DEFENSE BUDGETS AND CIVILIAN OVERSIGHT

Occasional Paper #9

Jeanne Kinney Giraldo

June 2001

Introduction

As countries democratize, newly-elected governments find that they must quickly assert control over the budget, especially the defense budget, if they are to meet their policy goals for all sectors and assert civilian control over the military. Fiscally, the process by which budget decisions are made should be structured so that defense spending does not "break the bank" by exceeding the government's capacity to pay. Sectorally, defense should compete with other government ministries so that the final overall budget is a monetary expression of the priorities of the nation, its choice between "guns and butter." Within the defense sector, the allocation of resources for training, personnel, and equipment should reflect the roles and missions for the armed forces established during a process of national security planning. Finally, the power of the purse provides civilians with a key lever of control over the military: government preferences are more likely to be taken into account when they are backed by the provision or withholding of resources.

In addition, the ongoing nature of the budget process, with planning for the next year beginning soon after the current year's budget is approved, can contribute to civilian control over the military and effective policy making in a number of ways. Ideally, it encourages a yearly debate on defense policy and forces the making of hard choices that might otherwise be postponed. It can also provide a key venue for legislative control of the defense sector. The writing of laws that affect the defense sector and the military is a periodic event, but the need for the legislature to approve and review expenditures is a permanent source of influence. Similarly, to the extent that a public debate over the defense budget occurs, a measure of control and participation by the civilian population is introduced into a policy arena that is often closed.

The budget process consists of four stages, analyzed in the body of this paper: (1) the formulation of the budget by the executive, (2) its enactment into law by the legislature, (3) the disbursement and spending of the funds, and (4) an evaluation of the efficiency and effectiveness with which the money was spent. This paper examines the different ways in which civilian and military participation is structured at each step of the process, based on the experiences of a wide range of countries at various stages of democratic development. (1) The impact of these variations on civilian control of the military and transparent and effective policy-making is evaluated.

1) Budget Formulation

For all sectors, the formulation of the budget usually takes place within the executive branch and requires two sets of decisions: one about the distribution of funds between sectors (e.g., defense, transportation, health) and another about resource allocation within sectors. In most countries, the "center" -- some combination of the Ministry of Finance (MOF), the head of the government and/or the cabinet -- is responsible for the first decision.(2) This decision should reflect both the macroeconomic constraints on spending as well as the policy priorities of the government of the day. The second decision, about the allocation of resources within a sector, is the responsibility of the relevant line ministry (e.g., the Ministry of Defense). In making this decision, the Minister is guided by input from subordinate service agencies (e.g., the armed forces) and the government's policy agenda for the sector (e.g., national security "white papers").

Neither decision can be made in isolation from the other. The Minister of Finance will not be able to distribute resources between sectors unless s/he has some sense of the needs of each sector. Similarly, the sectoral minister, who always faces a situation where needs exceed limited resources, will not be able to make the hard choices about spending unless s/he has some realistic estimate of the amount of money that will be devoted to the sector. As a result, the writing of the budget is usually an iterative process in which the center consults ministries about their spending needs and sets spending ceilings, ministries submit detailed spending requests based on these numbers, and then the center reevaluates their ceilings based on these requests and may choose to reallocate funds.(3) Any disputes between spending ministries and the MOF are typically arbitrated, either formally or informally, by the head of government or the Cabinet. The ministries are then expected to modify their spending estimates to match the revised ceilings set by the center. The final figures from each ministry are compiled into a draft budget bill that is then sent to the legislature for approval.

Setting Spending Limits: The case of defense

In many democratizing countries, civilian elites have been successful in asserting control over the level of the defense budget, often greatly reducing spending (e.g., Argentina, Nicaragua, South Africa, the Philippines). In most cases this can be attributed to the general strengthening of the Ministry of Finance relative to all spending ministries as countries facing domestic economic crises and the requirements of international lending institutions put their fiscal houses in order. For example, a survey of Latin America and the Caribbean showed that 18 of 20 countries in 1993 possessed Finance Ministers whose authority was "considerably greater than that of the spending ministries on budgetary issues."(4) In addition to this motive for reduced spending across the board, democratically elected officials often have incentives for cutting the defense budget in particular. In many cases, excesses in military spending under authoritarian rule led to a bloated defense budget and it is the task of the newly-elected civilians to rationalize spending. In all cases, democracy brings with it compelling demands for increased social spending, which are often answered at the expense of the defense sector.

While fiscal discipline is a welcome and necessary change, the ascendancy of the MOF within the executive branch brings with it the danger that excessive or arbitrary cuts will be imposed on the defense budget, based solely on fiscal criteria and disregarding national security needs. As David Pion-Berlin argues in the case of Argentina:

national security considerations and fiscal considerations are largely divorced from one another, with the latter taking precedence. That means that the military is left 'out of the loop.' The flow of budgetary decisions does not involve a constant mix of defense and fiscal strategists. It is only after the budget is assembled and approved that the military moves to center stage, with planners and programmers figuring out how to spend the scant funds delivered to them. Economists within the secretariat of finance are not obligated to consult with defense experts on how the size of the budget would impact national security objectives.(5)

A number of procedures can help ensure that larger policy considerations shape decisions about spending levels. In most countries, the president or cabinet uses policy criteria to resolve disputes between the fiscally-oriented MOF and the line ministries. A politically-connected civilian Minister of Defense with a clearly articulated vision of national defense is more likely to be able to influence the MOF and the President than a military minister.(6) In some countries, policy concerns are incorporated into the process of setting spending limits by consulting relevant stakeholders inside and outside of the executive branch prior to establishing ceilings. In the US, Congressional leaders, motivated by a combination of fiscal and policy concerns, often play a more important role than the Office of Management and Budget in setting the parameters of initial planning in the Defense Department.(7) The OMB, in turn, consults the Office of the Secretary of Defense about the needs of the sector before any financial decisions are made.(8)

Allocating Resources within the Defense Sector

Although civilians in democratizing countries have had success in setting the levels of defense spending in many countries, on the whole there has been little civilian control over the allocation of resources within the defense sector.(9) Civilian control over this second set of decisions (and effective policymaking) depends upon two related factors: a strong civilian ministry of defense and civilian participation in a process of national security planning. In the absence of these factors, the allocation of resources within the defense sector is likely to be the product of a power struggle among the services pursuing their narrow corporate interests, largely bereft of strategic considerations on how best to promote national security.(10)

It is the task of the Ministry of Defense (MOD) to produce a policy statement outlining policies and expected outcomes, the way in which objectives will be achieved, and performance indicators to measure success. In most developed democracies, the Minister's statement of policy ("white paper") is circulated through the services so they can draw up their budgets in line with government policy. For the most part, civilians within the MOD provide strategic assessments and military officers provide the operational details. As Pion-Berlin noted in the case of military reform: "The broad strokes of institutional reorganization must be painted by the president and his defense staff. Only then can the detail be filled in by the military."(11)

The creation of a civilian Ministry of Defense is a necessary but not a sufficient condition for civilian leadership in the defense sector. Civilians within ministries of defense must take an active role in developing the legal framework, bureaucratic organizations, and expertise necessary to perform their leadership role within the sector. For example, before the transition to democracy in South Africa, soldiers had great influence within the tripartite budget working groups (consisting of MOF, MOD, and congressional personnel) and the chief of the armed forces was the accounting officer for the group. After the 1994 creation of a civilian MOD, the Secretary of Defense was made the accounting officer, thus giving personnel from the Defense Secretariat more influence than the military within these groups.(12) In Nicaragua, in contrast, the 1997 creation of a Ministry of Defense headed by a civilian did little to change the previous system in which the budget was the product of an accord between the government's economic team and the armed forces. Although the MOD was charged with directing budget formulation and supervising its execution, it failed to create the centralized bureaucratic unit that would perform these tasks.(13)

In any policy area, executive leadership is often particularly important when major reform is required, changes are controversial, or the line ministry is weak relative to the subordinate service agencies. All three of these conditions apply to defense reform in democratizing countries, where national priorities and threat environments have changed, downsizing is often necessary, and ministries of defense are of recent origin. As Pion-Berlin has noted:

Ultimately, it is the president that must make defense reform a priority. It is he who must instruct his minister to demand from the services full compliance with national defense policy. And it is he who must instill in the defense ministry a real sense of purpose and direction.... [O]rganizations in general and certainly state agencies in particular are not necessarily oriented toward the fulfillment of goals unless they receive a clear mandate from above. In the absence of such a mandate, civilian appointees within the agency feel less motivated to bring themselves up to speed on issues of national security or to hire outsiders with the expertise necessary to confidently push through controversial programs.(14)

Unfortunately, in many democratizing countries, a sustained debate on national security has not taken place and executives find themselves besieged with other problems that require their attention. In Nicaragua, for example, the lack of a debate on national security leaves the civilian ministry of defense without clear guidelines to provide to the armed forces.(15) In Brazil, eleven years after the transition to democracy, Fernando Henrique Cardoso issued a presidential directive outlining a national security strategy, but this seemed to have little impact upon the military-led Ministry of Defense;(16) economic crises subsequently diverted the President's attention from the issue. In Brazil and elsewhere, however, there are encouraging signs of a movement toward national debates on defense issues. In Brazil, the appointment of a civilian Minister of Defense in 1999 has initiated a civilian-led and inclusive discussion on national defense. In Spain, plans are underway for the government to issue the country's first white paper on defense, more than two decades after the transition to democracy.(17) Argentina and Chile both issued their first white papers in the mid 1990s. And in South Africa after the transition to democracy, a defense white paper was written and a defense review conducted with the widespread participation of political parties, non-governmental organizations, the defense industry, defense analysts, the Armed Forces, and the public.

In sum, in the absence of a strong civilian Ministry of Defense and a debate on national security, decisions about the allocation of resources between sectors are often left by default to a fiscally-oriented MOF and, within the defense sector, to the military. A national debate on security issues is necessary not only to ensure that funding levels are adequate for the sector, but also to contribute to the ability of civilians to control and rationalize spending within the sector.

2) Enacting The Budget

In most countries, the legislature is not a key player in the budget process until the second stage, when the executive's proposal must be reviewed, revised, and enacted into law. In some countries, however, the legislature does have some involvement at the formulation stage. In the United States, for example, congressional leaders play a key role in setting initial spending limits. In addition, the executive must take legislative preferences into account when formulating the budget, given the Congress' broad powers to rewrite the executive proposal. In Germany, the ministries work closely with their respective committee in the Bundestag during budget formulation. In South Africa, a special budget working group brings together representatives from the Department of State Expenditure (accountable to the MOF), the Department of Defense, and the parliamentary defense committee.(18)

Once the executive's proposed budget is submitted to the legislature, it is usually discussed in two stages: first, there is a debate and vote on the overall amount of the budget and then later a detailed discussion of the allocation of resources among and within ministries. This detailed discussion usually takes place within the Budget Committee, which calls on Ministers to defend the budgets for their sectors and sometimes invites members of the relevant congressional committees to participate in the proceedings.(19) The centralized control of the Budget Committee helps ensure that the "big picture" of the budget is taken into account (i.e., fiscal limits respected and reallocation between sectors coordinated) while consultation of ministers and committees permits sectoral input.

The ability and willingness of the legislature to modify the executive's budget proposal varies greatly between countries. It is greatest in the US, where the Congress has complete freedom to modify executive proposals and the incentives to do so (e.g., the prospect of obtaining resources for constituencies). It is weakest, perhaps, in Great Britain (and other parliamentary systems where the government is backed by a majority party) where a rejection of the budget is seen as a vote of no confidence in the government.(20) Most legislatures fall between these two extremes, with many clustering closer to the British end of the spectrum. The role that the legislature plays is limited by the factors that shape the overall role of the legislature in the making of defense policy and which have been discussed in depth elsewhere.(21) In this section, I focus on how the specific rules that govern the budget process -- the amount of time the legislature is allotted for debating the executive's proposal, the amount of budget information to which the legislature has access, and the powers of the legislature to change the size and allocation of the budget - affect the ability and willingness of legislators to shape the defense budget.

Time period for budget review

In most countries, budget debates are charged affairs carried out under great time pressure and the knowledge that the proposed budget must be passed before the current one runs out, or else. The amount of pressure the legislature is under and the "or else" varies from country to country and has an impact on the ability of the legislature to make meaningful changes to the budget. Typically, the executive submits the budget two to four months before the start of the new fiscal year.(22) In some countries, however, the legislature is given less time to examine the executive's budget requests. In Nicaragua, for example, the executive is only required to present the budget by November 15 and it must be approved before the end of the session on December 15.(23)

Most of the close examination of the budget occurs in congressional committees; consequently, a large portion of the time the budget is in the legislature should be devoted to the committees. Where it is not, the legislature is unlikely to have effective input into the budget. In South Africa, for example, there is little time for committee deliberations, with the Portfolio Committee on Finance allotted only 7 days to submit a report on the Budget to the National Assembly, less than one tenth of the four months allocated to the entire legislative Budget process.(24) In South Korea, the time period for budget review is usually less than two weeks of the ninety-day regular session.(25) In Germany, in contrast, the legislative stage of the budget process lasts four months and the committee is allotted several weeks for its work.

What happens if the budget is rejected or not passed by the legislature within the constitutionally established time frame? In most countries, the previous year's budget is put into force. If this is done on a month-by-month basis while legislative approval is awaited, the legislature still has some leverage over the process (e.g., Brazil). If the decision to use the prior year's budget puts an end to the budget discussion, and the government is allowed to redistribute some of the funds, the government has the upper hand in negotiations (e.g., Argentina).(26)

Access to Information

Meaningful congressional input into the budget process depends upon the legislature having information about the contents of the budget and the policy implications of their budget decisions. For this information, the legislature relies on data provided by the executive branch as well as research services housed in the legislature.

The most important piece of information from the government is the budget proposal itself, which should have enough details that the legislature understands what they are approving or rejecting. In many countries, however, this is not the case, particularly with respect to the defense budget, where claims of the need for secrecy still shroud the release of information. In Mozambique, for example, the government budget did not differentiate between defense, police and intelligence allocations until 1999.(27) After the transition to democracy in South Korea in 1987 the defense budget was deliberated as a lump sum until 1993.(28) In Russia, from 1992 to 1996 the State Duma approved a defense budget that was generally one to two pages long and consisted of only six to nine spending categories. In 1997 this was expanded by law, but civilians in the Duma still tend to consider only the 17-category declassified budget request since they lack the security clearances necessary to view the new 300-line format.(29) In contrast, the US military budget contains 3-4,000 declassified line items and all the relevant legislators have security clearances to view the classified items.(30)

In addition to the official budget proposal, information from the executive can also be obtained through regular contact between legislators and government ministries. In Germany, for example, the legislative budget committee interacts regularly with government departments through regular department briefings and expenditure reports.

Within the legislature, members of the Budget Committee can rely upon their own expertise, their committee staff, congressional research services, and input from sectoral committees in making their decisions. Legislators in the US, for example, can rely upon a Congressional Budget Office in addition to well-staffed committees. Unfortunately, most legislatures lack access to this kind of specialized knowledge and expertise.(31)

Defense committees, even in countries with low levels of staffing, are likely to be more knowledgeable about the requirements of the defense sector than the members of the Budget Committee, and it is important that there is a mechanism for their input into the budget process.(32) Defense committees in East European legislatures, for example, influence the budget either by making recommendations to the Budget Committee, having a decisive vote, or presenting to the floor amendments that were rejected by the Budget Committee.(33) In Russia, the Defense Committee is only given a one-hour hearing by the Budget Committee, but it lobbies on the floor for increased spending for the armed forces.(34) In contrast, in Great Britain the budget does not have to be passed to the Defense Committee and, as one member of the Committee has noted: "As far as the budget is concerned in the UK, the Defence Committee is almost irrelevant."(35)

Powers to amend the budget

The legal powers of the legislature to amend the budget vary from country to country. At one extreme is the US Congress, which has the power to increase or decrease both revenue and expenditure.(36) In some countries the legislature is forbidden to increase expenditures (e.g., Great Britain), while in others the legislature cannot increase expenditures in one sector without taking from another or raising taxes (e.g., post-1993 Argentina, Brazil, Germany, the Philippines, Spain).(37) In many presidential systems, the president has line item veto power (e.g., Argentina, Brazil, the Philippines).(38)

In countries where the legislature is unable to increase spending levels (or reallocate funds), a military interested in enlarging its share of the budget would focus most of its attention on the executive branch and the formulation stage. In other cases, it might make sense for the military to attempt to lobby the legislature for increased funds. The success of this strategy will depend upon the interest of legislators in participating in the debate over defense policy and, in particular, in raising defense spending.

Interest of Legislators in Modifying the Defense Budget

When evaluating the interest of "legislators" in modifying the defense budget, it is necessary first to identify which legislators control the budget process. It is the interests of these legislators, shaped by their electoral and partisan concerns and the choices provided to them by the budget rules, that will determine the role of the legislature in modifying the defense budget.

Where parties (or factions) are highly organized and disciplined, they are likely to control the budget process within the legislature. Under these circumstances, the defense budget will be shaped by the policy positions of the parties (or factions). In some democratizing countries, parties might be committed to "the collective goal of democratic stability and, therefore, to support higher defense expenditures as part of a policy to safeguard corporate military interests."(39) In other cases, liberal parties (such as the Partido Popular in Spain) might support increased military spending for ideological reasons. Often, however, parties will support defense cuts to the extent that they permit increased spending in areas important to the public.

Party control over the budget process in the legislature can contribute positively to the making of defense policy, particularly when parties are willing and able to debate defense objectives and their relation to the budget (and bring to the table a perspective that differs from that of the executive branch). For example, in Germany, many spending decisions, especially big-ticket procurement items, are the subject of heated debate in the legislature. In Spain, in contrast, the legislature in 1982 rubberstamped the executive's ten-year budget plan for the military without debate, despite the fact that defense objectives had not yet been defined. Debate is more likely where the party in charge of the executive does not have a majority in the legislature and where defense issues are politically salient and parties possess defense experts. Unfortunately, in many democratizing countries, parties frequently do not have "working teams specifically assigned to develop positions on security and defense issues."(40) Despite an initial interest in defense issues during the transition to democracy, with the passage of time political parties have generally focused on traditional economic and social issues, with which they are more familiar and which are electorally more important.(41)

Where parties are less disciplined, as is the case in many democratizing countries, individual legislators will usually control the budget process and will "follow their own immediate political interests, for better or worse."(42) The interest of individual legislators in modifying the defense budget will depend on the rules of the budget game and the impact defense spending has on their chances for reelection. In some cases (e.g., the United States) there is no conflict between attending to constituencies and increasing the military share of the budget. This is because (1) the rules of the budget game give individual legislators on defense-related committees a great deal of power over the budget; and (2) these individuals often benefit politically from the resources provided by the defense budget, because of the presence of military bases, government agencies, or private defense companies within their districts.(43) In other countries, however, the choice that legislators face between attending to constituents in their home district and the military is much more stark. In Brazil, the defense budget provides few resources for legislators since the defense industry is geographically concentrated in a small area and the congress lacks control over the location of military bases.(44) (Many other democratizing countries lack a defense industry altogether or it is owned at least in part by the government and key decisions about its operation are outside the purview of the legislature.) In addition, since the rules of the budget process force legislators to make choices about reallocating funds between sectors, legislators are more likely to favor shifting funds from the defense ministry to more electorally-profitable ministries. In Brazil, for example, individual legislators are allowed to submit amendments to the budget and usually request shifting funds from the military budget to public works.(45) In the Philippines, these considerations, as well as strong public opinion against militarization after years of martial law, have led congressional committees to slash the executive's defense budget requests.(46)

In these cases, military efforts to lobby legislators are likely to be met with indifference or hostility. In Brazil, legislators view membership in defense committees as electorally unimportant and presentations by the military have been canceled because of a lack of legislative interest.(47) In such cases, it is likely that only when nationalist passions are inflamed will legislators perceive electoral gain in increasing the military budget. This was the case, for example, in the Philippines when Chinese aggression against Filipino claims in the Spratly Islands led the legislature to approve a long-delayed bill granting funds for military modernization.(48)

In sum, legislators in many countries face incentives to cut the defense budget but have little motivation to spend their time debating defense policy. This reinforces the trend at the level of the executive, where a strong Ministry of Finance can enforce budget cuts and a weak Ministry of Defense fails to articulate a politically persuasive national defense strategy.

3) Implementation Of The Budget -- Disbursing Funds

Once the budget is approved by the legislature, funds for defense should be distributed in the manner indicated in that document. However, this may not be feasible if unforeseen needs arise (such as a natural disaster) or there is a shortfall of revenues. In these cases, rules need to be in place for reallocating funds between spending categories or for the spending of emergency or reserve funds. In Argentina, for example, the budget can be modified on the government's initiative, with congressional approval. In Brazil, the government can modify up to 20% of the budget without congressional approval.(49) In Russia, the President's Security Council makes changes to the defense budget via decrees, which escape the scrutiny of the Duma.(50) Regardless of legislative participation in these decisions, all changes in resource allocation should be promptly reported to the legislature.

The amount of "rectification" of the budget that occurs will vary from country to country, with a great deal occurring in France, for example, to little or none in Germany.(51) Where budget shortfalls are common or inflation erodes the value of assigned funds, lobbying by spending ministries for the early dispensing of funds is commonplace (e.g., the Philippines).(52)

This can result in a modification of the allocation decisions made during the previous two stages of the budget process. In Russia, where few rules of the budget process are followed, it has been said that there are three budgets: the budget that is enacted by the Duma, the monies that the MOF disburses to the MOD, and the amount that the MOD actually spends (frequently three to four times the amount budgeted to it).

4) Auditing And Outcome Assessment

The evaluation or oversight stage of the budget process has two main purposes: (1) to determine whether the money is spent in the way the budget says (auditing); and (2) to evaluate whether policy goals have been met (outcome assessment). This section discusses each of these purposes and the variety of institutions that usually participate at this stage, ranging from those closely linked to the activity being examined (e.g., auditing institutions within the armed forces and the Ministry of Defense) to governmental institutions independent of the executive, to oversight mechanisms in the legislature, and watchdog groups in civil society.

Auditing of the budget can take two forms: a detailed review of individual transactions that results in holding individuals responsible for misuse of government funds or an examination of procedures and work practices to determine whether or not they are effectively designed to prevent errors and to get the most value for the money. The former approach is important, especially in countries where government corruption has traditionally been carried out with impunity, because it holds individuals accountable for abuses. However, the latter is especially important, and most auditing institutions have shifted to focusing on this, because it leads to changes in the procedures designed to prevent wrongdoing in the first place and introduces reporting requirements that increase the likelihood of uncovering individual wrongdoing after the fact.(53) In practice, the two approaches are likely to go together: a scandal uncovered by the review of individual projects or transactions often leads to changes in the procedures that govern decision making and the spending of money. In South Korea in 1993, for example, a legislative committee for audit and inspection revealed wrongdoing in a major defense project (the Yulgok project, which comprised up to 40% of the defense budget, and whose budget had been approved by the legislature in an unitemized format). This led to changes in the procedures for defense decision-making: a special interdepartmental committee was established within the MOD, designed to oversee decisions and reduce the opportunities for corruption.(54)

Outcome assessment is another key component of oversight, but one that is even more difficult to carry out than auditing. This approach focuses on determining whether resources spent on defense are in fact contributing to national security; if not, resources should be reallocated in the following year's budget to projects or missions that will contribute more to national defense. Such an evaluation of policy is notoriously difficult to make in the field of defense, where a collective good is being provided with few indicators of effectiveness. Whereas a Ministry of Public Works can measure its effectiveness in terms of miles of roads built or paved, or a Ministry of Health in terms of clinics built and patients served, national defense is more difficult to measure (e.g., "number of enemies deterred"?) Despite the inherent difficulties of measuring the effectiveness of defense policies, this issue needs to be debated, policy objectives and indicators of effectiveness established, and spending designed to achieve consensually agreed upon defense goals. Unfortunately, in many democratizing countries, this kind of debate has not taken place, making it difficult to evaluate the effectiveness of defense spending.

Controlling spending and evaluating its effectiveness are enormous tasks that need to be carried out by a wide variety of actors, starting with the institution that is charged with spending the money. Internal audit procedures must be in place within the military to hold their members accountable for their handling of resources. In the United States, for example, Inspectors General of the individual services carry out this function. In Nicaragua, the state Comptroller has helped the military establish internal audit procedures, which seem to operate effectively.(55) The Ministry of Defense should have a centralized office that monitors and evaluates spending by the individual services.

In addition to these agencies within the executive branch, most countries have national auditing offices that are independent of the government of the day and are responsible for reporting on government spending. In some countries, their task is limited to auditing spending; in others, it extends to evaluating whether policy goals have been met or not. In Great Britain, the National Audit Office, carries out both tasks. The effectiveness of independent auditors depends on some minimal level of accountability within the spending ministry itself. In Russia, for example, the State Comptroller has complained that the military does not produce enough receipts for sale of military property and therefore it is impossible to monitor and control military spending.(56)

Another source of external audit is the legislature. Few legislatures have the resources to carry out oversight of budgets independently. The US Congress is a noticeable exception, with a strong Congressional Budgeting Office, congressional staffers who monitor spending on projects of interest to their legislator, and a General Accounting Office that has evolved from an organization focused on detailed audits to one that plays a broader role in evaluating policy effectiveness.(57) In many countries, including the US, the legislature depends heavily upon information generated by the executive's internal audits and the national auditing office. Through legislation, assemblies require the executive to report on their spending. In Great Britain, for example, the Ministry of Defense produces accounts, which are audited and sent to the legislature in January. (58) The Budget Law passed in 2000 in Nicaragua requires the executive to send spending reports to the Assembly each trimester.(59) In addition, there are often provisions for the legislature to review reports of government and military activities produced by independent auditing agencies. Within the legislature, these reports are usually reviewed by specialized committees. In parliamentary systems a Public Accounts Committee frequently performs an auditing function; however, its lack of expertise in different subject areas prevents it from carrying out an outcomes assessment. This typically falls to the sectoral committees (e.g., the defense committee). For the findings of these committees to have an impact on the subsequent year's budget decisions, there needs to be a mechanism for these findings to be reported to the Budget Committee (such as the integration of the Public Accounts Committee into the Budget Committee) or to the executive during the formulation stage. In addition, reports from the executive and independent auditors need to be received in a timely fashion so that they can be analyzed and findings integrated into the next year's budget.(60)

In many policy areas, organizations in civil society often play a watchdog role and call attention to government abuses of power or to policies that need rectifying. These groups, however, tend to be weakest in the area of defense and, in particular, they lack the expertise and access to information often necessary to monitor budget decisions.(61) In Russia, for example, the defense budget was openly published and accessible to citizens only beginning in 1998, and the amount of information supplied was minimal.(62) Just as the legislature is dependent on a certain level of executive accountability to carry out its oversight responsibilities, non-governmental organizations (NGOs) depend on the ability of political parties and legislative committees to secure information from the executive. Unfortunately, parties and legislators often have few incentives to secure this information, for reasons discussed in the preceding section.

In addition to domestic NGOs, the international community has begun to play an important role in overseeing government spending in the defense realm. Adherence to treaties sponsored by international organizations like the United Nations or Organization of American States often requires providing data on military spending and arms transfers.(63) In other cases, bilateral or multilateral agreements designed to lessen regional tensions might lead to increased transparency in defense spending (e.g., a 1998 agreement between Argentina and Chile to fund jointly a comparison of defense spending). To the extent this information is made public, domestic NGOs can use it in carrying out their watchdog role. In still other cases, unilateral donor-aid recipient requirements might play a role in forcing a measure of responsibility in defense spending. US legislation, for example, requires a civilian audit of military receipts and expenditures (with results reported to a civilian authority) before US executive directors at international lending institutions can vote in favor of non-humanitarian assistance to any country.(64) (The audit need not be publicized, however, and therefore contributes little to the ability of other organizations to monitor the government.)

The daunting task of overseeing the defense budget is made even more difficult by the existence of secret funds earmarked by the executive for defense spending but largely free from legislative oversight. Sometimes the secret funds are clear line items in the budget (e.g., for intelligence programs) and the legislature acquiesces to them without knowing their content (e.g., Argentina, Brazil). In other cases, the executive hides military expenditures in civilian portions of the budget (e.g., Argentina, Russia).(65)

While these secret funds are beyond legislative control, off-budget funds earned directly by the military escape civilian oversight altogether and can undermine state policy and democratization. In Nicaragua, for example, officers are "major agricultural producers and highly competitive building contractors."(66) In Indonesia, it is estimated that the military receives 75% of its income from its participation in off-budget, profit-making enterprises.(67)

Conclusion
In sum, despite the success of many countries in reducing the amount of money spent on defense, democratically-elected civilians and their representatives have been less capable of controlling how the money is spent or ensuring that the allocated funds meet the nation's security needs. How can this uneven performance be explained? In part, it can be attributed to extensive debates countries have had on economic policy and government spending and the relative lack of attention to national security issues. A sustained debate on national security issues can make a number of contributions to the budget process: it permits indirect public participation in what is otherwise a rather closed process of budget formulation;(68) it informs the MOF's decision about allocation of resources between sectors; it guides MOD's decisions about spending within the defense sector; and it establishes the criteria necessary to evaluate whether a given resource allocation has contributed to policy objectives.

The pattern of control over spending levels but not choices can also be explained by variations in institutional strength: many democratizing countries have been able to develop strong Ministries of Finance, responsible for fiscal discipline, but the institutions responsible for the content of defense spending are noticeably weaker. Civilian Ministries of Defense are of recent origin in most democratizing countries and have yet to develop the expertise, the procedures, or the legal powers to orchestrate a national security planning process that will inform budget decisions. Executives often find their energies devoted to solving economic or social problems (or putting out fires in the defense sphere) and lack the time or political capital to expend on the comprehensive reform of the defense sector that is often necessary in democratizing countries. Legislatures are often unable and unwilling to participate meaningfully in a national debate on defense issues. Finally, organizations in civil society, which could add a different perspective to the national debate on defense and have an interest in monitoring government spending, are weak or non-existent. Actors interested in increasing civilian control over the military and in improving the quality of national defense should contribute to a debate on national security and work to strengthen the institutions responsible for linking defense priorities to the budget process.

---

(1) Countries representing a broad, geographical cross-section are examined: Argentina, Brazil, Germany, Great Britain, Nicaragua, the Philippines, Russia, South Africa, South Korea, Spain, and the United States. Information on the defense budgeting process is not readily available for most countries, with the exception of the United States.
(2) In most presidential systems, with the notable exception of the United States, the MOF sets spending limits and the President mediates any disputes between the MOF and line ministries. In parliamentary systems, countries that adhere to norms of "collective responsibility" might have a Council of Ministers, rather than the MOF alone, set ceilings.
(3) Ceilings can either be identified at the very beginning of the process or after a first stage when preliminary budget requests are made. It is difficult to set ceilings at the very beginning if the budget formulation process is a lengthy one because fiscal constraints are frequently not known with any certainty a full year in advance. However, in countries where spending ministries are not accustomed to moderating their requests to some reasonable level (e.g., an incremental increase over the previous year's spending), it is probably advisable to set a firm ceiling from the very beginning. [See Salvatore Schiavo-Campo and Daniel Tommasi, "The Budget Preparation Process," Asian Development Bank, available at http://www.adb.org/documents/manuals/govt_expenditure/Chap4.PDF. Accessed May 4, 2001.]
(4) Budget directors in each of the 20 countries responded to the survey. [See Alberto Alesina, Ricardo Hausmann, Rudolf Hommes, and Ernesto Stein, "Budget Institutions and Fiscal Performance in Latin America," (Inter-American Development Bank, Office of the Chief Economist, Working Paper Series 394), Table A4.]
(5) David Pion-Berlin, Through Corridors of Power: Institutions and Civil-Military Relations in Argentina. (University Park, PA: The Pennsylvania State University Press, 1997), pp. 138-39. Similarly, the Russian military has criticized the MOF for making across-the-board cuts of 5% without considering increased costs in the defense sector. Guedes da Costa notes that the Planning and Budget Ministry in Brazil often makes "arbitrary" cuts to the defense budget to bring spending in line with funds. [Thomaz Guedes da Costa, "Democratization and International Integration: The Role of the Armed Forces in Brazil's Grand Strategy," in David R. Mares, ed., Civil-Military Relations: Building Democracy and Regional Security in Latin America, Southern Asia, and Central Europe. (Boulder, CO: Westview Press, 1998), p. 228.]
(6) For more on the importance of a civilian Ministry of Defense, see Thomas C. Bruneau, "Ministries of Defense and Democratic Civil-Military Relations," Center for Civil-Military Relations, Occasional Paper # 11, (Monterey, CA: Naval Postgraduate School, August 2001).
(7) Personal communication with United States Air Force Major General (Ret.) Richard Goetze, April 3, 2001. See also Pion-Berlin, 1997, p. 134, and A. Premchand, "Budgetary Management in the United States and in Australia, New Zealand, and the United Kingdom," in Roy T. Meyers, ed., Handbook of Government Budgeting, (San Francisco, CA: Jossey-Bass Inc., Publishers, 1999), p. 110.
(8) The OMB is much weaker than the MOF in most countries, leaving room for some ministries to bypass it and discuss their budget proposal directly with the President. Historically, the Department of Defense is one of the sectors that has most often negotiated its budget directly with the President (and the Congress). [Harry S. Havens, "Budgeting and Policy-Making by the Legislature in the United States," in Budgeting and Policy Making. (SIGMA Papers No. 8. Support for Improvement in Governance and Management in Central and Eastern European Countries), p. 182. Available at http://www.oecd.org/puma/sigmaweb/pubs/pubnos/pubno8.htm. Accessed May 4, 2001.
(9) This was the case at least until 1982 in Spain where the Chief of Staff of the Armed Forces decided, with complete freedom, which weapons systems were to be acquired. [Antonio Marquina, "Spanish Foreign and Defense Policy Since Democratization," in Kenneth Maxwell, ed. Spanish Foreign and Defense Policy. (Boulder, Co: Westview Press Inc., 1991), p. 28.] Martins and Zirker cite a Brazilian admiral who acknowledges that the military has complete freedom to spend their budget as they see fit, as long as spending limits are respected. [Joao R. Martins Filho and Daniel Zirker, "The Brazilian Military under Cardoso: Overcoming the Identity Crisis," Journal of Interamerican Studies and World Affairs 42:3 (Fall 2000), p. 155.] Cruz and Diamant note for Latin America as a whole that "elected officials, in the main, favor streamlining military establishments; but after making resources allocations, they leave the armed forces to their own devices." [Consuelo Cruz and Rut Diamant, "The New Military Autonomy in Latin America," Journal of Democracy 9:4 (1998), p. 116.]
(10) Jelezov notes the historical inability of the Soviet armed forces to formulate broad strategic perspectives. [Boris Jelezov, Defense Budgeting and Civilian Control of the Military in the Russian Federation. October 1997. (Alexandria, VA: Center for Naval Analyses), p. 66.] Pion-Berlin makes a similar observation about the Argentine armed forces. [David Pion-Berlin, "The Limits to Military Power: Institutions and Defense Budgeting in Democratic Argentina," Studies in Comparative International Development, 33:1 (Spring 1998), p. 102.](11) Pion-Berlin, 1997, p. 142
(12) Peter Batchelor and Paul Dunne, "The Peace Divided in South Africa," manuscript, October 1997, pp. 11-12.
(13) Javier Meléndez Quiñónez, El gasto de defensa en Nicaragua: La toma de decisiones en la asignación de recursos. (Centro de Estudios Estratégicos de Nicaragua and the National Democratic Institute, 2000), pp. 33-35.
(14) Pion-Berlin, 1997, p. 169
(15) Meléndez.
(16) Martins and Zirker, pp. 151-152.
(17) On Brazil and Spain, personal communication, Professor Thomas C. Bruneau, August 1, 2001.
(18) Batchelor and Dunne, p. 12.
(19) This represents the process in most parliamentary and presidential systems; the United States and Great Britain are exceptions. In the United States, the process is extremely decentralized and a great number of committees are involved, even after a 1974 reform to create more coherence in the process. In Great Britain, financial powers are held directly by the House and not delegated to the Budget Committee, which consequently plays a reduced role.
(20) Schiavo-Campo and Tommasi. In Great Britain, the Parliament votes on the defense budget in its entirety, but there is no tradition of amending expenditures. This is very different from the United States and Germany where individual procurement decisions are frequently the subject of great controversy. [Tom Dodd, "Parliament and Defence: A summary of Parliament's role in scrutinising and controlling defence policy and the Armed Forces," RUSI Journal, 143:3 (June 1998), pp. 29-35.]
(21) See Jeanne Kinney Giraldo, "Legislative Control of the Military: The Comparative Experience," Center for Civil-Military Relations, Occasional Paper # 8, (Monterey, CA: Naval Postgraduate School, June 2001).
(22) The budget is usually presented three months in advance in Argentina, Germany, and Spain. A 1986 survey of developed and developing countries noted that the average was usually two months. ["Timing of the Budget," Volume I, Inter-Parliamentary Union, Parliaments of the World: A Comparative Reference Compendium, Second Edition. (New York: Facts On File Publications, 1986), p. 1068.]
(23) Meléndez, p. 32.
(24) Warren Krafchik and Joachim Wehner, "The Role of Parliament in the Budget Process," (Institute for Democracy in South Africa: Budget Information Service. No date.) Available at http://www.idasa.org.za/final/publications/Parliament1.htm.
(25) Yeon-Seob Ha, "Public Finance and Budgeting in Korea under Democracy: A Critical Appraisal," Public Budgeting and Finance (Spring 1997), p. 64.
(26) Alesina et al.
(27) Martinho Chachiua, "Internal security in Mozambique: Concerns versus policies," http://www.iss.co.za/Pubs/ASR/9.1/%20Security%20Mozambique.html, accessed January 4, 2001.
(28) In 1993, three categories were created: category A is an aggregated accounting presented to the National Assembly as a whole; Category B items are disaggregated and revealed without restrictions to members of the Committee on National Defense in the National Assembly; and Category C items are further disaggregated and revealed with certain restrictions to committee members. [Jong Chul Choi, "South Korea," in Ravinder Pal Singh, ed., Arms Procurement Decision Making. Volume I: China, India, Israel, Japan, South Korea, and Thailand. SIPRI. (Oxford University Press, 1998), p. 196.]
(29) Instead, military officers elected to the Duma, who have the appropriate clearances, oversee the classified details of the budget. [ David Betz, "No Place for a Civilian: Russian Defence Management from Yeltsin to Putin," paper presented to the International Studies Association, 41^st Annual Convention, Los Angeles, March 14-18, 2000.]
(30) Marybeth Peterson Ulrich, Democratizing Communist Militaries: The Cases of the Czech and Russian Armed Forces, (Ann Arbor: The University of Michigan Press, 1999), p. 92. The US intelligence budget, in contrast, was classified as secret until 1997. An excessively detailed budget proposal also can have its shortfalls, if budget lines are not grouped together into programs (or in some other fashion related to policy) so that the policy consequences of budget decisions are understandable.
(31) For more on this, see Giraldo.
(32) The kinds of legislative hearings which accompany the debate over other pieces of legislation are usually not possible, given the time constraints and pressures of the budget process. However, sectoral committees within the legislature should hold consultations and debates throughout the year, which would inform their input into the budget process. The holding of hearings outside the scope of the budget process itself avoids delays in getting the budget passed, and might increase the chances that general policy concerns and not particular lobbies inform spending decisions.
(33) See Bruce George and Alison Graham, "Defence Committees in Democratic and Democratising Legislatures," paper presented to the Workshop of Parliamentary Scholars and Parliamentarians, Berlin, August 1994, pp. 23-24.
(34) Joel M. Ostrow, "Procedural Breakdown and deadlock in the Russian State Duma: The problems of an unlinked and dual-channel institutional design," Europe-Asia Studies 50:5 (July 1998), Table 2. See also Jelezov, p. 53.
(35) George and Graham, p. 18. Bruce George is the longtime Chairman of the House of Commons Defence Committee.
(36) The Nigerian Constitution seems to give similar powers to that country's legislature. However, when the legislature recently tried to exercise these powers, the Executive argued that it was not entitled to do this and the legislature backed down. (Personal communication with Dr. Eva Busza, 24 July 2001.)
(37) Alesina et al. On Germany, see Krafchik and Wehner, p. 3. On the Philippines, see Gabriella R. Montinola, "Parties and Accountability in the Philippines," Journal of Democracy, 10:1 (January 1999), p. 136.
(38) Lisa Baldez and John M. Carey, "Presidential Agenda Control and Spending Policy: Lessons from General Pinochet's Constitution," American Journal of Political Science 43:1 (January 1999), p. 43
(39) Wendy Hunter, Eroding Military Influence in Brazil: Politicians Against Soldiers. (Chapel Hill: The University of North Carolina Press, 1997), p. 99.
(40) Cruz and Diamant (p. 121) argue that this is the case in most Latin American countries.
(41) For example, the legislature in Nicaragua played a role during the transition in writing the Código de Organización Militar, but since then have had little input into the national security planning and budget process (Meléndez, p. 38). Similarly, during the transition to South Africa, defense issues were important but have since taken a back seat.
(42) Hunter, p. 99.
(43) The resources provided by the military budget need not benefit a large number of legislators in order to have an impact on policy making. For example, in the United States the creation and findings of the special commission on intelligence in 1994 were shaped largely by the interests of one Senator in maintaining open the intelligence agencies in his state. [Loch K. Johnson and Kevin J. Scheid, "Spending for Spies: Intelligence Budgeting in the Aftermath of the Cold War," Public Budgeting & Finance. Winter 1997, p. 23.]
(44) Hunter, pp. 97-98.
(45) Hunter, pp. 98; pp. 108-109.
(46) Renato Cruz De Castro, "Adjusting to the Post-U.S. Bases Era: The Ordeal of the Philippine military's modernization program," Armed Forces and Society, 26:1 (Fall 1999), pp. 119-137.
(47) Martins and Zirker, p. 148; Hunter, p. 107.
(48) Cruz De Castro.
(49) Alesina et al.
(50) Jelezov, p. 44.
(51) Budgeting and Policy Making, pp. 85-86.
(52) Interview with high defense official in the Philippine government, January 14, 2001.
(53) Larry O'Toole, Anatomy of the Expenditure Budget. (SIGMA Policy Brief No. 1: Support for the Improvement in Governance and Management in Central and Eastern European Countries). Available at http://www.oecd.org/puma/sigmaweb/pubs/PBNO1.htm. Accessed April 25, 2001.
(54) The effectiveness of these reforms, however, is questionable (Choi, pp. 203-204). Procurement scandals are particularly common in the defense realm, given the large amounts of money involved and the high-profile nature of both the projects and the individuals making the decisions. On how procurement scandals have affected decision-making procedures in Great Britain and the US, see Andrew Cox and Stephen Kirby, Congress, Parliament and Defence: The Impact of Legislative Reform on Defence Accountability in Britain and America. (NY: St. Martin's Press, 1986). On Germany, see Regina H.E. Cowen, Defense Procurement in the Federal Republic of Germany: Politics and Organization. (Boulder, CO: Westview Press, 1986).
(55) Meléndez, pp. 40, 46.
(56) Ulrich, p. 92.
(57) For a short history, see "The Background and History of GAO," available at http://www.gao.gov/about/history.html, accessed May 15, 2001. For a more detailed analysis, see Harry Havens, "From Auditing to Policy Analysis: The Work of the General Accounting Office (GAO) of the United States," in Budgeting and Policy Making.
(58) Dodd, p. 3.
(59) Despite this requirement, legislative monitoring of public spending in Nicaragua tends to be sporadic (Meléndez, p. 39).
(60) Since 1982, the National Audit Office in Great Britain has published an annual Major Projects Report, focusing on arms procurement (comparing costs and dates with those projected). However, this is not received by the Public Accounts Committee until twelve months after the end of the fiscal year and is not available to the rest of the Parliament and general public for another five months (Dodd, p. 5).
(61) Even in the United States, where non-governmental organizations are strong and information on the budget relatively accessible, the monitoring of the defense budget is a formidable challenge. For a discussion of the challenges encountered by two leading NGOs to monitor just one portion of the defense budget -- spending in Latin America - see Latin American Working Group and Center for International Policy, "Findings and Recommendations from the 1998 edition," Just the Facts: A civilian's guide to US defense and security assistance to Latin America and the Caribbean. (Available at http://www.ciponline.org/facts/find99.htm. Accessed May 5, 2001.)
(62) Ulrich, p. 92.
(63) For example, countries provide arms transfer data to the United Nations Register of Conventional Arms. In June 1999, members of the Organization of American States signed an Inter-American Convention on Transparency in Conventional Weapons Acquisition. Unfortunately, many of the efforts of international organizations to collect data of military expenditures are largely unsuccessful. For example, the annual questionnaire on defense spending distributed by the UN Center for Disarmament Affairs received responses from only 20 countries in 1996, down from 33 in 1992. [See Paul George, "Defence Expenditures in the 1990s: Budget and Fiscal Policy Issues for Developing Countries," Bonn International Center for Conversion Conference Paper. Available at http://www.bicc.de/general/events/devcon/george.html. Accessed May 4, 2001.]
(64) Legislation referenced in Department of State, Annual Report on Military Expenditures, 1999: Indonesia. Available at http://www.state.gov/www/global/arms/99_amiex2.html. Accessed March 12, 2001.
(65) This is a common practice in Russia, where it has been said that specific budget data was more readily available in the late 1980s under communist rule than it is today (Ulrich, pp. 86-87). In Argentina, money spent on arms in 1995 was hidden under general Treasury expenditures, since the government's official position was that it wasn't making weapons purchases. [Thomas Scheetz, "Transparency, Accountability & Rational Decision Making in Defense Expenditures: The Argentine Case," (Bonn International Center for Conversion conference paper, February 12, 1998). Available at http://www.bicc.de/general/events/devcon/scheetz.html. Accessed on March 4, 2001.]
(66) Cruz and Diamant, p. 118.
(67) Theodore Friend, "Indonesia's Year of the Blue Carpet, Plus Several Pathologies and Five Personalities," Foreign Policy Research Institute E-Note, December 6, 2000, p. 3.
(68) As Falk and Shapiro note in their advice to NGOs attempting to influence the budget process, formulation is the most closed phase but also the most important. Decisions made at this stage can be influenced, however, by establishing relationships with policy makers or carrying out analyses that influence their decisions. [Stefan Falk and Isaac Shapiro, "Appendix 1: Background on Budgets," A Systematic Overview of the Different Aspects of Effective Budget Analysis, September 1999. Available at: http://www.internationalbudget.org/resources/guide/guide-06.htm. Accessed on March 4, 2001.] This work is part of the International Budget Project of the Center on Budget and Policy Priorities based in Washington, D.C. which has an extensive website and is oriented especially toward groups in civil society interested in influencing the budget process. See http://www.internationalbudget.org for more information.

---

PDGS - Documents