Wray Rives
SEC using technology to identify fraud earlier

New API software is designed to mine big data to help detect big fraud.

March 18, 2013
By Wray Rives, CPA, CGMA

In 2010, after a very public failure to detect Bernard Madoff’s $17 billion Ponzi scheme, the SEC was mandated by Congress under the Dodd-Frank Wall Street Reform and Consumer Protection Act, P.L. 111-203, to develop methods to prevent future large-scale fraud.

A great deal of publicity was given to the SEC’s new whistleblower program, which included increased awards for individuals who report fraud to the agency and the establishment of a Whistleblower Office within the agency. Some saw this as an admission by the SEC that it did not have the knowledge or technology to identify fraud independent of outside assistance. Furthermore, despite these stepped-up efforts, the SEC inspector general issued a report in 2010 criticizing the agency for making too few awards to whistleblowers.

The SEC traditionally has had a reputation of being behind on technology. But for the past few years the regulator has been developing a computerized system to analyze performance data of the entities it monitors in order to find early warning signs that might indicate possible fraud.

The effort has been aided by the almost universal implementation of extensible business reporting language (XBRL), a format that makes financial reports readable by computer. In 2009 the commission began phasing in XBRL filing requirements over a three-year period. By 2012, it had accumulated a huge warehouse of computerized data.

A data analysis program known as the aberrational performance inquiry, or API, also will be fully released by the end of 2013. The API targets investment firms and hedge fund managers, though some of the same philosophies are being applied to the broader stock market through real-time tracking of transactions on the public exchanges.

The SEC has not indicated if the results of its new testing will be publicly available, but CPAs who offer personal financial planning services will want to watch this program develop. The API also will be of particular interest to CPAs who audit investment firms and hedge funds, as API will analyze the data in their reports.

Using big data to identify big fraud

Some investment managers have expressed concern that they may be targeted simply because they do their job well and provide above-average returns for clients. Yet the software doesn’t try to detect suspicious activities merely by picking out top performers.

The API is designed to identify potentially suspicious investments by searching for returns that seem to be too consistent or ones that are outliers compared with the rest of the returns being analyzed. The basic concept of big data analytics such as this is to take huge amounts of data and uncover hidden patterns and unknown correlations through various data-mining techniques. As such, the API data analysis also will take into account risk factors such as frequent changes in auditors, performance that is inconsistent relative to the investment strategy of a fund, or frequent delays in the release of earnings. The goal is to identify characteristics that are common to fraudulent investment vehicles.

After the software identifies funds that should be reviewed, an agency team combs through the information to see if the numbers can be explained. As part of the enforcement division of the SEC, the review team can recommend that an investigation be initiated. Such an investigation may ultimately result in administrative procedures or even federal court action against a firm or fund.

SEC tracking real-time trading

By extending the technology initiative and using some of the same evaluation models from the API, the SEC is also implementing the ability to track real-time trading. Rather than start from scratch and hire developers to build its own application, the SEC purchased software known as Market Information Data Analytics System (Midas) from Tradeworx, a high-frequency trading company in New Jersey.

High-frequency traders already use data analytics and sophisticated computer applications to trade securities, often trading in and out of securities positions thousands of times a day. Midas will combine the strategies from the aberrational performance inquiry and algorithms developed by high-speed traders to monitor real-time activity in the stock market.

CPAs who provide audit services for investment firms, hedge funds, and publicly traded companies need to prepare their clients for these SEC initiatives because the agency will eventually be analyzing performance results from every company, firm, and fund traded in the United States.

Rate this article 5 (excellent) to 1 (poor). Send your responses here.

Wray Rives, CPA, CGMA, helps startup and small business owners compete and succeed in their local or global market as founder and manager of NeedaCFO.com and through his public accounting firm Rives CPA.