ETF portfolios versus active fund portfolios - who wins?

justETF.com

Less than 20% of active fund managers beat their benchmark index according to the persistent findings of the SPIVA report. But the reality is that most portfolios are exposed to multiple benchmarks because they contain multiple funds. So how does the active vs passive debate stack up when you take it to the portfolio level?

ETF portfolios versus active fund portfolios - who wins? The landmark study of active vs passive portfolios (i.e. the type that include ETFs) was undertaken by Rick Ferri, the renowned US investment advisor, best-selling author and passive investing advocate.  

Ferri partnered with Alex Benke, product manager for US robo-advisor Betterment to answer the portfolio question using actual performance figures over a meaningful time period. 

Ferri and Benke documented their methodology in their 2013 white paper, A Case For Index Fund Portfolios. To ensure a realistic result, they used a survivorship bias free database. Survivorship bias occurs when the results of closed or merged mutual funds are wiped from historical results. Because large numbers of active funds are eliminated over time due to poor performance, survivorship bias makes past results look better than they really are, if closed and merged funds are not counted. The reality is that real-world investors don’t know which funds will close and are therefore exposed to their poor performance. Ferri accounted for this by using the respected Center for Research in Security Prices (CRSP) Survivor-Bias-Free US Mutual Fund Database, maintained by the University of Chicago Booth School of Business. 

Ferri investigated the chances of active portfolios outperforming under a range of conditions as outlined below. He uses US data but studies have repeatedly shown that US patterns usually hold true for other developed markets, too.   

 

Test 1: 3-fund passive portfolios vs 3-fund active portfolios

A portfolio of 3 index funds was compared against 5,000 possible active portfolios, each consisting of 3 randomly selected active funds drawn from the same asset classes. The asset classes were US equity, international equity and US investment-grade bonds, split 60:40 in favour of equities. If a fund closed or merged then it was replaced with another fund that was available at the time. 

The result: 83% of active portfolios underperformed the index fund portfolio. The median underperformance of the losing active portfolios was -1.25% per year over 16 years. The median outperformance for the 17% of active portfolios that won was 0.52%. Given that we can’t tell which funds will outperform in advance, the balance of risks tilts heavily against choosing an active portfolio. 

 

Test 2: Do passive portfolios win across all time horizons?

The passive portfolios were pitted against the active portfolios across 3 independent 5-year horizons and a complete 15-year period. 
 
The result: The passive portfolios were superior in every period, outperforming 66% to 86% of the time. Ferri’s conclusion: the longer your timescale, the more likely it is that a passive portfolio wins. 

 

Test 3: Do active fund portfolios outperform in niche markets?

Ferri tested the often repeated hypothesis that active fund managers add value in less efficient markets like emerging markets and small caps. In this scenario, 5-asset class and 10-asset class passive portfolios were lined up against equivalent active portfolios. The asset classes were equally weighted. 
 
The result: 88% of the 5-asset class active portfolios underperformed their passive rivals. 90% of the 10-asset class active portfolios underperformed. The median annual underperformance for 5 and 10 asset class active portfolios was -1.1% and -0.93% respectively. The median annual outperformance for the few active winners was 0.44% and 0.29% respectively. The results dispel the myth that active managers add value in niche markets. 

 

Test 4: What if active investors diversify their fund manager risk?

Many active investors typically hold multiple funds per asset class to offset the risk of bad bets by fund managers. Ferri assessed this strategy by ranging the passive portfolio versus active portfolios that used 1, 2 or 3 active funds per asset class. 

The result: Fund manager diversification is a bad strategy. The more active funds held by a portfolio, the worse the results:
 
1 active fund per asset class = 83% chance of outperformance by the passive portfolio. 
2 active funds per asset class = 87% chance of outperformance by the passive portfolio. 
3 active funds per asset class = 91% chance of outperformance by the passive portfolio. 

 

Prefer to choose only the best funds?

Some people cling on to the belief that they can outperform by selectively choosing the right funds or the right markets for their active bets. But the odds are stacked against them. The data from Ferri (and SPIVA) shows that active managers don’t succeed in difficult markets where skill is reputed to give them an edge. 

Sure, some people will get lucky in the active fund lottery but the marginal advantage gained hardly seems worth the risk of the much larger downside measured by Ferri. It’s better to play the odds. 

Besides, how do you predict which active managers will succeed in the future? There’s a reason why every mutual fund product must carry the warning: “Past performance is no guarantee of future results.” That’s because it’s true. The SPIVA report consistently reveals very few active funds maintain their top tier status over 5 or even 3 year periods. The effort and knowhow required to select and audit active managers, their company and investment approach remains the preserve of large institutional investors, and even they are regularly disappointed. 

 

ETFs have data on their side

Ferri's paper comes to the same conclusion as many other studies into mutual fund performance: most investors are better off with passive portfolios. The advantage only increases as you add funds, asset classes and time factors to the analysis because a passive portfolio is more than the sum of its parts.