Budget is a foundation of economic management of a country, and therefore, in the context of developing countries, it is primarily an issue of development. It is often pointed out that the Philippine budget institution is characterized by its high degree of politicization. Such politicization is said to be one of the major causes of corruption, opaque plutocratic policy-makings, and frequent delays in budget approval and implementation. Politicization in the budget process, in concrete, is brought about by budget allocations aiming at providing benefits only to some regionally bounded constituency, which are often collectively expressed as “pork-barrel”. Presumably, there have always been several fundamental pitfalls in the discussions surrounding pork-barrel. First, to start with, the concept of “pork-barrel” has never been made clear, and thus the definition of the term is heavily dependent on the speakers or writers.
In many cases, the concept is simply too narrowly interpreted. Secondly, related to the first point, although under-the-table political deals between the legislators and the President related to pork-barrel have been often criticized, the way in which the political intentions of the President play a role in pork-barrel politics has not been sufficiently elaborated. Simplistic understanding of the pork-barrel does not recognize the fact that pork-barrel is not merely a possession of the local politicians, but rather an outcome of mutual interactions between the local politicians and the President. Thirdly, the previous discussions on pork-barrel politics seem to have failed to give a sufficient answer to the question as to why it matters at all; several important issues on both substantial and procedural aspect including the economic inefficiency of resource allocation involved in pork-barrel politics have attracted very little attention. Fourthly, and perhaps most importantly, talks on pork-barrel with little orientation to the actual budget figures are so prevalent – such abstract discussions may both exaggerate or underestimate the true magnitude of pork-barrels of different forms. Down-to-earth observations on actual budgetary documents are called for in order to capture the true behaviors of pork-barrel politics. The paper intends to shed light on pork-barrel in its entirety putting together the previously fragmented discussions from an institutional and economic perspective.
STRUCTURE OF THE PAPER
To preempt the main discussion of the paper, the author claims that pork-barrel politics takes place on four major stages during the course of budget formulation; a) lump-sum allocations, b) congressional insertions, c) disbursement specification/impoundment, and d) initial basic allocation. Corresponding to these stages of pork-barrel are the issues from different aspects to be discussed in the Philippine context, which could be graphically presented in the chart below: Universalism Common -Pool a. Lump-Sum Allocations Delayed Approval b. Congressional Insertions
Procedural c. Disbursement Specification/ Impoundment d. Initial Basic Allocation
The horizontal axis explains whether the issue belongs to a substantial dimension – namely, the fact that the pork-barrel politics would lead to an economic inefficiency by one way or the other, or a procedural dimension – such politicization raises concerns on unconstitutional practices, notable
delays in the congressional approval process, and most fundamentally, the plutocratic opaqueness wherein the important policy-makings between the President and the legislators take place outside of the public arena. Corruptive practice of the local legislators reportedly receiving kick-backs in implementing the projects in their electoral districts, which historically has been regarded as the central issue related to pork-barrel, is to some extent also a substantial issue since the resources lost in corruptive practice are considered to result in economic inefficiency, although the issue primarily emerges as a procedural matter. The vertical axis pertains to yet another classification. The issues related to budget politicization may stem from two opposite natures of the distributive politics.
On one hand, when budget it distributed universally across the districts regardless of their true fiscal needs, it naturally leads to an economic inefficiency which is characterized as a “common-pool” issue. On the other hand, when budget allocation is made in such a way that the shares each district receives are disproportionally weighed beyond the justifiable differences in fiscal needs as the result of political negotiations between the local legislators and the central government (or the President), it is assumed that there also exists resources lost in the lobbying and compromised deals. As opposed to the above “common-pool”, this issue is sometimes expressed as “common-agent” problem. The shades show which issues are most relevant to each stage of pork-barrel politics in the Philippine context. It should be noted that the way of illustration on this chart is never solid and inflexible – one type of pork-barrel can be interpreted as either universalistic or particularistic depending on the aspect one focuses on as well as how it emerges in actual budget process. Given the above, in Chapter II, the paper starts by explaining the regular budget cycle in the Philippines and various potential forms of pork-barrel which kick in at each stage of such cycle, including their historical paths of development. Chapter III shows in detail two economic presentations for the effects of budget politicization – common-pool issue and common-agent problem. Chapter IV presents the
procedural issues; constitutionality, corruptive practice of kick-backs and delays in congressional approval, while Chapter V examines the budget allocations on infrastructure projects in FY2009 and FY2010 budgets to indentify the characteristics of pork-barrel under Arroyo administration. Chapter VI concludes.
STAGES OF PORK-BARREL POLITICS
The term “pork-barrel”1 usually refers to the budgetary spending intended to benefit limited groups of constituents in return for their political supports. In many cases, the targeted beneficiaries are geographically defined voters, and therefore, pork-barrel politics is usually more prominent in countries with single-member electoral system as in the case of Philippine Congress (Lower House). It should be noted, however, that in some instances the beneficiaries are non-territorial basis, the typical examples of which exist in Europe. In the Philippine context, the head-aching fact is that at each stage of budget formulation, the chance of distributive politics kicks in, and therefore, the term is often used with different connotations. 1 The precise origin of the term “pork-barrel” is not entirely clear. One interpretation is that the term originated in a pre-Civil War practice in the US of giving slaves a barrel of salted pork as a reward to motivate them to compete among each other. The term is now used in numerous countries around the world with slight variations in nuance.
It is therefore critical to be conscious on what definition the writer or speaker is using the term in their statements. In order to capture what type of forms the “pork-barrel” may possibly take, it is also essential to understand the fundamental composition of Philippine budget.
1. HISTORY OF PORK-BARREL
The primitive form of what could be denoted as “pork-barrel” in the
Philippines emerged in 1922, when Public Works Act was passed. From 1950, it was made clear that the Congress was allowed to specify the projects which were to be funded under the Act, and from 1955, those legislature-sponsored projects were segregated from all other items under the name “community projects”. Eventually, these “community projects” were further subdivided into “miscellaneous community projects”, those specified by the Congressmen, and “nationwide selected projects”, those specified by the Senators2. Throughout the Marcos regime, the major arena for pork-barrel funding was Public Works Appropriations under this Act, a multi-year appropriation for infrastructure projects set aside from the rest of the national budget. Regarding this Appropriations, the Congress had complete freedom to change President’s proposal or even enact new and larger appropriations. Because Public Works Appropriations was a multi-year appropriation, and since not all parts of the appropriations coming out of the Congress are backed up by available funds, there was always a substantial accumulation of unimplemented projects from previous years3. After the democratization in 1986, pork-barrel was eventually incorporated into the general budget. Currently, there is no separate budget appropriation outside of the regular budget proposal. This, on one hand, improved the transparency of the budget system, at least superficially. Pork-barrel nevertheless did not cease to exist. Rather, they took different forms of budgetary items within the general budget and thus became even more invisible.
2. CURRENT BUDGET CYCLE
The budget cycle in the Philippines today is shown as the chart below. On the right hand side of the chart is the stages of pork-barrel politics which correspond to each phase of budget cycle, based on the classification shown earlier:
Eric Gutierrez, “The Public Purse” in Sheila S. Coronel (ed.) Pork & Perks: Corruption & Governance in the Philippines, Philippine Center for Investigative Journalism, 1998 3 Felisa D. Fernandez, “The Budget Process and Economic Development – The Philippine Experience”, College of Public Administration, University of the Philippines, Manila. 1973
Stages of Pork-Barrel Politics
Submission of Budget Proposals to DBM by Line-Agencies
Formulation of Budget Proposal under DBM
Congress: Committee on Appropriations / Other Committees Congress: Approval of Budget
Senate: Committee on Finance / Other Committees Senate: Approval of Budget
Bicameral Conference on Committee Congress: Final Approval Senate: Final Approval
President’s Veto Message
The Department of Budget and Management (DBM), the agency in charge of budget formulation/execution, formulates the budget proposal based on the requests it received from the line agencies, and submits it to the legislative body by July. The budget will be discussed first in Congress, and then in Senate. Even when the budget is approved in the two Houses, there are usually differences in the two versions of the approved budget, since Congress and Senate make changes to the proposed budget independently, as also in the case of other regular bills in the Philippines. Such differences are negotiated during the Bicameral Conference Committee. As a matter of terminology, the proposed budget by the President is called National Expenditure Program (NEP), the initially approved budget by both Houses are
called General Appropriations Bill (GAB), and the final approved budget is referred to as General Appropriations Act (GAA). After the two Houses finally approved a uniform budget, then it will be subject to the veto message of the President, where he refuses to implement particular items and clauses which he considers to be inappropriate. Ideally the budget is supposed to be ready for disbursement by the end of the year. Even after the budget is approved, whether and when each budgetary item is actually going to be disbursed is yet another question. Disbursement of any budget item is dependent on the “go” signals by DBM. Depending on the size and the importance of the item, such decisions may be made by working level officials, the Secretary (the head of the Department), or by the President himself.
This is thus said to beanother occasion where the President is able to make interventions into budget based on his political intentions. For the sake of intuitive understanding, the author would dissect the budget into three components; first, there is a group of “lump-sum allocations” in which the Congressmen and the Senators are given fixed amounts which they are entitled to allocate practically with a complete freedom. This category comprises of Priority Development Assistance Fund (PDAF) and fixed slot for infrastructure allocation under Department of Public Works and Highways (DPWH). Secondly, there is a group of “special purpose funds”, the use of which is highly discretionary by the President and by the executive branches under him4. Lastly, all the other items which do not fall under either of the above two categories are considered to be regular budgetary items.
3. STAGES OF PORK-BARREL POLITICS
Given the above structure of budget, as mentioned earlier, it is possible to enumerate four potential stages of pork-barrel politics in the Philippine budget; a) lump-sum allocations, b) congressional insertions, c) disbursement specification/impoundment, and d) initial basic allocation. In most cases when the concept of “pork-barrel” is at stake, it has been primarily the lump-sum allocations which were assumed to be the basis of
discussion, while other potential stages of pork-barrel politics have attracted far less attentions. In any case, it has to be underscored that the four stages do not necessarily have a one-to-one relationships with certain budgetary items. For instance, although a lump-sum allocation by itself is one form of pork-barrel, when it comes to the point where the President exercises his power to control its disbursement for his political motivations, it turns into what should be characterized as a disbursement impoundment by the President. In this sense, the major departure of this paper from the traditional understanding on the pork-barrel is that the author captures pork-barrel not as a group of chunks of money allocated to specific districts, but as a series of dynamic process by which the political interactions between the President and the local politicians kick in at each stage of the budget cycle.
Calamity Fund, Contingent Fund, Economic Stimulus Fund, E-Government Fund, International Commitments Fund, Miscellaneous Personnel Benefits Fund, National Unification Fund, Pension and Gratuity Fund, Tax Expenditure Fund. In fact, technically, PDAF is one of the special purpose funds. On the other hand, the fixed slot for infrastructure allocation under DPWH is among the regular budgetary items. However, by nature of those items, they could be grouped together.
The history of lump-sum allocations is an epitome of pork-barrel politics of the Philippines. Lump-sum allocation was first introduced during the Aquino administration when the Mindanao Development Fund (MDF) and Visayas Development Fund (VDF) were established in 1989. These funds were created to alleviate poverty in relatively less developed areas and thereby mitigatingregional gaps within the country; each Congressman from these electoral districts was entitled to implement projects totaling to 10 million pesos5.
A year after the implementation of the two Funds, legislators from Luzon requested that they too be given equal share. In response to this, the Countrywide Development Fund (CDF) was created to cover all electoral districts in the Philippines with the incorporation of the MDF and VDF6. This is a fundamental change in nature of the Funds – the original two Funds were created to equalize the levels of regional economies, which supposedly had an ample policy rationale, but now that the coverage of the integrated fund was expanded to nationwide, the Fund turned into a mere cash dispenser for the legislators. During the Ramos administration, the disbursement control of CDF is said to have been extensively utilized by the President to sustain his political influence on legislative body. The result was a growing criticism on pork-barrel reflecting the widely-spread recognition that the Fund was a ground for non-transparent under-the-table deals between the legislators and the President7. During the campaign period for the succeeding presidential election, one of the policy pledges that the President Estrada put forward was “abolishment of pork-barrel”. In fact, for his first year in office, he proposed a budget with deletion of CDF. The Congress responded by inserting equivalent funds for compensation, namely, Lingap para sa Mahirap Program Fund and Rural/Urban Development Infrastructure Fund, amounting up to 2,500 million pesos and 5,458 million pesos, respectively. The former is to be expensed to soft projects, while the latter is to be used for hard infrastructure spending. President Estrada had to accommodate the creation of such funds as his political compromise. From the year 2000, CDF was virtually revived under the name “Priority Development Assistance Fund (PDAF)”.
The total amounts of the Funds were 480 million pesos for MDF and 240 million pesos for VDF. The initial amount of the Fund was 2,300 million pesos. 7 Earl Parreno, “Pork” in Sheila S. Coronel (ed.) Pork & Perks: Corruption & Governance in the Philippines, Philippine Center for Investigative Journalism, 1998 6 Mindanao Development Fund (MDF) Countrywide Development Fund （CDF） Visayas Development Fund (VDF) Priority Development Assistance Fund （PDAF）
The members of the Congress and the Senate are given a wide-range of discretion for project identifications for PDAF. Under the Arroyo administration, it is stated in the GAA that the projects should be selected in accordance with Government’s prioritized policy areas called “Ten-Point Legacy Agenda”8. In practice, however, fairly a wide range of projects can be qualified with some rhetorical efforts. Typical examples of “soft projects” are human resource development or such in-kind support as provision of medicine to local hospitals, but as is identified in the said Agenda, small infrastructure projects may also be accommodated. The other existing lump-sum allocation is a part of infrastructure budget under DPWH. The amount of allocation for each legislator under PDAF and the DPWH lump-sum allocation is merely a gentlemen’s code between the legislators and the President as well as among the legislators themselves. For FY2010, such amount is agreed to be 200 million pesos for incumbent 24 Senators (80 million for PDAF and 120 million for DPWH, respectively) and 70 million for 228 Congressmen (30 million for PDAF and 40 million for DPWH, respectively)9. During the congressional process, there is usually a series of negotiations among the legislators to transfer one’s share to another within these allocations or barter the PDAF share with DPWH lump-sum allocation (on one way or the other), until they finally come up with the final amounts of allocations. It is also almost always the case that the legislators make an effort to increase the size of the Funds themselves by form of congressional insertions, which itself is another form of pork-barrel10.
1. creation of six to ten million jobs by a) tripling loans for small business owners, and b) development of one to two million hectares of land for agricultural business, 2. education for all, thru a) the construction of new schoolbuildings, and classrooms, b) provision of books and computers for
students, and c) scholarships to poor families, 3. balance the national budget in view of the country’s debilitating deficit woes, 4. decentralization of progress and development across the country through the development of transportation networks like the roll-on, roll-off ferries and digital infrastructure, provision of power and water supply to all barangays, 6. decongestion of Metro Manila by forming new cores of government and housing centers in Luzon, Visayas and Mindanao, 7. development of Clark and Subic as the logistics center in Asia, 8. automation of the electoral process, 9. peace agreements with rebel groups; and 10. “closure of wounds” caused by division due to EDSA 1, 2 and 3. 9 Aside from this, there is a separate allocation for the 57 party-list members (30 million pesos for PDAF and 40 million pesos for DPWH, respectively). 10 As shown in the following chapter, however, when those insertions are vetoed by the President, it is not supposed to be disbursed as in the case of other congressional insertions.
It is a long-standing practice in the Philippines that the legislators insert additional budgetary items for their respective constituency. One lawmaker reportedly testified anonymously that “CDF is loose change compared to what one can get from congressional insertions…. Through the (congressional insertions)…, we are able to get additional powers to decide on how billions more in the national budget will be spent”11. As elaborated in details later on, congressional insertions usually take the form of “reallocations” from the existing budgetary items in the proposed budget to avoid an expansion of total budget size. Whether such inserted budgetary items are in fact disbursed regularly is not entirely clear. So long as FY2010 is concerned, the President in her veto message made a clear statement that all insertions
are practically rejected12. This sort of explicit statement has not existed at least since 2000, however, the author was not able to make it clear whether the existent or non-existent of such statement makes a practical difference in terms of the disbursement of the inserted items. As shown in the following chapter, as of the end of June, most of the congressional insertions in FY2010 budget have not been disbursed consistent to the veto statement, although it is somewhat puzzling why some of the insertions made by a selective group of Congressmen were actually disbursed. In any case, it is obviously told that in previous years, the congressional insertions were disbursed on a massive scale as one representation of pork-barrel politics. In this sense, there is no doubt that the practice of congressional insertions is another occasion where the pork-barrel politics arises. Aside from the Congress and Senate, Bicameral Conference on Committee (BICAM) usually turns out to be another battle-field of pork-barrel politics when it comes to congressional insertions. In spite of its mandate to compromise the two versions of budget bills, the Committee presumably by and large adds up the insertions made in the two Houses, and may even accommodate additional requests from both groups of the legislators. Evidently, once the budget passes the BICAM, it is further fattened from the approved budget bills as it was also the case for FY2010 budget13. From an institutional point of view, it must be also pointed out that there is no constitutional ground for this institution. Furthermore, the discussions in BICAM are not made open to the public but rather kept in a black box. For these reasons, the Eric Gutierrez, ditto.
The President’s will to reject the disbursement corresponding to the congressional increases is presented in the veto messages, usually by conditioning it to future additional revenue measures. The veto message for 2010 budget states as follows; “…Moreover, release of the increased items of appropriations is subject to the identification by Congress of new revenue measures in support thereof.” 13 See again the Table1. What could be observed there is a general increase across budgetary items from GAB, the only distinctive exception to which is the interest payments. 12 11
Committee is often criticized as another opaque factor in budget process, which is also a cause of further delays in the fiscal cycle of the country14. c. Disbursements Specification/Impoundment by the President
The distribution of national budget is determined as the equilibrium of the game among the stakeholders, namely the legislators and the President. On one hand, it is the legislators who make efforts to acquire additional funding to their electoral districts, but on the other hand, it is ultimately the President who gives away those requested funding in political bargaining. It is especially convincing when one thinks of the budget cycle in the Philippines wherein the President, in the end, has a final say on actual disbursement of the approved budget. In this sense, it is fair to say that all budget items including those which are characterized as different types of pork-barrels eventually end up being subject to the disbursement control by the President at the final stage of the budget process. The paper categorizes this process as another occasion for pork-barrel politics. First of all, it has to be noted that there is a certain group of fiscal sources which are specifically referred to as the “President’s pork-barrel”. One such pork-barrel is “special purpose funds” as introduced earlier, over which the President enjoys a close-to-perfect liberty for their disbursements. The purposes of expenditure the funds are vaguely defined in most cases, and thus the President has a wide discretion as to what and, more importantly, where the money will be spent on. Similar to these funds are the “savings” in various budget categories15. Such “savings” are often utilized for discretionary spending in the middle of the fiscal year under the control of the President. Naturally, the legislators will lobby the President to get grants out of these funding sources.
The problem of these “President’s pork-barrel” is that their expenditures are highly invisible – the author was not able to identify such specific expenditures, which remains to be an issue for future studies. Secondly, it is generally claimed that the President often tries to enhance his political influences by withholding the fund releases in one district or another. This is said to be especially true when it comes to the abovementioned lump-sum allocations (PDAF and DPWH allocations), which are, by nature of the allocations, directly attached to a particular district. It is precisely for this reason that a specific provision was embedded in the FY2010 GAA, which calls for a prohibition against impoundment of appropriations16. It has to be nevertheless made 14 15 16
Benjamin E. Diokno, “Where Deep Secrets Lie”, Business World, February 11, 2010 Benjamin E. Diokno, “Busted Budget”, Business World, June 24, 2010 Christian V. Esguerra, Philip Tubeza, “Aquino Won’t Control Pork”, Daily Inquirer, May 21, 2010 ether such impoundment truly exists in the first place. For instance, as it will be presented in the next section, usual practice of congressional insertions is also extended to the lump-sum allocations, and these insertions are not necessarily what must be delivered from the institutional point of view. Careful study has to be made as to what extent these allegations to the President for the arbitrary manipulation of fund releases are valid. Also, assuming such impoundment in fact does exist, it is doubtful that such provision in GAA is really needed beyond a political performance – it may be just a matter of the will of the President. In fact, this provision is already a déjà-vu of the similar provision under the Ramos administration17 or the rule which was called as “what-you-see-is-what-you-get (WYSWYG)” under the Estrada administration18, and the new provision seems to have little additional value to what is already in place. d. Initial Allocation of Basic Budgetary Items
In the Philippines, as in any other countries, it is initially the line-agencies which first file their budget requests to the fiscal authority. The DBM, on the other hand, examines all the submitted requests, and formulates them into a consolidated budget plan. Political interventions may also take place at any point in this process. Despite the fact that the initial distribution of basic budgetary items constitutes the fundamental of the budget, this phase of budget cycle has been almost ignored until now in the context of budget politicization. There is an inherent difficulty in observing the existence of pork-barrel politics at such initial stage of
budget formulation. Since this type of skewed budget distributions are, by definition, already “embedded” in the original proposed budget, if ever – they are not visibly described in any of the published budgetary documents. Therefore, in order to demonstrate its existence, one has to resort to the indirect approach on the basis of econometrical work, which the paper presents in the later chapter.
Earl Parreno, ditto. Benjamin E. Diokno, ditto., June 24, 2010
ECONOMICS OF PORK-BARREL
Pork-barrel has two mutually-conflicting characteristics, both of which lead to economic inefficiency of budget distribution. One is universalism, whereby the distribution of budget is made in a uniform fashion across districts. The inefficiency brought about by such practice is that the districts will be engaged in a cartel to receive the equal amount of shares, regardless of whether such level of funding is truly necessary in their districts. At the same time, economic inefficiency may be also caused by particularism of budget distribution, when a regional differentiation in funding is a result of lobbying activities by the districts to the central government (or to the President). In this case, inefficiency exists in the sense that some resources are lost because of such lobbying activities. 1. Universalism: Common-Pool Problem
One of the main characteristics of a pork-barrel project is that the costs are dispersed through general taxation while those who profit from it are only limited in number19. It is usually pointed out that the underlying problem in pork-barrel is the so-called “common-pool” problem. Sato (2006)20 described the model of common-pool issue in the case of redistributive
politics based on the preceding work by Inman and Rubinfeld (1996)21. Suppose a local public good g is provided in a specific district, with a marginal benefit MB ( g ) . The marginal cost for the provision of this public good is MC(g ) . Out of this marginal
cost, only a portion 0 1 is to be shouldered by that particular district. Given that the federal funding is evenly spread among all districts in the country, the share of cost for each district would be:
1 MC ( g )
where n is the number of districts. For the district p which receives the benefit of the public good, their own marginal cost could be written as: 1 MC p g MC g . n
If the local legislator of this district is purely motivated to increase local benefit, which in Thomas D. Lancaster, “Electoral Structures and Pork Barrel Politics”, International Political Science Review, 1986 Motohiro Sato, “Seifukan Zaisei Kankei no Seiji Keizai Gaku (Political Economy of Intergovernmental Fiscal Transfers)”, Financial Review, Policy Research Institute, Ministry of Finance of Japan, 2006 21 Robert P. Inman, Daniel L. Rubinfeld, “Designing Tax Policy in Federalist Economies: An Overview”, Journal of Public Economics, 1996 20 19
many cases is what a congressman who aspires to be reelected does, then he shall try to receive a grant at the level where MB p MC p . Naturally, however, this would not yield the socially optimal solution. Another scenario may also be hypothesized where the provision of such local public good, although it is primarily addressed to a specific district, has an unintended spill-over effect to other regions such that only a part of the benefit (0 1) of the public good rests in that district while the remaining benefit in fact goes to the other region. In such case, if and
only if the share of the primary beneficiary district is determined as such that: MB ( g ) MC p ( g )
1 MC ( g ) n
and therefore the following holds:
n 1 , n 1
then the optimality would be achieved. Needless to say, by nature of pork-barrel, it is not practical to expect this to happen. It is therefore not hard to predict that in most cases what happens is an exceeding marginal social cost to the marginal benefit precisely because of the transfer of the cost from the beneficiary district to other districts. Despite this seemingly inefficient outcome, legislators are inclined to make a mutual commitment to approve provisions of such local public goods, which is commonly perceived as “universalistic norm”. It is sometimes colloquially expressed as “You scratch my back, and I’ll scratch yours”. Once all legislators adhere to the norm, then it provides a stable Nash equilibrium and no one tries to deviate from it. Here lies the fundamental problem in redistributive politics – budget allocations are not designed for the welfare maximization of the society as a whole but rather it is materialized as a result of the equilibrium of the interactions by the stakeholders taking part in the game. In that sense, pork-barrel distribution may potentially exist in all societies with an inherent implication of welfare inefficiency. It could be also pointed out that the lump-sum allocations in the Philippine budget – PDAF and fixed amount allocation under DPWH – are nothing but a straightforward institutionalization of such universalistic behavior. 2. Particularism: Common-Agent Model
The above is a horizontal model describing the implicit cartel among the legislators. If one tries to incorporate the interaction between the central government and local governments, some other model has to be introduced. Persson and Tabellini (1994)22 and Sato (2006)23 22
Torsten Persson and Guido Tabellini, “Does Centralization Increase the Size of Government?”, European Economic Review,
described this using common-agent model with competing local units as the principals and the central government as their common-agent. The central government can be interpreted as the President in the Philippine context who has the authority for final budgetary decision-making, while local governments are considered to be represented by legislators whose primary interest is to satisfy their geographical constituents so that he will have a better chance to be re-elected; in the case of the Philippines, it is the Congressmen who play the role as principals. Assuming the utility function of an anonymous average individual in the locality j as u j , the political agenda of the local agent is shown as: maxu j c j H ( g j ) s.t. c j y j t j where c j is the consumption of goods, y j is the level of production in this locality, t j represents a local lump-sum tax, while H is a concave function with the local public goods g j . In the case where there is no transfer from the central government, naturally t j g j holds.
On the other hand, if one supposes a case where there is a fund transfer from the central government as the result of fiscal redistribution, then the new budget constraint would be denoted as: cj yj 1 J gi n i 1
where n is the size of population in the country, while J is the number of localities in the country 1,2,3,…, j,…J . In such case, the utility function is now derived as: uj yj 1 J g i H g j n i 1
Furthermore, to take lobbying activities into consideration, we further import a contribution schedule to the central government as the consequence of lobbying: z j g j
which provides the new budget constraint:
c j 1 t j z j 1 1 J gi z j n i 1
A concavity is also assumed for . For the sake of simplicity, the local production is assumed to be unity. Rather than the utility of a local individual, now we turn our attention to the utility function of the central government, which is basically the sum of utilities of each locality as well as the profits the central government enjoys in return for the grants it gives away. Such profits may take various forms, both tangible and non-tangible, but in the context of this paper, it primarily assumes legislative supports of Congressmen to the Presidents. The utility function
1994 23 ditto
of the central government is denoted as:
v u i z i s.t. ui ci H g i 1 t zi H ( g i ) J J i 1 i 1
A realistic assumption should be adopted regarding the utility arising from the lobbying practice, that is 0 1 . Then, from the first order condition of such utility function: v H Z 1 1 0 g j g j g j
H Z 1 (1 ) 1 g j g j
could be derived, which shows the potential under-provision of the public good at a sub-optimal level. Of course, along with such sub-optimality, certain amount of resources are lost from the economy by the consequence of the lobbying activities as we assume to be smaller than one.
In the Philippines, it is actually the procedural matters which have been the central issues of debate when it comes to pork-barrel. The issues are mainly the constitutionality of both lump-sum allocations and the congressional insertions, corruptive practice of kick-backs whereby some local politicians are arguably receiving in implementing their pet projects, and the chronicle delays in budget approvals. 1. Constitutionality
In 1994, four years after the creation of CDF, the constitutionality of the Fund was put at stake by the Philippine Constitutional Association24. Their claim was that the CDF infringes the executive power by allowing the legislators to specify and implement public projects25. It was concluded that CDF was constitutional, the significance of which is that the judiciary admits the superior power of legislative body over executive power on annual budget26 This is in fact Philconsa vs. Enriquez, et al. Kenichi Nishimura, “Estrada Seikenka no Pork-Barrel Seido (Pork-Barrel under Estrada Administration)”, University of Osaka, 2002 26 “Under the Constitution, the spending power called by James Madison as ‘the power of the purse,’ belongs to Congress, subject only to the veto power of the President. The President may propose the budget, but still the final say on the matter of appropriations is lodged in the Congress. The power of appropriation carries with it the power to specify the project or activity to be funded under the appropriation law. It can be as detailed and as broad as Congress wants it to be.” 25 24 somewhat counter-intuitive in a sense given the high degree of discretion of the President on annual budget in the Philippines, but in any case, the basic tone of this ruling was practically replicated in a relevant case in the Supreme Court in 200127. The above judicial decision also provides theoretical background for those who claim the general superiority of the legislative power over administrative power concerning national budget28. In fact, constitutionality is also an issue in terms of congressional insertions during the budget cycle, and it again involves the issue of superiority among powers. In the practice of congressional insertions, usually the debt-services (interest payment of the national government) are trimmed down to be “reallocated” to some pet projects of the legislators29.
There is a reason as to why the legislators do not insert their new budget items in a more straightforward fashion; under the Philippine constitution, augmentation of budget by the legislative body from the budget proposal is explicitly prohibited30. Such fiscal rule aiming to prevent ruthless expansion of the size of budget, although it is apparently a strong constraint to the democratically elected decision-making body, is not particularly queer and unique to the Philippines but also exists elsewhere around the world, and the rule itself seems to have enough policy rationale31. As a loophole for this restriction, however, it has from some point become an established practice that the legislators cut the debt-services in return for the additional items they inserted to equalize the total size of budget. In regard to this practice, the following concerns are raised: While the legislators claim that the interpretation of the constitutional provision “…may
not increase…” only refers to the total size of budget, the wording is more commonly understood to refer to each budget item. This issue is analogous to the question as to whether an act of reallocation should be allowed in the first place. During Aquino administration, the legislators have limited themselves based on the stricter interpretation, but then it was gradually loosened. 27 28
Moreover, particularly regarding the debt-services, it is actually deemed to be one of the Sarmiento et al. vs. The Treasurer of the Philippines, et al. In 2008, a paper titled “Understanding the ‘Pork-Barrel’” was published by Prospero C. Nograles, then speaker of Congress and Edcel C. Lagman, then chairman of Committee on Appropriations aiming at justifying and advocating PDAF. The paper also cites this decision by the Supreme Court as its main point of argument. 29 See Table 1 for the figures for 2010 budget. Interest payments are cut down from 340,812 million pesos to 276,212 million pesos, which more or less pays-off the increase caused by the insertions. 30 “The Congress may not increase the appropriations recommended by the President for the operation of the Government as specified in the budget.
The form, content, and manner of preparation of the budget shall be prescribed by law.”(Article VI, Sec.25(1), The 1987 Constitution of the Republic of the Philippines) 31 According to OECD Budget Practices and Procedures Survey (2008), approximately 65% of the countries surveyed around the world have some sort of restriction on legislative bodies in budget process in place. Also, one study shows such institutional limitation on legislative power over budget is an effective measure to maintain fiscal disciplines especially in countries with high degree of partisan fragmentation (Joachim Wehner, “Institutional Constraints on Profligate Politicians: The Conditional Effect of Partisan Fragmentation on Budget Deficit”, London School of Economics, 2008). Doubtlessly, this argument fits well in the Philippine context.
“automatic appropriations” along with such other items as Internal Revenue Allotment (IRA) to local government units or contributions to Government Service Insurance System (GSIS) whereby the technical computation by the administrative branch (DBM) is basically respected. Irrelevant to the decreases in the congressional process, once the fiscal year begins, the government usually disburses these budget items as much as the originally proposed amount based on such understanding. This is precisely the reason why the cut in debt-services is often called an “imaginary cut”, which may bring about a consequential increase in total size of budget if any of the “reallocated” items are actually disbursed. 2. Corruptions – Kick-Backs / Plutocracy
As mentioned earlier, the topic of pork-barrel has been traditionally seen as an issue of corruption more than anything especially in relation to lump-sum allocations. While there is no hard-core evidence as to how exactly the pork-barrel related corruptions take place, there are numerous anecdotes which indirectly imply the existence of rebates and misuses of funds. Although the ground for estimation is not entirely clear, one study suggests that in the case of PDAF allotment for a certain region, around 40% of the
expenditure goes to the suppliers, while the rest is set aside and divided up for the legislator (approximately 45%), head of the implementing agency (approximately 10%) and for congressional aide (approximately 5%). Historically, DPWH allocation has been put under discussion less often than PDAF. The reason for this is presumably that the latter is meant to be directed on “hard” infrastructure projects whereas PDAF is meant to be spent for other “soft” projects, the implementation of which is more obscure and thus raises the doubt for corruptive practice. Moreover, the study also claims that it is only about 64% of the disbursement that is spent for the project implementation proper, and the rest is shared by the legislator (approximately 16%), provincial, city or municipal engineers (approximately 10%), mayor (approximately 7%) and barangay captain32 (approximately 3%). It is said to be a frequent observation that the people put in charge of pork-barrel projects are relatives or close confidantes of the legislators, and also, the public biddings are just ceremonial since the projects are usually already divided among the contractors even before the release of funds33. Moreover, aside from the aspect of actual disbursement, some argue that the under-the-table deals between the President and the legislators per se constitute the basis of plutocratic
The chief of barangay, the smallest local administrative unit in the Philippines. Earl Parreno, ditto.
opaqueness in the policy-making process. According to them, ideally, those who are engaged in power should exchange their views and opinions in public arena of policy-making. 3. Delayed Approval
The budget process in the Philippines is usually delayed as pork-barrel politics comes into each phase of the above budget cycle. It is rarely the case that the budget process is completed on time – aside from the most recent budget for FY2011, the last budget which was processed by the year
end is the FY1999 budget (approved on December 30, 1998). For the other years in the past ten years, the budget formulation continued in the new fiscal year34. For the FY2001, FY2004 and FY2006, Congress failed to approve the annual budgets, thus the budgets from the previous years were reenacted based on the relevant constitutional clause35. Needless to say, succession of budget from the previous year is supposed to be only temporary in nature, and it is clearly a deviation from the original intention of the law that the government relies on obsolete budgets for one entire year. Another drawback for such practice is the lack of accountability. Technically, reenacted budgets also go through congressional hearings etc, but naturally, such process attracts less attention from the public compared to the case of a brand new budget, and thus it is pointed out that both the legislators and the President are practically given wider discretions in terms of the actual allocation of funds. Because of the general delay in budget process year after year, formulation of supplementary budget is rarely successful in the Philippines, which may be referred to as another major negative repercussion of pork-barrel politics. From the macroeconomic management standpoint, a large amount of fiscal expenditure is sometimes required for rapid counter-cyclical action, but under a permanent delay in budget cycle mentioned above, by the time a supplementary budget is in place, most likely the timing is already lost. In last ten years, it was only after the eruption of Mt. Pinatubo, which brought a huge disaster and rescue needs, that a supplementary budget came into existence36. For FY2009, the government announced “Economic Resiliency Plan (ERP)” whereby they claimed they have implemented a fiscal stimulus operation with a size amounting up to 330 billion pesos. However, unlike many other
34 Each budget approved on FY00: February 16, FY02: January 21, FY03: April 1, FY05: March 15, FY07: March 22, FY09: March 13, FY10: February 9, respectively. 35 “If, by the end of any fiscal year, the Congress shall have failed to pass the general appropriations bill for the ensuing fiscal year, the general appropriations law for the preceding fiscal year shall be deemed reenacted and shall remain in force and effect until the general appropriations bill is passed by the Congress.”(Article VI, Sec.25(7), The 1987 Constitution of the Republic of the Philippines) 36 One of the special
purpose funds in the national budget is “Unprogrammed Fund” whose detailed disbursement is left entirely open to the future needs. It has to be underlined that sometimes the disbursement from this fund in the middle of a fiscal year is referred to as a “supplementary budget” as it was the case in 2009. This, however, it not a supplementary budget in the true meaning of the term, and the size of expenditure is relatively limited.
countries that implemented additional fiscal measures with a supplementary budget, ERP was little more than rhetoric in the sense that such fiscal disbursement is merely a conceptual figure computed as a gap between FY2009 and FY2008 budgets. For FY2009 alone, it was fortunate that by the time the annual budget was under formulation the slowdown of the economy was already clear. However, in order to be able to truly catch up with fiscal needs for future macroeconomic fluctuations which may take place in the middle of fiscal years, a rapid budget process without excessive interventions of pork-barrel politics would be the bottom-line.
EMPERICAL ANALYSIS ON 2009/2010 BUDGETS
This chapter attempts an empirical analysis on the budgets of the past two years. It is unfortunately rare to see arguments on pork-barrel in the Philippines built-up on even the most primitive observations on the budget figures, and the author has set such effort as the point of departure. Capturing the pork-barrel politics is easier on the stages of lump-sum allocations and congressional insertions; the former is already solidly institutionalized including the amount to be distributed, and the latter can be observed by comparing the approved budget to the proposed budget (Table 1). Trickier to observe is the remaining two stages of the pork-barrel politics; President’s disbursement specification/impoundment, and initial basic allocation – especially, when it comes to initial allocation, econometrical approach seems to be the only feasible way in capturing the
true picture of it. 1. Previous Literatures
Analysis on budget politicization initially developed in the United States. Grossman (1992)37 conducted an empirical analysis on 49 states out of the 50 states in the US using four data points; 1974, 1977, 1980 and 1983. The results showed that several political characteristics including the similarity of party affiliation between federal and state politicians are statistically significant to the size of grants that one state receives from the federal government. Bickers and Stein (1994) 38 developed a critical view showing that neither all legislators nor constituents are so much aware of the flow of new awards. Arguably, the legislators who are electorally vulnerable are most likely to seek an increase in projects for their districts, while politically attentive constituents as members of interest groups are expected to be more aware Philip J. Grossman, “A Political Theory of Intergovernmental Grants”, Public Choice, 1994 Kenneth N. Brickers and Robert M. Stein, “Congressional Elections and the Pork Barrel”, The Journal of Politics, 1994 of increase in new awards. Bickers and Stein (1996)39 found out that the flow of new awards early in a congressional term is higher in districts where the incumbent was elected in an open seat race by a narrow margin. Bickers and Stein (2000)40 compared the distribution of federal outlays of the Republican-controlled 104th Congress with the 103rd Congress under the Democrats, observing the outlays by type of program and recipient category. Alvarez and Saving (1997) 41 , focusing the period during 1984 to 1988, showed that the incumbent legislators who obtained federal largess were rewarded by their constituents and more successful in succeeding elections. There are studies which focus on particular spending items of pork-barrel. Evans (1994)42, based on the anecdotal as well as quantitative evidence on “demonstration” projects under the omnibus 1987 highway reauthorization in the US House of Representatives, presented the fact that those projects were used to help committee leaders to construct supporting coalitions for legislative packages.
Del Rossi (1995)43 investigated the federal financing of water resource development. Contrary to the common understanding that pork-barrel is purely politically motivated, the study argues that economic factors have been important in determining water source spending. Wilson (1986)44, on the other hand, examined the district-level data on rivers and harbors appropriations from 1889 through 1913 when the structural changes occurred within the House over relatively short period of time. It is claimed that the distribution pattern shows somewhat the nature of universalism, and is not driven by “need-based” computation. The research by Thompson (1986)45 is unique in the sense that it focused primarily on pork-barrel to one limited state – North Carolina, during 1980’s. Over the studied period, the pork-barrel system in the US has evolved, which is largely due to the change in political dynamics. Past literatures in this field also include studies on various other countries outside of the US. Costa-i-Font, Oreggia and Lunapla (2002)46 suggested empirical evidence which supports the existence of “local pork-barrel politics” in regional allocation of public investment by Mexican Kenneth N. Brickers and Robert M. Stein, “The Electoral Dynamics of the Federal Pork Barrel”, American Journal of Political Science, 1996 40 Kenneth N. Brickers and Robert M. Stein, “The Congressional Pork Barrel in a Republican Era”, The Journal of Politics, 2000 41 R. Michael Alvarez and Jason L. Saving, “Deficits, Democrats, and Distributive Benefits: Congressional Elections and the Pork Barrel in the 1980s”, Political Research Quarterly, 1997 42 Diana Evans, “Policy and Pork: The Use of Pork Barrel Projects to Build Policy Conditions in the House of Representatives”, American Journal of Political Science, 1994 43 Alison F. Del Rossi, “The Politics and Economics of Pork Barrel Spending: The Case of Federal Financing of Water Resources Development”, Public Choice, 1995 44 Rick K. Wilson, “An Empirical Test of Preferences for the Political Pork Barrel: District Level Appropriations for River and Harbor Legislation, 1889-1913”, American Journal of Political Science, 1986 45 Joel A. Thompson, “Bringing Home the Bacon: The Politics of Pork Barrel in the North California Legislature”, Legislative Studies Quarterly, 1986 46 Joan Costa-i-Font, Eduardo Rodriguez-Oreggia and Dario Lunapla, “Political
Competition and Pork-Barrel Politics in the Allocation of Public Investment in Mexico”, Public Choice, 2003 39 central government. A positive relationship is shown between the regional allocation of public investment and support for the central ruling party. Another Latin American nation which has been covered in a relevant research is Brazil. Ames (1995)47 found that the voting patterns of Brazilian Congressional deputies are profoundly influenced by the President-controlled pork-barrel. Specifically, deputies receiving pork benefits are inclined to vote to weaken the legislature and strengthen the executives. In European context, Golden and Picci (2008)48 used public works expenditures between1953 and 1994 in Italy, and showed that when districts elect more powerful individuals from the governing parties, they tend to successfully secure more infrastructure investments. Also, Limosani and Navarra (1999)49 presented that in pre-election periods in Italy, local administrators favor redistribution policies through an increase in investment spending beyond their standard growth rate that raises the re-election chances of their party leaders. As for the case in Germany, Stratmann and Baur (2002)50 shed light on the question as to whether the legislators’ behaviors differ across different electoral rules, namely the “first-past-the-post” and the proportional representation. It is pointed out that the former is more opted to be the members of committees that allow them to serve their geographically based constituency.
Denemark (2000)51 confirms distinctly that the distribution of Australian pork-barrel is clearly a partisan phenomenon wherein the distributions are made disproportionately to districts held by their most vulnerable colleagues in marginal seats. Unfortunately, relevant literatures in Philippine context are not abundant. Kawanaka (2007)52 exercised an empirical analysis on “pork-barrel” and claimed that the results do not support the hypothesis that the legislators’ proximity to the Presidents is a determining factor in distributions, although some attributes of legislators (such as memberships of certain committees) are statistically significant. However, the implication of such results is not quite rich since various different forms of pork-barrel are
not necessarily sufficiently captured in the analysis.
Barry Ames, “Electoral Rules, Constituency Pressures, and Pork Barrel: Bases of Voting in the Brazilian Congress”, The Journal of Politics, 1995 48 Miriam A. Golden and Lucio Picci, “Pork-Barrel Politics in Postwar Italy, 1953-94”, American Journal of Political Science, 2008 49 Michele Limosani and Pietro Navarra, “Local Pork-Barrel Politics in National Pre-Election Dates: The Case of Italy”, Public Choice, 2001 50 Thomas Stratmann and Martin Baur, “Plurality Rule, Proportional Representation, and the German Bundestag: How Incentives to Pork-Barrel Differ Across Electoral Systems”, American Journal of Political Science, 2002 51 David Denemark, “Partisan Pork Barrel in Parliamentary Systems: Australian Constituency-Level Grants”, The Journal of Politics, 2000 52 Takeshi Kawanaka, “Who Eats the Most?: Quantitative Analysis of Pork Barrel Distributions in the Philippines”, Institute of Developing Economies, JETRO, 2007
Previous studies largely support the view that the districts which are politically tied to the incumbent president are more prioritized in terms of the budget allocation they receive than the regions which are not. It is however not entirely clear whether that is truly the case for the Philippines. The story would depend on how one conceptualizes the marginal cost of lobbying – some may argue that if the local politicians are affiliated to the same political fragment as the president, then he should have a better access to the president, as a consequence of which the marginal cost of lobbying is lowered. The others, however, could claim that the marginal cost is in fact lower for those who are not politically connected to the president since all he has to do to entice the president is to agree on the bills which the president proposes, or whatever form of tangible support in political arena where they technically do not have anobligation to do so.
Based on such view, the politicians who are political allies of the president, on the other hand, have a difficulty in making a visible and meaningful contribution to the president since supporting the presidential bills is nothing special in the first place. Perhaps both theories could hold and they may partially off-set each other, but one factor may have a dominant effect over the other. The paper examines the budget allocation of DPWH, the biggest budget spending branch among all the executive line agencies (Table 1). With the exception of Metro Manila, the Philippines comprises of 80 provinces, each of which is headed by a Governor and a Vice-Governor. One province is usually divided into several legislative districts from which one Congressman is elected with a three-year term of office. There are, however, provinces which form one legislative district by itself, and also, there are single cities, group of municipalities etc. that collectively form one legislative district. While the provincial breakdown of annual budget allocation could be easily obtained across the years, further detailed breakdown on legislative districts was available only for FY2009 and FY2010. Furthermore, FY2009 data does not include the figures on congressional insertions and lump-sum allocations, and therefore is not suitable to study those items. The author excluded Metro Manila (NCR: National Capital Region) from observation since the area has a peculiar division of legislative districts as well as exceptionally high concentration of budget allocation. As a consequence, the paper focuses only on 190 legislative districts. The fundamental interest is whether, and if ever, to which direction the budget allocation is dependent on the political affiliations of the legislative district. Political affiliation is judged on whether or not the Congressman, Governor and Vice-Governor attached to the district belong to the same political segment as the President.
Pork-barrel politics takes place on different stages of budget cycle, and the study should be carried out for each of those stages. First, the author tried to capture whether there is any arbitrary maneuvering in terms of the disbursement of regular budgetary items. It is usually pointed out that the
President uses his authority for budget execution as one way to exercise his de-facto political influence. Since the budget breakdown for FY2010 was available only as of June, the author could only know to what extent the allocation has been released during the first half of the year. Given that limitation, observation on the FY2010 budget data is two-folded. When it comes to the basic budgetary items, whether a certain district is politically characterized as for or against the President seems to be irrelevant to the level of disbursement. Although historically there may have been some cases where the President actually utilized his power for budget execution as a carrot to the local politicians, the data could not factually show that this is in fact the case for the ongoing fiscal year as regards the initial allocation53.
Politically Allied with President Congressmen Unreleased Budget Exist Not Exist Total Yes 61 62 123 No 30 30 60 Governor Yes 64 64 128 No 27 28 55 Total 91 92 183
On the other hand, it is a strange finding that some of the congressional insertions have been released, in spite of the fact that all insertions were basically vetoed by the President. There are five of such districts, with four of the Congressmen, three of the Governors and Vice-Governors affiliated to the President. It is nonetheless difficult to derive any persuading conclusion from this observation, since the size of the sample is quite small. While this requires a more in-depth study in terms of the character of the expenditures, one potential hypothesis is the following; the legislators are engaged in congressional insertions, knowing that these insertions will be in principle vetoed by the President and thus most likely will not be implemented. However, they still find it meaningful to put these additional budgetary items on the “wish-list”, so that one day, they may bear fruits as a consequence of political deals with the President.
In FY2010, seven districts did not receive infrastructure budget allocation in the first place, and therefore, are excluded from the above table. For further preciseness, taking the same explanatory variables as the empirical model in the following section, I ran the logit model on binary estimate for
the existence of unreleased budget. The model returns insignificant results for the political inclinations of Congressmen, Governors and Vice-Governors.
Secondly, the paper examines whether the disbursement of the lump-sum allocation is politically influenced. As mentioned earlier, the lump-sum allocation on DPWH infrastructure spending for FY2010 was 120 million pesos for the Senators and 40 million pesos for Congressmen. The author solely focuses on the allocations of Congressmen since it is their allocations which are allocated on district-wise projects, while these allocations for the Senators are basically cross-district. Examining the FY2010 budget in detail, it becomes clear that the 40 million allocation for Congressmen in fact comprises of 20 million initial allocation which has been already a part of executive budget proposal, and another 20 million of congressional insertion. Again, a correlation was not found between the release of funds and the political affiliation. As of June, 35 districts still have their original allocations unreleased, and these districts are completely evenly split by 18 to 17 for those with the Congressmen affiliated to the same political fragments as the President and those who are affiliated to the opposition parties. This makes it hard to believe that there is an arbitrary control by the President to determine the release of basic lump-sum allocations. Meanwhile, the other 20 million which is added on as a result of congressional insertions shows exactly the other way around – absolutely none of the districts has got the fund released as of June. Although there is still a possibility that the fund releases may come during the latter half of the year, at least at that point in time, there is no positive reason to believe that the inserted portions are released to favor a certain political inclination of the district. From the series of observations described above, it seems that the disbursement phase of the budget is not influenced by a political affiliation of the district so long as the FY2010 is concerned. However, there is still a chance that there is a political influence at the initial stage of the budget distribution on basic budgetary
items. This is an issue which requires an extensive econometrical work. 3. a. Econometrical Work Intention
Skewed allocations of budget embedded at the initial stage of budget formulation as a result of lobbying by local politicians cannot be evidently observed on the surface of the budgetary documents. In this respect, the author hypothesized that the budget allocations that each legislative district receives in infrastructure spending is correlated with the political inclination of the key politicians of the district, namely, Congressman, Governor and Vice-Governor. Although the absence of such correlation does not necessarily guarantee the non-existence ofpork-barrel under this definition, if such statistical linkage is proven, on the other hand, one can rationally assume that there must have been an opportunity of pork-barrel politics which kicked in before the National Expenditure Program was submitted to the Congress. Of course, we would be very much interested in knowing the direction of the correlation in case it in fact does exist. Based on the previous literatures as well as people’s general perception, when the local political figures are allied with the President, there is presumably a positive effect on the level of grant that the district would receive. However, such correlation could be in fact the other way around, as presented by Jones, Sanguinetti and Tommasi (2000) 54 in their study on Argentine case – although such finding belongs to the absolute minority, as shown in the following section, the author’s work presents that this is in fact the case in the Philippines. b. Methodology
Since we are working on two fiscal years, FY2009 and FY2010, a panel data analysis would have been ideal in this context. However, under the constraint on data availability wherein all data points for the explanatory variables are 2008 or before which naturally means the variables are
homogeneous across the two sample years, panel analysis is not a dependable method. Instead, the author adopted a simple OLS with an incorporation of year dummy variable to distinguish one year from the other. Conceptually, the regression equation is expressed as; Y P X 1 1 X 2 2 X 3 3 X 4 4
where the dependent variable Y is the log of the infrastructure budget allocations of each district divided by its land-size. The author initially ran a regression taking the per-capita allocation as the dependent variable, but it did not return a statistically meaningful result55. Taking the log for the dependent variable was indispensable for the meaningful output, which implies the diminishing effect of the explanatory factors. In terms of the budget allocation per se, one has to pay attention to the fact that substantial amount of special grant allocation is made by the President during the formulation of budget proposal for large-scale infrastructure projects which are supposed to be, at least ideally, in line with the growth strategy of the Mark P. Jones, Pablo Sanguinetti and Mariano Tommasi, “Politics, Institutions and Fiscal Performance in a Federal System: An Analysis of the Argentine Provinces”, Journal of Development Economics, 2000 55 As mentioned later on, as an independent variable, the population in fact turned out to be negatively correlated to the adopted dependent variable, which could explain the fact that the population could not become a contributing factor when integrated into the dependent variable. 54 country.
These projects are usually announced during the Presidents’ speech in the Congress at the outset of the year, which is called “State of the Nation Address”, or more commonly know by its acronym “SONA”. Needless to say, the author is interested in knowing the nature of these pet projects of the President. The chances are two-folded; the skewed budget allocation, if ever, may be made either visibly by the form of SONA projects, or with more subtlety – it takes place outside of publically advertised SONA projects. For this purpose, the paper examined both the budget allocation with and without SONA projects. On the other hand, P represents a vector of dummyvariables pertaining to the political affiliation of the local politicians – Congressmen, Governors as well as the Vice-Governors, to show whether they are affiliated to the same or allied political parties of the President (=1 if they are, =0 if not). In order to appropriately control the other relevant factors, vectors of explanatory variables X 1 to X 4 are added to the model (see Table 2 for details): (i) basic determinants – year dummy to distinguish the two fiscal years (=1 for 2010, =0 for 2009) and the log of population, (ii) proxies for the degree of development of the district – log of per capita income, proportion of urban citizens, rice yield as well as corn yield per hectare, proportion of irrigated service area to the total irrigable area, number of installed telephone lines per capita, share of households with electricity, the number of establishments divided by the land-size, (iii) proxies for local development of transportation infrastructure – road density, average number of ports and airports as well as dummy variable for the existent of railway transportation (=1 if exists, =0 if not), and finally, (iv) irregular events – number of typhoons and armed conflicts in the region56. The error term is represented by . c. Result
The result of the work is shown in Table 3. As the finding of the utmost importance, it is observed that, for both the Congressmen and the Governors, their political alliance with the President is negatively correlated with the infrastructure budget allocation which their districts would receive. This result is quite robust for all patterns the author attempted although to a lesser degree in the model with SONA projects. While there are conflicting predictions regarding whether such alliance would work favorably or unfavorably in acquiring funding from the central government, this result underwrites the latter – ironically, if one wishes to receive a preferential treatment in a budget process, he would be better-off by politically distancing from the President until the President gives in. It has to be reiterated that even more
The variables which could be obtained only on provincial basis are divided by the number of legislative districts in the province.
significant implication is drawn from the very fact that such correlation exits. Setting aside the direction of the correlation (positive or negative), the existence of the correlation itself is an indirect evidence that the pork-barrel politics takes place even at the initial stage of budget formulation presumably as a consequence of a series of political bargaining. What adds complication to this seemingly straightforward outcome is the opposite sign the Vice-Governor dummy shows. It seems odd that only the Vice-Governor shows a positive correlation when it comes to their political attachment to the President. While it is true that their political influence is considered to be much smaller than that of Governors or Congressmen, a further study is called for in order to explain this odd result in a convincing manner. In any case, inclusion or exclusion of SONA projects does not bring about substantial changes in the result – in fact, the budget allocation without SONA projects generally presents more statistical significance as well as better fit of the model. This implies that the political factors, if ever, were taken into consideration prior to the inclusion of SONA projects and the SONA projects themselves are not so much politicized in the context of this study. Aside from the major variables related to political affiliation, it is somewhat a counter-intuitive finding that the population is negatively correlated with the budget allocation. With regard to this point, one has to recall, first of all, that the NCR is excluded in the model, and thus the areas for truly high concentration of population are not in the picture. That said, perhaps a fair assumption would be that more budget per land is being allocated for less populous areas since constituent can be more politically vocal because of the heavier weighs on their ballots. Some variables related to regional development are positively correlated to the share of budget allocation, which suggests that budget allocation in the Philippines is “gap widening” rather than “gap adjusting”. On the other hand, the existing level of development of the infrastructure per se seems to have weak correlations with the dependent variable. Pertaining to the irregular events, the conflicts have an unambiguously clear, negative correlation to the budget allocation. Mindanao, the nation’s most troublesome conflict affected area,
is said to be permanently under-budgeted. The above finding underwrites such view.
The paper has shown the concepts and the development of budget politicization in the Philippines and the problem which such pork-barrel politics entails. The Philippine budget system has a long history of pork-barrel, and it is deeply rooted in its institution. Sustaining such system, however, comes with a substantial cost in various aspects. In the empirical part, it is evidently shown that pork-barrel politics takes place even at the very initial stage of the budget cycle. In other words, pork-barrel distribution is nothing exogenous but is already embedded when the initial budget proposal takes its form. Such skewed distribution of fund favors, contrary to the anticipation of many, the districts led by Congressmen or Governors who are not politically allied with the President. This probably stems from the fact that, for the Presidents, it makes more sense to favor those who may otherwise vote against the bills he will propose. From the perspective of the Congressmen or the Governors, the marginal cost they have to pay to get the funding is lower when they are not allied with the President. Several points should be pointed out as the caveats especially related to the empirical part of this paper – the study is not free from criticism for its fragility due to the limited data availability. First and foremost, the study could only look into the last two years of the administration under President Arroyo. One may be able to draw a completely different picture when looking at the previous fiscal years under other Presidents. In fact, Arroyo is commonly characterized as a “weak” leader in regard to the political support from the legislative branch, and this may have caused some differences in the way she is involved in pork-barrel politics. Secondly, another drawback related to the data availability is the lack of precision in explanatory variables. For one, there is a substantial overlap in
explanatory variables across the examined two years, and for two, many data sets could be obtained only on provincial basis, not on each district. These flaws may naturally have caused an inaccuracy in the model. Thirdly, it is a frequent phenomenon in the Philippine politics that the Congressmen easily hop around different political parties – in other words, political parties are quite weakly established. It is, therefore, a valid suspect that political affiliation itself may not be an essential element to reflect the political relationship between the local politicians and the President in the first place. Fourthly, one cannot entirely eliminate the possibility that the local politicians may be getting something other than the budget allocation in return for whatever action they may take to satisfy the President, in which case the model also may not have come up with an accurate estimation. Last but not least, the tangible mechanism by which the Congressmen and the Governors influence the initial budget allocation has not been entirely made clear. Especially,
unlike the Congressmen who have votes in the legislative body, what Governors can offer to the President seems to be relatively limited. The theoretical thinking on pork-barrel politics has primarily assumed the members of legislative body as the President’s counterpart on political negotiations, and not much elaboration has been made on such other actors as Governors, Mayors etc. As much as there is still a room for elaboration in the Philippine context, the extent to which the findings of this paper can be generalized in other nations can also be an interesting topic to pursue. Would there be differences in presence/absence, degrees, and the forms of pork-barrel politics depending on the governance institution, level of social development, cultural background and so forth? These are the issues which future studies may explore.
Alvarez, R. Michael, and Saving, Jason L., “Deficits, Democrats, and Distributive
Benefits: Congressional Elections and the Pork Barrel in the 1980s”, Political Research Quarterly, 1997 Ames, Barry, “Electoral Rules, Constituency Pressures, and Pork Barrel: Bases of Voting
in the Brazilian Congress”, The Journal of Politics, 1995
Brickers, Kenneth N., and Stein, Robert M., “Congressional Elections and the Pork
Barrel”, The Journal of Politics, 1994
Brickers, Kenneth N., and Stein, Robert M., “The Electoral Dynamics of the Federal Pork
Barrel”, American Journal of Political Science, 1996
Brickers, Kenneth N., and Stein, Robert M., “The Congressional Pork Barrel in a
Republican Era”, The Journal of Politics, 2000
Costa-i-Font, Joan, Rodriguez-Oreggia, Eduardo and Lunapla, Dario, “Political
Competition and Pork-Barrel Politics in the Allocation of Public Investment in Mexico”, Public Choice, 2003 Del Rossi, Alison F., “The Politics and Economics of Pork Barrel Spending: The Case of
Federal Financing of Water Resources Development”, Public Choice, 1995 Denemark, David, “Partisan Pork Barrel in Parliamentary Systems: Australian
Constituency-Level Grants”, The Journal of Politics, 2000
Diokno, Benjamin E., “Where Deep Secrets Lie”, Business World, February 11, 2010 Diokno, Benjamin E., “Busted Budget”, Business World, June 24, 2010 Esguerra, Christian V., Tubeza, Philip, “Aquino Won’t Control Pork”, Daily
Inquirer, May 21, 2010 Evans, Diana, “Policy and Pork: The Use of Pork Barrel Projects to Build Policy
Conditions in the House of Representatives”, American Journal of Political Science, 1994 Fernandez, Felisa D., “The Budget Process and Economic Development – The Philippine
Experience”, College of Public Administration, University of the Philippines, Manila. 1973 Golden, Miriam A. and Picci, Lucio, “Pork-Barrel Politics in Postwar Italy, 1953-94”, American Journal of Political Science, 2008 Grossman, Philip J., “A Political Theory of Intergovernmental Grants”, Public Choice, 1994 Gutierrez, Eric, “The Public Purse” in Sheila S. Coronel (ed.) Pork & Perks: Corruption
& Governance in the Philippines, Philippine Center for Investigative Journalism, 1998
– 30 –
Inman, Robert P., Rubinfeld, Daniel L., “Designing Tax Policy in Federalist Economies:
An Overview”, Journal of Public Economics, 1996
Jones, Mark P., Sanguinetti, Pablo and Tommasi, Mariano, “Politics, Institutions and
Fiscal Performance in a Federal System: An Analysis of the Argentine Provinces”, Journal of Development Economics, 2000 Katayama, Yutaka, “Philippine Yosan Seido (Philippine Budget System)”, Unpublished, 2009 Kawanaka, Takeshi, “Who Eats the Most?: Quantitative Analysis of Pork Barrel
Distributions in the Philippines”, Institute of Developing Economies, JETRO,
2007 Lancaster, Thomas D., “Electoral Structures and Pork Barrel Politics”, International Political Science Review, 1986 Limosani, Michele and Navarra, Pietro, “Local Pork-Barrel Politics in National
Pre-Election Dates: The Case of Italy”, Public Choice, 2001 Nishimura, Kenichi, “Estrada Seikenka no Pork-Barrel Seido (Pork-Barrel under Estrada
Administration)”, University of Osaka, 2002
OECD, “OECD Budget Practices and Procedures Survey”, 2008 Parreno, Earl, “Pork” in Sheila S. Coronel (ed.) Pork & Perks: Corruption & Governance
in the Philippines, Philippine Center for Investigative Journalism, 1998 Persson, Torsten and Tabellini, Guido, “Does Centralization Increase the Size of
Government?”, European Economic Review, 1994
Prospero C. Nograles and Edcel C. Lagman “Understanding the ‘Pork-Barrel’”, House of Representatives, 2008 Sato, Motohiro, “Seifukan Zaisei Kankei no Seiji Keizai Gaku (Political Economy of
Intergovernmental Fiscal Transfers)”, Financial Review, Policy Research Institute, Ministry of Finance of Japan, 2006 Thompson, Joel A., “Bringing Home the Bacon: The Politics of Pork Barrel in the North
California Legislature”, Legislative Studies Quarterly, 1986 Stratmann, Thomas and Baur, Martin, “Plurality Rule, Proportional Representation, and
the German Bundestag: How Incentives to Pork-Barrel Differ Across Electoral Systems”, American Journal of Political Science, 2002 Wehner, Joachim, “Institutional Constraints on Profligate Politicians: The Conditional
Effect of Partisan Fragmentation on Budget Deficit”, London School of Economics, 2008 Wilson, Rick K., “An Empirical Test of Preferences for the Political Pork Barrel: District
Level Appropriations for River and Harbor Legislation, 1889-1913”, American Journal of Political Science, 1986
– 31 –
Table1: Comparison of FY2010 Budget at Each Stage (in Million Pesos) Proposed Budget Congress Budget (NEP) (GAB) 622,974.7 649,157.8 7,018.4 7,972.8 4,259.4 4,259.4 185.0 185.0 19,427.8 19,487.8 35,899.1 37,261.1 933.0 933.0 159,283.1 160,571.0 19,592.0 20,353.9 665.7 665.7 10,660.2 11,097.2 9,378.3 9,611.5 12,499.6 12,656.6 27,873.9 28,120.1 64,547.4 65,197.4 6,060.4 6,293.2 6,351.6 6,109.2 57,528.1 57,540.3 96,607.1 113,030.5 4,768.5 4,788.5 14,398.5 14,850.5 1,487.1 1,497.6 2,232.2 2,409.7 14,492.5 15,797.0 4,030.9 4,050.9 1,104.5 1,119.5 4,148.6 4,965.8 1.3 1.3 12,308.9 12,712.7 610.5 610.5 3,766.1 3,820.3 10,432.7 10,423.7 967.5 1,064.3 240.9 255.9 9,263.7 9,435.1 222,105.6 20,918.1 24,811.9 5,000.0 2,000.0 800.0 2,000.0 1,000.0 1,000.0 1,595.2 76,496.2 50.0 79,494.2 6,940.0 695,920.0 340,812.0 1,541,000.3 225,775.5 22,206.8 31,221.9 5,000.0 2,000.0 800.0 2,000.0 600.0 500.0 1,595.2 73,167.4 50.0 79,494.2 7,140.0 666,067.0 310,959.0 1,541,000.3 BICAM (GAA) 674,790.4 8,830.0 4,259.4 185.0 20,767.8 39,241.6 933.0 161,405.9 22,402.3 665.7 12,190.6 9,471.5 12,746.1 28,686.1 65,602.0 6,426.3 6,421.4 57,670.3 126,931.0 5,233.9 15,314.4 1,584.6 2,443.7 16,577.0 4,014.8 1,119.5 5,498.8 1.3 12,662.7 610.5 3,825.3 10,432.7 1,064.3 285.9 9,285.1 234,490.0 24,283.4 31,686.6 5,000.0 2,000.0 800.0 2,000.0 1,000.0 1,000.0 1,677.2 74,167.4 50.0 79,964.2 10,861.2 631,319.0 276,212.0 1,540,599.4
Departments Congress of the President Office of the President Office of the Vice-President Department of Agrarian Reform Department of Agriculture Department of Budget and Management Department of Education State Universities and Colleges Department of Energy Department of Environment and Natural Resources Department of Finance Department of Foreign Affairs Department of Health Department of Interior and Local Government Department of Justice Department of Labor and Employment Department of National Defense Department of Public Works and Highways Department of Science and Technology Department of Social Welfare and Development Department of Tourism
Department of Trade and Industry Department of Transportation and Communication National Economic and Development Authority Office of the Press Secretary Other Executive Offices Joint Legislative-Executive Councils The Judiciary Civil Service Commission Commission on Audit Commission on Elections Office of the Ombudsman Commission of Human Rights Autonomous Regions Special Purpose Funds Budgetary Support to Government Corporations Allocations to Local Government Units AFP Modernization Program Calamity Fund Contingent Fund DepEd School Building Program E-Government Fund General Fund Adjustments International Commitment Fund Miscellaneous Personnel Benefits Fund National Unification Fund Pension and Gratuity Fund Priority Development Assistance Fund Automatic Appropriations Interest Payments TOTAL NEW APPROPRIATIONS
Source: House of Representatives of the Philippines (Modified by the author.)
– 32 –
Table2: Summary of Variables Variable Year Dummy Population Congressman Description Source Dummy variable of fiscal year of the budget eaual to 1=if 2010, 0=if 2009. Total Number of Population Census of Population and Housing (CPH) Dummy variable of political affiliation equal to 1=if Congressman of the district is in the same political party as Commission on Elections (COMELEC) the president; 0=otherwise Dummy variable of political affiliation equal to 1=if Governor of the district is in the same political party as the Commission on Elections (COMELEC) president; 0=otherwise Dummy variable of political affiliation equal to 1=if ViceGovernor of the district is in the same political party as the Commission on Elections (COMELEC) president; 0=otherwise Per-Capita Income Family Income and Expenditure Survey (FIES) Share of Urban Population Family Income and Expenditure Survey (FIES) Palay yield per hectare (in MT) Bureau of Agricultural Statistics (BAS) Corn yield per hectare (in MT) Proportion of irrigated area to total irrigable area National Irrigation Authority (NIA) Total number of installed telephone lines National Telecommunication and Transportation (NTC) Proportion of households with access to electricity Family Income and Expenditure Survey (FIES Total Number of Establishments National Statistics Office (NSO) Total
Road Density Department of Public Works and Highways (DPWH) Number of Railways Philippine National Railways (PNR) Number of Airports Avation Authority of the Philippines (AAP) Number of Seaports Philippine Ports Authority (PPA) Disater Response Operations Monitoring and Information Center Number of Typhoons (DROMIC) Number of Armed conflict
Vice-Governor Per-Capita Income Urban Population Palay Yields Corn Yields Irrigation Telephone Lines Electricity Establishments Road Density Railways Airports Seaports Typhoons Conflicts
– 33 –
Table3: Sumamry of OLS Regression Allocation w/o SONA R2 R3 -0.408*** (0.098) -0.629*** (0.147) -0.236** (0.114) -0.321** (0.128) 0.352*** (0.128) 1.349*** (0.402) 0.182 (0.451) 0.297*** (0.085) -0.045 (0.060) 0.563** (0.243) 13.024*** (2.559) -0.676 (0.633) -0.010 (0.021) 0.829 (0.531) 0.876 (0.628) 0.094 (0.139) -0.042 (0.045) -0.408*** (0.098) -0.551*** (0.136) -0.253** (0.113) -0.399*** (0.122) 0.374*** (0.125) 1.379*** (0.389) 0.292 (0.424) 0.277*** (0.084) -0.019 (0.057) 0.645*** (0.237) 13.244*** (2.508) -0.509 (0.592) -0.004 (0.018) Allocation w/ SONA R2 R3 -0.489*** (0.104) -0.686*** (0.157) -0.217* (0.121) -0.348** (0.136) 0.214 (0.136) 0.928** (0.427) 1.050** (0.480) 0.060 (0.090) -0.002 (0.064) 0.841*** (0.258) 9.423*** (2.722) -0.066 (0.674) -0.019 (0.023) 0.167 (0.565) 0.115 (0.668) -0.132 (0.148) 0.053 (0.048) -0.489*** (0.103) -0.745*** (0.143) -0.214* (0.120) -0.369*** (0.130) 0.251* (0.130) 0.882** (0.411) 0.892** (0.449) 0.066 (0.089) -0.017 (0.060) 0.876*** (0.250) 9.666*** (2.652) 0.176 (0.626) -0.007 (0.019)
R1 Basic Determinants Year Dummy Population Politican Affiliation Congressman Governor Vice-Governor Regional Development Per-Capita Income Urban Population Palay Yields Corn Yields Irrigation Telephone Lines Electricity Establishments Infrastructure Road Density Railways Airports Seaports Irregular Events Typhoons
R1 -0.489*** (0.096) -0.502*** (0.147) -0.215* (0.112) -0.373*** (0.126) 0.177 (0.128) 0.744* (0.397) 1.026** (0.450) 0.053 (0.085) -0.031 (0.059) 0.721*** (0.240) 6.507** (2.552) -0.470 (0.626) 0.002 (0.021) -0.161 (0.525) 0.034 (0.620) -0.212 (0.141) 0.094** (0.046)
-0.408*** (0.092) -0.472*** (0.141) -0.235** (0.107) -0.342*** (0.120) 0.318** (0.123) 1.193*** (0.379) 0.169 (0.429) 0.290*** (0.081) -0.070 (0.056) 0.459** (0.229) 10.541*** (2.437) -1.020* (0.598) 0.007 (0.020) 0.552 (0.502) 0.804 (0.592) 0.031 (0.135) -0.006 (0.044)
0.097 0.129 (0.130) (0.136) Conflicts -0.258*** -0.303*** (0.041) (0.043) Adjusted R-Squared 0.516 0.453 0.451 0.454 *** Significant at 1% level, ** Significant at 5% level, * Significant at 10% level (standard deviation)