Throughout the twentieth century, U.S. military capabilities were sufficient to protect the United States and its allies. For many decades, the United States has been the global leader in military spending, and it continues to be so today. In the current fiscal year, U.S. defense outlays will total roughly $550 billion, reportedly more than the defense expenditures of the next 40 nations combined. Yet this spending is probably not enough to ensure the security of the United States -- and for something as critical as national security, even "probably enough" is inadequate. Handling the new threats facing the United States will require a significant rise in defense spending; the real questions are how much more is needed, what the new funds should be spent on, and how the money can be raised.

There are undoubtedly excesses in some parts of the defense budget and many examples of wasteful spending, but those are inherent features of any government activity. And unlike some areas of government spending where it is possible to consider privatization or a shift of responsibility to states and localities, defense must remain a federal responsibility, unchecked by the disciplining forces of competition. So one cannot expect to achieve dramatic gains from increased efficiency alone.

Deterring other great powers, such as Russia and China, will require Washington to maintain its dominance in conventional warfare and therefore at least to maintain its current level of military spending. But in addition, the United States now faces three new types of threats for which its existing military capacity is either ill suited or insufficient. First, there are relatively small regional powers, such as North Korea, Iran, and Pakistan, that can or will soon be able to strike the United States and its allies with weapons of mass destruction (WMD). Second, there are global nonstate terrorist networks, such as al Qaeda, with visions of re-creating the world order. And third, there are independent terrorists and groups motivated less by a long-term vision of global conquest than by hatred, anti-Americanism, and opposition to their own governments. Each of these threats is exacerbated by the relative ease with which crude WMDs can be developed due to the diffusion of modern technology and the potential emergence of a black market in fissile material.

Furthermore, there seems to be general agreement that the United States has committed so much of its war-fighting capacity to Iraq and Afghanistan that it could not fight in Iran or North Korea or elsewhere if that were deemed necessary. That limit on capacity encourages U.S. adversaries to behave in ways that are contrary to U.S. interests. Those adversaries would be less likely to do so if Washington had the extra manpower and equipment that were once assumed to be the goal -- and perhaps the reality -- of the U.S. military structure.


Dealing appropriately with these diverse enemies will require an increase in intelligence capabilities, both human and electronic. And the new threats mean that the United States will have to expand its intelligence activities at home as well as abroad, as the United Kingdom has done over the years in its struggle against the Irish Republican Army.

The new types of enemies will also require increases in the size of the U.S. Army and the Marine Corps. In the 1950s, President Dwight Eisenhower put forward a strategy emphasizing nuclear weapons because he concluded that maintaining a conventional army large enough to deter the Soviet Union would simply be too expensive. That strategy worked against the enemies of the day, but the new threats are not ones that nuclear weapons can be relied on to deter. And although conventional weapons have an important role to play, they, too, are insufficient for challenges such as urban warfare, which requires large numbers of combat troops and equipment that is sophisticated enough to protect and enable them.

The role of the Reserves and the National Guard in national security also has to change. These forces have been called on to serve for long periods in Iraq, in some cases for more than one tour of duty. It is not going to be easy to attract recruits if they face the prospect of that kind of service in the future. Nor is it appropriate to disrupt the civilian economy in such a way. The U.S. military of the future, therefore, will require larger numbers in the active army and a National Guard trained to play a significant role in homeland security.

A larger budget will also be needed to replace materiel used up in Afghanistan and Iraq, modernize military equipment more generally, and make equipment better suited for rapid deployment and the kinds of warfare the country increasingly faces. Today's weapons have much greater accuracy and lethality that those of the 1960s or even the 1980s, but they are not well suited to urban warfare: for that, the U.S. Army and the Marines need ground equipment that protects soldiers in close combat from land mines and grenade launchers.

But the delays and difficulties plaguing recent attempts at modernizing weapons systems, such as the army's experience with the Future Combat Systems program, are not encouraging. The basic idea of the FCS -- to create an integrated system of modern high-tech weapons that can be rapidly deployed -- is generally accepted as an appropriate shift for the army, despite disagreements over technical details. When it was proposed back in 1999, the plan was to begin producing weapons in 2006 and start getting them into the field in 2008. But production has yet to begin, and the army now estimates that the first fully equipped brigade will not be deployed until 2014 and that all units will not be fully outfitted until 2035.

Waiting almost 30 years to fully implement today's best technology is hard to fathom. The time from now until the projected first deployment of the FCS is more than the duration of World War II. Why was the United States able then to mobilize vastly more resources in so much less time? How many better ideas will come along over the next three decades, and how will military challenges change in the interim? Delays such as those which plague the FCS will ensure that the army misses what it needs during the next decade and gets saddled with outdated equipment later on.

Some analysts have voiced concern over the FCS' price tag -- $10 billion a year. But the program's cost represents less than 2 percent of today's defense budget. And since the defense budget is only 4 percent of GDP, the FCS plan for modernizing the army involves an annual cost of less than one-tenth of 1 percent of GDP.

The right kind of international aid could be a useful complement to increased military capacity. Although traditional economic assistance is less useful now than it was before the expansion of global capital markets, U.S. political and military assistance designed to help other countries fight terrorists within their own borders could serve the interests of foreigners and the United States simultaneously. An expanded U.S. foreign aid program could also counter the funds being used by Venezuelan President Hugo Chávez to promote anti-Americanism in Latin America and by several oil-producing countries in the Middle East to finance the spread of radical Islamism in Pakistan and elsewhere.

Ever since 9/11, it has also been clear that enhancing national security requires more spending to address security concerns within the United States. But achievements in homeland security have been limited. The U.S. government recently canceled its sole contract to produce an anthrax vaccine because the small company on which it had relied failed to meet its performance goals. The Department of Homeland Security recently abandoned a program to use electronic bioscanners to check for foreigners who had stayed beyond their visas' expiration dates. The government has been warning of risks to chemical plants and Internet infrastructure since before 9/11 but has not adopted federal programs to deal with these risks. And U.S. airlines have not acquired the technology that would enable them to elude surface-to-air shoulder-fired missiles due to its expense.


Despite these needs, U.S. national security expenditures today remain low relative to national income (the best measure of a nation's capacity to spend). Defense spending is now 4 percent of GDP, including the funds for the operations in Afghanistan and Iraq. Without those operations, defense outlays would now be about 3 percent of total economic output, whereas in 1962, before the Vietnam War, defense spending was 9.3 percent of GDP. By 1979, that percentage had been cut nearly in half, to 4.7 percent, in order to make room for the growth of domestic social programs. Defense spending rose rapidly during the presidency of Ronald Reagan, rising to 6 percent of GDP by 1986 -- a trend that helped bring about the collapse of the Soviet Union. The end of the Cold War led to the so-called peace dividend in the 1990s, and by 2000, defense spending had fallen to just 3 percent of GDP. Foreign assistance programs, meanwhile, only amount to 0.3 percent of GDP, essentially where they have been for most of the past 40 years.

Although the new threats confronting the United States imply that more funding for national security is in order, there is no way to say precisely what percentage of GDP should be devoted to it. The appropriate level of outlays can be determined only through a detailed budget process that examines each potential use of funds and arrives at an aggregate figure. But to frame the discussion about how an increase could be financed, it is useful to consider the 6 percent figure achieved during the Reagan years as a specific target.

Returning to that point would require devoting an additional 2 percent of GDP to defense. The United States clearly could afford to do so if it wanted to. Since real GDP can be expected to grow by about 15 percent over the next five years, a 2 percent rise in the share of output devoted to defense would account for only about one-sixth of this additional national income. The decision to increase national security spending is therefore a question of politics and budgets rather than available resources.

The current budget situation in the United States is healthier than often presumed, with a fiscal deficit -- 2 percent of GDP -- low enough to produce a slight decline in the size of the national debt relative to GDP. This favorable fiscal situation will not last, however, because an aging population will cause government spending on programs for the elderly to rise faster than the revenue used to finance those programs. The Congressional Budget Office projects that the combined costs of Social Security, Medicare, and federal Medicaid will rise from 8.7 percent of GDP in 2006 to 10.8 percent in 2016 and 16 percent in 2036. By that point, these three programs plus the interest on the national debt would use up almost all federal tax revenues (if tax rates remained where they are now).

Of course, this will not be allowed to happen. A political solution will be found to the problem of funding these important programs through some combination of tax increases, slower benefit growth, and a shift to investment-based accounts for Social Security. The fiscal challenge posed by defense spending can therefore be met at the same time that independent solutions are devised to finance future benefits for the aged.

But raising personal income tax rates to fund increased defense spending would be a serious mistake. Since individual income taxes now account for revenue equal to 8 percent of GDP, increasing them enough to raise defense spending to 6 percent of GDP would require the equivalent of a 25 percent tax hike across the board. And that assumes the increased taxes would not cause behavioral changes of a kind that reduces taxable incomes. In fact, given how people respond to incentives, financing such a rise in defense spending through an across-the-board tax-rate hike would probably require an increase of more than 40 percent.

Fortunately, it is not necessary to raise tax rates in order to increase the share of national income collected in taxes. Because of the graduated structure of tax rates (with higher incomes taxed at higher marginal rates), the general rise in incomes that happens every year automatically causes total personal tax revenues to rise faster than GDP. More specifically, under the current tax structure, economic growth in the United States causes personal tax revenue as a share of GDP to rise by about 1 percent a decade. That means that personal tax revenue will rise from 8 percent of GDP now to 9 percent in 2016 and 10 percent in 2026. (These estimates, by the Congressional Budget Office, assume that the tax cuts implemented by President George W. Bush in 2001 and 2003 will be renewed in 2008 and 2011.) This alone could finance an increase in the defense budget to 6 percent of GDP by 2026 if other uses of the tax revenue do not rise faster than GDP.

But although such a route would lead to the desired destination, the trip would take too long. One way to accelerate the pace would be to reverse the rise since 2000 in the share of GDP being spent on nondefense discretionary programs. That would not require actual cuts in these programs but rather simply holding spending on them constant in real terms. Combining the money thus saved with the automatic rise in the tax share of GDP would increase the funds available for defense by one full percentage point of GDP in five years and by two full percentage points by 2016.

A final and important source of incremental funds can be found in the long list of "tax expenditures": the revenue lost to special tax rules that exclude certain sources of income from being taxed (such as cash paid to employees to reimburse them for parking expenses) or that allow special deductions (such as for state and local income taxes). Some of these tax expenditures have a long history, whereas others have been introduced or expanded in recent years. Together, the largest 19 of these account for more than $700 billion of lost revenue, more than 5 percent of GDP. Eliminating tax expenditure rules may seem difficult, but they have been reduced in the past and could be again.

In short, a substantial increase in defense spending is necessary to protect and enhance U.S. national security, and the revenues to finance such an increase could be generated without resorting to higher marginal tax rates. What it would take is political leadership from an administration that recognizes the importance of national security and is committed to acting accordingly.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Martin Feldstein is George F. Baker Professor of Economics at Harvard University and President and CEO of the National Bureau of Economic Research.
  • More By Martin Feldstein