Exchange-traded funds (ETFs) have seen enormous growth in assets under management over the past two decades. And trading frequency has kept pace. From July 2014 to July 2015, $18 trillion of ETF shares exchanged hands (to put things into perspective, the U.S. GDP stands around $17 trillion). Considering that the total net assets of ETFs as of year-end 2014 was around $2 trillion, the average turnover of shares in ETFs was more than eight fold.

ETFs provide a convenient and cost-effective way to gain exposure to both broad market indexes and specific sectors. However, some ETFs are more efficient than others. One of the most important ETF distinctions is the active versus passive ETF classification; passive ETFs track a market-weighted index, whereas active ETFs allow managers to deviate from the underlying benchmark with the goal of outperforming the index.

But even that classification gets more complicated; just because something is called a “passive ETF” does not necessarily mean it is best for an investor who subscribes to a passive investing philosophy.

For example, some active managers use passively managed sector ETFs to make sector bets. Although sector index ETFs are themselves technically passive ETFs, they are typically used for active portfolio management. Achieving a broad passive allocation using a set of highly concentrated ETFs (e.g., biotech ETF) is a common “wolf in sheep’s clothing” approach to asset management, also known as “fund stuffing.” Sector ETFs not only rack up more fees when it comes to expense ratio, internal fund turnover, and bid ask spread, but they also require more work when it comes to rebalancing.

The Spectrum of ETF Asset Coverage

All ETFs sit on a spectrum in terms of coverage, or how many different assets they include. On one end are broad, market-capitalization-weighted ETFs that passively replicate an index with thousands of constituents. An example is Vanguard’s total market ETF, VTI, which tracks the entire U.S. stock market.

On the other end of the spectrum are highly concentrated and focused ETFs that cover niche markets. One example is the Fidelity MSCI Consumer Staples ETF, FSTA, which specifically tracks the consumer discretionary sector. Whereas VTI represents 100% of the investable companies in the U.S. equity market and has 3,886 stocks, FSTA has only 386 constituents, or about 10% of the stocks in Vanguard’s total market ETF. While FSTA may be classified as a “passively managed” ETF because it tracks an index, a sector ETF is rarely the ideal candidate for a passive broad market tracking portfolio.

An active investment manager may decide that rather than holding the inexpensive VTI, which doesn’t allow her to underweight or zero-out exposure to market slices she doesn’t like, she’ll build out a portfolio using sector ETFs. This allows her to tactically shift allocations based on her subjective views.

Of course, for sector ETFs to be chosen, they must make up for the additional costs incurred. For example, U.S. sector ETFs have a median expense ratio of 0.45%, about nine times the expense ratio of Vanguard’s total stock market ETF, VTI which costs 0.05%.

U.S. Equity Sector Expense Ratios Compared to VTI

sector-etf-expense-ratio-02

So right off the bat, the portfolio manager starts with a guaranteed 40bps headwind.

ETF Turnover Ratio

While costs such as the expense ratio can be easily found, other costs, such turnover ratio are more difficult to discern. Unless you pore over an ETF’s prospectus, the turnover cost is usually hidden. Even if you’re able to find historical turnover, current turnover may deviate significantly from historical data as turnover varies from year to year. While fund providers can attempt to estimate the turnover cost for that year, they may not be able to do so accurately considering that turnover may change substantially in volatile markets.

Turnover costs are seen both at the ETF level and the portfolio level. Typically, index funds have lower turnover rates than actively managed funds, and this is true at both the fund and portfolio levels. At the fund level, funds that track broad indexes such as the Wilshire 5000, which has more than 5,000 companies as constituents, will see less turnover than a sector ETF because of its stricter constraints and the higher hurdle required to meet each of the constraints.  

For comparison, VTI, the Vanguard total market ETF, has a turnover of 3% while the median turnover ratio of U.S. equity ETFs is around 20.5%. Sector ETFs, which by nature are more concentrated, are associated with a higher level of turnover than a typical broad market tracking ETF. For some narrow ETFs that track niche markets, the fund turnover is especially high. NASH, or the Nashville Area ETF, for example, targets companies specifically headquartered in the Nashville, Tennessee region and has a fund turnover of 38%.

U.S. Equity Sector Turnover Ratios Compared to VTI

sector-etf-turnover-ratio-02

In contrast to fund turnover which is the result of selling and buying the underlying securities in a fund, portfolio turnover is the result of an ETF changing hands frequently. Each time the ETF is bought and sold, a transaction cost is incurred. Turnover includes brokerage commission costs but also bid ask spreads. On average, more actively traded ETFs see higher turnovers at the portfolio level.

One source of portfolio turnover seen with portfolios designed to track a market cap weighted broad market but constructed using sector ETFs is the need for regular rebalancing. Over time, the value of each individual ETF in a diversified portfolio drifts from their target weights. For example, a random sector may outperform, resulting in a greater allocation than its original target weights. As a result, the allocation to consumer discretionary may need to be reduced to maintain the target weights. This rebalancing requires portfolio turnover to increase.

Here at Betterment, we place a great deal of emphasis on the importance of diversification. Holding a broadly diversified portfolio is essential to sound portfolio management. For this purpose, we rely on market cap-weighted, broad-market-tracking ETFs. While we do ascribe to value investing, all of our ETFs are passive ETFs that benchmark to broad-market-tracking indexes.We don’t use sector ETFs to construct our portfolios because not only are they more expensive to use, but they are also designed to provide a concentrated exposure to a very specific niche of the market.

If you subscribe to passive investing, sector ETFs may not be the most ideal investment choice. While it may be possible to construct a broadly diversified portfolio from sector ETFs, the extra cost in terms of expense ratio, turnover, and extra work required when it comes to rebalancing make sector ETFs inferior instruments for building broad market index tracking portfolios.

This article originally appeared on ETF.com.

More from Betterment: