Editor’s note: Here is a second article by the wealth management technology firm Temenos, examining the issue of performance attribution. The authors are Lynne Landau, product manager for private banking, Temenos and her colleague, Susanne Kukkuk, product management in private banking.  

The 2011 Capgemini/Merrill Lynch Financial Advisor Survey reports that the financial crisis has created new priorities for high net worth investors, with 93 per cent of those surveyed stating that transparency on fees and statements, along with more specialist advice were now important requirements. The 2011 PricewaterhouseCoopers Global Private Banking and Wealth Management Survey also revealed that improving technology tools for client reporting will be one of the top three priorities for private banking and wealth management institutions in the next two years.

Today, the ability to analyse and explain performance results is an important requirement for financial institutions to be competitive. With more internal and external regulatory pressures and increasingly savvy clients demanding more information on investment performance, portfolio managers need daily insight of profit and loss and performance according to business area, market, geography, and so on.  

Performance attribution provides valuable insight into the sources of portfolio performance and is a useful tool to internally assess the investment decisions of portfolio managers. It helps to identify the most successful investment strategies and to understand why they succeed so that they can be replicated across more portfolios. Intermediate benchmarks are a powerful method to isolate and measure the impact of specific investment decisions individually and independently.

Performance attribution will also become increasingly important for external client reporting. Today, performance attribution results are primarily provided directly to institutional clients and family offices. This is mainly due to the fact that to understand performance attribution figures it requires a certain level of financial knowledge of the client and the market value of the portfolio must be high enough to allow frequent adaptation to the strategy. As the know-how of private investors on the financial markets is increasing, we expect to see a rising demand for in-depth performance analysis from private banking clients also.

Passive vs active

A prerequisite of performance attribution is contribution analysis, i.e. the breakdown of the performance into market segments. Performance attribution compares the returns of the individual market segments against the respective market index performances.

In a passively managed portfolio, the portfolio manager will replicate a market index, i.e. the weights of the instruments in the portfolio are identical to their respective weights in the index. This means that the portfolio’s performance will closely follow the performance of the benchmark. The costs of this strategy are generally low, because no research work is required but there is very little chance to earn more than the market return.

With an active mandate, however, the portfolio manager’s aim is to beat the performance of the portfolio’s benchmark by taking investment decisions to deviate from the composition of the benchmark. Each investment decision can be seen as a “bet” to invest into the “right” markets and securities. If the decision of the portfolio manager was right, he wins his bet and therefore adds a positive contribution to the total over-performance.

The Brinson model

The history of performance attribution started in 1986 with the Brinson Model (as developed by portfolio manager Gary P Brinson). Brinson et al showed that the portfolio return in excess of the benchmark return could be broken into two components: the market allocation (market selection effect) and the security selection (stock picking effect). If the portfolio manager deviates from the benchmark by overweighting the market segments for which he expects an above average performance, this is known as the market selection effect. If he gains a higher performance per market segment by selecting individual securities, which outperform the benchmark for this market, this is known as the stock picking effect.

To measure these effects, Brinson and others used the concept of an intermediate benchmark, which has the same market segment weights as the portfolio, but market segment returns as the original benchmark.

Constructing intermediate benchmarks

Investment decisions are often taken in a top-down approach. For example, at the highest level of the market structure, the weights for stocks and bonds are determined. The stock and bond markets are then independently managed by dedicated teams, which take the decisions on the weights of the different sub-markets (by country, currency, sector etc). Finally, a specialist for each sub-market is responsible for the selection of the securities. The Brinson model can be extended to this case by constructing intermediate benchmarks at the different levels, which isolate the various effects. This allows measuring the impact of each decision individually and independently of the other decisions to assess which team and sub-market generated the highest contribution to the total over-performance.

While stock portfolios are usually managed by weights, the portfolio managers of fixed income portfolios mainly target the duration of the benchmark. These investment decisions are based on the expectations about the rise or fall of the interest rates in a market in general (parallel shift of the yield curve) or on certain maturity bands (non-parallel shifts). The approach of intermediate benchmarks can be extended to this situation by adapting the construction of the intermediate benchmarks accordingly.

Performance attribution is a valuable tool to improve the investment process by identifying key sources of returns. The concept of intermediate benchmarks allows the analysis of consecutive investment decisions of different natures. As private banking clients become more informed and their expectations higher, there is an increasing need to present returns, describe the investment policies used and to quantify the risks taken within those strategies.

By providing this information financial institutions can show knowledge and differentiate themselves from the competition with in-depth client reports which helps them to retain existing customers and attract new ones. Banks who fail to implement advanced performance attribution capabilities risk losing clients as they don’t achieve the necessary insight in the success factors of their investment strategies and cannot evaluate and demonstrate their investment performance to their end clients in an appropriate way.