Using Indices as Benchmarks

Chances are, if you ask a stranger where the Dow Jones Industrial Average (the Dow) is trading, they'll come back with an answer that's reasonably accurate. Why? Because index results are reported every-where-in the public media and in the information systems used by professional investors like Toron.

Given the prevalence of indices in the investing world, we need to examine what’s behind them. Why Does Index Construction Matter?

Index construction matters because people use indices to make investment decisions -either directly (via index investing strategies) or indirectly (using them to benchmark investment performance). They are appealing because they provide certainty and simplicity in a market characterized by uncertainty and complexity.

But indices are imperfect measuring devices. To illustrate this point, take a stab at the following question: How did the US equity market perform over a 5 month period?

  1. 7.2%
  2. 10.4%
  3. 11.3%
  4. 11.8%
  5. 16.0%
  6. 30.4%
  7. all of the above

Believe it or not, the answer is g). These figures represent the total return (i. e. including dividend reinvestment) for the following indices respectively:

  • the Dow
  • S&P 500
  • Wilshire 5000 (Value Weighted)
  • Value Line Composite (Geometric Average)
  • Value Line Composite (Arithmetic Average)
  • Wilshire 5000 (Equal Weighted Geometric Average)

You may not be aware of some of these indices but investors use all of them as barometers of the broad US stock market's performance. A 420% difference between high and low returns for the same market (supposedly) should be sufficient to illustrate that index construction matters. ↑ top

The Market vs. The Market Sample

When you use an index as a proxy for "the market", it is important that you understand the sample you are using. Breadth of sample is key. The Wilshire 5000 index is the most comprehensive measure of the U. S. stock market. This index measures the performance of all U. S. headquartered equity securities with readily available price data.

The following schedule compares a few of the key properties of this index to the S&P 500 and the Dow:

  Companies Total Market
Capitalization		 Smallest Company
Wilshire 5000 6,500 $11 trillion < $1 million
S&P 500 500 $9 trillion $340 million
the Dow 30 $3 trillion $7.9 billion

As you can see, the number and size of companies that underlie these "broad" indices are quite different.

It is also important to know that the companies that comprise an index change over time - one should remember that Enron was once a prominent S&P 500 stock. ↑ top

From Sample To Index

Having chosen a sample, the next step is to compute the index. The index calculation transforms the performance of each security in the sample into one aggregate performance number.

There are three primary ways to compute an index:

Description Example Issues
arithmetic average of prices the Dow high priced stocks dominate companies that split their stock (growth companies) have less impact -results in downward bias
weighted average of prices based on market capitalization of each company S&P 500, Wilshire 5000, S&P/ TSX Indices large companies dominate (remember Nortel's influence on the TSE ?)
geometric / arithmetic average 1 of % price changes Value Line Composite all stocks are weighted equally regardless of price or market value

↑ top

Beware Of The Bias

Index construction and maintenance generally results in the introduction of bias. "Success" and "survivorship" are two common biases that exist in equity market indices. Any index that is based on "large" or "the largest" capitalization stocks (e.g. the Dow, the S& P 500, etc.) have a built-in success bias because only the companies that grow large enough for inclusion in the index are represented -the companies that fail to grow do not make the cut. Similarly, most indices have inclusion criteria such as "financial viability", "reasonable price" etc. Survivorship bias is similar – indices only reflect the performance of the companies that survive – those that fail are dropped from the sample. Both success and survivorship bias result in higher performance figures than those actually delivered by "the market". ↑ top

Benchmarking Performance Using Indices

The issues arising when using indices as benchmarks in the evaluation of investment performance are numerous:

The benchmark should be consistent with your investing universe and philosophy. It is important to choose an appropriate combination of indices to match your portfolio structure. A broad value-weighted index such as the Wilshire 5000 is a reasonable benchmark for the overall stock market. An equal-weighted index is likely a better but not perfect benchmark for properly constructed equity portfolios.

Returns are important but so is risk. Indices (particularly value-weighted ones) can be dominated by particular sectors and/ or companies and hence can be more volatile than properly constructed portfolios. In addition, currency return, while usually a minor factor, must be considered. Recent strength in the Canadian dollar has clearly illustrated this point. ↑ top

Key Issues To Consider When Using Indices

The indices we use to measure investment performance are complex and imperfect. In fact, popular indices can become self-fulfilling prophecies -for a while (remember, a manager trying to match a value-weighted index must often buy the stocks that have become most expensive). While indices necessarily play a role in the evaluation of investment performance, their characteristics should be understood and they should not be the only tools used. The focus of investors and the managers they employ should always be on structuring and prudently managing portfolios that satisfy individual objectives and constraints (including risk).

After all, should you buy a stock because it satisfies the inclusion criteria for a particular index or should you buy a stock because it fits in your portfolio, you understand the company's business and prospects and its price is reasonable? Toron's focus is very much on the latter.

↑ top