The High Cost of Fraud Operations

November 11, 2009 by Paul McCormack

11 comment(s)

How much money do banks spend to detect and investigate fraud? Based on my experience and discussions with various banks – a lot! Most banks have at least one department that is dedicated to fraud. The bigger the bank, the higher the probability that they have departments focused exclusively on certain types of fraud (mortgage, bank card, check etc). The fully loaded cost of each employee can easily reach $60,000 for junior staff and well over $100,000 for senior management. It is not unusual for a large bank to have anywhere from 100 to 500+ staff dedicated to fraud detection and investigation.

Which begs the question – given the significant costs generated by a fraud department how many staff members should a bank employ in order stop fraud? Regardless of the industry and size of fraud losses, I have found that senior management will tolerate only so much investment in fraud detection and investigation. In all honesty, I have yet to meet a senior executive that takes pride in the size of their fraud department budget. It is a cost to be managed, not celebrated.

When fraud spikes, instead of adding staff to a fraud department, most senior executives will look for every excuse not to do so. After all, money invested in fraud staff means that less money is available for sales and marketing etc. It is not unusual to be asked "Can't we do something else instead of adding fraud staff?"

Whenever possible, fraud should be eradicated at the source. Whether a loan product is redesigned, or a checking account discontinued, removing the source of the fraud makes perfect sense. However, banks are in the business to make money. In order to remain competitive, banks must offer competitive products. Sometimes, that means offering a product that is susceptible to fraud.

If additional fraud staff is hard to come by, or you are being pressured to reduce your headcount, how can a fraud department stop more fraud? Consider improving the department's process as well as investing in technology. Process improvements can yield moderate improvements, and be relatively inexpensive. However, they take time to realize and need constant reinforcement. Technology can be expensive and take time to implement, but if used appropriately can dramatically improve the bank's ability to detect fraud.

So, when requests for additional staff fell on deaf ears, how did you respond? What did you do to improve your department's processes? Did you invest in new technology?

Make a Comment

* = Required

* Name:

* Email:

Website:

* Comment:

Fill in the code shown on the left

Recent Comments:

Hammclov
November 16, 2009 - 10:45 AM

"I think there are two issues here, a physical cost and a credibility cost First of all, as someone with a great deal of experience in securities litigation, I can guarantee that fraud absolutely has a cost to upper management (a physical cost, as well as a credibility cost) and an extensive cost to its shareholders. Most securities cases settle, usually on the order of millions of dollars. 2008 saw a huge uptick in the number of mortgage fraud cases, as did the number of subprime fraud cases. Litigation is extremely expensive, and even if the bank does win the suit, the costs associated with defending (or settling these claims) is enormous, and is often listed in the footnotes of a public filing. The credibility cost is even more important. The success of a product line isn't really material in this discussion. The real success of a large company, is it's ability to generate financing for it's activities. If a potential investor or loan officer sees little or no control at a product--they won't invest. Exhibit A is Bear Stearns. When their hedge funds collapsed over the summer of 2007, it called into question their entire business. That confidence is crucial to business. Confidence can only be built through public and private safeguards. Therefore, fraud controls at banks and other companies, are absolutely essential to the health and continued growth of businesses. Executives who bemoan the cost and cancel fraud detection programs, will get fired by the board of directors when their company is investigated or sued. "

Greetings
November 17, 2009 - 8:30 AM

"Paul, l am currently looking for a trainer to conduct a three day training for my organization on Fraud Prevention and Detection. I have searched for your contacts in all the possible social networks but l cant find your information. Kindly let us know how we can reach you. "

Becky
November 17, 2009 - 9:39 AM

"Good question. When analyzing losses, I look first to determine if their was any failed operational procedure. These could occur in either the origination OR servicing area. If none of the processes in these areas failed, then I determine it to be consumer abuse. If I find that any of the processes failed, I investigate to determine why the failure occured. If it occurred because of a people, process or technology issue within the organization it is a process failure. If the processes performed as they were supposed to, I look for fraud and 99.9% find some type of fraud involved. This evaluation is not always easy to determine since so many companies have bifurcated the origination and sertvicing processes to the extent that it is almost impossible to identify where a process failure occurs. The value with analyzing the processes is that if it is true fraud then it is easy to isolate. However, if it is a process failure, then typically you will find patterns of an issue or issues that are the drivers of the problem. My concern is that by labeling everything fraud a company is losing the chance to find a serious flaw in their process that could be corrected and prevent further problems. As for classification,in my opinion fraud is an operational loss. I know of no company whose operational handbook says it is OK to use fraudulent information. Let's look at an example. I have a borrower who qualifies for a loan based on income that is 10% greater than the actual dollar amount. However, he has only 1 month payment reserve and the guideline required 3. The loan goes into default. Was it because he falsified his income or was approved without the proper reserves. If it was because of the income, then it should be classified as fraud. If it was due to a lack of reserves, it is a process fail. There are no yes/no answers here. I think layered risk in process failures is as significant or more significant that a borrower's layered risk. As for classification of these losses, I believe it is a game we are playing until we begin using risk models that can tell us which of the layers actually drove the loss. That is what I am trying to determine through my risk model. Hope this answers your question. If not, let me know. Becky "

Becky
November 17, 2009 - 9:39 AM

"It is understandable that management is unwilling to pay fraud prevention costs that are in excess of fraud losses. Unfortunately rather than try to determine if "fraud" actually costs money, the industry has assumed that all fraud results in losses without even validating the concept. Through the use of my financial services process analysis risk model it is apparent that many issues currently labeled "fraud" have no impact on product performance. Wouldn't it be more cost efficient to revise processes and focus only on those issues that truely drive losses? "

Becky
November 17, 2009 - 9:59 AM

"Paul, My organization provides in-depth training on mortgage fraud. If we can help you support clients by conducting this training, I would be glad to help out. "

Jaspal singh
November 18, 2009 - 4:10 PM

"Its true any management will always try to minimize the cost on fraud prevention as it is direct hit to the profit of the organisation, but its the on Management how they take it, as in today's world fraud crime are increasing, to prevent fraud loss , fraud control cell in the organisation is must, but the working style should be toward fraud saving by busting fraud rackets and recovery of money and assets from fraudsters. Most important is that sending deterrent in market, the departmental cost should not be more than 35% considering fraud savings by department, and the fraud loss should not more than the targets fraud loss of the organisation. Employees working in fraud control department should be more action oriented along with suggesting policy changes for lending, which should be postively accepted. Regular changes in the apporach for fraud prevention is needed with help in reducing fraud loss "

Paul McCormack
November 18, 2009 - 7:18 PM

""Greetings" aka KAVAQ, Thanks for reaching out to me. I have responded to the email that you sent to the firm. If you have not received, please feel free to email me at pmccormack@innovarpartners.com Thanks! Paul "

Paul McCormack
November 18, 2009 - 7:18 PM

"Hammclov, You added two very important elements to the discussion - credibility and confidence. Capital markets are in large part based on credibility and confidence. Leaving fraud unchecked, or being less than focused on the prevention and detection can directly impact credibility and confidence. I have limited experience with Securities litigation but agree 100% that the costs can be astronomical (millions in professional fees a month!). The time and effort needed to manage Securities litigation at the C-level is somewhat hidden, but a cost nonetheless. Personally, as an investor, I would prefer that the CEO, CFO etc are looking forward in order to grow the company rather looking back to defend its actions (or inaction). Again, great comments. I look forward to reading more! Paul "

Paul McCormack
November 18, 2009 - 7:18 PM

"Becky, Great answer. The process is very well thought out. I completely understand your approach. The example you provide does a good job explaining why there are "yes/no" answers. In fact, I believe that you touched on one of the more frustrating elements of fraud detection, how much does a bank's (or company's) policies signal to the fraudster what they need to do during the approval process for example to commit fraud? Where does the bank place emphasis, and what does the fraudster need to do to pass that test, yet still commit fraud? As for training, I do not have a client currently that would benefit from your training. However, I will definitely remember your post and approach should you need arise in the future. I look forward to seeing more of your posts! Paul "

Paul McCormack
November 18, 2009 - 7:18 PM

"Hi Becky, Thanks for taking the time to post! Can you expand on your comments regarding "fraud" that does not result in loss may be with some examples? By way of clarification, when discussing fraud, I am focused on an event that DOES results in a loss. That said, I am aware that some experts anaylze losses and potential losses by classifying them as either misuse, abuse or fraud. To what extent do you conside intent when analyzing, and potentially reclassing a loss? Once you determine that fraud has NOT taken place, how do you reclass the event? I know that some banks default to "operational loss". I look forward to reading your comments! Paul McCormack "

Paul McCormack
November 19, 2009 - 9:12 AM

"Jaspal, Thanks for taking the time to share your thoughts. I am very curious. You note the following: "Most important is that sending deterrent in market, the departmental cost should not be more than 35%" How did you determine that cost should not be more than 35%? Is that a rule of thumb that you use in your bank? Look forward to reading your comments. Paul "

Search:

RECENT NEWS

MOST POPULAR

Blog Authors

The High Cost of Fraud Operations

Make a Comment

Recent Comments: