Property rate in Ghana : a poor local revenue source or underexploited potential ?

This article aims to assess whether or not property rates in Ghana are a good potential source of local revenue. Through detailed analysis of six local government case studies, we find that present prospects for most rural local areas to raise substantial rate revenue are circumscribed, but in urban councils they are more promising. Nevertheless, no council is able to collect rates fully and from all rateable properties. This is attributed to several factors: the politicisation of taxation; ethnic homogeneity; intergovernmental transfers; partisan local government elections; resistance caused by elite design; and the denial of public information. Although these factors have been identified in previous literature, the study includes new findings which challenge received academic thinking on how they affect local tax collection in developing countries.


Introduction
Property rate performance as a percentage of Gross Domestic Product (GDP) is far lower in transition economies (0.69%) and the global South (0.6%) than in the USA (4.0%), OECD (2.0-3.0%) and developed countries generally (2.2%) (ICDT 2017;Norregaard 2013). Further, when we look across Africa we find an average yield of only 0.38% (African Property Tax Initiative, 2017;McCluskey and Franzsen 2017). This is regrettable, as property rates are the most promising sources of income to accelerate the socio-economic transformation of cities in Africa (Moore et al. 2018). Among sources of local revenue (e.g. licences, fees, fines and charges), levies on the possession of immovable assets are one of the most viable avenues for raising income and are critical to achieving local financial stability and growth (Asiama 2006;Biitir and Assiamah 2015).

CJLG 2019 2
This disjunction is exemplified in the Ghanaian context where property rates have the potential of garnering 30% of municipal income (Kuusaana 2015). Yet country-wide, yield from levies on properties is less than 2.0% of local government earnings, and in 2010 represented just 0.03% of GDP (Kwakye 2010). Nor are trends moving in the right direction. Kwaye (2010) reports that revenue from taxes on properties was 0.2% and 0.19% of Ghana's GDP in 1997 and1999 respectively, which are tiny fractions, but still much larger than in 2010. Recent estimates indicate that the share of property rates of GDP is less than 0.5% and accounts for about 14% of the total revenue envelop of local governments in Ghana (Biitir and Assiamah, 2015). This strongly suggests that the huge potential of property taxes in Ghana is not being harnessed, and raises many questions as to why and how this situation arises.
Previous research on property rates in Ghana has examined individual aspects: challenges in its design and implementation (Boamah 2013); the effectiveness of the government's Land Valuation Division in discharging its mandate of property assessment (Petio 2013); and the prospects and problems of levying immovable assets and their impact on internally generated funds (IGF) (Kuusaana 2015). This paper considers all these variables, alongside others such as ethnic homogeneity, the impact of national and local elections, elitist approaches to tax design, and the impact of intergovernmental transfers on local tax efforts. It also provides insights into different urban and rural approaches and outcomes

Theoretical framework
The theory that is used to guide this study is the political economy of local government taxation (Jibao and Prichard 2015). The role of municipal tax performance and reform in underpinning an effective decentralised local authority system has been progressively acknowledged and is predicated on three interrelated variables (Jibao and Prichard 2015). First, a local authority that earns income by imposing levies on inhabitants will predictably be more financially self-determining, and will be better able to match council outlays with local most-felt needs. Second, such a council will have greater motivation to promptly address citizens' problems and this will in turn encourage residents to be tax-acquiescent.
Finally, residents' honouring of their local levy obligations will give traction to their calls for mutuality and answerability, and in turn will drive wider consultation on tax policy development and implementation (Jibao and Prichard 2015;Weingast 2009). However, the literature on decentralisation has surprisingly paid only scant attention to the practical details of developing local tax policy and implementation (see e.g. Faguet 2014).
Related studies that have been undertaken, for example, the politics of local income generation, have tended to look at the obstacles that the central government decision-makers face (Eaton et al. 2013;Jibao and Prichard 2015), and have rarely considered the political subtleties operating at local level (Smoke 2013). Paradoxically, however, it is acknowledged that these local subtleties are critical factors in determining success or otherwise (Jibao and Prichard 2015), and it is evident that notional decentralisation has led to only marginal, if any, progress in the collection of local taxes (Bird 2011).

CJLG 2019
3 Key reasons for poor local tax-raising in developing-country contexts include reliance on national government allocations; risk of local government capture by influential local economic elites; and weak capacity (Bird 2011;Jibao and Prichard 2015). Commentators have blamed this state of affairs on a lack of political will at local level to overcome community opposition, and low motivation to collect taxes (Jibao and Prichard 2015). This approach however can be criticised as being reductionist; since other important dynamics are at play, although research seems to have not yet been undertaken (Smoke 2017(Smoke , 2019. The present study attempts to bridge the gap by analysing a number of other variables that influence local-level tax efforts and reform.
A literature review of this field reveals a striking number of largely untested assumptions which have built up about these under-analysed factors. This paper, therefore, critically examines both the factors and the assumptions, and attempts to provide evidence on whether this 'received wisdom' is correct.
The factors addressed to see whether they impact on local authority property rate collection are: 1) the origin of local politicians' mandate (i.e. elected locally, or nominated centrally?) 2) the ethnic composition of communities (i.e. homogeneous or diverse?) 3) the presence or absence of substantial fiscal transfers from central government 4) individual elections (i.e. whether their conduct and proposed policies influence local attitudes) 5) the quality of municipal services that residents receive from their local authority.

Property tax in Ghana
The concept of a property tax has existed in Ghana since colonial days, although the structure of the tax has changed many times over the years. When Ghana gained independence in 1957 it sought the assistance of the United Nations (UN) to develop a uniform, equitable and sustainable method of property rating. The UN responded positively and a system based on taxing each property according to its replacement cost was recommended and adopted (Kusaana, 2015).
Today, property rating in Ghana is governed by the Local Government Act 1993 (Act 462), which upholds this 'replacement cost' approach to levying properties (Commonwealth Local Government Forum 2018). The approach calculates the rate impost as "the amount it would cost to provide buildings, structures and other developments as if they were new on an undeveloped land or site at the time the premises are being valued" (Section 96(10)[a]). Local authorities in Ghana are legally empowered to collect rate revenue due to them fully and promptly. This empowered position has not, however, translated into improved local tax performance.
The reasons offered in the existing literature for sub-optimal property rate performance in Ghana include: 1) outdated or non-existent property registers and valuation rolls 2) manual, not automated, rate implementation systems 3) limited administrative capacity and equipment 4) a narrow tax base caused by extensive legal exemptions 5) inadequately described taxable properties in the fiscal registers.
CJLG 2019 4 A number of reforms have been introduced to deal with these challenges. The government has mandated that local governments undertake street-and house-naming exercises to ascertain the type and total housing stock and other immovable properties in each council (Government of Ghana 2007). Some metropolitan authorities have contracted out the collection of property rates and other taxes to private firms, and have introduced automated revenue tracking systems. Many councils are also working to revalue the land and taxable properties in their area, and to create digitised maps of land (Farvacque-Vitkovic et al. 2008;McCluskey and Franzsen 2017). Nevertheless, property rate performance remains very poor. This paper investigates the reasons for the poor performance and provides suggestions as to how they might be resolved.

Methodology
The study gathered evidence in two ways: content analysis of secondary information, and semistructured interviews. The secondary sources were analysed to enable a better understanding of the historical context of Ghana's property rate performance and the issues faced. It also allowed comparison between councils and financial years. Such comparison enlarged the scope of generalisations and provided complementary insights to the interview findings.
The interview method was preferred to a questionnaire method because the national illiteracy rate is high at 21% (in 2018) for 15 years and above (World Bank 2020). The illiteracy rate is even higher in rural areas, so many respondents, particularly residents/property owners would not be able to read and respond to a questionnaire. Questions were devised in English and translated into relevant local languages (Sisale, Ga, Dagbani). Semi-structured interviews were also felt to be a methodological choice as they provided a means for exploring complexity and details of process which could not be examined by less flexible instruments such as questionnaires or quantitative tools (Liamputtong and Ezzy 2005). This enabled respondents to have considerable scope to express their views in their own words. Moreover, the face-to-face interaction afforded by the interviews allowed the author to obtain valuable contextual 'thick' description such as details of personal reactions, specific emotions, and so on, that added depth to the responses. Metropolitan Assembly (AMA), urban metropolis in southern Ghana. These local authorities were chosen to ensure regional balance, and to enable comparative analysis of urban and rural districts in property tax performance.

CJLG 2019 5
In each district, 31 respondents from the local administrative and political elite and ten property owners/residents were interviewed. These included the District Coordinating Director (bureaucratic head of district), District Finance Officer, District Planning Officer, District Budget Officer, and Land Valuation Officer, six revenue supervisors, ten revenue collectors and ten elected councillors. The respondents were chosen because their roles are connected to the formulation and implementation of property tax in their respective jurisdictions.

Findings
This section outlines the research findings, and enables comparison of the viability of the strategies adopted, and challenges faced, by each local authority in their implementation of property taxation.

Property rate performance
It may be helpful to first consider the collection performance of the councils studied. During the period under review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) the two urban councils, AMA and TaMA, budgeted to collect property rate revenue of GHC 74.4 million and GHC 3.1 million, respectively. AMA in actuality collected GHC 41.8 million, which was GHC 32.5 million short of its target. In the same period, TaMA collected GHC 1.6 million less than the budgeted revenue by GHC 1.5 million (see Figure 1). The budgeted revenue is calculated based on the rate of growth of properties, the inflation rate for the year, the total stock of properties, as well as changes in the tax rate. Although AMA did slightly better than TaMA in percentage terms for the entire period (56.2% as against 50.8%), in absolute terms AMA raised almost 26 times the amount attained by TaMA. AMA as the capital of Ghana has 119,850 rateable properties, compared to just 54,000 in TaMA (AMA 2007; Issah 2011). Moreover, TaMA property owners have to pay more per property. To illustrate, a firstclass residential property in TaMA is taxed at a rate of 0.04% of the rateable value (they are required to pay 0.04 of a cedi for each cedi of value that the property has been judged to have), while in AMA the tax is 0.0017%. A second-class residential property in AMA is taxed at 0.0013% while in TaMA the rate is 0.01% (AMA 2014; TaMA 2012).

CJLG 2019 6
In any event, as Figures 2 and 3 show, the total property rate revenue of both AMA and TaMAalthough fluctuating over the yearsis rarely more than a small proportion of their income. For example, for the entire period under review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015), at AMA property rate revenue was equivalent to 16.8% of its total external transfer funding, and at TaMA to just 3.3% of total external funding. As proportions of IGF the property revenue of the two urban councils is also not appreciable. As Figure   3 shows, the lowest proportion for TaMA occurred in 2015 when the property revenue raised was 1.2% of its internally generated revenue for that year. The highest proportion was attained in 2009 when property revenue made up 17.2% of the council's internally generated funds. For the entire period under review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015), TaMA property revenue as a proportion of its total internally generated revenue was 16.1%. In the case of AMA, the lowest performance of property rate as a proportion of its internally generated revenue was recorded in 2015 which was 8.3%. The highest proportion occurred in 2004 which was 39.3%. For the entire period under review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) property rate performance as a proportion of AMA's internally generated revenue was 27.6%.

CJLG 2019 7
The story of the rural district is similar. Not only are there discrepancies between targets and achievements in property revenue, but there are also wide swings in yield from year to year. A target is a performance appraisal yardstick that was introduced as part of the New Public Management inspired public sector reforms in the Ghanaian public service (Ayee 2013). It is thus an accountability tool that compels public organisations to not only set targets for individual employees but also the organisation as a whole. Oversight institutions can then use them to demand to know why there are discrepancies between targets and achievements. It is also required for budget preparation and it is a prerequisite for budget approval. Moreover, citizens use them to determine whether their councils have performed to the level anticipated.
GEDA projected to collect a total of GHC 6.7 million during the period under review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) but only succeeded in raising GHC 3.1 million. GWDA budgeted to collect GHC 5.2 million but actually netted GHC 1.9 million (see Figure 4). As indicated earlier, the projections are based on the rate of property growth, inflation rate, the total stock of properties, as well as changes in the tax rate. Probably councils also post ambitious targets to placate assessors of the District League Table (   The picture in SWDA and SEDA is even more concerning. SEDA aimed to collect GHC 616,097.00, but even so it actually collected 62.1% with GHC 382,521.04underperforming its target by GHC 233,575.98 (37.9%). Its total actual property rate revenue of GHC 382,521.04 was equivalent to 1.4% of its external transfer income and 29.5% of its total internally generated funds (see Figures 6a and 6b).  Moreover, its fluctuations in yearly performance were even larger than those of SEDA, as Figure 4 shows. Such wide fluctuations in revenue do not provide predictable streams of revenue over time. As one councillor in AMA bemoaned, "This distorts planning for local service delivery" because, he explained, "demand for services grows over time and outstrips revenue deriving from erratic property rate income." It is clear from the above findings that small and rural districts are performing dismally in terms of both IGF and property rate collection. They therefore depend more on external transfers than their urban counterparts to meet their obligations.

Causes of gap between property rate potential and performance
Property rate potential is determined by the respective values of the total stock of the different categories of rateable properties multiplied by their respective tax rates and then summed up. The Ghana Statistical Service (GSS) conducts censuses of not only the population but also the property stock (residential, industrial and commercial) (GSS 2010). This enables the GSS to project the population growth rate and the property stock growth rate. Local governments then use the growth rate of the property stock to assess the total stock of rateable properties in the local government jurisdictions. The property rate potential of the local governments is therefore a realistic gauge. Respondents in this study were asked what they thought was responsible for the gap between property rate potential and performance. The reasons suggested include, in descending order: the politics of taxation (95.0%); ethnically homogeneous communities (93.0%); elections and tax collection (92.0%); intergovernmental transfers (91.5%); lack of appreciation of justification for property rate (90.0%); lack of geo-property information (89.9%); resistance due to elitist design (88.2%); lack of transparency and accountability (86.6%); access to public information (85.0%); mismatch between tax income and service provision (83.3.%); and weak monitoring and evaluation (78.2%). In several cases new information was gleaned about the reasons for property tax non-payment; reasons which have not previously been discussed in the academic literature and which call into question the assumptions mentioned in the 'theoretical framework' section above. These issues are discussed below.

The politics of taxation
In regard to the politics of taxation, respondents were asked the following question: Do you think politics influence the design and implementation of property rates in your council? to provide the goods and services the people desperately need….Once services are provided in the right quantities and quality this will increase government legitimacy, shore up political support, and ultimately win votes.

Ethnically homogeneous communities
Against the backdrop of the ethnically diverse communities in the districts (see box 1), respondents were asked whether they thought ethnic homogeneity affected tax compliance. Ninety-three percent (229) of the respondents believed that ethnically homogeneous communities were another cause for the gap between property rate potential and actual performance. The general view was that indigenous populations tend to believe that land is given to them free of charge by God, and bequeathed to them by their ancestors, and as such should not incur taxes or rates. As a finance officer in TaMA  The The full picture is, however, more nuanced. The possibility of deference by tax collectors was highlighted by one AMA revenue inspector, who suggested that wealthy or influential indigenous residents were not always even asked to pay: There is a perception out there, especially among revenue collectors, that economic elites have both the economic and political muscle to resist tax payment … [revenue collectors] therefore ride on that perception and fail to present them with demand notices although they are willing to pay.
There were also other respondents who felt the picture of indigenes as wilful non-payers was too harsh.
One said, "Many of the well-educated indigenes are property tax compliant and come willingly to pay rather than wait to be served with demand notices." Another commented, about poorer indigenous residents, that "their non-payment of property taxes has less to do with resistance but more to do with their inability to pay" adding that, "this is particularly the case with the urban poor who although acknowledge their tax obligations are unable to honour them as regularly and without default as would be desired." One residential property owner in a porch suburb confirmed that he is tax compliant:

I always honour my property tax obligation when it is due. Although no tax collector has ever presented me with a demand notice, I go to the offices of the local authority and request to know the amount of my property tax obligation and when the information is given and I am convinced that the amount is correct I pay without hesitation.
A commercial property owner also said that: I own this property and have been operating business here for the past ten years but no revenue collector has ever presented me with a demand notice in relation to my property Generally, however, migrant residents were perceived much more willing to pay property tax than indigenous residents. As a councillor of TaMA put it:

Migrant residents are largely compliant with residential, industrial and commercial property rates … [they] fear that if they default or resist tax payment they could become the target of repressive local authorities' policies or incur the wrath of the indigenes who would accuse them of benefiting from the city's opportunities but failing to contribute to its development.
This perception was borne out by a migrant property owner who disclosed that:

I am an Akan and I have lived in Tamale for over a decade. I own two houses here and a wholesale mother and baby care shop as well as a retail shop. I pay taxes on these properties promptly. Tamale has given me a lot and I must reciprocate by honouring my tax obligations.
CJLG 2019 13 A Labanese resident of Tamale intimated that:

I own these sawmills and I always pay property rate on it promptly. Not only do I feel that as a good resident I should honour my tax obligation but I also fear that failure to do so may be penalised by tax authorities.
This seems to be borne out in AMA, which is the biggest, most cosmopolitan metropolis (where indigenes constitute 29.7% of the population and migrants take up a disproportionate 70.3%) and the seat of the national government, and which faces much less tax resistance. One Indian businessman said: I own a chain of supermarkets in different parts of the city. I pay property tax on these assets regularly. Tax evasion is a serious offence in India and I fear that it is equally so in this country.
Another migrant (Ewe) residential property owner asked: Why should I delay payment of property rate until the tax authorities come clamping down on me? I fear that by delaying or defaulting tax payment I will incur reputational costs in the eyes of fellow residents and the tax authorities.
According to a revenue inspector of AMA, "The migrant population of the city is not only overwhelming but also they own a disproportionately large portion of industrial, commercial and residential properties." He added that "The migrant population is comparatively more acquiescent to authority compared with their indigenous counterparts." This deference of migrants, explained a councillor, "is responsible for their lower tax resistance compared to indigenous residents."

Election and tax evasion
Respondents were asked: Do you think elections have an effect on tax collection? Two hundred and twenty-six respondents (92.3%) said yes, while 12 (5.0%) said no and 7 (3.0%) said they did not know.
One chief bureaucrat thought that being appointed, rather than elected, increased the power of a local leader to be able to enforce tax compliance: "The appointment rather than election of mayors in Ghana implies they can embark on austerity and aggressive tax collection and reform measures since their tenure of office is not determined by voters." However, it was felt by others that it was national, not local, elections which influenced local leaders when they considered their local tax-raising strategies.

Lack of appreciation of justification for property rate
On the issue of lack of appreciation of the justification for property rate, 222 (90.2%) of the interviewees said that ratepayers could not understand why they should be obliged to pay property tax. The remaining 24 (9.8 %) respondents appreciated the need to pay taxes irrespective of whether they received services or not. One residential property owner at AMA queried that, "I do not get any service in return for the taxes that I pay." A similar sentiment was expressed by a TaMA residential property owner when he charged that, "the chorus we property owners sing is that our properties are largely not supplied with water and waste water reticulation, storm water drainage, paved intra-city roads and household garbage collection." In the absence of these services, she added, "there is no justification for the local authority to continue to collect property rates from residents." Similar sentiments were expressed by respondents in the other districts. For example, one SEDA resident said, "Why should I pay property rate without water, sewage and storm water and drainage facilities to my house." A GWDA resident queried that, "I dislike the property tax because it offends indigenous land rights." A revenue collector confirmed these concerns of residents when he said that: we often encounter ratepayers who angrily retort upon notice to pay their rates that we don't know whether the local government exists at all because there is even no place of convenience in our crowded public places like the markets.

Lack of geo-property information
Concerning lack of geo-property information, 221 (89.9%) of the respondents agreed that it is responsible for poor rate performance. The remaining 25 (10.1%) could not express a view. One of the respondents who conceded that the absence of geo-property information is a problem averred that, The fiscal cadastre forms the information base for property taxation in every country. It comes in the form of an electronic, digitised or paper-based map. So the lack of fiscal cadastre undermines the efficiency and effectiveness of property tax administration in Ghanaian local governments. To compensate for this deficiency, the local governments use the growth rate of property stock computed in the population and housing census by the Ghana Statistical Service (GSS). Since the GSS conducts the census at long intervals (four years at least) this may not adequately reflect the reality.

Resistance due to elitist design
Pertaining to resistance due to elitist design, respondents were asked the following question: Is tax design an inclusive process in your council? In response, 217 (88.2%) of the informants said no, claiming rather that it is elitist and exclusionary. As one resident of AMA lamented, despite government guidelines "in practice the councils do not consult citizens in determining the tax structure and rates." Many entrepreneurial respondents also felt that properties and businesses were often wrongly or unfairly classified; this view was expressed, for example, by a hotelier in AMA, a private high school proprietor in TaMA and a wholesaler in SEDA.
However, not all respondents agreed. A senior bureaucrat at AMA pointed out that "notices of invitation are usually sent out well in advance of the scheduled dates for the consultations but few people actually honour the invitations." He said that, "Proposed tax structures, classification of taxable units and the respective tax rates agreed at the consultative sessions are usually sent out to taxpayers for comments before their final ratification in the assembly." A Municipal Coordinating Director at TaMA concurred, stating that, "In the case of our council not only do we consult with taxpayers but the proposed levies are sent to the taxpayers with ample time to make comments before reverting to the council." Some respondents, however, claimed that such consultations were 'tick-box' exercises only.
A resident of TaMA noted: The council rarely consults taxpayers, and on the rarest occasions when it does, it conducts manipulative consultations because decision-makers come to the table with prepared proposals and participants are merely asked to endorse them … The authorities do not allow for a process of genuine exchange of ideas, negotiations, bargains and consensus building at the consultative forums.

Lack of transparency and accountability
Pertaining to transparency and accountability, 213 (86.6%) of respondents thought it is one of the major factors responsible for the poor property rate performance of councils. Twenty (8.1%) respondents did not subscribe to this view, whereas the remaining 10 (5.3%) could not express a view. Those informants who agreed that opaqueness and lack of accountability are culpable for poor tax efforts intimated that rate impost is usually set arbitrarily without opportunity for public input. A TaMA resident charged that, "the inability of local governments to provide even basic services heightens suspicion surrounding the accountability deficit." A GEDA resident said that, "some local governments have had bad press for impropriety and fund mismanagement." A property owner in AMA expressed similar misgivings, saying that, "despite the huge sums of money local authorities collect from property tax, household garbage suffers infrequent collection." The accountability problem is confirmed by Mohammed (2019)

Access to public information
Interviewees were asked the following question:

Mismatch between tax income and service provision
The 60 respondent property owners were asked the following question: As a property owner do you pay property rate? Fifty (83.3%) respondents answered no, 7 (11.6%) answered yes and 3 (5.1%) did not In rural GWDA, a resident charged that "although rural water supply together with sanitation and storm water drainage services is the responsibility of the council, it does not deliver those services".
When asked about the lack of services to properties, a number of causes were cited. Taking the example of water supply and sewage management, a budget officer of AMA pointed out that "urban water supply is the responsibility of the Ghana Water Company and not the AMA." He acknowledged that sewage reticulation and discharge service "has recently been added as a function of local government" but pointed out that "the transfer of sewage functions to councils was not accompanied by a commensurate transfer of means and capacity to discharge that mandate." Interviewees were then asked: Have you put pressure on your council to deliver the required services?
In response 18.3% (11) of the respondents answered yes. A TaMA respondent noted: "We have on several occasions organised protest marches and demonstrations, a few of them violent, against the lack or poor delivery of services … I remember, we stormed the offices when the assembly was in session and the meeting was temporarily disrupted." Despite this citizens' pressure, he confirmed that "service provision has not improved due to lack of resources, requisite machines and tools as well as capacity."

Weak monitoring and evaluation
Weak monitoring and evaluation were also identified by 192 (

Discussion
Several lessons can be drawn from these findings. First, tax efforts and tax reform require both technical action and political will. However, this study suggests otherwise: the evidence here indicates that tax resistance rather emanates mainly from ordinary citizens whose influence derives from the fact that Ghana's de facto two-party political system has turned their national-level votes into tax resistance cards. This evidence appears to contradict prevailing assumptions (e.g. Labonte 2012). It reflects the power of ordinary citizens who have no economic clout but who as a group, constitute an electoral powerhouse that political leaders ignore at their peril.
Second, ethnic homogeneity does not necessarily promote tax compliance, as the literature would suggest. Instead, the evidence here shows that indigenous ethnic homogeneity rather provides a source of strength to oppose local taxation. It appears to enable indigenes in TaMA (who are Dagombas), and their counterparts in SEDA and SWDA (who are Sissale) to mobilise and resist payment of residential property rates. This resistance stems from the belief that land is given free by God, so its development for residential purposes should also be free of taxes. Equally persuasive is the finding in cosmopolitan AMA, with its huge ethnic diversity, that migrant residents are disproportionately tax compliant partly because they fear reprisals otherwise. This is consistent with previous findings by Prichard (2010) and Jibao and Pricharch (2015) who report that it is easier to overcome tax resistance within a more mixed, and less cohesive, community.
The received academic wisdom that homogeneous groups are easier to raise taxes from is therefore challenged in TaMA, SEDA and SWDA who are rather tax resistant. In cosmopolitan AMA, however, with greater heterogeneity there is greater tax compliance particularly from migrants. As migrants are compelled to shoulder a disproportionate part of the tax burden because they lack political influence and local connections, then this can further exacerbate inequality, and a rise in local tax revenue can be seen as regressive rather than redistributive.
Third, this study found that intergovernmental transfers do not necessarily disincentivise local government tax efforts, contrary to previous analysis (e.g. Weingast, 2009). While they may, sometimes they may also spur tax efforts or have no effect on them. The interplay of factors is complex, and depends on contextual conditions such as levels of local natural resources, motivation of leaders, and CJLG 2019 20 the size of the local economy. Councils like SEDA and SWDA lack natural resources and their economies are also very small. They thus have limited property tax bases, and are likely to be unable to raise more than 10% of their total income from own-source revenue, however hard they try.
Intergovernmental transfers thus have no effect on their tax efforts. By contrast TaMA has a large and vibrant economy, and a much wider tax base, particularly for propertyand yet it still generates less than 25% of its total revenue from internal sources. TaMA's case therefore supports the claim that central transfers weaken tax efforts. AMA, meanwhilethe national capital and also the most populous and cosmopolitan city -is in the third category, and seems to benefit from a 'virtuous circle' in which its wide tax base enables it to derive on average more than 55% of its total revenue from internal sources, while still receiving large intergovernmental transfers.
This study does, however, bear out the existing academic consensus across the developing world (see, for example, Brollo and Nannicini 2012) that politicisation of intergovernmental transfer impacts local tax efforts adversely. A good example is the directive by central government to re-route DACF transfers to fund the 'One Village One Dam' project as described above, which respondents believed was directly linked to a national government campaign promise in the previous election. Respondents in AMA also expressed a fear that further politicisation was likely at critical juncturesand this fear could have spurred it to intensify own tax revenue mobilisation, but such an effort will be thwarted by the national government because of its adverse electoral consequences.
Finally, it seems clear from the study that elitist and exclusionary approaches to the design and implementation of tax systems, and lack of transparency in how tax revenue is spent, enrage local residents and contribute to tax resistance. This is especially true if citizens feel they do not benefit in terms of service provision. This speaks of the enormous barriers to true citizen participation that still exist at local level in developing countries, despite the 'third democratic wave' dating to as far back as the 1990s.

Conclusion
This paper set out to investigate whether there a disjunction between property rate potential and performance of Ghanaian local governments. Interview information and secondary evidence from six local authorities were used to determine whether there is a gap between the two variables (potential and actual rate performance). The gap which was found between the two variables is blamed on a multitude of complex causes (political, cultural, legal, administrative and educational), many of which are interrelated and which are discussed in detail and supported by sector and citizens views. The author concludes that unless the local context is fully understood and addressed, and all the detailed factors discussed above are taken into account in tax design, implementation and reform efforts, the emergence of an efficient and effective property rate regime in Ghanaian local governments will remain elusive.

Declaration of conflicting interest
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The authors received no financial support for the research, authorship, and/or publication of this article.