How to...

• Overview
• Relative Sector Performance Analysis
• Sample Text Report
• Sector Weighting Constraints
• Sample Limit Input Screen
• The BIRR Advantage

Traditional methods of risk exposure control and performance attribution have often relied upon partitioning a universe of stocks into different categories. For example, an index fund manager might seek to match the benchmark index in terms of market valuation (so much in large stocks, so much in mid-cap, so much in small stocks) or industry groupings (retail merchandising, manufacturing, etc.). A more active investor would attempt to buy better-than-average stocks in each category and might also choose to depart somewhat from the percentages of the benchmark portfolio in order to tilt toward categories expected to do well and away from those expected to do poorly.

In addition, a growing number of fund sponsors make use of so-called "completeness funds" to control the level of bias in their portfolios -- i.e., toward or away from certain styles or sectors -- that can result from their managers' overall pattern of stock selection. The idea is not necessarily to eliminate style bets, but to ensure that such bets do not become excessive and, also, that they are not undertaken inadvertently. (For a discussion of completeness funds, see the article "Doing without Style" in the February 1996 issue of Institutional Investor, pages 79-85.)

Controlling category weightings is an approach that is easy to understand and seems intuitively reasonable, and (especially in the case of a large, diverse portfolio) often gives acceptable results. On the other hand, by itself it can be potentially risky in that it implicitly assumes that all firms in a particular category are similarly exposed to economy-wide surprises, and experience shows that this is not the case. A given stock may be far more sensitive to interest rates or the business cycle or other large forces influencing the stock market than another stock even with the same SIC code and comparable market size.

(Part of the problem is that industry groupings can be misleading. A firm whose primary business is in one area may have substantial involvement in another. Even a company concentrating in one field may have financial fortunes that are dependent on another, as in the case of a food service company that sells mainly to airlines, for example.)

The BIRR software includes tools for examining portfolio returns and risk exposures in a variety of ways, including on the basis of between-sector and within-sector performance. In addition, the optimizer enables you to control sector allocation while limiting exposure to economy-wide surprises at the same time. The following condensed excerpts from the BIRR User's Manual explain the capabilities in more detail.

Relative Sector Performance Analysis

The BIRR Risk and Returns Analyzer® enables you to see how much of a portfolio is invested in a given sector and to compare that with the sector weights of a benchmark portfolio of your choice. This enables you to examine a portfolio's performance on the basis of between-sector and within-sector investment decisions.

In addition to using BIRR's standard economic sectors, you can use the economic sectors defined by Standard & Poor's, or you can easily create your own. These sectors can be defined any way you like and are not necessarily restricted to industrial classifications. You could, for example, partition stocks based on market capitalization, on the basis of growth versus value, or in any other manner you find useful. Alternatively, you might choose to treat each industry analyst's area of responsibility as a sector in order to identify superior or inferior stock selection for each analyst.

To get an analysis of performance for the past n months based on default or user-defined sectors, choose the entry Performance by Sector Composition from the BIRR Historical Analysis Menu. In order for the analysis to be performed, both the primary asset and the comparison asset must be user portfolios other than return series portfolios. (That is, they must be portfolios defined by their stock holdings.)

Selecting Performance by Sector Composition takes you to the Sector Definition Menu which lets you choose the sector definition you prefer.

After choosing the sector definition, you will see the following menu:

Sector Performance Output Menu:
Text Report
Bar Graph of Sector Returns
Selection Returns Bar Graph
Contribution to Returns Bar Graph
Number of Months in Analysis

Selecting Text Report results in a report such as this:

Sector Weighting Constraints

BIRR enables you to control how much of a portfolio is invested in a given sector during optimization. You can set upper and lower investment limits on each sector to control diversification or to tilt toward or away from a particular sector or set of sectors while simultaneously limiting your exposure to economy-wide surprises.

Again, you can use BIRR's standard economic sectors, or you can use those defined by Standard and Poor's, or you can easily create your own. You might, for example, choose to treat each industry analyst's area of responsibility as a sector in order to control the funds allocated to that analyst. (That determination could in turn be based, for example, on BIRR's mean-variance asset allocation feature.) You could then simultaneously limit exposures to economy-wide surprises, control sector allocation, and maximize your analysts' stock rankings.

Economy-wide surprises result from unexpected changes in various long-term and short-term interest rates, inflation, real economic growth, and market sentiment. See the "The BIRR Approach to Controlling for Economy-Wide Surprises" for more information. Pages 26-28 of the Getting Started Handbook provide an example of how to use optimization to control for these economy-wide surprises.

Once you have set acceptable limits on your exposures to economy-wide surprises, in addition you can control sector weightings. Choosing Sector Weighting Constraints from the Linear Programming Menu takes you to the Sector Definition Menu which lets you choose the sector definition you prefer. Once you have designated a sector definition, you will see the following input screen:

If your current comparison asset is a portfolio defined by composition, its sector weightings will appear in the Comparison column. As a convenience, you can press the F8 key to set weight constraints of plus and minus 10 percent around the weights of the comparison asset, then adjust the individual values as you like. You can also use the F9 key to adjust the "tightness" of the constraints. When you press F9, you will be prompted to enter a multiplier. If you enter a value greater than 1, the ranges of the constraints will expand proportionately; a value less than 1 causes them to contract. More precisely, the distance of each limit from the corresponding value in the Comparison column is multiplied by the value you enter. If the value in the Comparison column is NA, the midpoint of the current range will be used as the basis for adjustment.

The BIRR Advantage

In summary, the BIRR Risks and Returns Analyzer® optimizer allows you to control sector allocation while simultaneously limiting your exposures to economy-wide surprises and, therefore, it provides you with important insurance against inadvertent risks and unwanted surprises.

Copyright © 1996 by BIRR Portfolio Analysis, Inc.