Designing and managing asset allocation strategies based on factors is promoted in some corners as a better way to build portfolios. Not surprisingly, there’s no shortage of studies that support this view. But the jury’s still out on whether it’s prudent to throw out the standard asset-class buckets. Factor-based investing can play a productive role in enhancing a conventionally designed asset allocation, but it’s debatable if a pure factor-only strategy is a true solution.

There are two problems to consider. First, real-world factor investing has a short history. There are exceptions, such as small-cap, value, and momentum strategies for the US equity market. Otherwise, the track record is short, at least in terms of mutual funds and ETFs that represent proxies for putting theory into practice. As I discussed recently, you can slice up the US stock market into eight different factor exposures via ETFs, but the common start date for this lot only stretches back to July 2013.

Four years is hardly enough time to pass judgment, but let’s ignore that glitch for now and create 1000 randomly allocated portfolios using the following eight ETFs:

* iShares Edge MSCI Min Vol USA (USMV) – low-volatility
* Vanguard High Dividend Yield ETF (VYM) – high-dividend yields
* Guggenheim S&P 500 Equal Weight ETF (RSP) – small-cap bias within large-cap space
* iShares Edge MSCI USA Quality Factor (QUAL) – so-called quality stocks
* iShares Edge MSCI USA Momentum Factor (MTUM) – price momentum
* iShares S&P Small-Cap 600 Value (IJS) – small-cap value stocks
* iShares S&P Mid-Cap 400 Value (IJJ) – mid-cap value stocks
* iShares S&P 500 Value (IVE) – large-cap value stocks

The benchmark for comparison: the S&P 500 Index, based on the SPDR S&P 500 (SPY). The question before the house: How do 1000 randomly designed factor strategies (using the eight ETFs above) compare? For some insight, I fired up R to generate 1000 wealth indexes based on randomly changing the initial allocations to the eight funds. Any one fund could have a weight in the portfolio ranging from 0% to 100%. Most of the weights were a mix across the funds. All the strategies were rebalanced at the end of each calendar year to the initial randomly generated weights. The results are shown in the graph below.

The main takeaway: besting the S&P 500 on a buy-and-hold basis hasn’t been easy. A $100 investment in SPY on July 19, 2013 grew to just under $159 after four years (through July 25, 2017). By comparison, the median portfolio for the random strategies was worth slightly less, at just under $158. The worst performer increased to a bit more than $152; the top performer grew to nearly $166.

In other words, the random portfolios more or less hugged the S&P’s performance. Breaking free of market-cap-based equity beta is no mean feat. It can be done, but the eight ETFs listed above may not be the ideal mix of products for earning excess risk premia while minimizing the stock market’s general influence. That’s also a clue for thinking that a rough patch for equities overall will also take a similar bite out of most factor-based strategies via the eight funds above.

Note, too, that the results are also comparable when measured in volatility risk (standard deviation of daily returns). SPY’s annualized volatility over the sample period: around 12%. That’s roughly equivalent to the median vol of the random strategies, which had a range of risk profiles from 11% to 13%.

These results don’t mean much, of course, since the analysis only reflects four years of data. Meantime, the S&P has enjoyed an unusually strong performance run in recent years, which suggests that recent history isn’t a reliable guide for managing expectations for a buy-and-hold portfolio for US stocks.

Let’s also recognize that building a factor-based portfolio has merits beyond performance chasing. Rebalancing and risk-management opportunities may be richer with factors vs. standard asset class buckets.

In short, there’s more work to be done for profiling a real-world test of factor investing beyond the standard results, such as demonstrating that value beats the market over the long haul. Nonetheless, our toy backtest serves as a reminder that leaving Mr. Market in the dust by holding a broader set of factor funds may be tougher than the academic research implies.