Interview with Gus Sauter, Vanguard Index Fund Manager By John Spence October 16, 2002

As managing director of the quantitative equity group at Vanguard, Gus Sauter oversees the management of $200 billion in 24 index funds. He is also responsible for the management of $4 billion in Vanguard's actively-managed funds, and $1 billion in Vanguard VIPERs exchange-traded fund shares. Mr. Sauter recently celebrated his 15-year anniversary at Vanguard - he was hired just two weeks before the October 19 crash of 1987.

Q: We're seeing index providers slice and dice the universe ever finer with new investable benchmarks, and new players like Morningstar are entering the field. Also, Vanguard's move to potentially switch to new indexes for its passive funds is unprecedented. Will this foster more competition among the index providers, and if so how is this a good thing for passive investors?

A: A lot of the flood of activity in the index creation business is the result of a surge in exchange-traded fund development, starting about five years ago. Before that time, development of new indexes was done in a workmanlike fashion, where you'd see something introduced every several years. As ETFs became more popular in the last few years, many index providers wanted to supply benchmarks for new ETFs, which resulted in an explosion in the creation of indexes. I think one benefit in this is that there have been new ideas on how to build indexes, and we've seen some well-constructed benchmarks being created.

At Vanguard, we've developed our own thinking internally over the years, and we've incorporated a lot of other people's ideas into our thinking. And perhaps we developed a couple of ideas of our own. So we compiled those ideas in my article on ideal index construction methodology [available in The Journal of Indexes archives].

None of the existing indexes satisfy all of our criteria, but all of them satisfy some of our criteria. I think creating indexes that get closer and closer to what we modestly think of as ideal is beneficial for investors. At the same time, creating a "me too" index probably doesn't advance the ball that much.

We've seen a lot of the new indexes - I'll cite the Dow Jones [style] indexes - incorporating some good new ideas. We've seen changes in the existing indexes that have been by and large good, although with some pain. For example, international indices like the MSCI series adjusted for float weighting. There was some pain when Standard & Poor's announced they were dropping nine non-U.S. stocks from the S&P 500. But I do believe these moves enhanced the quality of the index. So we have seen the evolution of indexes to a higher ground.

In a market that's providing in excess of 20 percent returns, do you really care if one or two percent is siphoned off to pay a mutual fund family or pay for transaction costs? So the issue of costs gets lost.    - Gus Sauter, The Vanguard Group

Q: Do you think best practices for indexes are still emerging?

A: I do, but it happens in stages. Is there going to be something that absolutely changes the way indexes are thought of? No, I don't think so. You can, in our view, improve upon what is out there today. You can push the envelope, but don't expect a complete revolution.

Q: Do you maintain a dialogue with the index providers?

A: We maintain relationships with index providers. But I wouldn't say that we have open dialogue with them, for example we don't participate in index advisory councils. We think it's beneficial to avoid any appearance of conflict of interest. We maintain an arm's length relationship with them. An index provider doesn't want to necessarily create indexes just for indexers. There are other purposes for indexes, so they don't want to muddy the water with just what an indexer prefers, and that's fine with us.

We do communicate with index providers constantly, but we're not telling them how we think they should maintain the index. This was a reason why I was compelled to write my article. We did want to get our ideas out in the public domain without sitting on index advisory councils. We wanted to maintain our objectivity, but still perhaps stimulate some thought in the industry.

Q: Vanguard has taken a unique approach by offering exchange-traded funds as separate share classes of its existing funds. Was this done to keep market timers out of the index funds, and how will investors in those funds benefit from the structure?

A: The original idea for the VIPER [Vanguard Index Participation Equity Receipts] share class was developed in the beginning of 1998. At that time, a lot of money was going into the various existing index funds. We were worried that this money was chasing past performance, and that if performance turned around it could leave as fast as it came in. We thought that investors who might be inclined to leave might be attracted to the VIPER class because it gives them more trading flexibility. And that would actually insulate the investors in our more conventional class of index fund shares from any negative activity that departing investors or short-time horizon investors might create in the fund.

The VIPERs had to be another class of shares of our existing funds. Otherwise, we couldn't have enabled these investors to migrate over to VIPERs. Later we started recognizing there were other advantages that we didn't originally anticipate. By adding a VIPER class to the existing class, when we have redemptions we can redeem our lowest-cost shares out of the fund and enhance its tax efficiency.

At the same time, ETF shares benefit from the conventional class of shares that have a natural cash flow using real cash. Since everything is done in-kind in ETFs, there is no cash inside the ETF to rebalance as the target index changes. In an ETF, you have to sell off a slice of every stock in order to buy a new stock being added to the index, for example. In our structure, we have cash flow coming through the conventional class shares, and we can divert that cash to buy the stock that's being added to the index. We don't have to sell off a slice of all the other stocks. So portfolio management is greatly enhanced by this structure.

Q: How is managing an index fund really an "active," or maybe more accurately "proactive," process?

A: Well, we're still compared to monkeys. But seriously, your question is right on. In an index fund, there's a lot of blocking and tackling - it's definitely a full-contact sport. If you aren't constantly managing the fund, it's going to wander away from the index . . . or I should say the index will wander away from the fund. Indexes are changed more frequently than people might realize.

The shares outstanding of a company might be adjusted on any given day, and those changes have to be incorporated in the portfolio on the same day. At the same time, you've got constant cash flow coming into the fund or leaving it. That cash flow has to be transacted in an efficient manner. We're very conscious of transaction costs, and we're quick adopters of new trading techniques.

Q: Vanguard has managed net inflows in the face of a bear market. Do you think modest outlooks for equities moving forward, and therefore an emphasis on low costs, has had anything to do with this?

A: In a market that's providing in excess of 20 percent returns, do you really care if one or two percent is siphoned off to pay a mutual fund family or pay for transaction costs? So the issue of costs gets lost. In a low-return environment, the difference between a net 4% return and a net 5% return can be quite significant.

We always felt our strongest suit was in a low-return environment. I think we're continuing to see strong cash flows for two reasons. The primary reason is that we're known as a low-cost provider, and people are acutely aware of costs at this point in time. The second reason is that we attract a different type of investor, one that has a longer time horizon than most. Our redemption ratio is half that of the industry.