"When you’ve finished changing, you’re finished".
I am not sure when the first performance team was put together, probably in the late 70s, but its role and function has changed and will continue to change. A big part of that change lies with technology which has both made things more efficient and also far more complicated. The original style of performance team had a basic set of tasks. Extract accounting data from the accounting system. Transform that data into performance numbers. Send a report to those that needed the information. When they first started, most teams gathered data over a quarter and sent it out eight weeks after quarter end. These days, things go a little faster, but not much. Reporting cycles are typically monthly with reports going out three to ten days after the month end.
Some teams are able to produce daily performance numbers, but these are “flash” reports as there is always the worry that some trades will be reversed or that some data has been omitted. Thus the end of month reconciliation is where the loose ends are tidied up. The analysis required has also expanded enormously so that performance teams also have to manage external data such as index data and market classifications and terms and conditions of bonds and exotic derivatives. Suppliers such as StatPro and our many competitors have provided large and complex software systems to manage these huge amounts of data. In winning a client’s business, the focus has tended to be on the calculation functionality (can you do top-down, single level attribution?) However, the day to day reality of the performance team member is that they spend 90% to 95% of their time fixing data or checking data.
This does seem a rather poor use of the undoubted expertise of most professional performance measurers. A performance analyst ought to be a value added job where insight and understanding of the nature of the performance, its sources and its risks and combination of risks of a portfolio are the prime responsibility, not endlessly reviewing data in spreadsheets to check for errors. Of course it is important that a new performance measurer “gets his hands dirty” by delving into the data to understand how different assets are computed; how else can you develop a profound understanding of the issues? It doesn’t matter how good your understanding of analytics is, garbage in means garbage out. However, understanding the issues and having to spend hours every day knee deep in data issues are two very different things.
In essence, performance measurement can be divided into two separate processes. Step one, create the performance raw data. Step two, produce a wide range of analytics based on this basic input data. Step one is about extracting accounting data and transforming it into performance data. This might appear obvious, but the problems of dealing with data are significant. An accounting system just needs to be right today. If a trade is put in on the wrong date, that might not change the balance sheet. However, such an error causes problems with performance as the data must be accurate for each day of the history of the portfolio. Step two today is all about the mass production of standard reports. This step is largely automated, but problems arise when the process experiences a problem like the data is late or a critical server falls over.
At StatPro our solution to step one, the data management problem, is our new service StatPro R+. It will automate most aspects of data management, making the task of getting accurate performance data much easier. It will also enable a daily process as its speed and scale makes this feasible. It is also fully audited so that all changes are always logged, meaning that compliance becomes simple.
The solution for the distribution of the performance and risk analytics is called StatPro Revolution. This service means that the performance team can share interactive analysis on any portfolio with anyone, anywhere (all controlled to restrict views, portfolios and dates as required). This means that fund managers, sales teams, distributors, clients and anyone else can access information about their portfolios when they want. You can still mass produce reports, but why? Instead of getting monthly paper reports that are out of date, you can get up to date analysis anytime you need (someone invaded the Crimea recently, what happened to my portfolio?) This is a profound change in the way that asset managers process information and communicate with their clients and distributors. Online banking has been around a long time and now it is time for online performance and risk analysis. Self-service is the best service.
So what does the performance team do? Well, apart from the fact that data still has to be managed and checked, we think that it means they can focus on their real job which is providing insight and support around the performance and risk numbers. It is all very well regurgitating huge volumes of analytics and sending it out to everyone, but what does it all mean? Perhaps performance measurers have become so used to being bogged down in data issues that they have forgotten about this vital role, but essentially, helping people understand what the data says, is far more important than just sending it to them. It is no different than sending someone the public accounts of a large bank and assuming that they will understand them. If however, you have a specialist on your side who can guide you through the morass of numbers and highlight the key drivers of performance (good or bad) and risk (too high or too low), then you can have insight.
Technology will once again change the performance team’s role, but it will also expand it and make it more interesting, adding more value. Going forward, performance measurers will spend much less time managing data and much more time analyzing portfolios. Now that is a change for the better.