Reporting a portfolio’s comparative performance against representative benchmarks over various periods of time has been an widely implemented technique of investment advisors for decades. Firms do this in a variety of different ways, and some methods are more meaningful than others.
Advisors may show the CPI as a baseline index against the portfolio and other indexes in an effort to illustrate how their investments have performed against inflation and hopefully point out the most basic benefit of investing with their firm versus keeping it under the mattress.
If a firm manages equity they may show the the Dow Jones or the S&P 500 indexes. If they manage international equities, or have a bond component they will likely show other representative indexes. They may further isolate performance by asset class and show it with the corresponding indexes to facilitate a client’s ability to draw a direct comparison between the performance of like asset classes and underscore the value they add to the investing process of specific asset classes.
PROBLEMS WITH TOTAL PERFORMANCE BLENDS
Firms typically show total performance too, but many run afoul when trying to show a representative index to match total portfolio performance. Portfolio management systems like Advent Axys and APX have long supported the ability to create blended indexes, but the actual implementation of the feature falls short of real-world requirements. In the past, many advisors were content to create a blended index with static weights, but firms soon recognized that these ballpark blends weren’t good enough for their requirements because they didn’t accurately reflect the asset mix of their portfolios over time.
Advent Software initially supported the creation of blended indexes in their portfolio management systems (Proport and Axys) through a report that was engineered to create simple index blends. This method of creating indexes was somewhat limited because index weights couldn’t be changed over time, but it allowed advisors to manage many indexes centrally through the use of scripts and macros, and it also created separate index (DEX) files. Having separate files allowed clients to track and show a number of blends if they chose to.
Advent improved the feature to some degree by creating the synthetic index, which allows users to change index weights over time on an individual portfolio basis. The synthetic index feature works best when used to create ad hoc indexes for selected portfolios, but when users try to scale the feature to be used for a large number of portfolios its usefulness tested. The ability to change the index weights over time is a plus, but updating the info in an automated fashion is not a feature of the system.
AUTOMATED BLENDING SOLUTIONS
In the early 90s, we started working with advisors to create custom blends with dynamic index weights in order to match the ever changing asset mix of portfolios. This was done by categorizing assets as fundamental asset types, and assigning those asset types to like indexes. Once the infrastructure was created we created an extract containing the historic month end asset type weights for all portfolios and used those weights along their corresponding index returns over time to create a true blended index. The process was automated with a combination of Axys reports, macros, scripts and Visual Basic.
Years ago, we helped firms address the some of the short-comings of the synthetic index feature by creating a product dedicated to creating blended indexes and managing them. This allowed users to manage the indexes centrally rather than at the portfolio level.
Last year, we revisited the problem of effectively managing custom index blends for Axys and APX users again. This time we automated the process of refreshing the index weights stored in each portfolio’s corresponding performance file through a process that updates the synthetic weights by stripping the performance files of their synthetic weight values and rewriting them based on each portfolio’s asset mix and correspondingly assigned indexes. The process builds monthly synthetic index weights for the inception-to-date period.
This most recent update to our index blending automation tool leverages Advent’s synthetic index capability, and adds the much needed automation that users require to implement the existing synthetic index feature in a way that makes more sense. Unfortunately, there are still limits to the built-in blending functions of Axys and APX. For example, synthetic indexes create a blend called simply “Blend”, and users trying to show multiple blends need to create those other blends somewhere else and then generate index files, so that they can be shown alongside synthetic blended indexes.
Thankfully, Advent’s import/export functions, scripting and macros allow savvy users to workaround these limitations as long as they are willing to roll up their sleeves and do a little work or know someone who can help.
–
About the Author: Kevin Shea is President of InfoSystems Integrated, Inc. (ISI); ISI provides a wide variety of outsourced IT solutions to investment advisors nationwide.
For details, please visit isitc.com, contact Kevin Shea via phone at 617-720-3400 x202 or e-mail at kshea@isitc.com.