A Hidden Trap for Bond Investors

Index funds revolutionized the way people invest. But at its core, an index fund is only as good as the index it tracks. If an index doesn't have criteria that match up with the investment objectives that most investors have, then it won't yield the best results. Unfortunately, that may be the case for some bond indexes and the investments that track them.

Stocks vs. bonds
Over the years, there's been substantial controversy over the way stock indexes are calculated. Many indexes, such as the S&P 500 and the Russell 2000, use market capitalization as a weighting factor. That means that for the S&P 500, the biggest company, ExxonMobil, has an impact that's several hundred times greater than the smallest stocks in the index have. You can find indexes that use other methods, such as equal weighting of every constituent stock, or a price-based average like the Dow Industrials.

For the most part, though, weighting market capitalization makes intuitive sense. It means you end up with a bigger part of your portfolio in big, arguably more stable companies, and less in small, speculative plays. As a company grows, it gains more influence; if it shrinks, it gradually loses its power to have a major impact on your overall returns.

For bond investors, however, the best way to determine weighting for an index isn't as clear. Market capitalization may speak to the perceived security of the issuer -- the more a company's equity is worth, presumably the more secure bondholders should feel. But since it's an equity-based measure, market cap doesn't seem to apply as intuitively to calculating a bond index.

The perils of issuance-based weighting
The alternative that some bond indexes have used is to use a weighting that's based on the amount of bonds each issuer has outstanding. Put more simply, the more bonds a given government entity or company issues, the greater the weight it has in the bond index. And while that may reflect the composition of the overall bond market, it can lead to the unattractive result that the issuers whose bonds are most heavily represented in the index are those that have the most debt outstanding. That's not always a great measure of how successful those issuers will be.

For instance, take a look at the iBoxx USD Liquid Investment Grade Index, which underlies the iShares iBoxx $Investment Grade Corporate Bond ETF (NYSE: LQD). According to the company that maintains the index, the weighting of the index depends on the market value of each issuer's qualifying bonds. That calculation results in the following companies being the ones with the biggest exposure in the ETF:

Source: iShares. As of May 17.

Now note that the fund doesn't own all the bonds in the index; instead, it uses a sampling technique to choose around 400 of the 600 component bonds. But since the goal of the fund is to match the index, it's reasonable to assume that these weights are representative of the index as a whole.

Are those the companies you want the most exposure to? Notice how the index ends up overweighting fairly low-rated debt; over three-quarters of the fund's value is tied up in bonds with a rating of A+ or below, while less than 1% gets a AAA rating.

Know what you own
The trouble with weighting bond indexes actually gives credibility to the argument that active management may bear more benefits in bonds than in stocks. Unfortunately, because returns on bonds tend to be lower than stocks over long time periods, the higher expense ratios of active funds take a proportionately larger bite out of returns, giving even flawed bond index funds a potential advantage.

For investors, though, it underscores why owning individual bonds rather than bond funds may be the better move. It also serves as one more reminder that even with supposedly "safe" asset classes, you have to be on your toes. Otherwise, risk can show up in the most unexpected places.

Are stocks rising only because interest rates are so low? Morgan Housel looks at what Warren Buffett and Charlie Munger have to say about exploding markets.