social Security side, however, the biggest policy change in the past thirty years, the Social Security Amendments of 1983 (a compromise between President Reagan and Democratic House speaker Tip O’Neill), actually reduced spending growth, primarily by gradually raising the full retirement age and making some benefits taxable.22 (Social Security benefits generally are quite modest: in 2011, the average annual benefit was only about $13,000.)23
This means that the biggest factor behind the long-term growth of federal spending is not reckless, wasteful spending by politicians and bureaucrats in Washington—whether warmongers or environmental zealots, depending on your perspective—but demographic and economic trends that are largely beyond anyone’s control. We can argue about whether the government should be providing old age insurance, disability insurance, and health insurance (and we will, in chapter 6). Less insurance would mean a smaller federal budget. But to a large extent, it would simply transfer the burden of saving for retirement and paying for health insurance back from the government to individuals, making some people better off (mainly the rich, but also the heirs of people who die young) and some people worse off (certainly the poor and the long-lived, but arguably much of the middle class as well).24
The growth of Social Security and Medicare also means that it would be difficult if not impossible to balance the budget by cutting spending outside of these two popular programs—despite the widespread belief that unneeded or wasteful spending is the cause of the deficit. Many people think that there is too much government regulation and want to get rid of agencies like the Consumer Financial Protection Bureau; others think the government has no place in the cultural sphere and want to get rid of the National Endowment for the Arts. Again, we can argue about whether those agencies are worth their costs, but that’s not where the money is: the CFPB’s budget is about $300 million and the NEA’s is about $150 million—each less than one-hundredth of 1 percent of total federal spending.25 In 2010, by contrast, Social Security, Medicare, and interest on the debt cost over $1.3 trillion, while tax revenues were less than $2.2 trillion. Balancing the budget at that level of taxes would have required cutting all other federal spending—national defense, immigration control, federal courts and the federal prison system, Medicaid, food stamps, student loans, everything—by over 60 percent.
Social Security and Medicare are also the most visible examples of a major shift in the nature of government spending over the past half-century. In the 1950s, payments for individuals (such as Social Security checks or, today, Medicare reimbursements) made up barely one-fifth of total federal spending. Today, they account for more than three-fifths. Cutting federal spending means taking money directly out of people’s pockets; there just isn’t enough governmental waste and bureaucratic empire building to make a big difference. Any spending cuts that significantly reduce deficits over the long term would directly affect tens of millions of people, making them politically difficult to enact; the other option, tax increases, is at least as unpopular. But the idea that we can balance the budget solely by eliminating unspecified “bad” government programs is a fantasy.
Add it all up, and many Americans believe that federal spending is out of control. At over 23 percent of GDP, total spending in 2009 and 2010 was the highest since 1946, when the country was demobilizing after World War II. But by international standards, our government is relatively cheap. Over the past decade, total government spending in the United States, including state and local governments, was 37 percent of GDP—the second-lowest figure among the twenty-seven advanced economies (according to the International Monetary Fund) in Europe or North America.26 We may not get value for our money, but our government does not spend a lot compared to its peers. And in the long term, federal spending has not been growing—at least not yet. In the postwar period, spending peaked in the 1980s, when it averaged 22 percent of GDP, largely thanks to the Reagan defense buildup. In the 2000s, after the end of the Cold War but before the 2007–2009 recession, it averaged only 19 percent. The problem is that in the next few years, the retirement of the baby boom generation will trigger a long climb in mandatory spending that will be difficult to match with cutbacks in discretionary spending.
Higher spending obviously means that the budget can only be balanced through higher taxes. But to many people it means something more: that the government is too big. This is a core tenet of the tax revolt, which is focused not on balancing the budget, but on cutting government spending. Grover Norquist has been repeating this to anyone who will listen; in a 2011 interview, he said, “The way we could screw up the pending Republican [success] in 2012 . . . is to lose focus on spending and get distracted on chasing the deficit.”27 As government spending increases, the thinking goes, so does its influence over the private sector, hurting economic growth and reducing personal liberty.
This line of argument is based on a couple of big assumptions and a basic fallacy. The first assumption is that government actions invariably reduce liberty. But the opposite is often the case. As economist Bruce Bartlett pointed out, the countries with the lowest taxes and the smallest governments are often very poor precisely because their governments fail to protect property rights, enforce the rule of law, and provide basic public goods such as transportation infrastructure. Meanwhile Denmark, where total government spending is well over 50 percent of GDP, provides more economic freedom than the United States, according to the Heritage Foundation and The Wall Street Journal.28 The idea that government reduces liberty also depends on a peculiarly narrow conception of liberty. According to this viewpoint, the Americans with Disabilities Act infringes on the liberty of business owners, who now (in certain circumstances) must make accommodations for people with disabilities. But that same law enhances the liberty of disabled people, who now have greater opportunities to work, attend cultural or sporting events, and otherwise lead fulfilling lives. Medicare may force people to pay a payroll tax during their working careers, but it also assures them of a basic level of health care in retirement, with all the additional opportunities that depend on having decent health. As Franklin D. Roosevelt said in 1941, freedom from want is also a kind of freedom.29
The second assumption is that shifting responsibility from the government to the private sector is always a good thing. But spending cuts usually come with a price. If we eliminate funding for the Hurricane Hunters, someone will have to fly planes into hurricanes, unless we want to live with less hurricane data and more uncertainty about when and where they will strike. That someone could be private weather services like the Weather Channel and AccuWeather, but they can’t do it for free; instead, their costs would get absorbed by shareholders or passed on to customers. It is possible that private competition could eventually provide hurricane data with higher quality and at a lower cost than the government can, although that’s by no means certain. We can debate what services should be provided by the private sector rather than the government. But in any case, someone has to pay.
The same thing goes for insurance. In some areas, like auto insurance, the private sector seems to do a decent job of meeting people’s needs (although, even here, we rely on state governments to make sure high-risk drivers can get insurance). In other areas, however, private insurance markets often fail to provide coverage to many people who want or need it, at least in the absence of government encouragement. At the beginning of the twentieth century, for example, most workers had little or no insurance against workplace injuries. In response, most states adopted workers’ compensation laws that ensured benefits to injured employees; these laws were even supported by some employers, who preferred the cost certainty of workers’ compensation insurance to the risk of losing a lawsuit and having to pay a large damage award.30 Today the most glaring failure is the individual market for health insurance (where people buy insurance for themselves instead of getting it through their employers or a government program), which has left fifty million people without coverage.31 And for some risks, only the government is a credible insurer, because it is the only institution with the ability to absorb large, unexpected costs if necessary. (This is why many private insurance markets, including auto, home, life, and workers’ compensation insurance, are backstopped by state guaranty funds.) Throughout American history, both businesses and individuals have seen the government as the “ultimate risk manager,” in the words of historian David Moss, in part because it has the power to raise taxes, ensuring that it can make good on its promises.32 Shrinking government insurance programs is one way to reduce government spending, but either many of us will have to pay private insurers for the same insurance—or we won’t be able to buy it at all.
The fallacy is the idea that total government spending is an accurate measure of the size of government. The best definition of a big government is one that has a large influence on society: in economic terms, it causes major changes in people’s choices and in the allocation of resources. But the number of dollars collected and spent by the government doesn’t tell you how much it interferes in the private sector in any meaningful sense.33 To begin with, most government policies can be accomplished at least three different ways: spending, tax credits (provisions that let you reduce your taxes), and regulation. For example, let’s say politicians in Washington want to help poor people afford rental housing. Among other things, they can build and manage public housing projects; they can give tax credits to developers who build affordable housing; or they can write a regulation saying that a certain percentage of all new housing units must be rented at affordable rates. The first increases the amount of cash flowing in and out of the government; the second decreases it; and the third leaves it the same. Yet all increase the impact of government on society.
The discovery that government policies can be pursued through tax loopholes has led to the extraordinary proliferation of what are known as “tax expenditures”—laws that use the tax code as a means of distributing money to people.34 Tax expenditures allow individuals or companies to reduce their taxes if they do certain things that the law is trying to promote. For example, if you take out a loan to buy a house, you can take a tax deduction for the interest on your mortgage, which will save homeowners a total of $94 billion in 2012—ostensibly promoting homeownership.35 Alternatively, the government could have eliminated the tax deduction (increasing tax revenues by $94 billion) and instead mailed $94 billion in checks to homeowners, which would have had essentially the same effect for all parties involved. That, however, would count as higher taxes, which are politically unpopular. Tax expenditures allow politicians to pursue their policy objectives (or provide favors to important interest groups) while lowering taxes, which accounts for their tremendous popularity in Washington: today, no matter how you count them, they amount to over a trillion dollars in “spending.”36 Much of that money goes to the wealthy, both because they have more to deduct (bigger houses and bigger mortgages) and because they pay income tax at higher rates, so deductions are worth more to them.37
Like other subsidy programs, tax expenditures can have perverse effects on people’s economic incentives. The mortgage interest deduction encourages people to buy bigger houses and borrow more money, which increases household indebtedness, increases the risk of default and foreclosure, and makes the economy more vulnerable to the collapse of a housing bubble. Similarly, the fact that businesses can deduct interest payments on their debt but cannot deduct dividends paid to shareholders encourages them to take on more debt.38 Since tax expenditures influence the choices that people and companies make, they both reduce the amount of revenue flowing into the Treasury Department, increasing deficits, and increase the influence of the federal government on society, making government bigger. Yet tax expenditures only contribute to many Americans’ confusion about just what it is that government does. People who benefit from traditional programs run by government agencies, such as unemployment insurance or Social Security, are more likely to feel that government has helped to provide them with opportunities. By contrast, people who benefit from government subsidies in the form of tax breaks, such as the mortgage interest deduction or tax-preferred educational savings accounts, are actually less likely to believe that the government has increased their opportunities.39
Even ignoring tax expenditures, the total cost of a program has little to do with its real size. For example, which has a bigger impact on society, Social Security or the Environmental Protection Agency? Social Security cost $701 billion in 2010; the EPA, only $11 billion. But the EPA writes regulations that affect virtually every major business in the country, forcing them to incur costs not counted in the federal budget, with the goal of protecting all Americans from harmful toxins in the air they breathe and the water they drink. Social Security shifts income from your working years to your retirement years—perhaps more income than you would shift otherwise—and insures you against outliving your savings. There’s no meaningful way to say that one has more of an impact on the country than the other, even though Social Security “costs” more than sixty times as much as the EPA.
In the end, we can debate whether the EPA should regulate business more or less, and we can debate whether Social Security should shift more or less income between our working years and our retirement years. But if there is an optimal balance between private sector activity and government intervention, it cannot be fully measured in dollars. When it comes to social insurance programs, the real question is how much risk we want to shoulder ourselves and how much we want to pool with our fellow citizens, with the government acting as the administrator. The total dollar level of government spending is a red herring.
The more money the federal government spends, the more money it has to come up with, one way or another. In 2010, the federal government collected $2.2 trillion in revenues, the vast majority of that from taxes. The much maligned individual income tax (for which returns are due on April 15) was the largest source of government revenues, accounting for $899 billion, or 42 percent of the total. Each household has to pay a percentage of its income, ranging from 0 percent for the first few thousand dollars it earns to 35 percent for income that exceeds several hundred thousand dollars per year.a (Because of various exemptions, deductions, and tax credits—as well as basic poverty—almost half of all households pay no federal income tax at all.)40 This structure means that the individual income tax is progressive: rich people generally pay a higher percentage of their income than poor people.
Today, 40 percent of Americans say they are “angry” about the level of federal income taxes they pay.41 (To be fair, that was in a poll taken in early April 2011, and other polls consistently show support for some tax increases, especially to reduce budget deficits.)42 Yet individual income taxes have only fallen over the past thirty years, from 8.5 percent of GDP in the 1980s and 8.4 percent in the 1990s to 7.4 percent since the 2001 tax cut and 6.2 percent in 2010—the lowest level since 1950.43 Even if we ignore the recent recession (which lowered income taxes, since people have been making less money), individual income taxes were significantly lower as a share of the economy during the 2001–2007 economic expansion than during the 1991–2000 expansion, largely thanks to the Bush tax cuts.44
As individual income taxes have been falling, the government has come to depend more on social insurance contributions, which brought in $865 billion in 2010, or 40 percent of total revenues. The vast majority of this money comes from the dedicated payroll taxes for Social Security and Medicare, which are levied only on income from work, not income from investments.45 Under current law, 12.4 percent of each person’s wages goes to the Social Security trust funds.b That money is used to pay benefits currently owed to retirees and disabled people; in previous years, when payroll taxes exceeded benefit payments, the surpluses were invested in Treasury bonds, meaning that they were lent to the rest of the federal government. Another 2.9 percent of wage income goes to another trust fund that pays for the Medicare Hospital Insurance program (Part A). Medicare’s Medical Insurance and Prescription Drug Coverage programs (Parts B and D), however, are not funded by the payroll tax; instead, they are paid for by beneficiaries’ premiums and copayments and by money from general government revenues (that is, other taxes). Together, the payroll tax, premiums, and copayments barely cover half of Medicare’s total expenses, which means that the program is heavily subsidized by the rest of the federal government.
The Social Security tax is only collected on each person’s wage income up to a cap, which was $106,800 in 2011 (and is indexed to inflation), while the Medicare payroll tax is collected on all wage income.46 This means that payroll taxes on the whole are regressive: poor and middle-income people pay a higher percentage of their income than rich people.c As social insurance contributions have grown (due to rising tax rates), from 2 percent of GDP in the 1950s to more than 6 percent in the first decade of the 2000s (see Figure 4-3), an increasing share of government revenues has come from regressive rather than progressive taxes, shifting the overall tax burden onto lower-income people.47
The third major source of federal revenues is the corporate income tax, which in 2010 brought in $191 billion, or 9 percent of total revenues. Businesses routinely complain that the United States has one of the highest corporate tax rates in the world, with a federal tax rate of 35 percent; including state corporate taxes, we have the second-highest tax rate among all advanced economies.48 But the effective tax rates that U.S. corporations actually pay have been falling for decades as powerful business interests successfully lobby for tax loopholes and companies become more aggressive at taking advantage of those loopholes.49 Figuring out ways to book profits in overseas subsidiaries where corporate tax rates are lower has become a lucrative pastime for many companies from General Electric to Google, which have claimed billions of dollars in tax benefits.50 While the corporate tax amounted to 4.8 percent of GDP in the 1950s and 3.8 percent in the 1960s, it fell to 1.9 percent by the first decade of the 2000s and only 1.3 percent in 2010.51 Including state taxes, corporate taxes in the United States are among the lowest as a share of GDP in the industrialized world.52
Tax loopholes not only reduce government revenues but also lead large companies to expend time and money on activities that serve no purpose other than reducing their taxes. They also put smaller companies, unable to afford expensive lawyers and accountants, at a competitive disadvantage. Our low effective corporate tax rates have led to calls from liberals for corporations to pay higher taxes, but it is important to remember that companies are not real people. If they did pay higher taxes, it is hard to identify how that burden would be spread across employees (as lower wages), shareholders (as lower profits), or other capital owners (as lower rates of return).53
In addition to individual income taxes, social insurance taxes, and corporate income taxes, the federal government brought in another $208 billion in 2010 from sources such as alcohol, tobacco, and gasoline taxes, customs duties, and estate taxes. From any perspective, the total revenues of $2.2 trillion (less than 15 percent of GDP) were remarkably low. Not since 1950 were total federal taxes such a small part of the economy as in 2009–2010. Leaving aside the recent recession, federal taxes averaged 17 percent of GDP during the 2001–2007 economic expansion; the last time taxes were so low during a period of growth was in 1958–1960.54 Including federal, state, and local taxes, the total tax burden in the United States was 24 percent of GDP in 2009, the second-lowest level among the thirty-four industrialized countries in the Organisation for Economic Co-operation and Development (OECD).55 Compared to other rich countries, we are not an overtaxed nation.
In total, in 2010 the federal government spent $3.5 trillion and brought in $2.2 trillion, producing a $1.3 trillion deficit. Our recent deficits have been big by any measure: the largest ever in dollar terms, the largest peacetime deficits ever as a share of the economy, and among the largest in the industrialized world.56 Repeated trillion-dollar deficits have also been a major factor (along with more urgent sovereign debt crises in Europe) in shifting political debate in Washington away from unemployment and economic growth and toward the deficit and spending cuts. But are those deficits really worth all the attention?
Large deficits clearly increase the national debt. If tax revenues are not enough to cover spending, the federal government has to make up the difference by borrowing money. If a country has a history of not paying back its debts, it may not be able to find anyone willing to lend it money, which has been a big problem for many national governments throughout history. But this has not been a problem for the United States, thanks to a relatively strong record of fiscal responsibility dating back to Hamilton and Gallatin and, more recently, the dollar’s status as the world’s reserve currency. Throughout recent history, the federal government has always been able to borrow money by selling Treasury bonds. Every year that there is a deficit, the government has to borrow more money, adding to the national debt; the larger the deficit, the larger the increase in the debt.57
If the Treasury Department never had to pay interest (and could always borrow as much as it needed), the national debt would not matter very much. Like most borrowers, however, the federal government usually has to pay interest on the money it borrows, even if it pays less interest than just about anyone else. This is a major reason why the national debt matters. In 2010, interest rates were historically low—partly because the stagnant economy meant that few households and businesses were borrowing money and partly because nervous investors around the world wanted the safety of Treasury bonds. The federal government had to pay only $196 billion in interest as the national debt grew from $7.5 trillion (at the beginning of 2010) to $9.0 trillion (at its end), an average interest rate of about 2.4 percent. But under ordinary economic conditions, interest rates are closer to 5 percent.58 At that rate, each additional trillion dollars in borrowing means an additional $50 billion in interest payments every year, indefinitely. That is not necessarily a bad thing: as we discuss in the next chapter, if the money is invested in something that generates more than $50 billion of value to society each year, then the borrowing makes sense—just as it makes sense for a business to borrow money to make profitable investments. If instead the money is simply distributed to people as transfer payments, it is harder to justify as an investment, but it can still make sense on other economic grounds. In any case, however, a larger national debt means more money that must be set aside from each year’s tax revenues to pay interest on the debt, leaving less for current spending obligations and new investment priorities. By the end of the decade, interest payments alone will probably consume close to 3 percent of GDP.59
The impact of a trillion-dollar deficit on the national debt is real, and will continue to be felt for years to come. Beyond that, however, the large annual deficit numbers we have seen recently can be misleading. In fact, they both overstate and understate America’s real debt problem.
The recent deficit numbers overstate the debt problem because they have been temporarily inflated by the financial crisis, the resulting recession, and the government’s responses. Deficits automatically go up when the economy is weak. A recession means that individuals and companies are making less money, so they pay less in taxes; it also means that more people qualify for safety net and social insurance programs such as Medicaid, food stamps, and unemployment insurance, so government spending goes up. Under these conditions, the government should let the deficit go up instead of trying to balance the budget. The Hoover administration’s efforts to balance the budget no matter what (which was standard operating procedure at the time) were one of the factors that prolonged and deepened the Great Depression, and the Roosevelt administration’s attempt to balance the budget in 1937 also triggered a severe recession.60 In addition, many (though not all) economists think that the government should take additional steps to boost the economy during a recession; whether spending or tax cuts, these policies will increase the deficit further. Conversely, when the economy recovers, tax revenues go back up, spending on the safety net falls, and stimulus programs end, reducing deficits.
One way to measure our fundamental deficit problem is to look at the structural deficit—what the deficit would be if the economy were operating at full capacity. In 2010, over one-quarter of the total annual deficit was due solely to the automatic drop in revenues caused by the economic crisis and the automatic increase in spending that helped mitigate the crisis.61 More than one-third was due to the 2009 stimulus bill, which has since largely faded away.62 Without the financial crisis, in other words, the deficit would have been only $490 billion, or 3.4 percent of GDP: not pretty, but not a front-page national emergency, either. This is not to say that trillion-dollar deficits can simply be ignored. For one thing, all that money does get added to the national debt, which means that we will have to pay interest on it indefinitely. But the recent record deficits in themselves do not necessarily indicate a fundamental problem with current policies.
At the same time, the current deficit numbers understate the future debt problem. The biggest problem with focusing on the annual deficit is that it ignores the future—even when we have a decent idea about what will happen. We already saw this problem at the end of the 1990s, when the annual budget went into surplus even as the government’s long-term financial health was continuing to deteriorate because of demographics and rising health care costs. Today we face the same problem—only we have gotten older and health care has been getting more expensive for more than a decade. We need a long-term picture that takes into account future spending and tax revenues, insofar as they can be predicted. And this picture looks considerably worse than it did a decade ago.63
A few words of caution are in order before entering the realm of budgetary projections. Like most analysts (and politicians), we begin with the projections issued by the Congressional Budget Office (CBO). The CBO periodically issues “baseline” projections, which forecast spending and revenues in detail over the next ten years, as well as long-term projections, which go out twenty-five years or more (with less detail). Since no one can see the future, these projections will always be wrong, for two sets of reasons.
The first is economic uncertainty. Any budget projection depends on a forecast of how the economy will perform. Over the long term, economic growth depends largely on productivity—what amount of goods and services can be produced with a given amount of labor and capital investment. If the economy does worse than expected, deficits will be bigger; if productivity growth takes off and the economy booms, deficits will be smaller.
The second is political uncertainty. Forecasts require assumptions about how policy will change (or not) in the future, which are likely to be wrong. For example, it is common to assume that the Social Security payroll tax and Social Security benefits will remain the same indefinitely, even though that leads to an apparent contradiction: at some point, the Social Security trust funds will not have enough money to pay benefits in full. This is why long-term projections sometimes produce what seem like nonsensical results. In that case, they are less predictions of what will happen than an indication that current policies are unsustainable. In addition, the CBO baseline forecast generally must assume that current law will not be changed—for example, that tax cuts will expire when scheduled, even if political factors make that unlikely.64 For this reason, it is common practice to adjust the official CBO baseline when attempting to predict what is most likely to occur.
With those warnings, we are ready to peer into the future. And over the next ten years, despite the rhetoric swirling around Washington and on the campaign trail, we do not see a deficit crisis. Figure 4-4 shows one possible path for the federal government’s primary balance (excluding interest on the debt) and the total deficit (including interest).d From 2017 through 2021, the federal budget shows a primary surplus between 1 percent and 2 percent of GDP. Because the national debt has already grown so large, the government has to pay over 2 percent of GDP in interest each year, producing deficits around 1 percent of GDP.65 This is not an ideal situation, since our accumulated debts will cost us hundreds of billions of dollars per year in interest payments. In addition, discretionary spending will fall to its lowest level since the Hoover administration because of the spending caps agreed to in the 2011 debt ceiling compromise. But an annual deficit of 1 percent of GDP would be well below the average for the fifty years before the financial crisis and would be easily sustainable given even mediocre economic growth.66
This relatively rosy picture comes with one big caveat: it assumes that the income and estate tax cuts of 2001 and 2003 are allowed to expire on schedule, as dictated by current law.67 If, instead, these tax cuts are made permanent, we will be in the world shown in Figure 4-5: a primary deficit indefinitely, interest costs over 3 percent of GDP, and deficits close to 4 percent of GDP and rising.68 In other words, over the next ten years, we do not have a deficit crisis so much as a tax cut crisis. And if you are worried about deficits over the next ten years, you should tell your representatives to let the Bush tax cuts expire. (Alternatively, if you think the tax cuts should not expire because that would increase unemployment, then what we really have is an unemployment crisis.)
In the short term, then, the deficit picture is better than what reading the headlines would lead you to believe. In the long term, though, the picture becomes murkier. On the one hand, the further out a forecast tries to go, the more likely it is to be wrong. On the other hand, the fundamental forces behind our long-term budget problems are simple. Our population is aging, turning more and more Social Security and Medicare contributors into beneficiaries. The increase in Medicare enrollees means that the government is responsible for a growing share of national health care spending. This is a problem for two reasons: first, we already spend much more on health care than any other comparable country; and second, health care costs are growing faster than the economy, meaning that spending on government health programs will grow faster than tax revenues.
Figure 4-5: Deficit Projections: Income and Estate Tax Cuts Do Not Expire
The United States pays almost twice as much per person on health care as the typical industrialized country, yet our actual outcomes are mediocre at best; for example, our life expectancy is three years worse than you would expect given our general prosperity.69 There are many explanations for why our health care is so expensive and not particularly good. One is that our insurance system gives providers (doctors, hospitals, and so on) the incentive to order too many tests and conduct too many procedures, since they will be reimbursed for them whether or not they are needed or effective.70 This system encourages doctors to become specialists, who provide more expensive care (and make more money), rather than primary care physicians, who are relatively less expensive and are crucial to integrating care effectively. Another, related explanation is that since individuals bear only a small part of the costs of their treatment,e they tend to “overconsume” health care; higher cost sharing (deductibles and copayments) could encourage people to use fewer unnecessary services.71 One story is that major hospital chains dominate local markets, giving them the power to boost the prices they charge to health plans; a contradictory story holds that major insurance companies use their market power to boost premiums and pocket the profits.72 In any case, fragmentation among insurers leads to high administrative costs compared to countries with more consolidated payment systems, both because many insurance plans lack economies of scale and because health care providers have to deal with a large number of complex plans. According to one study, in 1999, administrative costs came to more than $1,000 per person, more than three times the level in Canada; while there may be several reasons for that difference, billing- and insurance-related expenses account for more than 20 percent of health care spending that is paid for through private insurers.73 High income inequality in the United States means that we have to pay our doctors (at least specialists) more than in other countries (because they would otherwise become hedge fund managers).74 Another argument is that large medical malpractice awards drive up malpractice insurance costs (which get bundled into the price of health care) and motivate doctors to engage in “defensive medicine”—ordering unnecessary tests and conducting unnecessary procedures to shield themselves from lawsuits.
Whatever the reason, health care in the United States is expensive, and the aging of the population means that the federal government will have to buy more and more of it. This problem is compounded by the fact that health care spending has been growing faster than inflation and faster than the economy for decades, even after accounting for demographic shifts, and there is little reason for it to slow down. The most common explanation of rising health care costs is increasing spending on medical technology: medical innovation generates new procedures, devices, and drugs that can help patients, but at an ever-increasing price.75 Rising incomes have also contributed to higher spending by increasing demand for health care.76 Again, it’s not clear that our increasing spending on health care is buying us very much in the way of health. From 1990 to 2007, life expectancy at birth increased from 75.3 to 77.9 years; this gain of 2.6 years was the second-lowest among the thirty-four countries in the OECD (even though we were only in the middle of the pack in 1990).77
In the future, average incomes are likely to continue rising (although perhaps not median incomes), and there is no reason to believe the pace of medical innovation will slow. The Affordable Care Act of 2010 (the Obama health care reform bill) does contain several measures designed to reduce the growth rate of government health care spending, ranging from formulas that limit the growth of Medicare payment rates to pilot programs trying out new ways of paying for health care. There is considerable debate about how effective these provisions will be (and debate about whether the entire thing should be repealed), but in any case health care costs are still likely to grow faster than the economy in the long term.78
As health care becomes a larger and larger part of the economy and of federal government spending, it will become increasingly difficult to balance the budget. In 2011, federal spending on health care, including Medicare, Medicaid, and the Children’s Health Insurance Program, came to 5.6 percent of GDP; by 2035, it is likely to be around 10 percent of GDP, more than half of that because of Medicare.79 It is true that health care costs cannot grow faster than the economy forever, but in the absence of structural change in the health care industry, there is also little reason to expect their growth to slow down dramatically.80 In addition, half of the projected growth in government health care spending over that period is due simply to the aging of the population and not to the assumption that health care costs grow faster than the economy.81
Alongside growing health care costs, Social Security, which reduced federal deficits over the past decade (because payroll tax revenues exceeded benefit payments), will instead start contributing to the overall budget deficit.82 The main factor pushing Social Security into the red is the aging of the population and the resulting decrease in the ratio of workers (who make contributions) to retirees (who collect benefits). The other factor is increasing income inequality. Because the payroll tax is only levied on earnings up to a cap, the more money the very rich make, the more income escapes the payroll tax. This is exactly what has happened: 10 percent of wage earnings were not subject to the payroll tax in 1983, but that figure climbed to 16 percent by 2010 and is expected to continue rising to 17 percent by 2020.83 Because of these factors, in 2010 Social Security payroll taxes for the first time were not enough to pay benefits, and that gap will only grow as the baby boom generation retires, the ratio of workers to retirees falls, and life expectancy increases.84 If both payroll taxes and promised benefits remain unchanged, the gap between the two will climb to 1.4 percent of GDP by 2035.85
The fundamental problem is that government spending on Social Security and health care will grow faster than tax revenues. In our “optimistic” scenario, where all the recent tax cuts expire and the rest of the government (other than Social Security and health care) remains at the early 1930s levels mandated by the 2011 debt ceiling agreement, the deficit will grow to 5 percent of GDP around 2030 and continue upward. (The national debt will be 68 percent of GDP in 2030 and rising, as shown in Figure 4-6.) If, instead, the income and estate tax cuts are extended, the deficit will be over 8 percent of GDP in 2030 (with the national debt at 103 percent of GDP) and climbing rapidly.86 In the latter scenario, the economy will not be able to grow fast enough to keep pace with both rising interest payments on the debt and rising primary deficits—meaning that the debt will only continue to grow.87 Before we reach that point, however, something will have to give; the only question is what.
If the short-term deficit problem is due to tax cuts, the long-term deficit problem is largely due to health care. If the tax cuts were allowed to expire and, in addition, government health care spending could magically be kept at 2021 levels, there would be no long-term deficit problem. In that case, government spending other than interest payments would be about 20 percent of GDP in 2030, revenues would be about 21 percent (assuming Congress simply did nothing except index the alternative minimum tax to inflation), and the primary budget would show a slowly growing surplus.88 There is no magic wand, however, that can freeze health care spending. So the federal government does have a long-term deficit crisis—but it is the product of a national health care crisis.
Figure 4-6: National Debt Projections
As the largest buyer of health care in the economy, the government has some influence on overall health care prices, but that influence is limited. Congress can dictate how much Medicare will pay doctors for a given medical procedure, but Congress cannot force doctors to accept Medicare patients; so for Medicare to be a viable health insurance plan, it must pay something reasonably close to the market price.89 While the cost of Medicare and Medicaid has been growing over the past few decades, the cost of health care that is not paid for by the government has been growing as fast or faster because it is subject to the same pressures: technological innovation and rising incomes.90 And without structural change, there is no reason to expect private health care costs to start falling. In 1985, health care spending accounted for one-tenth of the national economy; in 2009, it was about one-sixth; and by 2035, it is likely to be more than one-quarter.91 This means that Americans will have to devote a larger and larger share of their incomes to paying for health care, whether they pay for it in the form of lower wages (for employer-paid health insurance), higher premiums and copayments (for all types of health insurance), or higher taxes (for government-paid health insurance)—reducing the amount of money available for everything else.
Our health care problems are broader than our deficit problems. The long-term deficit could be eliminated, if you were willing to eliminate Medicare, Medicaid, the Children’s Health Insurance Program, and government subsidies for buying health insurance. But that would only shift costs onto individual families and, in all likelihood, increase total health care spending. In 2011 the House Budget Committee, chaired by Representative Paul Ryan, proposed converting Medicare into a program where beneficiaries buy health insurance on their own with subsidies from the government; but this would only increase total spending on the same health care by 41 to 67 percent.92 Such a solution makes sense only if we care about the budgetary health of the United States government but not the real health (and financial health) of the people it serves. Our goal must be finding a way to prevent the federal budget deficit from spiraling out of control without abandoning the elderly and the poor to pay for health care on their own—just as health care becomes harder and harder to pay for.
Looking at the projections, balancing the long-term federal budget may seem like a hopeless task. But from an economic standpoint, it is certainly possible. The best solution would be one that reduces the long-term growth of health care costs for everyone, including the government. There’s no law of nature that says we have to pay so much for health care. The Affordable Care Act was a first step in the right direction, but few people expect it to solve the problem completely.
Even without an effective solution to health care costs, however, we should still be able to afford everything that the federal government does—if we want to. The United States is one of the richest countries in the world.93 And, as mentioned before, we pay less in taxes than most rich, industrialized countries. If we really want to balance the federal budget, there is no economic reason why we can’t raise taxes enough to keep pace with growing health care spending. As long as health care is becoming more expensive, we all pay real money for it, one way or another. It is true that, at some point, higher tax rates can harm economic growth by reducing people’s incentive to work. But the highest tax rate today is lower than at any point since World War II except the 1986–1993 period, and there are many other ways around that problem, such as closing tax loopholes or broadening the tax base (which we discuss later).94
Now, you may not want to pay more taxes. Instead of paying higher taxes, you may prefer receiving lower Social Security and Medicare benefits—in effect, betting that you don’t need as much insurance as the government currently provides. That is also a reasonable solution to the long-term deficit. The bottom line is that the only way to reduce the deficit is raising taxes, reducing spending, or some combination of the two. From an economic standpoint, any of these approaches will work. We have plenty of money; we just have to decide how we want to allocate it between individual bank accounts and social insurance programs.
In a low-tax/low-benefit world, your bank account is a little bigger (if you make enough money to pay taxes), but you face more risk of running out of money in retirement or not being able to afford health care; in a high-tax/high-benefit world, your bank account is a little smaller, but you face less risk. Since rich people are better able to self-insure, they gain less by pooling their risk with other people, so they might be better off in a low-tax/low-benefit world; poor people cannot self-insure, so they gain the most from risk pooling, and they will be better off in a high-tax/high-benefit world.95 Compared to current policy, reducing benefits so we can keep our low tax rates is a form of redistribution from the poor to the rich; raising taxes so we can maintain today’s benefit levels is a form of redistribution from the rich to the poor (assuming that the tax increases are progressive). But again, either way will work. All we need to do is decide, as a country, what we want from our government.
This is fundamentally a political decision, which is why solving the long-term deficit problem seems so hard. Today, Republicans are dug in against tax increases and are demanding benefit cuts in the name of fiscal discipline (and, thanks to the antitax pledge, have even managed to sink their ships behind them, eliminating the possibility of retreat96). Democrats, led by President Obama, have shown the willingness to combine tax increases with benefit cuts, but are adamantly opposed to a solution that only includes benefit cuts. As long as this political stalemate continues, the national debt will continue to grow.
a These are marginal tax rates, meaning that even if you are in the 35 percent tax bracket, you pay 35 percent only on income above a certain threshold, now around $400,000. The current tax brackets were set by the 2001 tax cut; if it is allowed to expire, the top marginal rate will go back up to 39.6 percent. Technically speaking, there is no 0 percent tax bracket, since taxes start at 10 percent on any taxable income. In practice, the personal exemptions and the standard deduction ensure that some of your income is not taxable.
b Technically speaking, the employee pays half of each payroll tax and the employer pays the other half. In addition, the December 2010 tax cut lowered the Social Security payroll tax by 2 percentage points for 2011 (since extended through February 2012).
c Someone who makes $50,000 a year ordinarily pays 15.3 percent of her salary, or $7,650, in payroll taxes. Someone who makes $200,000 a year, however, pays $19,043 in payroll taxes because of the cap on the Social Security tax; this works out to a tax rate of only 9.5 percent.
d We begin with the CBO’s official baseline projection and make three adjustments, all among the CBO’s “alternative policy assumptions”: (1) we assume that deployments for Iraq, Afghanistan, and similar operations will decline as planned; (2) we assume that Medicare payment rates will be maintained rather than being allowed to fall by 30 percent, as would occur under current law; and (3) we assume that the income threshold for the alternative minimum tax will continue to be increased to account for inflation. CBO, The Budget and Economic Outlook: An Update, August 2011, Tables 1-4, 1-8, pp. 4–5, 26–27. The CBO baseline assumes $1.2 trillion in deficit reduction over ten years due to the actions of the “supercommittee” created by the Budget Control Act. The committee failed to reach an agreement, but instead the act requires $1.2 trillion in automatic spending cuts. For further details, see the Appendix.
e That is, the incremental cost for a given doctor’s visit or procedure (the copayment) is relatively low compared to its actual cost; we do end up paying the full cost in the form of insurance premiums, lower wages, and higher taxes.
