One of the many key features of ETFs is that they provide significant diversification benefits; prior to the days of ETFs, investors had to rely on either expensive money managers or their own stock cherry-picking methodologies. Though exchange-traded funds give investors cheaper and more efficient means of portfolio diversification, a closer look under the hood of some of these products reveal several noteworthy biases. Below, we highlight what investors should be looking for to find out what really is driving their ETF

In the case of ETFs that have a relatively small number of holdings, investors should always look to see how much is allocated to the top security. Sometimes, a single stock makes up roughly a quarter of the fund’s total portfolio, meaning the performance of the ETF is heavily dependent on a single security. While this is not necessarily a bad thing, investors should be comfortable with the risk/return profile of such a top-heavy ETF.

Below are two examples of ETFs that have heavy allocations to a single stock. Please note that returns data is based on monthly adjusted prices from December 2012 to January 2014:

This chart features the performance of the Biotechnology ETF (BBH B-) versus its top holding Giliead Sciences (GILD). The fund allocates more than 14% to the stock (as of 4/22/2015). In this example, BBH benefited from a heavy allocation towards GILD, gaining nearly 80%.

Unlike the BBH example, this chart depicts how a single stock can actually negatively affect bottom line returns. The MSCI Global Silver Miners ETF (SLVP B+) allocates roughly 24% to Silver Wheaton Corporation (SLW), and as seen above, the performance of the stock and ETF essentially mimic each other.

While single-stock biases are sometimes easier to spot, investors should also be on the look out for those “broad-based” funds that feature a significant bias towards a handful of companies. Often, when investors see that an ETF has a relatively large amount of individual holdings, they assume that there is a reasonable amount of diversification. A closer look, however, reveals that 2-5 individual securities make up a significant portion of the fund’s total assets.

Here is an example of an ETF that features a heavy bias towards three stocks:

The Consumer Staples Select Sector SPDR ETF (XLP A) holds roughly 40 individual holdings (almost double the holdings of BBH). The top three holdings—Procter & Gamble (PG), Coca-Cola (KO), and Philip Morris (PM)—together account for more than one-third of XLP’s total assets. The chart shows how these three securities weighed heavily on XLP, with the ETF gaining only 23%. Though a 23% gain is by no means insignificant, comparing the fund to an equal-weighted ETF paints a slightly different picture:

Here we see the performance of XLP compared to the equally weighted Consumer Staples ETF (RHS A). Over the same period, the equally-weighted fund managed to outperform XLP by roughly 8%. Investors should note that this example does not mean alternative weighting methodologies always outperform traditional market-cap weighted ETFs, but that they do sometimes help mitigate the risk associated with single or multi-stock biases.

As is the case seen in multi-stock biases, sometimes “broad-based” funds are heavily skewed towards a particular industry or sub-sector of the market. Take for example the China Large Cap ETF (FXI A-), which is designed to measure the performance of the largest companies in the China equity market. Though the fund does offer exposure to a wide array of sectors, it invests more than half of its assets in financial services. Tech stocks, however, are given an allocation of less than 10%.

Throughout 2013, Chinese equities struggled to stay out of the red, but as the chart above shows, one particular corner of that market managed to deliver stellar returns – Chinese tech firms. Because FXI has a relatively small allocation to the sector, it, along with Chinese financials (CHIX C), lost almost 10%, while the China Technology Fund (QQQC C) gained over 60% during the same time frame.

For those investors looking to add exposure to foreign equities, ETFs are by far the most efficient means to tap into markets overseas. Investors have a variety of options when it comes to foreign exposure, including global, region-specific, and single-country ETFs. A closer look at some of the geographically “broader” ETFs sometimes reveals significant biases towards a particular country or region.

The EAFE Index ETF (EFA A) is designed to measure the performance of equity markets in European, Australasian, and Far Eastern markets. Its portfolio, however, allocates more than half of its total assets towards European equities. Looking at the chart above, the performance of EFA and the Europe ETF (VGK A) very closely mimic each other. An even closer look under the hood of EFA also reveals that Japanese equities account for roughly 20% of total assets, but because European stocks dominate the portfolio, EFA is not able to capture all of the stellar returns from Japanese equities.

Though ETFs certainly do offer key diversification benefits to investors, it is important to take a closer look under the hood of these products. ETF biases can occur in a number of ways, including heavy allocations to a particular stock, sector, country, or region. Depending on an investor’s risk/return profile, these biases are not necessarily bad, but the realization that a particular bias can essentially drive an ETF is crucial.