Public Finance and the Challenges of Autonomy: The Case of the Cordillera Administrative Region

SANTOS JOSE O. DACANAY III

University of the Philippines Baguio

Abstract

This paper examines the relationship between fiscal autonomy (the assignment and responsibility of revenues and expenditure functions across government levels) and public finance (or the sources and uses of public funds) in the Cordillera region in northern Philippines. The study reviews briefly the Cordillera autonomy literature and organic laws to contextualize fiscal autonomy. Financial analysis and regression are employed to explore the determinants and drivers of local public finance using the local government units’ (LGUs) income statement accounts for the period 2001 to 2006. Results show that poverty incidence as a proxy of average income and tax base in the LGUs is negatively related to per capita real property tax revenues, pointing to the inadequacy and unpreparedness of the LGUs to raise their own funds. The size of LGU expenditures, however, exerts pressure to improve tax revenue collections, leading to increased government accountability at the lower levels. National government transfers or internal revenue allotments (IRA) provide a disincentive for LGUs to increase their local tax and non-tax efforts, perpetuating the LGUs’ so-called IRA dependency. The study concludes that there is the need to revisit revenue and expenditure assignment, responsibility and functions across government levels for efficient public goods production and provision.

Keywords: Development finance, fiscal autonomy, regional development, public goods, local government, government transfers.  

1. Introduction

There has been a large empirical literature on fiscal decentralization initiated by Oates (1972, 209-213), though an empirical consensus has not been reached. While the experience of industrialized countries were studied and documented (Feld et al. 2003; Fiva 2005), scant attention has been given to fiscal decentralization and regional autonomy movements in developing nations and transition economies (Meloche et al. 2004). This paper hopes to shed light on the Philippine experience by investigating whether the establishment of an intermediate layer between the national and local government units is feasible from a fiscal policy standpoint, using data from the Cordillera region. In addition, since the 1987 Philippine Constitution provides for the establishment of autonomous regions in the Cordilleras and Muslim Mindanao, this echoes the decentralization theorem propounded by Oates (1972, 209) which maintains that “each public service should be provided by the jurisdiction having control over the minimum geographical area that would internalize benefits and costs of such provision.” The constitutional premise for the creation of a subnational autonomous government includes the distinctive historical and cultural heritage, economic and social structures in the two regions, and the possibility of finding a solution to the regional conflicts that have risen partly from cultural diversity, regional income disparity, and national government exploitation and neglect. The idea behind the constitutional provisions for autonomous regions is to allow the separate development of peoples with distinctive cultures and traditions (Rood 1989, 379-391). So far, only the Autonomous Region for Muslim Mindanao has been created. This reasoning follows the spirit of various studies that posit decentralization as a mechanism to make policy more responsive to local needs and to involve the local populace in processes of democratic governance (Feld et al. 2003).

As for the Cordilleras, the region is poised once again to draft, enact and ratify an organic act for the third time creating a regional government after two failed plebiscites in 1990 and 1998 (RDC-CAR 2007). Within this context, the paper explores whether the establishment of an intermediate layer between the national and local government units is feasible from a fiscal policy standpoint using data from the region’s component LGUs’ income statement accounts for the period 2001 to 2006, employing financial analysis and regression techniques. Results show that poverty incidence as a proxy of average income and tax base in the LGUs is negatively related to per capita real property tax revenues, pointing to the inadequacy of the LGUs to raise their own source of funds. The size of LGU expenditures, however, exerts pressure to increase tax revenue collections, leading to increased government accountability at the lower levels. National government transfers or IRA provide a disincentive for LGUs to increase their local tax and non-tax efforts, perpetuating the LGUs’ so-called IRA dependency.

The remainder of the paper is organized as follows. Section 2 relates the notions of public finance to fiscal autonomy in the context of regional autonomy and the creation of the Cordillera region. Section 3 presents the data, provides the empirical specification, and discusses the method. Section 4 explains the obtained results. Section 5 provides some concluding remarks.

2. Fiscal Autonomy and Public Finance in the Cordillera

2.1 Creation of the Cordillera Region

On July 15, 1987, then Philippine President Corazon C. Aquino issued Executive Order (EO) No. 220 creating the Cordillera Administrative Region, carved out of the provinces of the Ilocos and Cagayan Valley regions. The executive fiat actually envisions the consolidation and coordination of the delivery of basic services of line departments and agencies of the national government in the areas covered by the administrative region as a step preparatory to the grant of autonomy to the Cordilleras. The Cordillera Administrative Region is at present composed of six provinces (Abra, Apayao, Benguet, Ifugao, Kalinga and Mountain Province), with 76 municipalities and the City of Baguio (see Fig. 1). Subsequent to the issuance of EO No. 220, the Congress enacted Republic Act (RA) No. 6658 which created the Cordillera Regional Consultative Commission, tasked to prepare a draft organic act which subsequently became the first organic act, RA No. 6766. Passed on October 23, 1989, this organic act provided for the creation of an autonomous government headed by a regional governor, with a regional assembly to enact laws of regional application, and a regional judiciary composed of a supreme court and lower courts. The act also provided for the proposed autonomous government in the Cordillera to exercise governmental functions, including the raising of taxes, but excluding defense, foreign affairs, and monetary functions. However, as required by the Constitution, the act had to be ratified by the people, and a plebiscite was held for this purpose on January 30, 1990. In this plebiscite, it was rejected by all except the electorate of the province of Ifugao (Brillantes and Garming 1998, 737). The second organic act, RA No. 8438, was approved on December 22, 1997. In the plebiscite held on March 9, 1998, it was endorsed only by the electorate of the province of Apayao. Since the Supreme Court had earlier ruled that the Cordillera Autonomous Region could not consist of a single province (Buendia 1998, 706), the Cordillera remains a regular administrative region. Among the reasons cited for the rejection of the organic acts were: the threat to some sectors with the devolution of national functions to the autonomous region; the lack of faith and confidence of voters in the capability of future leaders to take over fiscal functions particularly on infrastructure projects; the sentiment from Baguio City and Benguet voters that they would end up subsidizing poorer LGUs; and the misinformation regarding possible increase in taxes (RDC-CAR 2007). Casambre (2010, 95), employing hermeneutics as framework and method of interpretation, suggested that disjuncture the failure to meet point-to-point characterized the autonomy debate which led to the second organic act’s overwhelming rejection.

2.2 Regional Autonomy as Political Autonomy

The creation of autonomous regions, a unique feature of the 1987 Philippine Constitution, contemplates the grant of political autonomy and not just administrative autonomy to such regions. The Constitution provides for an autonomous regional government with a basic structure consisting of an executive department and a legislative assembly and special courts with personal, family and property law jurisdiction in each of the autonomous regions (1987 Philippine Constitution, Art. X, Sec. 18). As envisioned in the two Cordillera organic acts:

Autonomy ensures for the people of the Cordilleras the right to secure for themselves their ancestral domain, develop their economy, promote their cultural heritage, and establish a system of self-governance within the framework of the Philippine Constitution and national sovereignty, as well as the territorial integrity of the Philippines. (Section 2, Article II of both RA Nos. 6766 and 8438)

However, the creation of autonomous regions does not signify the establishment of sovereignty distinct from that of the Republic, as it can be installed only within the framework of the Constitution and the national sovereignty as well as territorial integrity of the Republic of the Philippines. Regional autonomy is the degree of self-determination exercised by the local government unit vis-à-vis the central government. It refers to the granting of basic internal government powers to the people of a particular area or region with least control and supervision from the central government (Brillantes and Cuaresma 1990, 29). It also implies the cultivation of more positive means for national integration. It would remove the wariness among the Muslims (followers of Islam in southern Philippines) and the Cordillerans, increase their trust in the government, and pave the way for the unhampered implementation of the development programs in their regions (Tanggol 1998, 651).

A necessary prerequisite of autonomy is decentralization (Tapales 1998, 13). Decentralization is a decision by the central government authorizing its subordinates, whether geographically or functionally defined, to exercise authority in certain areas. It involves decision-making by subnational units. It is typically a delegated power, wherein a larger government chooses to delegate certain authority to more local governments. Federalism implies some measure of decentralization, but unitary systems may also decentralize (Stiglitz 2000, 728). Decentralization differs intrinsically from federalism in that the sub-units that have been authorized to act (by delegation) do not possess any claim of right against the central government.

Decentralization comes in two forms deconcentration and devolution. Deconcentration is administrative in nature. It involves the transfer of functions or the delegation of authority and responsibility from the national office to the regional and local offices. This mode of decentralization is also referred to as administrative decentralization (De Guzman and Reforma 1998, 24). Devolution, on the other hand, connotes political decentralization, or the transfer of powers, responsibilities, and resources for the performance of certain functions from the central government to local government units (Tapales 1998, 12-13). This is a more liberal form of decentralization since there is an actual transfer of powers and responsibilities (Brillantes and Cuaresma 1990, 29). It aims to grant greater autonomy to local government units in cognizance of their right to self-government, to make them self-reliant, and to improve their administrative and technical capabilities (De Guzman and Reforma 1998, 60-61).

The Supreme Court elucidated the concept of autonomy in Limbona v. Mangelin (G.R. No. 80391, 28 February 1989, 170 SCRA 786, 794-795). It said that autonomy is either decentralization of administration or decentralization of power. There is decentralization of administration when the central government delegates administrative powers to political subdivisions in order to broaden the base of government power and in the process to make local governments “more responsive and accountable,” and “ensure their fullest development as self-reliant communities and make them more effective partners in the pursuit of national development and social progress.” At the same time, it relieves the central government of the burden of managing local affairs and enables it to concentrate on national concerns. Decentralization of power, on the other hand, involves an abdication of political power in favor of LGUs declared to be autonomous. In that case, the autonomous government is free to chart its own destiny and shape its future with minimum intervention from central authorities. According to the framers of the constitutional provision, decentralization of power amounts to “self-immolation,” since in that event the autonomous government becomes accountable not to the central authorities but to its constituency (G.R. No. 80391, February 28, 1989, 170 SCRA 786, 794-795).

In Cordillera Broad Coalition v. Commission on Audit (G.R. Nos. 79956 and 82217, January 29, 1990, 181 SCRA 495, 506), the Court of Appeals ruled without any dissent that the creation of autonomous regions contemplates the grant of political autonomy an autonomy greater than the administrative autonomy granted to LGUs. It held that “the constitutional guarantee of local autonomy in the Constitution (1987 Philippine Constitution, Art. X, Sec. 2) refers to administrative autonomy of LGUs or, in more technical language, the decentralization of government authority. On the other hand, the creation of autonomous regions in Muslim Mindanao and the Cordilleras, grants political aside from administrative autonomy to the two regions.”

To this end, Section 16, Article X of the 1987 Constitution limits the power of the President over autonomous regions to general supervision. The President exercises “general supervision” over them, but only to “ensure that local affairs are administered according to law.” He has no control over their acts in the sense that he cannot replace their judgments with his own. In essence, the provision also curtails the power of Congress over autonomous regions. Congress will have to re-examine national laws and make sure that they reflect the constitution’s adherence to local autonomy. And in case of conflicts, the underlying spirit which should guide its resolution is the Constitution’s desire for genuine local autonomy. The diminution of the powers of Congress over autonomous regions was confirmed in Ganzon v. Court of Appeals (G.R.  Nos. 93252, 93746, 95245, August 5, 1991, 200 SCRA 271, 281) in which the court held that “the omission signifies nothing more than to underscore local governments’ autonomy from Congress and to break Congress’ ‘control’ over local government affairs.” This applies to areas over which autonomous regions have powers, as specified in Sections 18 and 20, Article X of the 1987 Constitution.

2.3 Fiscal Autonomy and Public Finance

It can thus be argued that autonomy has varied levels. LGUs have administrative autonomy. The 1987 Constitution asserts that the state shall ensure the autonomy of local governments (Section 25, Article II). This constitutional guarantee of local autonomy (Art. X, Sec.2) refers to the administrative autonomy of LGUs or the decentralization of government authority (GR No. L-31004, January 8, 1971, 37 SCRA 1). Regional autonomous governments, on the other hand, have, in addition to administrative autonomy, political autonomy. Political autonomy implies that the regional government is directly accountable to the people and allows the people more influence in the formulation and implementation of policies. Hand in hand with political autonomy is fiscal autonomy. Fiscal autonomy entails the responsibility of revenue raising, allocation, and expenditure (public finance). The constitution vests the judiciary and Congress, and the constitutional commissions with fiscal autonomy which means that their approved appropriations shall be automatically and regularly released. Fiscal autonomy is also discussed in Article XIII, Section 1 of the first organic act, RA No. 6766, to wit:

The Regional Government shall have the power to create its own sources of revenue and to levy taxes, fees and charges, subject to guidelines and limitations as the Constitution and this Act may provide, consistent with the basic policy of local autonomy. Such taxes, fees and charges shall accrue exclusively to the Autonomous Region.

In the second organic act or RA No. 8438, Article VIII, Section 1, likewise discusses sources of revenues and other fiscal matters, to wit:

The regional government shall have the power to create its own sources of revenues and to levy fees and charges except the power to impose taxes, subject to such guidelines and limitations as the Constitution and this Act may provide, consistent with the basic policy of local autonomy. (emphasis mine)

The curtailing of taxing powers of the proposed regional government in the second organic act is glaring. The possibility of imposing new taxes was one of the reasons for the rejection of the first organic act, hence, the prohibition of tax powers was included.

Musgrave and Musgrave’s (1989, 445) classic distinction about the different tasks of government in an economy is used as guide in Tables 1 to 4. Musgrave and Musgrave contend that allocation is better done by local governments while stabilization and redistribution are better done by central governments. Table 1 gives the comparative provisions of the two Cordillera autonomous region organic acts vis-à-vis the Autonomous Region for Muslim Mindanao (ARMM) law as amended. The stark difference between the second and the first Cordillera organic acts and also that of the ARMM law on the sources of revenue is that the second organic act removes the power of the regional government to impose taxes. This is in response to one of the major reasons for the rejection of the first organic act, that is, that new or increased tax rates will be levied to fund the development of the region. Another explicit statement in the second organic act is the sharing of the proceeds from the development and utilization of national wealth, an attempt to respond to the clamor of the people to make them partners in and beneficiaries of development. On the sharing of revenues, a specific provision or pre-determined formula on the sharing of national taxes in the second organic act is conspicuously absent. The continuing allotment to devolved agencies as well as the generous additional allotment of Php 4 billion for the next five years promised in the second organic act is directed to voters to approve the measure.

Table 2 gives the tax assignments of the various LGU levels. In the Local Government Code (Book II, Local Taxation and Fiscal Matters, Section 129), the power to create sources of revenue is detailed such as the power to levy taxes, fees and charges, consistent with the basic policy of local autonomy. Such taxes, fees and charges accrue exclusively to the LGUs, which is not diminished under the second organic act. These taxes will be the primary and internally-generated source of revenue for the LGUs which will ensure and measure their degree of self-sufficiency. On the other hand, the tax powers exclusive to national or central government are: income tax on both individuals and corporations; value added tax; excise taxes on alcoholic beverages, tobacco and petroleum products; custom duties. Part of these national taxes will be plowed back to the regional government in the form of IRA as specified in the local government code, or as an additional share and additional allotment as promised in the two organic acts. Figure 2 illustrates the tax assignment and transfers in CAR for the year 2006. As shown in the figure, the tax base which includes individuals and corporations is levied separately by the national and local governments, with the former taxing income and consumption, and the latter taxing real properties, businesses and other immobile sources. The LGUs, on the other hand, receive IRA, other shares from national taxes, and grants from the national government.

Table 3 gives the shares from national tax collection of LGUs. The table includes various types of transfers with examples. First among these is the IRA which is a mandatory, automatic, regular and unconditional block grant that represents the shares of the LGUs from national taxes. The IRA is the lifeblood of the LGUs. Resource-rich Cordillera also receives its just share of excise taxes from national wealth such as mines, hydro-energy, and quarries. Share on the tobacco excise tax is only applicable to the province of Abra as it is the only tobacco-growing area in the Cordillera. As for economic zones, only Baguio City has these established enclaves so far the Baguio City Ecozone in Loakan, a 119-hectare government-owned and operated economic zone; John Hay, a 288-hectare special tourism economic zone in the former US recreation base facility, under the Bases Conversion and Development Authority, a government-owned and -controlled corporation; and, at the SM Baguio Cyberzone Building, a half-hectare private economic zone catering to business process outsourcing (BPO) companies. The Baguio Ecozone remitted PhP 1.031 B for 2001-2005 to the national government, PhP 412M of which was Baguio City’s share, representing 2% of gross receipts. Other shares include the Local Government Service Equalization Fund (LGSEF) which was earmarked from the IRA while the Local Government Empowerment Fund (LGEF) was used to assist the 21 poorest provinces and 5th and 6th class towns that were identified during the Ramos administration’s Social Reform Agenda. Included in the LGEF is the Cordillera Highland Agricultural Resource Management (CHARM) project of the Department of Agriculture and the Department of Environment and Natural Resources. Finally, there is the Priority Development Assistance Fund (PDAF) or ‘pork barrel’ of congressional representatives in the Cordilleras. Senators can also channel their ‘pork’ to local projects such as Sen. Flavier’s direct assistance in the infrastructure development projects of the Baguio General Hospital and Medical Center, and the Baguio City National High School. The PDAF, however, is an opaque source of funds, which makes it controversial as it has both hard (infrastructure) and soft (services) components that may or may not be fully disclosed depending on the commitment of the legislators to transparency.

Table 4 details the extraordinary receipts, grants, and aids to the LGUs and provides examples. This last category is ‘extraordinary’ in the sense that this item is not regular and continuing, and non-recurring in the LGU’s income statements. Grants, usually in the form of Official Development Assistance (ODA), are unilateral fund transfer from foreign governments and multilateral institutions with a national government counterpart to depressed provinces and cities. The funds are either in the form of a loan/equity/grant mix for the financing of projects. National aids in times of calamities are likewise transferred from national coffers to the affected LGUs, though LGUs earmark 5% of their legislated budgets as calamity fund. LGUs which allow lottery (legally-sanctioned) in their areas also share in the proceeds of the operations from the Philippine Amusement and Gaming Corporation (PAGCOR). Lotto is meant to replace ‘jeuteng,’ an illegal numbers game which only benefits the operators and their protectors.

3. Data and Method

3.1 Data

The study uses statements of income and expenses (SIEs) of the six provinces, 76 municipalities, and one city in CAR for the fiscal years 2001 to 2006. The data, sourced from the Bureau of Local Government Finance (BLGF) under the Department of Finance, are actually quarterly submissions of the municipal, city and provincial treasurers of each LGU to the bureau. The SIEs detail the income of LGUs from internal local sources, broken down into tax and non-tax revenues. Tax revenues include real property taxes, business taxes and others as detailed in Table 2. Non-tax revenues, on the other hand, include regulatory fees, service or user charges, receipts from economic enterprises and others. External income sources are largely from shares of the LGUs from national tax collections which include the IRA, shares from national wealth; tobacco excise tax, shares from Reformed Value Added Tax; economic zones and others as discussed earlier and presented in Table 3. Other external fund sources include extraordinary receipts, grants and aids which include shares of LGUs from lotto sales; loans and borrowings; and, inter-local transfers as illustrated in the previous section and presented in Table 4. A schematic presentation of the components of the income accounts for LGUs as contained in the SIE is given in Figure 3. The figure shows the distribution of income for all LGUs in CAR (collectively for all the provinces, municipalities, and Baguio City) in nominal values and in percent for 2006. It can be gleaned from Figure 3 that local sources, which include both tax and non-tax revenues, only account for 16.2% of the total income of the LGUs in CAR. The balance of 83.8% comes from external sources 81% from IRA and other shares in national taxes and 2.8% from extraordinary receipts, loans and interlocal transfers.

The expenditure items are categorized into major sectoral activities such as general public services; education; health, nutrition and population control; labor and employment; housing and community development; social security, social services and welfare; economic services; debt servicing; and a catch-all item, others. Figure 4 illustrates the distribution of the major expenditure items for all LGUs in CAR from 2001-2006. The lion’s share of the expenditures goes to general services (42%), followed by others (21%) and health (16%).

Income class bracketing of LGUs are evaluated and changed every four years as mandated under EO No. 249, s.1987. Income classes are set by the BLGF as standards to measure the performance of each LGU. It provides the basis for the maximum amount expendable for salaries and wages, salary scales and rates of allowances, per diems and other emoluments that the LGU officials and personnel may be entitled to. It also dictates the number of Sanggunian members; predetermines the LGUs which are qualified for foreign and local loans and grants; and serves as the basis for increase or decrease of national allotment. Table 5 gives the income cutoffs for income classifications of LGU types for the year 2005.

The 2000 population and land area statistics per municipality used to compute for population density as well as the 2000 small area poverty estimates are sourced from the National Statistics Coordination Board (NSCB). As indicated in Figure 5, poverty incidences are generally high in areas located in the middle portion of the region where Abra, Kalinga, and Mountain Province are situated. Among all municipalities in CAR, Tanudan in Kalinga registered the highest poverty incidence in 2003 with 88.1% of its total population classified as poor. Next in the poorest list are Tinglayan (also in Kalinga) and Boliney (in Abra), which registered 82.1% and 73.5% incidence, respectively. According to the 2003 poverty study, variables found to be significantly correlated to income in CAR include those related to education, presence of amenities in the barangay (e.g., presence of a market, health center, street pattern), and accessibility to national highway. In consideration of the differences in the 2000 and 2003 NSCB poverty models, the former was used in this study as it provides a better baseline for the 2001-2006 LGU SIE data. Besides, the 2003 poverty estimates only became available upon their release in March 2009.

3.2 Empirical Specification and Method

To examine the local public finance (revenue side) and fiscal autonomy capability of municipalities in CAR, multiple regressions are employed for the period 2001 only due to insufficiency of data. This aims to investigate the effects of various revenue and expense accounts, control variables such as poverty incidence, and change in income status of municipalities on tax and non-tax revenue sources. In Equation 1, the dependent variable revenue (Rev) can be a tax or non-tax source following Feld et al. (2003, 12-13) and Manasan (2004, 40): real property tax (RPT); business tax (BusTax); Non-tax revenues (NontaxRev); and registration fees (Reg_fee).

The dependent variables are the ones of interest in explaining the independent or predictor variables. Regressions offer an indication of the size of any observed relationship and the likelihood of such relationship occurring by chance. Control variables are those which are explicitly held fixed when studying the effect of the independent variables on the dependent variable. The reason for controlling for poverty and change in income class is that the other independent variables may be correlated with other factors that also influence the dependent variable.

The right-hand side of Equation 1 includes explanatory and control variables. The explanatory variables or predictors include: registration fees per capita (Reg_fee) as a measure of own-source, non-tax revenue potential; total transfers from the national government or IRA per capita; total expenditure per capita (TotExp) as a measure of public goods provision; and population density (PopDen) as a measure of demand for public goods provision. Control variables include poverty incidence (Pov) as defined and operationalized by the National Statistics and Coordination Board for the year 2000 (baseline poverty estimate) and change in the qualitative income status of municipalities (Dum1) as determined by the BLGF. Dum 1 is zero if there is no change in the income status of the municipality from 2001 to 2005 as per BLGF classification, and 1 if there is a positive change (e.g., from a third to a second class LGU). No municipality exhibited a decline or negative change in status from 2001 to 2005. Control variables help determine or test whether the predicted relationship between the dependent and independent variables remains the same or varies under the different categories or conditions.

(Equation 1)

Following Fiva (2005, 12-13), various 2001 municipal revenue accounts are also regressed with different expenditure items as explanatory variables to investigate the public finance (uses of funds and provision of public goods side) with population as a control variable as shown in Equation 2. These major expenditure accounts include education per capita (EducExp); health per capita (HealthExp); social services per capita (SocSerExp); and economic services per capita (EcoSerEx). The independent variables include: revenue accounts such as real property tax per capita (RPT), business tax per capita (BusTax) and non-tax revenues per capita (NonTaxRev); total transfers from the national government or IRA per capita; and, population (Pop) as a direct measure of demand for public goods.

(Equation 2)

The various income and expenditure accounts referred to in the two models are also examined horizontally (through time) and vertically (across government levels) as part of financial analysis, following Berne and Schramm (1986). Hence, distribution of incomes and expenses across levels of government in CAR and by source are analyzed from 2001 to 2006. The growth rates of these income and expenditure accounts are also computed. Insights and highlights of the financial analysis are interwoven in the discussion of regression results.

4. Results and Discussion

In terms of the distribution of incomes across levels of government, the average total income for all LGUs in CAR from 2001 to 2006 is PhP 4.72 trillion, reaching a high of PhP 5.78 trillion in 2006 from a figure of PhP 3.70 billion in 2001 (see Table 6). Baguio City accounts for 15.32%, the six provinces for 37.51%, and the 76 municipalities for 47.17% of the average total income. Total income grew by 9.38% on average annually from 2001 to 2006 (see Appendix A for specific income account growth rates). Municipalities’ total income grew on average by 11.03%, outpacing the 8.94% growth registered by provinces and 6.21% of Baguio City for the same time period. The marked improvement in terms of absolute figures in municipalities’ total income is supported by the fact that majority of them retained their income classifications. Improving a notch higher are 24 and 19 out of 76 municipalities or 32% from 1997 to 2001, and 25% from 2001 to 2005, respectively. No municipality in CAR regressed in terms of income classifications from 1997 to 2001 and from 2001 to 2005. Hence, the non-significant dummy variable in the regression results in Table 6. All the municipalities are only able to generate 8.70% of total income from internal sources (see Table 7). The assignment of revenues has effectively shifted the distribution of own-source revenues from municipalities and provinces in favor of Baguio City in the region. The income from local sources of all the municipalities registered an average growth rate of 7.64% annually from 2001 to 2006 (see Appendix A).

The imposition of new taxes and the increase in local tax rates have become unpopular modes to improve the income of LGUs in CAR. The income from tax sources of all the municipalities accounts for 4.59% of total income. The lopsided distribution can be explained by the fact that only cities are allowed to impose all the taxes that provinces and municipalities are authorized to levy (see Table 2). It also gives cities a greater discretion in setting the tax rates. This is shown by the City of Baguio accounting for 63.78% of the average local tax revenue in the Cordillera of PhP 402 million 2001 to 2006, followed by municipalities at 25.15% and provinces at 11.07% (see Table 6).

For the different tax income sources, business tax remains the biggest contributor to total income with PhP 187.17 million followed by real property taxes (RPT) with PhP 177.22 million and other taxes with PhP 38.36 million (see Table 6). So, across levels of government, Baguio accounts for 59.84% on average of the total RPT collected on average of PhP 177 million from 2001 to 2006, followed by the municipalities at 21.51% and the provinces at 18.65%. Business tax revenues, or the annual tax imposed on the act of doing business within the LGU, continue to be dominated by Baguio City which accounts for 68.41% on average of all business tax collected from 2001 to 2006, followed by municipalities at 28.09% and provinces at 3.50%. Provinces are not assigned business tax powers under the LGC (see Table 2). Other tax revenues, averaging PhP 38.36 million from 2001 to 2006, include financial charges in the form of taxation imposed on transactions related to transfer of property/ ownership, practice of profession requiring government examinations and others.

Non-tax, own-source revenues amounting to PhP 317 million on average from 2001 to 2006 in CAR are financial charges in the form of fees for the rent of government property and purchase of government permits and forms. Baguio City accounts on average 42.67% of non-tax revenues, the provinces at 27.95% and municipalities at 29.38% for the region (see Table 6). Registration fees amounting to PhP 45 million on average from 2001 to 2006 are collected by Baguio City, the provinces and municipalities on average with the following distribution: 39.62%, 8.84% and 51.54%, respectively. User fees amounting PhP 54 million on average from 2001 to 2006 are collected across LGUs levels as follows: Baguio City, 53.38%; provinces, 23.37%; and, municipalities at 23.25%. Receipts from economic enterprise amounting to PhP 139 million on average from 2001 to 2006 are impositions from the operation of economic enterprises in connection with the LGUs’ exercise of their proprietary functions. Across levels of government, provinces dominate with 38.48% share in the average total receipts from economic enterprises from 2001 to 2006, followed by Baguio City at 33.88% and municipalities at 27.64%. Other non-tax receipts are revenues not falling in any of the non-tax revenue categories. This catch-all category accounts for PhP 77.67 million on average from 2001-2006, 53.50% of which are accounted for by Baguio City, 28.60% by municipalities, and 17.90% by the provinces. Non-tax revenues are the fastest growing own revenue source of the LGUs at 15.30% annually from 2001 to 2006, with the provinces leading the pack in all subcategories: 141% growth in registration fee collections; 116% growth in user fees; 185% increase in receipts from economic enterprises; and, 22.55% increase in other receipts (see Appendix A). This shows that with the limited taxing powers of provinces under the Local Government Code, the provinces have shifted their revenue generating measures to non-tax activities by fully exploiting their corporate powers and proprietary rights.

The total shares or transfers from the national government to all LGUs in CAR from the national government averaged PhP 3.86 billion from 2001 to 2006, reaching a high of PhP 4.68 billion in 2006 from its PhP 2.94 billion figure in 2001. In terms of government levels, municipalities get the lion’s share of these transfers at 52.30%, followed by provinces at 39.33% and Baguio City at 8.37%. Municipalities and provinces continue to be dependent on national government transfers with 90.18% and 85.31% of their total incomes derived from such allotments. Baguio City gets 44.19% of its total income from national transfers on average from 2001 to 2006. Average annual increase or growth rates from 2001 to 2006 of national government transfers stood at 5.65% for Baguio City, 9.20% for the provinces and 11.73% for the municipalities.

IRA received by CAR LGUs averaged PhP 3.66 billion from 2001 to 2006. IRA accounts for 65.5% of the national government transfers and 77.91% of total incomes from 2001 to 2006 on average. Under the LGC, the IRA is divided amongst the different levels of local government as follows: 23% each to the provinces and cities; 34% to municipalities; and, 20% to barangays. The IRA share of each tier of government is then apportioned to individual LGUs on the basis of population (50%), land area (25%) and equal sharing (25%). Per level of government, the IRA-share of municipalities and provinces are 53% and 40%, respectively on average of the total received by CAR LGUs from 2001 to 2006 while that of Baguio City is 7%. The IRA share to total income, or IRA dependency ratio, is highest for municipalities at 87.61%, followed by provinces at 83.95% and Baguio City at 33.34% on average from 2001 to 2006. The IRA dependency ratio of 76 CAR municipalities from 2001 to 2006 yield 456 observations, ranging from 38.70% to 99.85% with an average of 90.89% and a standard deviation of 11.43%. Table 8 shows the strong inverse relationship between IRA dependency and income class level. First class municipalities have an average of 70% IRA-dependency ratio which increases to 82% for second class municipalities. Third to fifth class municipalities have IRA-dependency ratios of 92% while it is 97% for sixth class municipalities. These are illustrated in Figure 6 where the distribution of CAR municipalities per income class as presented vis-à-vis their respective level of IRA-dependency ratios.

For the provinces, Benguet, a 2nd class province, has an IRA dependency ratio of 77.57% while Abra, Ifugao and Kalinga, all 3rd class provinces have IRA dependency ratio of 69.81%, 94.08% and 94.92%, respectively. Apayao and Mt. Province, both 4th class provinces, have IRA-dependency ratio of 86.30% and 94.92% respectively. Abra is the least IRA-dependent of all CAR provinces because it has increased its own-source revenue position by improving its receipts from user charges, economic enterprise and others. Apayao, on the other hand, even if it is on the same income class as Mt. Province, has lower IRA-dependency ratio because it has improved its own-source income particularly business tax collections. Table 9 and Figure 7 show the IRA dependency ratios of Baguio City and the six CAR provinces from 2001 to 2006.

With the above discussion of local government revenues or sources of public finance, regression is employed to investigate the factors influencing the various accounts for municipalities for the year 2001. Regression results on Table 10 show that poverty incidence as a proxy of average income and tax base in the municipalities is negatively related to per capita real property tax revenues, pointing to the inadequacy of the LGUs to provide for their own source of funds. Poverty incidence remains a limiting factor in the LGUs’ effort to raise revenues, whether tax or non-tax based. For example, the higher the incidence of poverty in the municipality, the lower the real property tax revenue. Per capita registration fees collected, as a non-tax internally generated fund, is positively related to real property and business tax per capita, implying the complementary nature of the fees collected to the two tax sources. Total expenditures per capita as a proxy for the municipalities’ budget size is a highly significant determinant of real property and business tax revenues. This implies that the size of local government expenditures exerts pressure to increase tax revenue collections. The relatively high, negative and highly significant coefficient of IRA per capita seems to suggest that national government transfers provides a disincentive for LGUs to increase their local tax and non-tax efforts. This finding points towards a bias for the status quo, perpetuating the LGUs’ IRA dependency. Further, the LGUs’ dependency on the IRA seems to point to the non-development and non-expansion of the local tax bases as well as the lack of innovation involving non-tax internal revenue generating schemes at the municipal levels. Non-tax revenues which include registration fees and user charges are negatively correlated with population density. The unexpected sign of population density as a proxy of non-tax revenue base seems to confirm the relatively underdeveloped non-tax internal revenue sources of municipalities. Real property tax complements the non-tax revenue collections as shown in the positive and significant coefficient.

For the uses of public finance, total expenditures for all the LGUs in CAR averaged PhP 4.07 trillion from 2001 to 2006, reaching a high of PhP 4.85 trillion in 2006 from a PhP 3.45 trillion. Across levels of government, all municipalities accounted for almost half or 48.90% of this amount, followed by the provinces at 38.26% and Baguio City at 12.84% for the period 2001-2006 as shown in Fig. 8.

Education covers the expenditures for the support of schools and education facilities, planning and manpower development, sports and cultural preservation and enrichment. Education is the lowest expenditure item in social services accounting only for 2.41% of total expenditures. Baguio spent 12% of its budget on education, but municipalities and provinces only spent 1% each from their education budget on average. Health covers the expenditures for the provision of housing and sanitary services, promotion of community development, zoning and control of pollution. Health services are the third highest expenditure item for CAR LGUs, accounting for 16% of total expenditures at PhP 665 million annually on average. Provinces spent 26% of their budgets on health while municipalities and Baguio only allotted 10% of their expenditures on such. Provinces account for 61% of total expenditures in health of all LGUs in CAR as the operation of district hospitals, for example, has been devolved to the provincial governments. Social services covers the expenditures for the upliftment of disadvantaged families and children, the rehabilitation of the physically and socially handicapped, assistance to distressed and displaced individuals and families, care of the aged and other welfare services and payment for retirement, pension and other social security benefits. This only accounts for 2.85% of average expenditures of all LGUs. Municipalities account for 62% of total expenditures on average, followed by provinces at 21% and Baguio with 17%. Economic services refer to activities directed to the promotion, enhancement and attainment of desired economic growth. This accounts for almost PhP 600 million of average expenditures from 2001 to 2006. Economic services expenditure is accounted for by provinces at 46% followed by municipalities at 35% and then Baguio at 18%.

With the above discussion of the uses of public finance at the local government level, regression is again employed for the same set of CAR municipalities for the year 2001 to investigate the drivers of the levels of the four major expenditure items. Various revenue accounts are regressed with different expenditure items with population as a control variable with results shown in Table 11. Real property tax per capita positively and significantly correlates with the following expenditure per capita items: education, health, social security services and economic services. Quite noticeable is the highly significant coefficient of real property tax with education expenditures. This can be explained by the fact that 50% of real property tax collections form part of the special education fund that LGUs use as support for the national government’s provision of free primary and secondary education. High real property tax collections would translate to higher education per capita expenditures. Business tax per capita is positively related to economic services expenditure per capita as the former feeds the latter. Non-tax revenues per capita positively and significantly affect health expenditures per capita at the municipal level. The negative and significant coefficient of per capita non-tax revenues on education and economic services expenditure per capita seems to point to the non-substitutability of tax revenues, particularly real property and business taxes, for non-tax revenues as fund source for the two expenditure items. Per capita IRA is positively and significantly correlated with education and social security services per capita.

5. Conclusion

The establishment of an autonomous regional government in the Cordilleras is premised on both development and decentralization imperatives that create opportunities for indigenous peoples to move forward. This could result in better outcomes, since delivering services to disadvantaged groups is more cost-effective at the local level. However, the high IRA-dependency ratio of LGUs due to their thin tax bases as a result of their low income levels (or poverty) in the Cordillera region would suggest that fiscal autonomy at the lowest levels (municipalities and provinces) should be addressed first before regional and political autonomy are granted and affirmed. Self-determination begins and is confirmed at the lowest levels of government. The creation of a regional government would mean siphoning tax and non-tax revenues from the LGUs as well as pushing for increased allocation from the national coffers before it can perform its redistributive and reallocation functions.

In conclusion, the paper demonstrates that there should be an identification of what public goods, as proxied by various expenditure items, need to be provided at the national, regional and local government levels, their catchment areas and spillover effects. Public goods expenditure on education and health, for example, are shown to be LGU-level specific. If and when there are regionally-provided public goods as opposed to national and local (provincial, city or municipality-based) public goods, then the assignment of the sources of funds, as illustrated by the tax and non-tax revenue items, should follow. Financing instruments currently available to local government units as provided for by various laws remain untapped and unutilized such as loans, bond flotation, build-operate-transfer (BOT) and its variants, privatization schemes, etc. A clear match between current and recurrent expenditure needs and revenue means remains problematic as a principle of fiscal autonomy. Good public finance and governance practice, matching long-term capital needs with clear earmarked revenue sources, continue to be a challenge in the region. Changes in the governance of public finance can theoretically lead, directly or indirectly, to improving the status of the Cordillera region and its people.

An earlier version of this paper appeared as “Public Finance and Fiscal Autonomy: The Case of the Cordillera Region in the Philippines” in Selected Proceedings of the Second International Conference on Social Sciences, Volume 3: Public Economics, Public Management & Public Policy, ed. Asim Balci, Fatih Savasan and Ozlem Ozkivrak, 41-45. Organized by the Social Science Research Society, the conference was held on September 10-11, 2009 in Izmir, Turkey.

REFERENCES

Berne, R. and R. Schramm. 1996. The financial analysis of governments. Englewood Cliffs, NJ: Prentice Hall.

Brillantes, Alex B. Jr. and Maximo B. Garming. 1998. The continuing quest for regional autonomy in the Cordilleras. In Tapales, Cuaresma, and Cabo, ed., vol. 2, 729-748.

Brillantes, Alex and Jocelyn C. Cuaresma. 1990. Local governments, local autonomy and decentralization. Quezon City: National College of Public Administration and Governance, University of the Philippines.

Buendia, Rizal G. 1998. The Cordillera autonomy and the quest for nation-building: Prospects in the Philippines. In Tapales, Cuaresma, and Cabo, ed., vol. 2, 689-728.

Casambre, Athena Lydia. 2010. Discourses on Cordillera autonomy. Baguio City: Cordillera Studies Center, University of the Philippines Baguio.

De Guzman, Raul P. and Mila A. Reforma. 1998. Decentralization towards democratization and development in the Asian Pacific region. In Tapales, Cuaresma, and Cabo, ed., vol. 1, 21-66.

Feld, Lars P., Gebhard Kirchgassner and Christoph A. Schaltegger. 2003. Decentralized taxation and the size of government: Evidence from Swiss state and local governments. CESifo Working Paper No. 1087.

Fiva, John H. 2005. New evidence on fiscal decentralization and the size of government. CESifo Working Paper No. 1615.

G.R. No. 80391, February 28, 1989, 170 SCRA 786, 794-795.

G.R. Nos. 79956 and 82217, January 29, 1990, 181 SCRA 495, 506.

G.R.  Nos. 93252, 93746, 95245, August 5, 1991, 200 SCRA 271, 281.

GR No. L-31004, January 8, 1971, 37 SCRA 1.

Llanto, Gilberto M. 2004. Rural finance in the Philippines: a continuing saga in policy challenges. PIDS Policy Notes, No. 2004- 06. Makati: Philippine Institute for Development Studies.

______. 2007. Reforming the BOT Law: a call of the times. PIDS Policy Notes, No. 2007-01. Makati: Philippine Institute for Development Studies.

Manasan, Rosario G. 2000. Improving tax administration: A new view from the theory of tax evasion in a corrupt regime. PIDS Policy Notes, No. 2000-11. Makati: Philippine Institute for Development Studies.

______. 2002. Explaining the decline in tax effort. PIDS Policy Notes, No. 2002-14. Makati: Philippine Institute for Development Studies.

______. 2004. Local public finance in the Philippines: In search of autonomy with accountability. PIDS Discussion Paper, Series No. 2004-42. Makati: Philippine Institute for Development Studies.

Musgrave, Richard A. and Peggy B. Musgrave. 1989. Public finance in theory and practice, 5th ed. New York: McGraw-Hill.

National Statistics Coordination Board. 2000 Municipal level small area estimates of poverty in the Cordillera Administrative Region.

______. 2003. Municipal and city level poverty estimates.

National Tax Research Center. 1998. Approaches to development in the autonomous regions. In Tapales, Cuaresma, and Cabo, ed., vol. 2, 747-768.

Oates, W.E. 1972. Fiscal federalism. New York: Harcourt Brace Jovanovich.

______. 1993. Fiscal decentralization and economic development. National Tax Journal 46 (2): 237-243.

Republic Act No. 7171. An Act to Promote the Development of the Farmers in the Virginia Tobacco-Producing Provinces, approved January 9, 1992.

Republic Act No. 8424 .Tax Reform Act of 1997 or the National Internal Revenue Code of the Philippines.

Republic Act No. 9337. Amending the National Internal Revenue Code of 1997.

Republic Act No. 6734. An Act Providing for an Organic Act for the Autonomous Region in Muslim Mindanao.

Republic Act No. 7160. Local Government Code of 1991.

Republic Act No. 6766. An Act Providing for an Organic Act for the Cordillera Autonomous Region, approved October 23, 1989.

Republic Act No. 8438. An Act to Establish the Cordillera Autonomous Region, approved December 22, 1997.

RDC-CAR. 2007. White paper on advancing the Cordillera agenda for regional autonomy. Baguio City: National Economic and Development Authority (NEDA)-Cordillera Administration Region (CAR).

Rood, Steven. 1989. Intergovernmental relations in a Cordillera autonomous region. Philippine Journal of Public Administration 33 ( 4): 379-391.

Shah, Anwar. 2005. A framework for evaluating alternate institutional arrangements for fiscal equalization transfers. World Bank Policy Research Working Paper, No. 3785. Washington, DC: The World Bank.

______. 2006. A practitioner’s guide to intergovernmental fiscal transfers. World Bank Policy Research Working Paper, No. 4039. Washington, DC: The World Bank.

Stiglitz, Joseph E. 2000. Economics of the public sector, 3rd ed. New York: W.W. Norton.

Tanggol, Sukarno D. 1998. Regional autonomy and social development: Some notes on the case of Muslim Mindanao. In Tapales, Cuaresma, and Cabo, ed., vol. 2, 631-662.

______. 1998. Managing the Muslim problem for Philippines 2000: Issues and prospects. In In Tapales, Cuaresma, and Cabo, ed., vol. 2, 663-680.

Tapales, Prosperina D. 1998. The nature and state of local government. In Tapales, Cuaresma, and Cabo, ed., vol. 1, 5-20.

Tapales, Prosperina D., Jocelyn C. Cuaresma, and Wilhelmina L. Cabo, ed. 1998. Local government in the Philippines: A book of readings, vol. 1 (Local Government Administration). Quezon City: Center for Local and Regional Governance and National College of Public Administration and Governance (NCPAG), University of the Philippines.

______, ed. 1998. Local government in the Philippines: A book of readings, vol. 2 (Current Issues in Governance). Quezon City: Center for Local and Regional Governance and National College of Public Administration and Governance (NCPAG), University of the Philippines.

Figure 1. Map of the Cordillera Region and the Philippines. Cordillera map courtesy of Alicia Follosco, Cordillera Studies Center.

Table 1. Comparative fiscal provisions of the Cordillera organic acts and the ARMM law.

Table 2. Tax assignments of local government units. Source: Sections 134 to 151 of RA 7160 or the Local Government Code of 1991.

Figure 2. Tax assignments and transfers in CAR, 2006. Source: Bureau of Internal Revenue and Bureau of Local Government Finance.

Table 3. Shares from national tax collections (transfers). Source: Bureau of Local Government Finance.

Table 4. Extraordinary receipts, grants and aids. Source: Bureau of Local Government Finance.

Figure 3. Distribution of incomes of all LGUs in CAR, 2006. Source: Bureau of Local Government Finance.

Figure 4. Distribution of Average Expenditures of All LGUs in CAR, 2001-2006

Table 5. Income classifications of LGU types as of 2005. Source: Bureau of Local Government Finance.

Figure 5. Poverty incidence per municipality and city in CAR, 2003.

Source: 2003 City and Municipality Level Poverty Estimates, National Statistics Coordination Board, p. 15.

Table 6. Distribution of incomes across levels of government in CAR by Source, 2001-2006. Source: BLGF, Statements of Income and Expenses, 2001-2006.

Table 6 continued.

Table 7. Distribution of LGU incomes in CAR by source, 2001-2006 (in %). Source: BLGF, Statements of Income and Expenses, 2001-2006.

Table 7 continued.

Figure 6. Income Class and IRA-Dependency Ratio, CAR Municipalities 2001-2006.

Table 8. IRA dependency ratio of CAR municipalities as to income class, 2001-2006.

Table 9. IRA Dependency Ratio (%) of Baguio City and CAR provinces, 2001-2006

Figure 7. IRA Dependency Ratio of Baguio City and CAR Provinces, 2001-2006

Table 10. Determinants of sources of public finance (tax and non-tax revenues)

Figure 8. Distributions of average expenditures across LGU level by source, 2001-2006.

Table 11. Determinants of local public finance uses (LGU sectoral expenditures).

Appendix A. Income account growth rates, 2001-2006, in percent. Source: BLGF, Statements of Income and Expenses, 2001-2006.

Appendix A continued.

Appendix B. Distribution of expenses across levels of government in CAR by source, 2001-2006. Source: BLGF, Statements of Income and Expenses, 2001-2006.

Appendix C. Distribution of LGU expenses in CAR by source, 2001-2006, in percent. Source: BLGF, Statements of Income and Expenses, 2001-2006.

Appendix D. Expense account growth rates, 2001-2006.

Source: BLGF, Statements of Income and Expenses, 2001-2006.