The value of military intelligence

Abstract

This study evaluates the role of military intelligence in an arms race between two countries. The intelligence apparatus of each country evaluates the rival’s capabilities and intentions, and enhances its military capability by increasing the effectiveness of its own weapon systems and reducing the effectiveness of the rival’s weapon systems. Intelligence superiority also yields an advantage in deterrence and preemption. This study shows the following. (a) Investment in intelligence is much less beneficial for small government budgets, low intelligence efficiency, and a low degree of conservatism on the part of the policy-maker. (b) The expenditure on intelligence increases when intelligence efficiency is low and rising, and decreases when intelligence efficiency is high and rising. (c) Being very conservative may be costly to the country. (d) High-quality human capital substantially improves the country’s security and welfare, particularly when the rivals are engaged in a knowledge race in addition to the usual arms race. An application of the model to the Israeli–Syrian arms race demonstrates its validity and usefulness.

Keywords:

• Intelligence
• Knowledge race
• Arms race
• Budget allocation

• JEL codes
• D74
• D78
• D90
• H56
• H68

Notes

¹ Two instances of Sun Tzu’s insights on the importance of intelligence and knowledge should be studied by all civil servants: ‘A hundred ounces of silver spent for intelligence may save ten thousand ounces spent on war’ (cited by Hughes-Wilson 1999); ‘If you know the enemy and know yourself, you need not fear a thousand battles. If you know yourself and not the enemy, for every victory you will suffer a defeat. But if you know neither yourself nor the enemy, then you are a fool and will meet defeat in every battle.’

² The decisive role of information is demonstrated in the following three examples. (1) During World War II, the Allies gained strategic and tactical advantages when they were able to decipher the ‘Enigma’ code and when they used feints as part of the Normandy invasion in June 1944. (2) High-quality intelligence helped Israel gain a significant military advantage during the 1967 Six-Day War. The three-hour air strike on its enemies’ air force bases at the beginning of the war was made possible, among other factors, by precise intelligence and out-of-the-box thinking. (3) The US intelligence community played an important role in the capture and killing of Osama Bin Laden; see the remarks of President Obama to the public on May 2011 (US, WH 2011).

³ A comprehensive bibliography of the intelligence literature can be found in Clark (2007).

⁴ See, for example, the mathematical analysis by Blackwell (1951), the assessment of the value of information by Ahituv (1989), the analysis of the value of information within an economic model by Shapiro and Varian (1999), the game theoretic approach by Solan and Yariv (2004), and the review of the literature by Repo (1989).

⁵ We do not focus here on the optimal organizational structure of the intelligence community. Rather, we consider all the intelligence agencies as one entity.

⁶ Some of the official data on US military intelligence expenditures appear in US, DCI (1997, 1998) and US, DNI (2007, 2008, 2009). Additional data are available from the final report of the 9/11 Commission and some unofficial estimates, or are based on these data (see Global Security 2009).

⁷ The UK data include only the direct expenses of its intelligence agency, MI5 (UK, MI5 2009). The shares of the other countries are estimates of both direct and indirect investments in intelligence.

⁸ A similar interpretation is based on a scenario analysis. Government and military decision-makers work with a set of assumptions on future (short-, mid-, and long-term) potential war scenarios. It is a reasonable assumption that higher expenditure on intelligence narrows the set of potential future scenarios and, therefore, reduces the need for excessive expenditure on weapon systems, infrastructure, and personnel.

⁹ The Cobb-Douglas utility function was employed, for example, by Rogerson (1990) and Hirao (1994).

¹⁰ We present here a deterministic model. Allowing uncertainty about the effectiveness of the intelligence process or about some other element of the model may yield additional insights into the effects of intelligence on security and welfare (although the added complications may mask some of the generality that is obtained by the integrated model), but do not change the main results and intuition that are obtained in this paper. Models with uncertainty are developed and assessed using simulations and, sometimes, real-world data, in Pecht (2013).

¹¹ Almost all the R&D aimed at the country’s intelligence is ‘in house’ and, hence, country specific.

¹² This formulation is similar to that of Johnson (2000).

¹³ The decision-maker’s degree of conservatism is a proxy for her uncertainty about the country’s ability to accurately estimate the rival’s military capability and intentions.

¹⁴ As is shown later on, some characteristics of the general form of the integrated model can be obtained analytically, but an explicit solution of the general model cannot be obtained. The general integrated model can be calibrated with real-world data, and special cases of the model can be solved analytically. These solutions yield some thought-provoking results. Using other welfare functions, such as the flexible CES function, a linear or quasi-linear function, does not change the main conclusions of this paper (see Pecht 2013). More complicated models employing dynamic specifications of the integrated model yield additional insights into the effects of intelligence on security and welfare, but do not change the main results and intuition that are obtained in this paper. These models are developed and assessed by simulations in Pecht (2013).

¹⁵ We assume and require

¹⁶ Assessment of the long-term effect of the degree of conservatism on welfare requires accounting for the expected war damages in case a war erupts between the rivals (see Pecht 2013; Shabtay and Tishler, forthcoming).

¹⁷ Our assumptions about the shares of intelligence expenditures in Syria’s defense expenditures are for different intervals (sets of years) than for Israel. In the case of Syria, significant changes in the item ‘other expenditures’ took place in the government budget during the three sets of years.

¹⁸ The robustness of our assumptions is assessed later on in this section.

¹⁹ Estimates of non-linear 2SLS or weighted non-linear 2SLS yield very similar results.

²⁰ For further information see Zellner and Theil (1962) and Amemiya (1977).

²¹ We believe this problem is caused by the relatively small variance of the exogenous left-hand side (LHS) variables.

²² Similar trends are obtained with the N3SLS estimates.