Regnskabsmanipulation: Kendetegn, forebyggelse og opdagelse

Claus Nielsen

Student thesis: Master thesis


This thesis focuses on the problem of financial statement fraud. The theoretical part of the thesis can be split into these sub questions: Who commits this type of fraud and what is their incentive? How is the fraud typically conducted, prevented and detected? The empirical part of the thesis concerns HP’s acquisition in 2011 of the British software company Autonomy. Shortly after the purchase HP accused the management of Autonomy of large scale fraud, and the case now has several pending investigations and lawsuits. The theoretical part begins with a discussion of the incentive to commit financial statement fraud. The main theory concerning this type of fraud is the traditional fraud triangle, which states that there must be three elements present simultaneously before this type of crime is committed. These elements are constituted by a pressure, an opportunity and a rationalization. Other models include the fraud diamond and predator behavior. The most likely person or group of persons to commit financial fraud is the management of the company. There are two main reasons for this. Firstly, the management is evaluated on a day-to-day basis by the stakeholders of the company according to the financial performance of the particular company. This puts a pressure on the management and gives it a motive to manipulate the accounting of the company so it seems more attractive. Secondly, the management is in a position to override all the company controls that has been implemented to prevent manipulation of the accounting. As a result of this, the management could have both a motive and an opportunity to carry out manipulations. What are the normal signs of financial statement fraud? A company must choose the relevant accounting framework, for example IFRS. An obvious deviation from the framework is either an honest mistake or a manipulation of the correct recognition. The thesis discusses several types of manipulation of the framework. Revenue-based schemes normally involves timing schemes, misclassification schemes and fictitious or inflated income. As an example revenue might be shifted from the next period to this period. Asset-based schemes could be carried out through improper capitalization of costs, inflation of assets and manipulation of fair value accounting, for example through understated impairment losses. Expense and liability schemes often involves shifting expenses to future periods or simply omitting or underreporting liabilities. Another type of fraud is manipulation of the disclosures, which usually means that some disclosures represents the company in more positive light than the framework permits. How is financial statement fraud prevented and detected? There are several steps in this process, and every company must find their own path. Strong governance provides the foundation for an 2 effective fraud risk management programme. This governance starts at the board level and permeates the entire company structure through a zero tolerance towards financial statement fraud. The next step is to identify and asses the fraud risk. Impact and likelihood of the risk is calculated, and a proper response according to the risk appetite is chosen. The most effective way to manage the fraud risk seems to be through preventive controls. This is however not always the case, since these controls can be overrided by the management, which also has the main motive for this type of fraud. An alternative strategy is to change the structure of salary and bonus systems and thereby reducing the fraud incentive. Detective controls might be very effective in the fraud search process and inconsistencies are often found via a tip. Another important investigation technique concerning financial statement fraud is analysis of the company accounting data. The thesis discusses three basic types of analysis, which is horizontal, vertical and ratio analysis. There are other ways to analyze this type of data often involving the use of computer software. A relevant test could be to check the data according to Benford’s Law. Some of the mentioned tools are used in the case described below.In 2011 HP acquired the fast growing British software company Autonomy. It was a friendly takeover and the big shareholder and founder of Autonomy, Michael Lynch, was to remain CEO. Half a year later, in the spring of 2012, the situation of the company was quite chaotic. The original management was fired and HP accused Michael Lynch and the CFO of manipulating the financial statement in the period leading up to the purchase of Autonomy. As a result of this HP wrote of the main value of Autonomy of which around 5 billion USD could be contributed to manipulations. In 2015 HP sued the old management and several investigations and lawsuits are now pending. The main fraud accusations concerned undisclosed sale of hardware, manipulated recognition of revenue from resellers as well as inflated revenue from hosting deals and Autonomy’s main product IDOL. Some of these elements cannot be investigated further in this thesis. However, emphasis is put on Autonomy’s high pressure bonus system and a management, that doesn’t seem to tolerate criticism from market analysts. Most likely HP and Autonomy was on a corporate cultural collision course from the beginning. Different analysis have been conducted in this thesis, but they don’t seem to confirm any of the many red flags that were raised in connection with the acquisition. The most relevant of these warning signal are probably Autonomy’s very high growth rates and a low receivable turnover. The conclusion of the still ongoing investigations and lawsuits will most likely reveal some more interesting details in this complicated case about financial statement fraud.

EducationsMSc in Auditing, (Graduate Programme) Final Thesis
Publication date2016
Number of pages87