Benchmarks & Data Migrations
What is an Investment Benchmark?
A benchmark is a standard or measure that can be used to analyze the allocation, risk, and return of a given portfolio. A variety of benchmarks can also be used to understand how a portfolio is performing against various market segments. Using an index, it is possible to see how much value an active manager adds and from where, or through what investments, that value comes. The S&P 500 index is often used as a benchmark for equities.
Benchmark Data Migration Issues
During any Data Migration or System Conversion, it is easy to overlook the complexities surrounding the index/benchmark data. Typically during a system conversion, the firm must load all historical benchmark data (returns or values) from the legacy system to the new system. The new system vendor will most likely have the benchmark data, but will not populate the historical benchmark data for you. It is up to the firm to provide the historical benchmark data to be populated into the new system, of which, will then need to be linked to the newly provided benchmark feed in the new system.
There are numerous nuances regarding to converting benchmark/index data. Shown below are just a few to keep in mind:
Data Validation
Determine if the new vendor will have reports available to re-validate the newly uploaded historical returns.
Daily vs Monthly
In today's environment, it is assumed that all benchmarks would be populated daily. However, a periodicity assessment should be performed to ensure data consistency between the two systems.
Mapping Issues
Determine if the benchmark get mapped individually at the account level OR if there is some sort of mapping table where a field is populated on an account, and then references the benchmark table and assigns the benchmark accordingly? The identification of how the benchmark data appears on reports is a crucial mapping item.
Manual Input
Investigate whether or not any benchmarks are manually inputted into the legacy system, and determine the new process going forward.
Benchmark Similarity
Ensure that all benchmarks assigned to all strategies/funds are the SAME everywhere throughout the firm (marketing, research, statements, etc.) During your research you may find that one of your firm's strategies utilizes a different benchmark depending on what report is being utilized (client statement vs composite vs one-off report).
Naming Conventions
Ensure that all benchmark naming conventions from the legacy system are available to be transferred and populated into the new system.
Benchmark Determination and Availability
Determine ALL benchmarks that need to be populated from the legacy to the new system. Research client statements, composites, internal reports, presentations, etc., for all possible benchmark reporting. During this process, a firm may identify benchmarks that are no longer used, etc. Therefore, this identification process can also serve as a 'housekeeping' task. There is no need to convert and populate benchmark data that is no longer used.
Data Formats
When extracting benchmark data from the legacy system for population into the new system, this data must be formatted to 'fit' into the new system's benchmark data scheme. Some systems use prices to generate returns, and some systems populate just the returns. A conversion scheme for this data must be identified.
Calendar Variables
When converting benchmark data (especially if it is a price-driven calculation), all calendar date variables must be factored into the data conversion. Some factors to consider are:
Determine the month end date methodologies between the 2 systems (ie all months are considered 30 days vs actual # of days in the month)
Determine the impacts to the overall returns when the month end date is on a holiday or a Saturday/Sunday
Especially when using price-driven calculations, determine if weekend/holiday prices need to be populated in the historical returns file