Project topics and materials
Get Complete Project Material File(s) Now! »
REPORT 2: DETERMINANTS OF INTERNAL TAX COMPLIANCE COSTS: EVIDENCE FROM SOUTH AFRICA
INTRODUCTION
The South African small business sector faces various challenges, one of which is the regulatory and legislative burden imposed on small businesses in the form of tax legislation (SBP, 2005:44, SBP 2011:28). One of the elements of the tax burden is the tax compliance burden – the amount of time and money (compliance costs) spent in order to comply with tax laws (Charron, Chow & Halbesma, 2008:iv; Coolidge, Ilic & Kisunko, 2009:4; Guyton, O’Hare, Stavrianos, & Toder, 2003:676; OECD, 2010:5). Three broad components have emerged as the indisputable core elements of tax compliance costs (Evans, 2006:3; Turner, Smith, & Gurd, 1998:96; Tran-Nam, Evans, Walpole, & Ritchie, 2000:229; OECD, 2009:16), namely taxpayers’ and unpaid helpers’ time (internal tax compliance costs); tax practitioners’ fees (external tax compliance costs); and incidental expenses.
It is the first component (internal tax compliance costs – taxpayers’ and unpaid helpers’ time) that will be considered in this article – the second component is considered in a separate article (the third component, as its name implies, is not considered material and is thus not considered further). Internal tax compliance costs include the cost of labour or time devoted to tax activities, for example, the time taken by a business person (an owner), the person’s employee (a manager/internal bookkeeper/accountant/other employee handling taxes) or an unpaid friend or relative to learn and understand the tax law and the obligations the law imposes, or the time taken to obtain documents and data to complete a tax return (Evans, 2008:451; Klun & Blazic, 2005:418; Turner et al., 1998:96).
Smulders, Stiglingh, Franzsen & Fletcher (2012) found that internal tax compliance costs increase as the size of the business – based on turnover – increases. Validation of these findings was considered necessary because a failure to address the validity of the responses obtained from instruments (such as online questionnaires as used in that study) raises issues of trust in research findings (Murphy, Hashim & O’Connor, 2007:1). Furthermore, it was considered appropriate to identifying any other possible determinants that could influence internal tax compliance costs and if so, which of these factors have the largest influence on the levels of these costs (Eichfelder & Schorn, 2008:2).
This study aims to provide insight into the key drivers of internal tax compliance costs (per tax) for small businesses by using regression analyses. A regression analysis was performed on the results obtained from the Smulders et al. (2012) empirical study conducted in 2011 on the tax compliance costs incurred by small businesses in South Africa. This survey was conducted by means of an electronic questionnaire distributed by the SARS to 88 057 small businesses (turnover of R14 million or less) registered with SARS and for which SARS had an e-mail address at the time the questionnaire was distributed (Murugan, 2011a).
As the whole target population (as described above) was selected, no statistical sampling techniques were used. The number of usable questionnaires received amounted to 5 865, representing a response rate of 6.7%. Saunders, Lewis & Thornhill (2007:358) indicates that internet based surveys are likely to have a response rate of 11% or lower, however, it must be mentioned that the electronic survey platform used to distribute the questionnaire could unfortunately not determine how many of the e-mails that were sent out were undeliverable (Murugan, 2011b:2). This could have had a major effect on the response rate and consideration should be given to this fact before concluding on the response rate. Although one can therefore not come to any definite conclusions about how representative and statistically reliable the sample was, 5 865 responses should nevertheless provide invaluable information and insight into an area where there is currently no reliable and up to date statistical information available.
The remainder of the article will first describe the methodology used for the regression analysis. The results of the regression analysis will be presented next. Thereafter the conclusions will be documented and the need for future research highlighted.
METHODOLOGY USED FOR THE ANALYSIS
Regression analysis
Field (2009:198) describes a regression analysis as a statistical tool used to examine the relationship between variables (anything that can be measured) by ascertaining the casual effect of one variable upon another single variable (simple regression) or upon more than one variable (multiple regression). A multiple regression analysis was considered an appropriate tool for the analysis of external tax compliance costs because analyses of this nature are typically used to show the applied value of research findings (Murphy et al., 2007:3).
In order to determine the variables (called the independent variables or explanatory variables – predictors) that have an effect on the internal tax compliance costs (called the dependent variables), the results of the small business tax compliance cost study were considered and a review of the literature was performed in order to develop hypotheses about the independent variables that could possibly have an effect on internal tax compliance costs, as suggested by Eichfelder and Schorn (2008:5). The hypotheses and choice of independent variables were selected based on the findings of the small business tax compliance cost study, as well as on past research (Field, 2009:212).
However, before considering these independent variables and hypotheses, it must be mentioned that the time (hours) was used as the dependent variable for the internal tax compliance costs (activities), rather than the costs (Rand values), because no singular Rand value amount was provided by the respondents in the survey. Instead, a series of calculations had to be performed by the researcher using various answers provided by the respondents to various questions in the questionnaire (such as who performs the task, how this task is divided between these persons and what the value of these persons’ time is) to obtain the overall internal tax compliance costs. It is argued that the use of hours is more suitable than the Rand values as the valuation of the hourly rate used to calculate the overall costs of internal tax compliance is regarded as a somewhat contentious issue (Evans, Ritchie, Tran-Nam & Walpole, 1997:11).
Having established hypotheses about the determinants (the independent variables) of internal tax compliance costs from the current study’s results and the literature, the assumed influence of these variables on the tax compliance costs were investigated using the General Linear Model (GLM) procedure in SPSS. Although other international tax compliance costs studies have used other forms of multiple regression – such as the Ordinary Least Squares method (Blaufus, Eichfelder, & Hundsdoerfer, 2011:8; Hasseldine & Hansford, 2002:380) and the logarithmic GLS model (Eichfelder & Schorn, 2009:11) – to estimate the effect of various independent variables on the tax compliance costs, it was decided to use the GLM in the current study because it readily accommodates both categorical and continuous predictors.
The internal hours provided by the respondents were provided per tax, so each tax was separately analysed by means of its own regression model. Following the approach used by Hasseldine and Hansford (2002:381), once the initial regression analysis had been performed and the results analysed, the analyses were re-run but this time including only those independent variables that were found to be significant in the first regression analysis. This was done to determine whether the model specifications were robust or not. The results from all these analyses are discussed after the hypotheses and regression models used have been explained.
Hypotheses
A hypothesis is a prediction about the state of the world (Field, 2009:787). A regression analysis tests hypotheses – by determining which predictor variables (independent variables) contribute substantially to the regression model’s ability to predict the outcome (dependent variable), or explain the variability in the dependent variable. Predictor variables should only be included in a regression analysis if there are sound theoretical reasons for expecting them to influence the dependent variable (Field, 2009:225). In order to determine the variables that could influence internal tax compliance costs, a review of the literature was performed.
The literature revealed that most of the studies that used a regression analysis to determine the influence of hypothesised variables on tax compliance costs did so in general and not per tax (DeLuca, Greenland, Guyton, Hennessey, & Kindlon, 2005; Eichfelder & Schorn, 2008; Reekmans Simoens, 2010). However, the Hasseldine & Hansford (2002) study only dealt with variables that influence VAT tax compliance costs and the Blaufus et al. (2011) study only dealt with variables that influence income tax compliance costs. As no detail is available on the influence of these variables on other taxes (such as PAYE, CGT, turnover tax, customs duties and excise levies considered in the current study) their application to the other taxes was considered appropriate. The same rationale applied to the other studies that used regression analyses for total tax compliance costs rather than for each individual tax, and hence their hypotheses were regarded as appropriate for each separate tax.
Using these studies it is argued that the following variables have an influence on internal tax compliance costs:
business size (Coolidge et al., 2009);
sector (Eichfelder & Schorn, 2008; Hasseldine & Hansford, 2002; Reekmans & Simoens, 2010); legal form (Blaufus et al., 2011; Coolidge et al., 2009; DeLuca et al., 2005);
business age (Eichfelder & Schorn, 2008);
use of small business tax concessions (Freedman, 2006, 2009; Pope, 2008); level of education of business owners/employees (Blaufus et al., 2011); accounting knowledge of business owners/employees (Blaufus et al., 2011); use of an external service provider (Blaufus et al., 2011);
type of accounting system used (Coolidge et al., 2009; Hasseldine & Hansford, 2002); province (Eichfelder & Schorn, 2008);
gender and marital status (Blaufus et al., 2011);
psychological factors (Eichfelder & Schorn, 2008; Hasseldine & Hansford, 2002);
administrative strategy (capital-intensive or personnel-intensive) (Eichfelder & Schorn, 2008, 2009);
use of electronic data interchange (Eichfelder & Schorn, 2009).
The information in relation to the last five abovementioned points was not considered in the current study. The primary reason for this being that the questionnaire was already considerably long without their inclusion. Research in these areas can therefore be considered in the future.
Using the information in the remaining bullet points and the results obtained from the small business tax compliance cost study (Smulders et al., 2012) – that the time (hours) and costs involved in complying with tax legislation increase as business size increases – the following hypotheses were derived:
Business size – As the size of the business increases, the absolute internal tax compliance costs also increase (Coolidge et al., 2009:3). Three measures of business size can be used – turnover, number of employees and gross asset value (South Africa, 1996). As the information relating to only two of these measures was available from the data of the current study, only turnover and number of employees could be considered. To determine which one of these variables has a stronger impact on internal tax compliance hours/costs, both variables were included in the analysis.
Sector – The sector in which a small business operates is not a significant determinant of internal tax compliance costs (Reekmans & Simoens, 2010:36). However, Hasseldine and Hansford (2002:382) found that some sectors (such as manufacturing, services and dealing in goods sectors) incurred lower VAT compliance costs than other sectors; Eichfelder and Schorn (2008:14) also found that the services sector had higher tax compliance costs (including the time burden) than the building sector. The results of the current study found that the transport, postal and warehousing sector followed by the public administration and safety sector spend the most time on internal tax compliance activities, but reasons for this were not immediately evident. There was no consistency in the local and international literature regarding the significance of sector on internal tax compliance costs. In addition, no other research in South Africa has indicated that the sector has a significant effect on tax compliance costs. Hence, the hypothesis that the sector in which a small business operates is not a significant determinant of internal tax compliance costs was retained for the purposes of this study’s regression analysis.
Legal form – Sole proprietors spend more time on internal tax activities than CCs and companies that tend to outsource their tax compliance activities (Blaufus et al., 2011:10).
Business age – Younger businesses incur lower internal tax compliance costs than more established businesses. This is due to a lower degree of tax complexity compared to more established businesses (Eichfelder & Schorn, 2008:11).
REPORT 1: TAX COMPLIANCE COSTS FOR THE SMALL BUSINESS SECTOR IN SOUTH AFRICA — ESTABLISHING A BASELINE
REPORT 2: DETERMINANTS OF INTERNAL TAX COMPLIANCE COSTS: EVIDENCE FROM SOUTH AFRICA
REPORT 3: DETERMINANTS OF EXTERNAL TAX COMPLIANCE COSTS: EVIDENCE FROM SOUTH AFRICA
