«Metropolitan City Finances in India: Options for A New Fiscal Architecture Roy Bahl International Center for Public Policy Working Paper 12-33 ...»
Comparable data for individual urban local governments are not readily available, but special surveys and case studies for individual cities have shown a weakening balance between own revenue capacity and expenditure needs. In a survey of the finances of 34 cities for 2004, Mohanty, et. al. (2007) found that nearly all faced annual revenue shortfalls. When revenue inflows were compared to expenditure needs based on a normative approach to identifying minimum requirements for service levels, the finding was that the revenue gap could not be covered with the current revenue structure. This study finds that on average, large cities finance about 45 percent of their expenditure from own sources, which is not a low number by international standards, but this statistic masks the problems that stem from the currently low level of spending.
Revenue effort. There are two routes to increasing metropolitan government capabilities to finance budgets from own sources: raising the revenue effort that is made by local governments, and strengthening the revenue raising powers of local governments. Both routes can be taken in India. There is considerable room for increased revenue effort, but vertical balance for large urban governments almost certainly will require enhanced revenue-raising powers.
The goal of raising the revenue effort by local governments in metropolitan cities might be addressed in a number of ways. User charges are rarely levied at cost Metropolitan City Finances in India: Options For A New Fiscal Architecture 13 recovery levels (Rao and Bird, 2011). With increased rates and stronger enforcement, revenues could be augmented significantly. Mohanty, et.al. (2007) studied 25 municipal corporations for 2004 and found that 16 recovered less than 20 percent of costs from charges and only 2 recovered as much as 75 percent.
The High Powered committee (2011, page 140) reports later data to show that only 8 of the 63 JNNURM cities recovered water and sewerage O&M costs, and only 6 recovered O&M costs for solid waste management services. The main bottlenecks to full cost recovery with user charges are political resistance, dissatisfaction with the level of services provided, and the complications that arise because tariff increases for urban services require state government approval.
There also is room to increase local tax effort in India. The mainstay of local government revenue systems is the property tax, but this remains underutilized with an effective rate against GDP at about one half the international average for developing countries (Mathur, et. al., 2009).
Measured by almost any yardstick, the Indian property tax is badly administered.
A survey of the property tax practice in the 36 largest urban local governments, carried out by Mathur, Thakur and Rajadhyaksha (2009), revealed that 44 percent of all parcels are excluded from the tax net, properties are assessed at about 30 percent of market value, and the average collection rate is about 40 percent.
They show that simply raising the coverage rate and the collection rate to 85 percent will increase the urban property tax in India to a level above the international average for low and middle income countries, though the average effective rate would still be lower than one percent of GDP.
In some metropolitan areas, the problem is legal or structural, suggesting that a different type of reform is needed. One example is the case of Mumbai where rent control legislation has eroded the tax base. To try and get around the problems of valuation and revaluation, and legal constraints on tax base growth such as rent control, some large cities have moved to a form of area based valuation, sometimes with success (Rao, 2008). In other cases the problem is an unwillingness to reform the property tax, arguably because it is so unpopular with voters. Two Indian states have abolished the property tax.
1314 International Center for Public Policy Working Paper Series
Revenue structure. Improved revenue effort on the existing tax base is not likely to be enough to establish vertical balance in the Indian system. In a study of urban local governments, Bandyopadhyay and Rao (2009) find evidence of a revenue gap that would exist even if these governments realized their full revenue potential. This problem can be addressed by assigning a broad-based revenue source to metropolitan local governments. From the international practice, we can observe that cities in low and middle income countries that have succeeded with developing a significant role for local revenues have done so with a broad-based tax as a mainstay of the revenue structure. A package of narrower based levies such as is common for local governments in developing countries – professions tax, entertainment tax, licenses, and the like – will not allow cities of 10 million plus populations to efficiently address their financing problems.
The fiscal instrument of choice in most countries seems to be a turnover tax. It has a broad base, high buoyancy and a relatively simple administration. The problem is that it is distortive because of its cascading feature, and it can give a great advantage to headquarters locations. Some cities – Buenos Aires and the Brazilian cities, for example – have attempted to address these disadvantages by including value added features, but others (Bogota, for example) seem willing to live with the flaws because of the revenue benefits and wait for the time when subnational government value added taxes become more feasible. The South African metropolitan cities relied heavily on a combination payroll-turnover tax, but this was abolished in 2010 and has not been replaced.
In the case of Indian metropolitan cities, the Octroi has been revenue productive, but its distortive features and its all-but-impossible administrative arrangements have led to its elimination in most places. This has significantly weakened the rate at which large cities finance services from own revenue sources. In Mumbai where it still is levied, octroi accounts for 45 percent of revenues, compared to about 25 percent for the property tax (Pethe, forthcoming).
Metropolitan City Finances in India: Options For A New Fiscal Architecture 15 There is no doubt that octroi is a highly distortive levy that imposes sizeable welfare costs on society, and on economic efficiency grounds its elimination is welcome. Complaints from the business community have heightened the calls for its complete elimination. But how will the lost revenues be replaced? To date, there has not been much progress on this front. Replacement with a turnover tax could restore lost revenues, but the distortive features would be retained. A derivation- based sharing of the VAT with large urban governments (with a higher local sur-rate in those cities) would be a better approach, even though it would have some distortive features and would create some incentives for protecting local producers. Rao and Bird (2011, p28) estimate that a one percent surcharge on the GST base could yield an amount equivalent to 0.34 percent of GDP.6 Note that if a metropolitan city had state government status (see below), it would automatically receive a share of the envisioned central-state value added tax.
Arguably a better option for a broad-based local tax – at least from a point of view of economic efficiency -- would be a payroll tax. To the extent that workers in formal sector employment are also residents in the city, the benefits test is passed. Such a tax can be easily collected on a place of work basis, and can be revenue productive though much less so than in the case of a turnover tax. It might also be levied as a surcharge on the federal income tax, so as to avoid additional administrative costs. The problems are the informal sector, which is not reached by such a levy, the imposition of a tax on labor in metropolitan areas where unemployment is at very high levels, and the situation in some cities where people are working in one jurisdiction but living in another. The ways around the latter problem are either a piggyback arrangement on the individual income tax, with a higher local sur-rate, and with distribution on a derivation basis, or the levy by a large metropolitan government rather than by a city whose taxing jurisdiction does not cover the entire area. Another major drawback to be resolved is the headquarters problem, which exists when a firm that operates nationally files a single return making it hard to sort out payroll amounts for specific local areas.
6 They also caution that movement away from a uniform rate structure would complicate the administration of the system.
There are other forms of taxation that could significantly increase revenue mobilization in large metropolitan cities. A better use of motor vehicle taxes could be greatly revenue productive. Licenses, parking, tolls, and even motor fuels are viable tax bases that can more of less pass the benefits test for local taxation if they are correctly levied (Bahl and Linn, 1992). The base of all of these taxes is growing. The High Powered Committee (2010, p56) pointed out that the motor vehicle population in India increased by 100 times between 1951 and 2004.
Moreover, taxes on motor vehicles, properly structured, might be justified as congestion charges. The problem is that this would be unpopular with certain voter groups, would probably raise transportation costs, and it would draw some revenues away from the state government.
Finally, there is the great potential of benefit charges, particularly those levied on land. In addition to the annual property tax, cities could be empowered to levy the transfer tax on real property and perhaps bring both under a uniform (or consistent) valuation approach. Various forms of betterment charges and impact fees could also contribute markedly to the revenue base.
A problem with all of these revenue options for metropolitan areas is the fragmentation of the local government structure. When there are multiple local governments operating in the same metropolitan area, the place where consumption or employment takes place may be different from the place of residence, and a mismatch occurs. In such cases, the property tax and benefit charges can work effectively but broad based taxes lead to significant interjurisdictional inefficiencies (Bahl, forthcoming).
The financing gap for urban local governments can be reduced by an increased flow of grants from higher to lower level governments. The justification for grant financing of local governments is usually to redress an imbalance between assigned expenditures and assigned taxing powers, to equalize, to correct for some external effect, or to encourage spending in some high priority area. There also are political reasons. Some elected local leaders see grants as a less risky political strategy than local taxation, and some elected state and central leaders Metropolitan City Finances in India: Options For A New Fiscal Architecture 17 see grants as a way to preserve their near-monopoly on broad based taxes and their degree of control over how the revenues are spent. Whatever the reason, intergovernmental transfers usually play a large role in the financing of subnational governments in most low and middle income countries. (Bird, 2006, Boadway and Shah, 2009).
Countries vary widely in terms of how they structure their grant systems, and how they treat large urban governments. Consider the following examples drawn from the unitary countries. The Chinese emphasize economic development and the main transfer instrument is shared taxes that are allocated on a derivation basis.
The provincial and local governments have no independent taxing powers. South Africa, on the other hand, allocates its transfers by an equalization formula and the largest metropolitan local governments receive little in grant revenues. Both Indonesia and the Philippines share a substantial portion of national taxes with local governments, with the allocation being by an objective formula that seems to favor local governments where expenditure needs are greater. Colombia operates a system of conditional grants to local governments, where the central government strictly controls the use of the funds.
There is also a lot of variation in the practice among federal countries. Pakistan and Argentina channel grants through one large program with a formula that emphasizes population. The grants are made directly to the provincial governments. Mexico’s grant system has a conditional and an unconditional sector, of about equal size.
What these few examples show is that countries structure their grant systems according to what they intend for these transfers to accomplish. In some countries, the metropolitan governments receive a significant share of the transfers (e.g., China), but in others the metropolitan governments raise most of the revenues from their own sources (South Africa and Argentina).
Transfers to state governments in India are made through three channels: Finance Commission grants, Development grants, and centrally sponsored schemes. The Finance Commission has recommended heavier allocations to local governments, but the decision is left to the state governments, with advice from their planning commissions, to decide whether to do this and if so, how it should be done. In all of this, there would not appear to be an urban strategy.
The JNNURM is a program of direct federal aid to large cities. This program does appear to be a strategy to address the infrastructure needs of urban areas, and also to provide local governments with incentives to undertake needed structural reforms. The amounts of assistance involved under this program are significant, and the policy reforms sought are important. But in its first iteration, the program suffered from design flaws. Most importantly, the proposed policy reforms did into materialize in many cases, and there was no conditionality built into the grant flow. Disbursement of the funds was slow, in part because of the matching requirement and in part because of the limited capacity of the local governments to implement the projects and the policy reforms.