Discussion Board & Replies/Forensic Accounting

(Original Content Only) (400 words) (APA citations)

(Discussion Board Post)

You are the new controller of a major US manufacturing firm. In your previous employment, you deteced multiple varieties of fraud. The CFO of you new company informs you that top management is concerned about possible fraud in the organization and is interested in taking a proactive approach both to detecting and deterring fraud. After noting that he has recommended you for the fraud detection assignment, the CFO tells you that he is a little nervous about hom much this investigation approach will cost and asks you to keep your choices simple and inexpensive. You and th CFO are good friends, an dyou’ve never had a problem suggesting ideas about upcoming projects in the past. You know that to be most effective in completing your new assignment, you should do some extensive data analysis because of the large size of the company and its databases.

Maria and Angela take the position of the CFO wo wants to keep analyss simle and Andrew, Afayi, Rebecca take the position of the controller who believes more expensive, data-driven approaches are necessary. Debate the appropriates of the various detection approaches, including the deduction approach, for your company. Explain why traditional approaches may not be sufficient.

(Discussion Board Replies) (3 Replies) (150 words each) (2 APA citations) (in-text apa citations are a must)

1. The first thing I would discuss with the CFO is the concerns about cost and how expensive and costly fraud can be: The company would have to sell $10 of goods to make up for every $1 in fraud if the profit margin is 10% (Albrecht et al., 2019). So the choice is really between unknown lost revenue, reputation, poor publicity, and blatant dishonesty (if fraud is found) or trying to be proactive in seeking out fraud or areas of where fraud might occur, even if that proactive behavior has upfront costs the company might not want to spend.

Maria and Angela and CFO wo wants to keep analysis simple

Since the company is a public, large manufacturer, there should be ready access to the financial statements to pursue a vertical and horizontal analysis. Many such companies have to provide public quarterly filings, and while the financial statements might not be instructive, comparisons of the different ratios could reveal changes in the ratios across different periods. The vertical and horizontal analysis would also reveal when two financial statement items (i.e., increased revenue and increased accounts receivable) do not behavior normally (Albrecht et al). If more money is coming into the company, but the accounts receivables are declining , it could indicate a symptom of fraud–or that more customers are paying cash up front and not by invoicing. The disadvantage of this method is that the investigation can miss the periods in which fraud is being perpetrated.

Andrew, Afayi, Rebecca take the position of the controller who believes more expensive, data-driven approaches are necessary.

It is likely the company already has access to computer programs/databases/software that can perform data-driven searches. The challenge is the integration of the different departments, internal employees, and outside investigators who need to collaborate together to use the data-driven approach. There is a likelihood that, while the initial set-up for the D-D approach is more time consuming, it can look at multiple factors the financial statement analysis cannot, such as addresses/telephone number crossover with employees and vendors (Albrecth et al.). There are also multiple methods for searching with the data: for outliers, using Benford’s Law, summarization or stratification, or fuzzy matching (Albrecht et al.). These methods use the probability of a particular event from occurring that highlights a particular troublespot. Unlike the financial statements method, the D-D approach is not limited to identifying the correct period in which the fraud occurred, and can look at the data with almost a multifaceted approach.

Explain why traditional approaches may not be sufficient.

The traditional method is kind of like calling the fire department after a house is engulfed in flames: The damage is already done, and it is hard to come up with the right location, item, and accelerant to determine the cause of the fire. A fire investigator may know the blaze was likely arson, but the investigator may not be able to prove who did it. The traditional approach needs “smoke” to sound the alarm: tips or symptoms to cause rise to an investigation. The D-D approach is like if the fire inspector was doing an inspection on a home, found a badly wired outlet and called in an electrician to prevent the damage done by the wire sparking: Already there, looking in all directions for everyone’s safety, and alerting to danger in a timely fashion.

2. In this debate about the appropriate fraud detection approach for the major US manufacturing firm, there are two distinct viewpoints: one favoring a simple and inexpensive approach, and the other advocating for more expensive data-driven methods. Let’s explore both perspectives.

Maria and Angela (CFO’s viewpoint – Simple and Inexpensive Approach):

The CFO’s primary concern is cost, and in favoring a simple approach, they are focused on minimizing expenses. They argue that traditional fraud detection methods are generally more cost-effective and manageable. Traditional approaches like internal audits and manual checks are less disruptive to the daily operations of the company. Extensive data analysis may disrupt ongoing processes. Traditional approaches have been used for years and are well-established. They can effectively identify common types of fraud without requiring significant investments in technology or specialized personnel.

Andrew, Afayi, Rebecca (Controller’s viewpoint – Data-Driven Approach):

The Controller team argues that a data-driven approach is necessary to be proactive in fraud detection. Fraudsters continually evolve their techniques, making it crucial to stay ahead of them. Data analysis can help identify patterns and anomalies that may go unnoticed with traditional methods. Given the large size of the company and its extensive databases, traditional approaches may not be sufficient. Massive amounts of data require advanced tools and algorithms to detect sophisticated fraud schemes. Data-driven approaches can help reduce false positives. Traditional methods often rely on simple rule-based alerts, which can generate numerous false alarms. This can be a waste of resources and distract from legitimate issues. Data analysis allows for deeper forensic investigations into fraud, enabling the identification of root causes and potential vulnerabilities that can be addressed to prevent future fraud.

Explanation of Why Traditional Approaches May Not Be Sufficient:

Traditional approaches like manual audits and rule-based checks are generally narrow in scope. They are designed to catch known fraud patterns but may miss new or sophisticated techniques that are constantly emerging. The company’s size and data velocity may overwhelm traditional methods. Identifying fraud in large datasets in a timely manner can be challenging without automated data analysis. Traditional methods often generate a high number of false positives, which can be costly to investigate and divert resources away from genuine fraud cases. Traditional approaches may not adapt quickly to changes in fraud tactics. Criminals are constantly finding new ways to defraud organizations, and data-driven approaches are more flexible in this regard.

In conclusion, the choice of fraud detection approach depends on the specific needs and circumstances of the company. A balance might be struck by implementing a hybrid approach that combines cost-effective traditional methods with data-driven solutions to cover a broad spectrum of fraud detection requirements while optimizing cost-effectiveness. The key is to tailor the strategy to the company’s risk profile and the sophistication of potential fraudsters, ensuring that resources are allocated effectively to mitigate risks.

3. As the new controller of a large, major US manufacturing firm, I would discuss the fraud detection methods and provide the company with the best option. I would inform the CFO that “full-population analysis is often the preferred method in fraud investigation” (Albrecht et al., 2019, p. 167). The data-driven approach is often the most expensive and time-consuming. However, it can be one of the main approaches that will identify fraud while it is still small and allows the investigation to target fraud before it becomes large and causes the most damage. Taking into consideration that the CEO is concerned with expense, the use of computer technology can aid in the analysis without “significant increases in cost or time” (Albrecht et al., 2019, p. 167). Other simpler methods, such as statistical sampling, may be less expensive but will not have the same capabilities to find fraud when looking at such a large amount of data. The traditional approach is reactive and should be implemented when an anomaly is found, or a tip is received. If this had been the case, it would be easier to pinpoint where the fraud was without extensive data analysis.

The deductive approach follows five steps. For this approach, a team would need to be created to examine and gather information about the company. After better understanding the business as a whole, the team would start to identify areas where fraud could be present. Next, the team would identify “red flags or fraud symptoms” (Albrecht et al., 2019, p. 169). We would then use technology to find data in the company’s databases that can be analyzed for errors that need further analysis. The final step in the data-driven approach is the investigation of fraud indicators.

An inductive approach starts with observations, looking for patterns in the observations, and then creating theories about the observed patterns. Even though the observations are valid, the theories or conclusions can still be wrong. A more simplified approach, such as sampling, can be used but is less effective than data-driven processes when the investigation deals with large amounts of data from a large firm.