Sage Journals: Discover world-class research

Abstract

The question of how much of the potential tax revenue is actually obtained remains very critical in Ghana’s efforts to improve domestic revenue mobilization. This paper computes and examines tax gap and compliance burden among micro and small businesses which constitute more than 90% of Ghana’s informal economy. Using data on 485 businesses in Ghana, the study finds that while some businesses underpay their taxes, surprisingly others actually pay more than expected. The average tax gap is computed to be about GHC 109.2 (US$19.2). Small businesses are associated with a higher tax gap of 24.9% compared to their micro counterparts. Moreover, we find that as compliance burden increases, the tax gap also increases, albeit significant variation in the effect of compliance burden on tax gap across regions and different tax handles.

JEL Classification: H25, H26, H27, H32

Keywords

compliance burden compliance cost micro and small enterprises potential tax tax gap Ghana

Introduction

The need to strengthen domestic revenue mobilization continues to receive global attention. This concern is echoed in almost all the global development program documents—The United Nations Agenda 2030 for sustainable development of leave no one behind, the sustainable development goals (specifically SDG 17.1), the African Union Agenda 2063 (dubbed “The Africa We Want”) as well as the Ghana Beyond Aid agenda all recognize that increasing domestic resource mobilization remains indispensable in delivering economic development and improved welfare. Improving domestic revenue mobilization is particularly critical in Ghana, as public debt has assumed alarming proportions. Specifically, Ghana’s debt-to-GDP rose to 79% in 2021 with the fiscal deficit including energy and financial sector costs worsening to 15.2% of GDP (International Monetary Fund [IMF], 2021). This puts the nation among the economies in sub-Saharan Africa (SSA) classified as being at high risk of debt distress (World Bank, 2018).

Ghana’s high fiscal deficit reflects its low tax revenue performance. For example, the country’s tax-to-GDP ratio has remained less than 14% since 1983 (World Bank, 2020). While the average tax-to-GDP ratio for Africa increased by 0.3% points between 2018 and 2019, that of Ghana decreased by 0.6% points from 14.1% in 2018 to 13.5% in 2019, falling below the 16% average of 30 African countries which is even far below that of Asian and Pacific economies (21.0%), Latin America and the Caribbean (LAC) (22.9%), and the OECD (33.8%) (OECD, 2021). Using national aggregate data, World Bank (2020) found the tax gap in Ghana to be very high. Specifically, the corporate tax gap was estimated to range from 81.6% to 85.6% of potential corporate tax revenue, which was equivalent to 9.4% and 12.6% of GDP in 2014. The import tax gap was found to be around 32.5% on average between 2012 and 2016 while the VAT compliance gap increased from 18.2% to 39.3% between 2011 and 2016.

As part of efforts to awaken a national consciousness on the need to pay critical attention to domestic revenue, the President of the Republic of Ghana in 2017 launched the “Ghana Beyond Aid” agenda. Primarily, the agenda seeks to build a transformed Ghana that is prosperous enough to be beyond needing aid. The Ghana Beyond Aid, thus, emphasizes domestic resources mobilization as the primary source of finance for the country’s growth as well as economic and social transformation (Government of Ghana, 2019). Schneider (2020) and De Mooij et al. (2018) stressed that preference should be given to domestic revenue mobilization in developing countries because there exists tax revenue potential. This untapped tax potential results in a low tax to GDP ratio in most developing countries (Bogetić et al., 2022; Odhiambo & Olushola, 2018).

An important step in a country’s effort to improve its domestic resource mobilization, therefore, must begin with a clear understanding of the country’s tax gap that is the difference between what should be paid and what is actually paid. In most African countries, tax gaps are considered to be high. The inability of countries to reduce the tax gap results in slow growth and inadequate provision of public goods and services (Danquah & Osei-Assibey, 2018). Globally, tax gap estimation is gaining a lot of attention for its importance in setting feasible revenue targets and designing appropriate revenue mobilization strategies. Currently, more than 23 tax administrations calculate the gap for some of their respective taxes. However, most of the tax gap studies have focused on developed countries and also on the formal sector and large taxpayers (Jansen et al., 2021; OECD, 2018; Poniatowski et al., 2020; Whicker, 2017). Thus, very little is known about tax gaps in developing countries, particularly the informal and small taxpayers. For Ghana, the burgeoning literature on tax gap analysis is not only based on old data, but also concentrates primarily on aggregates tax (Danquah & Osei-Assibey, 2018; World Bank, 2020). While such studies provide important insights, their estimates are at the macro level. Gemmell and Hasseldine (2012) indicates that macro-level estimates are at high level of aggregation and are susceptible to huge margin of error which may affect estimates and inferences made.

This paper fills the gap by estimating the tax gap among small tax payers who are predominantly micro and small businesses operating within the informal sector of Ghana. It further examines the effect of compliance burden—monetary cost of the average time spent honoring tax obligations—on the tax gap for different tax handles. Lastly, we explore the correlates of compliance burden. We concentrate on small tax payers for three reasons. Firstly, they are found in the informal sectors of developing economies. In Ghana, the informal sector employs about 86% of labor force (Anuwa-Amarh, 2015). As a large and growing share of GDP, informal taxation constitutes a greater potential source of tax revenue in developing countries (Schneider & Klinglmair, 2004; Schneider et al., 2010). Secondly, nascent evidence suggests that payment of taxes by small firms operating in the informal economy promotes social legitimacy and political accountability (Ayee & Joshi, 2008; Prichard, 2009). Thirdly, informal taxation boosts “tax morale” and builds tax compliance among larger firms (Alm et al., 2004; Terkper, 2003; Torgler, 2003).

On a whole, we found the average tax gap to be GHC109.2 (US$19.2) per month. However, the figure varied across different locations and firm structure. Tax gap was found to be higher for urban areas, firms operating as companies and partnerships as well as enterprises with movable structures. In terms of tax handles, we found small enterprises have a higher corporate tax gap (28.6%), income tax gap (25.9%), and VAT gap (25.6%) compared to micro-enterprises. Finally, compliance burden was found to be a key determinant of tax gap. Factors such as complexity of the tax system, size of firm, and tax knowledge are among the key factors that affect compliance burden.

The rest of the paper proceeds as follows. Section “Conceptual Issues: Tax Gap, Compliance Burden and Micro and Small Enterprises” explores the conceptual definitions of the tax gaps, small businesses and small tax payers, paying attention to the various tax handles, Section “Review of Literature” provides a brief review of the literature on tax gap and compliance, Section “Data and Methodology” presents the methodology and data, Section “Results and Discussions” presents and discusses the results, and finally the conclusions and policy implications are drawn from the results and discussions for policy.

Conceptual Issues: Tax Gap, Compliance Burden, and Micro and Small Enterprises

Potential tax is the amount of tax to be collected when tax laws are fully complied with. The difference between the potential tax and the actual tax collected is known as tax gap (Khwaja & Iyer, 2014) which literally means revenue loss. Tax gap, according to Dare et al. (2019) and Toder (2007), can emanate from three main sources: non-filing of tax returns, under-reporting of tax liability, and underpayment of tax liability. Tax gap is also linked to the uncollected tax revenue from taxpayers who file their tax returns but intentionally or unintentionally do not pay the amount they report. Underreporting of the tax base results in paying an amount of tax less than what the firm ought to pay. The computation of the tax gap could vary depending on the different tax handle being investigated. Minh (2007) explains that it is possible to multiply the tax base for each sector by the average VAT rate to get the VAT revenue potential which then aids in estimating the VAT gap. That notwithstanding, policymakers must consider the three components of the tax gap (non-payment, underpayment, and non-filing).

Compliance burden on the other hand, is the monetary value of the average time spent honoring tax obligations. The Standard Cost Model (SCM) is the main approach used in computing the costs associated with administrative activities in preparation, paying, and filing of tax returns. The costs are categorized into external and internal denoting the aspects of the services outsourced or insourced. The Standard Cost Model (SCM) is based on the belief that compliance burden resulting from the imposition of public policy (tax policy in our case), rules, or laws affects the overall cost efficiency of the firm. The reason is that firms’ productive resources are used in complying with those tax obligations rather than using them to produce outputs. In the SCM, the cost incurred during preparation, paying, and filing returns is the Price (P). The duration (measured in hours) required to complete preparation and filing returns is Time (T) and the repetition of the activities to be carried out per year (frequency) by the taxpayer that is subject to the information obligations is the Quantity (Q) (Dunlop & Radaelli, 2016). This is expressed mathematically as: C = P × T × Q, where C denotes the compliance burden. Since compliance burden forms part of the firm’s overall cost, it has the potential of affecting the firm’s compliance level.

In Ghana, the classification of small business enterprises has evolved over time. In defining Small Scale Enterprises in Ghana, Danquah and Osei-Assibey (2018) used an employment cut off point of 30 employees to indicate Small Scale Enterprises. This was further dis-aggregated into three categories: (i) micro—employing less than six people; (ii) very small, those employing 6 to 9 people; (iii) small—between 10 and 29 employees. Oppong et al. (2014) have classified enterprises into micro, small, medium, and large. According to Agyapong (2010), there is no generally accepted definition of businesses and that the definition of businesses varies depending on the size and kind of economic activities the firm engages in. The recent classification makes use of both the turnover criterion by the Ghana Revenue Authority (GRA) and number of employees’ criterion by the Ghana Enterprises Agency (GEA), formerly National Board for Small Scale Industries (NBSSI). For a business to be in the category of Micro and Small Enterprise (MSE), it must have an annual turnover of GHC 90,000 or less and also, have a workforce of less than 30. The disaggregation by the GEA which is also consistent with the Integrated Business Establishment Survey (IBES) by the Ghana Statistical Service (GSS, 2015) are as follows: enterprises with staff strength of less than six persons are classified as micro enterprises while those with 6 to 30 employees are small enterprises.

Three major sources of taxes dominate the country’s tax revenue: Value Added Tax (VAT), Corporate Income Tax (CIT), and Personal Income Tax (PIT) (Andoh, 2017). Evidence from the 2020 Revenue Statistics for Africa shows that in 2017, the share of PIT to tax revenue is 17%, 19% for CIT and 30% for VAT, whiles 16%, 23%, and 27% for PIT, CIT, and VAT in Ghana respectively. In 2018, VAT alone contributed 29% to the total tax revenue and this illustrates the assertion made by Danquah and Senahey (2020) that PIT, CIT, and VAT have high revenue elasticities.

Review of Literature

The dominant economic theory underpinning tax compliance is the economic deterrence theory which is traced to Becker (1968)’s seminal work on the economics-of-crime. Allingham and Sandmo (1972) first applied this theory to tax compliance in which they assume that an individual tax payer who receives a fixed amount of income, chooses how much of this income to declare to the tax authorities and how much to underreport. Tax is paid on the amount of income declared while no taxes are paid on underreported income. However, the individual may be audited and if the underreported income is discovered, a penalty is paid on each amount. Thus, the economic deterrence model views taxpayers as utility maximizers who are motivated by varying economic motives and choose to evade tax whenever the expected gain of evasion exceeds the cost (Atawodi & Ojeka, 2012; Ishola et al., 2020; Mangoting et al., 2020). High cost of compliance (high gain from evasion) will increase evasion and therefore create tax gaps. Since an increase in the probability of detection and the penalty rate reduce the gains from evasion (increase the cost of compliance), they tend to increase declared income, thereby reducing tax gaps. Slemrod (1989) introduced the variable of complexity into the model and established that complexity increased the cost of tax compliance and therefore increased non-compliance. Falkinger and Walther (1991, pp. 67–79) went beyond punitive factors to consider persuasive factors. They argued that a tax system that combines both penalties and rewards tends to be more effective in maximizing compliance than one that focuses solely on penalties. In view of this, positive inducements are also critical for compliance hence low tax gaps.

The subject of tax compliance and tax gaps has also been examined within fiscal psychological models. Unlike economic deterrence model, the fiscal psychological models focus on social psychological variables such as moral values and the perception of fairness of the tax system and the tax authorities. Taxpayers possess an array of attitudes and beliefs which interact and respond to social norms. As argued by Schmölders (1959, pp. 184–193), taxpayers have separate mentality for self-interest and contributing to community interests. Taxpayers with more positive attitude toward paying tax and working with the tax authorities, have greater willingness to pay tax, hence low tax gaps. Schmolders concluded that taxpayers’ attitudes were a reflection of cultural differences. Schmölders and Strümpel (1969)’s version of the fiscal psychology model stresses two main variables as critical in tax compliance. These are “rigidity of assessment” which measures the amount of tax fines, the assessment process, and the level of “red tape” involved in dealing with the tax authorities and, secondly, the “willingness to co-operate” which relates to individuals’ attitudes and perception of the tax system. Spicer (1974) particularly focused on the concept of exchange equity, arguing that the perceived inequity between taxes paid in return for public goods and services supplied by government was critical in determining tax compliance and tax gap.

Fishbein et al. (1980), offered an alternative version to the fiscal psychology model based on the Theory of Reasoned Action (TRA). According to the TRA, taxpayers’ compliance behavior depends on two key factors: personal attitude toward the behavior and subjective norms. An individual with positive personal judgment about tax compliance will comply while those with negative attitude will evade. The subjective norm is the social pressures on the person to perform or discard the behavior. In this case, taxpayers are concerned with the opinions of referent groups such as family members, friends, and colleagues regarding their non-compliant actions. Taxpayers will evade taxes if their friends and family members are in favor, otherwise they may not. Thus, approval of significant others has a key role to play in creating tax gaps.

Empirical studies have found several determinants of tax compliance. For example, Carsamer and Abbam (2020) found institutional, firm, and entrepreneurs’ characteristics to be critical determinants of SMEs’ tax compliance in Ghana. Tee et al. (2016) noted that the perceived adverse impact of existing tax policies was a major determinant of compliance among SMEs in Ghana Danquah and Osei-Assibey (2018) established that Ghana continues to lose US$56,951,573 from the informal sector after estimating the aggregated potential tax to be US$81,974,846 as against the actual tax payment of US$25,023,273. The study explained that business type, owners’ age, and sex affect tax gap. Okpeyo et al. (2019) found the inability of small enterprises to do proper bookkeeping of business transactions and file tax returns were the key determinants of tax compliance and tax gaps among SMEs. Akinboade (2015) identified the location of the business affects filing compliance in Cameron. In Yemen, Helhel and Ahmed (2014) found females to be more compliant with tax policies compared to their male counterparts.

Data and Methodology

Data

Data for the study was sourced from the “Ghana Beyond Aid” project which was funded and coordinated by the Directorate of Research, Innovation and Consultancy (DRIC) of the University of Cape Coast in Ghana. The data contains information on small taxpayers registered with the Ghana Revenue Authority (GRA) selected from three regions of Ghana. The sampling technique was a multistage. The country was first stratified into three belts: the northern, the middle, and the southern. One region with the highest tax capacity was selected from each belt. Consequently, the Greater Accra region was selected from the southern belt, Ashanti region from the Middle Belt, and the Northern region was selected from the northern belt. From each region, three Small Tax Payers Office (STO) were selected based on tax capacity. The STOs then provided a list of 12,596 registered small tax payers which was used as the sample frame. The Yamane (1967) sample size determination formula was used to obtain the sample of 388. The actual valid responses were 497 small tax payers. Specifically, from the Greater Accra region, a total of 143 small taxpayers were selected from the Kasoa and Kaneshi STOs. In the Ashanti region, a total of 230 were selected from the STOs in Konongo, Obuasi, Ashanti Mampong, and Suame. Lastly, a total 124 from the Tamale and Yendi STOs in the Northern region. However, due to missing observations, this study used a sample size of 485. The data contained information on the background of respondents, identification of the firms, employment, payable taxes, and knowledge on tax system, tax compliance, compliance cost, and firm’s investment decision.

The survey instruments were first validated at stakeholders’ workshop which brought together tax officials from the Tax Compliance Department of the small taxpayer office (STO) of the Ghana Revenue Authority (GRA), the revenue officers from the local government offices, representatives from the Ghana Enterprise Agency (GEA), tax experts form School for Development Studies, School of Business and School of Economics. After the validation workshop, the instruments were refined and submitted to the University of Cape Coast Institutional Review Board (UCC IRB) for ethical clearance. The instruments were then pre-tested in the Cape Coast Metropolis from 12th November 2019 to 15th November 2019. The pre-testing was needed to ensure internal validity and consistency of items. Actual field work was conducted among sampled businesses in the selected regions from 25th November 2019 to 10th January, 2020.

Estimation Strategy

In line with Chattopadhyay et al. (2002) and Danquah and Osei-Assibey (2018), OLS regression is used to examine to effect and correlates of compliance burden and tax gap. We estimate equation (1):

InTaxGa p_{i} = β_{0} + β_{1} InBurde n_{i} + β_{2} Exp + ε_{i}

(1)

where lnTaxGap is the log of the tax gap computed, lnBurden is the compliance burden, and Exp is a vector of the other explanatory used as controls. These variables are sectors, location, level of complexity, age of firm, type of MSE, business structure, legal form of business, sex of business owner, serviced used to prepare tax, tax audit, tax knowledge, number of taxes paid, and distance to tax office. Several variants of equation (1) were estimated for the three regions (Greater Accra, Ashanti, and Northern region) and the different tax handles (CIT, PIT, and VAT). All variables in equation (1) have been defined in Table 1

Table 1.

Definition, Measurement, and A Priori Signs of Variables.

Variable	Type	Definition and measurement	Expected sign
lnTaxGap	Continuous	Potential tax–actual tax
M_Sex	Categorical	A dummy of 1 for female and 0 for male managers	Positive
M_Edu	Categorical	A dummy of 1 if the owner has at least primary education and 0 otherwise	Negative
Location	Categorical	Location of the firm (0 = rural, 1 = urban)	Negative
MSE	Categorical	Dummy of 0 for micro enterprises and 1 for small enterprises	Positive
Age	Continuous	Number of years of business since registered with GRA	Negative
Complexity	Categorical	Complexity about tax system (0 = more complex, 1 = less complex)	Positive
Sectors	Categorical	Sectors (0 = industry, 1 = service)	Uncertain
lnBurden	Continuous	Compliance burden = P × T × Q (P is the cost per hour, T is the hours per day, Q is the days per week)	Positive onlnTaxGap
Audit	Categorical	Whether the firm is audited or not (0 = yes, 1 = no)	Positive
NTaxes	Continuous	Total number of taxes paid by the firm	Positive
BusinessSize	Continuous	Total number of employees in the firm	Positive
Legal_form	Categorical	The legal form of the business (0 = sole-proprietorship,1 = partnership, 2 = company)	Positive
ServiceUsed	Categorical	The kind of service used in honoring tax obligation (0 = internal,1 = external, 2 = both)	Positive
TK	Continuous	Tax knowledge score (percentage)	Negative
Distance	Continuous	Distance to the tax office measured in minutes	Positive

Source. Authors’ Construct (2020).

Measurement of variables

Tax gap: To estimate the tax gap, we ﬁrst identified and computed the potential tax for all the Micro and Small Enterprises. The specific tax rates were then applied on the firm’s turnover. The applicable rates during the period of data collection were 22% for CIT, 25% for PIT, and 3% and 15% for VAT flat and VAT standard respectively. This approach gives a more realistic information since all the businesses in the sample are registered with GRA. The difference was estimated by subtracting the actual amount of annual taxes paid from the potential amount to get the tax gap.

Compliance burden: It is the monetary value of the time spent honoring tax obligations. This was calculated using the Standard Cost Model (SCM). As describe earlier, this sum both the internal and external services costs associated with the administrative activities in preparation, paying, and filing tax returns. The cost incurred during preparation, paying, and filing returns is the Price (P), the time (hours) required to complete preparation and filing returns is Time (T), and the repetition of the activities to be carried out per year (frequency) by the business are multiplied and summed for each MSE. The currency is Ghana cedis (GHC) per month.

Sectors: This is a categorical data which classifies the MSEs into sectors of the economy. Although traditionally, there are three sectors in the economy: agriculture, industrial/manufacturing, and services, in this study, the MSEs were found to operate in two sectors—services and industry. The variable was therefore made a binary with 1 being services and 0 being industrial.

Location: This is a binary variable indicating whether or not the MSE operates in an urban area. In the study 1 denotes urban while zero denotes rural. Danquah and Osei-Assibey (2018) concluded that firms in urban areas were more likely to reduce the tax gap compared to their counterparts in rural areas.

Level of tax complexity: This variable was measured from a scale. Specifically, on a scale of 1 to 5 (1 represents the lowest while 5 represent the highest), respondents were asked the following statement: Ghana’s tax system is complex. All respondents who indicated 1 and 2 were in agreement that Ghana tax system was not complex while 3, 4, and 5 were reclassified as complex tax system.

Age: This is a continuous variable that measures the number of years the firm has registered with the Ghana Revenue Authority. Although the year the firm started operating should have been used to measure a firm’s age, the data available were not consistent. The higher the number of years the firms have been operating, the more it is expected to be abreast with tax regulations and hence less likely to be burdened and default compared to firms that currently register with GRA.

Type of MSE: This variable classifies the MSE as either micro or small enterprise. This study uses both the GRA (turnover) and NBSSI (number of employees) to classify businesses. For a business to be in the category of MSE, it must have an annual turnover of GHC 90,000 or less and also, have a workforce of less than 30. MSEs with workforce of less than size (6) are Micro-enterprises and from 6 to 29 workforces are Small-enterprises.

Business structure: This variable is measured as binary; 1 representing MSE operates in movable/mobile structure while zero indicates that the firm operates in immovable structure. The mobile businesses are usually hawkers, a situation which is predominant in Ghana (especially in urban areas). Movable firms are likely to evade tax compared to a non-movable firm.

Legal form of the business: This variable measures the registered business structure of the firm. The firms are registered as sole proprietorship, partnership, and companies. Being sole proprietor was assigned zero, partnership was assigned 1 while firms that registered as companies were assigned 2. The sole proprietorship was used as the base for the categorization in the estimation. Smulders (2013) and Blaufus et al. (2011) argue that sole proprietors typically spend more time internally when dealing with tax-related activities compared to partnership business and companies.

Service used: Firms may either hire external person or use their own staff to prepare their tax activities. This variable has three categories: internal only, external only, and both internal and external. Internal service is used as the base category. External services include hiring of personnel from outside the organization whereas internal service includes the use of the firm’s own accounting staff in preparing their tax schedules and returns.

Tax audit: Chattopadhyay et al. (2002) indicate that tax scrutiny (tax audit in this study) raises the firm’s compliance burden but reduces time compliance cost. In view of this, tax audit is captured on a categorical scale with a code of 0 if the firm is audited and 1 otherwise.

Tax knowledge: Tax knowledge is an index computed from a set of question. It is computed as:

\begin{matrix} Tax knowledge score \\ = \frac{Number of each correct questions}{Number of questions (25)} \times 100 % \end{matrix}

This index ranges from 0% to 100%. Higher values imply higher tax knowledge. Higher tax knowledge is expected to reduce compliance burden and hence the tax gap because a well-informed taxpayer is likely to understand the tax system and apply them in their tax-related activities.

Number of taxes paid: This is the total number of taxes MSEs are expected to pay. This includes the CIT, PIT, VAT, etc. It is measured on a continuous scale. The number of taxes that these MSEs pay could influence the amount of time allocated to tax-related activities. Distance to tax office: these measures how long (minutes) it takes the MSE to drive to the nearest tax office. Using the Google app, the distance between the MSEs and the tax office was captured. Sex/gender of the manager: Zero was assigned for male and 1 for female.

Results and Discussions

In order to provide deeper insight, we first provide a discussion of the descriptive results, after which the econometric results are presented and discussed.

Characteristics of Micro and Small Enterprises

Following from the application of the prevailing tax rates (as at 31/12/2020) on the firms’ turnover, the tax gap was computed. Table 2 presents the summary statistics on the 485 MSEs. The tax gap as indicated, is the difference between potentials and actual tax for the enterprises.

Table 2.

Summary Statistics.

Variable	Obs.	Mean	SD	Minimum	Maximum
Compliance burden	485	109.6 (19.2)	178.6 (31.3)	5 (0.9)	955 (167.5)
No. year	485	11.4	7.5	1	50
Employees	485	6.3	5.5	1	29
No. taxes	485	2.4	1.2	1	4
TaxGap	485	109.20 (19.2)	491.9 (86.3)	−809.4 (−142)	3,498 (613.7)
Turnover	485	10,577.7 (1,855.7)	12,157.4 (2,132.9)	400 (70.2)	80,000 (14,035.1)

Source. Authors estimates (2020).

Note. Obs. represents observation, Dollar ($) equivalence in parenthesis based on the exchange rate 5.7 as at 31/12/2019.

It has a minimum amount of negative GHC 809.4 (US$142) and a maximum of positive GHC 3,498 (US$613.7). The negative GHC 809.4 (US$142) tax gap indicates an over-payment of tax while the positive GHC 3,498 indicates an under-payment of tax (revenue loss). However, the average amount of tax gap per MSEs is about GHC 109.2 (US$19.2). This means that the GRA can generate an additional GHC 109.2 (US$19.2) more if it has an efficient tax collection system with a 100% compliance level. This estimate varies largely compared to the annual aggregated tax gap of US$14,9087.67 reported by Danquah and Osei-Assibey (2018). The average turnover for these MSEs is GHC 10,577.7 (US$1,855.7) . They have a minimum turnover of GHC 400 (US$70.2) and a maximum of GHC 80,000. The average amount of compliance burden is GHC 109.6 (US$19.2). This suggests that, on average, MSEs spend close to GHC 109.6 (US$19.2) per month when complying with tax laws. The lowest compliance burden is GHC 5 (US$ 0.9) whereas the maximum is GHC 955 (US$167.5). The compliance burden estimate is consistent with the tax incidental cost component of the tax compliance cost estimated by Bruce-Twum and Schutte (2021). This study found the average incidental cost of GHC 121 (US$21.2). Again, Eichfelder, Evers, Gläser, Heinemann, Jenzen, Kalb, and Misch (2010) as cited by Eichfelder and Vaillancourt (2014, p. 8) found that the average compliance burden associated with Germans’ thin-capitalization rules is 18 Euro. Converting the 18 Euro to USD, the Germans’ thin-capitalization rules spent about US$20 (using 1.12 as the exchange rate as at 2010).

In Table 3, it is observed that majority (66.8%) of small taxpayers perceive it as more complex. Almost all the enterprises (95.57%) are owned by Ghanaians. Only about 3.02% of them are owned by non-Ghanaians with just 1.41% owned by both Ghanaians and foreigners. This is consistent with the Integrated Business Establishment Survey report by GSS (2015). The report indicates that firms owned by only Ghanaians constitute 98.6%, non-Ghanaian 1.2%, and both Ghanaians and Non-Ghanaian own businesses constitute 0.2%. Again, the number of movable firms constitutes just about 18.51% while 81.29% for non-movable businesses, showing that the data is skewed toward non-movable firms.

Table 3.

Descriptive Statistics of the MSEs.

Variables	Categories	Frequencies	Percentages
Complexity	Less complex	161	33.2
Complexity	More complex	324	66.8
Nationality	Ghanaian	475	95.57
	Non-Ghanaian	15	3.02
	Ghanaian and non-Ghanaian	7	1.41
Structure	Movable	93	18.71
Structure	Non-movable	404	81.29
M_Edu	At most JHS/Middle	92	18.51
M_Edu	At least secondary	405	81.49
Legal_Form	Sole proprietorship	288	57.95
	Partnership	55	11.07
	Company	153	30.78
Location	Urban	467	93.96
Location	Rural	30	6.04
MSE	Micro	310	63.92
MSE	Small	175	36.08

Source. Authors’ estimates (2020).

Additionally, Table 3 shows that micro-enterprises constitute 63.92% while small enterprises constitute 36.08%. This is consistent with the GSS (2015) report which indicated that micro businesses constitute 79.76% of all establishments and 18.38% for small businesses. This indicates that the data used in this study mimic a nationwide data on firms.

Potential Tax, Actual Tax, and Tax Gap by Firms’ Characteristics and Tax Handles

Tax gap estimates could vary across different firm characteristics such as sectors, location, type of structure, legal form of the business, and for the different classification of businesses. The results are presented on Table 4. The average potential tax and actual tax for the industrial sector is GHC 3,084.3 (US$541.1) and GHC 2,235.52 (US$392.2) respectively. This shows a significant difference (tax gap) of about GHC 848.78 (US$148.9). The service sector recorded an annual potential tax of GHC 3,797.10 (US$666.2) and the actual tax of GHC 2,659.17(US$466.5). Therefore, the tax gap, is GHC 1,137.93 (US$199.6). This shows that there exists substantial revenue loss in the service sector compared to the industrial sector. The high amount of tax revenue collected in the service reiterates the dominance of the service sector’s contribution to the economy compared to the industrial sector (GSS, 2015).

Table 4.

Potential Tax, Actual Tax, and Tax Gap by Firm Characteristics.

Variables	Categories	Potential tax	Actual tax	Tax gap (GHC)
Sectors	Industry	3,084.30 (541.1)	2,235.52 (392.2)	848.78 (148.9)
Sectors	Service	3,797.10 (666.2)	2,659.17 (466.5)	1,137.93 (199.6)
Location	Urban	3,744.02 (656.8)	2,596.81 (455.6)	1,147.21 (201.3)
Location	Rural	2,772.63 (486.4)	2,517.60 (441.7)	255.03 (44.7)
Type of structure	Movable	3,798.00 (666.3)	2,539.09 (445.5)	1,258.91 (220.8)
Type of structure	Non-Movable	3,196.68 (560.8)	2,817.52 (494.3)	379.16 (66.5)
Legal form	Sole-Proprietorship	2,352.49 (412.7)	2,541.65(445.9)	−189.16 (−33.2)
	Partnership	4,012.05 (703.9)	2,673.02 (469.0)	1,339.03 (234.9)
	Companies	6,024.52 (1,056.9)	2,655.23 (465.8)	3,369.29 (591.1)
MSE	Micro	2,482.08(435.5)	2,517.47 (441.7)	−35.38 (−6.2)
MSE	Small	5,812.94 (1,019.8)	2,723.78 (477.9)	3,089.16 (542.0)
Regions	Greater-Accra	3,163.32 (555.0)	2,545.72 (446.6)	617.60 (108.4)
	Ashanti region	3,668.32 (643.6)	2,567.12 (450.4)	1,101.20 (193.2)
	Northern	4,018.22 (705.0)	2,635.70 (462.4)	1,382.52 (242.5)
Tax handles	Corporate	6,024.52 (1,056.9)	2,655.23 (465.8)	3,369.29 (591.1)
	Income	3,692.81 (647.9)	2,595.99 (455.4)	1,096.82 (192.4)
	VAT	4,138.54 (726.1)	2,632.68 (461.9)	1,505.86 (264)

Source. Authors’ estimates (2020).The values in parentheses are the US dollar equivalent of the Ghana cedis.

Additionally, the average potential tax and actual tax for the MSEs in the urban areas are GHC 3,744.02 (US$656.8) and GHC 2,596.81 (US$455.6) respectively. This creates a tax gap of GHC 1,147.21 (US$201.3) per annum, Enterprises located in the rural areas record an annual average potential tax of GHC 2,772.63 (US$486.4), actual tax of GHC 2,517.60 (US$441.7), and a tax gap of GHC 255.03 (US$44.7). It is evident that the difference in the tax gap for enterprises located in the urban areas is greater than that of the rural areas. This difference may be attributed to tax avoidance in urban areas. Owners and managers of urban enterprises may be more abreast with new developments relating to tax planning and management which may be lacking in enterprises located in rural areas. They are therefore able to exploit the loopholes in the tax system to their advantage.

Another important characteristic of enterprises is the structure in which they operate. Enterprises with movable structures have higher tax gap (GHC 1258.91 [US$220.8]) relative to non-movable enterprises (GHC 379.16 [US$66.5]). Meanwhile, the differences in the tax gap for enterprises with movable structure and unmovable structures can be explained in the context of evasion and availability of the enterprises to be located at the same location when tax officials go round to collect tax. Thus, firms with movable structures have a higher probability of evading tax anytime tax officials go around to collect taxes compared to enterprises that are non-movable.

Furthermore, significant variation in tax gap was observed among sole proprietors, partnership, and companies. The average potential tax for sole-proprietorship is GHC 2,352.49 (US$412.7), GHC 4,012.05 (US$703.9) for partnership, and GHC 6,024.52 (US$1,056.9) for companies. Their corresponding actual tax paid is GHC 2,541.65 (US$445.9), GHC 2,673.02 (US$469.0), and GHC 2,655.23 (US$465.8). Thus, there is a higher (GHC 3,369.29 [US$591.1]) tax gap for companies compared to partnerships (GHC 1,339.03 [US$234.9]) and sole-proprietorship (-GHC 189.14) [US$33.2]). The difference in the tax gap for companies, partnerships, and sole-proprietorship may be attributed to the varying degree of tax obligations these forms of businesses ought to honor. For instance, companies pay corporate tax while partnerships and sole-proprietors pay income tax. These types of obligations differ in magnitude. The World Bank (2020) found that although large firms (firms with 50 or more employees) constituted only about 8.8% of total corporate establishments in Ghana, they accounted for about 52.5% of total potential tax revenues. This means the tax gap is likely to be high among the large firms registered as companies

Table 4 shows an annual average potential tax of GHC 2,482.08 (US$435.5) and the actual tax of GHC 2,517.47 (US$441.7), giving a tax gap of negative GHC 35.381 (US$6.2) for the micro-enterprises. The small enterprise taxpayers on the other hand have an annual potential tax of GHC 5,812.94 (US$1,019.8), actual tax of GHC 2,723.78 (US$477.9), and a tax gap of GHC 3,089.16 (US$542.0). This suggests that just like the sole-proprietorship, micro-enterprises are overburdened with tax payment. This is in consonance with the 2018 Ease of Doing Business report which indicates that SMEs find it difficult to pay tax.

In terms of political regions, we found the estimated potential tax of GHC 3,163 (US$555.0) for firms in Greater Accra, GHC 3,668.32 (US$643.6) for firms in Ashanti region, and GHC4,018.22 (US$705.0) for the firms selected from the Northern region. With actual tax collected amounting to GHC 2,546 (US$446.6), GHC 2,567.12 (US$450.4), and GHC 2,636.70 (US$462.4) for Greater Accra, Ashanti, and Northern regions respectively. The study suggests that there is more revenue potential in the Northern region compared to the Greater Accra and Ashanti regions. Table 4 further shows that there is more revenue potential among companies than the other businesses. This is consistent with World Bank (2020) tax gap analysis in Ghana and the World Bank (2018)‘s Ease of Doing Business report. Finally, for the different tax handles, the estimated tax gap was highest for corporate income tax, amounting to GHC 3,369.29 (US$591.1), followed by VAT (GHC 1,505.86 [US$264]) and the least was income tax which was GHC 1,096.82 (US$192.4). This is consistent with the study by World Bank (2020) where corporate tax recorded the highest tax gap (85.6%)

Compliance Burden and Services Used in Honoring Tax Obligation

This section presents the estimates of the compliance burden for the different kinds of services used in honoring tax obligations. The blue color indicates the percentage of firms that use a particular service to honor their tax obligation while the red color shows the average estimated compliance burden measured in Ghana cedis and its dollar equivalent.

Figure 1 indicates that half (55.67%) of the enterprises make use of internal staff to prepare their tax activities. However, 31.75% employs external service while 12.58% employs both internal and external services. The MSEs that utilize internal staff incur GHC 91.29 (US$16.0) compliance burden, GHC 100.50 (US$17.6) for external service, and GHC 213.39 (US$37.4) for those that employed both internal and external services. These categories are very important when considering compliance burden because outsourcing the service of someone that prepare tax activities comes at a cost (Dunlop & Radaelli, 2016). Smulders (2013) explains that outsourcing tax service does not only increase the number of hours spent on internal VAT compliance but also leads to duplication of work and effort since the professionals have to check what the firms have been doing before making the necessary corrections and suggestions.

Figure 1.

Preparation, payment and filing of tax returns.

Differences in Tax Gap and Compliance Burden by MSEs

An important information for policy is the differences in tax gaps and compliance burden among MSEs. A t-test was conducted to examine the difference in tax gap and compliance burden between micro and small businesses. The results provided in Table 5 show that the difference in the tax gap for micro businesses and small businesses is negative GHC 312.45 (US$54.8). The p-value (.000) associated with the t-statistic (−5.756) indicates statistically significant difference at .01 alpha level. This suggests that the difference in tax gap between micro enterprises and small enterprises actually exist.

Table 5.

Differences in Tax Gap and Compliance Burden.

Group	Observation	Tax gap		Compliance burden
Group	Observation	Mean	SE	Mean	SE
Micro	310	−3.54 (−0.6)	159.156	79.622 (14)	8.440
Small	175	308.92 (54.2)	518.631	162.617 (28.5)	16.059
Combined	485	109.20 (19.16)	223.338	109.569 (19.1)	8.112
Difference	135	−312.45 (54.8)	542.502	82.994 (14.6)	18.142
T-statistics		=−5.760		=−4.575
Prob (\|T\| > \|t\|)		=0.000		=0.000
Satterthwaite’s df		=207.28		=271.73
Ho: Difference		=0		=0
Ha: Difference		≠0		≠0

Source. Authors’ estimate (2020).

Moreover, the difference in compliance burden of micro businesses and small businesses is GHC 82.99 (US$14.6). The p-value of .000 associated with the t-statistic of −4.575 indicates that the difference is statistically significant, suggesting that there exists substantial difference in compliance burden between micro and small enterprises.

Effect of Compliance Burden on Tax Gap Among MSEs

The econometric results on the effect of compliance burden and other factors on the tax gap among micro and small businesses are presented in Table 6. The overall model shows a p-value of .000 and F-statistics of 5.190, indicating that all the independent variables jointly and reliably predict the tax gap estimate with a predictive power of 12.5%. The coefficient of managers’ secondary education (Edu) is −0.080. Thus, compared to managers with at most basic education, the findings suggest that attaining at least secondary education is associated with 8% reduction in tax gap. Intuitively, individuals who are more educated are more likely to appreciate the importance of honoring tax obligation Kasipillai et al., 2003). We further found that compared to micro-enterprises, being a small enterprise is associated with higher tax gap of 24.9%. This result emphasizes the statistical difference in the magnitude of tax gap between micro-enterprises and small enterprises reported in Table 6. Increasing compliance burden by 1% increases tax gap by 7.5% suggesting that increases in compliance burden reduces tax compliance This is consistent with the findings of Tee et al. (2016) that existing tax policies have adverse impact on SMEs and that affect revenue mobilization.

Table 6.

Effect of Compliance Burden on Tax Gap.

lnTaxGap	Overall model	Regional estimates			Different tax handles
lnTaxGap	Overall model	Greater Accra	Ashanti region	Northern Accra	Corporateincome tax	Personalincome tax	Value added tax
Sex of manager (base = males)
Females	0.010 [0.046]	−0.063 [0.066]	−0.092 [0.082]	0.077 [0.078]	0.041 [0.155]	0.029 [0.046]	−0.006 [0.064]
Managers education (base = at most basic)
At least secondary	−0.080** [0.032]	−0.081 [0.054]	0.027 [0.100]	−0.148*** [0.044]	−0.268*** [0.098]	−0.069** [0.031]	−0.054 [0.054]
Location (base = rural)
Urban	−0.057 [0.080]	0.116 [0.163]	−0.097 [0.120]	−0.281** [0.120]	0.225 [0.231]	−0.160*** [0.053]	−0.232*** [0.060]
MSE (base = micro)
Small	0.249*** [0.051]	0.163 [0.121]	0.311*** [0.094]	0.243*** [0.078]	0.286*** [0.103]	0.259*** [0.051]	0.256*** [0.068]
Age Firms	−0.003 [0.002]	−0.003 [0.003]	0.005 [0.006]	−0.007** [0.004]	−0.008 [0.006]	−0.002 [0.002]	−0.001 [0.004]
lnBurden	0.075*** [0.020]	0.100*** [0.026]	0.063 [0.045]	0.109*** [0.035]	0.0483 [0.058]	0.078*** [0.020]	0.059** [0.0271]
Complexity (base = less complex)
Complex	0.020 [0.044]	−0.061 [0.094]	−0.024 [0.160]	0.081 [0.066]	0.047 [0.132]	0.016 [0.046]	0.027 [0.073]
Sectors (base= industry)
Service	0.008 [0.054]	0.081 [0.053]	0.009 [0.117]	−0.145 [0.121]	0.031 [0.131]	−0.007 [0.056]	−0.021 [0.097]
Constant	0.261*** [0.091]	0.379*** [0.117]	−0.342 [0.242]	−0.296* [0.169]	−0.008 [0.273]	0.272*** [0.093]	−0.172 [0.130]
F-Statistics	5.190	3.390	2.520	3.320	2.18	5.80	4.34
Prob > F	0.000	0.002	0.015	0.001	0.032	0.000	0.000
R-squared	0.127	0.136	0.136	0.184	0.067	0.141	0.114
Observation	485	138	223	124	167	465	255

Source. Authors estimation (2020).

Robust standard errors in parenthesis: ***p < .01, **p < .05, *p < .1.

Regional Analysis

Columns 3 to 8 of Table 6 present the effects of compliance burden on tax gap in the various regions (Greater Accra, Ashanti, and Northern region) and among different tax handles (CIT, PIT, and VAT). Specifically, the results of the regional analyses are presented in the second, third, and fourth columns of Table 7 while columns 5, 6, and 7 provide results for the three different tax handles. The third column of Table 7 shows the effects of compliance burden on tax gap in Greater Accra. The coefficient of the compliance burden is 0.1 and is statistically significant at .05 alpha level, suggesting that 1% increase in the amount of compliance burden increases the tax gap by 10% ceteris paribus. In Ashanti region, the results show that the classification of businesses has a significant effect on tax gap. The coefficient of ME is 0.311 and it is significant at .05 alpha level, implying that compared to micro-enterprises, small-enterprises record higher tax gap of 31.1%. Interestingly, the compliance burden is statistically insignificant in explaining tax gap in the Ashanti region because the p-value associated with its coefficient, .063, is greater than the .05 alpha level. This could be due to the multiplicity of business outlets of the same firm in the Ashanti region. In the Ashanti region, it is common to find that one person owning multiple businesses all of which render accounts at only one outlet. The effect of the burden is spread among all the other outlets, making its effect on the tax gap insignificant.

Table 7.

Correlates of Compliance Burden.

Lnburden	Coef.	Robust standard error
Complexity (base = less complex)
Complex	0.047	0.097
Audit (base = yes)
No	0.398***	0.091
NTaxes	0.185***	0.044
BusinessSize	0.033***	0.008
Legal_Form (base = proprietorship)
Partnership	0.127	0.134
Company	0.504***	0.115
Age	0.005	0.006
ServiceUsed (base = internal service)
External service	0.202**	0.087
Internal and external service	0.699***	0.119
Tax knowledge	0.021***	0.003
M_Edu (base = at most basic)
At least secondary	−0.688*	0.408
Distance	−0.007**	0.003
Constant	1.844***	0.239
F-statistics	24.52
Prob > F	0.000
R-squared	0.384
Number of observations	485

Source. Authors estimation (2020).

***

p < .01, **p < .05, *p < .1.

Column 5 of Table 6 shows the effect of compliance burden on tax gap in the Northern region. The model shows a statistically significant negative association between managers with at least secondary education and tax gap. It indicates that the tax gap among businesses managed by at least senior high school graduates reduces by 14.8% compared to businesses managed by at most basic school graduates. Again, businesses located in urban areas have a negative and statistically significant coefficient of 0.281, implying that businesses located in urban areas have a reduced tax gap of 28.1% compared to enterprises located in rural areas. This is attributable to the fact that more STOs are located in the urban areas compared to the rural areas and therefore. This makes enterprises located in the urban areas undergo regular monitoring and audit from tax officials compared to those in rural areas. This makes the urban businesses comply with tax laws.

Small businesses have a coefficient of 0.243 suggesting that the tax gap among small enterprises is 24.3% higher compared to tax gap in micro businesses. The number of years the enterprise has registered with the GRA is an equally important determinant of the tax gap as it was found to be statistically significant in reducing the tax gap. It has a coefficient of 0.007, implying that an additional increase in the years the business has registered with GRA reduces the tax gap by 0.7%. This is consistent with Danquah and Osei-Assibey (2018) that firms’ age negatively affects tax gap.

The coefficient of compliance burden in the Northern region is 0.109 and it is statistically significant. This implies that a percentage increase in the compliance burden increases tax gap among businesses in the Northern region by 10.9%. On a whole, the study found that compliance burden significantly affects tax gap in the two out of three regions (Greater Accra and the Northern regions). The coefficient of the Ashanti region was however, not statistically significant.

Different Tax Handles

The fifth, sixth, and seventh columns in Table 6 present the result for the different tax handles. Education level of the manager has a statistically significant effect on the tax gap. Compared to businesses managed by at most basic education graduates, tax gap is lower for businesses whose managers have at least senior high (secondary) education (26.8% and 6.9% respectively). However, the effect of managers’ education on VAT gap is not found to be statistically significant. Similarly, there is 16% and 23.2% reduction in income tax and VAT gap respectively for enterprises located in urban areas compared to enterprises located in rural areas. This reduction could be attributed to the visible presence of STOs in the urban areas compared to rural areas hence. Therefore, it is more likely for the tax officials to supervise and ensure tax payment in those areas than in rural areas. Additionally, small enterprises are associated with a higher corporate tax gap (28.6%) than income tax gap (25.9%) and VAT gap (25.6%) compared to micro-enterprises.

Furthermore, the coefficient of compliance burden is not statistically significant, suggesting that it does not affect the corporate tax gap. In contrast, the coefficient is found to be statistically significant in explaining income tax gap and VAT gap. Specifically, a percentage increase in the compliance burden increases the income tax gap by 7.8% and the VAT gap by 5.9%. Thus, the effect of compliance burden on the income tax gap is greater relative to the effects on the corporate tax and VAT gaps.

Correlates of Compliance burden among MSEs

Owing to the fact that tax gap in the preceding models (in Table 6) are significantly affected by compliance burden, we extend the scope of the study to investigate the factors that actually determine compliance burden. Table 7 presents the results of the correlates of compliance burden. An omitted variable test (Ramsey RESET test presented in Appendix B) indicates that the model has no omitted variable. Again, a linktest to determine the goodness of fit (see Appendix B) indicates that the model is correctly specified. Furthermore, the multicollinearity test (reported in Appendix C) gives the mean variance inflation factor of 1.209 indicating no collinearity among the variables used. Finally, the Breusch-Pagan/Cook-Weisberg tests for heteroskedasticity shows the presence of heteroscedasticity. The heteroskedasticity-robust standard error approach was therefore employed to correct for the problem (the test result presented in Appendix D).

The R-Squared (38.4%) shows the proportion of variation in compliance burden explained by the set of independent variables. Thus, it shows the overall measure of strength of association. With an F-statistics of 24.52 and Prob > F of 0.000, all the independent variables jointly and reliably explain the compliance burden.

Table 7 indicates that complexity about the tax system (Complexity) is not statistically significant in explaining compliance burden. The coefficient of enterprises that are not audited (Audit) is 0.398 and is statistically significant at .05 alpha level meaning enterprises that are not audited spend more (39.8%) when complying with Ghana tax laws than firms that are audited. Firms that are not audited may find it difficult to seek advice from the revenue collecting agency (i.e., the GRA), hence, they will resort to hiring the service of an expert which increases their compliance burden. In the USA, Chattopadhyay et al. (2002) found that scrutiny (a form of tax audit) increased compliance costs by most measures.

The coefficient of number of taxes (NTaxes) was found to be statistically significant and positively related to compliance (18.5%). This indicates that an additional tax paid by an enterprise increases the firm’s compliance burden by 18.5%. This makes sense because for each tax paid, the enterprise has to make the necessary preparation toward its payment as well as filing of returns, all of which come with cost. The implication of this is that for governments to realize the full potential in tax revenue mobilization, the need to synchronize taxes is key. Too many taxes may be burdensome, cumbersome, and cost ineffective.

The size of the business (Business_Size) has a coefficient of 0.033 and is statistically significant at .05 alpha level. This means that an increase in the size of a business increases the compliance burden by 3.3%. As larger firms engage in more business activities they are more likely to spend more time complying with tax obligations compared with smaller businesses (Smulders, 2013).

Furthermore, the Table 7 reveals that compliance burden for companies is 0.504, indicating a 50.4% higher than that of sole proprietorship. Interestingly, since the sole proprietorship spends more time on tax compliance activities, one would have expected a higher compliance burden (Smulders, 2013) compared to Companies. It is argued that companies pay more taxes compared to sole-proprietorship, therefore companies are more likely to outsource experts for tax preparations hence increase their overall burden. Again, companies are mostly large and engage in more sophisticated businesses which come with additional tax obligations compared to sole-proprietorship.

In terms of type of service used by the MSEs in preparing, paying, and filing tax returns, the results show that sourcing for external services has a coefficient of 0.202 which is statistically significant at .05 alpha level. This means that enterprises that use only external service in dealing with tax-related activities have compliance burden increased by 20.2% more than enterprises that use only internal services. Similarly, we found that the coefficient of both internal and external is 0.699 and is statistically significant at .05 alpha level. This suggests that enterprises that use both internal and external services have 69.9% increase in their compliance burden compared to enterprises that used only internal service. This is consistent with our expectation because outsourcing (external services) comes with a high cost that adds up to the firm’s overall compliance burden. The implication of this result is that firms that are able to prepare their tax obligations activities in house stand the chance of reducing compliance costs.

Tax knowledge has a coefficient of 0.021 implying that one% increase in tax knowledge score increases the compliance burden by 2.1%. Though this contradicts with our a priori expectation, it is argued that an increased tax knowledge is likely to increase tax compliance (Saad, 2014) hence compliance burden. The coefficient of managers’ level of education (M_Edu) is 0.688. This means that MSEs managed by individuals with at least secondary education reduces the compliance burden of the enterprise by 68.8% compared to MSEs managed by individuals with at most basic education. This is significant at .10 alpha level and this suggests that educated managers facilitate the tax administration process and engage in more tax planning and management compared to managers with at most basic education, hence reduce the firm’s compliance burden (Danquah & Osei-Assibey, 2018; Saad, 2014). It must be emphasized that knowledge about tax is a form of education and as managers are educated, they are likely to obtain relevant and adequate knowledge about tax issues. Education also acts as efficiency parameter, enabling taxpayers more efficient in the preparation of tax obligation activities.

Finally, the coefficient of distance to tax office (Distance) of 0.007 and is found to be statistically significant. This implies that additional minute spent in traveling to the tax office significantly reduces compliance burden by 0.7%. Firms that are far away from the tax office find it difficult to seek tax advice from the GRA officials, hence are likely to evade or avoid tax payment which then reduces their overall compliance. Appiah (2016) shows that the distance between the tax office and businesses premises in the Kumasi Metropolis of Ghana poses a challenge to taxpaying firms, leading to non-compliance as a result of the cost associated with moving from one’s business to the tax office.

Robustness checks

For the estimates to be efficient and consistent, the error term (ε_i) in equation (1) must be identically and normally distributed such that the ε_i should be homoscedastic as well as satisfy the CLRM axioms. To test for this, the Ramsey RESET test for omitted variable and multicollinearity test were conducted (the results are reported in the Appendix A). The multicollinearity test result show that there is little or no correlation among the independent variables used. However, the study failed to satisfy the Breusch-Pagan/Cook-Weisberg tests for heteroscedasticity, hence, the heteroscedasticity-robust standard error approach was employed (test reported in the Appendix D).

Conclusions

The study examined compliance burden and tax gap among micro and small enterprises in Ghana. We found that compliance burden of small enterprises is twice as that of micro-enterprises suggesting that micro-enterprises overpay tax while small-enterprises underpay tax. The average amount of tax gap per MSE is about GHC 109.2 (US$19.2). Moreover, the MSEs spend between GHC 5 (US$ 0.9) and GHC 955 (US$167.5) per month when complying with tax laws. Higher tax gap is recorded among firms that operate in the service sector (US$199.6) compared to the industrial sector ($148.9). In terms of location, the studies notes that the tax gap for enterprises located in the urban areas is greater than that of the rural areas, suggesting the potential untapped revenue in urban areas. Moreover, higher tax gap (US$220.8) is found among movable enterprises compared to non-movable businesses (US$66.5). The study further reveals that there is a higher tax gap among MSEs registered as companies compared to partnership and sole-proprietorship. We further found the tax gap to be regressive as micro-enterprises are burdened with higher tax payment compared to small enterprises as tax gap among small enterprises is 24.3% higher compare to tax gap in micro businesses. Finally, tax gap was highest for corporate income tax followed by VAT with the least being income tax.

Another revealing conclusion from the study is that compliance burden has a deteriorating effect on tax gap for major tax handles (VAT and income tax) which are at the center of domestic revenue mobilization in Ghana. It is also heart-warming to note that, income tax gap and VAT gap will be reduced considerably if managers of MSEs have at least senior high (secondary) education. This gives credence to the policy of free high school education currently underway in Ghana. The number of taxes, the perceived complexity of the tax system, the engagement of external services for tax activities, and distance to tax office were all found to increase compliance burden. In terms of policy, the results show that while the efforts to reduce the tax gap and strengthen domestic resource mobilization should be sustained for the entire tax system, the GRA should pay more attention to firms operating in the service sector, urban areas as well as movable and small enterprises. Particular attention must also be paid to all the tax handles but more importantly corporate tax since it records the highest tax gap. It is also recommended that the current policy of making senior high school education accessible to all should be sustained as it has the tendency to reduce compliance burden and tax gap. The key limitation of the study is that the sample involved MSEs registered with the Ghana Revenue Authority. Admittedly, unregistered MSEs constitute a majority in the informal sector and not capturing this category of taxpayers provides half of the story about tax mobilization of small tax payers in the country. It will be illuminating to conduct a study that explores the tax gap among unregistered MSEs to provide better insight into informal revenue potentials of the country.

Footnotes

Appendix A: Omitted Variable Test for the Overall Model

Ramsey RESET test using powers of the fitted values of lnTax_Gap

Ho: Model has no omitted variables

F(3, 473) = 1.26

Prob > F = 0.2874

Appendix D: Heteroskedasticity Test for the Overall Model

Breusch-Pagan/Cook-Weisberg test for heteroskedasticity

Ho: Constant variance

Variables: Fitted values of lnTax_Gap

χ²(1) = 255.11

Prob > χ² = 0.0000

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Directorate of Research, Innovation and Consultancy (DRIC), University of Cape Coast, Ghana.

ORCID iDs

James Atta Peprah

Francis Kwaw Andoh

Eric Atsu Avorkpo

References

Agyapong

(2010). Micro, small and medium enterprises’ activities, income level and poverty reduction in Ghana – a synthesis of related literature. International Journal of E-Business Management, 5(12), 196.

Akinboade

O. A.

(2015). Correlates of tax compliance of small and medium size businesses in Cameroon. International Research Journal, 13(4), 389–413.

Allingham

M. G.

Sandmo

(1972). Income tax evasion: A theoretical analysis. Journal of Public Economics, 1(3–4), 323–338.

Alm

Martinez-Vazquez

Schneider

(2004). ‘Sizing’ the problem of the hard-to-tax. Contributions to Economic Analysis, 268, 11–75.

Andoh

F. K.

(2017). Taxable capacity and effort of Ghana’s value-added tax. African Review of Economics and Finance, 9(2), 255–284.

Anuwa-Amarh

E. T.

(2015). Understanding the urban informal economy in Ghana: A survey report. Friedrich-Ebert-Stiftung Ghana. Retrieved January 29, 2017, from http://www.fesghana.org/uploads/PDF/Understanding%20the%20Urban%20Informal%20Economy%20in%20Ghana_A%20Survey%20Report.pdf

Appiah

(2016). Challenges of paying tax: A case study of small and medium scale enterprises in the Kumasi metropolis [Doctoral dissertation, Kwame Nkrumah University of Science and Technology].

Atawodi

O. W.

Ojeka

S. A.

(2012). Factors that affect tax compliance among small and medium enterprises (SMEs) in North Central Nigeria. International Journal of E-Business Management, 7(12), 87–96. https://doi.org/10.5539/ijbm.v7n12p87

Ayee

J. R. A.

Joshi

(2008). Associational taxation: A pathway into the informal sector?. Cambridge University Press.

10.

Becker

G. S.

(1968). Crime and punishment: An economic approach. In Fielding

N. G.

Clarke

Witt

(eds.), The economic dimensions of crime (pp. 13–68). Palgrave Macmillan.

11.

Blaufus

Eichfelder

Hundsdoerfer

(2011). The hidden burden of the income tax. (Discussion Paper No. 2011/6), Freie, Berlin.

12.

Bogetić

Ž.

Naeher

Narayanan

. (2022). Measuring untapped revenue potential in developing countries: Crossc-country frontier and panel data analysis. Journal of Public Finance and Public Choice, 1–39.

13.

Bruce-Twum

Schutte

(2021). Tax compliance cost of SMEs in Ghana. Journal of Accounting Finance and Auditing Studies, 7, 1–22.

14.

Carsamer

Abbam

(2020). Religion and tax compliance among SMEs in Ghana. Journal of Financial Crime. Advance online publication. https://doi.org/10.1108/jfc-01-2020-0007

15.

Chattopadhyay

Das-Gupta

Bhatnagar

D. B.

Mohanty

Mukhopadhyay

Surendra

P. S.

Prakash Singh

(2002). The compliance cost of the personal income Tax and its determinants (p. 199). National Institute of Public Finance and Policy. http://unpan1.un.org/intradoc/groups/public/documents/apcity/unpan014795.pdf

16.

Danquah

Osei-Assibey

(2018). The extent and determinants of tax gap in the informal sector: Evidence from Ghana. Journal of International Development, 30(6), 992–1005. https://doi.org/10.1002/jid.3361

17.

Danquah

Senahey

(2020). Tax reforms and revenue mobilization in Ghana. Journal of Economic Studies, 43(4), 522–534.

18.

Dare

du Plessis

Jansen

(2019). Tax revenue mobilisation: Estimates of South Africa’s personal income tax gap. South African Journal of Economic and Management Sciences, 22(1), 1–8.

19.

De Mooij

Nazara

Toro

. (2018). Implementing a medium-term revenue strategy (pp. 109–140). International Monetary Fund. https://books.google.com/books?hl=en&lr=&id=wK4ZEAAAQBAJ&oi=fnd&pg=PA109&dq=profit+as+a+tax+base+in+indonesia&ots=o17jAiXCtF&sig=7alJtJ5AiurxufpvXN-81M59EYg%0Ahttps://www.elibrary.imf.org/view/book/9781484337141/ch006.xml

20.

Dunlop

C. A.

Radaelli

C. M.

(2016). The politics and economics of regulatory impact assessment. In: C. A. Dunlop & C. M. Radaelli (Eds.), Handbook of regulatory impact assessment (pp. 3–18). Edward Elgar Publishing.

21.

Eichfelder

Vaillancourt

(2014). Tax compliance costs: A review of cost burdens and cost structures. SSRN Electronic Journal, 178. https://doi.org/10.2139/ssrn.2535664

22.

Falkinger

Walther

(1991). Rewards versus penalties: On a new policy against tax evasion. Public Finance Quarterly, 19(1), 67–79.

23.

Fishbein

Jaccard

Davidson

A. R.

Ajzen

Loken

(1980). Predicting and understanding family planning behaviours. In: Ajzen

Fishbein

(eds,). Understanding attitudes and predicting social behaviour. Prentice Hall.

24.

Gemmell

Hasseldine

(2012). The tax gap: A methodological review. Emerald Group Publishing Limited.

25.

Ghana Statistical Service. (2015). Summary report: Integrated Business Establishment Survey (IBES). Ghana Statistical Service.

26.

Government of Ghana. (2019). Ghana beyond aid charter and strategy document. Government of Ghana. https://scholar.google.com/scholar?q=Government+of+Ghana.+2019.+%E2%80%98Ghana+Beyond+Aid+Charter+and+Strategy+Document%E2%80%99.+Accra:+Government+of+Ghana

27.

Helhel

Ahmed

(2014). Factors affecting tax attitudes and tax compliance : A survey study in Yemen. European Journal of Business and Management Online, 6(22), 48–58.

28.

International Monetary Fund. (2021, July 20). IMF Executive Board concludes 2021 Article IV consultation with Ghana. Press release. https://www.imf.org/en/News/Articles/2021/07/20/pr21221-ghana-imf-executive-board-concludes-2021-article-iv-consultation

29.

Ishola

Bello

A. O.

Raheed

L. O.

(2020). Taxpayers Knowledge and compliance: Evidence from direct assessment tax in Lagos State. Indian-Pacific Journal of Accounting and Finance, 4(4), 28–40.

30.

Jansen

Steyn

Ngobeni

Sithole

(2021). The corporate income tax gap in South Africa: A top-down approach. Studies in Economics and Econometrics, 45(1), 53–69.

31.

Kasipillai

Aripin

Amran

N. A.

(2003). The influence of education on tax avoidance and tax evasion. EJournal of Tax, 1(2), 134–146.

32.

Khwaja

M. S.

Iyer

(2014). Revenue potential, tax space, and tax gap: A comparative analysis (World Bank Policy Research Working Paper 6868).

33.

Mangoting

Sumarno

V. A.

Gloria

Indriani

S. D.

(2020). Does coercive power influence the relationship of tax morale on tax evasion (Doctoral dissertation, Atlantis Press).

34.

Minh

T. L.

(2007). Estimating the VAT base: Method and application. Tax Notes International, 46(2), 203–210.

35.

Odhiambo

Olushola

(2018). Taxation and economic growth in a resource-rich country: The case of Nigeria. In J. Iwin-Garzyńska (eds.), Taxes and taxation trends. Intech Open.

36.

OECD. (2018). Revenue statistics in Africa 2018 – Ghana. oe.cd/revenue-statistics-in-africa

37.

OECD. (2021). Revenue statistics in Asia and the Pacific 2021: Emerging challenges for the Asia-Pacific region in the COVID-19 era. OECD Publishing.

38.

Okpeyo

E. T.

Musah

Gakpetor

E. D.

(2019). Determinants of tax compliance in Ghana: The case of small and medium tax payers in Greater Accra region. Journal of Applied Accounting and Taxation, 4(1), 1–14.

39.

Oppong

Owiredu

Churchill

R. Q.

(2014). Micro and small scale enterprises development in Ghana. European Journal of Accounting Auditing and Finance Research, 2(6), 84–97.

40.

Poniatowski

Śmietanka

Bonch-Osmolovskiy

(2020). Study and reports on the VAT gap in the EU-28 member states: 2020 final report (CASE Research Paper 503).

41.

Prichard

(2009). Taxation and development in Ghana: Finance, equity and accountability (Doctoral philosophy dissertation, University of Sussex).

42.

Saad

(2014). Tax knowledge, tax complexity and tax compliance: Taxpayers’ view. Procedia - Social and Behavioral Sciences, 109(1), 1069–1075. https://doi.org/10.1016/j.sbspro.2013.12.590

43.

Schmölders

(1959). Fiscal psychology: A new branch of public finance. National Tax Journal, 12(4), 340–345.

44.

Schmölders

Strümpel

(1969). Psicología fiscal comparada. Boletín de Estudios Económicos, 24, 277.

45.

Schneider

Buehn

Montenegro

C. E.

(2010). New estimates for the shadow economies all over the world. International Economic Journal, 24(4), 443–461.

46.

Schneider

Klinglmair

(2004). Shadow economies around the world: what do we know?http://ssrn.com/abstarct=518526.

47.

Schneider

J. L.

(2020). A fiscal policy dashboard for low-income economies (CGD Policy Paper 186). Center for Global Development. https://www.cgdev.org/publication/fiscal-policy-dashboard-low-income-economies

48.

Slemrod

(1989). The return to tax simplification: An econometric analysis. Public Finance Quarterly, 17(1), 3–27.

49.

Smulders

S. A.

(2013). An evaluation of tax compliance costs and concessions for small businesses in South Africa – establishing a baseline (Doctoral dissertation, University of Pretoria).

50.

Spicer

M. W.

(1974). A behavioural model of income tax evasion. The Ohio State University.

51.

Tee

Boadi

L. A.

Opoku

R. T.

(2016). The effect of tax payment on the performance of SMEs: The case of selected SMEs in Ga West Municipal Assembly. European Journal of Business and Management, 8(20), 119–125.

52.

Terkper

(2003). Managing small and medium-size taxpayers in developing economies. Tax Notes International, 29(2), 211–229.

53.

Toder

(2007). What is the tax gap? Tax Notes, 117(4), 367–378.

54.

Torgler

(2003). Tax morale, rule-governed behaviour and trust. Constitutional Political Economy, 14(2), 119–140.

55.

Whicker

(2017). The measurement of tax gaps. In Tax administration 2017: Comparative information on OECD and other advanced and emerging economies. OECD Publishing.

56.

World Bank. (2018). Doing business 2018. http://www.doingbusiness.org/reports

57.

World Bank. (2020). Ghana tax gap analysis. World Bank.

58.

Yamane

(1967). Statistics: An Introductory Analysis. Harper and Row.

Compliance Burden and Tax Gap Among Micro and Small Businesses: Evidence From Ghana

Abstract

Keywords

Introduction

Conceptual Issues: Tax Gap, Compliance Burden, and Micro and Small Enterprises

Review of Literature

Data and Methodology

Data

Estimation Strategy

Measurement of variables

Results and Discussions

Characteristics of Micro and Small Enterprises

Potential Tax, Actual Tax, and Tax Gap by Firms’ Characteristics and Tax Handles

Compliance Burden and Services Used in Honoring Tax Obligation

Differences in Tax Gap and Compliance Burden by MSEs

Effect of Compliance Burden on Tax Gap Among MSEs

Regional Analysis

Different Tax Handles

Correlates of Compliance burden among MSEs

Robustness checks

Conclusions

Footnotes

Appendix A: Omitted Variable Test for the Overall Model

Appendix D: Heteroskedasticity Test for the Overall Model

Declaration of Conflicting Interests

Funding

ORCID iDs

References