Regulatory Compliance in Finance and BI (part 2)
The previous blog entry on this topic provided an introduction to the realm of compliance and some of the key metrics with respect to BI. Click here to access the previous blog entry.
In this part, let us dwell on the importance of the data to be measured, extraction of such data and the standard KPIs for compliance.
Measurement of Data:
In view of the need to be vigilant of dubious transactions, any brokerage firm attaches immense importance to the analysis of data by its compliance officers. Without the aid of BI, the analysis of data is an onerous exercise for these officers. There could be millions of transactions daily, out of which a few transactions do not comply with the standards. In such a scenario, the firm should tap the BI infrastructure to analyze data precisely and make informed decisions.
Let us quickly understand the characteristics of a questionable transaction, through a few illustrative examples in the equity (stock) market. Very frequently, we encounter the term 'cross' trading. Before we delve into an example, let me also introduce the basic concepts of a trade.
A client, who wants to trade in the market, operates an account with an intermediary. The intermediary typically has a portfolio manager who manages the accounts of clients and places their orders to obtain the best prices for their transactions.
Now, in the 'cross trade' example, suppose there are two clients C1 and C2 who want to trade in the same security X. The client C1 wants to buy the security at a price P1 and the client C2 wants to sell the same security at price P1. The broker B1 manages accounts for both clients C1 and C2. The broker can execute these opposite orders on the same security between these client accounts without recording the trade on the exchange. This type of a transaction is illegal in major stock exchanges and is considered 'cross trading'. In such a transaction there is a high likelihood that the clients did not get the best price.
Another example pertains to wash trading in which the purchase and sale has occurred without a change in ownership. Such a transaction can be executed by an investor simultaneously across two different brokerage firms where he/she has an account to create an appearance of substantial trading in the security to attract market interest. This in the parlance of the stock market is called a bubble.
ETL Logic and Key Metrics:
The overriding purpose of the ETL Jobs is to identify dubious transactions for further analysis of the compliance team. We will not explore the algorithms to detect such records in detail here, since this write-up is more related to the function than the BI technology used commonly. Using such an algorithm which scans millions of records and isolates these suspect transactions using a variety of conditions, we load the reporting layer of the warehouse. The dimensions in this warehouse can be trader, broker, customer, time and security or stocks.
A standard key metric is the number of such trades (cross, wash etc) done on any exchange for a trader/broker on a day or week. Another metric is the absolute value of such transactions which is symptomatic of the health of the equities market. The number of such dubious trades negatively correlates with poor compliance.
The function of reporting in this exercise entails allowing the compliance officers to access the data, besides providing different perspectives to the data based on different dimensions (attributes). Compliance officers can analyze data for an individual trader or broker and chart the trends for that trader. Using a comprehensive ad-hoc reporting platform like Oracle BI, the compliance team would be mainly involved with reporting on important metrics and using the metrics for capturing trends and analyzing them. Conditional formatting, trend analysis, forecasting and improving the existing data entry checks are the attractive features provided by the BI infrastructure. One more interesting and effective method is the creation of scheduled alerts. This feature can be configured to detect dubious transactions with high dollar amount and immediately send notifications to senior compliance officers on daily basis. This gives an extra edge to the compliance officers in taking effective and timely action.
Thus far we have only scratched the surface on possibilities with BI. The pervasive usage of BI helps strengthen the effectiveness of compliance and enforcement in line with regulations such as SOX, GLBA, and HIPPA. In light of the growing demands on organizations for statutory compliance, a BI environment can significantly reduce the burden on staff concerned with this function and fortify their defense against costly repercussions on account of fraudulent or inappropriate behavior. It is the author's earnest belief that this discussion promotes the utility of BI in the area of compliance.
This article has been written by N P N Anand. Anand is a consultant with the Oracle Business Intelligence Practice at Infosys. His areas of interest include Packaged BI Applications.
