Essay: DETERMINANTS OF VALUE ADDED TAX REVENUE IN NIGERIA

ABSTRACT
This study investigated the determinants of VAT revenue in Nigeria. This study used consumption cost, Gross Domestic Product, Household cost and international trade values collected from the Central Bank of Nigeria Statistical Bulletin from 1994-2014 . The time series data was analysed using the Error Correction Model (ECM) method of analysis with the aid of E-views 8.1 statistical package. The results revealed that while consumption tax is a significant determinant of VAT revenue, GDP, Household cost and international trade are not significant determinants of VAT revenue. In view of our finding and in line with reality, government should intensify efforts to effectively exploit VAT as a source of revenue. Also, efforts should be made to block the leakages in the VAT collection process so as to ensure appropriate remittance to the Treasury. Furthermore, government should review the VAT rate higher as is obtainable in other countries. Finally, structure should be put in place to prosecute VAT evaders and their accomplices.
‘
CHAPTER ONE
INTRODUCTION
1.1 BACKGROUND TO THE STUDY
One of the means by which government increases its internally generated revenue is through the administration of Value Added Tax (VAT). This is a tax on the supply of goods and services which is eventually borne by the final consumer, but collected at each stage of the production and distribution chain. VAT as a concept was first introduced by France in 1954, and has over time been embraced by well over 100 countries. It has in recent time become a major source of revenue in many developing countries, including the sub-Saharan African countries (Ikpe & Nteegah, 2013).
Shalizi and Squire (1989) discovered that VAT accounted for over 30 percent of total tax revenue in Ivory Coast, Kenya and Senegal in 1982. Bogetic and Hassan (1993) found that Indonesia introduced VAT in 1983, and by 1988, the ratio of VAT revenue to GDP had risen to 4.5 percent. This impressive record in virtually all countries where it was introduced, according to Ajakaiye (2000) clearly influenced the decision to introduce VAT in Nigeria in January 1994 as a replacement for the existing sales tax. The Federal Inland Revenue Service (FIRS), the agency in charge of tax administration in Nigeria, pointed out that VAT is a consumption tax that is relatively easy to administer and difficult to evade, and has been embraced by many countries (FIRS, 1993a; 1993b; 1993c). In this context, it becomes necessary to empirically examine the likely determinants of VAT revenue.
Despite the prevailing rate of 5%, evidence so far supports the view that VAT is already a significant source of revenue in Nigeria. For instance, VAT revenue in the year of its inception (1994) was N8.194 billion, which was 36.5 percent greater than the projected N6 billion for that year (Ajakaiye, 1999). Similarly, actual VAT revenue for 1995 was N21 billion compared to the projected N12 billion. In terms of contributions to total federally collected revenue, VAT accounted for about 4.06 % in 1994 and 5.93% in 1995. As much as N404.5 billion was collected on VAT (5.1% of total revenue) in 2008 (Adereti, Adesina, & Sanni, 2011).
VAT has also become an indispensable component of the tax advice and tax reforms in developing countries. The growing practice of VAT is reflected in the extensive literature on technical, economic, and distributional dimensions of VAT, and there is a growing consensus on the “best practice” and desirable features required of a good VAT system (Zeljko & Fareed, 1993). VAT revenue is generated for distribution to the federal, state and local government of Nigeria. Unlike the oil revenue whose market government has no control over, VAT revenue is highly controllable and effective. VAT as an indirect tax, helps to reduce over dependence on oil revenue in Nigeria; assures a sustainable economic growth and development.
Tax revenue and its determinants have elicited considerable interest in public finance theory and empirical studies. Studies in this area include Harley (1965), Lotz and Morss (1967); Raja (1971); Roy (1979); Dehnashwar Ghura (1998); Gupta (2007) and Chaudhry and Munir (2010). Most of these studies used Ordinary Least Square method (OLS technique) to find the determinants of total tax to GDP ratio and most common exogenous variables used by the studies were share of agriculture sector, share of foreign trade and per capita income. The other variables tested are monetization, level of education and urbanization in the estimation of tax potential of different developing countries. Recent evidence for Nigeria on this argument is scarce and where available not without methodological gaps. On this note, this study aims to modestly contribute to the knowledge of value added tax revenue determinants.
1.2 STATEMENT OF THE PROBLEM
Nigeria’s reliance on oil revenue (which is based on global crude oil prices) has consequences for the growth and development of the economy. Since global oil prices are determined by factors beyond the control of the Nigeria government, uncertainty and unstoppable fluctuations of prices have further worsened the economic situation in Nigeria. In the same vein, the main shortcoming of Nigeria’s tax structure since 1986 has been its over-dependence on a small number of sources of tax revenue namely income tax VAT, and custom and excise duties. The trade taxes VAT on various imported products is vulnerable to external events because their prices are determined in the world market and tend to be volatile. This has resulted in inadequate tax revenue and continuous existence of budget deficit. The present tax structure does not raise adequate revenues thereby encouraging domestic borrowing and seeking external finance which are only temporary measures of deficit financing. Moreover, external funds can no longer be relied on due to donor conditions as a result of the recent global financial meltdown.
Furthermore, potential sources for domestic borrowing are few and external grants reduce autonomy and increase political and economic dependence. The alternatives are therefore to raise money through taxation, curtail desired government expenditure, or continuously revise the tax structure. This assertion has led a number of economies to conclude that the economic prosperity of a nation depends on its ability to raise revenue. (Gupta 2007, Johnson Et al 1998; & Mogres 2009).
Recognizing the need and effort by government to raise tax revenue in Nigeria, the study tries to investigate the factors affecting VAT revenue in Nigeria. The problem therefore is to examine how these factors have influenced the VAT revenue collected in Nigeria.
In line with the above, the present study poses a number of questions, objectives and hypothesis that will guide the conduct of this study.
1.3 RESEARCH QUESTIONS
This study raises the following questions:
1. What is the relationship between consumption cost and VAT revenue?
2. What is the relationship between GDP and VAT revenue?
3. What is the relationship between international trade and VAT revenue?
4. What is the relationship between household cost and VAT revenue?
1.4 OBJECTIVES OF THE STUDY
The broad objective of the study is to examine the determinants of value added tax revenue in Nigeria. The specific objectives are:
1. To find out the relationship between consumption cost and VAT revenue.
2. To find out the relationship between GDP and VAT revenue.
3. To find out the relationship between international trade and VAT revenue.
4. To find out the relationship between household cost and VAT income.
1.5 SIGNIFICANCE OF THE STUDY
The study contributes to the existing literature on tax revenue in Nigeria. It identifies the determinants of tax revenue which when properly understood, documented and captured in tax revenue models, would make it possible to estimate tax revenues within a specific period of time. The study also stimulates further research in the area of taxation which provides an informed basis for taking action on tax policy in addition to filling the gap about what is currently known about VAT revenue. The study is timely given the current effort to rationalize the budget, eradicate poverty and stabilize the economy.
1.6 STATEMENT OF THE HYPOTHESES
The following research hypotheses relevant to this study are stated in null form:
1. H0: There is no significant relationship between consumption cost and VAT revenue.
2. H0: There is no significant relationship between GDP and VAT revenue.
3. H0: There is no significant relationship between international trade and VAT revenue.
4. H0: There is no significant relationship between household cost and VAT revenue.
1.7 SCOPE OF THE STUDY
The study aims to research the determinants of VAT with careful examination of the above variables. It would involve the use of secondary data collected from Central Bank of Nigeria Statistical bulletin.
The study covers the period since the adoption of VAT in Nigeria till date.
1.8 LIMITATIONS OF THE STUDY
Limitations encountered in this study are;
1. The CBN statistical bulletin has not been printed for 2015.
2. The fact that results may not be generalized to other countries
3. Significant economic events occurring during the period of time under examination which may have affected the data.
1.9 DEFINITION OF TERMS
Tax policy: This is a broad statement of the government as regards to tax issues.
Taxation: This is the process of collecting the various taxes that have been levied upon the various groups of tax payers.
Tax reform: This is the way government revises how tax laws are imposed.
Tax advice: These are services rendered by a tax expert who is experienced in tax related issues.
Budget deficit: This is a budgetary system whereby government expenditure exceeds government income.
GDP: it is an acronym for gross domestic product. It is the total value of the goods and services produced by the people of a nation during a year not including the value of income earned in foreign countries.
‘
CHAPTER TWO
LITERATURE REVIEW
2.1 INTRODUCTION
This part reviews literature on different issues that assist as root for this work. This review is basically designed into theoretical and empirical reviews which explain the theory regarding VAT and its determinants.
2.2 CONCEPTUAL FRAMEWORK
2.2.1 The concept of tax
A tax is a compulsory levy imposed by a public authority on the income, profit or wealth of an individual, family, community, corporate or unincorporated body etc for the purpose of public use (Anyaduba, 1999). World Finance defines tax as a levy or other type of a financial charge or fee imposed by state or central government on legal entities or individuals. They went further to say that tax is the principal source of revenue for a country’s government.
Adebisi and Gbegi (2013) also agreed with them. They said that tax is a civic duty and an imposed contribution by government on her subjects and companies to enable her finance or run public utilities and perform other social responsibilities. Taxes therefore generate revenue to government to enable her meet the traditional responsibilities of maintaining law and other, general administration and other duties of providing social and infrastructural facilities to aid economic welfare and development of a country. It is not a voluntary payment or donation, but a compulsory, exacted pursuant to legislative authority and is any contribution imposed by government. It does not bestow on the tax payer the right to claim something equivalent in return for the tax paid from the government.
Taxes are classified into two broad categories. They are direct and indirect taxes. Direct Taxes are taxes that are extracted directly from the person who will bear the burden of the tax. It is collected directly from the income of the tax payer. It includes Petroleum Profit Tax, Pay as You Earn (PAYE), Capital Gains Tax.
Indirect Taxes on the other hand, are the taxes levied indirectly. It is levied on commodities before they reach the final consumer, but ultimately paid by the consumer as part of the market price. Here, the incidence is borne by both the producers and the consumers depending on the type of product. They are called indirect because the administering authorities which levy the taxes on goods and services do not collect the taxes from the consumer but do so indirectly through importers, manufacturers and other intermediaries. The shifting or passing on of the liability is effected by loading the tax element on the selling price of the commodities sold to the next person in the commercial chain until it is finally borne by the consumer. Value added tax, Excise duties and Mineral Royalties are examples of indirect taxes currently in use in Nigeria.
2.2.2 The concept of Value Added Tax
Value added tax is a consumption tax borne by all consumers both the government and the masses alike. It is levied at all stages of production and collected directly from the manufacturers who in turn levy it on the goods and services. Ola (2001) said that VAT is a tax paid at each stage of value added. It is a multi-stage tax which applies whenever goods and services are supplied by the producers. He also said that VAT is levied on the value gained or added on the products before being sold; VAT is the output less input value. Adereti et al defines VAT as a consumption tax levied at each stage of the consumption chain and borne by the final consumer of the product or service.
VAT is a multi-stage system of taxation, with tax being levied on value addition at each stage of transaction in the production/ distribution chain (Fathi & Esmaeilian 2012). They emphasized that VAT is a tax on the final consumption of goods or services that is ultimately borne by the consumer. IMF survey defines VAT as an indirect tax imposed on each sale beginning at the start of the production and distribution cycle and culminating in the sales to the consumer. They went further to create the impression that it is the consumer that absorbs the VAT as part of the sales prices, showing that VAT essentially is a consumption tax collected, throughout the production chain. In Nigeria, a charge of VAT 5% is expected to be remitted on all invoiced goods and services not exempted from paying VAT, under the Value Added Tax Act 1993 as amended. The items that have been exempted from value added tax include; Medical, veterinary and pharmaceutical raw materials and products, Basic food items, Infant foods, Books, newspapers and magazines, Educational materials, Baby products, Commercial vehicles and spare parts, Agricultural equipment and products, Fertilizers, Water treatment chemicals.
Based on Federal Government duty free concession, the following diplomatic goods are exempted from VAT: Aircrafts, their parts and other ancillary equipment, Films of educational, scientific and cultural character imported by United Nations or any of its Agencies, Fuel, lubricants and similar products to be used solely in the operation of aircraft of the armed forces, goods imported by voluntary organization, goods donated to charity, Goods imported for the President of the federal republic of Nigeria Goods imported for Consular Officers, Diplomatic Privileged Importations, Technical assistance (Based on Customs Personal Effect Standards, Passengers’ baggage, Patterns and samples, cut, mutilated, spoiled or otherwise rendered unmerchantable, Personal effects, Scientific specimens, imported for public exhibition, study or research 14. Foodstuff, Church and Mosque equipment, life-saving vehicles appliances including ambulances, Medicinal preparations, Specialised Hospital and surgical equipment, Military hardware and uniform, Arms and Ammunition imported by the Nigeria Police Force, Plant and machinery imported for use by companies located within the Export Processing Zone (EPZ), Imported goods intended for duty free shops, Goods and services consumed by Diplomats, Embassies, Foreign Missions which have reciprocal agreements with Nigeria, Goods donated or obtained free under a technical assistance program from donor international bodies or countries, Plant and machinery and equipment purchased for gas utilization in the downstream sector of petroleum operations, Tractors, ploughs, agricultural equipment and implements purchased for agricultural projects.
Services excepted from tax include: Medical and health care services, Services rendered by Community Banks, People’s Bank and Mortgage Institutions, Performance conducted by educational institutions as part of learning, Social services (orphanages, charities, and fire-fighting), Pure postal service, Overseas air transportation, Public telephone (coin operated) and telegram services. This does not include private telephone or telephone used for business or commercial purposes, Edible or table salt, Water, Salary or wages from employment or directors’ emolument from appointment, Hobby activities, Private transactions, House rent. (i.e rent on residential accommodation only). Where the VAT collected on behalf of the government (output VAT) in a particular month is more than the VAT paid to other persons (input VAT) in the same month, the difference is required to be remitted to the government, on a monthly basis, by the taxable person (Oserogho & Associates, 2008).
The operation of VAT can be viewed as a typical chain of transactions where goods produced by a manufacturer are sold to a wholesaler who sells to a retailer who in turn sells to a consumer. Everybody in the chain except the consumer gets a refund for the imput VAT he paid on his purchases. Since the consumer is not entitled to any refund in respect of the VAT included in the price he paid to the retailer, the tax can be said to form part of the net cost of the goods purchased by him. (Ezejelue, 2001).
From the above, it can be deduced that there are intermediaries through whom goods or services must pass before getting to the final consumer. At each stage, as the products or service pass from one person to another, a value is added to it. It is this value that is being taxed and borne at last by the final consumer. Therefore the value of the goods and services to the final consumer presents the aggregate of all the values added by successive producers or intermediaries in the chain. Since each trader pays only the VAT attributable to the value he added at his stage the final tax for any given final value is same, irrespective of the number of stages in the process or chain.
Fathi and Esmaeilian (2012) said that the regressive nature of VAT is a disadvantage. They agreed that VAT has a self-controlling nature, is neutral, increases economic growth, has an effect on inflation and is expensive to administer. They also comment that VAT has a positive effect on Balance of Payment. Their research on VAT and tax evasion reveals that there is no definite relationship between the VAT rate and tax evasion. Sologoub (2003) distinguished two main factor groups that influence VAT accumulation: VAT rules that is, rates, bases, etc and other variables of economic activities that condition tax bases and compliance of tax regulations.
The Institute of Chartered Accountants of India postulates that there are three variants of VAT; the gross product variant, income variant and the consumption variant; Gross Product Variant where taxes are levied on all sales and all deductions for tax paid on inputs excluding capital inputs is allowed, Income Variant allows set-off on purchases of raw materials and components as well as depreciation on capital goods. Consumption Variant where Taxes are levied on all sales with deductions for taxes paid on business inputs. Capital goods are included.
2.2.3 History of Value Added Tax
Value Added Tax was first adopted by France in 1954. Owing to the close economic relationship between France and its colonies, VAT was introduced almost immediately after 1954 in most of Francophone African Countries, beginning with Cote D’Ivoire in 1957. Within ten years of its administration, VAT in one form or the other was operational in most French speaking African countries. By the end of the twentieth century it had been adopted throughout the European Union and in many countries in Asia, Africa, and South America. Nigeria is one of those countries to adopt VAT in Africa.
In Nigeria, VAT was introduced in 1993 by the VAT Act No. 102 of 1993 as a replacement for the sales tax that was in operation in the Federal Capital Territory. It was the outcome of Dr. Sylvester Ugo’s study group on indirect taxation in November 1991. It was designed as a consumption tax payable on goods and services consumed by individuals, government agencies or business organizations. The federal government was not satisfied with revenue yield from the Sales tax which covered only nine categories of goods plus sales and services in registered hotels, motels, and similar establishment. She felt that the narrow base of the sales tax negates the fundamental principles of consumption tax, which by nature is expected to cut across consumptions of goods and services. Value Added Tax, in contrary has a broader base and includes, most professional services and banking transactions that are high profit generating. Only locally manufactured goods were targeted by the sales tax Decree of 1986; although this might not have been the intention of the law; VAT is neutral in this regard. Under VAT, a considerable part of the tax to be realized is from imported goods. The high yield from it boosts the fortunes of the state government with minimum resistance from the tax payers (Ola, 1999).
This resulted in setting up the Modified Value Added Tax (MVAT) committee on 1st June 1992 as recommended by the study group. This committee was set up to carry out a feasibility study on VAT. In January 1993, the Federal government agreed to introduce VAT. The Decree took effect from January, 1994 after the Act No. 102 of 1993 was promulgated.
Nigeria operates a VAT rate that does not synchronize with the Economic Community of West African States (ECOWAS) Protocol. ECOWAS adopted a uniform VAT protocol due to the constant movement of people and goods across the countries in the region, and the need to be subject to similar conditions. Under Nigeria’s influence, the ECOWAS advisory rate has been reduced to 10 percent, while Nigeria, despite being a signatory of the protocol, currently operates the lowest VAT rate across the sub-region, at 5 percent.
Since its adoption, VAT has remained a constant 5%. Currently, 17 categories of goods and 24 categories of services are VAT-eligible and all imports are VAT-eligible, whether raw materials or finished goods. The benefit of VAT in Nigeria since its inception cannot be over emphasized. It has increased the revenue of Nigeria over time, so that it is now the third highest revenue earner for the Federal Government, next to company income tax and petroleum profit tax. Secondly, it has also reduced the tendency of tax evasion. However, there are increasing complaints from various quarters, especially the organized private sector, about the effects of VAT on their operating costs and the prices of their products (Ajakaiye, 1999).
2.2.4 VAT administration in Nigeria
Every organization in Nigeria registered for VAT purposes is expected to pay the 5% VAT on all goods and services purchased or supplied to it. The goods and services are eventually sold or supplied to accredited agents, distributors, clients or customers, etc with the 5% VAT added to their respective selling prices (Anyaduba 2003). The prospective VAT payer obtains and completes the form and returns same to the nearest VAT office. Once, registered, the VAT proceeds are expected on monthly basis to be paid to the VAT office
The Federal Inland Revenue Service (FIRS) has made history by successfully administering the Value Added tax in Nigeria. Value Added Tax has been a major tool for a sustainable economic development in Nigeria since its introduction in January 1994. The tax has improved the social and macro-economic level in the Nigerian economy. VAT helps in accelerating economic growth by mobilizing privately held resources which will automatically boost public revenue enhance consumption patterns and generate savings. Despite the rate of 5%, VAT is a major contributor to tax revenue in Nigeria.
The net proceeds from VAT are shared among the federal, states and local government in the ratio of 56:24:20. This is done in consonance with the regulations guiding the bodies that are in charge of VAT in Nigeria.
2.2.4.1 The Federal Inland Revenue services (FIRS)
The Federal Board of Inland Revenue is the tax authority responsible for collection of tax accruable to the federal government; its operational arm is the Federal Inland Revenue Services (FIRS). This tax authority was established by the provision of section 3 No. 22 of the Companies Tax Act, CITA (1961). In 2007, the Federal Inland Revenue Service Board (Establishment) Act was enacted and it repealed that of 1961. Section 3(2) of the Board defined the membership. It is as follows:
(a) The Executive Chairman of the Service who shall be experienced in taxation as Chairman of the Service to be appointed by the President and subject to the confirmation of the Senate
(b) Six members with relevant qualifications and expertise who shall be appointed by the President to represent each of the six geo-political zones
(c) A representative of the Attorney-General of the Federation
(d) The Governor of the Central Bank of Nigeria or his representative
(e) A representative of the Minister of Finance not below the rank of a Director;
(f) The Chairman of the Revenue Mobilization, Allocation and Fiscal Commission or his representative who shall be any of the Commissioners representing the 36 States of the Federation
(g) The Group Managing Director of the Nigerian National Petroleum Corporation or his representative who shall not be below the rank of a Group Executive Director of the Corporation or its equivalent
(h) The Comptroller-General of the Nigeria Custom Service or his representative not below the rank of Deputy Comptroller-General
(i) The Registrar-General of the Corporate Affairs Commission or his representative not below the rank of a Director, and
(j) The Chief Executive Officer of the National Planning Commission or his representative not below the rank of a Director.
The members of the Board, other than the Executive Chairman are part time members. The chairman and any other seven members of the Board shall constitute a quorum. The Board shall have five years tenure of office.
Power and Duties of the Board:
Section 7 of the Federal Inland Revenue Service Board (Establishment) Act of 2007 contains provision with regards to powers and functions of the Board. They include:
(a) Provide the general policy guidelines relating to the functions of the Service
(b) Manage and superintend the policies of the Service on matters relating to the administration of the revenue assessment, collection and accounting system under this Act or any enactment or law
(c) Review and approve the strategic plans of the Service;
(d) Assess persons including companies, enterprises chargeable with tax
(e) Assess, collect, account and enforce payment of taxes as may be due to the Government or any of its agencies
(f) Collect, recover and pay to the designated account any tax under any provision of this Act or any other enactment or law
(g) Do such other things which in its opinion are necessary to ensure the efficient performance of the functions of the Service under this Act.
Technical Committee of the Board:
Section 9 (1) of the FIRS Act of 2007 established the value added tax technical committee. It comprises of:
a) A chairman who shall be the chairman of the Federal Board of Inland Revenue.
b) All the Directors in the Federal Inland Revenue Service (FIRS).
c) The Legal Adviser to Federal Inland Revenue Service
d) A Director in the Nigerian Customs service; and
e) Three representatives of the state Government who shall be members of the Joint Tax Board
Functions of the Technical Committee:
i) To consider all tax matters that requires professional and technical expertise and make recommendation to the Board.
ii) To advise the Board on its powers and duties as listed in the act.
iii) To attend to all such matters as may from time to time be referred to it by the Board.
Every state of the federation has a state internal revenue service. The state IRS assists the Federal Inland Revenue Service to co-ordinate all tax affairs within the state. The state IRS comprises of:
1. The Executive head of the State Internal Revenue Service is the Chairman and he is appointed by the Governor.
2. Three other persons nominated by the Commissioner of Finance in the State on their personal merit.
3. All Directors and Heads of Departments within the State IRS.
4. A Director from the Ministry of Finance.
5. A Secretary of the Board.
6. Legal Adviser of the Board.
2.3 LITERATURE REVIEW ON VARIABLES
VAT income is determined by a number of factors, including economic situation of the country, which is best characterized by gross domestic product. In addition, the higher the standard of life in the country, considered as GDP per capita, the higher the consumption, the better awareness of tax payments and lower tax fraud, consequently the relationship with GDP per capita is extra revised. Due to the fact, that VAT is consumption tax, primarily VAT income depends on the consumption level in the country. To identify what influences more VAT income ‘ governments or households’ consumption, the relationship not only with the general consumption, but also with both variables, is examined. On the one part, VAT income, in addition to the mentioned above factors, is determined by the level of export and import. (Bikas & Andruskaite, 2013)
2.3.1 VAT and consumption cost
One of the main rationales for taxing consumption rather than income is that it is believed that consumption taxes discourage consumption, encourage savings, and thus generate higher economic growth. We find that under all estimation methods and empirical specification that VAT is negatively related to consumption cost.
Very few literature provide formal empirical evidence on the relationship between VAT and consumer spending. Alm and El- Ganainy (2012) attempted to fill the void by examining the empirical linkage between VAT and consumption behavior using actual data and dealing appropriately with standard econometric issues. Using a sample of 15 EU countries, they found out that effective VAT rate is negatively correlated with the level of aggregate consumption. Using a two class model with uneven distribution, Matsuzaki (2003) showed that an increase in consumption tax rate leads to a decrease in effective demand. Bird (2005), Diamond and Zodrow (2008) conclude that the effect of VAT on consumption is an important unresolved issue. Tamaoka (1994) reported that value added tax is a regressive tax because it taxes consumption and the propensity to consume tends to decrease as income rise.
2.3.2 VAT and GDP
Eurostat considered tax revenue and its relationship with GDP in the European Union and the Euro area in 2016. It concluded that as tax revenue increased, taxes on production and imports accounted for 13.6% of GDP while current taxes on income and wealth stood at 12.8%of GDP. Wilford and Wilford (1978) estimated the income-elasticity and buoyancy of tax revenue in Central America. They found out that the income elasticity of tax revenue was less than unity which suggests that the tax structure was stable.
2.3.3 VAT and international trade
Desai and Hines (2005) estimated the impact of VAT introduction on international trade. Using a large data set of 143 developed and developing countries, they deduced that import and export performance are negatively affected by the VAT. Lutfunnahar (2007) identified the determinants of tax share and revenue performance in Bangladesh. The results obtained suggest international trade, broad money, external debt and population growth to be significant determinants of tax.
2.3.4 VAT and household cost
Household spending is the most important contribution to VAT revenue. Sekwati and Malema (2011) observed that a higher percentage of income of the rural household is spent on consumables as opposed to households in urban areas. This is due to the significant increase that retailers add to the prices as opposed to the dictated increase. Extending on the work of Barret and Wall (2006) Leahy et al (2011) assessed in Ireland, that households pay a higher percentage on their weekly disposable income. Cashin and Takashi (2011) conclude that VAT rate increases little or no impact on household spending in Japan.
2.4 THEORETICAL FRAMEWORK ON VALUE ADDED TAX
Several theories of taxation exist. The principle of economics is guided by two principles, who will benefit and who can pay. In the literature of public finance, there are two theories: the benefit theory and the ability theory.
2.4.1 The Benefit theory
The benefit theory has a modern name which is the voluntary exchange theory. This theory was developed by Erik Lindahl. This theory states that the tax levels are easily determined due to the fact that the tax payers pay proportionately for the benefit they receive from the government. Therefore, the individual who benefits more pays the most taxes. This is because VAT on goods and services are paid in order to benefit. Those who purchase more, pay more on their consumption and hence benefit most from the service. There are two approaches in analyzing, the benefit theory: The Lindahl model and the Bowen model.
The Lindahl model attempts to solve three problems.
‘ It tries to allocate total expenditure among goods and services.
‘ It attempts to allocate the tax burden.
‘ It determines the extent of state activity.
The Bowen model has more operational significance. This is because it demonstrates the opportunity cost of private goods is forgone when social goods and services are produced under conditions of increasing cost.
The benefit theory is advantageous because it highlights the correlation between revenue expected and expenditure in a budget. It also determines the market behavior of the consumers in the public sector. Despite these advantages, it also has its disadvantages; it limits the scope of government activities. Also it is only possible when the beneficiaries can be observed directly.
2.4.2 Ability to pay theory
This theory was propounded by Arthur C. Pigou and it treats the government revenue and expenditure separately. Taxes are based on the taxpayers ability to pay; effective demand. It is the most popular and commonly accepted theory. This theory is very hard to apply in practice as the difficulty arises in the definition of the ability to pay. There is no unanimous measure of a person’s ability to pay. Economists have advanced three principles.
1. The ownership of property: Some are of the opinion that ownership of property is a good basis for measuring a person’s ability to pay. This idea was rejected because not every wealthy person places importance in acquiring assets as it is a means of escaping paying taxes.
2. On the basis of expenditure: This principle relies on the level of expenses a person incurs; the greater the expenditure, the higher the ability to pay the tax. This principle was also rejected because some households are larger than others and would thus incur more expenses than the smaller households. Therefore using this measure would increase the burden of the taxpayers who are already incurring more expenses.
3. Most economists are of the opinion that the income of a person should be the measure of his ability to pay. This theory was accepted. This most likely gave rise to the progressive system of taxation. It is currently used in the modern world as a basis for measuring a person’s ability to pay.
Other theories include,
2.4.3 Principle of proportionality.
J. S. Mill and other classical economists suggested the principle of proportionate taxation in order to satisfy the idea of justice in taxation. This principle is of the opinion that if taxes were levied proportionately with income, it will exact equal sacrifice on taxpayers within the same income bracket. This principle is in consonance with the income principle of the ability to pay theory. However, the marginal utility of income increases.
2.4.4 The cost of service theory
Some economists are of the opinion that if the state charges the actual cost of the service, it will satisfy the idea of justice in taxation. This theory was rejected because most of the services incurred by taxpayers are not fixed, hence they cannot exactly be determined. Also the tax system does not operate on a quid pro quo basis which is a fundamental principle of this theory.
2.4.5 The Sacrifice theory
This theory attempts to determine the burden that rests on the taxpayer after he has paid his tax and how much of his disposable income is available for his sustenance. This theory believes that the payment of tax is a sacrifice the taxpayer bears in order to support the development of his economy. The applicability of this theory is difficult except is expressed in terms of income and consumption.
2.5 EMPIRICAL FRAMEWORK
Adereti et al (2011) investigated the contribution of VAT to the GDP in Nigeria from its inception to 2008. Time series data on both the GDP and VAT revenues were analysed using both simple regression analysis and descriptive statistical method. A positive and significant correlation was found to exist between VAT revenue and GDP though both variables fluctuated greatly over the period. VAT revenue was discovered to account for about 95% of the variations in the GDP. Ajakaiye (1995) analyzed the impact of value added tax on key sectoral and macroeconomic aggregates in Nigeria. A survey of VAT able Nigerian manufacturers, distributors, importers and suppliers of goods and services, organizations was conducted to gain insights on the way VAT is handled by these organizations. The survey shows that a majority of the VAT able organizations treat VAT in a price cascading manner by regarding it as cost contrary to expectations.
Aneta Kaczynska’s study (2015) on the determinants of VAT revenue in Poland used multiple regression analysis to analyse VAT collected as the dependent variable and the rate of inflation, GDP in current prices and standard VAT rate as independent variables. It indicated that only one independent variable can be used in linear regression model ‘ GDP; that implementing multiple regression model in the case of Poland was impossible. Following the work of Sansac et al(2010), Anthanosios Tagkalakis (2014) investigated the effect of cyclical economic activity development on VAT revenue efficiency. Regressing data by means of OLS over a period of 13 years, in order to study the effect of the shift in consumption pattern and the ability to control tax evasion on VAT C- efficiency, it was discovered that there is a negative but significant effect of consumption patterns of necessity goods on VAT C-efficiency.
The most extensive work that covers all the variables being discussed was done by Bikas and Andruskaite (2013). Their research covered the European Union member states from 2004-2013 which analyses the relationship between VAT revenue and GDP, GDP per capita, consumption, household expenditure, Government expenditure, import, export and unemployment. They premised that VAT income is determined by the economic situation of the country which is best characterized by GDP, consumption, export and import. Using correlation coefficient, the following model was constructed to determine the VAT revenue factors:
Y = -907222,59 – 0,42×1 – 0,06×2+ 0,50×3 + 0,28×4 – 0,26×5 + 158,80×6
Here: y ‘ VAT tollage
X1 ‘ gross domestic product
X2 ‘ consumption costs;
X3′ households’ consumption costs
X4 ‘ export
X5 ‘ income
X6 ‘ Gross Domestic Product per capita.
The result obtained showed that the strongest relationship exists between international trade, GDP, GDP per capita, consumption, import, export and VAT income.
Bogetic and Hassan (1993) used regression analysis to analyse 20 single rate countries and 14 multi rate countries. They deduced that the single rate group generates 0.503 VAT revenue of GDP for every percentage point of average as opposed to the multiple rate coefficient of 0.339. Though the difference is not statistically significant, they concluded that the VAT base, administrative capacity and other factors also determine the VAT revenue
Gupta (2007) investigated on the revenue performance of developing countries for the past 25 years. He discovered that the GDP per capita, share of agriculture and trade openness are strong determinants of revenue performance. Also, foreign aid was found to improve revenue performance while foreign debt does not. He concluded that with the adoption of VAT in many developing countries, the self-enforcing mechanism of VAT which induces greater compliance has a greater potential for improving the revenue performance.
Mullbacher et al (2013) researched on how many private households and VAT able persons spend on VAT able goods and services. They aimed to quantify for each EU member state, the amount households spend on goods and services when broken down into categories where,
‘ Zero and reduced rates were abolished.
‘ The rates were abolished and replaced with a new revenue neutral standard rate.
‘ The rates are abolished and low income households receive lump sum transfers in compensation for their higher
Using aggregate information from national household budget surveys provided by Eurostat, and applying the VAT rates in place in 2011 in each member state, they calculated how much each household paid in VAT as a proportion of their total expenditure. The opine that households account for 60% of all VAT tax liability across the 27 countries under the EU, that the average household faces a VAT bill of about 11% of their total expenditure. Romania and Hungary having the highest ratios of 17.8% and 17.5% respectively while United Kingdom having 8%, the lowest bill in proportion to expenditure.
They also studied the effect of changes in the VAT regime in GDP and consumption among others. Taking the same scenarios into consideration they concluded that:
‘ In scenario A the effective VAT rates increase, reduce manufacturing and production in agriculture respectively. Generally, it induces relatively small changes in macroeconomic indicators, GDP reducing by 0.4% in line with consumption (-0.7%).
‘ The second scenario showed no effect on GDP but a small effect on other macroeconomic variables.
Karran (1985) studied the determinants of taxation in Britain. He concludes that tax revenue is determined by three factors: the tax base, the tax rate and the tax collection efficiency. In developing countries, the tax administration was leads to pitfalls in the expected revenue.
Using MIMIC models Kasnauskiene and Krimisieraite (2015) examined the determinants of VAT Gap in Lithuania. They premised that general government consumption expenditure, inflation, gross fixed capital formation, additional amount of wage and money on deposit affect the VAT gap. Their result concludes that only two variables have a statistically significant influence on the VAT: general government consumption expenditure and inflation.
Taiwo (2016) researched on the determinants of VAT in South Western Nigeria. Using multiplicative functional form of tax revenue model, the results indicate that Ekiti state has the minimum (0.10) revenue generated for the period. This was attributed to the low population of the state and the limited industrialization of the state. Volume of trade and industrialization were also found to have a positive relationship with VAT. A negatively significant coefficient was observed between population and VAT. This was attributed to low demand for taxable goods in areas of low population and over reliance on oil revenue.
Teera (2002) examined the tax and tax structure of Uganda to investigate the factors effecting tax revenue in the country. He used the time series data of the period 1970 to 2000 and estimated a model which concluded that the agricultural ratio, population density and tax evasion affect all types of taxes. GDP per capita showed a negative sign. Tax evasion and openness (as measured by import ratio) gave the significant negative impact.
Wawire (2011) used Paul Samuelson’s fundamental general equilibrium analysis of the public sector to derive the determinants of VAT in Kenya. He tested the VAT structure to total GDP and international trade and concluded that a percentage change in GDP leads to a more than proportionate change in VAT revenue. That is, VAT revenue is elastic with respect to GDP; that the elasticity of VAT revenue with respect to volume of import is statistically significant. He concluded that VAT revenue responds with lags to changes in their respective bases. That is, the previous levels of tax bases (such as GDP, volume of trade and volume of imports) have significant influence on the present levels of VAT revenue. ‘
CHAPTER THREE
METHODOLOGY
3.0 INTRODUCTION
This chapter is aimed at stating the methodology intended to be used for the purpose of this research. It evaluates the system of explicit rules and procedures, the basis upon which this research is carried out and against which knowledge claims are evaluated. Also, it states the technique that is used in the collection and analysis of data and the type of data to be used. The data collected are related to the hypothesis, which are the problem of this study.
3.1 RESEARCH DESIGN
This study employs the correlational research design adopting the ex-post facto method of research because the data needed for the analysis already exists. The study covers Nigeria’s economy using a time series data analysis relating to the determinants of Value Added Tax (VAT) revenue for the years 1994 to 2014.
3.2 POPULATION AND SAMPLE SIZE
For the purpose of this study, the population is designed in terms of content, extent and time. To this end, the population relates tovalue added tax and consumption costs, gross domestic product, household cost and international trade in Nigeria from 1994 to 2014.
3.3 SOURCES OF DATA
The data to be used for this research are sourced from secondary sources. The data are got from Central Bank of Nigeria’s statistical bulletin published on the internet, the Federal Inland Revenue service (FIRS) online library and the Nigerian Bureau of Statistics (NBS) online library.
3.4 MEASUREMENT OF VARIABLES
For the purpose of this research, the following measures and proxies are adopted for the variables:
VAT = Value Added Tax (VAT) revenue
INT = International trade (Sum of oil and non-oil imports and exports)
HOC = Household cost (measured by the private consumption expenditure)
COC = Consumption cost (measured by the government consumption expenditure)
RGDP = Real gross Domestic Product (measured by GDP at constant prices)
3.5 METHOD OF ANALYSIS
This research uses time series data for a period of twenty (21) years applying the Error Correction Model (ECM) method of analysis with the aid of E-views 8.1 statistical package. The Augmented Dickey Fuller (ADF) technique will be applied on the variables to test for the level of stationarity of the variables.
3.6 MODEL SPECIFICATION
The model used for this work is adopted from the works of Bikas and Andruskaite, (2013) with some modifications. The econometric model as used by Bikas and Andruskaite, (2013) was stated as:
Y = ” + ”1×1 + ”2×2 + ”3×3 + ‘ + ”kxk + ‘
Where:
” = coefficients (constant)
x = independent variables
‘ = Stochastic error
The modified model is herein specified as:
VAT = f(RGDP, COC, HOC, INT)”'(1)
VATt = ”0+ ”1RGDPt+ ”2COCt + ”3HOCt+ ”4INTt + ”t ‘..(2)
Where:
”0 = constant
”1 to ”4 = Parameter Estimate
VAT = Value Added Tax revenue
RGDP = Real Gross Domestic Product
COC = Consumption Cost
HOC = Household cost
INT = International Trade
t = Time period (1994 ‘ 2014)
”t = Stochastic Error Term.
The model above specifies VAT as the dependent variable while RGDP, HOC, COC and INT as the independent variables.
‘
CHAPTER FOUR
Data Presentation and Analysis of Empirical Findings
4.0 INTRODUCTION
In this chapter, the empirical results which tested the systematic relationship between Value Added Tax (VAT) and the various determinants employed (Real Gross Domestic Product, Consumption Cost, Household Cost and International Trade) in Nigeria are presented. Firstly, because of the usual characteristics of time series data and the need for stationarity of variables in regression models, unit root tests are conducted on each of the variables to confirm the level of their stationarity. The study adopted Augmented Dickey Fuller technique using 5% significant level as benchmark. After that, the Johansen co-integration technique was employed to test for the existence of a long run relationship between the variables. Thereafter, a short-run dynamic model including the error correction term was specified.
4.1 DATA PRESENTATION
4.1.1 Descriptive statistics
Table 1: Descriptive Statistics for 1994-2014
INT HOC COC RGDP VAT
Mean 10149.11 10365.36 1892.104 14769.09 280.9005
Median 6589.800 8111.127 785.8194 527.5760 163.2977
Maximum 26232.60 22845.13 5849.987 67152.79 802.9647
Minimum 368.9000 610.3402 169.6692 345.2285 7.260800
Std. Dev. 9069.637 8164.507 1987.338 26278.34 278.2952
Skewness 0.626355 0.290739 0.825798 1.254294 0.775342
Kurtosis 1.887959 1.472453 2.070270 2.611874 2.108737
Jarque-Bera 2.455179 2.337576 3.143146 5.638196 2.799097
Probability 0.292998 0.310743 0.207718 0.059660 0.246708
Sum 213131.4 217672.6 39734.18 310150.9 5898.911
Sum Sq. Dev. 1.65E+09 1.33E+09 78990281 1.38E+10 1548964.
Observations 21 21 21 21 21
Source: E-views Output (2016)
The descriptive statistics of the variables used in the analysis presented in Table 1 explains the range, minimum, maximum, mid values, spread and normality of the variables. The mean value of VAT is N280.9 billion for the period under review while the minimum and maximum values are N7.26 billion and N802.96 billion respectively.RGDP had a mean of N14769.09 billion with a standard deviation of N26278.34 billion. The high standard deviation of RGDP may not be unconnected with the rebasing of its parameters in 2010 which was reflected in the increase of the activity sectors of the economy from 33 activities to 46 activities. All the variables used for the study including the dependent variable VAT are positively skewed and all have positive means.
4.1.2 Correlation Analysis
Table 2: Correlation Matrix
The correlation matrix was employed in this study to test for the level of association between the independent variables and the dependent variable and between the independent variables. The results are presented below:
Correlation
t-Statistic INT HOC COC RGDP VAT
INT 1.000000
—–
HOC 0.965562 1.000000
16.17696 —–
COC 0.976817 0.939379 1.000000
19.88963 11.94199 —–
RGDP 0.870454 0.764707 0.903079 1.000000
7.707931 5.172859 9.165687 —–
VAT 0.981507 0.945530 0.992338 0.890743 1.000000
22.34933 12.66064 35.01013 8.542566 —–
Source: E-views Output (2016)
From the table, it should be observed that the correlation of coefficient of a variable with respect to itself is 1.000 which implies a perfect correlation. All the independent variables were seen to have a positive relationship or association with Value Added Tax (VAT). Consumption Cost (COC) exhibited the strongest correlation or relationship with Value Added Tax (VAT) while Real Gross Domestic Product (RGDP) was seen to have the weakest relationship with Value Added Tax (VAT).
4.2 DATA ANALYSES AND RESULTS
4.2.1 Unit Root Testing
The variables employed in this study are tested at level, first difference and second difference. Value Added Tax (VAT) and Household Cost (HOC) were found to be stationary at second difference; Consumption Cost (COC) and International Trade (INT) were stationary at first difference while Real Gross Domestic Product (RGDP) was stationary at first difference second order. The results are shown in Tables 3.
Table 3: Unit root tests
Variables p-values Order Remarks
VAT 0.0001 2nd difference Stationary
HOC 0.0000 2nd difference Stationary
COC 0.0208 1st difference Stationary
RGDP 0.0005 1st difference 2nd Order Stationary
INT 0.0071 1st difference Stationary
Source: E-views Output (2016)
The implication of the above is that the variables are free from unit roots and are henceforth stationary. Hence, when we run the regression, we will not have a spurious result but a genuine and dependable one. ‘
4.2.2 Co-integration Tests
Table 4: Johansen Cointegration Test
Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None * 0.929290 123.6397 69.81889 0.0000
At most 1 * 0.871270 73.30553 47.85613 0.0000
At most 2 * 0.682579 34.35478 29.79707 0.0139
At most 3 0.469595 12.55178 15.49471 0.1323
At most 4 0.026158 0.503610 3.841466 0.4779
Trace test indicates 3 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegration Rank Test (Maximum Eigenvalue)
Hypothesized Max-Eigen 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None * 0.929290 50.33416 33.87687 0.0003
At most 1 * 0.871270 38.95075 27.58434 0.0012
At most 2 * 0.682579 21.80300 21.13162 0.0402
At most 3 0.469595 12.04817 14.26460 0.1089
At most 4 0.026158 0.503610 3.841466 0.4779
Max-eigen value test indicates 3 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Source: E-views Output (2016)
In order to conclude as to whether there is a long run relationship between the variables using the Johansen co-integration test at the 5% level of significance, the p-values of the Trace statistic and Max-eigen statistic must be greater than 5%. From the Trace statistic and Max Eigen statistic above, there is an agreement as to the number of co-integrated equations showing that there are at least 3 co-integrated equations in the model. The Trace statistic has a value of 12.55178 and a p-value of 0.1323 while the Max Eigen statistics has a value of 12.04817and its p-value is 0.1089. Therefore, this leads us to the acceptance of the alternate hypothesis that there is a long run relationship between the dependent variable VAT and the independent variables RGDP, HOC, COC, and INT.
4.2.3 Regression Analysis
Table 5: Error Correction Model result
Variable Coefficient Std. error t-statistic Prob. Status
C 15.48956 8.039947 1.926575 0.0746 Not Significant
D(INT) 0.001204 0.003400 0.354165 0.7285 Not Significant
D(RGDP) 2.25E-05 0.000602 0.037461 0.9706 Not Significant
D(COC) 0.071500 0.022446 3.185397 0.0066 Significant
D(HOC) 0.002243 0.005325 0.421279 0.6799 Not Significant
ECT(-1) -0.560203 0.233128 -2.402987 0.0307 Significant
R-squared 0.610757 Adjusted R-squared 0.471741
F-statistic 4.393444
Prob. (F-statistic) 0.012927 Durbin-Watson Stat 1.465754
Source: E-views Output (2016)
From the error correction model result in table 5 above, the coefficient of variation depicted R-squared is 0.61 which si gnifies that the model accounts for about 61% of the systematic variation in the dependent variable VAT while the remaining 39% is accounted for by variables that make up the stochastic error term. This coefficient of variation further reduces to 47% when it is adjusted for degrees of freedom. A look at the F-statistics which has a value of 4.393444with a p-value of 0.012927indicates that the model is statistically significant and the independent variables selected in the model can jointly influence our dependent variable VAT in the long run. The Durbin Watson statistics with a value of 1.465754indicates the absence of first order serial correlation and signifies that the model is free from bias and thus useful in decision making.
The result further reveals that all the independent variables (RGDP, HOC, COC and INT) have a positive relationship with Value Added Tax (VAT). This therefore suggests that an increase in any of them will lead to an increase in Value Added Tax (VAT). In terms of individual significance, only Consumption Cost (COC) had a significant relationship with VAT while the remaining variables (RGDP, HOC, and INT) had an insignificant relationship with VAT.The error correction term which is denoted ECM(-1) fulfilled the criteria for a long run relationship among the variables as the coefficient was negative and its p-value less than 0.05 at the 5% level of significance thus justifying the result of the Johansen Co-integration test. It also shows that the speed of adjustment to equilibrium is 3.07%.
4.2.4 Serial Correlation Test
Table 7:Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.966078 Prob. F(2,12) 0.1826
Obs*R-squared 4.936126 Prob. Chi-Square(2) 0.0847
Source: E-views Output (2016)
The Lagrang multiplier (LM) serial correlation test is used to test for higher order autocorrelation in the model and is proven to be more powerful than the Durbin-Watson statistics. The Obs R-squared of 4.936126with a p-value of 8.47% which is greater than 5% indicates the absence of serial correlation in the model. Therefore, the result of the regression analysis can therefore be depended upon for the purpose of decision making.
4.3 TEST OF HYPOTHESES
The decision rule for the rejection or acceptance of the hypotheses of the study is set at the 5% level of significance (critical value of 1.96); hence, the null hypothesis would be rejected if the probability value (P value) is less than 0.05. The following are the results of the tested hypothesis which are stated in the null form:
Hypothesis 1: There is no significant relationship between consumption cost and VAT revenue
From the results, it was discovered that the absolute t-statistic of COC was 3.185397with a p-value of 0.0066. This p-value is less than 0.05 therefore leading us to the rejection of the null hypothesis and then the acceptance of the alternate hypothesis that there is a significant relationship between consumption cost and VAT revenue.
Hypothesis 2: There is no significant relationship between gross domestic product and VAT revenue
From the results of the analysis, the absolute t-statistic of RGDP was 0.037461with a p-value of 0.9706. This p-value is greater than 0.05 therefore, the null hypothesis is accepted that there is no significant relationship between gross domestic product and VAT revenue.
Hypothesis 3: There is no significant relationship between international trade and VAT revenue
The results of the analysis as tabulated above, reveals that the absolute t-statistic of INT is 0.354165 while its p-value is 0.7285 which is more than 0.05 therefore leading us to the acceptance of the null hypothesis that there is no significant relationship between international trade and VAT revenue
Hypothesis 4: There is no significant relationship between household cost and VAT revenue
The results of the analysis as tabulated above, reveals that the absolute t-statistic of HOC is 0.421279while its p-value is 0.6799which is more than 0.05 therefore leading us to the acceptance of the null hypothesis that there is no significant relationship between household cost and VAT revenue.
4.4 DISCUSSION OF THE RESULTS AND FINDINGS.
The result presented earlier reveals that all the independent variables (RGDP, HOC, COC and INT) have a positive relationship with Value Added Tax (VAT), an indication that an increase in any of them will lead to an increase in Value Added Tax (VAT). In terms of individual significance, only Consumption Cost (COC) had a significant relationship with VAT while the other variables (RGDP, HOC, and INT) had an insignificant relationship with VAT.
From the analysis, it was discovered that the absolute t-statistic of COC was 3.185397 with a p-value of 0.0066. This p-value is less than 0.05 therefore leading us to the rejection of the null hypothesis and then the acceptance of the alternate hypothesis that there is a significant positive relationship between consumption cost and VAT revenue. This indicates that VAT revenue increases significantly as the consumption cost in the economy increases, thereby, creating more revenue to the government for infrastructural development
It was also observed from the analysis that the absolute t-statistic of RGDP, INT, and HOC were 0.03746,0.354165 and 0.421279 while its p-values were 0.9706, 0.7285 and 0.6799 respectively. The p-values are greater than 0.05 therefore leading us to the acceptance of the null hypothesis that there is no significant relationship between RGDP, INT, HOC and VAT revenue respectively. Although there is no significant relationship between these variables, the correlation was however found to be positive. An indication that an increase in RGDP, INT and HOC can also lead to an increase in VAT but not a significant increase as it is in the case of COC.
It should be noted that our findings is in contrast with that of Adereti et al (2011) who investigated the contribution of VAT to the GDP in Nigeria from its inception to 2008 and found that positive and significant correlation exist between VAT revenue and GDP. Also, Cashin and Takashi (2011) concluded that VAT rate increases have little or no impact on household spending in Japan while our finding negates this by revealing that there is a significant relationship between consumption cost and VAT revenue. The findings also negate that of Desai & Hines (2005) who estimated the impact of VAT introduction on international trade using a large data set of 143 developed and developing countries, and deduced that import and export performance are negatively affected by the VAT.
‘
CHAPTER FIVE
SUMMARY, CONCLUSION AND RECOMMENDATION
5.0 INTRODUCTION
This study examined VAT and its determinants in Nigeria. This chapter presents the summary of findings, conclusions drawn, recommendations made based on the research findings and suggestions for further research.
5.1 SUMMARY
This section summarizes the theoretical and empirical findings from this research. All variables were seen to have a positive correlation with VAT but in varying degrees.
At 5% level of significance, the result of the analysis showed that:
1. There is a significant relationship between consumption cost and VAT revenue.
2. There is no significant relationship between gross domestic product and VAT revenue.
3. There is no significant relationship between international trade and VAT revenue.
4. There is no significant relationship between household cost and VAT revenue.
5.2 CONCLUSION
It can be concluded that the importance of VAT as a source of revenue to government in this era of continuous drop in oil revenue owing to global oil price decline and the activities of the Niger Delta Avengers or militants in the oil rich region (Niger Delta) can never be overemphasized. VAT has a lot of advantages as discussed in the previous chapters. Therefore, the government of the day should intensify effort towards harnessing VAT revenue to boost its revenue generation power.
It is also worthy of note that the differences in our finding and that of other writers reviewed in our literature can be attributed to a number of factors such as; the size of data studied, the period covered and other country specific factors such as VAT rate differences in the various countries.
5.3 RECOMMENDATIONS
In view of our findings and conclusion summary above, and in line with the present day reality of the incessant decline in Nigeria oil revenue sources, the following are hereby recommended:
‘ The government of the day should intensify effort to effectively exploiting VAT as a source of revenue to fund its revenue needs owing to decline in our revenue sources.
‘ Efforts should be made to block the leakages in the VAT revenue collection process so as to ensure that the appropriate VAT is remitted to the Treasury.
‘ The government of the day should review the VAT rate higher as it is obtainable in other countries. However, the political will of effective utilization should be demonstrated.
‘ Structure should be put in place to prosecute VAT evaders and their accomplices.
5.4 TOPICS FOR FURTHER RESEARCH
Other researchers aiming to research further can study:
‘ VAT and Economic growth in Nigeria.
‘ VAT evasion and its effect on Gross National Product.
‘ Determinants of tax revenue in Nigeria.
‘
REFERENCES
Adebisi, J.F. & Gbegi, D.O. (2013). Effects of tax avoidance and tax evasion on personal income tax administration in Nigeria.
Adereti, S.A., Adesina, J.A. & Sanni, M.R. (2011). Value added tax and economic growth of Nigeria. European Journal of Humanities and Social Sciences. 10(1). 456-471.
Ajakaiye, D.O. (1999). Macroeconomic effects of VAT in Nigeria: Empirical analysis. International Research Journal of Finance and Economics. 95, 51-61.
Alm, J. & El-Ganainy, A. (2012). Value added taxation and consumption. IMF Working Paper 1203.
Anyaduba, J.O (1999). Companies Income Taxation in Nigeria. Nigeria, Edo, Benin City. United city press.
Anyaduba, J.O. (2003). An empirical survey of the value added tax system in Nigeria.
Bikas, E. & Andruskaite, E. (2013). Factors affecting Value added tax revenue. 1st Annual International Interdisciplinary Conference, AIIC. 41-49.
Bird, R. (2005). Value-added taxes in developing and transitional countries: Lessons and questions. Working Paper. 05(05).
Bogetic, Z. & Hassan, F. (1993). Determinants of value-added tax revenue: A cross sectional analysis. IMF Working Paper-series. 1203. 1-14.
Carrol. R, Cline. R. & Zodrow G. (2010). The Macroeconomic Effects of an Add-on Value added tax. A publication of Ernst &Young. 1-60
Cashin, D. & Takashi, U. (2011). The Intertemporal Substitution and Income Effects of a VAT Rate Increase: Evidence from Japan Discussion papers from Research Institute of Economy, Trade and Industry (RIETI).
Desai, M. A. & James, H. (2002). Value added taxes and international trade: The Evidence. Washington D.C.
Diamond, J. & Zodrow, G. (2008). Consumption tax reform. Fundamental tax Reform: Issues, choices and implications. Cambridge, MA. The MIT Press.
Ezjelue A. C. (2001). Primer on International Accounting. Port Harcourt, Rivers, Nigeria. Educational Books and Investment Limited.
Fathi, B. & Esmaeilian, M. (2012). Evaluation of value added tax (VAT) and tax evasion. Research Journal of Economic Theory. 4(1): 1-5.
Federal Inland Revenue Service (1993a) Value Added Tax Decree. 102.
Federal Inland Revenue Service (1993b). Value Added Tax: FIRS information circular. 9304.
Federal Inland Revenue Service (1993c). VAT on imports: FIRS information circular. 9305.
Gupta, A.S. (2007) Determinants of tax revenue efforts in developing countries. IMF working paper. 07(184). 1-38
Ikpe M. & Nteegah A. (2013). Value added tax and price stability in Nigeria: A partial equilibrium analysis. European journal of government and humanities. 2 (2). 137-147.
Johnson, S., Kaufmann, D. & Pablo, Z (1998). Corruption, public finance and the official economy. Retrieved from http://www.imf.org/external/pubs/ft/wp/2007/wp07184.pdf on 02/05/16
Kaczy”ska, A. (2015) The analysis of VAT revenue in Poland ‘ the size and determinants. (The 3rd Global Virtual Conference). European journal of Social Sciences- Economics and Business. 50-54
Karran, T. (1985). The determinants of taxation in Britain: An empirical test. Journal of Public Policy. 5, 365-386.
Kasnauskien”, G. & Krimisierait”, J. (2015). Using MIMIC models to examine determinants of VAT gap in Lithuania. Organizations and markets in emerging Economies. 6. 1(11). 107-126.
Lutfunnahar, B., (2007). A panel study on tax efforts and tax buoyancy with special reference to Bangladesh. Working Paper 715.
Leahy, E., Lyons, S. & Richard S. J. Tol., (2010), The distributional effects of value added tax in Ireland. Economic and Social Research Institute (ESRI). 42,(2) 213-235.
Legeida, N. & Sologoub, D. (2003). Modeling value added tax (VAT) revenues in a transition economy: Case of Ukraine. Institute for economic research and policy consulting working paper. 22. 1-21
Masayuki Tamaoka, (1994), The regressivity of a value added tax: Tax credit method and subtraction method ‘ A Japanese case fiscal studies. 15(2). 57-73.
Matsuzaki, D. (2003). The effects of a consumption tax on effective demand under stagnation. The Japanese Economic Review. 54 (1). 101-118.
Mogues, T., Samuel, B. & Cudjoe, G. (2009). Do external grants to governments discourage own-revenue generation? Retrieved from http://www.ifpri.org/sites/default/files/publications/ifpridp00934.pdf on 02/05/16
Mullbacher, S. Hyee, R ‘Adam, S. (2013) A Study on the economic effects of the current VAT rates structure. TAXUD/2012/DE/323. FWC no: TAXUD/2010/CC/104.1-172.
New world encyclopedia
Ola, C. S. (1999). Nigeria Income tax and practice. Ibadan, Oyo, Nigeria. Heineman Nigeria Plc.
Ola, C.S. (2001). Nigeria income tax law and practice. London, UK. Macmillan Publishers.
Oserogho & Associates. (2008). Legal alert March 2008 VAT & foreign non resident companies in Nigeria. Retrieved from: http://www.oseroghoassociates.com/pdf/2008
Sancak, C., Velloso, R., & Xing, J., (2010). Tax revenue response to the business cycle, IMF Working Paper 10/71.
Sekwati, I. & Malema, B. W. (2011) Potential impact of the increase in VAT on poor households in Botswania. www.ijeronline.com. 2(1) 47-54
Shalizi, Z & Squire, L. (1988) Consumption taxes in sub-Africa: Building on existing instruments in Gillis, Makolm, Carl S. Shoup and G. P. Sicat (eds) Value added taxation in developing countries. Washington D.C.The World Bank
Statistics explained Retrieved from https://ec.europa.eu/eurostat/statistics explained/mobile/index.php #Page?title=Tax%20revernue%20statistics&lg=en on 12/04/06
Tagkalakis, A.O. (2014). The determinants of VAT revenue efficiency: recent evidence from Greece. Bank of Greece working papers.181. 5-17
Taiwo, F. H. (2016) An empirical investigation of the determinants of Value added tax (VAT) in South western Nigeria. International Journal of Advance in Management and Economics. 5(2). 105-115
Teera, J. M. (2003). Could do better: An appraisal of Uganda’s tax performance relative to sub-Saharan Africa. Retrieved from www.wikepedia.com on 12/05/06
Wawire, N. (2000). Revenue productivity implications of Kenya’s tax system. CSAE Conference Paper. 1-41.
Wilford, S. D., & Wilford, W. T. (1978). Estimates of revenue elasticity and buoyancy in Central America. Taxation and Economic Development. 83-100.
World finance. Retrieved from http://finance.mapsofworld.com/tax/meaning.html on 02/04/16
Worlu, C. & Nkoro, E. (2012). Tax revenue and economic development in Nigeria: a macroeconomic approach. Academic Journal of Interdisciplinary Studies 1(2): 211-231.
APPENDICES
Appendix 1: Data from 1994 to 2014
Years VAT (N’B) INT (N’B) RGDP (N’B) HOC (N’B) COC (N’B)
1994 7.2608 368.9 345.2285 610.3 169.67
1995 20.761 1705.8 352.6462 1,387.4 242.74
1996 31 1872.2 367.2181 2,124.3 280.38
1997 34 2087.4 377.8308 2,091.1 377.78
1998 36.9 1589.2 388.4681 2,371.3 393.55
1999 47.1 2051.4 393.1072 2,454.8 231.29
2000 58.5 2930.7 412.332 2,478.8 393.55
2001 91.8 3226.2 431.7832 3,687.7 403.10
2002 108.6 3256.9 451.7857 5,540.2 478.29
2003 136.4 5168.1 495.0072 7,044.5 450.49
2004 163.2977 6589.8 527.576 8,111.1 785.82
2005 192.6565 10047.4 561.9314 10,099.4 1,003.10
2006 232.6972 10433.2 595.8216 11,834.6 1,283.40
2007 314.5455 12221.7 634.2511 16,135.9 2,131.81
2008 401.7367 15980.9 672.2026 17,166.5 2,871.38
2009 481.4073 14087 718.9773 17,930.8 3,269.93
2010 564.8916 20175.5 54612.2642 22,845.13 4,156.13
2011 659.1536 26232.6 57511.0418 22,840.83 4,979.90
2012 710.5551 24905.8 59929.893 19,536.05 4,852.81
2013 802.6835 24701.4 63218.7217 20,124.86 5,129.06
2014 802.9647 23499.3 67152.79 21,256.89 5,849.99
Appendix 2: Unit Root Tests
Vat at Level
Null Hypothesis: VAT has a unit root
Exogenous: Constant
Lag Length: 2 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -1.988486 0.2886
Test critical values: 1% level -3.857386
5% level -3.040391
10% level -2.660551
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 18
VAT at First Difference
Null Hypothesis: D(VAT) has a unit root
Exogenous: Constant
Lag Length: 2 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -1.589162 0.4662
Test critical values: 1% level -3.886751
5% level -3.052169
10% level -2.666593
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 17
Vat at Second Difference
Null Hypothesis: D(VAT,2) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -6.150909 0.0001
Test critical values: 1% level -3.857386
5% level -3.040391
10% level -2.660551
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 18
INT at Level
Null Hypothesis: INT has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -0.031432 0.9446
Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413
*MacKinnon (1996) one-sided p-values.
INT at First Difference
Null Hypothesis: D(INT) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -3.998385 0.0071
Test critical values: 1% level -3.831511
5% level -3.029970
10% level -2.655194
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 19
HOC at Level
Null Hypothesis: HOC has a unit root
Exogenous: Constant
Lag Length: 3 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -1.766739 0.3827
Test critical values: 1% level -3.886751
5% level -3.052169
10% level -2.666593
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 17
HOC at First Difference
Null Hypothesis: D(HOC) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -3.565000 0.0173
Test critical values: 1% level -3.831511
5% level -3.029970
10% level -2.655194
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 19
HOC at Second Difference
Null Hypothesis: D(HOC,2) has a unit root
Exogenous: Constant
Lag Length: 1 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -7.258006 0.0000
Test critical values: 1% level -3.886751
5% level -3.052169
10% level -2.666593
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 17
COC at Level
Null Hypothesis: COC has a unit root
Exogenous: Constant
Lag Length: 3 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -1.927901 0.3128
Test critical values: 1% level -3.886751
5% level -3.052169
10% level -2.666593
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 17
COC at First Difference
Null Hypothesis: D(COC) has a unit root
Exogenous: Constant
Lag Length: 4 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -3.564364 0.0208
Test critical values: 1% level -3.959148
5% level -3.081002
10% level -2.681330
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
and may not be accurate for a sample size of 15
RGDP at Level
Null Hypothesis: RGDP has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -0.022977 0.9455
Test critical values: 1% level -3.808546
5% level -3.020686
10% level -2.650413
*MacKinnon (1996) one-sided p-values.
RGDP at 2nd Order First Difference
Null Hypothesis: D(RGDP) has a unit root
Exogenous: None
Lag Length: 0 (Automatic – based on SIC, maxlag=4)
t-Statistic Prob.*
Augmented Dickey-Fuller test statistic -3.969888 0.0005
Test critical values: 1% level -2.692358
5% level -1.960171
10% level -1.607051
*MacKinnon (1996) one-sided p-values.
Warning: Probabilities and critical values calculated for 20 observations
Appendix 3: Cointegration Tests
Date: 07/01/16 Time: 03:12
Sample (adjusted): 1996 2014
Included observations: 19 after adjustments
Trend assumption: Linear deterministic trend
Series: INT HOC COC RGDP VAT
Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None * 0.929290 123.6397 69.81889 0.0000
At most 1 * 0.871270 73.30553 47.85613 0.0000
At most 2 * 0.682579 34.35478 29.79707 0.0139
At most 3 0.469595 12.55178 15.49471 0.1323
At most 4 0.026158 0.503610 3.841466 0.4779
Trace test indicates 3 cointegratingeqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegration Rank Test (Maximum Eigenvalue)
Hypothesized Max-Eigen 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None * 0.929290 50.33416 33.87687 0.0003
At most 1 * 0.871270 38.95075 27.58434 0.0012
At most 2 * 0.682579 21.80300 21.13162 0.0402
At most 3 0.469595 12.04817 14.26460 0.1089
At most 4 0.026158 0.503610 3.841466 0.4779
Max-eigenvalue test indicates 3 cointegratingeqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Appendix 4: Ordinary Least Squares Estimation
Dependent Variable: VAT
Method: Least Squares
Date: 07/01/16 Time: 03:14
Sample: 1994 2014
Included observations: 21
Variable Coefficient Std. Error t-Statistic Prob.
C 3.283040 13.50762 0.243051 0.8111
INT 0.008274 0.005635 1.468285 0.1614
RGDP -0.000167 0.000892 -0.187165 0.8539
COC 0.105080 0.021906 4.796894 0.0002
HOC -0.000261 0.004709 -0.055498 0.9564
R-squared 0.988001 Mean dependent var 280.9005
Adjusted R-squared 0.985001 S.D. dependent var 278.2952
S.E. of regression 34.08302 Akaike info criterion 10.09973
Sum squared resid 18586.43 Schwarz criterion 10.34843
Log likelihood -101.0472 Hannan-Quinn criter. 10.15371
F-statistic 329.3538 Durbin-Watson stat 1.353621
Prob(F-statistic) 0.000000
Appendix 5: Error Correction Model
Dependent Variable: D(VAT)
Method: Least Squares
Date: 07/01/16 Time: 03:18
Sample (adjusted): 1995 2014
Included observations: 20 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
C 15.48956 8.039947 1.926575 0.0746
D(INT) 0.001204 0.003400 0.354165 0.7285
D(RGDP) 2.25E-05 0.000602 0.037461 0.9706
D(COC) 0.071500 0.022446 3.185397 0.0066
D(HOC) 0.002243 0.005325 0.421279 0.6799
ECT(-1) -0.560203 0.233128 -2.402987 0.0307
R-squared 0.610757 Mean dependent var 39.78520
Adjusted R-squared 0.471741 S.D. dependent var 33.92528
S.E. of regression 24.65737 Akaike info criterion 9.491354
Sum squared resid 8511.803 Schwarz criterion 9.790074
Log likelihood -88.91354 Hannan-Quinn criter. 9.549667
F-statistic 4.393444 Durbin-Watson stat 1.465754
Prob(F-statistic) 0.012927
Appendix 6: Serial Correlation Test
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.966078 Prob. F(2,12) 0.1826
Obs*R-squared 4.936126 Prob. Chi-Square(2) 0.0847
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 07/01/16 Time: 03:26
Sample: 1995 2014
Included observations: 20
Presample missing value lagged residuals set to zero.
Variable Coefficient Std. Error t-Statistic Prob.
C -1.362908 8.801580 -0.154848 0.8795
D(RGDP) -0.000694 0.000756 -0.917967 0.3767
D(COC) -0.045443 0.031808 -1.428657 0.1786
D(HOC) 0.008594 0.007056 1.218035 0.2466
D(INT) 0.004463 0.004795 0.930737 0.3703
ECT(-1) -0.721825 0.468796 -1.539740 0.1496
RESID(-1) 1.414302 0.885934 1.596397 0.1364
RESID(-2) -0.211059 0.649233 -0.325089 0.7507
R-squared 0.246806 Mean dependent var 4.26E-15
Adjusted R-squared -0.192557 S.D. dependent var 21.16577
S.E. of regression 23.11391 Akaike info criterion 9.407921
Sum squared resid 6411.037 Schwarz criterion 9.806214
Log likelihood -86.07921 Hannan-Quinn criter. 9.485672
F-statistic 0.561737 Durbin-Watson stat 1.897829
Prob(F-statistic) 0.773298