Are zero-fee funds truly “free”?

Is cheaper always better?

Investment services are becoming increasingly cheaper. Not just cheaper, but actually free! On August 1st 2018, Fidelity Investments — with over $7 trillion in assets — was the first investment company to launch no-fee Exchange Traded Funds (ETFs). On the same day, north of the border in Canada, Horizons ETFs mirrored Fidelity’s announcement and offered two zero-fee index funds of its own.

Competition on management fees is not surprising. After all, passive asset managers (such as index funds) generate virtually identical returns as they track the market. Indeed, ETF average management fees dropped by almost one third over the past decade, from 0.31% in 2008 to 0.22% in 2017. Part of the trend is due to better technology: operating costs dropped as funds embraced automation and digital marketing. However, it is still not free to run a fund — salaries and legal fees, at the very least, need to be paid.

So, if the fund still incurs some operating expenses, how can a management fee of zero be sustainable in the long run? One answer is that funds can earn additional revenue from lending out stocks in their portfolios. This revenue allows to offset, and in some cases more than offset, the cost of running the fund (Blocher and Whaley, 2016). Forbes also identifies a number of 16 (particularly small-cap) ETFs with security lending revenues that exceed operating costs. A big question, however, still remains.

Do investors benefit the most when proceeds from security lending are used to finance zero-fee funds?

Maybe not. In a new research paper, I find that investors would be actually better off if funds would pay any revenue from security lending as dividends (like this marijuana ETF did).

The key argument is that funds do not lend out stocks at random times. Traders want to borrow a stock, among other reasons, when they expect its price to fall in the short run. Hopefully, the borrower can profit by selling the security now and buying it back cheaper later. This means that lending securities generates extra return for the fund exactly at those times when the market performance is likely to be disappointing. Therefore, lending revenues can provide a hedge against market downturns and reduce portfolio risk.

Since ETFs carry the risk on behalf of investors, they do not fully internalize the hedging benefit. Competitive ETFs can charge discounted fees (indeed even zero fees), allowing investors to receive the entire expected revenue from security lending when they put money in the fund. However, it is the timing of payments that matters. Investors optimally receive lending revenues as dividends on an on-going basis: most likely, dividends will be higher when market returns are low. Free may be indeed a powerful word, but free funds are second-best in this context.

Zero-fee funds may be optimal, however, in an non-transparent environment. If investors can not fully observe how much revenues ETFs obtain from lending out securities, and cannot easily figure out when and how large the dividend should be, they are better off asking for a deep discount on management fees. After all, fees are (more) transparent.

At a zero-fee lower bound, technology can come to the rescue. A decrease in operating costs (as driven for example by digitization) gives funds the incentive to increase cash inflows and the size of their balance sheet, rather than maximize profit margins per dollar under management. To boost inflows, funds can improve revenue transparency, educate investors, and share a larger portion of lending revenues of dividends. Does this sound familiar? Indeed, the business model described comes close to what cost-efficient “robo-advisors” seem to be doing: competing primarily to attract new investors to the market by offering low-cost, transparent products and financial education tips.

Paper is available for download [here].