Intergovernmental fiscal transfers and geographical disparities in local government income in the Philippines.
Periodically, the primary vehicle for inter-governmental fiscal transfers in the Philippines has come under criticism for having a counter-equalizing effect on regional development and disincentivizing local revenue generation. The fundamental claim is that the distribution formula (population, land area, and equal sharing) does not include an adequate horizontal equalizing feature and has thereby widened geographic disparities in levels of economic development (Manasan 2007). Consequently, there have been calls to redesign the system of intergovernmental transfers (Martinez-Vazquez and Liu 2001). Before such policy redesign can take place, a comprehensive understanding of the underlying dimensions of spatial inequality and the associated impact of intergovernmental transfers is required.
To address this need, this research note asks: (1) what is the impact of intergovernmental transfers on fiscal disparities between local governmental units (LGUs); (2) how do these disparities manifest across geographic divisions; and (3) how does the effect of intergovernmental transfers differ across different geographic divisions? The first question is answered by comparing a Gini and Theil coefficient calculated from locally sourced per capita income with a coefficient calculated from per capita income inclusive of the primary central government transfer---the internal revenue allotment (IRA). The analysis is conducted utilizing fiscal data collected at the city and municipal levels. To answer the second and third question, a subgroup decomposition of the Gini and Theil indices is employed. The decomposition analysis allows us to identify the spatial attributes underlying the general pattern of inequality amongst LGUs, inclusive and exclusive of the IRA. The spatial divisions examined include Super Regions, Regions, Provinces, Congressional districts, and terrestrial Eco-Regions.
The remainder of this note is structured as follows: section 2 briefly discusses intergovernmental fiscal transfers, both generally and in the context of the Philippines; section 3 describes different geographies over which disparities might occur; section 4 describes the methods and data used to answer the questions outlined in the preceding paragraph; section 5 presents the results of the data analysis; and the final section concludes with a discussion of the effects of intergovernmental fiscal transfers on geographic disparities in the context of calls for reform and redesign.
2. Intergovernmental Transfers
Intergovernmental fiscal transfers exist in some form in nearly all countries, varying in structure and purpose with a country's political-economic history, its human and physical geographic characteristics, and "the extent of government intervention in the economy" (Fjeldstad 2011). We can characterize each system of transfers by its purpose (what the system is intended to do) and its design (the formula for revenue assignment). The two primary purposes are as follows: (1) to equalize vertical fiscal imbalances; and (2) to equalize horizontal fiscal disparities. Secondary interests include correcting for jurisdictional externalities (economic spillovers) and rectifying bureaucratic weaknesses (Schroeder and Smoke 2003).
With regard to the design of intergovernmental transfers, two taxonomic distinctions can be made --whether transfers are horizontal or vertical and whether they are revenue or cost equalizing (Blochliger and Charbit 2008). The first distinction refers to whether transfers are made between levels of government (vertical) or between sub-national governments at the same level (horizontal). Vertical transfers can be used both for the purpose of remedying fiscal imbalances between levels of government (the vertical fiscal gap) or for remedying disparities between sub-national units at the same level (the horizontal fiscal gap). On the other hand, horizontal transfers deal specifically with remedying fiscal imbalances between governmental units at the same level. Since both types of transfers require a national government as a central administrator, the main difference is "where" the revenue is generated. For vertical transfers, funds are generally produced through national taxation--commonly a value added tax. Horizontal transfers, on the other hand, require revenue to be collected from and transferred between local governments.
The second distinction refers to whether transfers are intended to equalize differences in local revenue generation or to reduce differences in the cost of service provision. Revenue equalization schemes vary in their design, but are most commonly tax-sharing arrangements where funds are redistributed on the basis of a predetermined formula. The primary criticism of revenue equalization, particularly when there is no conditionality attached to the transfers, is that it can have a negative effect on sub-national tax raising capacity. In short, it dis-incentivizes local governments from developing their own tax base. Revenue equalization can be either horizontal or vertical in design, though in OECD countries revenue equalization tends to be associated with horizontal transfers (Blochliger and Charbit 2008). This, however, is not the case in the developing world where central governments have significantly greater taxation capacity than do local governments and thus bear the bulk of fiscal administrative responsibility. Cost equalization, on the other hand, describes the transfer of funds on the basis of service provision costs. There are often geographic and/or socio-economic differences in the cost of public good provision that require rectification if a minimum level of service is to be met. The primary criticism of this type of design is that cost equalization can contribute to budgetary inflation. The other criticism is that the cost differences vary in scale and character, and these variations are not easily captured by a single allocation formula.
Intergovernmental fiscal transfers in the Philippines are made through: (1) an IRA; (2) resource wealth sharing; and (3) ad hoc conditional and extraordinary aid grants. Of central importance is the IRA, which is the primary vehicle for both vertical and horizontal equalization in the Philippines. In general, the governmental fiscal situation is characterized by vertical fiscal imbalances where revenues exceed expenditures at the central level and at the local level expenditures exceed revenues (Guevera 2004). The imbalance is understood to be a substantive problem for growth and an obstacle to regional convergence. Nevertheless, national spatial planning in the Philippines is mostly focused around horizontal inequalities in local government income. For example, the 1991 Local Government Code states that roughly 40 per cent of the national internal revenue taxes collected by the Bureau of Internal Revenue be reallocated to LGUs. Revenues from national taxes are transferred to LGUs in accordance with the following allocation formula: provinces (23 per cent), cities (23 per cent), municipalities (34 per cent) and barangays (20 per cent). The allotment for each LGU type is further determined on the basis of population (50 per cent), land area (25 per cent), and equal sharing (25 per cent). Importantly, the IRA is a significant source of revenue for most LGUs.
Table 1 shows both the mean percentage of total revenue contributed by the IRA and the percentage range for the period 2001-10. Cities and municipalities, on average over the decade, received more than 64 per cent and 84 per cent of their total income from IRAs, respectively. The range of reliance on intergovernmental fiscal transfers is noteworthy. Some cities and municipalities effectively receive all of their revenue from national transfers (>96 per cent), while IRA transfers comprise only a marginal fraction for other cities and less than 12 per cent for other municipalities. Given the extraordinary fiscal centrality of the IRA, the question of how the IRA contributes to or otherwise rectifies spatial inequalities is paramount.
3. Geographic Disparities, Which Geography?
Spatial disparities in levels of economic development have been the subject of considerable attention in many disciplines over the last several decades. A great deal of attention has been paid to the effects of spatial inequalities on regional and national growth, but perhaps the greatest concern beyond macroeconomics is that large disparities in levels of development between sub-national regions can lead to violent conflict (Murshed and Gates 2005; Stewart 2008). The latter is central to Gurr's (1970) relative deprivation thesis, which forms the theoretical core of a large portion of the conflict studies literature.
For both reasons, discussions of spatial income inequalities are commonplace in Philippine economic discourse (Balisacan and Fuwa 2005). There has also been an accumulation of papers throughout the discipline exploring regional inequalities in different contexts and from different perspectives. For example, Fedorov's (2002) study of regional inequalities and polarization in Russia describes differences in levels of economic development across: (1) a West-East configuration; (2) national status-ethnic identity configuration; (3) a capital city size configuration; and (4) an exporting-non-exporting regional configuration. Kanbur and Zhang (2005) addresses spatial inequalities in education and healthcare and Kanbur and Zhang (1999) addresses regional per capita expenditure rates in China across urban-rural and inland-coastal divides. Murshed and Gates (2005) explore differences in rural-urban disparities, disparities across geographic zones (e.g. East, West, etc.), and income inequalities across ecological zones (mountains, hills, and plains). Also, a significant portion of the literature is focused on the urban-rural divide. For instance, Balisacan and Fuwa (2005) provide evidence of inequalities across the urban-rural divide in the Philippines. Similarly, Kanbur and Zhang (1999) show that the contribution of urban-rural disparities to overall regional inequality in China far exceeds other differences.
Explanations for geographic disparities in wealth can be divided into policy-oriented explanations and market-oriented explanations. For example, Jian, Sachs and Warner (1996) argue that socialist planning in China during the 1960s and 1970s favoured rich industrial regions, leading to a divergence of urban and rural outcomes. Kanbur and Zhang (2005) likewise demonstrate that the evolution of regional inequality in China mirrors political-economic phases--from heavy-industry development, to decentralization, to an open economic transition. Thus, the causes appear policy-driven. The other, and certainly not mutually exclusive explanation, is that spatial inequalities are "due to the natural advantages of some regions relative to others and to the presence of agglomeration forces, leading to clustering of activity" (Venables 2005). In short, given homogeneous regions and perfect competition, incomes will eventually converge, but economic development is unequal due to natural environmental heterogeneity and the tendency for economic activity to cluster.
Generally speaking, we do not have a clear definition of what constitutes the "spatial" and which "geographies" underlay unequal patterns of economic development in less developed countries. Certainly the urban-rural income gap is an important issue and is widely discussed, but it is not unique to developing economies. Regional inequalities in developing countries also have roots in the legacy of colonial economies and in contemporary political processes. This highlights a point made by Stewart, Brown and Mancini (2010) in their review of tools for measuring horizontal inequalities, and echoed by Elbers et al. (2008) in their article describing a new method for interpreting between-group inequalities. Determining which groups are relevant requires in-depth investigation and consideration of multiple classifications (e.g. ethnic, religious, linguistic, physiographic, or ecological). The remainder of this research note follows this path to develop our understanding of the geographic patterns underlying the Philippines' spatial economy.
4. Methods and Data
The analysis presented in the remainder of this note relies on the Gini coefficient and Theil-T index (Theil index) of inequality and their subgroup decompositions to understand the impact of intergovernmental fiscal transfers on geographic inequalities in development, measured as the per capita income of local governmental units (cities and municipalities). First, the inequality indices were used to characterize the distribution of wealth across all LGUs, regardless of location or geography. Next, each LGU was grouped based on locational and geographic attributes. A sub-group decomposition of each index was then conducted. This two-step analysis was completed using wealth measures that include locally generated income only and also real income, inclusive of the IRA.
The Gini coefficient is widely used in the regional inequality literature, has a clear interpretation, and can be decomposed into a between, within and overlap components. The overlap component is particularly useful for identifying the degree of stratification within groups. General entropy class of measures of inequality, including Theil-T, are used in a near identical manner and have the added quality of being perfectly decomposable into between and within group inequalities. Both measures are utilized and compared throughout. I also utilize Elbers et al.'s (2008) method for indexing between-group inequality with the Theil index as the underlying statistic of interest.
The Gini coefficient, using LGUs as the unit of observation, is calculated using the following formula (Shankar and Shah 2003):
G = (1/2[[bar.y].sub.u]) x 1/n(n-1) [n.summation over (i)] [n.summation over (j)] [absolute value of ([y.sub.i] - [y.sub.j])]
where G is the Gini coefficient ranging from 0 to 1, [y.sub.i] and [y.sub.j], are GDP per capita for each LGU, n is the number of LGUs, and [[bar.y].sub.u] is the unweighted mean of the GDP per capita for LGUs. The Theil-T index takes the following form:
T = [1/n] [summation over (i)] ([y.sub.i]/[bar.y]) ln ([y.sub.i][bar.y])
where [y.sub.i] is the per capita income of local government unit i and [bar.y] is the mean of the GDP per capita.
A Gini coefficient and a Theil index value were calculated using the above formulas for cities and municipalities separately. Statistics were calculated using per capita income based only on locally sourced revenue (L) and per capita income based on locally sourced revenue inclusive of transfers from IRAs (L + T). Given that the calculations are made using the same observational units, the Gini and Theil values are perfectly comparable for identical years. However, due to jurisdictional changes and the failure of some municipalities to report their fiscal situation on a regular basis, the measures are not perfectly comparable across time, though a general trend can be ascertained. Notably, the empirical results presented here are stable over time. The numbers of observations for each year is reported.
Both the contribution of spatial and geographic disparities to overall inequality and whether those disparities are exacerbated by intergovernmental fiscal transfers is explored using a sub-group decomposition of the Gini coefficient. The decomposition method is that of Bhattacharya and Mahalanobis (1967). The sub-group decomposition takes the following form (Lambert and Aronson 1993), where G, the overall Gini coefficient, can be represented as:
G = [G.sub.B] + [SIGMA] [a.sub.k][G.sub.k] + R
where [G.sub.B] is the between-group Gini coefficient, [a.sub.k] is the product of group share and income share for subgroup k, [G.sub.k] is the Gini coefficient within subgroup k, and R is the residual overlap. Logically, the within-group coefficient describes the distribution of units within groups and the between-group coefficient describes the distribution of across groups. The overlap term describes the degree of separation between groups --stratification (Lambert and Aronson 1993). For example, R = 0 when incomes do not overlap between groups. In the results section that follows, it is shown that intergovernmental transfers often increase the overlap term, which can be interpreted as it having a de-stratifying effect.
I also decompose the Theil index, which is one of a number of general entropy measures that are perfectly decomposable (i.e., it can be decomposed into between and within components without a residual term). The decomposition assumes the following form (Bellu and Liberati 2006):
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where the WITHIN term is the weighted average of the Theil inequality indexes of each group ([T.sub.k]), with the weights representing the number of LGUs falling into each group. The BETWEEN term is the Theil index where unit-level incomes are replaced by subgroup means.
The results of each decomposition is reported as a fraction of overall inequality (I) explained by between-group differences ([R.sub.B] -[I.sub.B]/I), the fraction contributed by within-group differences ([R.sub.w] = [I.sub.w] /1), and for the Gini coefficient decomposition the additional contribution to overall inequality by between-group overlap ([R.sub.0] = R / G). Cowell and Jenkins (1995) describe these measures as being akin to the [R.sup.2] in regression analysis in the sense that they report how much a given partition of the data "explains" overall inequality.
An alternative index that describes between-group inequality is derived using the method of Elbers et al. (2008). The method is motivated by the fact that between-group inequality can only account for 100 per cent of overall inequality in the special case where groups are completely homogenous and variation exists only between groups. Moreover, the traditional between-group inequality measures described above cannot, by definition, decrease as the number of subgroup partitions increase (Cowell and Jenkins 1995). Elbers et al. (2008) argue that such a benchmark for the contribution of between-group differences to overall inequality is extreme and has few real world applications. They make the case that observed between-group inequality should be compared against maximal between-group inequality, calculated from the actual data, the number of groups, and the relative size of groups. The resulting index from Elbers et al. (2008) is given by the following:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where Max [I.sub.B] is between-group inequality when sub-groups occupy non-overlapping intervals. Note that the measure is not a substitute for [R.sub.B] = [I.sub.B] / I (per cent of overall inequality contributed by between group differences). Rather, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] can be thought of as a measure of between-group intensity that is indifferent to the resolution of the group partition. Thus, it can be used to make robust evaluations regarding the salience of alternative groupings of different sizes.
The main data source for the analysis that follows is the Statement of Income and Expenditures for Local Government Units, published by the Philippines' Bureau of Local Government Finance of the Department of Finance. The statements provide information on revenue generation and expenditures. Complete data is available for provinces, cities and municipalities in spreadsheet form from 2001 to 2010. Included in this collection are measures of total income earned from local sources, notably tax (real property and business taxes) and non-tax revenues (regulatory fees, service charges, etc.), and income earned from LGUs' share of national wealth, including IRAs, extraordinary aid and borrowing. At the time of writing, per capita income inequalities between LGUs have only been studied using provincial level statistics (Pede, Sparks and McKinley 2012). Thus, this analysis represents the first time city and municipal level income aggregates are utilized in such a study.
The following analysis utilizes per capita income as the primary measure of interest, a measure not directly included in the LGU fiscal dataset. To calculate a simple per capita figure, population estimates for provinces, cities and municipalities were taken from CityPopulation.de (Brinkhoff 2014). This data is attributed to the National Statistics Office of the Philippines. One difficulty with the use of census figures in this manner is that they are only available for 2000 and 2010, corresponding to the two most recent national census takings. To generate reasonable population estimates for cities and municipalities for the intercensal years, I extrapolated population figures by way of an arithmetic rate of increase (United Nations 1952). The extrapolated figures were then utilized to derive per capita income generated from local revenue and from revenue inclusive of the IRA.
Tables 2 and 3 report the Gini coefficient and Theil index values for both the per capita income earned from local sources (L) and per capita income inclusive of the IRA (L + T). The unit of observation is either cities (Table 2) or municipalities (Table 3). Several results are worth noting. First, inequality in local income generation across national space is modest. This is consistent with expectations. Municipal level measures tend to show relatively lower levels of inequality than measures of Gini and Theil based on household income surveys. Even with this being the case, Balisacan and Fuwa's (2005) research on household income in the Philippines also finds relatively low levels of spatial income inequality. Thus, the general characterization of extreme spatial income inequalities and the need to rectify those differences through policy is perhaps premature. Nevertheless, there are some caveats to this finding. First, the Manila metropolitan region as a whole is many times larger economically than the rest of the Philippines, but the effect of individual component cities and municipalities located in the metropolitan region on the indices characterizing differences across the entirety of the country is small.
Second, there is an equalizing effect amongst LGUs that occurs once the central government fiscal transfer is included in the calculations. Across national space, municipal and city income inequalities decrease by more than 50 per cent with the inclusion of the transfer. This result is corroborated by both inequality indices. Third, inequalities in both local income generation and total income receipts have remained consistent over the last decade. One exception is the large spike in the Theil index for municipal local income generation in 2009 (Table 3). The Theil index is sensitive to raw value changes, and in 2009 the perennial best-performing municipality--Alfonso Castaneda, Nueva Vizcaya--increased its income threefold. They received an award by the national government for their success, but their income returned to its normal reporting levels in 2010. Even with this outlier included, the Philippines' spatial economy is generally stable.
The question remains as to how the identified inequalities intersect with spatial and geographic differences in levels of economic development. Table 4 reports the ten year average of the decomposition from the Gini and Theil indices as percentage of overall inequality explained by between, within and overlap components. First, looking at the extended Gini decomposition for local revenue only, the minimum values are as follows: Eco-Regions contribute 50 per cent (cities) and 36 per cent (municipalities) to overall inequality, respectively. Conversely, differences between Congressional districts contribute 98 per cent and 77 per cent to general inequality in cities and municipalities, respectively. Between-group differences across other grouping, such as administrative Super Regions and Eco-Regions, contribute markedly less to overall LGU fiscal inequalities for both cities and municipalities. The results are suggestive of a national economic geography in which LGUs are differentiated within, but not across broad areas.
The results of the Gini decomposition, suggesting large spatial and geographic disparities, is not corroborated by the Theil index decomposition. More importantly, the Theil's within component is much larger for nearly all group types. For example, 76 per cent of inequality between cities and 86 per cent of inequalities between municipalities in local revenue generation is explained by differences within Eco-Regions. The source of the more general within-group inequality is clear from looking at the Gini decomposition. Little or no LGU inequality is contributed by distinct within-group differences, but rather it is the result of minimal economic stratification. In other words, there are very large raw differences in LGU income within spatial groups and a lack of stratification between groups. The results suggest that economic inequalities are not distributed sharply across national space (i.e., the notion of strictly poor South and strictly rich North is not supported).
Finally, the inclusion of the IRA modestly reduces overall inequality, but does not significantly impact the pattern underlying disparities between LGUs. For cities, the amount of inequality explained by differences between Congressional districts (98 per cent) and Provinces (85 per cent) decreases by 2 per cent and 12 per cent, respectively. For municipalities, the amount of inequality explained by differences between Congressional districts (77 per cent) and Provinces (69 per cent) increases marginally or does not change. There is no evidence that the primary vehicle for intergovernmental transfers exacerbates inequalities along spatial and geographic lines. Even where the Theil index identifies within differences in fiscal status as the primary contributor to overall inequality, the inclusion of the IRA reduces between-group inequalities or has a negligible effect.
The number of group partitions is known to have an effect on decomposition results. Between-group contribution to inequality often increases as the number of group partitions increases, but cannot decrease. Thus, I also report the result of Elbers et al. (2008) index's between-group inequality described earlier. Figure 1 shows the results from this analysis in graphical form for each of the ten years under study. First, the contribution of between-group differences to overall inequalities is reduced by the intergovernmental transfer for both cities and municipalities. The effect is larger for cities. Second, the greatest contribution to inequalities is the differences between Congressional districts and Provinces. This result corroborates those shown in Table 4. Thirdly, there is a systematic ordering of between-group contributions to inequality. For example, as one moves in size from Super Regions to Regions to Provinces and then to Congressional Districts, the between-group contributions to overall inequality increases. This suggests that economic development, at least in broad terms, forms a hierarchical arrangement and the primary source of inequality is an intra-regional urban-rural process. Further substantiation of this claim requires a somewhat different analytic approach, but the analysis is strongly indicative. Finally, between-group contribution to overall inequality has generally remained steady over the 2000s. The exception to this consistency are the contributions by Super Regional and Regional differences in local revenue generation to overall inequality between LGUs, which have both declined steadily over the decade. The decline predates the actual formation of Super Regions and the result is not robust to the inclusion of the IRA. Thus, the horizontal equalization in local revenue generation is less the result of a formal policy intervention than a rationalization of regional economic development over the last decade through improvements in local income generation. In fact, the majority of the broad-scale development planning of the Philippines has long been concerned with rectifying these major regional inequities and this appears to have had a modest effect.
The findings presented in this research note provide a renewed understanding of geographic disparities in per capita income in the Philippines and the role that vertical intergovernmental transfers play in remedying or otherwise exacerbating these income inequalities. Gini and Theil index of inequality statistics for local government per capita income, inclusive and exclusive of the primary intergovernmental fiscal transfers, were compared and decomposed for several different types of geographic groupings. The major findings are fourfold. First, the IRA does reduce inequality in LGU per capita income across national space. Second, between-group (interregional) differences in municipal income levels vary by group definition. The largest between-group contribution to national-level inequality is those contributed by differences across Provincial and Congressional districts, suggesting a political-administrative determinant of inequality beyond the specific effects of fiscal institutional design. Third, LGUs are not strictly stratified, indicating local economic formations, rather than a spatial economy stretched over national space. Fourth, spatial income inequalities, inclusive and exclusive of the IRA, have been remarkably stable over the decade spanning the years 2001 to 2010.
One objective of this research was to inform policymaking and to better understand how fiscal policy and institutions can be formulated
to address local developmental inequalities. On this point, the system of intergovernmental fiscal transfers enacted by the Local Government Code does not include specific horizontal equalizing features, but the nature of geographic disparities in the Philippines is such that the distributional formula associated with IRA does lead to an overall flattening of inequalities across national space. Some inter-regional/between-group differences, however, remain high. Thus, immediate concern for substantial between-group inequalities is warranted, but a complicated reconfiguration of the system of IRAs is not required to solve this problem. There is no immediate evidence that transfers are the cause of horizontal inequalities or contributing to a worsening of the inter-LGU fiscal situation. Rather, the extent and cause of uneven development requires further examination.
The largest income differences, though not explicitly explored here, occur over the urban rural divide. Currently, municipalities are allocated 11 per cent more revenue via the IRA than are cities. Because there are almost twenty times more municipalities than there are cities, the funds are malapportioned. Basic changes to the apportionment formula based on the number and types of LGUs can lead to an improvement in the income gap between rural and urban LGUs. Finally, how should we address the belief that transfers reduce the will to raise revenue locally? There is no easy solution to this problem and it is difficult to disentangle whether the IRA is the cause or the effect. There are considerable differences in local government revenue raising capacity. Some LGUs received nearly all of their income (~99 per cent) from the IRA while others receive less than 1 per cent. Further research is needed to understand the consequences of this reliance. Clearly, improving LGU capacity to raise revenue will reduce income inequalities at the governmental level beyond what can be accomplished with any intergovernmental transfer scheme.
Adam Yeeles is a research scholar examining issues related to international development and human-environmental systems. He has served as a Lecturer at the School of Economic, Political and Policy Sciences in the University of Texas at Dallas, 800 West Campbell Road, Richardson, TX 75080-3021, and National Louis University; email: firstname.lastname@example.org
Balisacan, Arsenio M. and Nobuhiko Fuwa. "Changes in Spatial Income: Inequality in the Philippines: An Exploratory Analysis". In Spatial Disparities in Human Development: Perspectives from Asia, edited by Ravi Kanbur, Anthony J. Venables and Guanghua Wan, pp. 207-32. Tokyo: United Nations University Press, 2005.
Bellu, Lorenzo Giovanni and Paolo Liberati. Policy Impacts on Inequality: Decomposition of Income Inequality by Subgroups. Food and Agriculture Organization of the United Nations, 2006.
Bhattacharya, N., and B. Mahalanobis. "Regional Disparities in Household Consumption in India". Journal of the American Statistical Association 62, no. 317 (1967): 134--61.
Blochliger, Hansjorg and Claire Charbit. "Fiscal Equalisation". OECD Economic Studies 44, no. 1 (2008): 1-22.
Brinkhoff, Thomas. City Population <http://www.citypopulation.de> (accessed 1 June 2014).
Cowell, Frank A., and Stephen P. Jenkins. "How Much Inequality Can We Explain? A Methodology and an Application to the United States". Economic Journal 105, no. 429 (1995): 421-30.
Elbers, Chris, Peter Lanjouw, Johan A. Mistiaen and Berk Ozler. "Reinterpreting Between-group Inequality". Journal of Income Inequality 6 (2008): 213-45.
Fedorov, Leonid. "Regional Inequality and Regional Polarization in Russia, 1990-99". World Development 30, no. 3 (2002): 443-56.
Fjeldstad, Odd-Helge. "Intergovernmental Fiscal Relations in Developing Countries: A Review of Issues". CMI Working Paper, Chr. Michelsen Institute Development Studies and Human Rights, Bergen, Norway, 2011.
Guevera, Milwada M. "The Fiscal Decentralization Process in the Philippines: Lessons from Experience". Paper presented at the International Symposium on Fiscal Decentralization in Asia Revisited, Hitotsubashi University Tokyo, 20-21 February 2004.
Gurr, Ted Robert. Why Men Rebel. Princeton, NJ: Princeton University Press, 1970.
Jian, Tianlun, Jeffrey D. Sachs and Andrew M. Warner. "Trends in Regional Inequality in China". China Economic Review 1, no. 1 (1996): 1-21.
Kanbur, Ravi and Xiaobo Zhang. "Which Regional Inequality? The Evolution of Rural-Urban and Inland-Coastal Inequality in China from 1983 to 1995". Journal of Comparative Economics 27, no. 4 (1999): 686-701.
--and Xiaobo Zhang. "Spatial Inequality in Education and Health Care in China". China Economic Review 16, no. 2 (2005): 189-204.
Lambert, Peter J. and J. Richard Aronson. "Inequality Decomposition Analysis and the Gini Coefficient Revisited". Economic Journal 103, no. 420 (1993): 1221-27.
Manasan, Rosario G. "IRA Design Issues and Challenges". Policy Notes 2007-09, Philippines Institute for Development Studies, Makati City, Manila, 2007.
Martinez-Vazquez, Jorge, and Yongzheng Liu. "Philippines: Designing a Local Government Enhancement Fund". Southeast Asia Working Paper Series, Asian Development Bank, Manila, Philippines, 2001.
Murshed, S. Mansoob and Scott Gates. "Spatial-Horizontal Inequality and the Maoist Insurgency in Nepal". Review of Development Economics 9, no. 1 (2005): 121-34.
Pede, Valerien O., Adam H. Sparks and Justin D. McKinley. "Regional Income Inequality and Economic Growth: A Spatial Econometrics Analysis for Provinces in the Philippines". Paper presented at 56th AARES Annual Conference, Fremantle, Western Australia, 2012.
Schroeder, Larry and Paul Smoke. "Intergovernmental Fiscal Transfers: Concepts, International Practice, and Policy Issues". In Intergovernmental Fiscal Transfers in Asia: Current Practice and Challenges for the Future, edited by Paul Smoke and Yun-Hwan Kim, pp. 20-59. Manila: Asian Development Bank, 2003.
Shankar, Raja and Anwar Shah. "Bridging the Economic Divide within Countries: A Scorecard on the Performance of Regional Policies in Reducing Regional Income Disparities". World Development 31, no. 8 (2003): 1421-41.
Stewart, Frances, ed. Horizontal Inequalities and Conflict: Understanding Group Violence in Multiethnic Societies. New York: Palgrave Macmillan, 2008.
--. Graham Brown, and Luca Mancini. "Monitoring and Measuring Horizontal Inequalities". CRISE Overview, no. 4 (2010): 46.
United Nations. "Estimates Derived by Extrapolation of Census Results". In Manual I: Methods for Estimating Total Population for Current Dates. United Nations Publications, 1952.
Venables, Anthony J. "Spatial Disparities in Developing Countries: Cities, Regions, and International Trade". Journal of Economic Geography 5, no. 1 (2005): 3-21.
TABLE 1 Summary of Internal Revenue Allotments, 2001-10 % of Total LGU Type Income Min. % Max. % Cities (a) 0.64 0.07 0.96 Municipalities (a) 0.84 0.11 1 NOTE: a. The number of cities and municipalities used to calculate the statistics varies by year (reported in Tables 2 and 3). SOURCE: Philippines' Bureau of Local Finance Government. TABLE 2 Gini and Theil Index of Inequality for LGU Per Capita Income Across Cities Income Index Source (a) 2001 2002 2003 2004 2005 Gini L 0.48 0.49 0.51 0.5 0.5 L + T 0.29 0.2 0.21 0.2 0.2 Theil L 0.46 0.49 0.53 0.52 0.49 L + T 0.14 0.08 0.09 0.09 0.09 N 115 115 115 115 116 Income Index Source (a) 2006 2007 2008 2009 2010 Gini L 0.5 0.5 0.52 0.5 0.5 L + T 0.2 0.21 0.2 0.21 0.21 Theil L 0.5 0.51 0.52 0.48 0.49 L + T 0.09 0.09 0.08 0.09 0.09 N 116 117 134 135 115 NOTE: a. L = Revenue generated solely from local sources; L + T = Revenue generated from local sources + IRA transfer. SOURCE: Author's estimation. TABLE 3 Gini and Theil Index of Inequality for LGU Per Capita Income Across Municipalities Index Income Source 2001 2002 2003 2004 2005 (a) Gini L 0.5 0.51 0.5 0.5 0.51 L + T 0.28 0.28 0.27 0.26 0.26 Theil L 0.52 0.57 0.53 0.52 0.55 L + T 0.25 0.23 0.22 0.22 0.21 N 1485 1485 1486 1493 1491 Index Income Source 2006 2007 2008 2009 2010 (a) Gini L 0.52 0.51 0.51 0.56 0.52 L + T 0.28 0.28 0.28 0.3 0.28 Theil L 0.56 0.55 0.56 0.92 0.59 L + T 0.27 0.27 0.28 0.35 0.29 N 1495 1499 1483 1405 1314 NOTE: a. L = Revenue generated solely from local sources; L + T = Revenue generated from local sources + IRA transfer. SOURCE: Author's estimation. TABLE 4 Group Decomposition of Inequality Indices, Mean Values 2001-10 Gini LGU Type Group Income [R.sub.B] Source (a) City Super Region L 0.62 L + T 0.27 Region L 0.74 L + T 0.49 Province L 0.85 L + T 0.73 Congressional L 0.98 L + T 0.96 Eco-Regions L 0.50 L + T 0.26 Municipalities Super Region L 0.42 L + T 0.39 Region L 0.57 L + T 0.55 Province L 0.69 L + T 0.69 Congressional L 0.77 L + T 0.78 Eco-Regions L 0.36 L + T 0.34 Gini LGU Type Group [R.sub.W] [R.sub.O] City Super Region 0.24 0.14 0.27 0.46 Region 0.07 0.19 0.08 0.43 Province 0.04 0.10 0.04 0.23 Congressional 0.00 0.02 0.00 0.04 Eco-Regions 0.34 0.16 0.34 0.41 Municipalities Super Region 0.23 0.35 0.24 0.37 Region 0.06 0.37 0.06 0.39 Province 0.01 0.29 0.01 0.29 Congressional 0.00 0.22 0.00 0.22 Eco-Regions 0.28 0.36 0.26 0.40 Theil LGU Type Group [R.sub.B] [R.sub.W] City Super Region 0.36 0.64 0.06 0.94 Region 0.52 0.48 0.22 0.78 Province 0.64 0.36 0.43 0.57 Congressional 0.98 0.02 0.96 0.04 Eco-Regions 0.24 0.76 0.08 0.92 Municipalities Super Region 0.15 0.85 0.14 0.86 Region 0.28 0.72 0.24 0.76 Province 0.41 0.59 0.43 0.57 Congressional 0.54 0.46 0.54 0.46 Eco-Regions 0.14 0.86 0.16 0.84 NOTE: (a) L = Revenue generated solely from local sources; L + T = Revenue generated from local sources + IRA transfer. SOURCE: Author's estimation.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||RESEARCH NOTE|
|Publication:||Journal of Southeast Asian Economies|
|Date:||Dec 1, 2015|
|Previous Article:||The impact of political regime and governance on ASEAN economic growth.|
|Next Article:||Narratives on Malaysia's contemporary issues.|