This article is part of a regular series of thought leadership pieces from some of the more influential ETF strategists in the money management industry. Today's article features Rusty Vanneman, CFA, CMT, chief investment officer of Omaha, Nebraska-based CLS Investments.
Recent surveys have shown that investors consider a variety of criteria when shopping for an ETF. But the most popular consideration—cost—is often overemphasized, while the most important is overshadowed.
When selecting an ETF, or any investment for that matter, the most important criterion should be desired market exposure. Investors should consider what exposure is needed for the portfolio and what is appropriate for the overall allocation. They should also consider the expected risk and return, and how the ETF will mix with other existing holdings.
Desired market exposure, however, is typically far down on many investors’ lists of key selection criteria. At the top, instead, we often find cost as a top consideration.
In reality, only once market exposure is decided, should other factors be considered, including low costs. Cost is overemphasized by many investors to the potential detriment of their portfolios.
For example, let’s say two market exposures are approximately the same, but one has lower costs and a less desirable expected return/risk profile. Many investors would select that ETF simply because it is cheaper, even though it is inferior as an overall investment.
Let me stress that costs are indeed important; I’m not arguing against that. At CLS Investments, we factor costs into our expected return calculations.
There are several components of cost to consider. The expense ratio is the annual cost for the ETF provider to manage the fund. The commissions and bid/ask spreads are costs associated with the fund being traded on an exchange. Commission refers to the broker’s transaction charges, and bid/ask spread refers to the spread between the highest price a buyer will buy for and the lowest price a seller will sell for.
If an ETF were bid (the highest price a bidder will buy for) at $100 per share and the ask (the lowest price a seller will sell for) were $101, then the bid/ask in dollar terms would be $1 (a wider-than-average spread). To convert that bid/ask spread into percentage terms, like an expense ratio, take the spread ($1) and divide by the bid/ask midpoint ($100.50). That comes out to nearly 1% (or just under 100 basis points).
Trading costs are also important to consider. The average bid/ask spread for all ETFs as of mid-August 2018 is approximately 25 bps. This compares to the average ETF expense ratio of just over 50 bps (or one-half of 1%).
Why Exposure Trumps Cost
Costs are, and should be, the deciding factor when two ETFs with identical market exposures are being considered. But market exposure with its anticipated return/risk characteristics trumps the relative costs. Let’s consider the data.
At CLS Investments, we recently reviewed all ETFs with three-year total return track records as of mid-August and examined their cost data: the latest average bid/ask spread, the current expense ratio and the sum of the two. CLS analyst Dustin Dorhout did the digging, and he helped us find some interesting trends.
First, we found a slightly negative correlation between three-year total returns and total costs for domestic equity ETFs and international equity ETFs. This means funds with lower costs had higher returns.
However, the correlation was low (below -0.2), and was, counterintuitively, slightly positive between three-year returns and expense ratios for international equity funds. In other words, on average, the more expensive international equity funds had higher returns.
When we reviewed top decile and top quintile funds, the correlations were basically turned upside down. In that case, both top-decile and top-quintile-performing funds in domestic and international equity funds had higher trading costs, expense ratios and total costs in their respective universes. Funds with higher costs and higher returns included smart-beta ETFs and sector-oriented funds.
Fixed income was also interesting. In this case, the correlations between costs and performance were positive, meaning that when looking at all fixed-income ETFs, the funds with higher costs had higher returns.
The top-decile and -quintile fixed-income funds also had the highest costs. This is probably because there are more actively managed ETFs that charge more and have lower trading volumes (typically meaning wider trading spreads), but that have added value versus their peers, at least historically.
In sum, costs do matter. They absolutely impact returns and are the most reliable factor when building an expected return for an ETF. That said, they are only one factor, and this factor is often simply overrun, and rightfully so, by the return and risk characteristics of the ETF.
This information is prepared for general information only. Information contained herein is derived from sources we believe to be reliable; however, we do not represent that this information is complete or accurate, and it should not be relied upon as such. All opinions expressed herein are subject to change without notice. 895-CLS-9/11/2018