Part of the difficulty of providing the required look-through for Solvency II reporting is the complexity involved in processing the fund data. In this article Josef Molvidsson, Product Manager, Licensed Data, Morningstar, offers a data provider’s perspective on the process.
Solvency II Portfolio Data
Solvency II look-through reporting requires insurers to detail their holdings across the entire investment portfolio. One of the difficulties this raises is the complexity of obtaining the relevant data, especially for funds of funds, where a single portfolio contains a number of funds. An insurer’s overall investment portfolio holdings could well consist of holdings of the same securities either directly – through a single fund – or indirectly through an investment in a fund of funds. One challenge, therefore, is to be able to isolate the holdings of each individual security held in the different investment vehicles.
What follows is a description of the key stages needed to extract data from the investment vehicles and prepare it for Solvency II reporting. It offers an insight into the complexity of the process.
Collecting the data
As in the usual collection of data for fund performance analysis or other measures, fund companies will submit their full holdings to the data provider. Data relationship managers and local country data teams will work with the fund group to establish data supply arrangements including timings, formats, and the implementation of data feeds.
Collecting data for a fund with a single portfolio of securities is relatively straight forward. The challenge arises when collecting data from a fund of funds, which holds a number of funds in its portfolio (each with an underlying portfolio of securities). Collecting the correct data from these funds, with which you may not always be familiar, poses a more significant challenge.
The chain of data extraction consists of numerous steps, including: sourcing of information, uploading data, processing data, internal output, packaging and delivery. Because of the multitude of funds involved there are many points where the chain could break or get disrupted. The most common causes are using the wrong source data for the holdings or mapping to the wrong fund. A quality assurance team is therefore required to monitor the process from start to finish, consisting of an acquisition team, a standardisation team, and proofing team.
The acquisition team will manage the relationship with each fund company, including managing the reminder process for portfolios, resolving issues of mapping portfolios to funds, and communicating with fund companies on data related issues, such as incorrect data.
For maximum quality assurance, data should be sourced directly from each individual fund company or their appointed custodians or service providers. In most cases, retrieving the data via an FTP feed will provide the easiest and most straightforward route, but email or web interface supply is also possible.
The preferred file format, common with all major fund companies, typically includes: date of the holdings positions, unique identifier of the fund, fund name, identifier of each security, security description, shares/par value, market value number, coupon rate, maturity date, and portfolio base currency. Captured fixed income would typically include credit rating risk, interest rate risk, yield, and currency exposure.
The standardisation team performs control and adjustments on file formats and uploads the normalised data files to the data company’s independent master database. Additionally the team will develop system enhancements to increase the flexibility of receivable formats from fund companies.
The proofing team will validate the data received and resolve issues reported by the suite of quality assurance (Q&A) checks conducted on both the raw and derived statistical data. Typically, these Q&A checks are built on Lean and Six Sigma principles, and regular audits of the data should be performed to check and improve data quality. Such checks should be embedded throughout the operations process to ensure data is timely, accurate and complete.
Before the data can be considered clean for Solvency II reporting the full portfolio data must be validated. Pre-reporting checks should include: abnormal change of asset allocation, country exposure and Total Net Assets (TNA), consistency of TNA with performance figures, and implied stock price derived from number of shares and market value consistent with stock price at portfolio date. Data providers will also use historical portfolios, category averages and refined rules based on calculated key figures on the portfolios to evaluate the correctness of the data.
Enhancing the data for Solvency II
As part of Solvency II reporting, there is a requirement to uniquely identify all holdings against a set of Complementary Identification Codes (CIC) defined by EIOPA. The ‘CIC code’ (as it is commonly referred to) is a four digit number based on the type of investment for all funds and individual securities held by institutional investors. The solution requires the data provider to have in place rules and data validation processes and leverages real time global market data.
To map securities, each security type is grouped together to generate high level asset classes, with other layers of data attached to each security. These additional layers allow the data provider to generate different asset breakdowns and analyse portfolios from different risk perspectives. Once all holdings are mapped, all doors are open to further enhance the data by adding more data points.
The ability to analyse fund portfolios based on an insurer’s underlying securities (all securities held in the different investment vehicles) is substantially dependent on the scope, quality and maintenance of the securities Masterfile. The Masterfile contains two sets of data: information about the individual security (name, ISIN, price, etc.) and the enhanced data (CIC, ratings, etc.) which further defines the securities. The information in the Masterfile is then matched to the portfolio holdings. It is imperative, therefore, that insurers ask key questions relating to the data provider’s ability to map files, the number and type of securities in the data provider’s master database.
Once the complete core mapping of the underlying holdings is secured, it is possible to add all the other type of data, such as credit ratings, fundamental bond and stock data, pricing, etc., for reporting purposes.
The Solvency II Reporting Templates
Solvency II levels of portfolio coverage requires a substantial and long term commitment in terms of expert personnel covering markets as well as data processing, building, maintaining and evolving the necessary technical knowledge and infrastructure. The complexities of the process mean that data providers must stand ready to work with both insurers and asset managers to provide solutions.
Link to Symposium index page.