Abstract
Tax Fraud is a criminal activity done by a manager of a firm or at least one tax payer who intentionally manipulates tax data to deprive the tax authorities or the government of money for his own benefit. Tax fraud is a kind of data fraud, and happens every time and every where in daily life of households, business, economics, politics, science, health care or even in religious communities etc. Data fraud is extensionally characterized by the four fields: Spy-out, data plagiarism, manipulation and fabrication. Data manipulation takes existing data and manipulates the content encapsulated in tables, diagrams, documents or (historical) pictures.
Tax fraud manipulates book keeping figures and tax declarations either by increasing expenditures or decreasing income. There is no clear boundary to accounting and balance sheet policy of firms, especially if accounting and valuation latitude is utilized.
The tax fraud investigation by the tax fraud authority can be embedded into the Bayesian Learning Theory based on investigation and integration of partial information. The kick-off is an initial suspicion issued by a stage holder or insider like a fired employee, disappointed companion or wife, envious neighbor or inquisitive custom collector. This first step can be conceived as the fixing of the prior distribution p(θ) on the (complete but still in detail unknown) tax liability θ of the tax betrayer. The next step at the authority’s site is concerned with opening a new case, and getting access to the tax file of the suspect. Formally, the likelihood of the tax fraud, l(x|θ), is established. This allows updating of the initial suspicion for gaining the posterior distribution p(θ|x) ∝ l (x|θ) p(θ).
This cycle may be performed again if further step by step investigations deliver more information on the non-conforming suspect‘s life style related to the series of his annual taxable income. The necessary investigations are tricky for getting insight into the betrayer’s life style, and make use of criminal investigator’s good practice like, for instance, “Simple issues first!”.
The main step, however, of the tax fraud investigation is getting a search warrant from the court, and, consequently, starting inspection of business premises and home with “full power”. More formally, we take the former posterior p(θ|x) as a new prior p*(θ) and combine it with the new facts about the tax crime, y, using the likelihood l*(y|θ) getting the new suspicion facts p*(y|θ) as the updated posterior. The investigation stops when a general definition of the tax crime is formulated using p*. Then the charge is left to the judicial system to prosecute, judge and eventually arrest the accused people.
There is and will be no omnibus test available to detect manipulations of (even double-entry) book keeping data with high precision. However, a bundle of techniques like probability distribution analysis methods, Benford’s Law application, inliers and outlier as well as tests of conformity between data and BKI-indicator systems exist to give hints for tax data fraud.
Finally, investigators may be hopeful in the long run because betrayers never will be able to construct a perfect manipulated world of figures, cf. F. Wehrheim (2011).