Investor Loss Aversion
That investors hate losses more than they love gains is one of the most stable findings in behavioral finance. Research consistently shows that investors are more sensitive to losses than gains, a phenomenon known as loss aversion (Hwang 2010, Hwang 2003, Bordley 2017, Artavanis 2020). This is particularly pronounced during bull markets, like the current one, when investors become even more loss-averse.
Wall Street always loves a costless (for them) arbitrage, and investor loss aversion is a perennial favorite, from the days of portfolio insurance, to option strategies, and more. The latest variant is so-called "buffer" ETFs, which are seeing huge cash inflows—$46 billion since 2018— and now generating ~$430m in fees. They are driven by the promise of guaranteed return of your invested capital at one year, while still giving you exposure to equity markets.
There is no free lunch, of course. These ETFs cap returns at some number—around 10%, in the case of a new Blackrock variant—and charge much higher fees than broad-market ETFs from Vanguard, et al, on the order of 0.5% vs 0.05%. And you must hold for the entire one-year period to obtain the promised return of capital. These are significant restrictions, even if they may not appear that way.
Simulating Capped Return ETFs
To better show this, I did 10,000 simulations of a capped returns ETF against the market. The market is modeled with an average 11% annual return and 15% standard deviation. The hypothetical buffer ETF imposes a 0.5% fee, caps returns at 10%, and has a 0% floor. I modeled the market with a lognormal distribution, which, while imperfect, is a decent approximation.
The ETF has some appealing features. For example, it avoids all market crashes and drawdowns, which are lethal for investors, especially as they approach retirement age. In 5% of years the 10,000-year simulation had market returns under -15%, and three years with whopping 50%+ market crashes, perhaps corresponding to meteor strikes. With a capped ETF you avoided all of those.
Of course, you also "avoided" all years with more than 10% gains, and there were a lot of those. In the simulation, market returns exceeded 10% 52% of the time. This had implications, because, despite avoiding losses, and some crushing ones, you still only outperformed the market around one-quarter of the time, which you can see in the comparative mean returns of 6.2% for the capped ETF vs 11.6% for the broader market.
Simulation Results
- ETF Mean Return: 6.2%
- ETF Standard Deviation: 3.1%
- Market Mean Return: 11.6%
- Market Standard Deviation: 15.2%
Turning it around, it's worth comparing this buffer ETF to CDs and other less volatile instruments with higher yields. You can currently get 5%+ on such vehicles, so with a buffer ETF you are, in effect, getting 1.2% over that guaranteed yield. But it's not guaranteed, even if it has a lower standard deviation than the market, as shown above. Put differently, you are paying a sizable risk premium for a small return increment over low-risk CDs.
Conclusion
A simple buffer ETF underperforms the market 82% of the time, which is material. It charges a sizable risk premium over a 5% CD, making it not compare well. For most investors, direct market exposure or even a CD offers better return prospects at lower costs. The ETF’s lower volatility may appeal to extremely risk-averse investors, but the trade-off in returns does not justify its inclusion in most portfolios.
Buffer ETFs are, however, a terrific new fee-generation vehicle for ETF sponsors, now producing nearly a half billion dollars in revenues. Market may crash, and stocks go out of style, but monetizing investor biases will never go out of style.
Papers Cited
Nikolaos Artavanis, and Asli Eksi. (2020). Reference-Dependent Preferences and Mutual Fund Flows. Social Science Research Network. https://doi.org/10.2139/ssrn.3434779. Investors place a greater emphasis on losses than gains when they evaluate performance.
Bordley, R., and L. Tibiletti. (2017). Benchmark-Based Preferences Make Investors Loss Averse in Bull Markets and Gain Seeking in Bear Markets. https://doi.org/10.5539/ijbm.v13n1p46. Investors are more likely to be loss averse during bull markets than during bear ones.
Hwang, Soosung, and S. Satchell. (2003). The Magnitude of Loss Aversion Parameters in Financial Markets. https://doi.org/10.2139/ssrn.383942. Investors are more sensitive to changes in losses than changes in gains.
Hwang, Soosung, and S. Satchell. (2010). How Loss Averse are Investors in Financial Markets? https://doi.org/10.2139/ssrn.383942. Investors are more sensitive to changes in losses than changes in gains.