For the first time since the 2008 financial crisis, financial institutions have begun to reduce headcount in compliance. And while it is impossible to determine any one specific driver of this rollback, it is likely that as the volumes of fines levied in the sector abates—after reaching a staggering cumulative total of $321 billion—senior management may see an opportunity to reallocate resources elsewhere.
Growing automation in financial services firms is also obscuring this dynamic. However, while some jobs that feed into the compliance function—such as the manual completion of AML forms—are, indeed, being digitized, automation is a much broader concept encompassing big data, data lakes, machine learning and artificial intelligence. Combined, these technological advances are already revolutionizing how financial institutions and other large companies collect, manage and use data to inform strategy and make business decisions.
The implications for compliance are transformational and will ultimately change how this function works with the rest of the business, it is likely that it will increase, rather than lessen, a Financial Institution’s reliance on compliance professionals. Ultimately, the development and deployment of new software needs to be able to produce data that leads to actionable insights if it is actually going to meet the evolving demands of compliance departments. Although the challenge is significant, the opportunity is exponentially greater.
Judgement not monitoring
Compliance professionals argue their function has never been a matter of simply processing data. Rather, people such as Anne Murphy, head of U.K. financial services at executive-search firm Odgers Berndtson believe that the core function of compliance utilizes a company’s data to provide insight and judgement about how an individual company—with its individual structure, processes and objectives—optimizes the application of regulation within a commercial context. As a former compliance officer of nearly 20 years, Rachael Hutchins, Regulatory Business Manager, Bloomberg Tradebook, has witnessed the “expansion of compliance into a hybrid between the legal function and operational risk; an applied science that allows senior managers to proportionately focus their attention on the regulation that poses the greatest regulatory risks to the firm”.
To do this effectively, compliance ultimately seeks to ensure the quality of advice offered by an individual—which is also, in part, a function of the quality of information to which that compliance individual has access. This is where automation can significantly impact the compliance function.
As today’s large companies undergo a revolution in the sheer amount of data captured from both internal and external sources, new Hadoop technology developed by Google to index the huge amount of data on the internet (link: /) enables data lakes to be created that empower companies to combine unstructured and traditional structured data—such as spreadsheets. “Now audio files, movies, and social media excerpts can all be channelled into a large pool of data along with transactional data and other more traditional sources,” according to Rob Calichman, Head of Product, Bloomberg Vault and Directory. New cloud-based storage solutions also radically lower the cost of such data-driven approaches, so even relatively small companies are adopting best practices in this area.
The implications for compliance are clearly both enormous and challenging, and Calichman stresses that “compliance will be required to understand exactly what data their company is harvesting and storing, how the company collects and organizes this data, as well as the implications for existing and evolving regulations.”
A key issue for compliance professionals will likely be to avoid drowning in data. Which is to say, compliance systems need to be flexible enough to integrate information in various formats from different sources into meaningful analytics. Compliance will increasingly depend on “disruptive technology solutions that can integrate trading systems and Order Management Systems (OMS), as well as the company’s own data storage and analytics systems, including best execution and security programs,” says Michael Tirello, Global Compliance Product Manager, Bloomberg SSEOMS.
For years, many compliance professionals insisted that the function should be perceived as more than “a cost center”; that compliance can improve corporate performance by optimizing processes. Measuring that “value-added” in hard dollar terms, however, has been an elusive quest.
Building access to deeper and better data has the potential to change this. For example, as institutional visibility into cybersecurity improves, it is possible to measure—through penetration testing—the amount of time IT spends responding to hacking or ransomware attacks. The same is true of employees who are victims of ID theft. And because these have a measurable impact on employee productivity, it follows that reducing such incidents will likely have a measurable impact on company performance.
There are other case studies: An escalator manufacturer collated data on all its slip-and-trip claims, identifying a correlation between a spike of claims in December as retailers placed items close to the bottom and top of elevators. This discovery led to a procedural change that led to many fewer claims the following December—in other words, a solution.
As part of Bloomberg’s outlook for compliance, Hutchins sees that “it is entirely within the realm of possibility that these significant developments in automation will lead to many more such opportunities for compliance professionals to work as strategic business partners to improve many areas of operations. To that end, compliance teams should behave as such—experimenting with new data analytics and corresponding methods to identify workable solutions. If managed appropriately, these developments in data will ultimately elevate—not replace—the role of compliance.”