THE EFFECT OF TAXES ON DIVIDEND POLICY OF BANKS IN NIGERIA.

Mr. Matthias A. Nnadi1 & Dr. (Mrs.) Meg Akpomi2 Rivers State University of Science & Technology, Port Harcourt, NIGERIA.3

Mr M.A.Nnadi is an Assistant Lecturer in the Department of Business Education, Rivers State University of Science & Technology, Port Harcourt, Nigeria
2

Dr (Mrs.) M.Akpomi is a Senior lecturer in the Department of Business Education, Rivers State University of Science & Technology, Port Harcourt, Nigeria. Email: megakpomi@yahoo.com
3

All correspondence should be addressed to Mr. M.A.Nnadi, 22, Norfolk Street, CV1 3BX, Coventry, UK. Email: matthiasnnadi@yahoo.com Tel. +44(0) 78598 123 66

THE EFFECT OF TAXES ON DIVIDEND POLICY OF BANKS IN NIGERIA.

Abstract
The study explores the impact of taxes on the dividend policy particularly in banks in Nigeria. The study was set out to examine the relationship of profit, dividend and taxes especially in the banking sector. The research underscores the theoretical assumptions of the M&M theory. The standard multiple regression analysis was applied in testing the hypothesis. The study identified pattern of past dividends, concern about maintaining a target capital structure, current degree of financial leverage, shareholder need for dividend income, legal rules and constraints; such as impairment of capital, the desire to send favourable signals to investors, the desire to conform to the industrys dividend payout among factors influencing dividend policy of banks. The analyses of the study show a significant correlation between taxes and dividend structure of the banks and also suggest that profit is a major variable in the formation of dividend policy of the organisations. This is supported by the hypothesis, which showed significant effect of profit on dividend and a positive correlation between profit, tax and dividend. The finding corroborates the postulations of some financial theorists and recommends capital gains in lieu of dividend for high taxpayers and that an adoption of a dividend policy by banks particularly in Nigeria should be strictly considered based on the unique circumstances of the bank and not necessarily based on age long traditional factors often formulated by academics.

Operational Definitions of Terms

Dividend policy dividend policy is primarily concerned with the decisions regarding dividend payout and profit retention by companies. Capital gain ---- tax paid on profit made on sell of shares or stocks. Dividend income income derived from payment of dividend of companys shares Clientele effect -- - the term used to classify investors into particular groups based on their interest on dividend and the tax effect

Background of the research

A lot of controversies regarding taxes and dividend policy have attracted many academic interests. Financial theorists such as Brennan (1970), Masulis & Trueman (1988) have stipulated that taxes affect organisational corporate dividend policy. If this speculation were true, changes in corporate dividend payout would be expected whenever the government changes its income tax policy (Wu, 1996). However, this does not always apply especially in especially in the banking business. Linter (1956) had asserted that the major determinants of dividend policy are the anticipated level of future earnings and the pattern of past dividend. This discrepancy may have underpinned Modigliani & Miller (1961) theory, which provided a platform for the enormous debates and researches on dividend policy. The banking sector is of interest in this research because of the because of the structure of its dividend as revealed in the study. Dividends are usually paid to owners or shareholders of business at specific periods. This is apparently based on the declared earning of the company and the recommendations made by its directors. Thus, if there are no profits made, dividends are not declared. But when profits are made, the company is obligated to pay corporate tax including other statutory taxes to the government. This is an essential corporate responsibility particularly profit making companies. The taxes no doubt reduce the profits available at the disposal of the organisations, either to be retained or distributed as a dividend to shareholders of the company. For decades, several postulations and assumptions have been made regarding whether such taxes paid by organisations actually affect their pattern of dividend policy. Dividend policy is the trade-off between retaining earning and paying out cash or issuing new shares to shareholders. Some firms may have low dividend payout because management is optimistic about the firms future and therefore wishes to retain their earnings for further expansion. It is hard to deny that taxes are important

to investors. Although, dividend affects the shareholders tax liability, it does not in general alter the taxes that must be paid regardless of whether the company distributes or retains its profit (Brealey, Myers & Marcus 1999). Conscious of these assumptions, surrounding dividend policy, and this study is directed at evaluating the effects of taxes on the dividend policy of banks in Nigeria.

Statement of the Problem

The debate regarding dividend policy is yet unending. Academicians have postulated several theories on what an idea dividend policy should be. There seems to be a chasm with what is really obtainable in practice. There are extraneous factors dictating the tune of any policy to be applied by organisations. There is also the link of taxation and the impact it exudes on the formulation of dividend policy. Financial theorists such as Wu (1996) etc opined that evasion of taxes by companies is a key factor in the determination of the extent to which its dividend policy is affected. Miller & Scholes (1982) however admits that taxes weigh tremendous influence on corporate dividend structure. In turn, there are contentions on the effects of dividend policy on companys share price. Whether these are true has remained a matter of intense debate. The dividend irrelevant theory of Modigliani & Miller (1961) which assumes a perfect market is still very much held in contention but its principles underlines most companies policies. Three contemporary schools of thought have emerged with theories all attempting to explain the dividend structure and the impact of taxation. These include the dividend irrelevant theory, bird in hand theory and the tax preference theory. The dividend irrelevance theory assumes the absence of brokerages or taxes. This is may be seem as presumptuous and unrealistic in practice, as taxes are inevitable in the business world. The seemingly popular school led by Modigliani & Miller (M&M) claims that dividend policy makes no difference in a world without taxes, transaction costs, or other market imperfection. They further postulated that companies could raise shares whenever they need capital to finance the business. There are conservative theorists who believe that an increase in dividend payout increases the value of the firm. They are the bird in hand theorists who assert that investors will prefer dividends to capital gains generated by growth, as dividends are more certain. Another school of thought, the tax preference theorists agree that an increase in dividend payout reduces value. They reason that since dividends are taxed as ordinary income and capital gains, investors will prefer capital gains to dividends. This group strongly assert that whenever dividends are taxed more heavily than capital gains, the firms would pay the lowest cash dividend they can get away with. The available cash should be retained and reinvested or used to repurchase the shares.

The avalanche of opinion regarding dividend policy and taxes is not only examinable but mind probing for academic research. This research is therefore poised at making further contributions by examining the theories but with specific application to the banking sector in Nigeria.

Research Questions
The following research questions were drawn to guide the study. 1) What factors affect dividend policies of banks in Nigeria? 2) What is the relationship between taxes and dividends? 3) How does profit affect dividend policy?

Research Objectives
The research is set out to achieve the followings: 1) To investigate the factors that affect dividend policies of companies. 2) To examine the effect of taxes on dividend policy of banks in Nigeria. 3) To find out the influence of profit on dividend policy of banks in Nigeria. 4) To make recommendations on ways of achieving effective dividend policy.

Statement of Hypothesis
The hypothesis to be tested in this study is tripod as it relates to profit, tax and dividend. The research has as one of its objectives, the examination of the effect of taxes on the dividend of banks in Nigeria. The results of the previous researches seem to suggest a strong correlation among tax, profit and dividend and therefore streamline the focus of the study. However, despite the conflicting and mixed evidence in the literature concerning the extent of the relationship between profit and tax, the resulting impact is expected to be very significant. Therefore, a test of the null hypothesis will permit an examination of the significant level of the expectations.

H o : There is no significant effect of profit on dividend and tax of banks in Nigeria. In line with previous researches and body of literature reviewed, the present study has an expectation that tax impact strongly on the dividend policy of banks in Nigeria. In particular, the evidence tends to purport the expectation that dividend and tax are strongly correlated.

The Significance of the Study

The result of the study will be very useful to organisations in articulating a viable dividend policy. It will also enhance the proper understanding of the intricacies of the dividend theories and the effects of their application on the organisations. The study of dividend policy and tax are apparently one of unending debate and but with intriguing findings. Past findings and contributions to the subject have made inroads to shaping and providing a better understanding of how companies should view any policy relating or affecting their dividend. Enormous scholarly literatures have been developed as a result. This study is yet another contribution to the existing literature in the study of dividend policy.

The Meaning and Types of Dividend Policy

In discussing the meaning of dividend policy, it is important to highlight what a dividend is. A dividend is simply the money that a company pays out to its shareholders from the profits it has made (Doughty, 2000). Such payments can be made in cash or by issuing of additional shares as in script dividend. Davies & Pain (2002) however defined it as the amount payable to shareholders from profit or distributable reserves. Companies that are listed in the stock exchange are usually obligated to pay out dividends on a quarterly or semi annual basis. The semi annual or quarterly payment is referred to as the interim dividend. The final payment, which is usually paid at the end of the financial year of the company, is known as the final dividend. Dividends are normally paid after the corporate tax has been deducted. Dividend policy is primarily concerned with the decisions regarding dividend payout and retention. It is a decision that considers the amount of profits to be retained by the company and that to be distributed to the shareholders of the company (Watson & Head, 2004). Theoretically, there are different types of dividend policies. These include constant payout, progressive policy, residual policy, and zero policy and non-

cash policy. Investors are seen to belong to a particular group or clientele. This is because they tend to pitch their tent with a particular policy that might suite them. This is the clientele effect of dividend policy (Hutchinson, 1995; Kolb & Rodriguez, 1996). Constant or fixed policy: The Company pays out a fixed amount of its profit after tax as dividend. Thus, the company maintains a fixed payout ratio of dividend. A company may as a matter of policy, decide to constantly payout sixty percent of its after tax profit as dividend to its shareholders and retaining the remaining fraction. This type of policy allows the shareholders the opportunity to clearly know the amount of dividend to expect from their investments in the company. However as noted by Watson & Head (2004), the policy could be traumatic to companies experiencing a volatile or fluctuating profit earning. This is because of the uncertainty of its profit. If capital projects are to viable capital projects, the policy can be chaotic. Progressive policy: Payments on dividend is on a steady increase usually in line with inflation. This could result in increasing dividend in money terms. The firm uses the policy as a ratchet. Every effort is made to sustain the increase even though marginal. Seldom, the company may be constrained to cut down on dividend payout. This is to enable it sustain its operations. This though not a frequent action as it sends a wrong signal to investors. Firms operating this policy will opt to avoid paying dividends during the period rather than consistently cut down on the dividend (Kolb & Rodriguez, 1996). Residual policy: Dividends are just what is left after the company determines the retained profits required for the future investment. This policy gives preference to its positive NPV (Net Present Value) projects and paying out dividends if there are still left over funds available. Dividend becomes a circumstantial payment only paid when the investment policy is satisfied. There is a tendency therefore that this type of policy could give rise to a zero dividend structure. Firms may need to modify this policy to ensure that investors of the different clienteles are not chased out by a strict application of the policy (Kolb & Rodriguez, 1996). Zero dividend policy: Some firms may decide not to pay dividend. This is especially common in newly formed companies that rather require capital to execute its projects. All the profit is thus retained for expansion of the business. Investors who prefer capital gains to dividends because of taxation will naturally be lured by this kind of policy. This type of policy is quite easy to operate and avoids all the costs associated payment of dividends (Watson & Head, 2004). Alternative policies to paying cash: In order to give shareholders a choice between dividends or new shares, the company might choose to buy back shares. This is share or stock repurchase. This has a significant advantage in terms of tax to the shareholder. While the dividend is fully taxed just as ordinary income, the stock repurchase or buyback is not taxed until the shares are sold and the shareholder makes a profit or capital gain (Ross, Westerfield & Jordan, 2001). There is also the policy of stock dividends and split. Shareholders are given additional shares in lieu of cash to the shareholders (Brealey, Myers & Marcus, 1999).

Some Factors Determining Dividend Policy

The choice of a particular dividend policy by a firm is not usually accidental. It is tailored to either meet the firms need or the shareholders. Shareholders have different choice of dividend depending on their needs. Firms also adopt policies that suite their peculiarity. Thus some identifiable factors affecting dividend policy include but not limited to the followings: reinvestment required for new capital projects, availability of other sources of funds, existing cash levels and liquidity, market reaction to a change in dividend, tax, shareholders preference for income or capital gain etc. Market reaction to a change in dividend: Management often sticks to a pattern of dividend payout. This can be explained by the signal conveyed by an alteration in dividend payment of the company. A cut in dividends will give a negative signal to investors on the future prospect of the company (Hutchinson, 1995). This might affect the share price of the company. Thus since management would not give such weak impression about its company, it is likely to stick to its dividend policy that does not attempt to cut down on the current payout. Shareholders preference for income or capital gain: Investors are classified according to their preferences. They are grouped into clienteles, defined by their current income needs. The wealthy investors may not have an immediate need for cash dividend. Others groups may have strong need to receive cash as dividend. By targeting a particular clientele or group, management may wish to tailor its dividend policy to suite the needs of its target investors (Hutchinson, 1995). There are reasons why an investor may wish to avoid or seek dividends other than cash need. An investor who is concerned about taxes will opt for little or no dividend to be paid to him. This is because of the tax levied on dividend income. In attempt to avoid tax, some will prefer not to receive dividend but to pay a capital gain tax at a later time when their shares are sold and profit made. Thus, a firm will structure its policy to suit its investors (Kolb & Rodriguez, 1996). Existing cash and liquidity level: The tendency to pay dividend is strictly based on availability of fund. This is a residual dividend model, which holds that companies will usually base their dividend policy on its cash flow and capital structure. Thus, where the company requires capital for investment for a project, the retained earning is utilised and the amount left for dividend will be affected. Only the residual earning is used for payment of dividend (Samuels & Wilkes, 1987). As can be seen from the few factors considered so far, tax plays an important point in dividend decisions particularly for the shareholders in attempt to make tax savings. The firms may in turn structure their dividend policy to suite shareholders clientele. Thus weighing a substantial influence in dictating how the dividend policy should be.

There are however, other issues that are crucial in the dividend policy of any business organisations. These include the clientele effect, signalling hypothesis, dividend stability, dividend extras, residual dividend model, share prices etc. Some investors prefer no dividends while others may prefer large dividends. Investors will therefore invest in companies that have suitable policy for them. Changes in the amount of dividend sends a signal to the investors. When dividend increases, it is a signal of growth and a reduction signals otherwise. Thus, companies that have stabled or predictable dividend enjoys the goodwill of investors. The bank dividend policy is found different from other industries as it does not react to the Rozeff model elements such as past growth rate, beta, and other acute factors (Casey and Dicken, 2000). In a similar study, Baker, Powell and Veil (2001) segmented the factors influencing dividend policy and financial and non-financial firms. Their findings include pattern of past dividends, concern about maintaining a target capital structure, current degree of financial leverage, shareholder need for dividend income, legal rules and constraints; such as impairment of capital, the desire to send favourable signals to investors, the desire to conform to the industrys dividend payout ratio, investment considerations such as the availability of profitable investments, expected rate of return on firms assets etc. The findings conform to the factors analysed, which depicts the congruence of the dividend factors, irrespective of the industry.

Some Theoretical Models to Explain Dividend, Taxation and Share Price

Corporate dividend policy has captured the interests of financial economists and has been an issue of intensive theoretical models and empirical examinations over the last few decades (Frankfurter & Wood, 1997). The controversy probably began with the Modigliani & Miller (M&M)(1961) theory, which classified investors into dividend clientele. It was later found in a later research that tax is responsible for marginal alterations in the portfolio composition rather than the major differences predicted by Miller & Scholes (1978). At the onset, when the M&M theory was postulated, most financial practioners and many academics greeted the conclusion that tax had a marginal effect on dividend policy with surprise (Baker, Powell & Veit, 2001). Other models such as tax adjusted model, Masulis & Trueman (1988) Model of cash payments, Farrar & Sewlyn (1967) Model of after tax income of investors, Auerbach (1979) Model of shareholders wealth and Akerlof (1970) Signal Model, information asymmetric theories, Jensen (1986) free cash flow hypothesis, Feldstein and Green (1983) theoretical dividend behavioural models and Shefrin & Statman (1984) theory of self control are all measure to unmask firms dividend policy. The tax adjusted model group investors into tax dividend clienteles. It purports that investors are happy with shares that offer higher returns but will accept lower value for shares due to the tax liability of the dividend. However, because investors have different tax liabilities, their choice of investment based on companys dividend policy will be at variance with each other. This agrees with the propositions of

Modigliani (1982) and Masulis and Trueman (1988) models of grouping investors into tax clienteles. It is believed that dividend payment decreases or increases in opposite direction with tax liability (Frankfurter & Wood, 1997). The strategy of tax sheltering of income especially by high tax investors was instrumental to the Miller & Scholes (1978) and Miller (1986) attempt to inject rationality into the tax-adjusted model. An investor can choose to either buy or decline to buy shares with high dividend as a technique of avoiding the apparent tax liability on such shares. Investors could turn to tax free shares to counter balance it with the tax incidental on dividend paying shares. However, DeAngelo & Masulis (1980) disagreed on the effective effect of the tax sheltering strategy but rather advanced personal tax shelters to minimise or totally avoid tax. The study reaffirms Bradley, Capozza and Seguin (1998) hypothesis on dividend policy and cash flow uncertainty. They assert that when a shareholder is confronted with a greater tax rate on dividend rather than capital gain, then the coefficient associated with dividends will be significantly affected However, the finding of Black and Scholes (1974) which set out to examine the impact of profit on the dividend policy of companies fell short of any concrete findings as no correlation was established in the two variable. This contrasted with an earlier finding of Ramaswamy (1979) in which a significant relationship was established. But Miller and Scholes (1982) attributed any significant relationship between profit and dividend to dividend information effect rather than the effect exerted by tax (Watson and Head, 2004). . The after tax income model of investors also classifies investors into two clienteles. This is based on their preference for dividend or capital gains. Farrar & Selwyn (1967) agree that share repurchase rather than dividend should be implored in distributing the companys profits to shareholders. This model was further expatiated by Brennan (1970) into an equilibrium framework. But as stated by Frankfurter & Wood (1997), an equilibrium with dividend paying firms is not consistent with a zero required return per unit of dividend yield. The shareholders wealth model assumes that the shareholders are the major priority in wealth maximization rather than the firm. Where there is an option of capital gain or tax on dividends, the idea of wealth maximization seizes to be on the companys market values. Auerbach (1979) in this model sees dividend as a child of circumstance; which results following a long-term under valuation of capital of the firm and a reinvestment of the earnings after deducting of expected returns of the firm and investors. Akerlof (1970) came up with the signal model of corporate dividend policy. The crusaders of the signal model affirm that when dividend is used as a method of presenting the message of a firms quality across bears a lower cost than other available alternatives. It implies that other methods of putting across good performance of the firm are not reliable, thus the use of dividend. This is invariable because of the asymmetric nature of information regarding the firm. The asymmetric

of information between the owners of the business and the management warrants the payment of dividends to restore dividend and adds value to the firms equity (Frankfurter & Lane, 1992). It is an aid of allaying investors fears and helps the firm to remain competitive in the industry. Other contenders of the information asymmetric have developed thought on applying the agency cost in modelling the dividend policy. This is exponential on the reduction of costs incurred by management. It is firmly believed that huge dividend payments reduce the amount of capital at the disposal of the managers. Thus requiring management to seek fund in the capital market to finance investment projects. This will thus ensure that such funds are effectively monitored by the capital market and to ensure prudence in spending by management (Easterbrook, 1984). In line with this thought, Jensen (1986) postulated the free cash flow hypothesis, which is an extension of the agency cost idea and the signalling model. It hinges on the prudence of managers on the use of funds available after financing the capital projects of the firm. The remaining fund, he assumes, could cause conflict of interest between the shareholders and the managers. Thus, the separation of the ownership and management affords the managers the opportunity to demonstrate their worth. However, the funds left after dividend and interest payments may become too marginal that unscrupulous managers have little to squash. The Feldstein and Green (1983) formulated the theoretical behavioural model of dividend in attempt to prove that dividend is a product of several factors. The model asserts that the tax liability of dividend is marginal if compared with the transaction costs of trading shares when the firms profits are retained. Dividend payment reaffirms trust and confidence in management and thus serves as a signal factor. The theory of self-control is used by Shefrin and Statman (1984) to explain the investors preference for dividends and capital gains. The model posits that capital gains and dividends are not close substitutes. Investors are less patient with when dividend is delayed. This can be verified when dividends are reduced. Thus dividend payout provides more checks on spending level of management. Dividend provides opportunity to spend rather than save but the opposite applies to the capital gains when earnings are retained. In a similar study of banks dividend payout, Casey and Dickens (2000) implored Rozeff (1982) model to examine dividend payout factors in which he found five variables to be very significant in dividend payout policies of firms. The variables include beta, percentage of insider ownership, past revenue growth rate, forecasted revenue growth and the number of common stockholders. Those variables are equally very useful in articulating dividend policies of banks. Chang and Rhee (1990) added that financial leverage is a crucial factor in dividend policy of firms. A firm that has a high financial leverage tend to have a high dividend payout ratio. This however depends on the tax chargeable on the dividend income being higher than the capital gains.

The Structure of the Nigerian Tax System

The Nigerian tax is structured to reflect the nature of the business. Depending on the type of business, taxes are levied on businesses on an annual basis. This implies that all businesses, organisations and taxable persons are obligated to make a tax return to the Inland Revenue. Profits arising from transactions of companies constitute taxable income following their assessment to tax. This also includes personal income tax, which is duly imposed on individuals by the relevant tax authority in the territory where the company has its principal office, or the place of business on the first day of the year of assessment or year of commencement of business. The state Board of Internal Revenue is responsible for the administration and collection of the relevant tax in while the Federal Inland Revenue is charged with the responsibility of the company and other related taxes. These structures are governed by the Personal Income Tax Act (PITA, 1993) and Company Income Tax Act (CITA, 1994) respectively (Anyaduba, 2004). The Nigerian tax system is prudently organised in order to effectively enhance the collection of taxes and reduce the incidence of tax evasion and the subsequent loss of revenue to the government. The tax laws therefore provides for the collection of taxes at source of the taxpayers income. This is achieved through the Withholding tax system, which allows taxes to be deducted at the source of income. This is provided in sections 63 of CITA and 72 of PITA; where the laws recommend that income tax assessable on any company, whether or not an assessment has been made, shall if the Board (the relevant tax authority) so directs, be recoverable from any payments made by any person to such company. The income tax recovered under the provisions of the law by deduction from payments made to a company or person is usually set-off for the purpose of collection against tax charged on the company or individual by an assessment. This is however only to the extent that the total of the deductions does not exceed the amount of the assessment and in case of a company, the deduction is limited to the period in which the payment relates (CITA, 1994). The implications of the above provisions of the tax laws have been summed up by Anyaduba (2004) as follows: The provisions are applicable to both individuals and companies who are liable to tax. It is not limited to any particular source of income with specific reference to those mentioned in the CITA. Any tax deducted at source on withholding tax basis is regarded as payment on the account and it is set-off against any tax assessed, due and collectible from the company or individual. The Nigerian tax laws provide that where a dividend or such other distribution becomes due from a taxpayer; person or company, payable by a Nigerian company to

any other company or to any person, the company paying such dividend or making such distribution shall on the date when the amount is paid or credited, whichever comes first, deducts tax at the rate specified in the Act and shall forthwith pay over to the relevant tax authority the amount so deducted (CITA, section 62 and PITA, section 70). The relevant tax authority for the purpose of withholding tax on dividend is determined on the basis of the residence of the taxpayer and not the location of the company paying the dividend. The tax rate of withholding tax on dividend is 10%. The tax is deducted at source of the dividend. Once paid over to the relevant tax authority is final for both residents and non-residents. The Nigerian tax system however makes an exemption since the 1986 tax year towards the treatment of all withholding tax incomes received from abroad. Section 74(PITA) states that all incomes, including rent, interest, dividend, royalties, directors fees etc which are brought into the country through Government approved channels are exempted from tax and hence from any withholding tax. This means that such incomes, which include dividends, must be brought in and paid into a domiciliary account with an approved bank. The approved banks are those authorised to deal in Foreign exchange under the Second Tier Foreign Exchange Market.

The capital gains tax

Another interesting aspect of the Nigerian tax system is the treatment of the Capital gains tax (CGT). Capital gain is deemed to arise on the disposal of assets by a person where any capital sum is derived from sale, lease, transfer, an assignment, or disposition of assets including those assets acquired by the person paying the capital sum such as: Where any capital sum is derived by way of compensation for loss of office or employment Where a capital sum is received in return for forfeiture or surrender of rights or for refraining from exercising rights. Where a capital sum is received under a policy of insurance and the risk or any kind of damage or injury to or the loss or depreciation of assets Where any capital sum is received as consideration for use or exploitation of any assets and Where any capital sum is received in connection with or arises by virtue of any trade business, profession or vocation. The Capital Gains Act came into effect from the 1st of April 1967 and imposes tax liability on capital gains arising from the disposal of assets accruing to an individual or company. Such assets include shares and stocks, interest, royalties and any such investment incomes. The tax is charged at a flat rate of 20% (10% from 1st January, 1996). The Relevant Tax Authority is the State Board of Internal Revenue in respect of gains accruing to individuals, the Federal Board of Inland Revenue in respect of gains accruing to companies and the residents of the Federal Capital Territory or non-residents of the Federal Republic of Nigeria.

The chargeable gains are gains accruing on the disposal of assets whether situated in Nigeria or outside. Anyaduba (2004) outlines these to include: Options, debts and incorporeal property Any currency other than Nigerian currency and Any form of property created by a person disposing of it, or otherwise coming to be owned without being acquired. Stocks and shares of all descriptions (now exempted with effect from 1998) All qualifying capital expenditure under the Personal Income Tax Decree, the Companies Income Tax Act and the Petroleum Profits Tax Act.

Options in Capital gains tax relates to the right granted by one party to another to acquire specified assets. Where such a right is granted for a valuable consideration, the person to whom it is granted at a profit may dispose it. The gains accruing to the seller is subject to capital gains tax. Similarly, if the buyer of the option finally acquires the property, then on the eventual disposal, the capital gain shall be the difference between the sales proceeds and the total cost incurred in acquiring the property. Where an asset is acquired by a creditor in satisfaction of his debt or part of the debt, the debtor or an acquisition by the creditor for a consideration greater than the market value at the time of the creditors acquisition shall not treat it as a disposal. However, if the chargeable gains accrues to the creditor on a disposal by him of the assets, the amount of the chargeable gain, where applicable, shall be reduced so as not to exceed the chargeable gain which would have accrued if he had acquired the property for a consideration equal to the amount of the debt or part of the debt so satisfied. The Capital Gains Tax Acts (CGTA, 1996 amended) provides that where the shares and stocks of a particular quoted company are held before the Capital Gain Tax Act became operational, the acquisition cost for the purpose of CGT is the market value on the 1st day of April, 1967. This will usually be the price quoted on the Lagos Stock Exchange (LSE). For all quoted shares, the board is empowered to determine the value of the shares as at 1st of April, 1967.for the purpose of CGT. However, where shares of the same class in the same company are purchased in different lots at different prices, an average price is applied on the disposal or part disposal of the shares. Hence, the average price of the shares is computed for the purpose of determining the cost of the disposal. When a bonus issue is made, the existing shareholders acquire additional shares of the company at no extra cost. This is indeed an appropriation of profit. The effect of a bonus issue is to reduce the cost of the existing shares. On disposal of bonus shares, no chargeable gains accrue to the shareholder as he is deemed to be converting his dividend into cash. A right issue has the same effect with a bonus issue in reducing the average cost of the original holding but since the shareholder paid some money for the additional holding, the total cost of the shares is

increased. Where the shareholder makes a disposal of shares acquired on a right issue, a chargeable gain accrues and CGT is therefore payable.

The Banking System and Taxation in Nigeria

The banking system in Nigeria has undergone several reformatory processes. The Central Bank of Nigeria (CBN) regulates all the banks, which is the apex bank in the country. The 1969 banking decree required all banks to be locally incorporated and to publish their balance sheets on their Nigerian banking business only. There are also other regulations, which are aimed at stabilising the indigenous banks for their longterm existence and committing the foreign banks more into the countrys economic improvement. Ghosh (1990) notes that the statutory transfer of 25% of profit after tax and after deducting bad debt provisions to general reserve and the increase in the minimum paid up capital, which presently is two hundred billion naira, are indicators of control of the Central Bank of Nigeria to inject control in the banking arena of the country. The peculiar aspect of the taxation of banks is that in addition to the company tax payable at the normal rate of 30%, banks are required to pay the excess profit levy. This is a levy of 10% on the excess profit of the bank, which was introduced in 1978. However, with effect from 1989-tax year, the excess profit levy tax became 15%. According to Aguolu (2000), the excess profit is the difference between the total actual profit of the bank and the normal profit of the bank computed by applying the following specified percentages. These percentages are usually applied to the capital employed by the bank as at the end of the accounting year: 40% of paid up capital; 20% of capital reserves, 20% of general reserves and 20% of long-term loans. The total is regarded as the normal profit of the bank. The various components of the capital employed may have their divergent meanings. Thus, paid up capital may comprise of the banks paid up ordinary and preference shares where applicable. The capital reserves will also include the share premium account, surpluses on revaluation of fixed assets and amounts set aside out of profit for the issue of share capitals. While the general reserves are undistributed profit apart from the statutory reserves .The long-term loans are those with repayment periods in excess of five years (Banks and Other Financial Institutions Decree, 1991).

Correlation Between Taxes and Dividend

Several researches aimed at correlating tax and dividends have been conducted with intriguing results. Miller and Modiglianis (M&M) theory assumes that taxes do not affect dividend policy of the firm in a perfect market. Thus investors decisions are independent of dividend considerations. However, when the tax on dividend is higher than that of the capital gain of the investor, the issue of dividend policy of the firm becomes a provocative factor for the investors. Chu (1996) also studied the effect of

tax on personal income and posits that dividend tends to expose investors to more tax liability than capital gains. Therefore firms will opt to pay lower dividend to minimize the tax liability on investment. The findings also shows that the after tax income of shareholders is greater when the firm pays no dividends, where the capital gains is lower than the tax consequential on the dividend. Most of the researches were carried out in the US to investigate the effect of the various tax reforms on corporate dividend policy. Bolster & Janjigian (1991) researched on the 1986 tax laws in US but found no concrete evidence of any correlation of any increase in dividend following the reform. However, Papaioannou and Savarese (1994) found an increase in the dividend payout for the average firms after such reforms. This is attributed to the reduction in income tax, which the reform injected in the tax system. This aligns with the preposition that tax has a tremendous effect on the dividend policy of firms. One identifiable problem of inferring the tax effect on dividend is the variation in tax rates faced by different investors. While some investors whose personal tax liabilities are less than their capital gain tax liability, which underscores their interest in dividend rather than capital gains (Hutchinson, 1995). For such, the tax effect does not apply. Different tax situations would ignite diverse effects on the share price of the firm. The rise in dividend can inversely be attributed to the tax provisions on such dividends. One outstanding contribution at correlating dividends and taxes was made by Rozeff (1982) in his dividend payout model. He attempted to correlate the dividends payout with the effect of changes in tax policies using value lines. This model was later replicated in a study by Casey and Dicken (2000) in which banks were used in the research. The result indicated a series of sequence of effect following changes in capital gain tax, which consequently increased the likelihood of no change in dividend payout.

Summary of Literature
The literature review has attempted to identify crucial factors in dividend policy formulation by firms. It has also excavated some significant and recurring effects of tax on the dividend policy of firms as found by vast number of researchers in the subject. Dividend payments have been examined and interpreted by various institutions based on the impact it exudes on the firms. It is a diametric factor as it portends different signals to different interest groups. For managers, it is a way of maintaining or increasing the share price and thus attracting investors. The shareholders read different meanings to it as it signals a lot depending on the clientele. It could show an indication of future strength and builds confidence in investors. It could also be a nightmare because of the attendant tax liability.

Different factors have been examined as warranting various dividend dispositions by firms. It ranges from capital availability, the industry norm, expected earning of the firm, capital project execution, profit and liquidity of the firm and various other factors as amplified in the literature. Managers avoid reduction in dividend because of the sticky signal it sends to the investors and shareholders. It may be a hallmark of incompetent management or a tip of an iceberg of future failure. It is difficult to divorce dividend policy formulation of firms from the tax effect it attracts. Many literatures existing in the sphere of the impact of tax on dividend policy are superfluous as their diverse opinions are. The M&M theory, which posit on the irrelevance in a tax-less society is remarkable fictitious on its assumption of taxes. Tax is a recurrent factor in most economies. Taxes undeniable affect investors and the firm especially in the dividend policies. This research is thus another step in revealing the effect particularly in the banking sector of Nigeria.

RESEARCH METHODOLOGY Research method involves a systematic and orderly approach implored towards the collection and analysis of the data used in the study so that useful information and meaning can be derived from the data (Jankowicz, 2000). The methodology is subjugated into the following sections: a) The Design of the Study b) The Population and Sample of the Study c) Data Analysis Procedure/ Technique

The Research Design

The research is both qualitative and quantitative in its approach, which triangulates in a grounded theory. The qualitative approach adopts descriptive and inductive forms of research. Statistical models are used in analysing data obtained in the study. The methodology is empirical as it implores statistical tools in analysing data obtained for the study. Data obtained from the financial statements of the banks in Nigeria were used in the analyses of the study. The quantitative approach involves an extensive evaluation of the literature and statistical models. In considering a descriptive research, well-detailed information on the nature of the problem was carried out. In this case, dividend and taxes are correlated. This involves analytical interpretation and the application of some economics models. As with descriptive research, it is a fact-finding mission with interpretation and analysis of trends in dividend policy and facts in terms of the commonality of taxes and dividend, share prices and the potential of making predictions and generalisations (Smith, 1997).

The major source of data is through the financial statements of the banks. Though relevant literatures including periodicals and journal articles giving clues on the dividend policy, taxation etc also provide veritable data for the analysis. They also provide some esoteric and quantitative data. Other data sources include official publications such as those of government, Inland Revenue, company report, trade association data which gives some economic indices, private data services, information services, company internet websites and databases, newspapers and bibliographic information are all components of the secondary data.

Population and Sample of the Study

The study population consist of banks in Nigeria that are quoted in the Nigerian Stock Exchange (NSE). With the Federal Government of Nigerians capitalisation structure of the banks to capital requirement of twenty five billion naira (about hundred million pounds), banks are coarsed into merging; resulting in reduction in the number of banks operating in the country. The number of banks is therefore uncertain, as the mergers are still an ongoing process. The study sample was selected using the systematic sampling technique. The choice of the systematic sampling technique is in order to get convenient samples that will be an adequate representation of the study. This has the advantage of eliminating possible bias particularly in a survey (Smith, 1997). A major weakness of this technique according to Jankowicz (2000) is the inherent problems of changes that may occur during the study in which the study population is affected by those changes. The banks selected are based on the number of branches across the thirty-six states in the country. The banks have at least a branch in each state of the country and are grouped in the new generation banks whose financial statements are publicly available. They also have employees numbering over eight hundred and fifty each. These include Guaranteed Trust Bank (GTB), Standard Trust Bank (STB) and Zenith Bank (ZB).

Data Analysis Technique

The data analysis technique implored include the multiple analysis technique and descriptive statistics. Multiple regressions are a statistical tool used in addressing variety of research hypotheses. It has the potential of predicting a particular outcome. It provides information about the set of data and the contribution of each variable in the analysis. It is usually used as a control tool when exploring the prediction of a model (Tabachnick and Fidell, 1996). It is used in correlating the variables used in the study. However, for apparent clarity, ease of analysis and interpretation, both the qualitative and quantitative data were used in the data analysis of the research. The research methods chosen are therefore eclectic, as it combines quantitative and qualitative approaches.

. The choice of the standard multiple regressions for the study is justified on the merit of its strength in determining the variability of the variables in a study. A positive (r) correlation implies that as one variable increases in value, the other variable also increases in value. The reverse is the case in a negative correlation; while one variable increases the other has the tendency to decrease in value. The closer the correlation value is to 1, the stronger the relationship between the variables (Pallant, 2002). In a qualitative research, the issue is usually not the population size that matters or the frequency with which the action occurs but the nature and content of what is researched and the meaning that are most important. The method involves the identification of meaning, categorisation, and the evaluation of their relative importance, which need be listed. Jankowicz (2000) insists that the fact in issue is not the number and meanings, but on the basic assumptions on which the research method and techniques are based. Qualitative research develops a sense of coherence in the study, which is usually dictated by the study rather than any prior suggested structure (Meloy, 2002). Thus, the above principles justified the methodology applied.

PRESENTATION AND ANALYSIS OF DATA The analysis was based on the data obtained from the financial statements of the banks used in the study. They are presented and analysed in line with the research objectives. Factors affecting the dividend policies of banks Most of the factors have been reviewed in the literature, they are further highlighted here. The following factors have been identified and include; pattern of past dividend, stability of earning, level of current earning, concern of the effect on the share price, desire to maintain a given dividend payout ratio, concern about the effect of changes on investors, shareholders needs for income dividend, liquidity factor, investment considerations, availability of alternative source of capital etc. An outstanding factor in the determination of dividend policy is the liquidity factor. Banks that have good liquidity status find it easier to pay dividend. Studies show that a key factor in dividend policy of banks is determined by the liquidity position. Banks that are not highly liquid will apparently find it difficult to pay dividend to its shareholders (Casey and Dicken, 2000). Investment projects of the banks are usually given strong consideration in their dividend payout. Usually, firm that have good investment projects with positive NPV are bind to consider such projects before paying dividends to their shareholders. Such investment opportunities will range from capital investments, expansions of branches, mergers and acquisitions etc.

Stability and level of earnings are identified in the factors. These two factors are very crucial in the dividend policy of the banks. Earnings of the bank are the net income of the banks. When the earning of the banks falls, it is only natural that the dividend will be affected significantly. This is logical as the hub of the payments of dividend is centred on the current or earnings of the bank. Dividend policies are influenced by two outstanding factors. These include the effect on share price and the signal effect of any change on the shareholders. Non-payment of dividend adversely affects the share price of the bank. Shareholders may lose confidence in the stability and growth of the company especially if the dividend is reduced. This will inadvertently affect the investment in the banks share and a fall in the share. The desire to maintain a level of dividend payout is also a determinant factor in the dividend policy formulation of banks. Certain banks may wish to maintain a constant dividend payout ratio to its shareholders. There is also the desire to maintain a level of payout and thus shape the dividend policy formulation of the bank. Where banks have alternative sources of financing their operations and positive NPV projects, the amount left to pay dividend will increase and the bank will apply the resultant surplus earnings into dividend payment. A significant factor in the dividend policy of the banks is the preference of investors for either cash or shares. If the investors are not interested in immediate cash receipt, the companys dividend policy may be aligned to suite such interest. But if the majority interest is basically to have the cash dividend, it is reasonable to expect the banks dividend policy will be designed to reflect that. Shareholders who are interested in making tax savings or avoiding taxes on dividends will apparently seek non cash dividend and to opt for capital gains especially if that is lower than the tax payable on dividends.

The Relationship Between Taxes and Dividend. The section explores the relationship between taxes and dividend of the banks in line with the research question. The Pearson correlation coefficient is used in determining the extent of the relationship of the two variables in the study. However, the financial statements below are those of the banks used for the study. The financial statements, which are three years, comprise of the balance sheets and the profit and loss accounts of the three banks used in the study. The amounts are denominated in the Nigerian currency (naira); but the currency conversion is later made in the analysis to the pound sterling to bring it to terms with the study environment. The exchange rate of naira to pound is calculated at N250 to 1. The table below shows the mean and standard deviation of the tax and dividends the banks.

Table 1 The correlation between Tax and Dividend using Pearson correlation

Tax Tax Pearson Correlation Sig. (2-tailed) N Dividend Pearson Correlation Sig. (2-tailed) N 1 . 9 .942(**) .000 9

Dividend .942(**) .000 9 1 . 9

** Correlation is significant at the 0.01 level (2-tailed).

Descriptive Statistics Table 2 Tax Dividend Mean 439632.33 33 650707.77 78 Std. Deviation 412523.33934 680573.98587 N 9 9

The table above shows the correlation between tax and dividend of the three banks used in the study for a period of three years. The correlation shows a perfect positive correlation between the tax and dividend. This indicates that tax and profit weigh heavily on the determination of dividend policy of the banks. This implies therefore that a change in tax will significantly affect the dividend decision of the banks. As seen in the table, a (r) of 0.942 is significant. .

The Correlation Between Profit and Dividend. This section gives an analysis of the research question and uses the Pearson correlation coefficient and descriptive statistic in the analysis. The analysis is presented in the tables below
Correlations Table 3 Profit Profit Pearson Correlation Sig. (2-tailed) N Dividend Pearson Correlation Sig. (2-tailed) N 1 . 9 .927(**) .000 9 Dividend .927(**) .000 9 1 . 9

** Correlation is significant at the 0.01 level (2-tailed).

Descriptive Statistics Table 4 Profit Dividend Mean 2495203.3 333 650707.77 78 Std. Deviation 2190258.8838 1 680573.98587 N 9 9

The table4.3a above shows that the profit and dividend are positively correlated. A change in profit will significantly affect the dividend policy of the banks. The analysis also shows that the strength of the relationship is positive as the range of the figures is above 0.10 to 0.29. The correlations (r), 0.927 indicates that the value of one variable can be estimated by knowing the value of the other variable. The mean of variables is central with a little deviation. This indicates that the each of the factors is dependent on the others and therefore is related. This is indicated in table 4 above. The raw estimates of the profit, tax and dividend for the three years are shown in the appendix. Test of the Hypothesis Null Hypothesis: There is no significant effect of profit on dividend and tax of banks.

In order to verify the assumptions on the relationship among profit, tax and dividend, the standard multiple regression analysis was used and the results are shown in the tables below.

Standard Multiple Regression

Correlations (a) Table 5 Profit Pearson Correlation Sig. (1-tailed) Profit Tax Dividend Profit Tax Dividend N Profit Tax Dividend 1.000 .942 .965 . .000 .000 9 9 9 Tax .942 1.000 .927 .000 . .000 9 9 9 Dividend .965 .927 1.000 .000 .000 . 9 9 9

ANOVA (b) Table 6. Sum of Squares Df Regressio 128982330 2 n 0739.999 Residual 715807432 6 86.001 Total 136140404 8 4026.000 a Predictors: (Constant), Tax, Dividend b Dependent Variable: Profit Model 1 Mean Square 64491165036 9.999 11930123881. 000 F 54.057 Sig. .000(a)

Coefficients(C) Table 7 Unstandardized Coefficients Model 1 (Constant ) Tax Dividend B -327.565 .201 .124 Std. Error 58030.888 .151 .047 .332 .657 Standardized Coefficients Beta t -.006 1.329 2.631 Sig. .996 .232 .039 .140 .140 7.126 7.126 Collinearity Statistics Tolerance VIF

a Dependent Variable: Profit

In using the Standard multiple regression analysis in testing the hypothesis, a relationship was first established among the variables. Table 4 is the correlation table

and shows that tax and dividend have a positive correlation with profit at 0 .942 and 0.965. The Coefficient table shows the linearity of the variables. However, the t-figure of 1.329 for tax indicates an insignificant correlation of profit and tax. But the t-figure of 2.631 of dividend shows a significant correlation of profit and dividend. This is also shown in the beta values. The tolerance figure of 0.140, which is below zero, indicates that the variables are in linear. (Pallant, 2000). The standard coefficients indicate that the values of the variables have been converted to scale for ease of comparison. The beta value gives the contribution or relevance of each of the independent variables. The highest beta figure is 0 .657, which indicates that dividend has a strong correlation with profit rather than the tax, which has a less beta value of 0.332. However at 0 .05, the result is not statistically significant as shown in the ANOVA table, indicating that profit does not significantly affect tax of the banks. Therefore the null hypothesis is true, as the standard regression analysis shows no significant effect of profit on tax but establishes a significant correlation between profit and dividend of the banks. Discussion of Findings The discussion of the findings of the study is based entirely on the research questions and hypothesis, which are in line with the research objectives. The results of the data analyses obtained in the study show that the various factors affecting the dividend policies are interrelated and are therefore central to the decision of management in the formulation of the companys dividend policy. While each firm may structure its dividend ratio and payout, there is a crossbreeding of the factors as they are interwoven. This finding is in line with the submission of Frankfurter and Wood (1997) that the tradition of dividend policy by firms is influenced by same factors over the three hundred years of dividend evolution! Company managers are usually not in a hurry to reduce dividend payments even in periods of financial distress. This is to maintain the confidence of investors in the going concern of the company. Increase in dividend is only feasible when a company envisages sustainable growth. Managers are also abreast with the significant dividend payment expectation of shareholders. However, shareholders will prefer dividend to capital gain despite the tax implications because of they would opt for immediate payment depending on their level of cash needs. The forgoing findings show that increase and payment in dividend are very important to shareholders and that companies strive to achieve a sustainable level of dividend payout in the long-term. Profits are sinquanon and determinants of the level dividends. Managers are usually not happy with reduction or downward fluctuation in dividend as this reduces confidence in the firms investors.

While the examination of the factors are very revealing and a confirmation of established theories, there remains some intricacies in the evaluation and reconciliation of these factors. As noted by the Shiller (1986), the complexities of the factors affecting dividend policy can only be properly explained when a model is developed which incorporates the financial, behavioural and psychological logic into consideration. Such a model will ameliorate the incongruence of the existing theories that exclude other extraneous variables in considering dividend policy factors. Some of the identified factors dictating the dividend policy of a firm may be influenced by the nature of the firm. Collins, Saxena and Wansley (1986) examined these factors in the context of regulated and unregulated firms. Their findings were contained some potential difference. Thus, the nature of business is a crucial index in dividend formulation of firms. The banking sector, which is the centre of the study, is traditionally considered as an attractive element to investors and may be divergent in its dividend structure. The findings of the study on the impact of taxes on dividend was analysed using the Pearson correlation. The hypothesis formulated in the study and tested using the regression analysis, shows that though taxes correlate with dividends, they do not significantly affect the dividend structures of the banks. This is understandable from the viewpoint of capital gain tax. Ross, Westerfield and Jordan (2001) highlight that effective tax rates on dividend income are higher than the tax rates on capital gains. The dividends are taxed as ordinary income while the capital gains are taxed at a relatively lower rate and are taxed when the shares are sold. Thus, these options of accepting a cash dividend or differing it for a capital gain explains the significant effect of the tax on the dividend as obtained in the study. Thus, the findings support the idea of dividend clientele, as highly taxed individuals are attracted to firms that pay low dividends and would avoid companies that offer high dividend payout policy. The decisions of the banks in adopting dividend policies that are compatible with the tax expectation of their investors grossly underscore the significant result of the study. The findings of the study conform to Talmor and Titman (1990) on the variation of tax rates on dividend and capital gains. Their analyses reveal that if tax rate is expected to increase, the shareholders will prefer dividends to share repurchase. The investors will opt to accept the dividend and pay the tax at a later date. However, if the tax rates are constant, the amount of taxes due on dividend by the investors will be same, irrespective of whether it is a share or cash dividend. Timing of the tax payment is the crucial variable in explaining the difference in the tax impact on dividend. The share repurchase option will be most suitable to investors as the tax liability could be differed. Another veritable approach to ameliorate the effect of tax on dividend is for companies to pay script dividend. Watson and Head (2004) state that an outstanding advantage of this form of dividend policy is in the consideration of tax. Taxes are assumed to have been paid on the income but are not immediately deducted. This makes for tax savings on the dividend. This is similar to the capital gain, which allows the tax deduction on the dividend only when the share has been

sold, and the profit on the sale declared. Thus, to allay the harsh effect of tax on dividend, a paper dividend. The study also revealed the impact of profit on the dividend policy of the banks. The dividend structure of an organisation is a function of the profit available for distribution. Where no profit exists, the dividend of the company will be altered, as the company cannot effectively pay its shareholders.

Conclusions
The following conclusions can be deduced from the findings of the study. Dividend structures of companies are basically influenced by various factors. The factors though may be similar in most organisations or industries, but they are uniquely determined by the nature of the company. Due to the susceptibility of the profit of banks, their dividend policy is often to maintain a low but steady payout. Foremost among the determinants of the dividends structure are the liquidity position of the company. This is a function of the profitability of the company. Organisations that have investment projects will rely on their after tax profit. Thus, the availability of alternate investment opportunity shapes what the dividend will eventually be. Dividend clientele is a very formidable factor in the consideration of a dividend policy. Managers are more inclined to formulate policies that suit the investors interest. The dividend structure of organisations is often flexible, as it is based on the present circumstance of the organisation. However, most managers will shy away from adopting policies that will necessitate reduction in the dividend. Reduction is viewed as a signal of non-performance and eventual failure in the future. A constant policy is often, the most preferred. The profitability of a business is a major variable in the dividend formation of the organisation. Where a business does not have good performance indicators, its dividend policy will be twisted and hardly stable. Profit though, does not always determine the structure of the dividend. Companies may maintain a constant dividend to impress investors. Thus, dividend is considered as a hallmark of good performance. The financial statements of some of banks used in the study indicate that profit is not ultimately the crucial factor in the dividend formulation of the banks. The DPS does not consistently variant with the EPS of the banks. Thus, changes in profit do not necessarily warrant increase in dividend. This averse Taxes have a significant impact on the dividend policy of banks. The implication is that reduction or increase in taxes will make for a different dividend structure. Alternatives to cash dividend are optimally sought for by managers in order to alleviate the impact of the taxes on the dividend. Shareholders who are tax averse will opt for capital gains; scrip dividends and others tax avoidance technicalities in order to reduce the tax effect on their dividends.

Tax would generally favour use of debt in developed countries. Where capital gains are taxed, the investors will opt for a preference of shares. Investors preference will lead to relatively low equity capital costs in countries that do not tax capital gains. The study shows no significant effect of profit on tax but establishes a significant correlation between profit and dividend of the banks.

Recommendations The following recommendations are made based on the results of the study. Adoption of a dividend policy by the banks particularly in Nigeria should be strictly considered based on the unique circumstances of the bank and not necessarily based on age long traditional factors often formulated by academics. This is essential in order to maintain a steady and reasonable policy. Due to the susceptibility of the banks profitability to economic changes in the country, it is unrealistic to seek to formulate a dividend policy that follows a constant payout. However, a low, reasonable and realistic policy should be adopted. Hence differential taxation of dividend might affect the equilibrium assets price, rational investors should be interested in the profit after tax of the firm. Consideration should be given to the needs and expectation of the shareholders in streamlining a dividend policy. When shareholders are akin to tax saving or minimising tax liability, it is reasonable to consider policies that allows for tax savings such as scrip dividends. Banks should encourage capital gains and lay structure that will prompt their shareholders to opt out of cash dividend especially in a developing country like Nigeria where investors are not comfortable with investment income tax.

BIBLIOGRAPHY
Aguolu, O. (2000) Taxation & Tax Management in Nigeria. Enugu; Meridian Associates Akerlof, G. (1970) The Market for Lemon: Quality, Uncertainty and the Market Mechanism. The quarterly journal of Economics, 84, 488-500. Anyaduba, J.O. (2004) Partnership Taxation in Nigeria: Some Basic Issue. ICAN Student Journal 9,2, 15-17. Auerbach, A.J. (1979) Share Valuation and Corporate Equity Policy. Journal of Public Economics, 11, 291-305. Baker, H.K., Gall, E.F.& Edelman, R.B. (1985) A Survey of Management Views on Dividend Policy. Financial Management, 14,181-190. Baker, H.K., Powell, G.E. & Veit, E.T. (2001) Factors Influencing Dividend Decisions of NASDAQ Firms. Financial Review, 36,111-117. Banks and Other Financial Institutions Decree, 1991. The Federal Republic of Nigeria. Bolster, P.J. & Janjigian, V. (1991) Dividend Policy and Valuation Effects of the Tax Reform Act of 1986. National Tax Journal, 44, 511-518. Bradley, M. & Capozza, D.R. (1998) Dividend Policy and Flow Uncertainty. Journal of Finance, 6. 4. 23-43. Bryman, A. & Duncan, C. (1990) Quantitative Data Analysis for Social Scientists. London: Routledge Brealey, R.A., Myers, S.C.& Marcus, A.J. (1991) Principles of Corporate Finance 4th ed. London: McGraw-Hill Brennan, M (1970) Taxes, Market Valuation & Corporate Financial Policy. National Tax Journal, 23,417-427. Capital Gains Tax Act, 1996. The Federal Republic of Nigeria. Casey, M.K. & Dickens, R.N. (2000) The Effects of Tax and Regulatory Changes on Commercial Bank Dividend Policy. The Quarterly Review of Economics and Finance, 40, 279-293. Company Income Tax Act, 1994. The Federal Republic of Nigeria.

Collins, M.C., Saxena, A, K, & Wansley, W. (1986) The Role of Insiders and Dividend Policy: A comparison of Regulated and Unregulated Firms. Journal of Financial and Strategic Decision 9, 2, 1-8. Davies, T. & Pain, B. (2002) Business Accounting & Finance. Berkshire: McGraw-Hill. DeAngelo, H. & Masulis, R.W. (1980) Optimal Capital Structure Under Corporate and Personal Taxation. Journal of Financial Economics, March 3-29. Devore, J. & Peck, R. (1993) Statistics: The Exploration & Analysis of Data, 2nd. Calfornia: Duxury press. Droughty, M. (2000) The Joy of Money. London: Simon & Schuster Ltd. Easterbrook, F.H. (1984) Two Agency-Cost Explanation of Dividends. The American Economic Review 74, 650-659. Farrar, D. & Selwyn, L (1967) Taxes, Corporate Financial Policy & Returns to Investors. National Tax Journals, 17,444-454. Feldstein, M.S. & Green, J. (1983) Why Do Companies Pay Dividends? The American Economic Review 73, 17-30. Frankfurter, G.M.& Lane, W.R. (1992) The Rationality of Dividends. International Review of Financial Analysis. 1, 115-129. Frankfurter, G.M.& Wood, B.G. (1997) The Evolution of Corporate Dividend Policy. Journal of Financial Education, 23,16-32. Jankowicz, A.D. (2000) Business Research Projects 3rd ed. London; Thomson Learning. Jensen, M.C. (1986) Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. The American Economic Review 76, 323-329. Kolb, R.W. & Rodriguez, R.J. (1996) Financial Management 2nd ed. Cambridge: Blackwell publishers. Ghosh, D. (1990) money and Economic Growth in Nigeria. A PhD Thesis submitted to the University of Leicester. Gravetter, F.J. & Wallnau, L.B. (2000) Statistics for the Behavioural Sciences 5th ed. Belmont: Wadsworth. Hutchinson, R. (1995) Corporate Finance. Cheltenham: Stanley Thornes Ltd.

Lintner, J. (1956) Distribution of Income of Corporation. American Economic Review, 46,97-113. Masulis, R.W. & Trueman, B. (1988) Corporate Investment & Dividend Decisions Under Differential Personal Taxation . Journal of Financial & Quantitative Analysis, 23,369-386. Meloy, J.M. (2002) Writing the Qualitative Dissertation: Understanding by Doing 2nd ed. London; Lawrence Associates. McGregor, A. (2004) Retail Sector. Manchester, Graduate prospectus. Miller, M.H.(1986) Behaviour, Rationality in Finance: The Case of Dividends. Journal of Business 59, 451-468. Miller, M.H. & Modigliani, F. (1961) Dividend Policy, Growth, & the Valuation of Shares. Journal of Business, 34,411-433. Miller, M.H. & Scholes, M.S. (1978) Dividend & Taxes Journal of Financial Economics, 6,333-364. Modiagliani, F. (1982) Debts, Dividend Policy, Inflation and Market Valuation. The Journal of Finance 37, 255-273. Pallant, J. (2002) SPSS Survival Manual. Buckingham; OUP. Papaioannou, G.J. & Savarese, C.M. (1994) Corporate Dividend Policy Response to the Tax of 1986. Financial Management 23, 56-63. Petroleum Profit Tax Act, 1993. The Federal of Nigeria. Ramaswamy, T. (1979) Efficient Signalling with Dividends and Investment. The Journal of Finance 42, 321 343. Ross, S.A.; Westerfield, R.W.& Jordan, B.D.(2001) Essentials of Corporate Finance 3rd ed. Singapore: McGraw-Hill. Rozeff, M.S. (1982) Growth, Beta, and Agency Costs as Determinants of Dividend Payout Ratios. The Journal of Financial Research 5, 249-259. Samuels, J.M. & Wilkes, F.M. (1987) Management of Company Finance. 4th ed. Berkshire: Van Nostrand Reinhold. Sharp, J.A., Peters, J. & Howard, K. (2002) The Management of a Student Research Project, 3rd ed. Aldershot: Gowers Pub.Ltd. Shefrin, H.M.& Statman, M. (1984) Explaining Investor Preference for Cash Dividends. The Journal of Financial Economics 13,253-282.

Shiller, A. (1986) The Marsh- Merton Model of Managers Smoothing of Dividend. The American Economic Review 76, 499-503. Silverman, D. (1993) Interpreting Quantitative Data: Methods for Analysing Talk, Text and Interpretation. London: Sage. Smith, C.B. (1991) A Guide to Business Research, 2nd ed. Chicago: Nelson-Hall. Talmor, E, & Titman, S. (1990) Taxes and Dividend Policy. Journal of Financial and Qualitative Analysis 16, 413- 435. Watson, D. & Head, A. (2004) Corporate Finance: Principles & Practice 3rd ed. Essex: Pearson Education Ltd. Wisker, G. (2001) The Postgraduate Research Handbook. Palgrave. Wu, C. (1996) Taxes & Dividend Policy. International Journal of Economics & Finance. 5, 110-122.