Quant Investing


Factor based investing has become quite popular these days. Factors are characteristics of a group of stocks, the most famous being value and small cap, that are used to sort the overall universe of stocks. For quite some time certain factors have been shown to outperform the overall market over extended periods of time. The finance industry has jumped all over this and now offers many off the shelf funds and ETFs that aim to invest in these factors and outperform the market. There are about 400 Smart Beta funds now , totaling about $400B in assets. No need to do the work yourself. Just buy these simple off the shelf products and outperform the market. If only it were that simple. Today I want to briefly touch on a fundamental issue with smart beta funds, how it impacts returns, and compare smart beta to doing factor investing yourself through quant strategies.

The big issue with smart beta ETFs is the structuring of the product for scale. Patrick O’Shaughnessy at Investor Field Guide has a must read post on this topic. Basically, in order to make money off smart beta products, especially in the trend of lower fees across the industry, the finance industry needs to be able to scale these products to manage large amounts of dollars. The problem is that reduces returns. How much? Here’s the key chart from the post.

Screen Shot 2016-04-14 at 12.28.48 PM

Basically, the more stocks in the portfolio, the more cap weighting in the portfolio, the lower the returns. So much so that pretty soon the fund is just a closet index fund. There are lot more gems in the post but this is main point. If you’ve read my blog for a bit you know that there is a better way; doing your own factor investing through the quant strategies that I talk about here often. All of the strategies I’ve posted on are just ways to get exposure to the main factors that lead to market beating returns. Let’s see how one of those strategies does with various numbers of stocks and versus smart beta.

I used the Value 2 composite (VC2) quant strategy for this test but with one small change. I limited to stock universe to the SP500 stocks vs the All Stock universe to get an apples to apples comparison. Results are better with the All Stock universe. I then varied the number of holdings from the basic 25 to 50 to 100. I also compared the results to the index, SPY, and the SP500 pure value index which is basically a smart beta strategy. Backtests were run from the beginning of 1999 through yesterday. Below are the results.

Screen Shot 2016-04-14 at 12.48.07 PM

Smart beta is pretty darn good. The SP500 value index outperformed the SP500 by about 1.5% over the last 17.25 years. But that is not even close to what a better more concentrated value approach achieved. Even a 100 stock VC2 strategy almost doubled the performance of the smart beta approach. And results go up for lower numbers of stocks, not to mention feasibility (fees, slippage, etc…) and ease of implementation. Almost of 600 basis points of alpha over a smart beta approach and 730 basis points over the index is definitely worth a little effort I think.

In summary, quant stock investing far outperforms not only indexes but also smart beta strategies. Of course, the doesn’t mean it’s easy but it is certainly worth considering. In a future post I’ll talk about why investors should expect the outperformance of all factors and strategies to decrease in the future.



Tagged , , , , ,

6 thoughts on “ Quant stock investing vs smart beta strategies

  1. To be more fair, you really need to compare it to the SP500 equal weighted index, right? It’s still great outperformance at ~6.7% on the equal weighted.

    1. Sure. Equal weighting is a source of alpha. But the larger point in O’Shaughnessy’s piece is that Smart Beta funds kind of have to be market weighted to attract enough assets to make them investable and to keep their costs down enough to offer competitive fees.

  2. Paul,

    Great info! … more reason to consider the various Quant Portfolios you cover that use 25 holdings.

    In all your quant portfolios you detail the basic value criteria you are using (i.e. value composite2, shareholder yield, etc) and sometimes industry criteria for certain portfolios (i.e. Utility and Consumer Staples).

    However, I wanted to know if you are adding any other search criteria such as market not equal to Over the Counter or price >1 or share volume criteria, etc?

    I’m trying to do my screens in AAII Stock Investor Pro (still haven’t sprung for the portfolio 123 yet). I don’t have an opinion on this “extra” criteria either way, but am curious to what you are doing and why.


    1. B,

      The only filters I use on the stock universes, besides market cap or sectors for the relevant screens, is NO OTC stocks. This is the O’Shauhnessy methodology. For most screens I’ve found that the O’Shaughnessy criteria are goof enough and I don’t need to filter for volume or stock price. The only exception would be the microcap strategy. I’ve don’t use that strategy anymore but when I did I did filter for volume and price even though that reduces returns quite a bit.


  3. Hi Paul,

    Thanks for sharing your thoughts on the blog!

    I am curious which site(s) you use as a stock screener to build these quant portfolios?

Comments are closed.