To understand why the end of the Cold War did not produce a peace dividend, this report should be read in conjunction with two more recent reports, which I published in 1996 and 1997.

  1. "Defense Spending Time Bomb," Challenge: The Magazine of Economic Affairs, July-August 1996, pp. 23-33.

  2. "Porkbarrels and Budgeteers: What Went Wrong with the Quadrennial Defense Review, Strategic Review, Fall 1997, pp. 29-41.

While this paper was originally published as a Fund For Constitutional Government Report in the Fall of 1990, during the deployment but before the outbreak of hostilities with Iraq, it correctly predicted that the costs of that war would be astronomical and that there would be no peace dividend. The burden of these costs ($60 to $100 billion for 43 days of conflict) was hidden from the US taxpayer, because the Saudis paid for much of it. Moreover, our continued deployment in the Gulf shows that we are still paying for this war, and that burden is now hurting the combat readiness of our military forces.

Although the data in this report is now eight years old, the basic arguments describing the origin and conduct of the Defense Power Games and their relationship to the skyrocketing cost of defense are as valid today as they were in 1990.

Taken together, these three reports provide a good background for understanding the root causes of the readiness and modernization problems that are now engulfing the Pentagon, even though we are outspending all our potential adversaries combined by at least three to one.

Franklin C. Spinney
June 14, 1998


October 1990

The power politics practiced by the Pentagon and Congress continue to drag our nation deeper into a quagmire of spiraling weapons' costs, shrinking forces, and high defense budgets. This essay describes how bureaucratic gaming and political logrolling weaken our military, corrupt behavior in the Pentagon, and subvert the checks and balances of the Constitution. While events in the Persian Gulf temporarily divert the public's attention and seduce decision makers into postponing urgently needed corrective action, the defense power games are continuing as if the cold war never ended. Left unchecked, these politics guarantee that the much smaller post-cold-war military will require a cold-war budget to keep it running. The essay concludes by arguing that the key to reform is a Congress that actively asserts its constitutional prerogatives. It identifies four changes Congress could make today, if it wants to get serious about exorcising the self-destructive effects of the defense power games.



  1. Background: Two Perspectives of Defense Budgets in the Post-Cold-War Era

  2. IThe Defense Power Games: Definitions and Consequences

  3. Business as Usual: How the Defense Power Games Operate

  4. The Defense Power Games in Action (A): Political Blitzkrieg

  5. The Defense Power Games in Action (B): Meltdown

  6. Reform: Back to Fundamentals 


Figure 1: $167 BILLION REDUCTION, JAN 1990






Table I: Early Plans: Growth Rates Assumed by Decisions Made during Jan-May, 1981 

Myths have a way of becoming reality in Washington, particularly when they serve to further the agendas of politicians and bureaucrats. One of the most useful myths in recent memory is that Saddam Hussein's conquest of Kuwait robbed the United States of its hard-earned peace dividend. The ballooning costs of keeping our forces in the Persian Gulf, costs that will explode if war breaks out, will condition the American people to high defense budgets in the near term. But when the dust settles, and Saddam's gambit is history, taxpayers will be surprised and perhaps angered when they learn the real reason why there won't be a peace dividend: The much smaller, hi-tech military of the post-cold-war era requires a cold-war budget to keep it running.

This essay explains how the power politics practiced by the Pentagon and Congress put the United States into this predicament, and it describes a set of institutional reforms which can restore balance, discipline, and accountability to the decisions that shape our military forces.

I. Background:

Two Perspectives of Defense Budgets in the Post-Cold War Era

The bankrupting cost of business as usual is becoming evident in the deployment of our military forces to the Persian Gulf. While no one knows how much Operation Desert Shield will ultimately cost, the first estimates were staggering. In early September, the Pentagon estimated it would cost $11.3 billion to keep 100,000 men and women in the Persian Gulf for Fiscal Year 1991 (October 1 1990 through September 30 1991), assuming war does not break out. Put another way, each soldier standing guard in Saudi Arabia for a year would cost the taxpayer $113,000. This equates to 73 percent of the $155,000 it cost to keep a soldier fighting for a year in Vietnam between 1966 and 1971. (The effects of inflation have been removed in this comparison.)(1)

It is now clear this estimate was only the tip of the iceberg; consider the following: As early as September 6, according to Hobart Rowen of the Washington Post, Senator Sam Nunn was arguing that a full year's cost, plus the aid given to friendly nations, would be about $50 billion—or more than four times as high as the Pentagon's initial estimate. By early October, newspapers reported that the Pentagon had increased the deployment to over 200,000 troops, and by early November (as this paper went to press), President Bush announced yet another increase, bringing the planned commitment to about 400,00 troops. The Pentagon has yet to provide revised cost estimates for these subsequent increases. If fighting breaks out, some experts believe that costs could skyrocket to $1 billion a day—more than four times the cost of Vietnam.(2)

The harsh fact is that no one knows how much Desert Shield will cost, but the bill is going to be very high. Bear in mind, we are discussing the incremental costs of keeping about one quarter to two thirds (depending on the category) of our active-duty combat units in the Persian Gulf. These costs are limited to moving forces to the Persian Gulf, maintaining and re-supplying existing forces at a greater distance from their home base, training harder in a harsher environment, mobilizing some reserves, and if war breaks out—fighting. It has not been necessary to create any new combat units. Although the buildup in the Persian Gulf has been much more rapid than the initial buildup in South Vietnam (as many troops were sent to the Persian Gulf in three months as were sent to South Vietnam in the twelve months between June 1965 and June 1966), the commitment to Desert Shield is still smaller than Vietnam—an upper limit of 450,000 would be about 84% of the troop strength committed in Vietnam during the peak years of 1968 and 1969.

Notwithstanding the smaller commitment of troops to Desert Shield than to Vietnam, it is now clear the United States can not afford to pay for the operation. In September we witnessed the spectacle of our Secretary of State rushing around the world with the tin cup out, begging our allies for financial aid. Congress found it necessary to design legislation establishing the Defense Cooperation Fund so it could "appropriate" these "gifts," and thereby have a say in how the money is spent. Cynics might conclude from these bizarre activities that the United States is exporting troops to earn hard currency, and Rube Goldberg has replaced James Madison as our nation's chief designer of checks and balances. Can we no longer afford to defend our "vital" interests? Is the all volunteer army becoming the source of the twentieth century's Hessians or Janisaries? Must we pervert the power-sharing arrangements so carefully crafted into the Constitution by the Founding Fathers? How could the United States end up in such a humiliating position after "winning" the Cold War?

To answer such questions, we need to appreciate that the unaffordable cost of Desert Shield is but one consequence of a much deeper problem. Costs are out of control because the Pentagon and Congress will not make hard decisions, and because they will not or can not choose, the needs of a coherent defense policy have been preempted by the selfish desires of its individual components. The situation we faced before Saddam invaded Kuwait puts the consequences of this breakdown in a clearer perspective. Let's examine it.

In January 1990, President Bush announced that the reduction in world tensions permitted major cutbacks in the size of our military forces. He told Congress that he would reduce Defense Department's 1990 to 1994 plan by $167 billion.(3)

Three months later, the generals and admirals submitted the first draft of the new five-year plan to Richard Cheney, the Secretary of Defense. Newspapers reported that seven tactical fighter wings, 200 Minuteman II missiles, 75 strategic air refueling aircraft, 87 ships, and three active and two reserve Army divisions were put on the chopping block. The military chiefs also planned to slow the rate of modernization by stretching out or reducing the production of high priority programs like the single warhead strategic missile, the Advanced Tactical Fighter, the B-2 stealth bomber, the C-17 transport, the SSN-21 attack submarine, the DDG-51 guided missile destroyer, and the advanced cruise missile. By the mid-1990s, instead of having the 44 tactical fighter wings envisioned in President Reagan's first five-year plan, President Bush would have 28 wings; instead of a 600 ship navy and a 28 division army, the new plan provided for 488 ships and 23 divisions with proportionally fewer active duty maneuver battalions.(4)

With a few exceptions, the average age of our equipment would be older in the mid 1990s than it was in 1980, and this smaller force would be flown, marched, steamed, and driven less often than in 1980.

Figure 1

While this collage of cutbacks appears to be a decisive response to the end of the Cold War, its economic ramifications present a far different picture. Figure 1 shows that they merely brought future budgets down to a level that Congress probably would have appropriated, had the Cold War continued for another five years. The previous plan, approved in April 1989, assumed that the budget would grow from $295 billion in Fiscal Year 1990 to $369 billion by 1994.(5)

As Figure 1 shows, subtracting $167 billion brought the 1994 goal down to $308 billion, putting future budgets on a straight-line projection of the budgets Congress had been appropriating since 1985. Taking out the effects of inflation, this would reduce the annual budget by about two percent per year. Given these circumstances, calling the $167 billion reduction a peace dividend would be like saying you lost twenty million dollars when you didn't win the D.C. Lottery. That money was never there to begin with.

Therein lies the essence of our current predicament: if a cutback to the late-1980s, cold-war status quo resulted in the carnage described by the military chiefs in May, real budget cuts will trigger chaos in the military-industrial complex. Saddam's gambit in Kuwait may temporarily mask this reality, but it will not make it go away.

II. Defense Power Games:

Definitions and Consequences

How could the 2.4 trillion dollar spending spree of the 1980s bring us to the brink of chaos in the 1990s?

Part of the answer to this question, I think, can be found in the way the Pentagon and Congress play power politics. Two power games—known in Pentagonese as front loading and political engineering—help to explain how we set ourselves up for a crash in the 1990s, and why, if we continue business-as-usual, our nation will remain stuck in a quagmire of spiraling costs, shrinking forces, and high defense budgets, Cold War or no Cold War.

Front loading is the practice of planting seed money for new programs while downplaying their future obligations. This game, which is a clever form of the old-fashioned "bait-and-switch," makes it easier to sell high-cost programs to skeptics in the Pentagon and Congress. Political engineering is the strategy of spreading dollars, jobs, and profits to as many important congressional districts as possible. By making voters dependent on government money flows, the political engineers put the squeeze on Congress to support the front-loaded program once its true costs become apparent. Front loading and political engineering are about increasing the flow of money; the former starts the money flowing while the latter tries to lock the spigot open, and in American politics, control of the money spigot is power.

These games do more than reinforce each other in a political sense, however. They also bias the choice of technologies by creating powerful political motives to buy complex high-cost weapons in lieu of simple low-cost weapons. Complex hardware is easier to front load than simple hardware. The more complex a piece of equipment is, the more uncertain we are about its ultimate performance and cost. More things can go wrong, and often the existence of some uncertainties is not even suspected during the early stages of a program's life cycle. There is, therefore, an intrinsic tendency to overstate performance while understating problems and costs. The front loader cynically exploits this uncertainty by hiding the future consequences of today's decisions in a fog of overly optimistic predictions. Since complex weapons embody many intricate subsystems, they require more subcontracts in their manufacture than do simple weapons. Increasing complexity, therefore, creates more opportunities for funneling money through sub-contractors to crucial congressional districts. The temptation to use these opportunities as patronage to buy political support is difficult if not impossible to resist, given the competition for resources and the division of power between the President and Congress.

The Pentagon and its willing partner, the Congress, have been playing these games with increasing subtlety for the last thirty-five years. During peacetime, both sides benefit: the Pentagon gains money and power, and incumbents in Congress get pork and votes, but these benefits carry a heavy price:

  • Decision makers on both sides of the Potomac sacrifice the real needs of the military on the altar of porkbarrel politics. These games create a pattern of decisions that guarantees costs will grow faster than budgets, even when budgets grow at very high rates, as they did in the early 1980s. The inevitable long-term results are unworkable complexity, smaller forces, older equipment, shortages of spare parts and ammunition, continual pressure to reduce training for combat—and high defense budgets.

  • By making money the focus of decisions, front loading and political engineering encourage immoral behavior at all levels within the Defense Department. We exaggerate the threat to justify larger budgets. We use deceitful if not illegal accounting tricks to hide the true costs of programs. We reduce the chances of weapons being terminated for poor performance by designing success-oriented operational tests and by rushing weapons into production before they are fully tested. We obscure future costs behind the cloak of excessive secrecy. We tolerate cost overruns and bad management practices, some of which are spilling over into the civilian economy and damaging our international competitiveness.

  • By striving to hook specific regions and their representatives in Congress on the narcotic of defense spending, these games corrupt the political relationship between the Defense Department and Congress. Front loading and political engineering aim to neutralize Congress's power of the purse, and to the extent that they succeed, they subvert the checks and balances that are the heart and soul of our constitutional system of government. A craven Congress, paralyzed by its addiction to the President's checkbook, corrupted by the selfish actions it must take to keep the money flowing to its constituents, is not the guardian of individual liberty that the Founding Fathers had in mind.

That's a brief introduction to business-as-usual and the damage it causes. Let's examine these games in more detail to understand how their effects overwhelmed the unprecedented budget increases of the early 1980s and set the stage for chaos in the 1990s.

III. Business as Usual:

How the Defense Power Games Operate

The Defense Department's annual budget request is the first year of a continuously updated five-year spending plan. Long-range planning is necessary because most of the programs contained in the budget entail an obligation to spend money far into the future. When the Pentagon and Congress agree to buy a new aircraft carrier, for instance, they are committing the nation to a spending stream that could last as long as 50 years. In theory, the five-year plan is supposed to place such decisions in context of their future obligations. During the 1970s and 1980s, however, the planners in the Pentagon systematically downplayed these future obligations by basing their decisions on three unrealistic assumptions:

New weapons will cost less to operate because they will be more reliable and easier to maintain than the older weapons they are replacing.

These assumptions allowed the bureaucrats in the Pentagon to front-load the first year of the five-year plan (which is also the annual budget request made to Congress) with too many high-cost investment programs: Larger budgets in the later years (Assumption 1), coupled with the biased allocation toward investment (Assumption 2), increased the total money available for developing and buying new weapons. The assumption of sharply declining purchasing costs (1st part of Assumption 3) permitted even more weapons to be stuffed into the later years of the investment plan, known in the Pentagon as the "bow wave." Planners, therefore, could tolerate low production rates in the early years, because the rising bow wave coupled with the declining unit costs promised higher rates in the later years. By the fifth year of the plan, the Defense Department would get well: Higher complexity weapons would be produced at much higher rates, forces would be larger and newer, training tempos would be higher (because equipment is assumed to be more reliable and easier to maintain—2nd part of Assumption 3), and large quantities of spare parts and ammunition would be flooding into the stockpile.

Figure 2

There was only one problem; the fifth year never arrived. Congress appropriated less than the Defense Department requested, and costs were higher than predicted. Figure 2 helps to explain the annual unraveling process; it depicts a hypothetical five-year plan where the total budget and its four future years are broken down into two categories: "Operations" is the cost of maintaining a combat ready force; it buys salaries, training, maintenance of equipment and property, and the thousands of activities needed to run the military on a daily basis. "Investment" buys the programs needed to modernize and expand the force; it includes research, development, procurement, and construction—programs such as Star Wars, stealth bombers, Trident submarines, M-1 tanks, new hangers, wind tunnels, etc. Figure 2 makes the following typical assumptions: the total budget will grow at about 7 percent per year for the next five years; and within that total, the operating budget will grow at 3 percent per year, which will permit the investment budget to grow at 12 percent per year.

While the Pentagon decides what it wants to buy in the context of a five-year plan, Congress decides what the Pentagon actually buys one year at a time. Each year, Congress reduces the total budget by making marginal changes to hundreds of individual line items in the first year of the plan. Each change Congress makes causes a ripple effect to reverberate into the later years of the plan. The Pentagon must take all these changes into account—in effect, sliding its desires further into the future—as it prepares its budget for the following year (which is the first year of a new five-year plan). Over time, this action-reaction sequence creates a powerful external pressure to shift the top line depicted in Figure 2 down and to the right (Assumption 1 unravels—for a preview of how this happened in the 1980s, the non-believer may look ahead to Figure 3).

At the same time Congress is reducing the budget, operating costs are growing more rapidly than planned: The complex equipment entering the force is less reliable and harder to maintain than was originally predicted (the second part of Assumption 3 unravels), and this causes the operating budget to grow uncontrollably, thereby creating a powerful internal pressure to increase its share of the total budget (Assumption 2 unravels).(8)

The only way to offset this internal pressure is to hold down operating tempos, cut purchases of consumption items and support equipment, reduce depot maintenance, and ultimately shrink the size of the force.

The external pressure imposed by Congress and the internal pressure caused by unplanned growth in the operating budget combine to squeeze the investment budget, decreasing it in the near term while pushing the programs making up the later years further into the future, much like toothpaste being squeezed out of a tube. Meanwhile, the costs of the programs making up the investment budget—i.e., the tanks, airplanes, ships, etc.—are also increasing (the first part of Assumption 3 is unraveling). The three most important causes of cost growth—the buy-in, continual design change, and the stretch-out—are closely related to the defense power games.(9)

One reason, the program stretch-out, is a direct result of the unraveling process itself: The low-rate production in the early years of the five-year plan must be extended into the later years, because the investment dollars needed to finance the higher rates never materialize. Consequently, unit costs do not decline as much as predicted (Indeed, they often increase!), and the total cost of buying all the programs in the investment plan begins to build up, imperceptibly at first, then more rapidly, like an ocean wave before crashing on a barrier reef. As the wave rolls forward, year after year, the liabilities beyond the fifth year of the plan grow larger and larger, hence the term "investment bow wave."(10)

Summing up: The pathological logic of the front-loading game creates a five-year plan that short changes the operating budget while maintaining too many modernization programs in the investment budget. When the plan unravels, as it does year after year, decision makers in the Pentagon and Congress react to the unfolding welter of conflicting pressures by reducing training tempos, cutting purchases of spare parts, postponing maintenance, and stretching out production rates. Thus, action in the Pentagon and reaction in Congress reinforce each other, leading inevitably to reduced readiness for combat, slower rates of modernization, an aging inventory, and ultimately, a shrinking force size. Increasing the size of the defense budget may bring temporary relief, but it actually makes matters worse over the long term. When Congress buys into the increase, the bureaucracy in the Pentagon reciprocates by using the three assumptions to ratchet the game to a higher level, and the degenerative process starts anew.

The obvious way to cut the Gordian knot is to change the assumptions—make realistic projections of future Congressional appropriations, face up to the consequences of increasing technical complexity by properly financing the operating budget, and ruthlessly terminate the production of ineffective or low priority weapons in order to make enough money available in the investment budget to efficiently produce higher priority weapons. But as Mr. Cheney found out last year, when he tried to stop the production of the high-cost F-14 fighter and V-22 tilt rotor aircraft (minor cuts in terms of the total budget squeeze), it is difficult, if not impossible, to turn off production lines when complex weapons have subcontracts spread all over the country.

Workers need jobs, capitalists covet profits, and elected officials in Congress hunger for their votes. These are the implacable forces of democracy unleashed by Congress when it buys into the first year of a front-loaded plan. By designing overly complex weapons, then spreading subcontracts, jobs, and profits all over the country, the political engineers in the Defense Department deliberately magnify the power of these forces to punish Congress, should it subsequently try to reduce defense spending by terminating major procurement programs. The $60 billion B-2 Stealth Bomber illustrates the extent to which technical complexity and politics come together to preserve the flow of money. In July, 1989, when some members of Congress began to build a coalition aimed at canceling the B-2, Northrop Corporation, the B-2's prime contractor, retaliated by releasing data (which had previously been classified) showing that tens of thousands of jobs and hundreds of millions in profits were at risk in 46 states and 383 congressional districts. This brazen attempt at political blackmail prompted Ron Dellums, Chairman of the Research and Development Subcommittee of the House Armed Services Committee, to accuse the B-2's proponents of selling the program geographically. He said, "This thing starts to get shopped like a piece of meat."(11)

That meat had the sweet aroma of pork. Mr. Dellums was describing the after shocks of political blitzkrieg. The front loaders were the stormtroopers; they infiltrated the front by softening up resistance with a clever deception operation. The political engineers were the follow-on shock troops; they provided the brutal punch needed to paralyze opposition and expand the initial penetration into an unstoppable torrent. The strategic goal was a self-perpetuating patronage operation masquerading as a defense policy. In the next section, we will see how this blitz maneuvered Mr. Dellums and his colleagues in Congress to the edge of the abyss. Even if the House can convince the Senate to cap production of the B-2 at the irrational number of 15 aircraft, members of Congress will find it increasingly difficult to throttle back the big green spending machine. They are going to find themselves paralyzed by the imploding demands of frightened voters who are fighting to save their jobs, homes, and families, of frightened industrialists who are trying to safeguard their profits, and of frightened politicians and bureaucrats who are defending their power bases.

Let us now turn to the decisions and actions in the 1980s and see how the front loaders and political engineers stampeded Mr. Dellums and his colleagues into the political black hole.

IV. The Defense Power Games in Action (A):

Political Blitzkrieg

The spending surge of the early 1980s was a reaction to the frustrations of the 1970s. Recall how newspapers pounded the public with stories of equipment breakdowns, ships that could not steam and planes that could not fly, shortages of spare parts, low morale, and poorly trained crews. Beset by these problems, humiliated by defeat in Vietnam, captured by the exaggerated fear of an ever growing Soviet menace, the Defense Department saw itself as the beleaguered victim of arbitrary budget cuts imposed by a niggardly Congress—a victim of the so-called "Decade of Neglect." The pot boiled over in April 1980, when the Iranian rescue mission ended in a miasma of broken equipment and flaming wreckage. That disaster catalyzed the politics of defense spending. Ronald Reagan won the Presidency, in part, because he promised to end our national humiliation by "re-arming America." His relentless political momentum and the Pentagon's front-loading power game came together in the opening month of 1981, when the new director the Office of Management and Budget, David Stockman, thoughtlessly approved a plan for the largest peacetime spending spree in our nation's history.

Stockman's memoir, The Triumph of Politics, describes how Frank Carlucci, the wily new Deputy Secretary of Defense, duped him into approving an extravagant increase in the five-year plan. Ronald Reagan had campaigned on a platform that pledged to increase the defense budget by 5 percent per year after the effects of inflation were removed. Upon entering office in 1981, however, the new administration immediately declared an urgent need for a "get well" package and increased the previous year's budget (via the Fiscal Year 1981 Supplemental Budget Request) by 12 percent and amended the current budget request (for Fiscal Year 1982) to increase it by 15 percent over the 1981 supplement. All these increases are in inflation-adjusted terms.

At same time these near term increases were taking place, Stockman and Carlucci got into an argument over how much to increase the budget over the long term. Carlucci wanted the budget to grow at 8 or 9 percent per year. Stockman wanted to adhere to Reagan's promise of 5 percent, but he had a problem: the Carter administration left the Reagan team with a plan that matched their promise of 5 percent, and Reagan had said repeatedly that Carter was not spending enough. When Carlucci went on the offensive, asserting that an annual growth rate of 5 per cent would not buy the force expansion, modernization, and readiness that were needed, he maneuvered Stockman into the untenable position of appearing to defend the Carter budget. Doing his calculations late at night on a hand-held calculator, Stockman caved in, suggesting that they split the difference and settle for 7 percent. Carlucci accepted but said the 7 percent should be computed from 1982 after the "get well" package had been added to it. Stockman agreed on January 30, 1981; he had been in office just ten days.

In his haste to cut the deal, Stockman admits, he had forgotten about the size of the "get well" package. He was thinking in terms of increasing the 5 percent plan of the Carter Administration to 7 percent. By jumping off from the higher "get well" level in 1982, he agreed to add $245 billion when he thought he was adding 130 billion to the five-year plan (my calculations). This error resulted in a budget that would grow at 10 percent per year between 1980 and 1986, or double what Ronald Reagan promised during the election campaign.(12)

Nowhere in his story does Stockman suggest that he had the slightest idea of how the Pentagon would respond or what the money would buy. To Stockman, as to most Washington politicians, the crucial policy question was one of input, not output.

Stockman's incompetence gave the stormtroopers in the Pentagon the opening they needed, and they poured through the gap. Within three and a half months, the first draft of the 1983 to 1987 five-year plan was in the computers, and information on where the money was going began to seep out to lobbyists in industry, staffers on Capital Hill, and sympathetic columnists (known in the Pentagon as "wholly-owned subsidiaries") in the fourth estate. Table I summarizes these decisions. It also compares them to the last five-year plan completed by the Carter Administration and to the growth rates eventually appropriated by Congress. All the rates are adjusted to remove the effects of inflation.


Table I

Early Plans: Growth Rates Assumed by Decisions Made during Jan-May, 1981

Planned Increases:

Total Budget



Last Carter Plan (1981-85):








Reagan "Get Well" Plan (1981-1982):








1st Reagan Five-Year Plan (1983 thru 1987) :








Actual Increases:




Fiscal Years 1981 and 1982:




Fiscal Years 1983 thru 1987:




Note: "Operations" includes Operations & Maintenance and Military Personnel after the accrual for military retirement has been removed


The first two assumptions of the front-loading game are starkly evident in Table I. The planned increases were unprecedented: Between 1976 and 1980, the budget had grown by less than 2 percent per year. Over the long term, the median growth rate between 1951 and 1980 had been less than 1 percent per year. Moreover, in the previous thirty years, the budget had never grown for more than three years in succession before experiencing a real decline. By continuing the growth started in the Carter administration, the Stockman-Carlucci deal assumed eight consecutive years of unequaled rates of growth. The bias toward investment (Assumption 2) is equally stunning: Table I shows that the Defense Department intended to increase investment six times as fast as operations during the "get well" period, and once this was done, the new plan would compound this difference by increasing it three times as fast for the next five years. Note that the bias adopted by the Reagan Administration differed from the Carter Administration in scale but not in pattern. Rather than correcting the distorted funding allocations that caused the readiness problems of the 1970s, the table suggests that the front-loading game was being ratcheted up to a higher level.(13) Other evidence supports this conclusion. Table I shows that Congress appropriated most of the increases requested for Fiscal Years 1981 and 1982. Investment grew almost five times as fast as operations during the "get well" period, but these increases did not yield increased efficiency in production. In fact, data compiled by the Congressional Budget Office (CBO) shows that average unit costs increased dramatically during this period.(14)

Of 39 major weapon systems procured in Fiscal Years 1980 and 1982, 26 increased in cost, and of the 13 that decreased in cost, the bulk of the reductions came from programs that had first entered production in 1979 or 1980 (for example, the F-18 fighter and the M-2 Bradley Fighting Vehicle). Costs naturally decline in the first two or three years of production as rates increase and the one-time set-up costs are absorbed, but the actual declines experienced in 1981 and 1982 were much less than had been predicted as recently as 1980. So, even in the case where costs declined, actual costs were higher than predicted costs. The CBO data also shows that the Pentagon packed 31 additional programs into its procurement market basket in 1981 and 1982. Thus, by the end of the "get well" period, a larger variety of higher cost weapons were positioned to enter the front end of the new five-year plan.

That plan predicted costs would immediately begin to decline, and production rates would accelerate. Fighter and attack aircraft procured by the Air Force and Navy illustrate this point. Between 1980 and 1983, the average cost increased by over 75% in inflation-adjusted dollars and total production plummeted from 452 to 299 airplanes per year. The new five-year plan then predicted an immediate reversal, with future costs dropping by 25% and total production increasing to 510 per year by 1987. More generally, the CBO data shows that (for a sample size of 69 weapons programs) the 1983 to 1987 plan predicted costs would decrease for 57 programs (the median decrease being 30% in constant dollars), and production rates would increase for 57 programs as well. It seems fair to conclude that the front-loaders used the euphoria of the "get well" period to ratchet up the cost base from which they then predicted new cost declines (that is, the first part of Assumption 3 was loaded in from a higher cost base).(15)

So, notwithstanding the adoption of the much higher cost base during the "get well" period, unprecedented budget projections, a heavy bias toward investment, and new predictions of declining unit costs permitted planners to pack the front end of the 1983 to 1987 plan with a larger variety of more complex weapons while at the same time predicting that these decisions would result in a larger and newer force by the late 1980s. There was, however, a crucial asymmetry in these decisions—the plans for the future were hopes and dreams, the higher costs were reality. By ratcheting up the front-loading game during the "get well" period, Defense Department planners ensured that costs would grow faster than budgets for the remainder of the decade.

Before describing how the new plan unraveled, it is appropriate to pause briefly and ask a question—one that leads us into the labyrinth of political motivation: To what extent were these early decisions a deliberate attempt to stampede Congress into making commitments it did not understand as opposed to being the product of a bureaucracy running mindlessly on its own momentum? In December 1982, at a seminar in the Brookings Institution, no less an authority than the Secretary of the Navy, John Lehman, affirmed the premeditated character of the Pentagon-Congressional power game. Responding to William Kaufmann, a widely-published critic of the naval build up, Lehman implied there was nothing Kaufmann or anyone else could do about it. He said that it was "too late" to stop the buildup to a 600 ship navy. "We've already accomplished it," he continued, "because we front-loaded (emphasis added) the budget."(16)

Of course, the 600 ship navy did not exist in December 1982. Lehman was referring to the fact that he and his colleagues used the political honeymoon of the "get well" period to plant the seed money in the shipbuilding budget for a large number of high complexity ships. His statement suggests that he believed this money would create enough constituent pressure on Congress to lock in support for increasing defense appropriations over the long term. As far as the motive behind the front loading operation is concerned, Mr. Lehman's braggadocio speaks for itself—front loading is nothing less than the first part of a comprehensive extortion strategy.

Events, however, soon proved Lehman wrong. While members of Congress often pander to their constituents like prostitutes to their tricks, they are not slaves. Besides, Congress must respond to the competing desires of many different constituents. The defense spendup was being financed by debt and by the assumption that Congress would reduce funding in other programs—a strategy guaranteed to mobilize increasing opposition over the long term

Figure 3

Indeed, the new plan began to unravel as soon as the ink was dry. Although Congress doubled the budget (in current dollars) between 1980 and 1987, Table I shows that the budget grew more slowly and with a different allocation between investment and operations. Figure 3 puts the mismatch between plans and reality into a long-term perspective. It compares the last 9 five-year spending plans (the thin lines) to the real world of Congressional appropriations (the thick line). Note that the effects of inflation are included in Figure 3.(17)

The Reagan buildup (including Lehman's 600 ship Navy) was based on the 1983 to 1987 plan, which is portrayed by the thin line furthest to the left. That plan presumed appropriations would rise to $401 billion by 1987. Actual appropriations, however, increased to $286 billion, resulting in a mismatch between plans and reality of $305 billion over the five years. Note that the front-loading game did not stop with the meltdown of the 1983 to 1987 plan. Subsequent plans rose to even more fantastic heights. By constantly shifting its hopes and dreams further into the future, the Pentagon created a surreal economic world where decision makers (in the Pentagon and Congress) were free to ignore the effects of a $300 billion shortfall. For example, only one major investment program was canceled between 1983 and 1987 (the Sergeant York air defense gun), and it was canceled for performance reasons, not budget reasons. Furthermore, the Pentagon continued to plant seed money (which Congress joyfully appropriated) for new programs, taking on new commitments of hundreds of billions of dollars—programs such as Star Wars, the Midgetman ballistic missile, the B-2 Stealth bomber, the C-17 transport, the Advanced Tactical Fighter, the A-12 attack fighter, and the Army's new light helicopter.

Given this kind of Alice-in-Wonderland behavior, the remainder of the decade became quite predictable: We did not increase operating tempos(18), we stretched out procurement rates and modernized only a part of the force while retaining older weapons to maintain the size of the force, and by the late 1980s, we began to cut purchases of spare parts(19), defer maintenance, and reduce the size of our forces. The next section examines how the seeds planted in the early 1980s blossomed into the bitter fruits we will eat in the 1990s.

V. The Defense Power Games in Action (B):


Perhaps the sharpest example of how the front-loaded plans of the early 1980s degenerated into what I will call the strategy of "selective modernization" is the case of land-based strategic missiles. In 1980, nuclear strategists believed that our 1054 silo-based missiles (Minuteman IIs, IIIs, and Titans), capable of delivering 2054 nuclear warheads, were vulnerable to a Soviet surprise attack. Their solution was to eliminate the so-called "window of vulnerability" by making land-based missiles hard to hit—we would replace all silo-based missiles with 200 mobile M-X missiles, capable of delivering a total of 2000 warheads with greater accuracy. But rising costs, including the unforeseen cost of environmental disruption, made a complete replacement of the Minutemen and Titans far too expensive, so we deployed only 50 M-Xs (retiring 50 Minuteman IIIs and 54 Titans). To add insult to injury, we placed the M-Xs in silos. Now if we accept the logic of nuclear theology, placing a larger number of more accurate warheads in fewer silos increased our vulnerability to a surprise attack. So, after spending $9 billion, 1000 US missiles are still in silos (950 of which are aging Minuteman missiles), and because the nuclear gurus still believe silos are vulnerable, we now have two mobile missile programs in development (the Midgetman and the M-X Rail Garrison), a massive strategic defense program (Star Wars) more oriented toward defending missile fields than cities, and large strategic budgets as far as the eye can see. This is not the outcome that was presented to the American taxpayer in the early 1980s.


Figure 4

Aircraft purchased by the Navy and the Air Force illustrate more generally how front loading and political engineering create a quagmire of spiraling costs that leads ultimately to aging and shrinking forces and the requirement for ever increasing defense budgets. Figure 4 is a depiction of the total number of helicopters and airplanes purchased between 1950 and 1990 by the Navy and the Air Force.



Figure 5

Figure 5 relates the quantities in Figure 4 to the money made available by Congress—the bars represent the sum of the Navy and Air Force budgets used to buy the aircraft; the budget totals are measured in billions of dollars on the left vertical scale. The rising line merely divides the annual budget by the number of airplanes procured in each year. Its index (budget dollars per airplane procured) is in millions of dollars and is measured on the right vertical scale of Figure 5. Note that the effects of inflation have been removed from Figure 5.

"Budget dollars per aircraft procured" is a general index of the cost pressure within the Pentagon's market basket, and it must be interpreted carefully. The aircraft procurement budget buys more than new airplanes; it also buys modifications to existing airplanes, as well as support equipment and spare parts. The index increases when the cost of existing airplanes increase (pure cost growth, caused by production stretch-outs, simple waste, etc.), when more expensive models of existing airplanes are introduced (for example, when F-15Es replace F-15Cs), and when the market basket is shifted to a more expensive mix of airplanes (for example, when low cost airplanes like the A-10 are dropped and high cost airplanes like the B-1, C-5, and B-2 are added). The measure also increases if budget dollars are shifted from the procurement of new airplanes to the modification of existing airplanes (the other categories are relatively minor).

Figure 5 shows that costs have been growing faster than budgets for the last forty years. More importantly, it shows that increasing the budget magnifies the divergence—the index skyrocketed to record levels when the budget increased sharply between 1980 and 1985. The front-loading operation in the early 1980s permitted the Pentagon to add very expensive new aircraft (particularly the B-1 and C-5B) and more expensive models of existing aircraft (F-16C, F-15E, F-18D, etc.) to its shopping list. Then, the unfolding mismatch between plans and reality (Figure 3) increased costs further by forcing the Pentagon to postpone and eventually cancel the build-up to high rates of production, (for example, the F-15, F-18, AV-8B, AWACS, A-6, and the F-14). There is also evidence that waste, fraud, and abuse increased in the early 1980s: the spare parts horror stories, exemplified by the $600 toilet seat and the $450 hammer, the widespread illegal trafficking in secret government budget documents now coming to light in the Ill Wind corruption scandal, and the acceptance of equipment that did not meet its performance specifications, the most spectacular being the defensive electronics in the B-1 bomber, all suggest a breakdown in management control that allowed contractors to ratchet up their prices. Note that the completion of the B-1 and C-5, which accounts for the sharp drop in the index during mid decade, did not change the general pattern; between 1980 and 1990, the index of cost pressure grew by 77 percent, while the budget grew by only 31 percent.

When costs grow faster than budgets, selective modernization becomes inevitable. Figure 5 shows that the combined aircraft procurement budget leaped from $15 billion in 1976 to a peak of $43 billion in 1985 (an increase of 187 percent) before dropping to $26 billion in 1989. The total number of airplanes and helicopters in the Navy and Air Force inventories, however, increased by only six-tenths of one percent (from 15,017 to 15,111) during this fourteen year period. Furthermore, the average age of this inventory increased by 49 percent (from 10.5 years to 15.6 years). Despite an immense infusion of investment capital, skyrocketing costs made it impossible to buy enough new airplanes to modernize all mission areas, and so some mission areas had to be ignored—for example, tactical airlift, strategic air refueling, and basic flight training in the Air Force.(20)

It therefore became necessary to divert money into modifications to extend the life of the older airplanes—a short sighted expedient which took money away from purchases of new airplanes while delaying and magnifying ultimate modernization burden. For example, we began an expensive program to replace the engines on KC-135 tanker aircraft (a militarized version of the 1950s vintage Boeing 707) rather than buying new tankers.

There were variations in the pattern of selective modernization. The size of the Navy fleet increased by about 17 percent between 1976 and 1989, but the average age of a Navy ship also increased (by about 30 percent). While the average age of armored vehicles in the Army decreased, the number of active duty heavy maneuver battalions decreased as well.

The dam burst in February 1988. Frank Carlucci, the Secretary of Defense, began the standard gambit of shifting the blame for the coming cutbacks to Congress. His annual report to Congress was peppered with claims that "budget constraints" made it necessary to begin an across-the-board retrenchment. Carlucci announced that the Navy would prematurely retire 16 frigates and cutback its sea-based air forces by one carrier air wing; the Army would cut the size of its helicopter fleet by 20 percent and reduce the number of its light infantry maneuver battalions; and the Air Force would reduce its tactical fighter forces from 38 to 35 wings.(21)

By claiming that these decisions were caused by "budget constraints," Carlucci insinuated that unexpected cutbacks by a capricious Congress were causing the retrenchment—an accusation that is partially correct (Congress can be capricious) but conveniently dodges the question of the Pentagon's contribution to the developing debacle. Given the chronic mismatch between plans and reality depicted in Figure 3, the idea that budget constraints could be unexpected in 1988 is preposterous.

Carlucci's cutbacks were the first penalty payment for the excesses perpetrated in the early 1980s. The much larger cutbacks proposed by the military services last May (7 more Air Force fighter wings, another carrier air wing, 52 additional ships, and 5 Army divisions, etc.) are another payment. There will be more payments as the 1990s unfold. As I pointed out in the introduction, the most recent round of cutbacks merely brings the five-year plan down to a level that would have been a realistic projection of future budgets, had the Cold War continued. The collapse of the Warsaw Pact threat and the growing need to bring our chronic fiscal imbalances under control suggest that the long-term assumptions of the current plan are still excessively optimistic. Saddam Hussein's gambit in Kuwait may confuse this issue but it does not change it. Put another way, Assumption 1 of the front-loading game is alive and well.

Moreover, the bias toward investment (Assumption 2 of the front-loading game) is not only alive and well, it appears to be increasing. The composition of the cutbacks reported by the New York Times and the Washington Post in May (footnote 4) indicates that our civilian and military leaders have chosen to sacrifice forces-in-being (that is, they will slash the operations budget) to save the weapons programs in the investment budget. The example of aircraft procured for the Air Force and Navy, once again, illustrates this point. At the same time the generals and admirals were putting together plans to slash the size of their air forces, the senior civilian leadership was conducting a review of the six aircraft development programs that are the key to modernizing our aircraft inventories through first decade of the next century. On April 26, Secretary Cheney announced the results of the Major Aircraft Review. He said "changed world conditions" and "increasingly tight fiscal constraints" were forcing the Pentagon to cut back all six aircraft modernization programs. The contrast between these decisions and the slash-and-burn approach to force planning could not have been starker. They are worth describing in detail.

The new plan cuts the peak production rate of the B-2 stealth bomber by 50% (from 24 to 12 per year) and reduces the total number procured by 43% (from 132 to 75). If there are no more engineering problems, this change will reduce the total program cost by only 19%, saving the taxpayer $14 billion but ratcheting up unit costs by 43% to $815 million per copy. Similarly, we will cut the peak production rate for the C-17 transport by 17%, the total number procured by 43%, and if all goes well, reduce the total cost by 28% while ratcheting up unit costs by 25%. The Defense Department announced substantial delays, stretch-outs, or cutbacks in the remaining four programs (all of which are high-cost, technically risky "stealth" designs) but declined to identify the total cost savings and the unit cost increases caused by these changes. These decisions will (1) reduce the peak production rate for the Navy's A-12 attack aircraft by 25% and cut total production by 28%; (2) extend the development time for the Air Force's Advanced Tactical Aircraft, keeping the program alive but shifting the production decision from 1993 until some time after 1997; (3) reduce the peak production rate of the Air Force's Advanced Tactical Fighter by 33%, delay the start of production by two years, and keep the total buy constant at 750 aircraft; and (4), delay production of the Navy's Advanced Tactical Fighter by two years, reduce total production by 13%, but keep the peak production rate at 48 planes per year.(22)

Not one program was canceled! While the force level cutbacks are hardball decisions for reducing the operating budget in the near term, the decisions made in the Major Aircraft Review are softballs. They will reduce budgets in the short term but lead to much higher procurement costs over the long term. If these decisions are carried out, the mismatch between cost growth and budget growth (Figure 5) in the 1990s will make the explosion of the early 1980s look like a minor blip on a steeply rising curve. The result of this kind of decision making should be clear by now: plummeting production rates mean that our forces will continue to shrink and get older as we move into the next century. The hardball cuts in the operations budget and the softball cuts in the investment budget prove that our leadership has forgotten the most important lesson of the 1970s—the front-loading power game also works while budgets are declining, and its effects are devastating.

While Defense Secretary Cheney deserves credit for scaling back the irrational budget projections of his predecessors, current projections are still too optimistic.(23)

Moreover, the recent round of force reductions and procurement stretch-outs indicate that the excesses of the Pentagon's power games are spinning out of control. If Cheney does not come to grips with this problem, and soon, the carnage will be far worse in the 1990s than it was in the 1970s. During the 1970s, we were able to live off of a large cushion of weapon inventories and spare parts left over from the Vietnam War, and so the damage caused by front-loading the budget during the early 1970s did not become intolerable until the late 1970s. No such cushion exists as we enter the 1990s. Rather than using the unprecedented budget increases of the early 1980s to re-build the cushion, the Pentagon and Congress chose to plant the seed money for a new generation of complex weapons that will be, by far, the most expensive weapons in US history. The political engineers then spread thousands of subcontracts all over the country, adding almost 1.4 million jobs to the industry payroll between 1980 and 1989. If procurement budgets continue to decline in the 1990s, and the Defense Department insists on buying all of these weapons, the mismatch between cost growth and budget growth will explode, causing a collapse in production rates as well as force structure. This conclusion closes the circle. The real reason there won't be a peace dividend is that we will have to keep defense budgets at cold-war levels and raise taxes to bail out the defense contractors. No doubt, the bail-out will presented in some kind of positive light, perhaps as a plan to maintain the competitiveness of the defense industrial base during a period of changing world conditions. The political engineers have sprung the trap.

VI. Reform:

Back to Fundamentals

How can we extricate our nation from this quagmire? If there is one thing the great spending experiment of the 1980s proved, it is that increasing the defense budget is no solution. Feeding the monster increases its voracity. The defense power games are not a management disorder; sterile management solutions such as blue ribbon panels, acquisition czars, lists of initiatives and reforms, and organizational streamlining may palliate the voters, but they do nothing to weaken the monster. The monster is political, and there is only one solution.

We must slay the monster.

We must acknowledge and confront the threat posed by the defense power games. Front loading and political engineering undermine the philosophical foundation of our system of government. That foundation is James Madison's ideal of checks and balances: the idea of setting up a compound republic with overlapping zones of power, each competing with others to prevent the rise of a single source of tyrannical power, the idea of pitting faction against faction and interest against interest. Foremost among these checks and balances is the constitutional stricture that the executive power vested in the President shall be counterbalanced by the restraining power of the purse vested in Congress.(24)

Front loading and political engineering aim to subvert the power of the purse, and to the extent that they succeed, they transfer power from Congress to the President, thereby corrupting the balance of power the Founding Fathers established in the Constitution.

The genius of the Founding Fathers was that they built a government that assumed men were not angels, to paraphrase Mr. Madison. They assumed men and women in power will always try to accrete more power. The Iran-Contra Affair, a conspiracy to neutralize Congress's power of the purse, is precisely the kind of power play the Founding Fathers anticipated when they designed the adversarial power-sharing arrangements into the Constitution. In the case of Iran-Contra, the checks and balances worked, although the machinery seemed rusty. The defense power games are a much more sophisticated and dangerous strategy for accreting power. They are dangerous because they generally conform to the letter, if not the intent, of the Constitution, they insinuate themselves slowly and unobtrusively into our nation's political fabric, buying the voters out from under Congress bit by bit, and they mesmerize Congress into enthusiastically participating in a game that leads to its own ruin. Their main weapon is the second oldest corrupting agent known to man. Money.

The defense power games are stratagems for increasing the flow of money; they are about transferring money from the taxpayer to a central bureaucracy that subsequently disburses the money to a socialistic industry, even if the transfer sacrifices the capabilities of our military forces. That is what slashing the operations budget to save the investment budget is all about. Costs grow faster than budgets and technologies grow more baroque because they are subsidized by a politically motivated central planning authority. Consequently, economic benefits flow to some sectors of the economy and nation at the expense of other sectors. This is the all too familiar politics of welfare and dependence. The base political nature of these games has been hidden, I think, by an alliance of primitive fear and secrecy engendered by the Cold War, a suffocating atmosphere of intellectual intimidation engendered by ever more complex technologies, and the coalition of private and political interests that benefits from the flow of funds.

If the monster thrives in darkness and stale air, the first step in killing it is to eviscerate it by throwing back the curtains and opening the windows; glasnost is the precondition for perestroika. The end of the Cold War provides a unique opportunity to increase the flow of information to the people without jeopardizing our nation's defenses. On the contrary, openness will make the government more accountable to the people; it will strengthen the Constitution and the institutions of our compound republic.

We should go back to the fundamentals that have served our nation well. The idea of checks and balances in a system of shared power can be used to rein in the excesses of the defense power games. The first step in raising the political debate out of the depths of the porkbarrel is to generate high-quality information. Here is a list of four corrective actions Congress could take today, if it wants to get serious about exorcising the destructive effects of front loading and political engineering:

First, every witness appearing before Congress, from the Secretary of Defense down, should be routinely put under oath before testifying on any subject. Swearing in a witness sets the proper legal tone—it tells the witness he could go to jail if he lies or willfully distorts the truth, and it reminds the members of Congress that adversarial hearings are not political circuses, they are the essence of the power-sharing arrangements set forth in the Constitution. Today, with the exception of rare public spectacles such as the Iran-Contra hearings, witnesses are almost never put under oath. When Congress does not put witnesses under oath, it makes it easier for the Pentagon to control the flow of information. Psychologically, Congress gives the bureaucracy more power to pressure lower level witnesses into suppressing their differences of opinion, and it provides less cover to protect witnesses, should they choose the path of integrity and fealty to the Constitution over loyalty to the bureaucracy. Some may argue that it is insulting to put high public officials under oath. This is a red herring. By making it a requirement that everybody be placed under oath, we are merely establishing a way of doing business that emphasizes the Constitutional character of Congressional proceedings.

Second, every two years, Congress should pass a joint resolution describing its assessment of the threats facing the United States. This resolution should be arrived at in an open, deliberately-paced, legislative process managed by the foreign relations committee of each house. For the entire period of the Cold War, the executive branch had a monopoly on threat analysis. The alliance of secrecy, intellectual intimidation, and fear fed the defense power games, making it easier to exaggerate Soviet capabilities in the interest of increasing the defense budget. While the bomber gap in the 1950s, the missile gap in 1960s, and the window of vulnerability in the 1970s are either outright falsehoods or absurdities in retrospect, they seemed real to the public and Congress at the time, and they generated large cash flows.(25

Although the Soviet Union bears the responsibility for starting the Cold War, it is also probable that our excesses fed their paranoia, magnified genuine differences, and heightened the tensions of the Cold War. Now that the Cold War is over, it is especially important that Congress have a countervailing voice to offset the executive monopoly. Some may argue that this idea is simply not practical, that Congress is incapable of making such an assessment.

Our nation owes a debt of gratitude to Senator Sam Nunn, Chairman of the Senate Armed Services Committee, and his counterpart in the House of Representatives, Les Aspin, for taking the initiative to prove that Congress can have a voice in threat analysis, if it chooses to do so. On March 29, Senator Nunn, in a speech on the Senate floor, outlined his analysis of the threats facing the United States. His conclusions, which went so far as to specify such technicalities as warning times in Europe, were derived from information provided to him in special Congressional hearings by executive department witnesses, independent specialists, and personal contacts—sources that are open to all members of Congress. With minor dressing up, Nunn's speech could serve as a strawman for a joint resolution expressing the sense of the Congress. Aspin broke new ground by inviting Soviet officials to appear as witnesses in open hearings before his committee. While Nunn and Aspin have taken the initiative and shown what is possible, I think that their work should be institutionalized under the auspices of the foreign affairs committees. The Armed Services committees are responsible for authorizing defense budgets. The ideal of checks and balances suggests that we should set up power-sharing arrangements within Congress as well as between Congress and the President. Separating the threat analysis function (which generates budget requirements) from the budget approval function in Congress would be such an arrangement.

Third, Congress should declassify the Five Year Defense Plan and insist that it be submitted in its present form to Congress and the American public each year as part of the President's annual budget message. Congress has the power to amend the US Code to make this a legal requirement. Article I, Section 8 of the Constitution grants Congress the power "To make Rules for the Government and Regulation of the land and naval Forces."

We have seen that the defense power games are intimately connected to the structure of the five-year plan. The Pentagon seduces Congress into approving front-loaded budgets, uses political engineering to hook a significant part of Congress and the country on the narcotic of defense spending, then blames Congress when it chokes on the final bill. Congress, desperate to save jobs and votes, discredits itself by putting the dictates of the porkbarrel ahead of the needs of national defense as it cuts back the Pentagon's budget. The Pentagon closes the trap by accusing Congress of meddling and micro management, and Congress becomes the clown in the theater of the absurd. This minuet thrives on secrecy. Declassifying the five-year plan is the single largest step that can be taken to expose the destructive effects of the defense power games.

Some may argue that, notwithstanding the end of the Cold War, declassifying the five-year plan will hurt national security; it will give our adversaries too much information. The following counter-arguments can be made: So much information is now delivered piecemeal to Congress in unclassified form, that competent foreign intelligence agencies, with access to large analytical resources, are already synthesizing accurate assessments of the five-year plan. Moreover, we must assume foreign agents already have the plan—if industrial spies working for defense contractors can obtain the plan illegally, it is absurd to think that the KGB can not. Finally, it does not matter whether foreign agents obtain the plan. The arguments presented in this paper show that it can not be executed. The plan is so infected by the defense power games that it is meaningless. Keeping it secret hurts our defenses because the only people kept in the dark are our own citizens.

On the other hand, powerful arguments can be made in favor of declassification: The competing players in the Washington policy-making establishment, public and private, advocates and watchdogs, would all be singing from the same sheet of music. Ideas would compete more freely, more alternatives would surface, and the policy debate would naturally rise to a higher intellectual plane. There are several reasons why this would occur.

Open books would reveal the accounting chicanery the Pentagon uses to avoid hard decisions. In January 1987, for example, the budget total projected in the Pentagon's five-year plan (for Fiscal Years 1988 to 1992) exceeded the total presented by the President in his budget message to Congress by about $80 billion. Two years later, the 1990 to 1994 plan coincided with the totals in the President's budget message, but only with the inclusion of a "negative planning wedge." (Negative money is explained in the reference cited in the third footnote.) The Pentagon would not dare to use these unethical accounting tricks if the five-year plan was a public document. It is impossible to have a serious policy debate when the different players are using different sets of books.

Open books would reveal creeping policy changes before Congress and the American people are presented with a fait accompli. The policy decision to apply special access security classifications to weapons development programs (in addition to their traditional use in intelligence-related activities) led to the explosive growth in "black programs" during the 1980s. This momentous decision greatly decreased the accountability of the Pentagon. If the five-year plan had been a public document, the details of the "black programs" would not have been revealed (they are entered as code words, and in fact, these code words are routinely submitted to Congress in unclassified budget documents), but the magnitude of the intended budget shift would have been visible and open to debate before it occurred. Surely, this would have improved the quality of the defense debate in the early 1980s.

Open books would create a more vigorous and lively defense debate. The personal agendas and competitive instincts of competent analysts in private think tanks and watchdog organizations, as well as government analysts in oversight agencies, would drive them to uncover and force a debate over the structural assumptions and the policy decisions contained in the five-year plan. If front loading and political engineering continued, it would be made obvious to all concerned. At least our political leaders would be forced to tell the American people about their policy decision to put industry and voters on the path toward welfare and dependence.

Open books, scrubbed and squeaky clean, would provide a common intellectual foundation from which policy analysts could debate the direction our military should take as we move beyond the Cold War. In the past, the policy gurus in the defense intelligentsia have proposed major shifts in policy without access to the information which described the options. Our nation faces the most profound defense decisions in a generation. We also face enormous domestic problems which must be addressed if we are to remain competitive in the emerging world. We will pay an incalculable price if we move in the wrong direction. We can no longer afford the luxury of a national defense debate based on data-free analyses.

Open books would also have the administrative benefit of reducing the onerous reporting requirements Congress and the media place on the Pentagon. Coupled with the information contained in the Defense Secretary's annual report to Congress, an unclassified five-year plan would eliminate the need for hundreds of documents the Pentagon routinely prepares to explain its position to Congress and the public.

Fourth, the last corrective action would introduce a new player and a legal framework into the defense debate. Information is power in Washington. One reason the defense power games thrive within the Pentagon, as well as between the Pentagon and Congress, is that the bureaucracy has the power to manipulate the content and flow of information on a vast scale. By the time information reaches the Secretary of Defense or Congress, it has been massaged by so many people, with so many different agendas, that its content often only bears a coincidental resemblance to the message the original sender intended to transmit. Bureaucratic scrubbing systematically squelches bad news and dissenting arguments.(26)

Today there is no offsetting mechanism for checking the bureaucracy's ability to stifle controversy and limit the terms of debate. Article II, Section 8 of the Constitution grants Congress the power to create an adversarial mechanism for providing the needed checks and balances. Congress should pass a law establishing a politically independent Defense Evaluation Board.

The Defense Evaluation Board would be a independent regulatory agency, modeled after the Federal Reserve Board. Housed in the Pentagon but separate from it, the Defense Evaluation Board would consist of five voting members, each appointed by the President for terms of ten years and confirmed by Congress—a chairman, vice-chairman, and three directors (for operational test and evaluation, program analysis, and cost analysis). The law would prohibit members from accepting any future employment, consulting, or lobbying relationships with any firm doing business, either directly or through a subsidiary, with the Defense Department. In return for this prior restraint, board members would retire on full salary after their tenure expires. The board would be supported by a professional staff of no more than 100 highly qualified, career civil servants. These men and women are going to be the skunks at the garden party, and they must be protected from political and bureaucratic harassment.

The board would use legal certification procedures to ensure that the current state of knowledge (what is known, what is not known, what is presumed, where the uncertainties exist, and what alternatives are available) is accurately presented to decision makers before they commit themselves to an irreversible course of action. It would perform three regulatory functions, each aimed at improving the quality of information flowing to the Secretary of Defense and Congress. The first function would be to certify whether or not the data contained in the five-year plan is distorted by the pernicious effects of bureaucratic gaming and inter-service logrolling. Each year, the board would produce an unclassified report documenting the structural assumptions and policy decisions embodied in the five-year plan. It would compare the new plan to the old plan and explain why it changed. The report would also determine whether or not the five-year plan provides the resources needed to execute the war plans produced by the Joint Chiefs of Staff. The board would be empowered to identify deficiencies and recommend changes to the plan. The board would submit the report to the President, the Secretary of Defense, the Governmental Affairs Committee in the Senate, and the Government Operations Committee in the House of Representatives in February of each year.

The second regulatory function would ensure that the information used to support decisions to proceed with the development and procurement of major weapon systems accurately portrays what is known and not known at the time of the decision-making milestone. Specifically, the board would be required to legally certify that independent cost estimates were arrived at independently, used the best available information, and identified where the estimates were uncertain and what the likely impact of the uncertainties would be. It would be required to similarly certify the results of development tests used to justify the decision to proceed to the next phase of a weapon's development. And finally, prior to a decision to start production, it would certify that the weapon met or exceeded all its specifications in a realistic operational test run by uniformed military personnel randomly selected from combat units. No weapon would enter production until the board certified the operational test. The board would be empowered to subpoena information and witnesses from government agencies and private contractors, and it would be empowered to obtain testimony under oath. These certifications, along with any minority views, would be presented to the Secretary of Defense and Congress prior to each decision-making milestone. Periodically, all certifications would be transmitted to the Archives of the United States for permanent preservation and storage.

The third function would provide the Secretary of Defense and Congress with the power to obtain independent views and analyses on major policy issues. Senators, working through the Government Affairs Committee, and Congressmen, working through the Government Operations Committee, could request special studies on any defense related matter. The Secretary of Defense could make similar requests by directly communicating with the Chairman of the board. A tripartite consultative process, involving the Chairman, Secretary of Defense, and the Chairmen of the Congressional committees would deal with the problem of selecting the most important studies to pursue, should workload become an issue. The board would also be empowered to initiate any studies it deemed to be important.

No doubt, some will argue that the idea of an independent regulatory board will add another layer of review and oversight to the oppressive weight of bureaucracy, it will interfere with efficient decision making, and it will needlessly slow the process by clogging the bureaucratic arteries. It will also be argued that effective independent test and evaluation, cost analysis, and program analysis functions already exist within the Pentagon, and therefore, the Defense Evaluation Board would duplicate their efforts.

These arguments are technically correct, but they are distractions aimed at shifting the terms of debate from a political premise to an efficiency premise. The B-1 bomber and the M-X missile, two of our most important strategic programs, illustrate the relationship between decision-making efficiency and raw political power. By every standard of measure, both weapons flew through their decision-making milestones during the 1980s, production was initiated before testing was completed, and all dissenting views were either suppressed or ignored. Notwithstanding this well-greased, bureaucratic efficiency, both programs produced defective products. In the case of the B-1, the electronic jammers, sold as being essential to the bomber's ability to penetrate Soviet defenses, did not meet operational requirements. In the case of the M-X, defective guidance systems were delivered late, and there were allegations of falsified test results. Yet both weapons passed acceptance tests and were declared operational. The internal, self-imposed, checks and balances did not work. The Founding Fathers understood this; that is why they devised the system of shared power.

Had an effective Defense Evaluation Board existed when the B-1 and M-X were being rammed through the bureaucracy, it could not have prevented the Secretary of Defense from approving these defective programs, but the independent power of the legal certification authority would have forced the Secretary and Congress to acknowledge that defects existed before they made their decisions. With the end of the Cold War upon us, the case for high-speed, "buy-before-you-fly" decision making can not be justified on military grounds. Surely, a little more time delay would be a small price to pay for ensuring that the Secretary of Defense and Congress receive the highest quality information. Who knows, perhaps they do not want to buy weapons that do not work or approve front-loaded plans that can not be executed.

The American people will continue to have a shrinking military without the economic benefits of a peace dividend until the Pentagon and Congress extricate themselves from the quagmire created by the defense power games. In The Imperial Presidency, Arthur Schlesinger made a plea for the restoration of comity between the different branches of government when he said, "The great powers of the American government resided in the area of joint possession. They therefore demanded active self-assertion of each branch, but always within a pervading frame of good faith, mutual respect, and self-restraint." That frame does not exist. Whether intentionally or not, the defense power games defile the ideals of good faith, mutual respect, and self-restraint. The Pentagon will not change the way it does business. Why should it? It is winning. Article II, Section 8 of the Constitution grants Congress the power put the necessary checks and balances in place. But when men are not angels, active self-assertion is not a bloodless academic concept, it means taking risks. If Congress wants the best military money can buy, manned with competently trained troops and equipped with weapons that work, and a peace dividend, it must have the courage and vision to exercise its constitutional responsibilities. Better information and higher moral character are the necessary and sufficient conditions for change. The soldiers at the pointy end of the spear and the American taxpayers deserve no less from their leaders on both sides of the Potomac.


1The cost of the first 100,000 troops comes from Stephen Dagget, "Costs of the Vietnam conflict compared to costs of current US military deployments in and around the Arabian Peninsula," September 10, 1990, a Congressional Research Service letter report prepared for Hon. Andy Ireland (R-Fla).

2 Hobart Rowen, "The Gulf Crisis: No Excuse," Washington Post, September 6, 1990, page A27. I do not know if Senator Nunn based his $50 billion estimate on DoD's September announcement, or if he based it on advance knowledge of the subsequent troop commitments.

3 The reduction of $167 billion illustrates how budgets drive strategy. It was part of the President's budget message to Congress, and presumably its effect was even incorporated in the President's estimate of future deficits, but it was not accompanied by the production of a new five-year plan detailing which programs would be cut back. Without this breakdown, it is impossible to determine how the program changes affect our defense strategy, or how the spending changes affect the deficit. For a discussion of this chronic problem, see my essay "A Defense Strategy That Works," Proceedings, US Naval Institute, January 1990.

4 Michael Gordon, "Military Services Propose Slashes in Existing Forces," New York Times, May 12, 1990, Page 1. See also Patrick Tyler, "Military Chiefs Detail Plans to Cut Troops, Weapons," Washington Post,. May 12, 1990, Page 1.

5 This estimate is derived from data contained in DoD's Budget Status: Fiscal Years 1990-94 budget Reduction Decisions Still Pending. US General Accounting Office. February 1990 (GAO/NSIAD-90-125BR). It is higher than the number released at the Defense Secretary's press conference in April 1989 because it assumes the more realistic inflation rates adopted by DoD in December 1989.

6 Assumption 2 depends in part on Assumption 3. During the 1970s and the 1980s, DoD's five-year plans routinely predicted the following outcomes by the fifth year: slightly larger forces equipped with newer, more complex weapons, and higher levels of combat readiness (including increased training tempos). If, at the same time, the operating budget is to grow more slowly than the total budget, the new equipment entering the force needs to be more reliable and easier to maintain than the equipment being replaced.

7 The defense department uses the theory of the learning curve to predict the future costs of a weapon's production. This theory predicts that unit cost decreases exponentially (at a decreasing rate) as the total quantity produced increases, even if the rate of production remains constant. When used as a budget planning device, this theory has a history of grossly underestimating the future costs of a weapon's production (see my book: Defense Facts of Life: The Plans/Reality Mismatch, Westview Press, 1985, pages 131-168 and the Appendix).

8Even when the new equipment is more reliable than the equipment being replaced, it is usually less reliable than it was predicted to be during the early stages of development. Although the new equipment may be easier to maintain at the organizational level of maintenance (e.g., on the flight line), the increased use of "black boxes" (equipment that is removed and replaced at the organizational level but repaired at the depots) makes it more dependent on high cost depot maintenance. Moreover, the increased use of high-value "black boxes" increases the costs of managing the flow of spare parts. Finally "black box" technology increases the demand for highly skilled, contract-maintenance labor. The Army's Apache helicopter illustrates all of these points: The General Accounting Office recently reported that key parts were breaking down much faster than predicted, that crews were cannibalizing each other's aircraft so often, it became necessary to install padlocks on the parked helicopters, and that during a recent exercise in Germany, employees of the prime contractor, McDonnel Douglas Corp., were forced to deliver spare parts to maintenance teams by driving around the countryside.

9The buy-in is an unrealistically low cost estimate made to get a program approved; it is an obvious front-loading gambit. Costs also increase when designs are changed. As equipment becomes more complex, technical difficulties multiply, unpredictable problems crop up more frequently, and designs consequently become less stable. Contractors have economic incentives to change designs because the changes, when accepted by the government, can be used to loosen contract provisions such as cost ceilings. In the defense business, using design changes to hide cost overruns is known as "contract nourishment."

10 The stretch-out of the F-15 fighter illustrates this point. The FY 1983 to 1987 five year plan predicted production would increase from 42 per year in 1983 to 96 per year in 1987 (for a five-year buy of 390 F-15s) and that unit costs (after removing the effects of inflation) would decrease by 26%. Congress, however, only appropriated funds for 207 F-15s during this period. In 1987, F-15s were being produced at 42 per year, and instead of being 26% lower, costs were 4% higher than in 1983. This stretch-out increased the total cost of the 390 F-15s, and it shifted the purchase of 183 (i.e., 390-207) F-15s into the "invisible" years beyond the fifth year of the five-year plan.

11 "Bombers Backers Cite Jobs," USA TODAY, July 26, 1989, Page 4. In the summer of 1990, the House of Representatives refused to appropriate production funding for additional B-2s in Fiscal Year 1991, while the Senate cut the administration's request in half. Even if the House wins in conference, which is by no means clear as of this writing, production will continue for the 15 B-2s already appropriated.

12 Stockman, David A., The Triumph of Politics, Harper & Row, 1986, pages 105-109.

13 I describe how the bias toward investment led to chronic readiness problems in Defense Facts of Life: The Plans/Reality Mismatch, Chapters 2 through 4 (see footnote 7).

14 "Total Quantities and Costs of Major Weapon Systems Procured: FY 1974-1988." Defense Cost Unit. Congressional Budget Office. March 15, 1988.

15 The cost data comes from Defense Facts of Life: The Plans/Reality Mismatch, page 160. The production data comes from "Selected Weapons Costs From the President's 1983 Program," Congressional Budget Office, March 18, 1982.

16 Michael Gordon, National Journal, January 22, 1983, page 157.

17I do not know how to remove inflation from this depiction. Congress appropriates current dollars, not inflation adjusted dollars. Five-year plans include the Pentagon's predictions of future inflation, whereas Congressional appropriations include the effects of actual inflation. When the Pentagon underestimated inflation in the late 1970s, Congress compensated it for the lost purchasing power with supplemental appropriations; however, when the Pentagon overestimated inflation in the early 1980s, it reaped a $40 billion windfall (according to the General Accounting Office) in excess purchasing power that was never returned to the treasury. An inflation adjusted depiction would hide the effects of our asymmetrical reaction to the "ravages" of inflation. Current dollars are, therefore, a better measure of the Defense Department's demand for and receipt of money from Congress.

18Despite a 32 percent increase in real spending for operations and maintenance between 1980 and 1989, flying hours per crew, steaming days per ship, training days per battalion, and miles driven per tank remained constant or declined. The main exception was a 20 percent increase in Air Force fighter-pilot flying hours, but even in this case, the current rate is still substantially less than the 1973 training rate of 26 hours per month.

19Shortages of spare parts became evident in the deployment to the Persian Gulf. According to Lt. Gen. Henry Viccellio, for example, shortages forced widespread cannibalization of Air Force aircraft remaining in the states (see Aviation Week and Space Technology, September 17, 1990, page 27.)

20 Congress did force the Defense Department to buy a few tactical transports (C-130s) for the reserves and the Air force bought a comparatively small number of tactical air refueling aircraft (KC-10s) for the active force.

21Annual Report to Congress by Frank Carlucci, Secretary of Defense. February 18, 1988.

22"Cheney cuts major Pentagon aircraft buys." News Release 203-90. Office of Assistant Secretary of Defense (Public Affairs), Washington, D.C. April 26, 1990 and "Cheney Suggests Cuts In Aircraft Programs." Washington Post. April 27, 1990. Page A12.

23 On April 20, Senator Sam Nunn, in the last of a series of widely acclaimed speeches, probably set the best case scenario for future budgets year when he suggested that between 100 and 130 billion dollars will probably be cut from the five year plan. The budget resolutions passed by the Senate and the House of Representatives would cut the five-year plan by substantially larger amounts.

24 Madison, in Federalist Number 58, said "This power over the purse may, in fact, be regarded as the most complete and effectual weapon with which any constitution can arm the immediate representatives of the people, for obtaining a redress of every grievance, and for carrying into effect every just and salutary measure."

25Threat inflation includes exaggerations in force size and capabilities. In 1971, for example, the Defense Intelligence Agency's 1980 projection for the Soviet naval order of battle included between 71 and 101 modern nuclear attack submarines; by 1980 however, the actual order of battle included only 44 of these submarines (source: Jan Beemer, Soviet Submarines: Design, Development, and Tactics, Jane's Information Group, page 124). In 1976, Soviet pilot Viktor Bilenko defected with a Mig-25 to Japan. The information gathered from this incident revealed that the American intelligence community had grossly exaggerated the capabilities of the Mig-25 (source: John Barron, Mig Pilot, Avon Books, pages 172-7). For a comprehensive discussion of this problem and the damage it causes, see Andrew Cockburn's The Threat: Inside The Soviet Military Machine (Random House, 1983). If the Soviets want to reduce tensions, it is in their interest to make it harder for us to exaggerate their capabilities. They can do this by releasing more information about their military to the West.

26 Two recent examples illustrate the power of the bureaucracy to hide bad news. The Wall Street Journal (June 11, 1990, page 4), reported that the Under Secretary of Defense for Acquisition, John Betti, was surprised by reports that the Navy's A-12 attack aircraft had serious technical problems and a schedule slippage. Sources said he questioned why the problems hadn't surfaced during the Major Aircraft Review which was completed and delivered to Congress by Mr. Cheney only six weeks earlier. Last December, in press conferences immediately after the Panama invasion, Mr. Cheney repeatedly cited the pinpoint bombing accuracy of the Air Force F-117 stealth fighter. When New York Times reporter Michael Gordon brought photos to the Pentagon showing a bomb crater about 1000 feet from the target, Mr. Cheney ordered an investigation. The Air Force Times (June 18, 1990, page 4) reported that the Air Force inspector general attributed the Air Force's failure to report the bad news to poor communications and special access security rules within the Air Force. Mr. Cheney's spokesman, Pete Williams, said he did not know who was responsible or whether any disciplinary action followed.



The views expressed in this article are those of the author and do not reflect the official policy or position of the Department of Defense or the US Government.

© 1990, 1998 FRANKLIN C. SPINNEY
All Rights Reserved

Author's Note