When is a Sell a Buy? (When the Sell-Side Says So)

Print Friendly, PDF & Email


Graham Copley / Nick Lipinski



October 24th, 2016

When is a Sell a Buy? (When the Sell-Side Says So)

  • Recommendation history in the Chemical sector in aggregate shows an inverse relationship with forward performance – this should not come as a surprise to seasoned investors
    • Performance improves as bearishness increases for our group of 25 Chemical companies
  • Counter-consensus performance trend is evidenced in several ways:
    • The 5 stocks with the most bearish sell side sentiment in any given month outperform the 5 most bullish stocks over the next year 66% of the time (since 2005) and by an average of 8%
    • At extremes in sentiment for individual stocks over the past five years, stocks are more likely to outperform after a bearish extreme (on 75% of such instances) and more likely to underperform after a bullish extreme (on 66% of such instances) – Exhibit 1
    • Breaking the historical recommendation history for each company into quadrants, the average 12 month forward performance in the most bearish quadrant is 15% compared to 1% in the most bullish quadrant (since 2005) – Exhibit 6
      • We show performance by quadrant for each company, as well as the current positioning, historical volatility and number of observations in Exhibit 10
  • Currently the five stocks with the most bullish recommendations are PPG, POL, SHW, OLN, and FUL
    • PPG and SHW recommendations are potentially a cause for concern as downgrades could trigger further weakness in the stocks, which have already started to show cracks
    • We are already counter-consensus on OLN and have been for some time
  • The better covered larger cap stocks with the most bearish sentiment are MOS, FMC, and CBT
    • MOS and FMC have upside tied to the ag cycle and possible ag consolidation – we like both longer-term
  • Additionally, LYB, PX and ECL are currently at 5 year individual bearish extremes and DOW is at an individual bullish extreme – but this is deal related and we agree with it.
    • If PX participates in further industrial gas consolidation this bearish signal is likely a buying point – if not, it is likely correct

Exhibit 1

Source: Capital IQ and SSR Analysis


In our Coatings report from earlier this year we highlighted that sector’s outperformance over the past five years by juxtaposing the underperformance of the Industrial Gas space and noting the divergence in sentiment that prevailed at that time. This is shown in Exhibit 2, where five years ago the Coatings and Gas subsectors were at opposite ends of the sentiment spectrum. Today the Coatings space has come full circle and analysts are now most bullish on this group.

Exhibit 3 summarizes these results at the sector level and Exhibit 4 shows this for the constituent stocks. The stock composition of the Chemicals subsectors is shown in the Appendix.

Exhibit 2

Source: Capital IQ and SSR Analysis

Exhibit 3

Source: Capital IQ and SSR Analysis

Exhibit 4

Source: Capital IQ and SSR Analysis

Value investors will confirm that one of the signals they look for is sentiment – when everyone hates a stock or a sector, it is probably time to take a look. Sell side recommendations tend to be lagging indicators – analysts like a stock for too long and hate a stock for too long and consequently peaks in bullish sentiment tend to be red flags and peaks in bearish sentiment tend to be green flags. There are likely many reasons for this and it is unlikely that any one situation can be explained by a single reason:

  • Over-confidence of management – telling a believable but overly positive story. We have written at length about corporate optimism. OLN is likely the best example of this issue today in our view, and has been for a while, but both DOW and DD were guilty in the past.
    • Analysts either buy into the corporate rhetoric or are not confident enough to offer a more negative view sometimes and opt for the consensus.
    • An example of where the sell-side is no longer listening is EMN – the story is too bullish but now no-one believes – if EMN could get the story right from here the stock could be interesting.
  • Macro consensus – this explains the bullishness in coatings and the bearishness in commodities.
    • There is belief that US housing can continue its growth – new homes and resale, spurred by low interest rates. This belief is propping up valuation and recommendations for PPG and SHW, especially give the sector’s focus in the US. We think this is now overplayed.
    • On commodities, there is a consensus view that the ethylene market is at peak, although while recommendations for LYB are at a bearish extreme, estimates are not and we believe that the bar in Exhibit 4 could extend further upwards for the balance of the year before creating a great entry point.
  • Inertia (or following rather than leading): It is very difficult and very contentious to stop backing a horse that is winning. The risk is that you will be too early and be remembered for that rather than the call itself.
    • The last research call I made at Bernstein was to downgrade one of the stocks I covered at the time to a sell – I was sure that the fundamentals were going to turn against the company. The stock went from $34 per share to $2 per share!! The problem was that over the six months following the downgrade it went from $34 to $50!

Performance Results

The counter-consensus performance differential between Coatings and Industrial Gas that we discussed is evident at a more granular level as well and on a number of analyses and over multiple time frames. Exhibit 1 showed the one year forward performance of individual stocks following extremes in bullish or bearish sentiment over the past five years. We have alternatively explored this counter-consensus performance trend by taking the five most bearish and five most bullish stocks at the end of every month and comparing the relative performance of these groups. The group of bearish stocks has outperformed the group of bullish stocks roughly 66% of the time since 2005 and by an average of 8% – Exhibit 5 – though the results have shown significant volatility (particularly in the ’08-’09 whipsaw). Since 2010, volatility has decreased and the hit rate has improved (bearish stocks outperform bullish stocks 73% of the time)

Exhibit 5

Source: Capital IQ and SSR Analysis

Quartile analysis also shows a trend of better performance as recommendations get more bearish. Exhibit 6 illustrates the average performance for our group of Chemical stocks broken down by recommendation range. When recommendations fall in the most bearish quartile, forward 12 month performance versus the S&P has averaged 15% – conversely in the most bullish quadrant, the equivalent performance figure is just 1%. Exhibit 7 shows the data at the company level and highlights the volatility – this analysis suggests the bearish side is more consistently associated with outperformance than the bullish side is associated with underperformance, though the differential is significant in its own right.

Exhibit 6

Source: Capital IQ and SSR Analysis

Exhibit 7

Source: Capital IQ and SSR Analysis

PPG currently has the most bullish recommendation in the Chemicals sector (seen in Exhibit 4). Exhibit 8 summarizes PPG’s recommendation history since 2005 and the performance results by recommendation quadrant. For PPG, bullish recommendations have not signaled stock success, and in fact the stock has most significantly outperformed when sentiment has been most negative.

Exhibit 8

Source: Capital IQ and SSR Analysis

Exhibit 9 represents a longer history (from 2005) than the five year period shown in Exhibit 4 and the table in Exhibit 10 summarizes each stock’s performance in its relative recommendation ranges as shown for PPG above. The current recommendation position for each stock is indicated by a gray highlight and boldface.

Exhibit 9

Source: Capital IQ and SSR Analysis

Exhibit 10

Source: Capital IQ and SSR Analysis

Note TROX and NEU are excluded from the performance results due to the low level of analyst coverage – most months these stocks have only 1 or 2 analyst recommendations, which does not reflect an accurate diversity of opinion. We have therefore restricted the performance analysis to recommendations comprised of at least 5 individual analyst estimates.



©2016, SSR LLC, 225 High Ridge Road, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

Print Friendly, PDF & Email