Intergovernmental fiscal relations
: a case study of Pakistan

  • Syed Wasti

    Student thesis: Doctoral Thesis


    This dissertation explores the effect of intergovernmental fiscal transfers on the fiscal operations of the federating units in Pakistan. It is the first attempt to carry out disaggregated analysis of the fiscal behavior of each of the four provinces of Pakistan, separately and jointly in response to various types and categories of federal transfers, grants and borrowings. Thus it is a noteworthy addition in the empirical literature in the context of a less developed and resource constraints country where the sensitivities are always attached in the determination of distributional criteria and allocation of transfers among federating units.

    The study explores whether federal transfers to provinces have been utilized for stimulating provincial public expenditures or have largely been substituted for fiscal efforts to collect taxes from provincial own resources. The study also investigates the impact of unconditional and conditional transfers to determine the varying effects on public spending. Moreover, the phenomenon of “Flypaper Effect” which hypothesizes that the federal transfers and the provincial gross domestic product (resident income) have similar accelerating or multiplier effect on provincial expenditures is also examined. In addition, it is also attempted to scrutinize the role of federal transfers in the process of fiscal equalization among provinces as regards to the provision of public services.

    The government expenditure method is applied to determine the quantity of public service provision. To estimate the provincial fiscal response to federal transfers, total provincial expenditure is modeled as a function of provincial gross domestic product at factor cost, several types of transfers and total borrowings. Various macro, fiscal and demographic variables are also used in the estimation process to remove simultaneity in grants and provincial expenditures and also to control for diverse socioeconomic characteristics of federating units. All variables are adjusted with the respective provincial population and measured in constant prices. The expenditure functions are estimated by using Ordinary Least Square and Two Stage Least Square estimation techniques through E-VIEWS software covering the period between 1973 and 2009. Various specifications of expenditure function are used in this study for testing different hypothesis.

    The empirical results establish the importance of our thesis as estimated fiscal behavior of individual provinces show significantly varied responses. The findings of the study clearly highlight that aggregate or joint provincial fiscal response is more similar to the behavior of two relatively developed Punjab and Sindh provinces. The direction, marginal effects and the level of significance of coefficients measuring fiscal response to federal transfers for these two provinces is significantly different as compared with the other two underdeveloped Khyber Pakhtunkhwa and Baluchistan provinces. Thus, any attempts to draw conclusions regarding provincial response to federal transfers based on combined data may mislead mainly due the diverse socio and demographic characteristics and also because of the varied levels of economic development of federating units. Therefore, to design transfer strategies in Pakistan disaggregate analysis of provincial fiscal behavior is imperative.

    It is affirmed that the conditional grant have a larger and more elastic effect on provincial expenditure compared to unconditional grants. Therefore conscious effort is needed to design appropriate transfer strategies by attaching conditionality to its spending. The findings also suggest that unconditional transfers are heavily utilized for substitution of effort to raise revenue from own resources. Hence for rewarding improved provincial fiscal effort, some premium may be attached on the achievement of certain level of social services. This reward should be in the form of close ended matching grant to avoid its likely misuse. Similarly conditional matching incentives may also be given for goods of high federal priority but attracts lower provincial investment. Conditional grants though compromise the objective of provincial expenditure autonomy but nevertheless these greatly enhance the multiplier effect of fund transferred.

    The significant and much higher provincial dependence on federal transfers are found in Pakistan. It is therefore vital that provinces may have access to some buoyant sources of revenue to finance adequately their fiscal needs. Alternatively provinces may be also allowed piggybacking on personal income tax, wealth tax or single stage sales tax as practiced in number of countries. The unconditional transfers however may be continued for meeting fixed cost of running provincial governments without compromising on the provincial priorities.

    The analysis will facilitate development practitioners, policy makers and planners in designing appropriate criteria and allocation strategies for future fiscal framework of federal transfers.

    Date of AwardSept 2013
    Original languageEnglish
    SupervisorShabbar Jaffry (Supervisor) & Guy Judge (Supervisor)

    Cite this