Peace Dividends Always Come Due, the 1990s Drawdown Reconsidered

Picture a family that cancels its fire insurance because the house has not burned down since Tuesday. Then picture the family putting the premium money in a jar, labelling the jar "dividend," and giving a press conference. That, folks, is a fair summary of American defense policy in the 1990s, and the remarkable thing is that it was bipartisan. A Republican administration started the cuts, a Democratic administration deepened them, and a Republican Congress in the middle of it all cheerfully voted for the invoice.

I want to be fair to the people who did it, because they were not fools. The Soviet Union had dissolved in December 1991. The Gulf War had just been won in about a hundred hours of ground fighting. The mood in Washington was that the long twilight struggle was over and the grown-ups could finally go home. The trouble is that the world was not consulted about the end of history, and it went right on having a history.

So here is the argument. The post-Cold War drawdown was not wrong because it cut. Some cutting was inevitable and right. It was wrong because it kept cutting past the point where the force matched the missions Washington kept handing it, and because it hollowed out the part of the defense establishment that takes years to rebuild: the factories, the shipyards and the skilled workers who run them. Those things cannot be turned back on like a faucet. And I will answer the strongest objection, which is that the savings helped produce the balanced budgets of the late 1990s.

The terms: what a peace dividend is

A peace dividend is the budget money freed when a threat recedes and the force built against it is reduced. The phrase was everywhere by 1990. It is a perfectly respectable idea. After every American war the country has demobilized, and it should. Nobody is arguing that a nation of 1991 should have kept the Army of 1989.

The question is never whether to reduce but how far, how fast, and what you keep. Defense planners separate three things that politicians blur together. Force structure is the number of divisions, wings and ships. Readiness is whether those units can actually deploy and fight. Modernization is the purchase of the next generation of equipment. You can cut the first and protect the other two. In the 1990s America cut all three, and then asked the smaller, older and less ready force to keep doing everything it had done before.

What the drawdown actually did

It started under the first Bush administration. General Colin Powell, then Chairman of the Joint Chiefs, drew up what became known as the Base Force: a plan to shrink the military by roughly a quarter by the mid-1990s. President Bush announced the new strategy in a speech at Aspen, Colorado, on August 2, 1990. That was the same day Iraq invaded Kuwait, which is the sort of timing that history seems to enjoy.

The Base Force was meant to be a floor. It was a deliberate, managed reduction that kept a force able to fight a major regional war and keep a presence overseas. Then the Cold War ended in earnest, a new administration arrived, and the floor kept descending.

In September 1993 Defense Secretary Les Aspin released the Bottom-Up Review. It sized the force to fight and win two nearly simultaneous major regional conflicts, the planning construct of the decade. On paper that was a respectable standard. In practice, critics inside and outside the Pentagon argued from the first day that the force it prescribed was too small and too thinly funded to meet the standard it set. A strategy that says "two wars" and a budget that pays for one and a half is not a strategy. It is a hope with a footnote.

Here is what the numbers looked like, in round terms:

• Active-duty strength fell from about 2.1 million in 1989 to about 1.4 million by 2000.
• Active Army divisions fell from eighteen at the end of the Cold War to ten by the mid-1990s.
• The Navy went from well over five hundred ships to about three hundred.
• Defense spending as a share of the economy fell from a little over 5 percent of GDP in 1990 to about 3 percent by the end of the decade.

Procurement fell even harder than the headline budget, because when you must cut and want to keep the troops fed, you cut the purchases. Planners called it a procurement holiday, a lovely phrase for a holiday during which the factories went quiet. The B-2 bomber was planned at 132 aircraft and ended at 21. The Seawolf submarine was planned at 29 boats and ended at three. The Army, flush with equipment bought in the 1980s, deferred its next generation of vehicles and upgraded what it had. Each decision, taken alone, had a respectable argument. Taken together they amounted to a decade in which the country stopped designing and building much of what it would need next.

Nobody voted on "let's stop making the future." It happened in a thousand line items, each defended as sensible, which is how most big mistakes in Washington happen.

The industrial base nobody thought about

This is the part of the story that cable news never covered, because there is no good footage of a missed production schedule.

In 1993 the Pentagon's leadership, led by Deputy Secretary William Perry, invited the heads of the big defense contractors to dinner and told them, in effect, that the budget was shrinking and the industry would have to shrink with it. The meeting later acquired a nickname: the Last Supper. The message was that the government expected consolidation, and it got it. Lockheed and Martin Marietta merged in 1995. Northrop acquired Grumman in 1994. Boeing absorbed McDonnell Douglas in 1997. A field of dozens of prime contractors narrowed to a handful.

Notice what that means. A market with five buyers and no more orders is a market in which the sixth firm does not exist when you need it. Below the primes, the damage was worse and quieter: the suppliers of castings, forgings, electronics, specialty steel and ship components. When a small firm loses the Navy as a customer, it does not sit patiently waiting for the next war. It switches to making something else, or it closes. The welder who retires is not replaced.

This is the lesson of How America Built 2,710 Liberty Ships and Why It Could Not Do So Today: industrial capacity is a stock, not a flow. You cannot call a shipyard back into existence in a quarter. The same logic runs through The Shipbuilding Gap with China Is an Industrial Story Before a Navy Story, which is the modern bill for decisions that were cheap to make and are dear to reverse. And the wartime tool that tries to repair the problem after the fact is explained in The Defense Production Act of 1950 Is a Wartime Tool with Peacetime Habits.

The argument in six steps

Let me lay it out the way an argument should be laid out.

1. The threat did not disappear, it changed. The end of the Soviet Union removed one enemy. It did not remove Saddam's army, the North Korean regime, a rising China, Iran's revolutionary government or the networks that were already attacking Americans abroad. The 1993 World Trade Center bombing came in the same year as the Bottom-Up Review, and the attacks on the Khobar Towers in 1996, the embassies in East Africa in 1998 and the USS Cole in 2000 followed. The war was already underway. It just did not look like the old one.
2. A force is sized to missions, and the missions did not shrink. In the 1990s American troops went to Somalia, Haiti, Bosnia, Kosovo, the Persian Gulf and the Korean Peninsula's edge, and they enforced no-fly zones over Iraq for years. A smaller force was spread across a larger map. That wears out people and machines faster, and it shows up as retention problems and aging equipment.
3. Capability is cheap to dismantle and expensive to rebuild. Cut a division and you save its budget next year. Re-create it and you pay for recruiting, training, equipping and, most of all, the experienced sergeants and officers who cannot be hired off the street. Cut a production line and the tooling, the workers and the supplier network disperse. This asymmetry is the whole case against treating the force like a thermostat.
4. Deterrence is a perception, and perceptions can be bought cheaply or lost cheaply. An adversary decides whether to test you by estimating your power and your will. Shrinking the first while advertising the second as a luxury is an invitation. I take this up at length in Deterrence Fails When an Enemy Thinks You Will Not Pay the Price.
5. Both parties share the blame, and so both parties share the lesson. The Base Force was a Republican plan. The Bottom-Up Review was a Democratic one. A Republican Congress elected in 1994 did not reverse the slide in procurement. A conservative who pretends this was all one party's doing has not read the appropriations bills.
6. The bill arrived with interest. After September 11, 2001, the country went to war in Afghanistan and then Iraq with a force that had been shrunk, a ground army that had to rotate the same units repeatedly, and industrial lines that had to be restarted under fire. The surge in spending that followed was in part the price of rebuilding what the 1990s had let lapse. A dollar saved in 1996 and spent with a premium in 2004 is not a saving. It is a loan from a very expensive lender.

The fair objection: the 1998 surplus

The strongest answer to all this runs as follows. The federal government ran a surplus in fiscal 1998, about $69 billion, the first in almost thirty years. It ran surpluses for several years straight. Defense was the biggest discretionary line in the budget, and its decline as a share of the economy was one of the main reasons a balanced budget became possible. Conservatives wanted balanced budgets. You cannot celebrate the surplus and mourn the cut that made it possible.

That is a serious argument, and part of it is correct. The surplus was real, the discipline behind it was good, and defense restraint contributed. I will grant all that.

Now check the arithmetic, labelled as an illustration and not an audit. In 1998 the economy was roughly $9 trillion. Defense had fallen by roughly two points of GDP from its 1990 share. Two percent of $9 trillion is about $180 billion a year. So, measured against the 1990 share, the defense decline was worth much more than the entire $69 billion surplus. That does not prove the surplus was an accounting trick. Revenues surged in the late 1990s boom, and other lines moved too. It does show that defense paid a bigger share of the deficit reduction than its slice of the budget would suggest, and that is exactly the point. Defense was the line that was easiest to cut, not the line that was the cause of the deficit. Then as now, the big engines of long-term spending were the entitlement programs, and they were left alone.

So the right reading is this. The country took the cheap savings from the Pentagon, did not touch the expensive ones in the entitlement programs, congratulated itself, and then discovered that the security savings had a due date. A budget that is balanced by deferring the maintenance on the national defense is like a household that balances its books by skipping the roof repair. The surplus was real. So is the water in the attic.

There is a second answer. The question is not whether the 1990s cut defense. It is whether the cut was sized to the world. The better test is the Weinberger standard, laid out in The Weinberger Doctrine Set Six Tests Before Sending Troops to Fight: if you intend to use force, resource it to win. The 1990s used force abundantly and resourced it grudgingly. If you will not pay for the force, do not send it. They did both anyway.

What the Founders said about the cost of readiness

George Washington told Congress in his first annual message, on January 8, 1790: "To be prepared for war is one of the most effectual means of preserving peace." That sentence is older than the Bill of Rights' ratification, and it is a better guide to the 1990s than every think-tank paper of the decade.

The Constitution takes the question seriously and with balance. In Article I, Section 8, Congress is given the power "To raise and support Armies, but no Appropriation of Money to that Use shall be for a longer Term than Two Years." The Founders feared standing armies and wrote a leash into the text: the Army must be funded afresh at least every two years so the people's representatives control it. But the same section says Congress shall "provide and maintain a Navy," with no two-year limit at all, because a navy cannot be improvised and does not threaten liberty at home the way an army can. They were pointing at the same lesson: some capabilities take a generation to build.

The preamble lists "provide for the common defence" among the reasons for the whole project. It does not say "provide for the common defense when convenient" or "until a favorable budget forecast appears." And note the constitutional discipline works both ways. Congress, not the Pentagon, owns the decision. The 1990s drawdown was not imposed by generals. It was a decision by elected officials and signed by presidents, so the remedy is not complaining about the brass. It is reading the budget, which brings us to what you can do.

Do not misread this as a case for permanent maximum spending. The same Founders who wanted a navy wanted accountable money, and the Pentagon's own books remain a scandal; see The Pentagon Has Never Passed an Audit, and Conservatives Should Mind and the longer look at spending levels in Defense Spending as a Share of the Economy, From Korea to the 2010s. A conservative should want the force sized to the threat and the money tracked to the penny. The 1990s managed neither.

What the bill looked like after 2001

The September 11 attacks were, among other things, a lesson in what a free country forgets when it has had too long a holiday. The commission that studied them named a failure of imagination, and I would add a failure of upkeep; the real lessons are in The 9/11 Commission Found a Failure of Imagination, Read Its Real Lessons.

The pattern after 2001 is easy to describe without a single statistic. The Army had to fight two wars with ten active divisions and a reserve component that was called up in numbers nobody had planned for. Units rotated back to combat zones with less time at home than the training cycle required. The Marine Corps, a smaller service, did the same. Equipment that had been bought in the 1980s was worn out by desert use, and the replacement programs had been deferred through the 1990s, so there was nothing in the pipeline. The Navy and Air Force found that aging fleets cost more to keep flying.

Do I say the 1990s caused every difficulty that followed? No. Wars have a way of producing their own surprises, and strategy mistakes after 2001 were made by people who had a full force to spend. The claim is narrower and harder to dodge: the country went into the 2000s with less depth than it needed, and depth is the thing that gives a president options. A thinner force narrows the choices of the commander in chief, and a nation that has fewer choices makes worse ones.

I will add a word about the optimism, because I am an optimist. America can rebuild. It has before: after the interwar neglect, after Korea, after Vietnam, when the volunteer force of the 1980s became the best the country ever fielded. The recipe is not mysterious. Keep the force matched to the mission, protect the training, keep at least one production line for each critical item alive, and never let a savings figure stand in for a strategy. The lesson of the 1990s is not despair. It is that a nation which remembers what the dividend cost can decline to take it next time.

The case in brief

1. A peace dividend is legitimate in principle, and the 1990s cut was not wrong because it cut but because it cut past the missions the country kept assigning.
2. The Base Force, the Bottom-Up Review, the loss of eight active Army divisions, the procurement holiday and the consolidation of the industrial base were bipartisan decisions, and each looked sensible in isolation.
3. Capability is cheap to dismantle and costly to rebuild, so savings taken from factories, shipyards and experienced soldiers carry a bill with interest.
4. The 1998 surplus was real, but defense paid far more than its share of it while the larger engines of spending were left alone, and the bill arrived after 2001.
5. The Constitution assigns the decision to Congress, which means the remedy belongs to voters, not to the generals.

What you can do: read the defense appropriations bill for your own representatives on Congress.gov, look at the procurement lines and not only the top number, and ask your member of Congress in writing which production lines will still exist in ten years.