# Introduction urrent scholarship, theory, and practice concur that the basic goal of decentralization is for central government to extend part of its political, administrative and fiscal management powers and mandate to lower levels of government. However, power and mandate delegation to sub-national entities is required to be accompanied by an equally sizeable level of operational autonomy and leverage. Autonomy is pertinent if decentralization basic goals of effective resource allocation, efficiency and economic growth have to be realized (Fessha & Kirkby, 2008). Sub-national and specifically local government fiscal autonomy; one of the most pivotal aspects of resource allocation, has been a focus of fiscal federalism research for quite a long time. Most studies (e.g. Akindele, Olaopa & Obiyan, 2002;Fessha & Kirkby, 2008;Oulasvirta & Turala, 2009) claim that majority entities in both the developed and developing world are rarely fiscally-autonomous in practical and real terms. For instance, Fessha and Kirkby (2008) observe that in Sub-Saharan Africa, central governments simply dictate local revenue mobilization, grant management, and expenditure terms. Unfortunately, none of the current studies nor theory and practice conclusively explains what precisely leads to fiscal autonomy gaps in local entities. Nevertheless, findings from some studies (e.g. Bird & Smart, 2002;Liu, 2014;Rao & Das Gupta, 1995) suggest that fiscal horizontal transfers management is the most dominant and problematic phenomenon to autonomy. A number of fiscal horizontal transfer attributes; notably, equalization, accountability, and risk management are said to predict fiscal autonomy dynamics (Bird & Smart, 2002;Liu, 2014). Moreover, other scholars (Lameck, 2009;Hood & Mcgarvey, 2002) associate local entity fiscal autonomy status in both developed and developing countries to public-private partnerships. Guided by the pure theory of local expenditure (Tiebout, 1956) and fiscal federalism theory (Musgrave & Musgrave, 1973;Oates, 2005), the current study examines fiscal horizontal transfers-fiscal autonomy linkages in 27 districts, 9 municipalities and 243 subcounties in Uganda's eastern region. It also investigates whether public-private-partnership actually mediates the relationship between fiscal horizontal transfers-fiscal autonomy in those entities. The East African country operates on a political-administrative machinery of seven regions; namely, south-western, central, northwestern, northern, north-eastern, and eastern regions. Over the years, Uganda has been commended for its fiscal federalism proficiency (Akindele et al., 2002;Albouy, 2012;Shankar & Shah, 2003). Whereas there is accumulated prior research on local entity fiscal autonomy in Western, South American, and South-East Asian countries (e.g. Akindele et al., 2002;Awortwi, 2011;Shankar & Shah,2003), few studies address the following related and salient concerns in regard to Sub-Saharan Africa and particularly Ugandanbased entities: 1. What specific roles do individual fiscal horizontal transfers components; namely, equalization, accountability, and risk management play in entity fiscal autonomy? 2. How do public-private-partnerships matter in the fiscal autonomy structure? 3. Do empirical results on fiscal horizontal transferspublic private partnerships-fiscal autonomy connectivity hold in the Sub-Saharan Africa region and particularly Uganda? In responding to the foregoing concerns, this study makes a number of contributions to conventional fiscal federalism literature and specifically that of fiscal autonomy. First, we examine the otherwise and often empirically-overlooked influence of fiscal horizontal transfers on local fiscal autonomy. The attention paid to the three fiscal horizontal transfers constructs (equalization, accountability, and risk management) enriches its research, theory and practice in a special way given that previous studies (e.g. Bird & Smart, 2002;Oulasvirta & Turala, 2009;Shankar & Shah,2003) largely focused on fiscal vertical transfers as an antecedent of fiscal autonomy. Second, the study identifies public-privatepartnership as a potential mediator to fiscal horizontal transfers-fiscal autonomy linkages. This provides a general yet novel input in fiscal autonomy analysis given that in the Sub-Saharan African setting (Akai & Sato, 2008;Fessha & Kirkby,2008)majority local entities derive enormous service support from the private sector. Third, by contextualizing this investigation to Sub-Saharan Africa and Uganda, the present research enhances empirical comparison and replication of investigation in other parts of the world. As variously noted in related studies (e.g. Awortwi, 2011;Oulasvirta & Turala, 2009;Rao & Das Gupta, 1995), most decentralizing countries in the developing world face almost similar fiscal autonomy challenges. However, majority of them tend to underscore fiscal vertical transfers activities as such transfers dominate the budgetary machinery (Oulasvirta & Turala, 2009). # II. # Theoretical Background and Hypotheses a) Fiscal Autonomy Whether in developed or developing countries, fiscal autonomy granted to local government jurisdictions is often perceived the most salient indicator of effective fiscal decentralization. Fiscal federalism theory (Musgrave & Musgrave, 1973;Oates, 2005) proposes that effective fiscal decentralization must ensure that local entities are free to impose their own taxes. Moreover, the entities should independently manage fiscal transfers from central government, and allocate any other funding with minimal external interference (Oates, 2005). Consistent with the Tiebout (1956) pure theory of local expenditure model, subnational units and especially local governments will then spend strategically and avoid resource waste. That level of fiscal leverage does not only enable the entities to competently set and execute their budgets but it also enhances accountability, transparency, and quality service delivery. Conventional research (e.g. Akindele et al., 2002;Oulasvirta & Turala, 2009) provides evidence that fiscal autonomy leads to budgetary efficiency and ultimately economic growth. Given the numerous tasks (mandatory and permissible) local entities are constitutionally required to execute; the fiscal autonomy concept can best be appreciated from the income-source and expenditure-mandate perspectives (Fessha & Kirkby, 2008;Oulasvirta & Turala, 2009). Traditional public finance literature (e.g. Musgrave & Musgrave, 1973;Oates, 2005;Tiebout, 1956) holds that central government is responsible for two major fiscal responsibilities: macroeconomic stabilization and income re-distribution. As noted by Musgrave and Musgrave (1973), the two assignments are of nation-wide nature and entail massive resource outlays to accomplish. Thus, local entities are excluded from handling the obligation (Oates, 2005). Besides, the pure theory of local expenditure model advanced by Tiebout (1956) posits that naturally, local entities operate limited economies of scale and their benefits spill-over to other jurisdictions is equally minimal. Thus, such units stop at merely providing goods and services to their communities (Awortwi, 2011;Oulasvirta & Turala, 2009;Tiebout, 1956). In various developing economies such as those of Africa, fiscal federalism policies are seemingly founded on the foregoing theoretical underpinnings (Albouy, 2012;Hood & Mcgarvey, 2002;Oulasvirta & Turala, 2009;Shankar & Shah, 2003). For instance, in Ethiopia, Ghana, Nigeria, South Africa, and Uganda; to cite but a few, local governments are denied total discretion to levy certain taxes. Hence, the entities rarely levy import duty on international trade or inter-entity trade transactions, employee pay-as-you-earn tax, and corporation tax (Hood & Mcgarvey, 2002;Shankar & Shah, 2003). Similar restrictions abound in relation to expenditure especially from conditional grants periodically remitted from central government. Accordingly, some scholarly (e.g. Akindele et al., 2002;Fessha & Kirkby, 2008) argue that given that open-ended policy of local fiscal autonomy is often restrictive, it is of little doubt that such economies frequently experience administrative complexities, inequalities, and persistent resource allocation distortions. In Uganda, government has resorted to creating more districts and other local governments with a view of addressing the inequality and resource allocation distortions. But consistent with empirical evidence (Albouy, 2012;Awortwi, 2011;Shankar & Shah, 2003), unnecessarily large numbers of districts have instead bred budgetary incapacitations and fiscal horizontal transfers problems. The country's fiscal decentralization also pays little attention to the socio-economic development and political maturity realities of its largely rural communities. Shankar and Shah (2003) noted that these factors are crucial for local fiscal autonomy in that they guarantee maximum benefit amidst rampant rentseeking, corruption, and fraud practices. # b) Fiscal Horizontal Transfers Also referred to fiscal equalization transfers, fiscal horizontal transfers are grants extended to subnational entities particularly local governments to achieve inter-entity fiscal efficiency and equity (Liu, 2014;Rao & Das Gupta, 1995). Both political and socioeconomic policies advocate for these transfers on the grounds that they assist recipient entities to provide public goods and services at comparable tax rates without being compelled to do so (Albouy, 2012;Liu, 2014). Accordingly, proponent literature (e.g. Akai & Sato, 2008;Bird & Smart, 2002;Shankar & Shah, 2003) claims that in resource-constrained localities such as those of Africa, equalization transfers do not only enhance entity preference diversity but also promote pursuit of preferred policies. However, Bird and Smart (2002) remarked that resultant fiscal ability is only feasible in entity fiscal capacity-fiscal needs equilibrium environments. Without fiscal capacity-fiscal needs symmetry, majority entities take advantage of outright lack of local entity fiscal autonomy, to operate inefficiently (Bird & Smart, 2002;Shankar & Shah, 2003). From a theoretical angle, both the fiscal federalism theory ( (Musgrave & Musgrave, 1973;Oates, 2005), and the Tiebout (1956) pure theory of local expenditure model, emphasize the relevance of always weighing fiscal horizontal benefits against associated entity-efficiency costs. In particular, Oates (2005) notes that central government policy intended to institute equalization grants must not overlook the fact that by nature, entities are resource-endowed differently. Thus, equalization support meant to correct fiscal inequalities in such diversity may not only spark-off inter-regional resource allocation inefficiency but compromises fiscal autonomy. Relatively, such experience does not only dominate Sub-Saharan African local entities (Rao & Das Gupta, 1995) but has also hit developed entities in German recently (Akai & Sato, 2008;Albouy, 2012;Shankar & Shah, 2003).Conventional research on local government fiscal autonomy generally associates fiscal horizontal transfers with three predictive attributes: equalization (Akai & Sato, 2008;Shankar & Shah, 2003), accountability (Bird & Smart, 2002;Rao & Das Gupta, 1995), and risk management (Albouy, 2012;Liu, 2014). # i. Equalization According to public finance norms (Bird & Smart, 2002;Rao & Das Gupta, 1995), the primary objective of fiscal federalism policy in majority decentralized jurisdictions is to achieve vertical fiscal balance (VFB). Moreover, the dream is that VFB be realized on a sustainable basis. Much as Bird and Smart (2002) consider vertical fiscal balance a largely illusionary notion, in practice it amounts to that fiscal state when revenue and expenditure at all levels of government are consistently equal to one another. Thus, any divergence from the revenue-expenditure equilibrium amounts to some fiscal gap. Given their capacity to generate their own revenue, more recent empirical evidence (e.g. Akai & Sato, 2008;Albouy, 2012;Liu, 2014), indicates that rich and well resourceendowed localities are often able to narrow down their fiscal gaps. On the contrary, fiscal gaps in poor and unfortunate less resource-endowed entities are rarely filled. Thus, they commonly constitute a very complex operational situation (Akai & Sato, 2008). Often accused of significantly contributing to the current fiscal autonomy complications in majority Sub-Saharan Africa localities (Albouy, 2012;Liu, 2014;Shankar & Shah, 2003), it is argued that fiscal gaps can only be appreciated from a horizontal fiscal balance context. In Africa, as is common elsewhere, horizontal fiscal balance is not easy to achieve. Also simply referred to as equalization (Akai & Sato, 2008, Shankar & Shah, 2003), horizontal fiscal balance is generally believed to be a multi-interpretational dimension. Thus, various countries hold differing preferences as far as its attainment is concerned. It is this multitude of often conflicting preferences, that Bird and Smart (2002) attribute to exacerbated local entity fiscal autonomy malaise in the region. For instance, equalization largely ignores actual entity revenue mobilization and expenditure capacity realities. Several countries tend to release more transfers to low tax-generating but high expenditureprone localities considering them naturally disadvantaged. Ultimately, this approach does not only discourage tax-raising efforts but also seriously promotes over-spending and rent-seeking practices (Albouy, 2012;Bird & Smart, 2002). Accordingly, the universal empirical stand (see Rao & Das Gupta, 1995;Shankar & Shah, 2003) is that: inter-entity equalization would be rational if central governments made adequate transfers to all localities. Most resource-constrained economies of Africa, Uganda inclusive, can rarely afford to make such transfers on a regular and sustainable basis. budgets but has further tightened grant conditionalities. Supplementary budgets are a reflection of transfers inadequacy while stiff conditionalities portray constrained fiscal autonomy (Akai & Sato, 2008;Bird & Smart, 2002;Liu, 2014). Based on the foregoing equalization-fiscal autonomy deliberations, the following is proposed: Hypothesis 1: In local government, equalization relates positively with fiscal autonomy. # ii. Accountability Both the pure theory of local expenditure (Tiebout, 1956) and fiscal federalism theory (Musgrave & Musgrave, 1973;Oates, 2005) stress the need for accountability in sub-national entities. According to Musgrave and Musgrave (1973), citizens should always be able to hold local entity authorities responsible for the way fiscal resources (local tax revenue, transfers, and donor funding) are expended. Thus, the grantor (central government) has to design and tailor fiscal transfers to local operational circumstances in a manner that enhances accountability efficiency. Accordingly, recipient entities must also be accountable to both the grantor and citizens for resource-utilization integrity and output. Expenditure output or results must then be evidenced by prompt and high quality service delivery (Musgrave & Musgrave, 1973;Oates, 2005). Fiscal autonomy proponents including (Akindele et al., 2002;Fessha & Kirkby, 2008;Oulasvirta & Turala, 2009) assert that theory-assumed accountability can never be realized in practice. This is more so in corruption-ridden and fiscal autonomy-deprived countries of the developing world. For instance, Fessha and Kirkby (2008) identified two critical factors that foster pessimism of fully realizing accountability efficiency in such countries. First, majority of their entities lack technical capacity. Local governments in various decentralized Sub-Sahara Africa countries; with no exception Uganda, largely rely on unqualified manpower. Often sourced on political, sectarian and tribal basis, the work-force can rarely deliver to required standards and cannot be disciplined for habits such as corruption. The resultant inefficiency has been aggravated by extremely weak or non-existant internal control systems operated by most entities (Fessha & Kirkby, 2008;Shankar & Shah, 2003). Second, various rural-based local governments operate manual accounting and budgetary systems. Much as such entities are the majority recipients of horizontal fiscal transfers, their manual systems inhibit operational performance and particularly compromise accountability efficiency (Fessha & Kirkby, 2008;Oulasvirta & Turala, 2009). On the basis of the foregoing empirical dialogue, the current study makes another novel contribution to the local fiscal autonomy literature. Specifically, it attempts to unlock the accountabilityfiscal autonomy relationship dilemma. For example by implication, some existant evidence (Akindele et (Bird & Smart, 2002). To beneficiary local governments, equalization transfers create unmanageable risk of high level accountability and efficient resource management. As noted Albouy (2012) and Liu (2014), the transfers do not only compound the internal control shortfalls in majority African localities, but also breed inter-entity disunity. Beneficiary entities face a big risk of operational discrimination and heightened competition. Ouite often, it is such a situation that compels central governments to restrain local fiscal autonomy (Albouy, 2012;Shankar & Shah, 2003). Finally, local communities and indeed the entire citizenry cannot easily escape the risk managementfiscal autonomy challenge. Citizens face a big risk of resultant local tax uncertainties, service delivery quality, and production inefficiency. In a nutshell, the communities risk the loss of obtaining what Rao and Das Gupta (1995) termed agglomeration or collective benefits. Against the risk management-fiscal autonomy debate above, it is proposed: # c) Public-Private Partnerships In the developing world, governments at both national and sub-national level are often faced with the challenge of extraordinary demand for public goods and services particularly related to social infrastructure. Thus, despite the tight fiscal and budgetary constraint environments they operate in, authorities are obliged to set-up educational, health, and water and sanitation facilities, and construct road networks (Awortwi, 2011;Oulasvirta & Turala, 2009;Shankar & Shah, 2003). Recent empirical evidence (e.g. Hood & Mcgarvey, 2002;Lameck, 2009) suggests that various sub-national and particularly local jurisdictions in Africa apply public-private partnerships (PPP) to address social infrastructure concerns. According to Hood and Mcgarvey (2002), PPPs are normally public entity-private organization contracts entered into to help in the construction and maintenance of public entity infrastructure on a value-for-money basis. The contracts; usually with a fixed lifespan, combine a fixed price or no payments-in-progress component with that of a long-term service value (Hood & Mcgarvey, 2002).However, depending on the contract terms, either the local entity owns the infrastructure outright or the contractor holds it until the contract is completed. Related service installment-payments are then made after construction (Lameck, 2009). Advocates of the public-private partnerships approach (Akindele et al., 2002;Hood & Mcgarvey, 2002) argue that PPPs constitute the most effective procedure to public-private sector cooperation. Relative to conventional procurement (design-and-construct) contracts, the partnerships create enormous local fiscal efficiency gains. Moreover, when properly designed and their terms of reference adhered to, the partnerships are also able to generate government-service provider arms-long relationships. Traditionally, arms-long relationships are considered important for their capacity to enhance effectiveness in contract enforcement (Akindele et al., 2002;Lameck, 2009). Research and practice also view well-structured public-private partnerships as the most ideal option for social infrastructure development for local entities with limited skills and technology. According to Hood and Mcgarvey (2002), today several African countries endeavor to promote PPPs as the only viable means of filling their local government manpower capacity gaps. In reality, however, restrictions in fiscal autonomy and associated public performance limitations render anticipated public-private partnership benefits unrealizable. # Methods a) Participants and Procedures In order to test the hypothesized model of the current study, 295questionnaires were distributed to 27 districts, 9 municipalities and 243 sub-counties (units of analysis) located in the eastern region of Uganda, East Africa. Two types of questionnaires were used, one for administrators and the other for employees and councilors. Bettis, Gambardella, Helfat and Mitchell (2014) recommend collection of survey data from a multitude of sources approach as it enhances effective predictor-mediator-criterion variable evaluation. Moreover, the separation significantly minimizes effects of potential same data-source bias (Bettis et al., 2014;Edwards & Lambert, 2007). The administrators: Chief administrative officers, resident district commissioners, local council 5 chairpersons and heads of department, were purposively selected. Employees and councillors were selected on a purerandom basis (Edwards & Lambert, 2007).From the 295 questionnaires distributed to the participants,only 257 of them were received back (response rate: 87%). Furthermore, another 2 instruments were eliminated due to un-matched responses and related problems. Thus, only 252 questionnaires were left for use in the final analysis (final response rate: 85%). Among the respondents, 56% were male with the mean age of 35 (SD = 7.69) and largely of married category. 34% of them were diploma holders and 43% held a bachelor's or higher degree qualifications. On average, 48% of the participants had worked in their respective entities for 6.47 years (SD = 2.14). # b) Measures All the study variables and constructs were assessed using established scales and subsequently measured on a five-point Likert type scale ranging from 1 = Strongly Disagree to 5 = Strongly Agree. i. Income Source In order to assess income source, the author employed three multiple sub-scales measures from the works of Akindele et al. (2002) and Fessha and Kirkby (2008). The comprehensive scale Cronbach's alpha was (0.93). Sample items included: "?income source is very reliable." and "?the source enhances entity fiscal flexibility." ii. Expenditure Mandate Expenditure mandate was assessed using a modified seven-item scale developed by Oulasvirta and Turala (2009). Sample items: "This local government has full authority to spend on relevant community-based areas and projects" and "All expenditure is often transparently accounted for." The corresponding Cronbach's alpha coefficient was (0.89). # iii. Equalization Employing scales developed by Bird and Smart (2002) but systematically modified, the eight items related to equalization used in the study exhibited (? = 0.91). Some of the items included the following: "Grants meant to achieve equalization in this entity have unnecessarily stringent conditionalities" plus "?equalization undermines entity competitiveness instead." # iv. Accountability Eight items (? = 0.96) were adopted for the scale that measured accountability as a construct of fiscal horizontal transfers in this research. The scale was derived from previous empirical work (Akai & Sato, 2008; Rao & Das Gupta, 1995) but adjusted accordingly. Typical item: "The approach used by this entity to achieve fiscal accountability requirements is very effective." # v. Risk Management The construct risk management was analyzed on the basis of measurement scales employed by Liu (2014) and Shankar and Shah (2003). The seven items (? = 0.97) used in the assessment included: "?fiscal autonomy cannot be attained in this locality due selfcreated risks." # vi. Public Private Partnership Measurement scales rooted in the work of Lameck (2009) and that of Hood and Mcgarvey (2002) but tailored to suit local circumstances of the current study, were used to evaluate the variable public private partnership. Related seven items embraced, including: "Public private partnership has compromised fiscal autonomy achievement drive in this local government" had a Cronbach's alpha coefficient of (0.85). # c) Control Variables Five demographic variables of participants' gender, age, marital status, education, and period served were included in the study model as control variables. Previous studies (e.g. Asparouhov & Muthen, 2009;Edwards & Lambert, 2007) have shown that demographic variables are closely related to local fiscal autonomy and thus have a potential influence on its undertakings. Accordingly, gender was measured as a dichotomous variable coded as (0) for male [141] and (1) for female [111] Furthermore, one latent variable was espoused to support analysis of the study instrument validity assessment.As recommended by previous research (Asparouhov & Muthen, 2009;Bollen & Stine, 1992), validity can effectively be verified by subjecting such a factor to Harman's one factor confirmatory factor analysis (CFA). The essence of CFA is to depress potential suppressive effect latent variables may have on hypothesis analysis results. The latent variable was therefore controlled for in order to achieve that goal. # d) Analytical Approach In order to test the study's proposed hypotheses effectively, structural equation modelling (SEM) using AMOS (v.10) was performed. Past simulation experience (Asparouhov & Muthen, 2009;Bollen & Stine, 1992;Edwards & Lambert, 2007) suggests that SEM approach enables a concurrent evaluation of all study variables in hypothesized models thus enhancing verification of model-test data consistence. This can certainly be attained if the hypothesized model is subjected to a bi-analytical (measurement model and structural model) strategy (Bollen & Stine, 1992;Edwards & Lambert, 2007). In the present study, the measurement model was the first to be examined. The verification was executed by means of CFA and as indicated by Edwards and Lambert (2007), no control variables was involved. Secondly, a structural model was instituted to establish the hypothesized model fit as per the measurement model output. Ideally, the structural model is meant to foster hypothesized mediation estimation (Asparouhov & Muthen, 2009;Bollen & Stine, 1992). Besides, error variances of latent constructs related all control variables were fixed at zero (0) value. Asparouhov and Muthen (2009) noted that fixing error variances at zero facilitates manifestation of only one item for each control variable and enhances control variable latent constructs to be loaded on mediation and dependent variables in designated paths more easily. # e) Results # i. Descriptive Statistics and Correlations Table 1 presents the descriptive statistics, reliability and inter-variable correlation coefficients of the study. While a number of relationships were in the expected direction, several of them were not. Notably, public-private partnership relates positively and significantly with risk management to the extent of (r = .28, p < 0.01) and also with accountability (r = .17, p<0.05) but negatively albeit significantly with equalization (r = -.33, p < 0.05). Moreover, while fiscal autonomy has a positive and significant association with public-private partnership (r = .16, p < 0.01), it relates negatively with accountability and equalization at levels (r = -.27, p < 0.05) and (r = -.35, p < 0.01) respectively. At variable level, fiscal autonomy relates positively and significantly with fiscal horizontal transfers to the magnitude of (r = .42, p < 0.05) and as indicated earlier, with publicprivate partnership (r = .16, p < 0.01). The results also show that all variable and construct items meet the reliability threshold (? ? .75) (Bettis et al., 2014;Hood & Mcgarvey, 2002). # f) Hypothesis Testing Before testing the proposed study hypotheses, presence of multi-collinearity threat to the data set was established. Past research (e.g. Asparouhov & Muthen, 2009;Bollen & Stine, 1992) notes that multi-collinearity can significantly compromise hypotheses results if not properly addressed. Data tolerance values (TV) and variable inflation factors (VIF) at (TV? 1.0) and (VIF? 10.0) levels respectively imply absence of the threat to data (Bollen & Stine, 1992). As shown in Table 2, the various TV and VIF results suggest that the study data set was safe from the multi-collinearity problem. Moreover, the results were confirmed by poor Harman's single factor model goodness-of-fit indices(?2 =9.382; df = 14; ?2/df = 0.670; IFI = 0.591; TLI = 0.763; CFI = 0.887; RMSEA = 0.218; L.052, H.236) relative to those in the measurement model(?2 =14.845; 0.983; RMSEA = 0.017; L.019, H.114) (Bettis et al., 2014;Bollen& Stine, 1992). The weak Harman's single factor model goodness-of-fit indices further suggest that data were safe from the common methods variance threat and had quite a robust discriminate and construct validity set-up (Edwards & Lambert, 2007). The study hypotheses were tested using the regression analysis-structural equation modeling combined approach consistent with previous research ( # i. Direct Effect The indices for the direct effect model were as follows: (? 2 df )=1.437; df=1; p=.231; (?2/df)=1.437; GFI=.988; NFI=.988; RFI=.884; IFI=.996; TLI=.962; CFI=.996; RMSEA=.043 (L.000; H.186) at 90. In reference to the various hypotheses, it was proposed in Hypothesis 1 that in the surveyed local governments, equalization has a positive relationship with fiscal autonomy. Table 2 results (? = .36, p < 0.01, t-value = 1.69) indicate that data actually offer support to that presumption. The results further suggest that available data support Hypothesis 2 (? = .16, p < 0.05, t-value = 1.83) position that in those entities, accountability activities relate positively with fiscal autonomy endeavors. In addition, Hypothesis 3 which predicted a positive association in the investigated localities between risk management and fiscal autonomy also received data backing. Besides, Adjusted R 2 = .697value affirms that the study predictor constructs explain changes in fiscal autonomy to a quite large magnitude of nearly 70%. In sum, what the foregoing results imply is that: The more the entities endeavor to attain equalization, accountability, and good risk management practices, the stronger the drive to achieve entity fiscal autonomy and independence also becomes. # ii. Indirect (Mediation) Effect The Conventional structural equation modeling (Edwards & Lambert, 2007) method results (Table 2) (? = .39, p < 0.01, t-value = 1.69), indicate that data support that proposal. Numerous studies (e.g. Bird & Smart, 2002;Bollen & Stine, 1992;Lameck, 2009) often cast doubt on regular SEM-derived mediation effects claiming they are not substantive enough. Thus, 2500 sub-samples were created and accordingly tested for mediation based on bootstrapping methodology (Bollen & Stine, 1992). The results (Table 2), suggest that the generated 95% bias-corrected and significant confidence interval (CI)[-0.084â??"-0.001] confirm Hypothesis 4; public-private partnership mediation in the fiscal horizontal transfers-fiscal autonomy relationship. Bird and Smart (2002) consider the indirect effect statistically-significant since the (CI)[-0.084â??"-0.001] contains no zero value within it. Such independence is manifest able in transparent horizontal fiscal transfers' management supported by public-private partnerships. In line with Hypothesis 1, the study results showed that the higher the equalization level, the more fiscal autonomy can be realized in local entities. As indicated earlier, equalization also referred to as horizontal fiscal balance, is registered when central government makes adequate and equal transfers to all its local entities in a particular period. In practice, however, and consistent with previous empirical evidence (Akai & Sato, 2008, Shankar & Shah, 2003), few countries in the developing world attain acceptable fiscal equalization standards on a regular basis. # IV. Discussion and Study Contribution According to Shankar and Shah (2003), relentless resource and technical capacity constraints in Sub-Saharan Africa, often render equalization in most local entities a myth. Resultant horizontal fiscal gaps create revenue imbalances and therefore complicate expenditure especially that related to social infrastructure. It is also further observed by Akai and Sato (2008) that this often unnoticed equalization-related frustration has serious repercussions on local fiscal autonomy. Rather than the commonly believed entity lack of accountability and transparency, it is that frustration that is largely responsible for central government denial of full fiscal autonomy mandate to most entities. The present study therefore suggests that central governments adopt realistic equalization policies capable of promoting local entity fiscal autonomy. This will enhance realization of fiscal federalism advocated for recently by Oates (2005) in the fiscal federalism theory. Results for Hypothesis 2 indicated that there is a direct relationship between accountability and fiscal autonomy. This relationship is expected given prior pure theory of local expenditure (Tiebout, 1956) position and empirical evidence (e.g. Akai & Sato, 2008;Rao & Das Gupta, 1995). For instance, Rao and Gupta (1995) posited that when local entities properly account for all resources granted to them particularly those arising from fiscal horizontal transfers; it tremendously enhances their fiscal autonomy flexibility. In Uganda, accountability in most local entities is derailed by partisan politics, rampant corruption, and rent-seeking behavior (Akai & Sato, 2008;Shankar & Shah, 2003 Furthermore, the model was founded on classical and contemporary theories, namely, pure theory of local expenditure (Tiebout, 1956) and the fiscal federalism theory (Musgrave & Musgrave, 1973;Oates, 2005). Accordingly, the investigation contributes to the pure theory of local expenditure by clarifying how the equalization-accountability-risk management networks in fiscal horizontal transfers are key mechanisms in promoting related fiscal expenditures in local entities. Understandability of the mechanisms may significantly influence fiscal autonomy decisions. Similarly, the model also impacts the conventional fiscal federalism theory (Musgrave & Musgrave, 1973;Oates, 2005). Previously overlooked by the theory, the study illuminates the contribution of fiscal autonomy flexibility in attaining effective fiscal federalism. specifically, Ugandan-based entities are yet to attain that level (Liu, 2014;Shankar & Shah, 2003). Finally, it had also been proposed as Hypothesis 4 that public-private partnerships (PPP) mediate the fiscal horizontal transfers-fiscal autonomy relationship. Coincidentally, that hypothesis was supported by the study data. Previous scholarly work (e.g. Lameck, 2009;Hood & Mcgarvey, 2002) stress the pivotal role PPP frequently plays in developing, construction, and management of major social infrastructures in Sub-Saharan Africa. Most of such projects are funded by conditional and equalization grants but entrusted to third parties in the private sector. Hood and Mcgarvey (2002) claims that local authorities are often sidelined due to questionable accountability and transparency capacities. This is a direct manifestation of constrained fiscal autonomy. V. # Theoretical and Practical Implications The present research advances theoretical and practical understanding of fiscal autonomy in local entities in at least two important ways. First, past theorizing on the topic largely focused on exemplar assumptions surrounding local fiscal autonomy in ideal economies of the developed world. This study emphasizes an in-depth investigation of fiscal autonomy parameters in resource-constrained countries of Sub-Saharan Africa and particularly those of Uganda. Since such economies operate generally weak fiscal policies often undermined by partisan politics, fiscal autonomy analysis on a prototype basis is significantly avoided. As observed by Musgrave and Musgrave (1973) and quite recently Oates (2005), effective fiscal federalism and by implication local fiscal autonomy, must always avoid prototypicality. Second, the study was developed to assess novel aspects of fiscal autonomy in local government particularly fiscal horizontal transfers and public-private partnership. Past studies (e.g. Awortwi, 2011;Bird & Smart, 2002;Fessha & Kirkby, 2008) emphasize local entity-central government balanced interactions in realizing local fiscal efficiency. Thus, local entity leaders are encouraged to place less focus on fiscal autonomy setbacks but rather appreciate the fiscal horizontal transfers-public-private partnership input to fiscal autonomy. Moreover, central government policy formulators need to adopt fiscal policies that are practical to local entity situations in order to realize the fiscal equilibrium (Bird & Smart, 2002;Fessha & Kirkby, 2008). VI. # Limitations and Future Research Direction Despite the abovementioned contributions of this study, some potential limitations should be noted. First, only five fiscal horizontal transfers-fiscal autonomy constructs (equalization, accountability, risk management, income source, and expenditure mandate) were adopted in the research analysis. Much as these five attributes are often considered the most influential (Akindele et al., 2002;Albouy, 2012;Oulasvirta & Turala, 2009;Rao & Das Gupta, 1995), it would be meaningful to investigate other factors that are not particularly main stream. Such factors may include technical capacity, entity location, politics, and tribalism. Second, the current study may have a potential reversal causality risk given that a cross-sectional research design was employed. It is recommended that future investigations employ more time-flexible designs such as longitudinal design to mitigate the reversal causality complications. Third, the study variables and constructs were rated by same sources. This remains a potential threat to the validity of the study results. Upcoming studies are therefore advised to gather data from a range of sources. # VII. # Conclusion The present results supported the notion that in local government, fiscal horizontal transfers impact fiscal autonomy. Moreover, public-private partnerships play a critical mediation role in fiscal horizontal transfers-fiscal autonomy linkages. We hope that this study will not only serve as a benchmark for future research, but is also suggestive of a number of local fiscal autonomy areas for future investigation. horizontal transfers recipient localities (Hood &Mcgarvey, 2002; Shankar & Shah, 2003).The preceding seemingly-endlessfiscalhorizontal transfers-public private partnership-fiscalautonomy debate suggests the following proposal:Hypothesis 4: Public-private partnership mediates therelationship between fiscal horizontal transfers and fiscalautonomy.III.Year 2019budgetary efficiency in less-competitive and fiscal horizontal transfers recipient entities (Shankar & Shah, Second, pre-mature partisan politics (Albouy, 2012; Bird & Smart, 2002) and lack of proper fiscal resource governance (Oulasvirta & Turala, 2009) undermine public-private partnership goals in African local entities. Accordingly, investors and the donor community (funders of most social infrastructure projects), doubt contract authenticity. The scholars especially Albouy (2012) and Oulasvirta and Turala (2009) assert that, in general, it is the politics-governance triangulation that has undermined fiscal equalization in such countries and therefore complicated the fiscal autonomy dynamics. Third, a combination of salient factors such as: weak administrative structures, inadequate project cost-benefit analysis capacity, corrupt contract bidding-awarding systems, and unpredictable regulatory environments, eventually generate contract cost over-runs and recurring re-negotiations. Thus, and in congruence with empirical evidence (Bird & Smart, 2003).Volume XIX Issue VI Version I Global Journal of Management and Business Research ( ) C2002; Hood & Mcgarvey, 2002; Lameck, 2009; Shankar& Shah, 2003), the foregoing factor combinationsurrounding majority local jurisdictions, prompts centralgovernment authorities to restrict local fiscal autonomy.This position is rational in local public entity-privatepartnership arrangements and particularly in fiscal 1#VariableMSD123456781.Equalization3.321.52.842.Accountability3.061.56-.37 **.813.Risk Management3.021.52-.26.39 **.794.FHT2.881.58.44.36.43 **.895.PPP3.001.54-.33 *.17 *.28 **.31 **.836.Income Source3.121.55.42.15 **.38 **-.24-.42 ** .777. Expenditure Mandate2.941.57-.14 **.39.17.29 **-.22 .31 **.868.FA2.901.59-.35 **-.27 *.24.42 *.16 ** -.28 * .32 ** .85Notes: FHT = Fiscal Horizontal Transfers; PPP = Public-Private-Partnership; FA = Fiscal Autonomy;**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level(2-tailed); Reliability coefficients are in parentheses; n=252. 2Dependent Variable: Budgetary EfficiencyParticulars?SEtTVVIFDirect EffectsEqualization? Fiscal Autonomy.36 **.15 1.69 .88 1.14Accountability?Fiscal Autonomy.16 *.17 1.83 .71 1.42Risk Management?Fiscal Autonomy.42 **.86 1.16 .73 1.38Indirect EffectFiscal Horizontal Transfers ?PPP? Fiscal Autonomy.39 ** 1.69 2.63 .86 2.75Adjusted R 2 [.697]Bootstrapping Results: Indirect Effect (CI) [-0.084â??"-0.001]Notes: SE = Standard Error; TV = Tolerance Value; VIF =Variable Inflation Factor; PPP = Public -Private Partnership; *p < .05;** p < .01; Bootstrap Sample Size = 2500;CI = Confidence Interval; Standardized Coefficients Reported; n = 252. © 2019 Global Journals Fiscal Horizontal Transfers and Fiscal Autonomy in Local Government: Evidence of Public-Private-Partnership Influence in Uganda * Too big or too small? A synthetic view of the commitment problem of inter-regional transfers NAkai MSato Journal of Urban Economics 64 3 2008 * Fiscal federalism and local government finance in Nigeria: An examination of revenue rights and fiscal jurisdiction STAkindele OROlaopa ASObiyan International Review of Administrative Sciences 68 4 2002 * Evaluating the efficiency and equity of federal fiscal equalization DAlbouy Journal of Public Economics 96 2012 * Exploratory structural equation modeling TAsparouhov BMuthen Structural Equation Modeling 16 3 2009 * An unbreakable path? A comparative study of decentralization and local government development trajectories in Ghana and Uganda NAwortwi International Review of Administrative Sciences 77 2 2011 * Quantitative empirical analysis in strategic management RBettis AGambardella CHelfat WMitchell Strategic Management Journal 35 2014 * Public sector governance and accountability series: Inter-governmental fiscal transfers, Principles and Practice RMBird MSmart Boadway, R. & Shah, A. 2002 The World Bank Washington Inter-Governmental Fiscal Transfers: International Lessons for Developing Countries * Bootstrapping goodness-of-fit measures in structural equation models KABollen RAStine Sociological Methods & Research 21 1992 * Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis JREdwards LSLambert Psychological Methods 12 1 2007 * A Critical survey of sub-national autonomy in African States YFessha CKirkby Publius, The Journal of Federalism 38 2 2008 * Managing the risks of public private partnerships in Scottish local governments JHood NMcgarvey Policy Studies 23 1 2002 * Making effective public private partnership in local government WULameck Research Journal of Social Sciences 4 2009 * Does competition for capital discipline governments? The role of fiscal equalization YLiu International Tax and Public Finance 21 2014 * Public Finance in Theory and Practice RAMusgrave PBMusgrave 1973 McGraw-Hill Kogakusha, Tokyo * Toward a second generation theory of fiscal federalism WEOates International Tax and Public Finance 12 2005 * Financial autonomy and consistency of central government policy towards local governments LOulasvirta MTurala International Journal of Administrative Sciences 75 2 2009 * Intergovernmental transfers as an instrument to alleviate poverty MGRao ADas Gupta Environment and Planning C: Government and Policy 13 1995 * Bridging the economic divide within nations: A scorecard on the performance of regional development policies in reducing regional income disparities RShankar AShah World Development 31 8 2003. 25 th March, 2019 -Sunday 28 th April, 2019