Credit Where Credit’s Due

February 13, 2006

The strong returns for PowerShares’ flagship fund has indexing traditionalists scurrying to explain away the performance. But their accusations don’t hold up.

The strong performance of the PowerShares Dynamic Market Portfolio (PWC) since its launch in 2003 is making the traditional indexing universe increasingly uncomfortable. The fund, which uses a quantitative strategy in an attempt to beat the market, delivered nearly 14% returns to investors in 2005, easily outpacing the 4.9% return on the S&P 500 Index. That helped it gain momentum on the assets front, and the fund now boasts nearly $800 million in assets. 

In part because of this strong performance, AMVESCAP ponied up big bucks last month to buy out PowerShares—a move that highlighted the increasing shift of investor interest toward "enhancing" or "active" indexing products. Now, the indexing traditionalists are on the offensive, arguing that the outperformance will be short-lived.

As a PWC shareholder, I have a very real interest in finding out who's right. So I decided to take a look under the hood of the fund—and the fund's detractors—to see where that performance was coming from.

A Brief History Of PWC

When the fund launched in May of 2003, PowerShares had backtested data on the underlying Intellidex Index (from the American Stock Exchange) showing that the fund would outperform the broad market. But backtested data is notably unreliable, and it was easy enough to dismiss that theoretical performance as the benefit of 20/20 hindsight.

In the first year the product was on the market (Yahoo! has data stretching from October 07, 2003, and that is the period used in this analysis, although the fund officially launched on May 1, 2003), PWC handily outperformed a comparable exchange-traded fund (ETF) tracking the S&P 500 Index (the Standard and Poor's Depository Receipts 1, or SPY), delivering 16.5% returns against 10.5% returns for SPY.[1] 

Ahh, but one year is such a short time, the critics said. Anyone can get lucky.

Now, two and a half years into the experiment, the critics are starting to worry. Through February 10, PWC has delivered cumulative returns of 48.1%, against just 26.6% for SPY. The outperformance is getting increasingly hard to ignore.

The latest wave of explanations comes in this Marketwatch article by John Spence, in which a variety of observers suggest that the startling performance stems from the fund's small/value tilt.

"Managers of ETFs tied to established indexes say the new instruments [like PWC] outperform in back-tested scenarios because they're biased toward small-cap and value stocks -- the market's leaders for several years," writes Spence. "If large-cap and growth shares regain favor, they contend, their funds will outperform the new approaches."

Spence then finds a number of experts who agree, including Gus Sauter, chief investment officer at Vanguard, and Jim Wiandt (publisher of this site), who says:

"They [PowerShares] were in the right place at the right time, and the market cooperated."

Is that all it is?

A Look At The Data

I turned to the data.

The capitalization bit is easy. Wiandt and Sauter are right: The PowerShares fund does have a small tilt compared with the S&P 500 Index. To use brute numbers, the average market capitalization of the PowerShares fund is $23.1 billion, compared with $84.1 billion for the S&P 500.

But that's not news. Although PWC uses the S&P 500 as a benchmark (a mistake—it should use an all-market bogey), the product literature makes it clear that it's an all-cap fund. As of September 30, 2005, the fund was 72.8% exposed to large caps, 14.4% exposed to mid-caps and 12.8% exposed to small caps. That's a particularly large exposure to small caps, which make up just 3 +/- percent of the U.S. equity market.

Could this explain away the performance? One way to test that idea would be to create a customized benchmark portfolio using the same 72.8%/14.4%/12.8% capitalization breakdown shared by the PowerShares fund. The comparison won't be exact, because the PowerShares' breakdown changes over time, but it should be good enough to get the gist.

To do that, I composed a portfolio using three ETFs: the SPY, the Mid-Cap SPDRs (MDY, which tracks the S&P MidCap 400 Index) and the iShares S&P SmallCap 600 (IJR). I weighted these indexes on the 72.8%/14.4%/12.8% scale, and ran the performance using data from Yahoo! Finance stretching back to October 7 (the last date for which PWC performance data is available).

Through February 13, this customized benchmark would have returned 32.3%, cumulatively. That's better than the S&P 500, but still substantially less than the 48.1% return of PWC. For reference, MDY returned 45.7% over the same time frame, while IJR returned 49.6%. In other words, the all-cap PWC would have beaten the mid-cap fund, and just lagged the small-cap ETF—in a period of small and mid-cap outperformance.

Clearly, size alone can't explain PWC's performance.

What About Value?

The indexing traditionalists reach for the value explanation because value has been beating the pants off of growth ever since the Internet bubble popped in 2000. This has created real differences in performance in the market, and funds that follow a value bent—like the index funds from Dimensional Fund Advisors—have done well.

The trouble is, when it comes to PWC, the value accusation just doesn't stand up.

While there are a thousand and one ways to separate growth and value stocks, most come down to two core variables: price/earnings ratio and price/book ratio. According to its literature, at year-end 2005, PWC boasted a price/earnings ratio of 18.69—almost a dead ringer for the S&P 500, which sported a price/earnings ratio of 18.70. For comparison, the P/E ratio for the S&P MidCap 400 was 23.54, and the price/earnings ratio of the S&P SmallCap Index was 21.22.

The price/book ratio of PWC at year-end 2005 was 5.56. I couldn't find the price/book ratio for the S&P 500 as of year-end, but as of November 30, it was 2.85. That number couldn't have changed too much in the last month of the year. In other words, the price/book ratio for PWC was much higher—much "growthier"—than the price/book ratio of the S&P 500. For comparison, the price/book ratio of the MidCap 400 was 2.62, and it was 2.37 for the S&P SmallCap 600. 

Here is how PowerShares breaks down the exposure of PWC in its own literature:

Large-Cap Growth



Large-Cap Value



Mid-Cap Growth



Mid-Cap Value



Small-Cap Growth



Small-Cap Value



If you do the math, the fund is 49.8% exposed to growth … and 50.2% exposed to value.  It's hard to find much of a tilt in there.

So What Are The Differences?

If there is no style tilt, and if the size tilt can't account entirely for the performance, how do you explain the strong outperformance of the PowerShares fund? The two real possibilities are either sector exposure and individual holdings.

The sector exposure is easy enough to compare:



















Consumer Services



Business Services



Financial Services



Consumer Goods



Industrial Materials









Data from Morningstar

There are some notable differences in the sector weighting, particularly in the Business Services segment, where PWC has a heavy weighting (11.91%) and SPY has a tiny weight (3.84%). The Business Services segment has performed well over the past five years, so this could contribute to PowerShares' performance.

It's a bit more difficult to gauge the impact of individual stock selection, as the components of the PowerShares fund change on a regular basis. One thing to note is that PWC is considerably more concentrated, with 26.4% exposure to its top 10 holdings, compared with 20% for SPY. What's more, only one of those holdings overlap—Proctor & Gamble. That leads me to believe that individual stock selection plays a major role in the performance.


Because the American Stock Exchange and PowerShares zealously guard the inner workings of PWC and related funds, it's impossible to suss out why the funds have been performing so well. But it's easy enough to eliminate some of the variables, such as style, and to measure the impact of other variables, such as size.

It's clear that these two factors alone can't explain away the performance of PWC; something else is working. And while the inner indexer in me keeps chanting "reversion to the mean," so far, you just can't argue with the performance.


[1]I choose to use the SPY ETF rather than the S&P Index itself to create a better comparison, as the SPY performance—like PWC—includes expense ratios.

Find your next ETF

Reset All