Effect of Tax Policy on Unemployment in Nigeria

This Study examined the effect of tax policy on unemployment in Nigeria, using aggregate time series data from 1981 to 2017. Relevant data was sourced from the Central Bank of Nigeria Statistical Bulletin and were estimated using the Ordinary Least Square technique. The results of the analysis showed that tax policy has significant long run impact on unemployment in Nigeria. Specifically Company income tax, Personal income tax and Customs and excise duty have inverse relationship with unemployment in Nigeria, while Value added tax showed positive relationship. Thus government should factor in tax policy when formulating policies that are meant to reduce unemployment in Nigeria. Based on findings of the study, the following recommendations were made; provision of adequate infrastructures by the government, elimination of multiple taxes as well as adjusting tax rates to encourage investments.


Objectives of the Nigerian Tax System
The Nigerian tax system is expected to contribute to the well-being of all Nigerians and taxes which are collected by government should directly impact on the lives of the citizens. This can be accomplished through proper and judicious utilisation of the revenues collected by government. These objectives include: -To promote fiscal responsibility and accountability -To facilitate economic growth and development. - To provide the government with stable resources for the provision of public goods and services -To address inequalities in income distribution -To provide economic stabilisation -To pursue fairness and equity -To correct market failures or imperfections 2.2 Keynes General Theory of Employment This theory was propounded by John Maynard Keynes, a British economist in 1936. His major work, entitled 'the general theory of employment, interest and money' was first published in 1936. According to Keynes, full employment is possible only when the level of total spending is adequate. If spending is inadequate, unemployment will result. Keynes rejected the classical contention that market economies automatically tend toward full employment; he focused attention on the level of demand or total spending as the critical determinant of an economy's health.
Keynes advocated the use of fiscal policy by central government to manage the level of aggregate demand to preserve full employment and avoid inflation. This involves the manipulation of government spending and taxation in order to guide the economy's performance. When unemployment exists, the federal government should increase its spending on goods and services (without increasing taxes). This will shift the aggregate demand curve to the right and increase the equilibrium level of real GDP and employment. A reduction in income taxes (without a reduction in government spending) will accomplish the same purpose because it will cause households to spend more at any given price level. When inflation exists, government spending should be reduced or taxes increased. These policies will reduce aggregate demand and thus reduce inflationary pressures. Gbosi (2006), defined unemployment as a situation in which people who are willing to work at the prevailing wage rate are unable to find jobs. Unemployment occurs when a person who is actively searching for employment is unable to find work. Unemployment is often used as a measure of the health of the economy and the most cited measure of unemployment is the unemployment rate. Employment Effect of Personal Income tax Many entrepreneurs start as self-employed individuals, who are subject to personal income taxes. Both the level and progressivity of Personal income taxes can influence the behaviour of potential entrepreneurs (OECD, 2002). Studies generally find that lower personal income taxes can increase rates of self employment. A highly progressive income tax structure can discourage entrepreneurs, who may be particularly sensitive to tax rates and schedules as income increases (Gentry & Hubbard, 2000).
According to Skinner (1989), payroll taxes in developed countries are thought to cause efficiency losses because workers vary their labour supply in response to the net wage rate. He views labour demand in developing countries as being consistent with a dual market: the modern or urban sector pays a higher wage to otherwise identical workers than the traditional or rural sector. As a consequence, rural workers migrate to cities as long as the expected value of getting a highly paid job exceeds their agricultural wage rate (Harris & Todaro, (1970). Higher urban wages cause a reduction in rural output (as migration to the cities occurs) and more unemployment. Hence taxes on labour in urban areas would have positive effects, since reducing the net urban wage, migration and urban unemployment could be reduced. Employment Effect of Company Income Tax The corporate tax affects the economy via a number of channels. First, it increases the cost of capital. This raises marginal production costs and through mark up pricing the output price. Accordingly the demand for domestically produced goods falls, which exerts a negative output effect on capital and labour demand (Bettendorf, Horst and Mooij, 2007). Secondly the higher cost of capital induces substitution from capital to labour. Corporate taxes can affect employment by reducing investment and production, and by reducing labour supply to the extent that firms pass on these taxes to employees in the form of lower wages (Cottarelli, 2012). Lower corporate tax rates generate more jobs. Changes in after-tax business profits are positively related to changes in the number of people employed. As after-tax profits increase showing reduction in tax rates, companies have more money to invest in growing their business or to distribute in dividends to investors who have financed business growth. When business grows, they hire more people; employment also grows. On the other hand, when after-tax, profit fall (increase in tax rates), business growth slows, and companies respond by hiring fewer people or eliminating jobs altogether. As a result employment growth slows down or begins to contract. (Myers, 2011) Any increase in the cost of capital including taxes can affect the labour market by reducing output, by inducing factor substitution and by reducing labour productivity. (Eduardo & Farjardo, 2012). The higher cost of capital due to a corporate tax will increase production costs and lead to a fall in output, thereby decreasing the demand for both capital and labour. The corporate income tax may also result in some production moving from the formal to the informal sector, increasing demand for labour in the later, and at least partially countering the negative effect on employment in the formal sector. The overall effect on employment in the economy will depend on the relative labour intensities between the two sectors and the substitutability of labour and capital (Eduardo & Fajardo 2012). Tvrdon (2011), focused on the characteristics of labour taxation and its effect on the labour market, the level of employment or unemployment especially. It uses a panel data regression model over the period 2000 to 2009, using the constant coefficient model, the fixed effect model and the Random effect model. Results show that an increase in the tax wedge decreases the employment rate. The study suggested that reducing the total tax wedge is more likely for the high income groups of the population, which consists of fraction of the total labour force. Garcia & Sala (2006), provides a detailed analysis on the incidence of the tax structure on the labour market for 21 OECD countries over the period 1965 -2003.They use the OLS estimation technique and the Lagrange multiplier serial correlation test. The study develops a model encompassing different wage bargaining systems and the incidence of different types of taxes. The results indicate that the payroll tax bias (PTB) plays a significant role in explaining unemployment in the continental, European countries but not in the Nordic or the Anglo Saxon ones. There is also no relationship between the incidence of the PTB and unemployment persistence, even though there is a positive one with respect to the level of the fiscal wedge. Hagedorn, Mariovskii & Stetsenko (2007), investigate the relationship between taxes and unemployment using data from a number of OECD countries from 1996-2000. It introduces ex ante heterogeneity between workers and two technology shocks, neutral and investment specific as the driving force into the basic mortensen pissarides search and matching model. Results show that countries with higher tax rates have higher aggregate productivity, lower skill premia, and higher unemployment rates among both high-and low-skilled worker. Jukka & Hakan (2011), examine the impact of the Swedish tax policy on employment, using data on individuals, instead of countries. The study focused on the micro-economic part of research, drawing on both applied theory and micro-econometric evidence and employed the difference in differences method for the data analysis. Results show that the current Swedish tax system has fairly favourable effects on employment. The study suggested as follows; better targeting the earned income tax credit at families and low income workers, lowering the top marginal tax rates and maintaining the tax incentives for old workers.

Review of Empirical studies
Carroll & Prante (2012), uses the EY general equilibrium (EYGE) model of the US economy to evaluate the impact of the increase in the top tax rates in the long-run. The report finds that the higher marginal tax rates result in a smaller economy fewer jobs, less investment and lower wages. Eduardo & Fajardo (2012), empirically explores the effects of payroll taxes, Value added taxes and corporate income taxes on a variety of labour market outcomes such as employment, unemployment, informality and wages. The study used national -level data on labour variables for 15 Latin American countries and adopted the OLS fixed effect estimation model. Results show that payroll taxes reduce employment and increase labour cost when their benefits are not valued by workers. Value added taxes increase informality and reduce skilled labour demand. Corporate income taxes may help reduce informality, especially among low education workers but when tax enforcement capabilities are strong may reduce labour participation and employment of mediumand high-education workers. Bettendorf, Horst & Mooij (2007), analyse the impact of corporate taxes on structural unemployment in a general equilibrium framework. It adopts a union bargaining model to explain equilibrium unemployment on the basis of several institutional variables. The bargaining framework is embedded in an applied general equilibrium model for the European Union that is designed for analyzing corporate tax policies. This so called CORTAX model encompasses various distortions of corporate taxation, including the marginal investment distortion, international spillovers from foreign direct investments and profits shifting by multinationals and distortions in the financial structures of companies. Results show that although the effects of corporate tax on unemployment may be smaller than the effects of labour taxes and value added taxes, the welfare costs of corporate taxation are typically larger for most European countries under plausible parameters, especially under strong international spillovers. Parmentier (2006), analyses the effect of the marginal tax rate on unemployment and economic efficiency in a matching model with homogeneous agents where wages and working hours are bargained over. Results show that the theoretical impact of a higher marginal tax rate on unemployment is ambiguous whatever the instantaneous utility in unemployment that is either fixed or perfectly indexed on net wages. Separovic & Zagreb (2009), examine the influence of the tax wedge on employment in OECD countries in comparison with Croatia. The study used descriptive, discriminate and cluster analysis with figures from OECD countries. Results show that Croatia has high tax wedge which has negative effects on employment and is partly responsible for the high unemployment. The study recommended that Croatian government should work towards reduction of the high tax wedge, by increasing personal allowances, reducing income tax, reducing contributions by employers and employees. Bassanini & Duval (2006), use cross-country/time series data from 21 OECD countries over the period 1982 to 2003 to explore the impact of policies and institutions on employment of OECD countries in the past decades. They employ the reduced-form unemployment equations, consistent with standard wage setting/price setting models. Results show that in the "average" OECD country, high and long lasting unemployment benefits, high tax wedges and stringent anti-competitive product market regulation are found to increase aggregate unemployment. Primo & Laporšek (2010), assess the characteristics of labour taxation for five different groups of workers and labour market performance (in terms of employment and unemployment rate) in the EU and to examine whether tax wedge affects employment growth in the EU. The descriptive empirical estimates show that the level of labour taxation varies greatly across EU Member States, by which the tax wedge tends to be higher among new member States (excluding Cyprus and Malta). Furthermore, the panel regression analyses confirm statistically significant negative relationship between tax wedge and employment growth in the EU as a whole.
Zirgulis & Sarapovas (2017), study the effect of corporate taxation on unemployment utilizing a dynamic panel covering 41 countries over 11 years. The study investigates how changes in the corporate income tax affect unemployment. The study employs system general method of moments (GMM) due to peculiarities of the data set and the endogeneity issues present in the research problem. Findings show that a rise in the effective average corporate tax rate significantly increases the unemployment levels.
Meyer (2018) studies the impact that lowering the corporate income tax rate can have on unemployment levels in a country, using an empirical study of 15 European Union member countries, mainly due to the availability of data and the unique mobility of capital. The study utilises the ordinary least square regression equation. Findings indicate a direct positive relationship between the tax rate and unemployment rates. Kugler & kugler (2001), investigate the effect of higher payroll taxes on employment and wages. They use a panel of manufacturing plants from Colombia and exploit the time series and cross-section variations in payroll taxes over the 1980s and 1990s. The study employed the basic differences and differences-in -differences (or differences -in -differences-in -differences when both production and non production workers are included). Results show that a 10 percent increase in payroll taxes lowers employment by about 4 percent. Pissarides (1998), examines the effects of employment tax cuts on unemployment and wages. Different models are used to evaluate the impact of policy by simulating the effect of tax changes in four partial equilibrium models: A competitive model, a Union model, a search model and an efficiency model. In general, tax cuts have smaller employment effects in the competitive model and larger one in the efficiency wage model (both of which imply that the tax structure is irrelevant).

Research Design
The time serial ex-post-facto and analytical research designs were used to investigate the relationship between tax policy and unemployment in Nigeria.

Data and Variable Description
This study was based on secondary data. A sample of annual observations on time series covering the period from 1981 to 2017 was employed. Series are in current domestic currency. Most series were collected from the Central Bank of Nigeria statistical bulletin (various editions) while the others were obtained from the Bureau of National statistics (abstract of statistics), and the World Bank (International Development Indicators). The specification of the econometric model adopted in this study, builds on theoretical propositions.

Estimation Procedures
-The characteristics of the time series data used in the analysis are first evaluated. The summary statistics of the various variables are estimated. The statistical properties of the variables provide information about the means, medians, standard deviations, skewness, kurtosis and jarque-Bera statistics of each variable. Standard deviation measures dispersion in the series. Skewness measures asymmetry of the distribution of the series around its mean and it is expected to be zero for normal distribution. Kurtosis measures the peakedness or flatness of the distribution of the series while Jarque-Bera is a test statistic for testing whether the series is normally distributed. The correlation analysis shows the extent of linear relationship that exist among the variables employed in the study. Thereafter, the stationarity properties of the employed data are examined. -linear combination of two non-stationary series may be stationary. The existence of a stationary linear combination from the non-stationary time series is referred to as co integration and it can be interpreted as a long-run equilibrium relationship among the variables. Given this consideration, Maximum Eigenvalue test of Johansen Co integration test is carried out based on the assumption of linear deterministic trend.

-
The long run analysis is estimated using Ordinary Least Square (OLS) estimation technique.

Statistical Properties of the Variables
The characteristics of the time series data used in the analysis are presented in Table 4.1. The variables are systematically distributed. The outcomes of each variable's mean, median as well as values for their maximum and minimum suggest that the variables are well behaved. The mean values of all the variables employed are not too different from their respective median values. This is an indication of absence of excessive outliers and stability of the variables employed, which are essential for the analyses carried out in this study. The value of the standard deviation of each of the variables is a further proof of the fact that the distribution of the variables is approaching normal distribution. In addition, the skewness, kurtosis and standard deviation statistics show that the variances of the variables are not unnecessarily large.
The probability values of the Jarque-Bera Statistics as presented in the table show that LUNE, LTRP, LINV, LPIT, LCIT, LCED and VAT are normally distributed.

Correlation Analysis
The correlation matrix of the variables employed in this study is presented in Table 4.2. The table presented all possible bivariate combinations of all the employed variables. The correlation analysis also gives insight into understanding the econometric results and other analyses that are later carried out in this study. The result as presented in Table 4.2 showed that most of the variables employed are highly correlated. The directions of the correlation for some are positive, while negative for others.

Unit Root/ Stationarity Test
The variables employed in the analysis are tested for stationarity using the Augmented Dickey-Fuller test to determine whether they are stationary or non-stationary series. The tested null hypothesis is the presence of a unit root. The results of the unit root tests as presented in Table 4.4 indicated that LTRP, LPIT, UNE, LCIT, LCED and LVAT are stationary at after first difference. This implies that they are integrated of order one. The other variable namely, LINV is stationary at level, indicating that is integrated of order zero.

Tests for Co-integration
The UNE model, which is specified to examine the relationship between tax policy and unemployment in Nigeria, is tested for the null hypothesis of no co-integration assuming linear deterministic trend. Comprised in the UNE model are UNE, LINV, LTRP, LVAT, LPIT, LCIT and LCED. Table 4.5 illustrates the outcome of the co-integration test for the UNE model. There are three co-integrating relations among the variables in the UNE model as indicated by the Max-Eigen Statistic. This implies that there are long run relations among the variables employed in the UNE model.  Table 4.6 above. The result show that LINV, LTRP, LPIT, LCIT and LCED exert negative influence on the UNE, while LVAT has positive impact on UNE. All the explanatory variables included in the model are significant except LPIT.
About the overall significance of the UNE model, the Adjusted R 2 coefficient (0.511) suggests that about 50 per cent of the total variations in UNE are explained by the variables included in the model, which are LINV, LTRP, LVAT, LPIT, LCIT and LCED. Similarly, the F-statistics shows that the UNE model is statistically significant.

Discussions
i. Findings of the study indicated a positive relationship between Value added tax and Unemployment. The impact of Value added tax on unemployment is also significant. This result means that increases in Value added tax worsen the unemployment rate. In theory, the burden of Value added tax is supposed to rest on the final consumers. It is the final consumer that pays the tax. But often, the incidence of the tax depends on the elasticity of demand and supply of the goods in question. Under competition and where demand is elastic, it will be difficult to shift the burden of the tax to the final consumer in form of higher prices. The informal economy and tax evasion is prevalent in developing countries like Nigeria. This makes it difficult for some companies to fully shift the tax burden to the consumer. In other words the producers end up bearing the burden. This results in higher cost of production and adverse effect on employment. Therefore increases in Value added tax revenue leads to increase in the rate of unemployment. The steady rise in the amount of revenue derived from Value added tax annually may have contributed to the rising unemployment rate in the country.
Ii The results of the analysis also indicated a negative relationship between the Company income tax, Personal income tax, Custom and excise duty and Unemployment Rate. This means that as these taxes increase, unemployment falls. The effects of Personal (Labour) Income taxes on employment (labour Demand) depend on whether gross wages are affected or not by the tax. If an income tax increase is completely absorbed by the workers or an increase in company income tax borne by the buyers , labour demand and unemployment will not be affected directly (Eduardo & Farjado 2012).

5.1.Conclusion
The major objective of the study was to examine the effect of tax policy on unemployment in Nigeria. The study covered the period 1981 to 2017 and used aggregate time series data to examine the effect of tax policy on unemployment.
The analysis carried out in the study began with the summary statistics of the variables employed. It also presented the correlation analysis of the employed variables. Thereafter, co-integration relations among the variables were verified using Johansen co-integration test. The specified model was estimated through the ordinary least square estimation technique to ascertain the extent of the impact of the independent variables on the dependent variable. The results revealed that Personal income tax, Company income tax and Custom and excise duty have inverse relationship with unemployment rate in Nigeria, with Value added tax showing a positive relationship.
The study therefore concludes that tax policy has significant long run relationship with unemployment in Nigeria and therefore imperious for the government to factor in tax policy when formulating policies that are meant to reduce unemployment in the Nigerian economy.

Recommendations -Provision of Infrastructures
There are peculiar challenges in the Nigerian investment environment, such as infrastructure which may negatively impact investment decisions as the cost of doing business in Nigeria, may be significantly higher than it would have been in countries with better infrastructure. Through policies that encourage domestic and foreign investment, there will be greater employment opportunities in Nigeria which will help distribute the benefits of economic growth. The Tax Strategy should therefore encourage investment that creates greater employment opportunities. -Adjusting Tax Rates to Encourage Investment The strategy of Government will be to reduce the rate of Companies Income Tax to 20% or less and to decrease the top rate of Personal Income Tax to about 17.5% of taxable income. Decreasing income tax rates, will encourage investments, create greater employment opportunities and increase tax compliance. -Import and Excise Duties In order to reduce costs of Nigerian manufacturers, and make their products more competitive, government should reduce import duties on raw materials to zero percent (where this would not conflict with international / regional obligations, such as ECOWAS commitments). This will encourage the production of both intermediate and finished goods and develop the Nigerian economy.

-Elimination of Multiple Taxes
Streamlining and harmonising taxes across the federation would increase Nigeria's productive output. Strict enforcement of the taxes and levies approved list for collection Act No 21 of 1998 is recommended as well as the abolition of the law on withholding tax on dividend, interests etc.