Representative HAMILTON. Thank you very much, Mr. Adams. Mr. Lee, please proceed.

STATEMENT OF L. DOUGLAS LEE, VICE PRESIDENT AND CHIEF ECONOMIST, COUNTY NATWEST, WASHINGTON ANALYSIS CORP.

Mr. LEE. Thank you, Mr. Chairman. It is a pleasure to be here today. I would like to focus my remarks on three broad areas. First, some general remarks on the nature of defense spending and its relationship to the economy. Second, some remarks on the timing and structure of the defense cuts that I believe reasonable. And finally, some observations about how these interact with the greater economy.

I thought that it might be useful at first to try to remove some of the often-repeated errors of fact and logic that cloud discussions about defense spending. Defense spending is often considered nonproductive and inflationary because there is no flow of useful goods and services. I don't think that is correct. From an economic perspective, the inflationary potential of defense spending has nothing to do with its usefulness.

In a private market transaction, the production and consumption of goods and services is a two-sided transaction. That is not true in a government transaction, however, as the Government pays for goods but it removes them from the private economy. So unless there is some other mechanism to soak up the purchasing power, such as taxes, it would result in inflation. The key fact here is that the inflationary impact depends on whether the aggregate purchasing power is being expanded more than the aggregate production of goods.

The second fallacy in this logic is that defense spending does not produce a stream of benefits. In fact, it provides something we call national security. An "adequate" amount of national security is difficult to define and more difficult to value, but that does not mean that it is worthless. Everyone will agree that without an adequate amount of national security, we would not be able to enjoy the other benefits of our economic system. It is this aspect that makes the changing cold war environment so exciting.

In the early 1980's, we believe that we had allowed our defense capabilities to run down during the decade following the Vietnam conflict and that rebuilding was required. Basically, this meant that we had not been providing an adequate amount of national security. To correct this imbalance, resources had to be shifted from the civilian to the defense sector. However, if you believe that a stream of benefits flows from providing an adequate amount of security, then you would conclude that this was a redistribution exercise with little net impact on the Nation's overall standard of living.

The situation today is different from the early 1980's. Today we believe that the nature of the Soviet threat has changed. The Soviets are less aggressive, less economically capable, and, due to opening the borders in Central Europe, less able to mount a surprise attack on Europe with short notice. Because the nature of the threat has declined, fewer resources are now required to provide an adequate amount of national security. This means that, rather

than simply redistributing resoures as we did in the early 1980's, resources will be freed for other purposes. Whatever these other purposes are, because the benefits provided by an adequate security will continue, it should mean a substantial net addition to our standard of living as a nation.

Saying that there is nothing inherently nonproductive or inflationary about defense spending as long as we are willing to pay for it with lower levels of consumption, it not, however, the end of the story. For most defense goods there is only one market—military. This is not true for most other goods that the Government buys. A $10 billion cut in defense spending will have a very important impact on industries such as small arms, ammunition, explosives, and nonferrous forgings where 15 to 25 percent of the industry output is purchased by the military. An equivalent $10 billion cut in transfer or interest payments or expenses would be spread across all of the goods and services produced in the economy with no single industry feeling a large impact.

A feeling for the concentracted nature of defense spending is revealed in the charts in my prepared statement. Chart 1 is a typical picture of defense as a share of the total economy. Over the past few years, defense has declined from a post-Vietnam peak of about 6.5 percent in 1985 to about 5.5 percent today. The next three charts, however, are much more helpful in describing the relationship of defense to the economy.

Chart 2 shows defense capital goods shipments as a share of total capital goods. As you can see, during the mid- and late 1980's, defense goods became increasingly important for the capital goods sector.

Charts 3 and 4 show the goods and service parts of the economy separately. DOD currently buys about 8 percent of the services produced in our economy and just over 5 percent of the goods. While it may be somewhat surprising that defense is more important to services than to the goods sector, one must remember that the salaries of the 3.3 million people directly employed by DOD are counted in services. During the 1980-85 period, total employment in the United States grew about 9 percent while employment in the defense sector grew about 30 percent. The industrial and geographic concentration of defense production is explored in more detail later.

CUTTING THE DEFENSE BUDGET

The administration is currently in the process of putting its fiscal year 1991 budget proposal in final form. At this point there are many decisions that have not been made, but there is also some useful information that is flowing from this process. My first observation is that the administration's decisions are being driven at least as much by budgetary considerations as by national security needs.

The DOD's budget-making process, the final spending number is the result of many individual decisions made over several years about weapon systems, programs, and personnel. If a weapon needs to be purchased and Congress agrees to fund it, then the flow of spending occurs as the weapon is built. This means that any one

year's outlay number is the result of many past decisions about the national security. Looking at the fiscal year 1991 budget, almost 40 percent of the defense outlays that will occur are the result of decisions that have already been made, even though Congress has not yet seen or approved the fiscal year 1991 budget.

The administration's approach to the fiscal year 1991 budget has been to start with outlays rather than to end with them. The only reason for doing this is to force decisions about programs and weapon systems to produce a desired spending total. While this approach is most likely to achieve a spending and deficit target, there is no reason to expect it to yield the best national security posture. Often it also results in an outlay estimate that is inconsistent with the recommended level of budget authority and outlays that are higher than planned.

A second observation is that the outlay target approach to budgeting will force certain types of decisions to be made. With the decisions made in fiscal year 1991 affecting only 60 percent of fiscal year 1991's defense spending, Congress and the administration will be quite constrained in where they make cuts if the desired spending target is to be achieved. In fact, if you look at that part of the budget which can be changed by this year's decisions, that over 70 percent of the dollars are for pay. There is no practical way for Congress to make significant cuts in fiscal year 1991 defense spending without reducing the number of DOD's civilian and military employees.

The rate at which budget authority translates into spending is shown in table 1 of my prepared statement. As you can see, about 68 percent of the authority for pay is spent in the same year that it is provided. About 56 percent of the operations and maintenance authority and 40 percent of the research authority is spent in the first year. If you intend to reduce the defense budget and have it be reflected in lower spending in the same year, these are the areas where spending must be cut. Stated differently, a dollar cut from the weapons procurement budget will lower outlays by only 20 cents, while the same dollar cut from the pay budget will lower outlays by almost 70 cents. Understanding this structure is necessary both to anticipate where the administration's cuts are likely to be concentrated, and to understand how those cuts are likely to impact the industrial structure of the U.S. economy.

A third observation I would make is that the administration is trying to play the old baseline game. The game is simple. First you create a baseline spending path; then you measure all changes relative to that baseline. If the baseline is high enough, you can make substantial cuts from that baseline and still have a generous budget. When Mr. Weinberger was Secretary of Defense, he regularly presented baseline budgets that contained 5 percent real growth in real terms.

Congress, however, stopped providing real growth in 1985. The defense budgets for 1986-90 fell between 1 percent and 4.8 percent in real terms each year. Over the last few years, Defense Secretaries have steadily been bringing the baseline down closer to what Congress was providing, but chart 5 shows that Mr. Cheney's last official baseline path was still anticipating 2 percent real growth. Of course, no one-including Pentagon analysts-really expected

this baseline to materialize. The optimists expected Congress to provide zero real growth, most people expected about zero nominal growth, and the pessimists expected nominal spending cuts.

It is this baseline against which the $180 billion cut proposal is being measured. To provide some perspective, I have calculated the amount of savings-relative to the same baseline-that would be produced by the zero real and zero nominal paths. To avoid worrying about the best forecast of inflation, I have simply used the assumption contained in the DOD baseline. Current, unpublished DOD estimates will vary slightly from the data I have used, but this will not change any conclusions of the analysis.

The calculations are shown in table 2 of my prepared statment. As you can see, zero real growth would reduce the baseline by about $124 billion over the 1992-94 period, while zero nominal growth would reduce it by $192 billion. Viewed in this context, Mr. Cheney's $180 billion proposal would only bring the Pentagon's plan in line with what most observers had already expected to see. Considering that these expectations have been formed over the past few years, as we have watched congressional behavior, and did not reflect any of the recent events in Central Europe or the Soviet Union, Secretary Cheney's proposal seems quite modest. In fact, an analysis of what spending cuts of this magnitude would mean for the defense program has already been done by the Congresional Budget Office (CBO).

Last March, CBO published an analysis of the implications of a zero real growth defense budget and a budget that declines 2 percent in real terms. The new plan that the administration submits next year is likely to fall somewhere in this range. Broadly speaking, two conclusions result from ths analysis:

One, Congress and the administration must decide whether the cuts are to be concentrated in military forces-people-or in investment spending-weapon systems-or to be divided among each.

Two, in a zero growth scenario, it is possible to concentrate the cuts in people while keeping the current weapons plans largely in tact. This would require a cut of about 14 percent or 462,000 people. In a budget that declines 2 percent in real terms, the cuts are too large for a realistic plan to achieve them with personnel cuts alone. Major weapon systems will also need to be reduced.

A cut of 462,000 people from the 3,300,000 military and civilian employees of DOD would be very large, but not unthinkable. A reduction of this magnitude would leave us with the smallest number of people in the military since the Korean war, but with more than we maintained between the end of World War II and Korea. President Bush has already proposed limits on troops stationed in Europe which would require the withdrawal and demobilization of about 30,000 U.S. troops. Under Bush's proposal, the total might grow to 40,000 if all support personnel are included, but it would still leave 275,000 air and ground personnel in Europe.

Even the relatively small personnel cut proposed by President Bush would result in corresponding weapon and operations expense cuts. For example, if the 30,000 troop cut were accomplished by eliminating one mechanized division and 1% air wings, we would expect first, to save about $2 billion per year in personnel and op

erations costs, and second, to eliminate the need for about 110 F-16 aircraft, 520 M-1 tanks, and assorted other pieces of equipment.

In table 3 of my prepared statement, I have identified the weapon systems most likely to be canceled, postponed, or stretched out in the coming defense cuts. Obviously, if the cuts are at the smaller end of the range and more concentrated in personnel, then fewer of these systems will be affected. However, some weapon cuts are likely in any event. The decisions about the particular systems to be reduced will also be influenced by any agreements reached in the Strategic Arms Limitations Talks (SALT) and the Conventional Forces in Europe (CFE) negotiations. For example, one item under discussion is limiting the number of cruise missiles. To accomplish this, we may also need to limit the number of bombers and submarines used to launch those missiles.

In deciding where to make the defense cuts that will produce a desired spending total, there are only a few simple rules that the administration must keep in mind. First, go where the money is. Chart 6 in my prepared statement shows how the typical defense budget is distributed among major accounts. When one starts to think about defense cuts, major weapon systems come quickly to mind. Chart 6 shows that this is not where DOD spends the bulk of its money in any given year. Certainly, cutting weapons results in large savings when cumulated over several years, but the same thing is even more true of personnel cuts, because this reduces training and equipment expense as well as pay.

The second rule, if you want to see the results of the cuts quickly, is to go to the accounts that spend out the fastest. These have already been identified in table 1 of my prepared statement. Closing unneeded military bases, for example, is a very intelligent policy. However, because of the costs of impact statements, environmental cleanup, adjustment assistance, and relocation of people and equipment, closing bases will actually add to defense spending for 2 to 3 years after the decision is made.

Finally, remember that if this approach to defense cuts provides the appropriate amount of national security, it will be a happy accident.

ECONOMIC IMPACTS

The economic effects of a declining defense budget should be examined from both a macro and a micro perspective. This should be done remembering that the adjustment process is not something that lies exclusively in the future. As mentioned earlier, defense budgets have been declining in real terms for the past 5 years. Thus, the real issue is not the direction of change, but the speed at which it is likely to occur.

MACROECONOMICS

Earlier I argued that there is nothing inherently inflationary in defense spending because other macroeconomic adjustments can fully offset any inflationary impact. This argument can be broadened to apply to economic measures other than inflation. The keys, of course, are the other macroeconomic adjustments and the timeframe examined. In the short term, if cuts in the defense budget

35-140 0 91 - 2