Managing Data Quality (DQ) has held a stable position within the top priorities of insurance companies, in particular for global employee benefit (EB) captives. Vittorio Zaniboni, EY Luxembourg Director and Captive & Insurance Excellence Leader, shares some practical advice for managers on how to implement data quality initiatives.
Not only do EU regulators consider Data Quality (DQ) a fundamental pillar of a successful Solvency II implementation, but the industry at large has developed a growing sensibility towards its importance in all the aspects of running business. When focusing on global EB, its role can become even more critical.
Still, research has shown that one third of insurance organisations either don’t have a DQ policy in force, or have it covering only regulatory purposes.While it is relatively easy to understand the importance of DQ in providing accurate underwriting and pricing decisions, it is also important to realise that it represents a continuous effort for companies to adopt best practices in data collection, management and analysis.
The fundamentals of Data Quality and its importance for the industry
DQ refers to the accuracy, completeness and consistency of the data sets collected and used by insurance companies. Accurate data is essential for effective decision-making, as incomplete or inconsistent data can lead to incorrect decisions.
Although estimations show that only 50% of insurance companies had a formal data governance committee in place in 2020, proper data management and governance are necessary to comply with regulatory requirements and avoid potential penalties.
Looking at the characteristics of EB related data, it tends to be more fragmented than data sets relevant to other lines of risks (in fact it spans across HR data, health data, financial data, personal data which often also involves the families of the employees, etc.).
The supply chains of EB data are also likely to be more complex and articulated, involving local HR departments, local brokers, third-party administrators, managing general agents, local insurers, fronting networks, etc.
This makes the challenges a global EB captive has to face to govern its DQ significantly more complex.
Implementing best practices to mitigate EB related risks
Considering that captives rely on data to underwrite and price risks accurately, inaccurate data can lead to mispriced policies, resulting in lost profits and even insolvency.
In addition, it is important to consider that the data received from the EB captive’s providers is not only used for underwriting and pricing, but for all regulatory needs, e.g., solvency capital requirement (SCR) calculations, cash and investment management, strategic decisions, risk appetite management, risk retention management and purchase of retro-protection, down to possible decisions on local benefits levels.
To efficiently ensure a proper DQ framework, EB captives can put several best practices in place:
1. Establish DQ Standards: Standards to assure data accuracy, completeness and consistency should be established in collaboration with all stakeholders, including data collectors, analysts and IT staff. They should be clearly defined, documented and communicated to all relevant parties, especially if the captive is using several fronting networks at the same time. Ensuring a consistent set of DQ standards across different fronting networks can prove to be particularly challenging.
2. Collect Relevant Data: The data collected should originate from reliable sources, such as plan administrators, medical providers, local insurers and other relevant parties, and be properly aggregated and organised by the various stakeholders along the value chain (fronting networks, global brokers, etc.). The data should also be collected on a regular basis to ensure that it is up to date.
3. Automate Data Collection: Manual data collection processes are prone to errors and inconsistencies. Captive insurance companies can use various solutions to automate data collection, such as electronic data interchange (EDI) or dedicated web-based portals. Very often EB captives tend to tame the complexity of the data sets they receive from their providers, by adopting an overly manual approach, and do not dedicate enough energy and resources to the digitalization of data-related internal processes. This approach, besides being inefficient, will rapidly show its limits, as soon as the volume and complexity of EB risk managed by the captive increases, making the scalability of this setup rather difficult.
4. Validate Data: Data validation involves checking data for accuracy and completeness. Validation processes can include data profiling, data cleansing and data enrichment. Captive insurance companies can use various tools and software solutions to do so, always focusing on preserving data integrity, while ensuring that the data flows they receive are consistent with the specifications negotiated with the data providers.
5. Integrate Data: Data integration involves combining data from different sources to provide a complete view of the risk being (re)insured. It can help identify correlations and patterns that would not be visible in individual data sets. A typical example of such integration is represented by the “medical claims reports” provided by several fronting networks to their captive clients, on their medical portfolios. In this case, the pure accounting data provided in the cession framework is not enough for captives to properly assess the performance of those schemes and needs to be integrated with non-accounting data sets. Some fronting networks have recently started to extend this approach to long-term disability (LTD) annuities, where the proper evaluation of mathematical reserves ceded to captives requires a wider range of information than that usually provided in the cession framework.
6. Monitor DQ: Monitoring should be performed on a regular basis to ensure that DQ standards are met, and issues should be addressed promptly. Regular complete audits can also help companies to i) ensure that standards are being met and ii) identify areas where further improvements are needed.
Implementing a holistic set of DQ initiatives clearly requires a multi-faceted approach that involves people, processes and technology. Diving further into the detail, here are some practical steps that captives can implement:
1. Appoint a DQ manager: The DQ manager should have the necessary skills, knowledge and empowerment to oversee and manage DQ processes in an effective way. Due to the limited headcount of most EB captives, this role does not necessarily need to be full time, but assigning DQ accountability in a clear way within the organization, is a fundamental step to ensuring an effective governance.
2. Establish a data governance framework: The framework should outline DQ standards and processes and be communicated to all relevant parties. Regular communication should be provided to ensure that all stakeholders are aware of their roles, responsibilities and interdependencies.
3. Use DQ tools: In order to identify and address DQ issues, captives should choose tools that align with their specific needs and requirements. The technology aspects mentioned above (alongside people and processes) are fundamental, but often neglected. Considering the volume and complexity of business data handled by captives, it is of critical importance to make use of the proper tools to support the DQ activity.
4. Implement DQ checks: Captives should establish a clear and articulated protocol of checks at key stages of the data lifecycle, including data collection, processing, and reporting, with acceptable ranges for every KPI, and a mechanism of escalation and alerts in case of deviations. Being concerned about DQ and “checking numbers” is not enough to execute an effective governance; it is important to lay down a complete set of checks and actions to be sure that the procedures do not remain an end in itself. Further, an important step is then to assign remediation measures to be implemented, in case a DQ issue is detected.
Ensuring DQ is essential for captives that (re)insure employee benefits: poor DQ can lead to inaccurate reporting, increased risk and poor decision-making, in so doing defeating the many advantages a corporate can achieve by consolidating its EB risk in its captive.
To achieve the goal of DQ, it is crucial to adopt a holistic approach, engaging all stakeholders, using the appropriate and relevant tools and including in the scope all the data along the value chain.