Russell 3000 Anomaly: Doubles 2017 Index Returns - 10 New Selections For 2018

Includes: INNT, ZFGN
by: JD Henning

The 2017 Stock Anomaly selections returned 39.92% compared to 18.63% for the Russell 2000 benchmark index over the same period.

5 of the original 10 selections returned over 70% annually with 2 stocks (HCC and ABEO) gaining over 150%.

Two sampled stocks are provided from among the 10 stocks selected for the 2018 Russell 3000 anomaly portfolio reserved exclusively for subscribers.

Stock additions to the Russell 3000 index leads to a "dramatic increase" in trading volume ratio in the month of June (Chang et al., 2013).

Measurable price effects of the Russell 3000 annual index reconstitution have been documented for both stock additions and deletions (Chang, Hong, & Liskovich, 2013).


This article marks my second annual report and selection of Russell 3000 anomaly stocks that exploit recurring patterns documented in the financial literature. However, rather than just validating the continuation of the original Russell 3000 anomaly, I have applied a practical trading test from last year's study to exploit this irregularity. The results have significantly outperformed the baseline Russell 2000 index (more than 2x) and the S&P 500 (nearly 4x) the returns over the past year.

This article is not a substitute for thorough academic research and peer-reviewed analysis. This informal annual study provides insight to my readers on work I am conducting to better understand and profit from stock price anomalies in the marketplace. The purpose of all the different portfolio studies I make available on enhancing trading algorithms and anomaly research is to provide more practical trading applications from the financial literature that may be highly profitable to my readers.

Substantially more value is provided with membership here at Value & Momentum Breakouts where all ten of the Russell 3000 anomaly stock selections are made available in full to subscribers along with many additional studies and portfolios.

Review of last year's performance results

The performance results beginning last year in the Russell 2000 gained +19.64% in the one year period from the June 23 reconstitution date of the index as shown on the chart below.

The official FTSE Russell annual return of the Russell 2000 through May 31, 2018 was +20.8% which differs in time by about one month from my measurement start.

Stock Anomaly Returns 2017-2018

The table below shows the returns from the 2017 Anomaly Picks and compares the results over time against 3 benchmark indexes. The 2017 All New Additions represents those stocks added to the Russell 3000 in 2017 not including the Anomaly Selections. The Russell 2000 and S&P 500 Index are the standard benchmark indexes representing the complete index performance.

Anomaly PerformanceResults Annual Return % Return from Selection Date (11 months) Return 1st 6 months Return 2nd 6 months
2017 Anomaly Picks (9 stocks) +73.46% +39.92% +63.34% +10.12%
2017 All New Additions (149 stocks)* +16.43% n/a +13.05% +3.38%
Russell 2000 Index +19.64% +18.63% +9.00% +9.76%
S&P 500 Index +12.08% +9.91% +10.73% +1.15%

*Does not include the 9 stock selections that were added to the index.

** Original 2017 selection included the Top Ten stocks. However, Stone Energy (SGY) merged with Talos Energy (TALO) in Nov 2017 and was dropped from the study.

Significant Outcomes

One of the most significant observations of the 2017 Anomaly Picks was that 55% of the stocks (5/9) from last year were among the top 10% biggest returns of all the stocks added to the Russell 2000 index last year.

It is also noteworthy that 83% of the gains from the Anomaly Picks last year were realized in the first six months of the year. The second half of the year was still profitable but the gains were only 10.12% compared to 63.34% in the first half. This pattern followed for all the stocks added to the Russell 3000 index with more than 80% of the gains in the first six months and much lower returns in the second half.

The Russell 2000 index on the other hand performed slightly better in the second half of the year than in the first half indicating a shift back to many of the earlier underperforming stocks that originally comprised the index. Because of this first half effect and my close review of last year's selections, I have decided to reduce the selection period from 1 month to 1 week. It appears that a significant amount of momentum contributing to annual returns occurs close to the reconstitution date for the stocks that qualify for the anomaly effect and a month delay may reduce the expected annual returns.

The details on the selection method, the financial literature and structure of the Russell index follow below.

Reconstitution and Structure of the Russell 3000

The FTSE Russell is a global firm wholly owned by the London Stock Exchange Group and among many other functions also manages a set of well known market indexes. The Russell 3000 rebalancing process, that publishes the reconstituted index lists and end-of-June weights, provides much more transparency for analysis by money managers and researchers in contrast to the S&P 500 index (Chang et al., 2013). From the FTSE Russell website:

June is the month that the preliminary reconstitution portfolio is communicated to the marketplace. Beginning on June 9, preliminary lists are communicated to the marketplace and updates are provided on June 16 and 23. The newly reconstituted indexes take effect after the close on Friday, June 23.

The Russell 3000 is comprised of all constituents in both the Russell 1000 (U.S. large capitalization stocks) and the Russell 2000 (small capitalization stocks) indexes."

The complete list of Russell 3000 additions and deletions can be found on their website: Russell Reconstitution. The 2018 summary recap of the index reconstitution are as follows:

  • Representing the US small cap market segment, the total market capitalization of the Russell 2000 Index increased 9% from $2.3 trillion as of last year’s reconstitution to $2.5 trillion.
  • The largest company in the index with banding applied is Entegris, with a total market cap of $5.0 billion, which is about 12% larger in size than last year’s largest company in the index (see Exhibit 1).
  • The smallest company in the index is FuelCell Energy with a total market cap of $159.2 million (see Exhibit 1).
  • There are a total of 212 companies joining the Russell 2000 Index. Thirty-five are migrating from the Russell 1000 Index / Russell Midcap® Index, and 95 companies shifting up from the Russell Microcap® Index — over one third of which are represented by the Health Care sector.
  • There are 25 IPOs (mostly Health Care and Financial Services companies) joining the index and 57 companies joining the index from outside the Russell US Indexes universe.
  • The largest addition to the small cap index by size is an IPO — Zscaler — with a total market cap of $3.1 billion. The largest addition to the index by index weight is Liberty Latin America, which is joining due to a change in country assignment from UK to US.
  • There are 152 companies departing the Russell 2000 Index. The largest company leaving the index is Nektar Therapeutics, which is moving to the Russell 1000 Index due to its new total market cap of $12.6 billion. Another 47 companies moving to the large cap index represent the largest portion of the deletions by weight (95.1%). There are 83 companies moving from the Russell 2000 Index to the Russell Microcap Index, and another 21 companies leaving the Russell US Indexes universe altogether.

The overlap depicted in the banding process between Russell 1000 and Russell 2000 stocks is due to certain complications in the assignment process. The timing of stock selection is based on May cut-off dates for the market capitalization size-based determination of inclusion to the R1000 or R2000 that are subject to change prior to the June 23 effective date.

Temporary vs. Permanent Effects

More than 18 years of published research on the effects of the annual reconstitution of the Russell 3000 actually details a more positive short-term effect for deleted stocks than for added stocks. This provides another opportunity for testing return anomalies. However the current focus is on additions only.These results are shown below in Table 2 for the period 2000 to 2015 as compiled by FTSE Russell Index in their report. On average, during the period from Reconstitution to Aug-end the added stocks lost -2.86% while deleted stocks gained +1.44% in the same period (FTSE Russell, 2016). One explanation that represents current theory is that this rebalancing anomaly comprises temporary and permanent effects.

Historically, added stocks become more negative on average through the end of August even while eventually returning as much as 20% for the year ended May 31, 2017. So the working theory that I tested last year was that if anomalous returns for added stocks are not well realized in the first two months from reconstitution, then the permanent annual effects must be significantly more positive in later months. Or put another way -- the better that stocks perform against the historical short term negative effects, the better returns they may generate on the year long (permanent) horizon when more positive returns are historically realized.

One could certainly argue that if added stocks have reacted well with significant gains during this post reconstitution period then they may have already contributed the full measure of both the temporary and permanent effects we should expect for 2018. However,1) Not only was that not the case for 2017 with more than doubling the baseline Russell index, but selected stocks gained as much as 250% throughout the year.

2) Price momentum theory is based on the well documented phenomenon, "where stocks with low returns over the last year tend to have low returns for the next few months and stocks with high past returns tend to have high future returns" (Fama & French, 2008, p. 1653). This momentum pricing effect typically consists of increasing momentum stocks that continue to increase (defying prevailing averages in Table 2).

So we shall test again whether the post-rebalance period may identify some of the strongest momentum growth candidates for 2018.

Top Two Stocks for 2018 Russell Anomaly portfolio

The two stocks from the top 10 selected stocks for the 2018-2019 Russell anomaly are as follows in descending order of one-week price performance:

  • Innovate Biopharmaceuticals (INNT) +59.18%
  • Zafgen (ZFGN) +34.91%

All ten stocks are available with membership here at Value & Momentum Breakouts

Neither this article nor any of the studies researched for this article suggest that performance gains are exclusively the result of being added to the Russell 2000. It is very clear from prior published research as illustrated in the charts above however, that the invested contribution of forced indexing provides significant trading volume increases for most of the highest performing stocks. The impact of index reconstitution is well documented, but the anomalous one year (permanent) effect on individual firms has room for additional explanatory studies.

We have seen a statistically significant result from last year's selection portfolio based on a one month performance period. This year I am interested to see how this anomalous effect may impact the most positively reactive stocks after only the first week of announcement drift. I hope it may also result in more positive outcomes for my readers who join me in this study.

Please click the "Follow" button at the top of the page to enjoy free updates on the progress of each of the different portfolios I offer that are outperforming the S&P 500 in each of their respective time periods.

To Subscribe to Value & Momentum Breakouts, click on the link HERE.

All the very best, and have a great week of trading!

JD Henning, PhD, MBA, CFE, CAMS


Chang, Y., Hong, H., & Liskovich, I. (2013). Regression Discontinuity and the Price Effects of Stock Market Indexing. doi:10.3386/w19290

FAMA, E. F., & FRENCH, K. R. (2008). Dissecting Anomalies. The Journal of Finance, 63(4), 1653–1678. doi:10.1111/j.1540-6261.2008.01371.x

Charts from Tables and Figures from

Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in INNT, ZFGN over the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.