Inclusion and ensuring equitable development are important themes in the emerging post-2015 Sustainable Development Agenda. Inclusion and equity are not only important features in the own right: poverty reduction and development progress can only be accomplished in a meaningful and sustainable manner if government efforts are able to reach the poor where they live. This often includes places that are rural, remote, hard-to-reach or otherwise disadvantaged.
To some extent, the public sector’s efforts to provide public service to people in rural and hard-to-reach places can be measured by the flow of government funds to these places. Thus, inequities in the availability of fiscal resources at the local level would raise a concern to the degree that resources fail to flow to places that need public services and development the most.
As such, a recent study by the Overseas Development Institute (ODI) analyzed the progress, achievements and challenges for addressing inequalities of (recurrent) grant allocations across Local Government Authorities (LGAs) in Tanzania. The study sought to provide guidance on how the declared Government policy of more equitably allocating LGA staff and funds, for the purpose of achieving more equitable service delivery, can be supported. The assignment included a desk review of available background documents, extensive analysis of fiscal and human resource data, as well as fieldwork in 11 LGAs, including some that are significantly underfunded compared to the national average as well as others that are significantly overfunded.
To understand the disparities in allocations that were uncovered involves an examination of both inequity and inequality. Inequality refers to the variation in the distribution of resources from one LGA or service delivery unit to another and in this study we will use variations in per capita allocations as our measure of inequality, while the term inequity refers to the unfairness of the distribution of resources, and correspondingly in this study we take the previously agreed formula as a proxy for a fair and equitable allocation of resources. The term disparity is used as an overarching term to refer to both inequities and inequalities in the distribution of transfers. In addition, the concept of ‘Hard To Reach and Stay (HTRS)’ defines areas where over time it has been difficult to ensure that staff actually report on duty and continue in their post for various reasons, including ‘remoteness’, and lack of access to electricity, water and social services.
The patterns of fiscal inequity in Tanzania
There are three main types of financial transfer for service delivery in LGAs: recurrent block grants—composed of specific allocations for personnel expenditures (personal emolument, or PE) and others for other charges (OC)—, subventions (including basket funds) and capital development grants. While subventions and donor basket funds form part of the ‘development budget’, in practice these funding streams often fund expenditures that are recurrent in nature. Capital development grants fund LGA infrastructure and include the discretionary Local Government Development Grant (LGDG) and sector development grants.
The Government of Tanzania started in 2004/05 to introduce a system of formula-based allocations in order to make the budget allocations more transparent and needs based. A formula was developed for each of the sector block grants as well as for the development grants under the LGDG system. It was not possible to include PE allocations as part of the formula-based system because of the continued centralization of the LGA staff allocation system. The formula has in practice only been applied to the development grant system under LGDG and to parts of the OC allocations.
The patterns of inequality and inequity have remained very significant over the 6 year period analysed. In other words there has been no substantial improvement in terms of developing a more equitable or ‘fair’ distribution of fiscal resources to LGAs. Some districts (such as Kibaha District Council (DC)) receive more than four times the allocation of relatively underserved LGAs (such as Sumbawanga DC). It is clear that these different allocation patterns are unrelated to objective measures of relative need based on poverty, disease burden, land area, etc. This is verified by the analysis of inequity.
Overall, LGA financing is characterized by an increasing wage share driven by the education and health sectors, and reduced operational funding. Education and health, being the largest sectors, drive patterns of inequity. The patterns of inequity for these main sectors are similar, in that the LGAs that are relatively under or overfunded in one sector tend to also be under or overfunded in other sectors.
In terms of inter-LGA fiscal inequity, there exists substantial fiscal inequity between LGAs that is largely driven by PE allocations. LGA revenues are dominated by salary payments (PE) and they constitute an increasing share of total transfers to LGAs. LGA revenues, excluding user fees, were on average TZS 70,000 per capita in 2012/13, 70% of which were recurrent transfers mainly earmarked to sectors and dominated by salary payments. These salary payments represented 78% of recurrent transfers and 55% of all LGA revenues, with other recurrent transfers amounting to only 22% of recurrent transfers and 15% of all LGA revenues. This pattern has become more prominent over time as salary allocations, driven by the introduction of recurrent transfers to secondary education, have continued to increase while other recurrent allocations have declined. By 2013/14 budgeted transfer funding for the running costs of health, water, and primary and secondary education was 40% lower than four years previously, at only TZS 10,700 (about US$6) per capita.
The remaining revenues for LGAs comprise development transfers (21% of LGA revenues in 2012/13) and own source revenue (only 8% in 2012/13). Budgetary outturns for transfers and own source revenue were similar over the review period, averaging 81%, with some deterioration in recent years.
In terms of intra-LGA fiscal inequity, there exists substantial fiscal inequity within LGAs that is often greater than the inequity between LGAs. These patterns of resource inequities are particularly marked for primary education but are also substantive in other sectors. Notably, the patterns of inequities within LGAs are most significant for those LGAs that are categorized as HTRS or underfunded. For example, within Sumbawanga District Council (DC) the number of primary pupils per teacher (PTR) ranges from 19 to 298. For secondary education, the greatest variation arises in science teacher allocations, which vary from 76 students per teacher to 529 students per teacher. The variation of staffing in health facilities is also very significant, with the number of staff per dispensary ranging from 9 to 1 within the same district. The distribution of agricultural extension staff is also very uneven within all LGAs, with a variation of average staff per village at a factor of ten.
Capitation grants for primary and secondary schools are the only major transfers made to service delivery units, with health facilities receiving no transfers. Secondary school fees are a far more important source of revenue than capitation transfers, and user fees are an important revenue stream for health services, making up as much as one-third of recurrent financing available at the LGA level.
The drivers of fiscal inequity
The drivers of inter-LGA fiscal inequity are primarily the sector block transfers, which, in turn, are driven by patterns of staff allocations and their salary payments. The current system of staff allocations has only addressed the inequities to a very limited extent. This is because (a) staff are not retained in targeted LGAs and (b) the current system provides no incentive for deployment of staff to the most needed areas within HTRS LGAs. Employees are, however, amenable to relocation if they have access to basic facilities and allowances amongst other incentives.
Fiscal inequities have a substantial impact on service delivery as a better allocation and use of staff has the potential for substantial improvements in service delivery. However, a continued increase in staff numbers without due attention to details of locational deployment and staff motivation may only have very limited impact on service delivery. Current staff management practices and the resulting disparities in resource allocation reduce the efficiency of service delivery both across and within LGAs.
The main drivers of intra-LGA fiscal inequity are staff preferences for working and living in areas in which they are most comfortable, in the absence of incentives for them to remain in the less comfortable areas where they may have been allocated. In other words, rather than staff being deployed to areas where the service need is greatest, they end up in areas where they feel most comfortable: near relatively developed urban centres, near roads and communication networks, near water and electricity and social facilities – and preferably near their spouse. In theory, LGAs have the discretion to recruit staff independently, but due to acute shortages of staff, sector ministries have since the mid-2000s in practice allocated education, health and recently agriculture staff directly to LGAs. Government has since 2008 made special efforts to target new recruits to the LGAs that are most understaffed. This targeting has stopped the direct deployment of primary school teachers for LGAs that were already over served. The health sector has also made efforts to target underserved areas, but has not adopted a similar strict approach as in primary education: central allocation of health workers to LGAs that are already staffed above average (such as Kibaha DC) has continued because they still are understaffed compared to national norms.
Once staff are allocated to a specific LGA it is the responsibility of the respective LGA to allocate the staff to facilities (schools, health facilities and ward/villages) within the LGAs and to ensure that staff are retained and motivated to work. Not all staff designated to respective LGAs actually report for duty as they perceive the LGA they have been appointed to as ‘unattractive’. In general it appears as if the percentage of staff that are allocated to an LGA and actually report on duty has been increasing, however it is clear that some LGAs still face challenges in attracting staff. Of those who do report, there are also some who either leave within a few months, or after a period of maybe three to five years request for a transfer because of medical reasons. 26% of staff in the survey had arranged for a transfer to a posting of their preference. In general, all the HTRS LGAs see a transfer out of their LGAs while the already well-staffed LGAs see a net increase of their staff through these transfer arrangements.
In addition to the problems of effective deployment to, and retention of staff in HTRS LGAs, there is also the added problem of distributing staff resources in accordance with needs within each of the LGAs. This is in particular a problem in HTRS LGAs.
Return on additional resources to the least funded LGAs is higher than additional funding for relatively well funded LGAs. In the education sector there are substantial variations in district efficiencies (measured as public expenditures per Primary School Leaving Examination (PSLE) pass). One of the key findings from the 2010 Public Expenditure Review (PER)1 study was that ‘there is strong evidence that shifting incremental resources to the worst served areas is likely to improve efficiency rather than reduce it’. The same study concluded that geographical re-allocations could lead to TZS 240 billion in efficiency savings. Similar findings were reported in the PER report on the health sector. In the current study the team found similar patterns across the years of correlation between staff inputs and incremental service delivery impact across LGAs.
Within LGAs some areas are significantly understaffed compared to others and these deliver fewer services (measured in pass rates etc.), but at the same time the workload of staff in the same facilities would very often be higher (measured in patients attended by each health worker or measured by PTR in schools). However, staff productivity was often very low in some of the better staffed facilities. Previous studies on primary school teachers have concluded that Tanzanian teachers on average spend less time teaching than teachers in other sub-Saharan countries. During fieldwork, the team found that many secondary schools were overwhelmingly staffed with arts teachers, who were de facto only assigned five hours of classroom teaching per week.
Only a few LGAs have developed any form of comprehensive scheme for staff retention and motivation. The most often quoted reason is the lack of funding. The most comprehensive schemes identified during fieldwork are being implemented in Kigoma DC (a focus on the health sector) and in Rukwa Regional Council (RC) (a focus on health and secondary education). Both schemes have been relatively successful in improving the attraction and retention of staff through a combination of various fiscal and non-fiscal incentives. The scheme in Kigoma proved to be the most sustainable as the management of the scheme was less complex. In addition, the Kigoma scheme effectively targeted the most HTRS areas within LGAs, just as it also attempted to document impact on service delivery to a greater extent than the Rukwa scheme. In summary, local incentive schemes can impact positively on the rational deployment of staff, staff motivation and service delivery, but few LGAs have developed any comprehensive strategies and none have the financial resources to target a large part of their workforce.
Conclusions and recommendations
The study found very uneven allocation patterns of fiscal resources across LGAs that are primarily driven by unequal allocations of salary expenditure and result in the inequitable allocation of staff. While there has been some progress in improvement in staff allocations in HTRS LGAs, patterns of inequity have persisted across LGAs, with some LGAs still being overstaffed relative to others.
There are also significant inequities and inequalities in resource allocations within LGAs, in particular the HTRS LGAs. The trend of declining OC allocations in LGAs limits and reduces the effective use of existing staff resources. Furthermore, low utilization levels of some staff categories and inefficiencies in the use of staff resources are compounded by the declining rate of return to increase staffing in relatively well staffed areas, compared to investment in staff in HTRS areas. This is an important conclusion as the Government strives to achieve Big Results Now (BRN) within a constrained fiscal framework. A better distribution of resources, in particular staff, could lead to substantive service delivery improvements.
This report recommends three mutually supporting main strategies to address the fiscal inequities across LGAs:
Strategy 1: Enabling and supporting local level initiatives for retention and deployment to address intra-LGA inequity. This aims to facilitate and incentivize local actions to improve distribution, retention and motivation in HTRS areas. This strategy would be initially focused on disadvantaged LGAs and then could be rolled out more widely. Under the leadership of PMO-RALG it would comprise: (i) the development of local strategies for improvement of staff distribution and motivation; (ii) the introduction of a special grant to understaffed LGAs to support the implementation of their strategies, with a recurrent and development component. Access to the special grant would be provided to only those LGAs that fulfil basic pre-conditions, such as the existence of a strategy that meets certain standards.
Strategy 2: Streamlining and prioritizing human resource allocation and deployment for addressing inter-LGA inequity. This strategy proposes that Central Government introduces a preferential allocation of staff to the most underserved areas and increases real OC allocations to service delivery. This requires adjusting the existing staff allocation system such that the HTRS LGAs should be prioritized for the allocation of new staff to all sectors in relation to their relative degree of understaffing; and there is some reallocation of staff away from the most well-staffed areas. Combined with this adjustment to the staffing allocation mechanism, the level of operational funding for service delivery should be increased in parallel through strict adherence to an agreed equitable and needs-based formula to address disparities.
Strategy 3: Enhancing operational funding and fiscal transparency around resource allocations and utilization. In addition to these two main strategies, it is recommended to support initiatives in support of greater transparency of LGA resource allocations and utilization at the central and local levels. This involves better data on LGA finances and human resources as well as greater transparency through improved publication of the data. In particular, data on staff deployment and existing staff established in all LGAs should be made publicly available. A more transparent system would allow LGAs to advocate for their fair share of staff and would allow policy analysts and the public to track commitments to a fairer, more equitable and effective deployment of staff. Transparency of staff allocation is almost entirely absent but could be greatly enhanced through the use of the newly developed tool for monitoring staff and the payroll as the basis for public reporting covering: existing staff in LGAs; annual deployment of new staff to LGAs by sector; and data on transfers across LGAs.
Read the full report from the ODI website:
Per Tidemand (team leader), Nazar Sola, Alloyce Maziku, Tim Williamson, Julia Tobias, Cathal Long and Helen Tilley. 2014. Local Government Authority (LGA) fiscal inequities and the challenges of ‘disadvantaged’ LGAs. Overseas Development Institute.