The Determinants Of Corporate Tax Avoidance In Tunisian Context Accounting Essay

Published:

Abstract

In this paper we explore how the tax avoidance can be considered as a practice of revenue management and the determinants of tax avoidance.

In this paper, we study the determinants of tax avoidance by taking inspiration mainly from the positive accounting theory. The results of this study, undertaken near 39 Tunisian listed firms from 2001 to 2006, reveal that tax avoidance is tributary mainly of the size, the age and the profitability of the company.

Keywords 

Revenue management, corporate tax avoidance, book-tax difference, tunisian listed firms

JEL Classification: H 25; H 26; H 32

Biographical notes: Mohamed Ali Omri is a Professor of Accounting at the Faculty of Economics and Management of Tunis. He is responsible for the research team tax accounting controls and auditing. Also, he is the author of two books on taxation of companies and several articles on taxation, auditing and corporate governance.

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

Ines El Aissi Ben Ameur lectures in Accounting at High Institute of Finance and Taxation of Sousse, Tunisia. She has recently completed her PhD Thesis on the Tax management of Tunisian firms. Her main research interests and publications are in accounting and taxation.

1. Introduction

The amplified use of revenue management (RM) has generated the need for researchs to identify the significant activities to successful RM performance. We can identify tow types of activities wich can implies the sucessful RM performance: increase the revenue and decrease the cost. Beck and al. (2010), identify and validate the underlying dimensions of activities important to successful RM performance, through increasing the revenue. Other researchers focused on the activities that can improve performance through deacreasing costs. In fact, Eldomiaty (2010) shows that high growth is associated with lower costs. To increase revenue each firm try to reduce its tax cost throw legal method, that's why we can consider tax avoidance as a pratice of revenue management. 

The tax avoidance is defined as legal activity that reduces the tax liability by lawful methods. The tax avoidance is legal and its aftermaths are certain, contrarily to the fiscal fraud which is illegal and its aftermaths are uncertain, dependent on the probability of the fraud detection.

The techniques of tax avoidance are examined in detail by Stiglitz (1985) and include two main modes: the taxpayer can use the shelters or loopholes which correspond essentially to the involvements in some societies whose profit is exonerated of tax, i.e.: he chooses from the beginning to invest in some firms of fiscal alleviated regime. Besides, he can invest abstraction made of fiscal regime and looks for reducing the taxable plate by benefiting from some fiscal advantages. It is this last mode that we will try to explain in seeking to know what the determinants of the corporate tax avoidance are.

The determinants will be dealt with only from an economical point of view disregarding the moral and political motives, of which the importance is acknowledged, but which are too difficult to measure (Gauthier, 2000).

The survey of the determinants of the corporate tax avoidance will permit, through its results and its findings, to illuminate the authorities on the development of the fiscal strategies aiming to reduce the corporate tax avoidance.

Besides, it permits to conceptualise a model of the shelter behaviour of some firms, a topic which, in our opinion, has not been until now dealt with.

To this end, we are going to identify the determinants of the corporate tax avoidance through a review of literature and then we will try to validate them empirically next to some Tunisian listed firms.

2. The determinants of the corporate tax avoidance

The literature on the corporate tax avoidance attempted to describe and to formalize the behaviour of the individual income tax (rather than corporate income tax) in using mainly the theory of the expected utility. It consists in determining the part of income to declare which maximizes the expected utility. Recently Slemrod (2004) emphasised the need of differentiating between the individuals and the corporations so as to explain the behaviour of fiscal regularity, and suggested that this behaviour should be analysed in the context of the agency theory.

Ahmed (2004) proposed to model the corporate tax liabilities based on the agency theory frame where the State is" the principal" which seeks to collect the owing tax entirely and the firm is" the agent" who seeks to maximize his wealth.

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

The economical theory of regulation can be considered as a reference frame since it considers the firm like an entity in relation with the whole society. The latter imposes, through his government institutions, some expenses which burden the benefits of the shareholders (Raffournier, 1990).

Our work is based on the positive accounting theory which is derived from the agency theory and the economical theory of regulation. The positive accounting theory studies the contractual, economical and political motivations and the impact on the result, of some accounting choices made by the firms. This theory doesn't distinguish between the book income and tax income, but in our work we focus on this distinction, since we consider the corporate tax avoidance as the action of minimizing the tax income (with regard to the book income), benefiting from some fiscal advantages. So, the determinants, which, according to this theory, are susceptible of pushing the firms to minimize the result, can be used also to explain the minimisation of tax income.

The positive accounting theory is limited to mention three determinants to decrease the result: the size, the sector of activity and the structure of the ownership. Some other empirical studies, mainly those of Tedds (2006), Giles (2000) and Ahmed (2004) cited some other determinants as the organizational structure, the age of the firm, the profile of the audit firm and the profitability of the firm.

2.1. The size of the firm

The hypothesis of the political costs driven from the positive accounting theory is based on the act that the big firms are more affected by the political pressures than the small ones because they are subject to one of the biggest transfer of wealth. Indeed, the bigger the firm is, the more it contributes to the state financing. The government, therefore, is interested in the performances of the big firms. According to this theory, the political costs influence the behaviour of the firms towards the result by minimizing it through the accounting choices (Deakin, 1989; Malmquist, 1990) or following a change of the methods (Labbell, 1990), or also through the earning management (Cormier and Magnan, 1995; Cormier and al. 1998). The recourse to one of these three methods allows the minimization of the result, but temporarily, since they consist in repatriating the expenses in the progressing year and deferring the profits in the following year (Stolowy and Breton, 2000).

Lisowsky (2010) show that tax avoidance measured by the book tax difference is positively related to size.

If it was demonstrated that the size of the firm is a determinant which pushes the firms to minimize their income temporarily through some accounting choices, we would, therefore, expect the big sized firms to have a tendency definitively to minimize the tax income through the corporate tax avoidance.

2.2. The sector of activity

Following the positive accounting theory, some sectors may represent a risk of political costs more than some others, such as, the energizing sector, which generally constitutes the target of the political pressures for the transfer of the wealth (Cormier and al., 1998).

Literature seems to show that the corporate tax avoidance varies from one sector to another. The empiric survey led by Giles (2000) reveals that the manufacturing; construction and wholesale trade, retail trade accommodation, cafes and restaurants; and communication services, exhibit "excessive" avoidance.

Ahmed (2004) aims to provide an insight into the tax behaviour of firms and examine factors which have an impact on corporate tax liabilities. The results reveal that the econometric Model has an explanatory power very significant for the firms of computer service and those belonging to the hotel sector. They show, besides, that the impact of some explanatory variable varies with the sectors of activity.

Probably, the sector of activity influences, according to his specificities and his opportunities, the propensity of the firm to the tax avoidance. We can deduce this when the government incites to the investment in some sectors of activity more than in some other. Indeed, to the reading of code of incitements to the investments, we can clearly note that the granted advantages varying with the sector of activity subject of investment. So, the article 7 of this code stipulates that" the subscription to the capital or to the increase of capital of the firms operating in the activities aimed by the first article of present code benefits of the deduction of the profits reinvested in the limit of 35% of net profits submitted to the profit tax." We appraise whereas the firm who operate in the activities represented on the list have a propensity to exercise the corporate tax avoidance more than the others.

2.3. The ownership structure

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

The positive accounting theory applies to the entrepreneurial firms (in which the manager possesses an important part of capital) preferring the methods which decrease the result. This is explained by the fact that these managers want to maximize the value of their action in minimizing the tax. However, the managerial firm prefer the methods witch increase the result.

Some empirical studies focused the impact of the ownership structure on the minimisation of tax through the choice of the accountant's methods and the earnings management (Dhaliwal and al., 1982; Cormier and al., 1998; Dyreng and al., 2010; Martin and al., 2010). Accordance to the political cost hypothesis Martin and al., (2010) confirm that Managers may under-state reserves to justify lower rates to regulators. In the same way, Dyreng and al (2010), show that individual top executives have incremental effects on the level of corporate tax avoidance that cannot be explained by the other characteristics of the firm.

In tunisian firm the manager is used to be a founding family members, we can then expect that firm managed by founding family members will be more tax planner than their non-family counterparts. However chen and al. (2010) found that firms owned and run by founding family members are less tax aggressive than their non-family counterparts. They explained this results by the fact that firms owned and run by founding family members are characterized by a unique agency conflict between dominant and small shareholders. That'why family owners are willing to forgo tax benefits to avoid the non-tax cost which can arise from minority shareholders.

This conflictual results leads us to test the relation between corporate tax avoidance qnd the ownership structure. We appraise that the ampleness of the corporate tax avoidance is more important to the entrepreneurial firms than to those managerial.

2. 4. The organizational structure

Ahmed (2004) focused on the impact of the organizational structure in minimising the tax liabilities. He demonstrated that the firms with complex structure have a tendency to exercise the corporate tax avoidance more than the others, through transfer pricing and tunnelling. He showed that this is of so much truer for the firms of service and the industrial ones.

According to Sikka and Willmott (2010), 'transfer pricing' is represented as a technique for optimal allocation of costs and revenues among divisions, subsidiaries and joint ventures within a group of related entities. They show how this practice is deeply implicated in processes of enhancing private gains by avoiding the payment of public taxes. They prove that transfer prices practices are used by corporations to avoid taxes in developing and developed economies.

We consider that a complex structure could encourage the behaviour of firm's tax avoidance. Firms with a large number of subsidiaries manage to reduce their future tax liabilities as, for example, reinvest in a subsidiary which allow the exoneration of profit.

2.5. The age of the firm

Ahmed (2004) put in evidence the positive tie between the age of the firm and its capacity to minimize the tax liabilities, in pulling profit of the fiscal advantages as foreseen by the legislation. Indeed, the older the firm is the more it is capable of seizing the opportunities of tax avoidance. Tedds (2006) argued that the effect of a firm's age is ambiguous: younger firm may be more tax avoider because they may have more competitors, may be struggling to turn a profit, and may view tax avoidance as a way to cut costs. On the other hand, young firm may be more reliant on external financing, which could cause them to be more tax compliant.

2.6. The profile of the external auditor

Although the audit mission is of legal order it is exercised in the setting of a contractual relation comparable to an agency relation. Some authors appraise (Watts and Zimmerman, 1986) that the auditor could develop opportunist behaviour in order to enter in a relation of cooperation with the manager.

In literature, the firms, which are attended by some auditors belonging to some BIG 5, have more tendencies to manage their result (Becker and al., 1998). However, some other authors showed that the adherence to some BIG 5 could constitute a brake to the management of result (Mohd Nor and al., 2010; Cormier and al., 1998). Mohd Nor and al. (2010), examine the relationship between fraudulent financial reporting and firms' characteristics audit quality, they find that audit quality have significant negative relationships with fraudulent financial reporting.

These contradictory results could be explained by the fact that the auditor could play two roles. He could provide some conservative counsels fearing that this avoidance is assimilated to a fraud and that, therefore, his reputation is blemished. He can, on the other hand, provide an aggressive counsel since he is supposed to have the expertise required in order to minimize the tax respecting, at the same time, the law so as to satisfy his proxies. The survey led by Carcello and al. (1992) reveals that the auditors in the Anglo-Saxon context offer more than a simple certification of the accounts and behave like a real counsellor for the service of the firm.

We estimate that the firms audited by an auditor belonging to an international cabinet has a tendency to exercise the corporate tax avoidance more than the others, since he is the best placed to seize the opportunities of avoidance.

2.7. The profitability of the firm

Rare are the researches which studied the impact of the profitability on the level of corporate tax avoidance. Giles (1998) showed that the profitable firms have a tendency to be more regular than the others. On the other hand, Manzon and Plesko (2002) appraised that the profitable and auspicious firms have a tendency to exercise the corporate tax avoidance more than the others. Lisowsky (2010) show that tax avoidance is positively related to profitability. In the same way, the study of Cai and Qiao (2009) shows that Chinese industrial firms in more competitive environments engage in more tax avoidance activities.

Following the same reasoning adopted by the positive accounting theory concerning the impact of the size on the result minimisation, we consider that the profitable firms are more appreciable to the political pressures than the others, and submitted to a biggest tax burden. So, in order to minimize their tax liabilities, they have more tendencies to exercise the corporate tax avoidance.

3. Explanatory factors of the corporate tax avoidance: Empirical Validation

The objective of this survey is to test empirically the hypothesis according to which the corporate tax avoidance depends on: the size, the sector of activity, the ownership structure, the organizational structure, the age of the firm, the profile of the auditor, and the profitability of the firm.

We present, in the following section, the firms' sample, the variables' measures, and the model and finally the results achieved.

3.1. The sample of the survey

Our initial sample is constituted of 41 listed Tunisian firms, for one active period from 2001 to 2006. The data is collected from the financial states and some information annexes, published by the stock market. We excluded, first of all the firms who benefit from a fiscal particular regime, and the observations which show that the fiscal income is superior to the book income, since, in this case, it can't be considered as tax avoidance (Frank and al., 2009). The total number of observations rises to 94observations.

3.2. Measure of variables

In this section, we are going to present at first the variables to be explained and second the research explanatory variables.

3.2.1 The variable to explain: The corporate tax avoidance.

AVOI: The corporate tax avoidance is measured by the permanent gap between the tax income and the book income. We adopted the same measurement that was conceived by Frank and al., (2009).

We are going to determine for every firm, and for every exercise:

* The gap between the book income and the tax income commonly called: [book tax gap]

BTGit= BIit- TIit,

Where

BIit = The book income of the firm i to the instant t = the net resultit+ the tax owing it TIit= The tax income of the firm i to the t instant= (tax owing it/ tax rate)- ΔNOLit

NOLjt = Change in net operating loss carryforward from year t-1 to year t for firm i.

* The permanent gap which measures the ampleness of the corporate tax avoidance

AVOIit= BTGit - BTGTit Where BTGTit, represents the temporary book tax gap = Deferred tax of the firm i to the instant t/ tax rate

AVOIit= BTGit - Deferred tax it/ tax rate

In order to measure the corporate tax avoidance, several researches are founded on the book tax gap (Hanlon, 2003; Desai and Dharmapala, 2006; Frank and al., 2009). This is justified by the fact that during the 90s, the growth of the practices of corporate tax avoidance was accompanied by the growth of this gap (McKinnon and Harwood, 2003; McKinnon, 2004).

The positive gap between the book income and tax income is assigned partly to the corporate tax avoidance. This difference could be divided in two compounds: the temporary gap and the permanent gap (Plesko 2003; Desai and Dharmapala, 2006).

The temporary gap emerged when the same expense or the same income is acknowledged similarly in the accountant system and in the fiscal system but at different periods. Some researchers showed that the temporary difference reflects the earning management activity (Phillips and al. 2003; Hanlon, 2003). However, the corporate tax avoidance is a permanent and definitive reduction of the tax (Desai and Dharmapala, 2006; Frank and al., 2009).

The permanent gap exists when an expense or an income is acknowledged in a system but never in another. It is this gap which could be assigned to the corporate tax avoidance (Desai and Dharmapala, 2006; Frank and al., 2009).

The amount of the corporate tax avoidance will be normalized by the balance of the capital of every firm and in every exercise. At this level we have to note that we will take the case only where the book income is superior to the tax income. In the opposite case, it can't be considered as corporate tax avoidance (Frank and al., 2009).

3.2.2 The explanatory variable

SIZE. The size of the firm is measured by the natural logarithm of the asset of the firm to the t instant. It is generally the adopted measurement by the authors studying the impact of the size on the earning management (Cormier and al., 1998). We appraise that the ampleness of the corporate tax avoidance increases with the size of the firm.

SECP. The adherence to a sector of privileged activity will be measured with a mute variable, which takes 1 if the firm operates in the activities aimed by the first article of code of incitements to the investments and otherwise 0. We appraise that the ampleness of the corporate tax avoidance increases for the firms which operate in the privileged sectors.

OWNE. The ownership structure is a variable which takes 1 if the leader of the firm possesses an important part of capital, and otherwise 0. Determining if the involvement is considered as important is not an easy task. Some researchers consider that the leader could have considerable influence if he possesses more than 20% of capital (Cormier and al., 1998). By referring to the article 283 of the commercial firms' code , which stipulates that" a or several shareholders representing, at least 5% of social capital could ask for the enrolment of supplementary projects of resolutions in the agenda of the general assembly." So, possessing at least 5% is considered important. We estimate that the level of corporate tax avoidance is much higher for the firms managed by a leader who possesses more than 5%.

SORG. The organizational structure is measured by a mute variable which takes 1 if the firm is a part of a group and 0 elsewhere (Ahmed, 2004). The organizational complex structure could influence the behaviour of tax avoidance since it increases the opportunities of minimisation of the tax burden. So, the level of corporate tax avoidance is more important for the firms belonging to a group of firms than that for others.

AGE. The age of the firm is measured by the number of years since its creation. We appraise that the ampleness of the corporate tax avoidance increases with the age of the firm.

PAUD. The profile of the auditor is measured by a binary variable, which takes 1 if the firm is audited by an auditor representing an international cabinet in Tunisia and 0 elsewhere. We expect that the ampleness of the corporate tax avoidance in the firms audited by an international cabinet is weaker than in others.

PROF. This variable is measured by the ratio of the return on assets. We foresee that in order to maximize their net result of tax, the profitable firms tend to demonstrate more tax avoidance.

The table 1 sums up the whole survey variables and their measurement.

Table 1: Presentation and measures of the variables

Variables

Symbols

measurement

Corporate tax avoidance

AVOI

Pretax book income less taxable income for firm j in year t less deferred tax expense / tax rate,

Size of the firm

SIZE

Natural Log of the total asset

Sector of activity

SECP

Adherence to a privileged sector : 1, otherwise 0

Ownership structure

OWNE

Entrepreneurial: 1, otherwise 0

Organizational

structure

SORG

Adherence to a group of the societies: 1, otherwise 0

Age of the firm

AGE

The number of years since the constitution.

Profile of the auditor

PAUD

Adherence to an international cabinet: 1, otherwise 0

Profitability of the firm

PROF

Return on assets

3.3. Presentation of explanatory model of the corporate tax avoidance

Inspired mainly by the positive accounting theory and some empirical studies led by Giles (1998), Ahme­d (2004), Tedds (2006), Frank and al. (2009), Mohd Nor and al. (2010), Lisowsky (2010), Dyreng and al. (2010), Martin and al. (2010) and Sikka and Willmott (2010), we present the explanatory model of corporate tax avoidance behaviour:

AVOIit= a0+ a1 SIZEit+ a2 SECPit+ a3 OWNEit+ a4 SORGit+ a5 AGEit+ a6 PAUDit+ a7 PROFit+ εit

The econometric model aims to identify the micro factors that potentially affect the corporate tax avoidance

3.4. Results of the research

3.4.1. Univariate analyse

The results of Student test represented in the table 2, shows that the average of the corporate tax avoidance is slightly more elevated for the privileged sectors by the code of incitements to the investments compared to others. However this difference is not statistically significant.

The results reveal, besides, that the average of the tax avoidance defers significantly at 10%, with the ownership structure (OWNE). This seems to show that the manager who possesses an important part of capital has recourse more to the exercises of corporate tax avoidance, in order to maximize the value of his actions after tax. On the other hand, the average of the corporate tax avoidance, though it is slightly more elevated for the firm belonging to a group, it's not statistically significant. As for the variable profile of the auditor (PAUD), the results attest that the average of the corporate tax avoidance at the firms audited by a government auditor representing an international cabinet is statistically more elevated than others.

Table 2: The means of the corporate tax avoidance of different binary variable

Characteristics

Average of tax avoidance of the group of which the characteristic is present

Means of tax avoidance of the group of which the characteristic is absent

T de Student

p-value

SECP

0,147

(n = 37)

0,126

(n = 57)

0,875

0,384

OWNE

0,148

(n = 63)

0,110

(n = 31)

1,681

*0,096

SORG

0,143

(n = 70)

0,111

(n = 24)

1,189

0,237

PCAC

0,165

(n = 47)

0,105

(n = 47)

2,680

***0,009

*** Statistically significant at 1%

* Statistically significant at 10%

The results of the correlation test represented in the table 3 reveal that the variable (AVOI) is significantly and positively correlated to the variables (SIZE, AGE and PROF).

Table 3: Evaluation of the correlation coefficients

TAIL

AGE

PROF

AVOI correlation coefficient

Signification

0,189*

0,073

0,376***

0,000

0,504***

0,04

*** Statistically significant Coefficient at 1%

* Statistically significant at 10%

3.4.2. Multivariate analyse:

The results of the multiple regressions, in the table 4 below reveal that the behaviour of corporate tax avoidance is explained at 49.4% by the model, which could be considered as relatively important.

Table 4: Evaluation of the relative parameters of the model:

The variable to explain: corporate tax avoidance (AVOI)

Explanatory Variable

Coefficients

t

signification

SIZE

0,389

3,769

***0,000

SECP

- 0,017

-0,175

0,861

OWNE

0,039

0,425

0,672

SORG

0,046

0,502

0,617

AGE

0,172

1,856

*0,067

PAUD

0,123

1,345

0,182

PROF

0,678

6,751

***0,000

*** Statistically significant at 1%.

* Statistically significant at 10%.

R²= 0.494 (R² adjusted= 0,451).

Three variables are significant: the size, the age of the firm and the profitability of the firm.

The positive sign of the variable SIZE allows us to validate the positive link between the behaviour of corporate tax avoidance and the size of the firm. This constitutes, besides, a validation of the hypothesis of the political costs of the positive theory, which stipulates that the firms of big size have a tendency to minimize their result.

As for the effect of the adherence to a privileged activity sector, the results show that the coefficient of regression of the variable SECP is negative and statistically not significant. This could be due to the used measurement: it is necessary to distinguish between the sector of adherence and the sector object of investment. For example the adherence to the financial sector doesn't permit to the firm to benefit from a fiscal particular advantage, but the investment done by this same firm in an activity aimed by the code of incitement opens right to deduction of reinvested profit.

The mute variable, (OWNE), is positive but not significant. This could be justified by the fact that managerial firms can, as well as the entrepreneurial firm, exercise the corporate tax avoidance if the remuneration for their leader is tributary of net result. Erickson and al. (2004) studied the features of some firms accused of fraud accountant, and demonstrated that the system of remuneration of leader has a significant impact on the fraudulent behaviour.

Elsewhere, the regression coefficient of the organizational structure (SORG) is positive but not significant. This can be justified by the fact that the mother firm of a group could decide upon the regime of result integration, which consists in making the algebraic sum of the tax income of different firms, members of the group. It is on this embedded income that the tax will be acquitted. So, the firms, members of this group, could not resort to minimize their tax income by benefiting from some fiscal advantages, since it is sufficient that one firm, member of a group, showing a deficit that the global result is minimized. Actually, most of the quoted firms are not mother firms but members of a group, and the tax avoidance won't be exercised at the level of the firm, member of a group, but rather at the level of the mother firm. The latter can seek to oppress the result of a firm, by different techniques such as the transfer pricing until the result shows a deficit. Ahmed (2004) showed that a group could minimize its tax liabilities through the transfer pricing and tunnelling.

The explanatory power of the firm age (AGE) is significant, at 10%. This result confirms the idea according to which the ampleness of the corporate tax avoidance increases with the number of activity years. Indeed, from one year to another, the firm acquires the latitude of seizing the opportunities of corporate tax avoidance. This result rejoins the one of Ahmed (2004) who showed that the positive tie between the age and the minimisation of the tax liabilities seems to be valid mainly at the first years of the firm's lifespan.

The results of regression model reveal, besides, that the profile regression coefficient of the auditor (PAUD) is not statistically different from zero. The positive sign of this coefficient shows, however, that the firms audited by an auditor belonging to an international cabinet have more tendencies to exercise the corporate tax avoidance, since they have the expertise required in order to seize the loopholes and the opportunities. The non significance of this tendency can also be explained by the distrust of the auditors to the avoidance, notably after the enactment of the code of right and fiscal procedures which holds the auditor for responsible in case of fiscal fraud. So, for fear that it is assimilated to an illegal avoidance and therefore to a fiscal fraud, the auditors, and mainly those representatives of an international cabinet, have a tendency to be more rigorous in the respect of the fiscal rules especially after the scandal of Enron who blemished the reputation of some international cabinets.

The results show that the profitability of the firm (PROF) and its prosperity play a major role in the demonstration of fiscal behaviour. Indeed, the positive sign of the coefficient, significant at 1%, show that the ampleness of the corporate tax avoidance increases with the profitability of the firm. Indeed in order to be profitable, it is necessary to be organized efficiently in all the fields including the fiscal one.

4. Conclusion

Following the adherence of Tunisia to the Worldwide Trade Organization and its engagement to enter in a zone of free exchange, it became imperative to count on the other feeding sources of the state's budget.

So, this survey tends to guide the fiscal authorities so as to optimise the returns got through the direct tax and mainly through the profit tax.

We don't pretend, to the term of this work, the conceptualisation of an explanatory model of corporate tax avoidance behaviour, but, we attempted merely to focus on the firm features which are susceptible of encouraging such behaviour.

To the light of the resulted achieved from 39 Tunisian listed firm samples, we could conclude that the size, the age and the profitability of the firm are the main features of corporate tax avoidance. This shows implicitly that the firms whose contribution to the budget of the state should have been the most elevated, is actually the least elevated, since these firms have the human and financial means to exercise the corporate tax avoidance. This raises the problem of the firms' contribution to the state budget.

One should be careful when generalizing these observations because of the research limits. One of them which can be evoked refers to the measurement of the chosen variables.

In order to measure the corporate tax avoidance, we constructed a proxy in adopting the methodology conceived by Frank and al., (2009). This variable should be appreciated by the permanent gap between the book income and the tax income. So, we need to calculate the temporary gap which is equal to the tax deferred on the tax rate. However, when we analysed data, we noted that no firm showed, in its financial states, the amount of deferred tax. This can be explained by the fact that the standard on the deferred tax is not again published and that the firms, though they can take advantage, don't take their count in consideration. This constitutes a limit to the measurement of the corporate tax avoidance, since it could be majored by the corresponding amount to the temporary gap.

Besides, the use of the binary variable is not advisable. Indeed, it doesn't permit really a seizing of the sense of the relation. It is, on the other hand, desirable to determine a quantitative measurement of the studied phenomenon.

This survey constitutes a little contribution to the literature on the behaviour of corporate tax avoidance and can serve as a starting point for some other later studies in the Tunisian context. Showing the significant impact of some characteristics of the firm on the corporate tax avoidance, we think it would be interesting to enlarge the survey by shedding light on the cultural, environmental, and sociological factors which can also influence the avoidance behaviour.