Foreign Relations of the United States, 1977–1980, Volume IV, National Security Policy

143. Memorandum From Fritz Ermarth of the National Security Council Staff to the President’s Assistant for National Security Affairs (Brzezinski) and the President’s Deputy Assistant for National Security Affairs (Aaron)¹

Washington, July 5, 1979

SUBJECT

• Defense Investment (U)

The attached chart, one you used at the 4 June NSC meeting, makes a point we shall have to take increasingly into account in defense planning and budgeting: Soviet defense investment has exceeded ours for a decade, has exceeded ours by about 75 percent for the past several years, and is projected to continue to do so well into the 1980s.² (S)

This is the conclusion of work done in Andy Marshall’s net assessment office in OSD. I’ve had a chance to review informally a draft of his full report, and want to pass on some additional highlights. Andy has sent copies to Harold Brown and other senior DOD people. After responding to some of their comments, he will distribute a finished version. Almost all of the analysis, incidentally, is unclassified. It derives from comparing published US and NATO defense data with unclassified CIA analysis of Soviet and Warsaw Pact defense outlays. (U)

The key message is in the chart itself. While Soviets outspend the US on defense by 25–40 percent, depending on how you measure, they out-invest us by even larger amounts. Defense investment is defined as spending on military R&D, procurement of hardware, and military construction. Unlike spending for current operations and maintenance (except for training), defense investment is aimed at future strength. It tends to be cumulative (although there is also the opposite tendency for investments to “decay” through obsolescence and wear-out of equipment and facilities). In other words, even if we somehow matched Soviet defense investment next year, we could not suddenly match the results of Soviet investment in earlier years. (U)

This is, of course, a highly generalized indicator of relative defense activity. Captured within it are such specific developments as those Soviet strategic force improvements that have made MINUTEMAN vulnerable. Andy’s study reaches the judgment that unless the USSR sharply alters very long-standing defense spending patterns, continuation of established US defense spending policies, including budget increases currently projected, will lead the US into a clearly inferior position in the 1980s. The trends and magnitudes depicted in this analysis have to be radically in error for any other conclusion to be reached. It follows naturally, to quote from the draft, “We will have to spend a lot more on defense if we are to remain a major power relative to the Soviets.” (U)

The Marshall study considers various arguments advanced to soften or discount so stark a judgment. For example:

— Methodological uncertainties in measurements of this sort may lead to an overly gloomy view. This may be so. Or the true picture could be even worse. CIA data probably understate Soviet defense investment somewhat because they do not measure all real investment in defense, e.g., construction for R&D facilities. At the same time, US projections of actual US defense investment in the 1980s may be optimistic. In any case, the gross picture of relative investment is not likely to be wrong.

— The role of our allies compensates somewhat. This is true. But taking allies on both sides into account shows the Warsaw Pact out-investing NATO by 15–20 percent. Moreover, as Victor noted in our pre-NSC discussions, NATO defense efforts are a “dependent variable”; they depend on US strength and leadership. If the Allies see the overall trends against the West, they will not take up the slack, but seek refuge in Soviet good will, knowing that they cannot compensate for Soviet power.

— Do we have compensating technological advantages? Yes, but they are dwindling. Our general level of military–technological innovation appears to be subsiding, and we put less of our available technological potential into deployed systems than the Soviets do.

— The Soviets continue to fear our demonstrated capacity to mobilize our industrial base for defense when we get excited. But they probably also notice that this US capability has declined in recent years for various reasons: (1) Due to our shrinking overall defense effort, the defense industrial base for mobilization has shrunk, e.g., many subcontractors have left the business. (2) Modern weapons systems are more complex and, therefore, expanding their production takes more time than past experience would suggest. (3) We pay less attention to national mobilization as an aspect of defense policy than does the USSR.

— In the 1980s, Soviet internal difficulties may persuade Soviet leaders to reduce defense effort. This is possible but not likely, and, according to CIA, not dictated by economic necessity. Soviet priorities are well entrenched. The opportunities and threats of the 1980s are likely to impress upon Soviet leaders the wisdom of past emphasis on defense. Defense is a “success industry” in the USSR from a political viewpoint. And, as Bialer keeps reminding us, today’s USSR is striking, not for the severity of its internal problems, which it has always had, but for its socio-political stability and its ability to manage despite these problems. (U)

It might be argued that investment is a measure of “inputs” to defense, but not necessarily a measure of useful military outputs. The investment imbalance may be thought to reveal Soviet inefficiency; they have to spend more because they get less out of it. In fact, the Soviets are probably a lot less efficient in the defense sector than we used to think. But this measure does not reveal it. Rather, this measure is largely the result of military outputs. Soviet defense costs are estimated by CIA on the basis of observed physical activity, e.g., missiles, ships, bunkers, etc. Occasional clandestine reports of real Soviet defense budget figures confirm that our totals are about right. (S)

A more significant deficiency of Andy’s analysis might be called “the problem of relative starting points.” If the Soviets were far enough behind the US in real military strength when, around 1968, they began to invest more than we in defense, then all their effort may amount to no more than closing large gaps. This merely reminds us that we cannot divorce gross investment and spending trends from real force balances, trends in those balances, and missions. To a large extent the Soviets have been closing gaps. They have also been investing in areas where we choose not to compete and may, in fact, be forcing them to waste investment, e.g., in air defense. And they must invest against the China threat. (U)

Yet it is hard to accept that a Soviet investment effort larger than ours for more than a decade will yield no more than gross military equality, given the starting point. In the late 1960s, the Soviets had functional strategic parity with the US by their own testimony. They enjoyed preponderance in theater forces in Europe. And they had largely completed the quantitative expansion of their forces opposite China, a power balance in which the US also has an interest. They suffered deficiencies in all these areas. Naval forces were a major area in which they lagged. Here differences between US and Soviet naval missions and doctrine are important. In terms of missions, the US needs naval superiority; the USSR does not. (U)

We should keep in mind that Soviet spending levels in the 1970s result from steady growth since the late 1950s, not any change of Soviet policy. It is we who changed our behavior, reducing real defense outlays since the late 1960s after a period of growth. One cannot easily escape the conclusion that, sooner or later, a superior effort by a competent performer will yield a superior position. (U)

If this overall message is approximately correct, it has far-reaching implications:

1. The increases in defense spending we now talk about are likely to be insufficient. In no case can we turn things around quickly. Substantial increases in defense spending as a percentage of GNP over a sustained period may be necessary to alter the prognosis in five years or so.

2. Increasing the efficiency with which we select and buy defense assets is a necessary, but insufficient, condition for redressing the balance. There must be more resources. But at the same time we probably do require radical reform of defense decisionmaking to achieve a new level of innovative capacity and to squelch time-honored forms of resource misallocation. Trying to find the “most cost-effective options” among many dozens of discrete choices tossed up by the established system will only exhaust the leadership and defeat it in detail. At best we may sub-optimize at the wrong level of performance. (S)

It would be useful for you to mention the defense investment study to Harold at one of your regular meetings. You should solicit his reaction to its basic messages, and urge him to send it over to us for comment. If it withstands critical examination, it could be then made “background reading” for the DOD budget review and possibly incorporated into another NSC meeting, as was CNA 78 in the last one.³ (C)

1. Source: Carter Library, National Security Affairs, Staff Material, Defense/Security, Ermarth, Box 3, Defense (Items in the System): 7/79–6/80. Secret. Sent for information. Copies were sent to Brement, Odom, and Utgoff.
2. Not found attached. For the June 4 NSC meeting, see Document 141.
3. The NSC discussed the 1978 Comprehensive Net Assessment on June 5. See Document 142.