Reading 19 Market Indexes and Benchmarks Flashcards
Fundamental-Weighted Indexes
The proponents of fundamental-weighted indexes argue that capitalization-weighted indexes overweight overvalued issues and underweight undervalued issues. Fundamental-weighted indexes seek to avoid this problem by using valuation metrics, rather than market value, to weight index constituents. Proponents also assert that these indexes will be more representative of an issuer’s importance in an economy because they weight by fundamentals, rather than by market prices subject to bubbles.
- A disadvantage of fundamental-weighted indexes is that they reflect the index creator’s view of valuation, which may or may not be correct.
- A second disadvantage is that they may be less diversified than capitalization-weighted indexes if the valuation screen is restrictive.
- Third, as mentioned previously, not all investors could hold a fundamental-weighted index because they are weighted by valuation metrics, not by available liquidity (market capitalization).
- Fourth, the construction methodology used by these indexes is usually proprietary because they are created to market a fundamental-weighted fund. In this case, these indexes would not serve as valid benchmarks because their composition and weightings are not fully known.
- Fifth, just because fundamental-weighted indexes have outperformed capitalization-weighted indexes in certain past time periods does not mean they will in the future and, more importantly, does not indicate that they are useful indexes. As a reflection of “how the market did,” an index should be judged on whether it is a representative sample, not on its performance. It is active management that is intended to deliver superior performance, not indexes. Index performance should be an outcome of index construction, not an objective. Furthermore, the reason why fundamental-weighted indexes have outperformed historically is that they are usually tilted toward small-cap value stocks, a well-documented effect in academic literature. For investors preferring a large-cap or growth emphasis, fundamental-weighted indexes would not serve as ideal benchmarks.
Index Construction
There are three primary choices in index construction:
Inclusion criteria: the first choice, the inclusion criteria, determines which specific population of securities the index represents.
Security weighting: the second choice, selecting the methodology for security weighting, is usually a choice among value, price, or another weighting scheme.
Index maintenance: the third choice relates to the index’s maintenance rules, which will influence the index’s performance and its applicability as a benchmark. One should be aware of such differences before comparing a manager’s portfolio to an index.
Equal-Weighted Indexes
- The primary advantage of equal-weighted indexes is that they give smaller weights to large-cap securities (and larger weights to small-cap securities) than indexes formed by cap weighting. By reducing exposures to the largest-cap securities, the index is thereby less concentrated in those securities and more diversified. Proponents of this approach suggest that market prices do not always reflect true value and that equal weights reflect a more naive or informationless weighting of a portfolio, resulting in superior returns during particular historical periods.
- Second, these indexes may better represent “how the market did” because they provide an average of all index security returns.
However,
- Some argue that this type of weighting results in a small-issuer bias, the primary disadvantage of equal-weighted indexes, because small issuers are weighted the same as large ones.
- Second, to maintain equal weighting, strong-performing issues must be sold and weak performers must be bought, resulting in frequent rebalancing and high transaction costs for a portfolio tracking such an index.
- Third, the inclusion of small issuers means that investors tracking the index may not be able to find liquidity in some of these issues.
Types of Benchmarks
Given the uses described above, benchmarks are an important part of the investment process for both institutional and private wealth clients. The seven types of benchmarks are:
- absolute (including target) return benchmarks;
- manager universes (peer groups);
- broad market indexes;
- style indexes;
- factor-model-based benchmarks;
- returns-based (Sharpe style analysis) benchmarks; and
- custom security-based (strategy).
Custom security-based benchmarks
Custom security-based benchmarks are built to accurately reflect the investment discipline of a particular investment manager. Such benchmarks are developed through discussions with the manager and an analysis of past portfolio exposures.
Custom security-based benchmarks are also referred to as **strategy benchmarks **because they should reflect the manager’s particular strategy. Custom security-based benchmarks are particularly appropriate when the manager’s strategy cannot be closely matched to a broad market index or style index.
Market Indexes as Benchmarks
Capitalization-weighted, float-adjusted indexes may have several limitations for use as benchmarks.
- First, a value-weighted approach might not be compatible with a manager’s investment approach.
- Second, some construction rules might be less transparent than desired. As an example, consider float adjustment. Although desirable, some index providers do not fully disclose their adjustment process, and there is also no standardization of the process among providers. Moreover, as an index is reconstituted, its composition changes over time, sometimes in non-predictable ways. As a result, an index’s style, sector, and risk exposures can change drastically over time.
- Market indexes are designed to be representative of a market segment, not a particular manager’s portfolio. The performance of a broad market index is usually influenced by many factors, few of which would be related to a manager’s style or skill.
In summary, market indexes are similar to Swiss Army knives; they serve many useful purposes but are sometimes not the best tool for the job. If market indexes do not capture a manager’s investment process, a custom benchmark is often constructed
Returns-based benchmarks
Returns-based benchmarks (Sharpe style analysis) are similar to factor-model-based benchmarks in that portfolio returns are related to a set of factors that do well in explaining portfolio returns. In the case of returns-based benchmarks, however, the factors are the returns for various style indexes (e.g., small-cap value, small-cap growth, large-cap value, and large-cap growth). The analysis produces a benchmark that is essentially the weighted average of these asset class indexes that best explains or tracks the portfolio’s returns.
Valid benchmarks, properties
Valid benchmarks will be:
- unambiguous,
- investable,
- measurable,
- appropriate,
- reflective of current investment opinions,
- specified in advance, and
- accountable (“owned”).
Market index, definition
A market index represents the performance of a specified security market, market segment, or asset class.
Defining index weighting
There are several weighting schemes commonly used:
1. Capitalization weighting, also known as market value weighting, market cap weighting, or cap weighting. The most common weighting scheme used is capitalization weighting. In this scheme, constituents are held in proportion to their market capitalizations, calculated as price times available shares. The performance of a value-weighted index represents the performance of a portfolio that holds all the outstanding value of each index security. By far, market capitalization weighting has the greatest acceptance by investment professionals.
Nearly all capitalization-weighted indexes are adjusted for the free float. The free float is the amount of shares outstanding for a given company that is available to the public. This adjustment is intended to exclude the capitalization of a company that is not widely available for purchase and thus is not part of the investable opportunity set. The resulting index is called a free-float-adjusted market capitalization index, or float-weighted index for short. A float-weighted index represents the performance of a portfolio that holds all the index securities available for trading.
Float adjustments result in the index being more investable.
2. Price weighting. In this scheme, constituents are weighted in proportion to their prices. The index value thereby can be interpreted simply as an average of the constituent prices. The performance of a price-weighted index represents the performance of a portfolio that holds one unit of each index security. Although the advantage of price weighting is its simplicity, the scheme offers little relevance to the way most investors weight their portfolios.
3. Equal weighting. In a pure equal-weighting scheme, all constituents are held at equal weights at specified rebalancing times. The performance of an equal-weighted index represents the performance of a portfolio that invests the same amount of wealth in each index security. Variations of this approach might weight groups of constituents (such as sectors or industries) equally. Equal-weighted indexes must be rebalanced periodically (e.g., quarterly) to reestablish the equal weighting because individual security returns will vary, causing security weights to drift from equal weights.
4. Fundamental weighting. This weighting scheme uses company characteristics other than market values, such as sales, cash flow, book value, and dividends, to weight securities. By forming weights based on variables considered important (fundamental) for valuation, these indexes seek to weight securities using true values, rather than the market prices of capitalization and price weighting. The performance of a fundamental-weighted index represents the performance of a portfolio that invests according to valuation metrics for a security.
Use of Market Indexes
- Asset allocation proxies
Used for asset allocation, an index constructed consistently over time provides the investor a tool to measure asset class ex ante return, risk, and correlations. It allows investors to determine the incremental expected return and risk from adding a new asset to a portfolio. These measurements can be used to design an investment policy suitable for different risk aversion levels.
- Investment management mandates
As a result of their effectiveness as asset allocation proxies, investment mandates can include a specified benchmark index.
- Performance benchmarks
Indexes are often used as ex post performance benchmarks, where they answer the basic question, did the manager beat the market?
- Portfolio analysis
In addition to benchmarking the manager’s performance, indexes can be used for more detailed portfolio analysis. For example, currency-hedged and unhedged versions of non-domestic indexes can be used to measure the effectiveness of a currency management strategy.
- Gauge of market sentiment
Possibly the most common use of indexes is as a gauge of public or market sentiment. They answer the question, how did the market do today?
- Basis for investment vehicles
Indexes are also used as a basis for investments, such as index mutual funds, many exchange-traded funds (ETFs), and derivatives.
Index Construction Tradeoffs
Completeness vs. investability
In principle, striving for complete coverage of a market would suggest including every possible security in the investment universe available to investors. Doing so, however, would include many small-capitalization securities that are too illiquid and could not be purchased in amounts relevant to institutional investors. Eliminating hard-to-trade securities improves the investability of an index. Index designers must decide how broad their indexes can be while maintaining adequate investability.
Investability is not, however, the same as liquidity, as emerging markets sometimes illustrate. A security may have excellent liquidity in its home country but may not be investable for foreign investors if the country’s government imposes restrictions.
In choosing an index, the fund creator should consider how the index has balanced the tradeoff between completeness and investability. A more complete index can provide broader, more diversified performance. However, investability is an important concern for managers facing frequent and uncertain withdrawals.
Another consideration is that portfolios tracking more popular indexes tend to have lower trading costs, owing to their greater liquidity.
Reconstitution and rebalancing frequency vs. turnover
The processes of reconstitution and rebalancing are designed to keep an index close to its intended membership and weighting criteria. Reconstitution refers to the process of adding and dropping securities from an index, whereas rebalancing refers to a readjustment in the weights of existing securities.
In principle, more frequent reconstitution and rebalancing would be desirable in order to improve representativeness were it not for the fact that such activities create turnover in the index. Turnover is costly for investors who wish to track an index because they must trade more often. Therefore, there is a conflict between representativeness and low turnover.
Turnover is a potential concern for managers who track particular size and style indexes because they must trade as companies migrate from one classification to another. As a result, these indexes are less frequently reconstituted by providers.
Rebalancing is of particular concern in international indexes, where float adjustments are more important.
Objective and transparent rules vs. judgment
Security prices rise when securities are added to an index and fall when they are deleted from an index. Index reconstitution results in decreased returns for managers tracking the index because they have to buy added securities at higher prices and sell deleted securities at lower prices. These price changes will be more acute for more popular indexes.
Transparency and objectivity are desirable characteristics of indexes because they allow investors to readily predict the changes in index constituents that might occur.
All index constructors, however, exercise some degree of judgment in applying their methodologies. Doing so allows providers to adapt to changing circumstances or special situations that might not have been anticipated by explicit rules. Some providers intentionally use committees to choose index constituents, for example, in order to obtain certain characteristics that might be difficult to describe in rules. Index designers must decide when to use judgment in their methodologies and how much to use.
Price-Weighted Indexes
The main advantage of price-weighted indexes lies in the simplicity of their construction.
There are several disadvantages of price-weighted indexes:
- First, price-weighted indexes are overly influenced by the highest-priced securities because they have greater weights in the calculation of the indexes’ returns. Because price-weighted indexes weight by price instead of total market value, they do not necessarily reflect the economic importance of issuing companies.
- Second, stocks that appreciate often experience stock splits, and their weights in the index will decrease. Successful companies will become underrepresented, creating a downward bias in the index return.
- Lastly, price-weighted indexes assume an investor holds one unit of each security in the index, which does not describe how most investors form portfolios.
Absolute return benchmark
An absolute return benchmark is simply a minimum target return that the manager is expected to beat. The return may be a stated minimum (e.g., 9%), stated as a spread above a market index (e.g., euro interbank offered rate + 4%), or determined from actuarial assumptions.
Liability-based benchmark
A liability-based benchmark will match the duration profile and other key characteristics of the liabilities. A liability-based benchmark typically consists of nominal bonds, real return bonds, common shares, and other assets. Unlike market indexes in which the components’ weights typically reflect relative overall market values, in a liability-based benchmark component weights are determined based on the requirement that the benchmark closely track returns to the liabilities.
