Contrarian Investing: It pays to be different

Thematic | Contrarian Investing

Contents: It pays to be different

It pays to be different ........................................................................................................... 3

Top contrarian

BUY

ideas ...................................................................................................... 6

Top contrarian

SELL

ideas ..................................................................................................... 9

Interesting charts from the note ......................................................................................... 11

Key quotes and references on contrarian investing ............................................................ 12

Contrarian investing: What it is; why it works .................................................................... 13

What is stock popularity? ................................................................................................... 19

Analyzing themes based on consensus recommendations .................................................. 21

Neutral to moderately popular stocks emerge winners ...................................................... 23

Up-capture and down-capture ............................................................................................ 27

Quintile-4 is best on risk-reward metric .............................................................................. 29

Final thoughts: Is consensus change significant? ................................................................. 30

Cases based on popularity theme ....................................................................................... 32

Consensus sells are significant underperformers ................................................................ 37

Consesus buys are not always great investments ............................................................... 40

Summary: Performance of the quintiles

............................................................................. 42

Contrarian strategies based on valuation multiples ............................................................ 44

Low P/E stocks outperform high P/E stocks ........................................................................ 46

Low P/B stocks dominate, but overall results mixed ........................................................... 54

Summary: Performance of contrarian themes

.................................................................... 56

Analysis of sectors on our framework ................................................................................. 57

Metals: A contra bet of CY16 .............................................................................................. 58

Pharma: Ripe for a contrarian bet ....................................................................................... 60

Information Technology: A contrarian bet? Not yet !! ........................................................ 63

June 2017

Thematic | Contrarian Investing

Contrarian Investing

It pays to be different

Identifying contrarian investment strategies that work

Contrarian investing is a time-tested investment tool, which

involves buying/selling

stocks that goes against the prevailing sentiment of crowd or the market.

It is

among the first lessons taught to budding investors and literature on the subject

dates back decades.

Contrarian Investment Strategies: The Psychological Edge

David Dreman,

Extraordinary Popular Delusions and the Madness of Crowd

Charles Mackay, and

The Triumph of Contrarian Investing

by Ned Devis are only

some of the literary works that help understand this subject. The objective of this

report is to develop an application toolkit for India. We have covered two major

themes based on consensus ratings (stock popularity) and valuation multiples, and

have demonstrated the success of contrarian investing in India.

What is stock popularity?

Bloomberg collects analyst recommendations on each stock and assigns a consensus

rating based on these recommendations. It assigns 5 points for every buy

recommendation, 3 points for every hold recommendation and 1 point for every sell

recommendation. A consensus rating is arrived at by taking the average of these

scores.



Popularity

Average



BSE100 popularity

4.3

4.0

3.8

3.5

3.3

3.0

Contrarian Investing:

It pays to be different

A stock with a consensus rating of 5 would have all buy recommendations.

A stock with a consensus rating of 3 would have an equal number of sell and buy

recommendations, apart from hold/neutral recommendations.

A stock with a rating change from <3 to >3 has a recommendation change from

a net sell to a net buy.

A stock with a consensus rating of 1 would have all sell recommendations.

Consensus sell rating (%) of a stock = number of sell recommendations / total

recommendations.

Consensus buy rating (%) of a stock = number of buy recommendations / total

recommendations.

Contrarian investing can generate disproportionate return



Gautam.Duggad@motilaloswal.com

91 22 3982 5404

Please click here for Video Link

Our analysis based on empirical evidence over the last decade suggests that

investing in out-of-favor stocks can generate disproportionate returns as

compared to the market.

Our findings suggest that neutral to moderately popular stocks deliver

significant outperformance, even bettering the most popular stocks. Our

findings are in harmony with the findings of

“Analyzing the analysts: When do

recommendations add value?”

Narasimhan Jegadeesh, Joonghyuk Kim,

Susan D Krische and Charles Lee, Journal of Finance 2004,

that in the absence of

favorable characteristics (value stocks and positive momentum stocks), the most

popular stocks are associated with the worst returns.

June 2017

Thematic | Contrarian Investing



Our findings prove that out-of-favor low P/E stocks deliver disproportionate

returns, significantly beating the benchmark. In contrast, the performance of

high P/E stocks is dismal. Our findings are in harmony with the findings of

“Cross-Section

of Expected Stock Returns”, by Fama-French,

Journal of Finance,

1992.

Great recommendation: Buy stocks with a consensus change from net sell

to net buy

Our findings suggest that a simple strategy of investing in stocks for which

analyst consensus has changed from “net sell to net buy” with a holding period

of one year has delivered 24.1% annual returns over the last 10 years.

In the study,

“Analyzing the analysts: When do recommendations add value?”

the

authors also find that the quarterly "change" in consensus recommendations is a

robust return predictor that appears to contain information not contained in a large

range of other predictive variables.



As a part of our analysis, we

have divided BSE100 in 5

groups of 20 – Quintile-1

(Q1), Quintile-2 (Q2),

Quintile-3 (Q3), Quintile-4

(Q4) and Quintile-5 (Q5),

respectively, with Q1 being

composed of the most

popular stocks and so on

60 winners (Q4, Q2, Q1): 40 losers (Q5, Q3)

Besides the superiority of neutral to moderately-popular stocks, we have

established that the most popular (Q1) and second-most popular (Q2) stocks also

beat the benchmark.

Predicting the winners: Popularity > Consensus Sell > Consensus Buy

The result seems fit, as popularity is an aggregate measure of sell, buy and hold

recommendations, and conveys more information than sell/buy recommendations

in isolation. Consensus sell recommendations, with their scarcity value, are better

predictors of stock performance than buy recommendations.

In each of our valuation

themes, we have made

groups based on P/E and

P/B

Low P/E (P/B) stocks are the

20 stocks in the BSE-100

having the lowest P/E (P/B)

ratios

Trailing multiples are better predictors than forward multiples

Our findings suggest that trailing valuation multiples are much better predictors of

future performance as compared to forward multiples.

Predicting the winners: P/E > P/B

Our findings suggest that the price-to-earnings (P/E) multiple is a better predictor of

returns, outperforming price-to-book (P/B), significantly.

Pharma ripe for Contrarian investing; IT not yet!!

The application of our framework on sectors suggests that opportunity for

contrarian investing is best in the Pharma sector. We have also back-tested the

thesis on the Metals sector based on numbers for FY16, when the sector checked all

the boxes for Contrarian Investing. However, we still do not believe that the IT

sector qualifies for Contrarian Investing yet.

June 2017

Thematic | Contrarian Investing

Exhibit 1: Price to Earnings> Price to Book in stock selection

600

500

400

300

200

100

Indexed Returns

Low PE

Low PB

BSE 100

Perfromance of different themes

571.1

324.8

269.6

Source: Bloomberg

Top Contrarian BUY ideas based on Popularity metric

Correction

Member

Popularity

% of Buy from 52w month

Company

ratings

High

return

score

Quintile

(%)

-1

Colgate

3.26

-20

-2

Axis Bank

3.22

-9

Bajaj Auto

3.35

-35

-21

Idea

3.00

-37

-27

Sun Pharma

3.45

-22

PNB

3.04

Top Contrarian BUY ideas based on Valuation metric

Company

Tech Mahindra

Coal India

Rural Elec.

Shriram Tran. Fin.

Contrarian SELL ideas

COMPANY

Asian Paints

Bajaj Finance

Eicher

Escorts

GAIL

IOC

LIC Housing Fin.

Voltas

Trailing

Forward

% of Correction

Popularity

month FY18E

P/E

Buy from 52w

P/E

score

return PE (x)

ratings high (%)

Quintile

(%)

5 Yr

Avg

PE (x)

14.0

15.8

4.4

15.1

PE -

Prem/

Disc to

LPA (%)

-14

-12

-15

FY18E

PB (x)

1.9

6.2

0.9

1.8

5 Yr

Avg

PB (x)

2.8

6.2

0.9

1.9

PB -

EPS

Prem/ CAGR

Disc to FY17-19

LPA (%)

(%)

-34

-8

FY18E

PE (x)

42.9

21.7

18.9

-6.2

21.2

14.2

5 Yr

Avg

PE (x)

38.3

16.0

18.0

21.1

30.6

10.5

PE -

Prem/

Disc to

LPA (%)

-31

FY18E

PB (x)

22.2

2.1

4.3

1.4

3.4

0.8

5 Yr Avg

PB (x)

24.7

1.9

5.1

2.1

5.7

0.8

PB -

Prem/

Disc to

LPA (%)

-10

-16

-31

-41

EPS

CAGR

FY17-19

(%)

Loss

3.94

4.03

3.43

3.68

-29

-30

-19

-25

-27

-22

116

-14

12.0

14.0

5.2

12.8

% of

Popularity

Buy/Hold

score

ratings

3.24

4.04

4.00

4.33

3.47

4.03

3.85

3.74

CMP

(INR)

1,164

1,401

27,653

710

360

394

770

465

Member Appreciation

PE -

Member

from

month Prem/Disc

of PE

return

Popularity 52w low

to LPA

Quintile

(%)

294

242

152

-8

-12

1 Year

Fwd

PB (x)

14.1

6.5

11.1

3.1

1.5

1.9

3.1

4.2

PB -

EPS CAGR

Prem/Disc

FY17-19

to LPA

(%)

125

266

-4

76%

91%

80%

100%

71%

84%

87%

81%

June 2017

Thematic | Contrarian Investing

Top BUY ideas based on the twin metrcis of Popularity

and Valuation multiples

Our top contrarian picks are:

Axis Bank, Bajaj Auto, Colgate, Coal India, Idea, PNB,

REC, Shriram Transport Finance, Sun Pharma & Tech Mahindra.

Axis Bank

The bank has made significant investments to ride the next growth cycle (post near-

term asset quality challenges), with support in the form of (1) strong capitalization

(13% tier-I) and (2) expanding liability franchise (3,400+ branches). For FY19, we

expect the bank to report 1.3-1.4% RoA and 16% RoE.

Axis Bank features in our list of Buy ideas owing to its presence in Quintile 4 of the

popularity metric.

Bajaj Auto

Valuations at 18.7x/15.8x FY18/19E (MOSL Est.) standalone EPS are attractive,

considering (a) scope of profitable market share gain, (b) high visibility of sustenance

of superior profitability and (c) lower capex requirement driving return ratios.

Bajaj Auto is also a member of Quintile 4 of the popularity metric – its valuations are

at long-period average (LPA), while the rest of the sector is at a premium to its

average valuations.

Colgate

Colgate is a strong play on an imminent rural volume recovery, led by a confluence

of positives such as likely normal monsoon, moderate inflation levels, government

schemes to boost growth (expansion of DBT benefits, increase in rural allocation in

the national/state budgets, and farm loan waivers) and a weak base of the past

three years. Over the medium-to-long term, margin drivers are operating leverage

on expanded capacity, lower A&P compared to elevated levels of the past three

years and continued premiumization. Colgate features in Quintile 4 of the popularity

metric.

Coal India

CIL has a strong balance sheet, healthy cash flow and net cash status. With volume

uptick apparent in the recent months, we believe the benefit of operating leverage

will result in EPS growth.

Coal India features in Quintile 5 of P/E valuation metric (quintile with lowest P/E)

and has valuations which are trading at discount to its LPA.

Idea

Idea’s merger with Vodafone will improve its spectrum and data network quality. In

the current hyper-competitive market, it will allow Idea to match RJio’s offering and

subsequently arrest market share dilution. We believe the industry should stabilize

in the next 3-4 quarters as RJio’s free usage offer concludes and its network reaches

optimum capacity. Idea has one of the lowest Buy ratings on the street and is part of

Quintile 5 of the popularity metric.

June 2017

Thematic | Contrarian Investing

Punjab National Bank

PNB remains highly levered to the RBI’s resolution mechanisms and an improvement

in recovery environment. Significant stress has been recognized over the last eight

quarters, and OSRL has also come off sharply in last one year. Concentration of Infra

(incl. power) and Iron & Steel is very high in stress loans (50%+), and any upgrade in

these segments can lead to a significant upgrade in earnings. In our view, RoAs and

RoEs have bottomed out in FY17, and we expect a gradual improvement hereon.

PNB is part of Quintile 4 of our popularity metric, and has valuations trading at LPA

with less than 30% of Buy ratings on the street.

Rural Electrification Corporation (REC)

REC’s near- to medium-term earnings were at risk following the implementation of

UDAY. However, progress so far has been encouraging both on the growth and NIM

front, and RECL has delivered good results this year. Despite repayment of DISCOM

debt under the UDAY scheme, loan growth is anticipated to be in positive trajectory,

with working capital and refinancing demand on strong turf. RECL features in

Quintile 5 of our valuation-based metric, with valuations at LPA.

Shriram Transport Finance (SHTF)

SHTF’s return ratios are at cyclical lows, with decadal high credit cost and NPLs. We

believe the worst of asset quality troubles is behind, and SHTF should witness

improving return ratios due to lower credit costs. Additionally, we believe margin

compression fears are overplayed, with the company yet to reap significant benefit

on CoF. SHTF is part of Quintile 3 of our valuation-based metric, with valuations at

discount to long-period averages.

Sun Pharma

Near-term earnings will be under pressure due to the weak business outlook in the

US and the GST impact in India. However, this is more than factored in the recent

stock price decline, in our view (valuations at ~30% below five-year average). We

believe the strategic investments in the specialty business over last 2-3 years,

coupled with stable growth in the domestic market and the enhanced focus on

complex generics in the US, will help drive earnings growth in the medium term.

Sun Pharma is part of Quintile 4 of the popularity metric, with valuations at 30%

discount to long-period average.

Tech Mahindra

Gradual recovery in Telecom and continued outperformance in Enterprise are likely

to lead to an improvement in revenue growth and profitability, subsequently

resulting in valuation catch-up with peers. Tech Mahindra is part of Quintile 5 of the

P/E valuation-based metric, with valuations at ~15% discount to LPA.

June 2017

Thematic | Contrarian Investing

Top Contrarian BUY ideas based on Popularity metric

Correction

Member

Popularity

% of Buy from 52w month

Company

ratings

High

return

score

Quintile

(%)

-1

Colgate

3.26

-20

-2

Axis Bank

3.22

-9

Bajaj Auto

3.35

-35

-21

Idea

3.00

-37

-27

Sun Pharma

3.45

-22

PNB

3.04

Top Contrarian BUY ideas based on Valuation metric

Company

Tech Mahindra

Coal India

Rural Elec.

Shriram Tran. Fin.

FY18E

PE (x)

42.9

21.7

18.9

-6.2

21.2

14.2

5 Yr

Avg

PE (x)

38.3

16.0

18.0

21.1

30.6

10.5

PE -

Prem/

Disc to

LPA (%)

-31

FY18E

PB (x)

22.2

2.1

4.3

1.4

3.4

0.8

5 Yr Avg

PB (x)

24.7

1.9

5.1

2.1

5.7

0.8

PB -

Prem/

Disc to

LPA (%)

-10

-16

-31

-41

EPS

CAGR

FY17-19

(%)

Loss

Trailing

Forward

% of Correction

Popularity

month FY18E

P/E

Buy from 52w

P/E

return PE (x)

score

ratings high (%)

Quintile

(%)

5 Yr

Avg

PE (x)

14.0

15.8

4.4

15.1

3.94

4.03

3.43

3.68

-29

-30

-19

-25

-27

-22

116

-14

12.0

14.0

5.2

12.8

PE -

Prem/

Disc to

LPA (%)

-14

-12

-15

FY18E

PB (x)

1.9

6.2

0.9

1.8

5 Yr

Avg

PB (x)

2.8

6.2

0.9

1.9

PB -

EPS

Prem/ CAGR

Disc to FY17-19

LPA (%)

(%)

-34

-8

June 2017

Thematic | Contrarian Investing

Top contrarian SELL ideas

Our top sell ideas are:

Asian Paints, Bajaj Finance, Escorts, GAIL, LIC Housing, IOC,

Eicher and Voltas.

Asian Paints

The slowdown in real estate is likely to inhibit growth prospects. Industry reports

suggest that demand in this segment, which is 20% of the paints business, has nearly

halved in recent months. The company has had tailwind of low crude costs in the

past three years. Any sharp reversal would make margins vulnerable. RoCE has

consistently declined from ~40% in FY11 to ~24% in FY17, as the non-paints

businesses have struggled. Valuations at 42x FY19E EPS are at multi-year highs (5-

year average is 39.6x, 10-year average is 31.2x, and 15-year average is 26.3x).

APNT features in Contrarian Sell owing to its presence in Quintile 1 of P/E based

valuation metric with valuations at 30% premium to LPA and 75% Buy ratings.

Bajaj Finance

BAF – a dominant player in the consumer durables financing segment – continues to

reap the benefits of healthy consumer demand, increasing its market share in

consumer as well as other businesses. While yields are likely to remain under

pressure due to lower share of CD financing, the stock looks expensive at 6.5x/5.2x

FY17E/18E BV (MOSL Est.).

BAF is also a part of Quintile 1 of valuation based metric with valuations at 100%+

premium to its LPA and 90% Buy ratings.

Eicher Motors

We expect Royal Enfield (RE) volume growth to moderate to ~18% CAGR over FY17-

20 (against ~55% CAGR over CY13-FY17). Despite factoring in EBITDA margin of

~33% for RE, consolidated EPS growth is estimated to moderate to ~31% CAGR over

FY17-20 (v/s 62% CAGR over CY13-FY17). This moderation in earnings growth

trajectory would drive moderation in valuations assigned to RE business (we assign

exit P/E of ~27.5x for FY20), which in turn would result in stock underperformance.

Eicher Motors is member of Quintile 2 of P/E based valuation metric with valuations

at 20% premium to its LPA and 80% Buy ratings.

Escorts

FY17 has been a robust year for ESC, with Tractors volume growth of 24% and

EBITDA margin expansion of 280bp. However, Tractors is dependent on monsoon

and undergoes business cycle; delivering growth on linear basis here thus appears

challenging. With sharp-run up in stock , stock doesn’t look attractive on valuation

front. Escort’s valuation are at 150% premium to its LPA with 100% Buy ratings.

GAIL

While we do build in 32%/15% growth in EPS in FY18/19, return ratios of 12% are 5%

lower than what they were five years back. There is no visibility on most part of

5.8mmtpa 20yr contract to import gas from the US. A hit of USD1/mmBtu could

wipe out 22% EBIDTA. The threat calls for lower multiple. We value it at INR357

using 9x FY19 adj EPS & adding value of investments.

Gail features in Quintile 3 of our Popularity metric with 70% Buy ratings.

June 2017

Thematic | Contrarian Investing

Indian Oil Corporation (IOCL)

Adjusting for inventory gains and one-time employee expense, we would see ~15%

EBITDA increase in FY18, with normalization of Paradip refinery. However, we do

not see more than 3-4% growth in FY19. Market share loss has become significant,

with private players achieving 3% share in petrol and 8% in diesel. Over the next two

years, free cash flow would be good. However, with the company committed to

INR1.75t capex over seven years, the balance sheet is likely to be strained. We see

very little, if any, probability of valuation re-rating from here.

IOCL has 85% Buy ratings with valuations at 70% premium to its LPA on P/B.

LIC Housing Finance

While margins have been on an uptrend over the past four quarters due to declining

cost of funds, growth remains an issue. Core home loan growth has been range-

bound at 9-10% over the past 4-5 quarters. Almost 50% of the incremental loan

growth in FY17 was driven by LAP and developer loans – this has also helped

maintain yields. However, we believe that with the share of non-core book

stabilizing, growth would slow down and margin expansion would be subdued. The

stock has performed well over the past one year and now trades at 2.7x FY19E BV.

LIC Housing Finance has 90% Buy ratings with valuations at 50% premium to its LPA.

Voltas

We expect the air conditioner (AC) industry to see a rapid shift to inverter ACs. We

expect inverter ACs to constitute 30% of industry volumes in CY18 and 50% in CY20.

Voltas has historically been weak in this category; it is likely to lose share to the

Japanese/Koreans. Margins in UCP segment, which were at peak levels of 14.5% in

FY17, are likely to decline over the next two years. Valuations are steep at 35x FY19E

EPS for the UCP segment v/s historical band of 20-25x.

Voltas with 80% Buy ratings and valuations at 50% premium to LPA fits in well in our

Contrarian Sell thesis.

Contrarian SELL ideas

COMPANY

% of

Popularity

Buy/Hold

score

ratings

CMP

(INR)

Member Appreciation

PE -

Member

from

month Prem/Disc

of PE

return

Popularity 52w low

to LPA

Quintile

(%)

1 Year

Fwd

PB (x)

PB -

EPS CAGR

Prem/Disc

FY17-19

to LPA

(%)

Asian Paints

3.24

1,164

76%

Bajaj Finance

4.04

1,401

91%

Eicher

4.00

27,653

80%

Escorts

4.33

710

100%

GAIL

3.47

360

71%

IOC

4.03

394

84%

LIC Housing Fin.

3.85

770

87%

Voltas

3.74

465

81%

Disclaimer:

The ideas highlighted in this section may

specialist analysts.

14.1

6.5

125

11.1

294

242

3.1

152

266

1.5

-8

-4

1.9

-12

3.1

4.2

not be in consonance with the ratings assigned to the respective stocks by our sector-

June 2017

Thematic | Contrarian Investing

Three interesting charts from this report

Exhibit 2 highlights the comparison of Trailing Price to Book vs. RoE for the markets and compares the historical

trend.

Exhibit 2: Herd behavior evident in 2007-2008 – Nifty P/B (trailing) v/s RoE

Market Price to Book Ratio(trailing) Vs Return on Equity

7.0

6.0

5.0

4.0

3.0

2.0

Dec-08

Undervalued

y = 0.1859x + 0.284R² = 0.661

Source: NSE

Valuations haven’t yet reached the

stratospheric levels of Dec’07. RoE is

lower today but so is Cost of Equity.

Current

Mar-15

Overvalued

In the earlier peak of Dec’07,

RoE was much higher.

Dec-07

Exhibit 3 demonstrates the key finding of our analysis based on popularity of consensus ratings – Quintile 4

(Neutral to Moderately Popular) stocks have given best returns.

Exhibit 3: Stocks on which consensus is neutral to moderately positive have performed best

Performance of different quintiles of BSE100 by popularity ( quarterly rebalance)

500

400

300

200

100

Absolute Total returns (Average

)

Most Popular(CAGR-14.8%)

Q3(CAGR-5.6%)

Q5(CAGR-6.9%)

Q2(CAGR-13.5%)

Q4(CAGR-16.0%)

BSE100(CAGR-10.2%)

458.0

405.6

372.0

269.6

198.0

174.8

Source: Bloomberg

Exhibit 4 chart juxtaposes the popularity of current constituents of Sensex based on consensus ratings.

Exhibit 4: Stock popularity of 30 largest companies arranged by consensus popularity ratings

Consensus rating of 30 largest stocks of BSE100

100%

75%

50%

25%

Sell(%)

Hold(%)

Buy(%)

Consensus rating(rhs)

5.0

3.8

2.5

1.3

0.0

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Key quotes and references on contrarian investing

Jesse Livermore, one of the greatest stock traders of all time, has remarked, “Wall

Street never changes. The pockets change, the stocks change, but Wall Street

never changes because human nature never changes”.

What he said decades ago

holds true today as well! Contrarian investing works on the same premise.

As early as the 1840s, in his remarkable book,

Extraordinary Popular Delusions and

the Madness of Crowd,

Charles Mackay observed: “We find that whole communities

suddenly fix their minds upon one object and go mad in its pursuit; that millions of

people become simultaneously impressed with one delusion, and run after it…Sober

nations have all at once become desperate gamblers and risked almost their

existence upon the turn of piece of paper …Men, it has been well said, think in herds

…go mad in herds, while they only recover their senses slowly and one by one.”

Benjamin Graham’s “Intelligent Investor”

cites a study involving the constituents of

Dow Jones Industrial Average.

The performance of the 10 lowest P/E stocks, 10

highest P/E stocks, and of all the 30 stocks in the index was measured over set

periods between 1937 and 1969. In each time span, the low P/E did better than the

market and the high P/E stocks did worse.

In their legendary paper,

“Cross-section of Expected Stock Returns”,

Journal of

Finance, 1992,

Eugene Fama and Kenneth French

reported results similar to

Benjamin Graham’s findings. Several studies before and after the Fama-French study

also point towards the outperformance of out-of-favor stocks.

In “Analyzing

the analysts: When do recommendations add value?”,

Journal of

Finance 2004, Narasimhan Jegadeesh, Joonghyuk Kim, Susan D Krische and Charles

Lee have highlighted that naive adherence to the consensus recommendations can

be costly, because the "level" of the consensus recommendation adds value only

among stocks with favorable quantitative characteristics (that is, value stocks and

positive momentum stocks). In fact, among stocks with unfavorable quantitative

characteristics, higher consensus recommendations are associated with worse

subsequent returns.

In “Can

Investors Profit from the Prophets? Security Analyst Recommendations and

Stock Returns”,

Journal of Finance 2002, Brad Barber, Reuven Lehavy, Maureen

McNichols and Brett Trueman, highlight that stocks with the most favorable

recommendations outperform stocks with the least favorable recommendations. Our

findings are also on similar lines.

Refer to page 13,

(Contrarian investing:

What it is; why it works)

for details

Refer to Page 46-53 “Low

P/E stocks outperform high

P/E” for details

Refer to Page 23-26

“Neutral to moderately

popular stocks emerge

winners

June 2017

Thematic | Contrarian Investing

Contrarian investing: What it is; why it works

Amongst the multitudes of investment frameworks and styles which have worked,

one of the most interesting and least popular is Contrarian Investing. Going against

the grain and herd is not a common natural human instinct, be it investing or any

other domain. In investment parlance, Contrarian investing, in simple terms, is an

investment style which goes against the prevailing market consensus and is

characterized by buying and selling against the prevailing market sentiment. It has

some similarity with value investing as both contrarian and value investors are

hunting for price-value mismatch. However, value investing focuses mostly on

valuations and its overarching priority is to look for valuation mismatch vs. intrinsic

value of the company. Contrarian investing can also encompass several qualitative

aspects (market sentiments, macros, expectations) along with the quantitative

parameters of valuations.

Jesse Livermore, one of the greatest stock traders of all time, has remarked, “Wall

Street never changes. The pockets change, the stocks change, but Wall Street

never changes because human nature never changes”.

What he said decades ago

holds true today as well! Contrarian investing works on the same premise.

“Contrarian

investing is an investment strategy that is characterized by investing in

stocks which are out of favor with the crowd or market. A contrarian investor

believes that certain crowd behavior like irrational exuberance, overreaction, and

human limitation as a forecaster can lead to certain mispricing’s in stock market

which can be exploited in a consistent and systematic manner”.

Our evidence from

India suggests the same.

Investor enthusiasm or crowd herding often causes popular stocks to become

overpriced. Similarly, lack of investor enthusiasm often dumps unpopular stocks into

the bargain basement. Earnings and other surprises result in re-evaluation of both

groups and more realistic pricing. Contrarian investing focuses on buying stocks

dumped into the bargain basement. Some contrarian strategies are:

1. Buying stocks with low price-to-earnings

2. Buying stocks with low price-to-book

3. Buying stocks with low price-to-cash-flow

4. Buying neutral-to-moderately-popular stocks (popularity based on consensus

rating*)

Evidence from India shows that the companies the market expects the best future

for (as reflected in high price-to-earnings or high price-to-cash-flow) often

underperform, while the companies the market expects the worst future for (as

reflected in low price-to-earnings or low price-to-cash-flow) often outperform.

As early as the 1840s, in his remarkable book,

Extraordinary Popular Delusions and

the Madness of Crowd,

Charles Mackay observed: “We find that whole communities

suddenly fix their minds upon one object and go mad in its pursuit; that millions of

people become simultaneously impressed with one delusion, and run after it…Sober

June 2017

Thematic | Contrarian Investing

nations have all at once become desperate gamblers and risked almost their

existence upon the turn of piece of paper …Men, it has been well said, think in herds

…go mad in herds, while they only recover their senses slowly and one by one.”

This is the case in the stock market, as well – investors get fixated with some

popular stocks. As investors crowd around these popular stocks, they become

overpriced, and as evidenced in this study, fail to deliver returns.

We delve into facets of human behavior, heuristics and biases, which explain the

success of contrarian investing.

Investor overreaction hypothesis (IOH)

The investor overreaction hypothesis (IOH) states that investors overreact to certain

events in a consistent and predictable manner. Some key predictions of IOH are:

1. Investors will continuously overvalue favored stocks and undervalue out-of-

favor stocks.

2. Investors will find themselves over-optimistic on forecasts of the “best” stocks

and too pessimistic on those of the “worst” stocks over time.

3. Absolute contrarian strategies provide superior returns over time.

4. Out-of-favor stocks as valued by at least three major fundamental

measurements – low price-to-earnings, low price-to-cash-flow and low price-to-

book-value – will outperform the market as a group over longer periods of time,

normally 5-10 years or more.

5. Favored stocks as measured by high price-to-earnings, high price-to-cash-flow

and high price-to-book ratios will underperform the market over the same time

periods.

6. IOH posits that the outperformance of out-of-favor stocks and the

underperformance of favorites are caused primarily by behavioral influences

(affect, cognitive heuristics, neuroeconomics and other psychological variables).

As implicitly stated in IOH, it is not only in manias, contagion and panic that this

predictable overreaction occurs. We see this consistently in normal market settings.

As we will see across themes, the “best” stocks (those with the most promising

prospects by various yardsticks) underperform as a group as compared to “out-of-

favor” stocks.

And that’s why contrarian investing works successfully!

Human limitation as a forecaster – overconfidence bias and over-optimism

The consistent underperformance of the best stocks comes from investors’ and

analysts’ over-confidence in their ability to predict earnings accurately, though

empirical evidence proves that this cannot be done.

Consider for instance, the Nifty constituents, all well-tracked companies on which a

plethora of information is easily available. We juxtaposed the forecasted aggregate

earnings growth for the Nifty against the actual earnings growth for the last eight

years. We found significant over-optimism on earnings growth towards the start of

the year, which tapers off towards the end of the fiscal year, and the forecasts are

quite off the mark, even towards the end of the fiscal year.

June 2017

Thematic | Contrarian Investing

If the magnitude of error is so high for the most actively tracked companies, the

error would probably be much larger for companies outside the Nifty. The market

would be more likely to punish high-multiple stocks, for which expectations are

much more than low-multiple stocks, from which expectations are muted.

Our study highlights that trailing multiples are better predictor of future

performance as compared to forward estimates, which emphasizes human

limitation as a forecaster.

Exhibit 5: Nifty earnings growth (%) – actual v/s consensus estimates

Estimate

(12 months prior)

18.9

11.9

25.1

19.3

17.2

23.9

17.1

22.0

20.7

19.6

Estimate

(3 months prior)

5.8

7.6

25.4

8.9

14.1

16.2

18.0

15.4

17.7

14.3

Actual

-13.4

7.8

25.8

7.1

2.9

14.5

-5.1

-0.7

10.8

5.5

Source: Bloomberg

FY09

FY10

FY11

FY12

FY13

FY14

FY15

FY16

FY17

Average

Combining this with IOH which states that crowd consistently overvalues favorite

stocks, such an overvaluation with the over optimism of analysts to predict the

earnings hits them much harder as compared to the other stocks. Simply put it,

“negative earnings surprise would affect these stocks much more as compared to

other stocks”.

Human biases and heuristics

Affect:

Our strong likes, dislikes and opinions, experienced feelings such as

happiness, sadness, excitement and fear, can consciously or unconsciously, heavily

influence our decision-making process. Affect can cause behavior to move many

standard deviations away from the norm. So, our own preference for sports might

prompt us to invest in a sports company as compared to a radio company, which

appears dull to us but the radio company might be a better investment.

Availability bias:

Availability bias is a behavioral concept that describes how our

environment can shape our perceptions. Our perceptions are heavily influenced by

what is personally most relevant, recent or dramatic, which represent only a part of

the complete economic picture. For example, a person whose home has lost 20% of

its market value and whose spouse endured a long period of unemployment is less

likely to see or feel an economic recovery even while housing markets show signs of

recovery and unemployment ticks down.

Anchoring and hindsight biases:

The other two systematic heuristic biases that tend

to cause investment errors are anchoring bias and hindsight bias. These biases also

reinforce the others.

Anchoring

is a simplifying heuristic that sets up a price for a

June 2017

Thematic | Contrarian Investing

stock that is not far removed from its current trading price as its anchor. The

heuristic process tends to drop the anchor far too close to the stock’s current price

and investors don’t seem to believe that price can go either much lower if they want

to sell or much higher if they want to buy; this often leads to missed opportunities.

The inevitability of what happened seems obvious in retrospect.

Hindsight

bias

seriously impairs proper assessment of past errors and significantly limits what can

be seen from experience.

2007-2008: Could the crash have been foreseen?

In his book, Extraordinary Popular Delusions and the Madness of Crowd, Charles

Mackay had remarked in 1840,

“Men,

it has been well said, think in herds ……go

mad in herds, while they only recover their senses slowly and one by one”.

Fast

forward 168 years, did a similar thing happen in December 2007? Can crowd

behavior led to bubbles and manias?

Consider Exhibit-6. The expectation (P/B) attached by the investor to return on

equity (RoE) was mindboggling in December 2007. A crash was imminent (in

hindsight?), as the valuations were extremely high. Even in the months prior to

December 2007, the valuations were stretched, but the herd didn’t budge!

Exhibit 6: Herd behavior evident in 2007-2008 – Nifty P/B (trailing) v/s RoE

Market Price to Book Ratio(trailing) Vs Return on Equity

7.0

6.0

5.0

4.0

3.0

2.0

Dec-08

Undervalued

y = 0.1859x + 0.284R² = 0.661

Source: NSE

Valuations haven’t yet reached the

stratospheric levels of Dec’07. RoE is

lower today but so is Cost of Equity.

Current

Mar-15

Overvalued

In the earlier peak of Dec’07,

RoE was much higher.

Dec-07

As things stand today, valuations haven’t yet reached the stratospheric levels of

December 2007. While RoE is lower for the aggregates vis-à-vis December 2007, the

P/B multiple is also lower. In the previous peak of December 2007, valuations kept

on rising, accompanied by expanding RoE. If the past is anything to go by, then we

can see expansion in P/B multiples even from the current levels, if earnings growth

recovers and capital efficiency ratios improve. Another difference vis-à-vis

December 2007 is in the cost of capital – it is lower today, and thus, supportive of

relatively rich valuation multiples.

Move to a different geography – the US, a year before crash! In the housing bubble

in the US, which evolved into the Global Financial Crisis, excessive use of leverage

June 2017

Thematic | Contrarian Investing

played a massive role in destruction of wealth. The world’s smartest investors and

rating agencies that rated the mortgages were all caught in mania!

In his book, Charles Mackay highlights the

Dutch Tulip Mania.

In 1637, investors

scrambled to buy tulips, paying the equivalent of USD75, 000 for a Semper

Augustus, a rare bulb. Then, suddenly, the price collapsed to a miniscule fraction of

this. The question is: how does something so valuable suddenly become worthless?

There was no disruptive technology or substitute! Rationality definitely goes for a

toss!

Human behavior has been remarkably similar across different geographies and time.

In 1637, investors overreacted to events and then corrected their original reaction,

causing a major reversal in price. In December 2008, the Nifty was at deep discount

– investors had overreacted; normalcy returned thereafter.

June 2017

Thematic | Contrarian Investing

Consensus recommendations

Popularity as the selection criterion

Does stock popularity with analyst has a meaningful

impact on stock’s performance?

Consensus Sell rating (%) as the selection criterion

Sell recommendations are not as common as buy/hold

recommendations, and possess scarcity value.

Consensus Buy rating (%) as the selection criterion

With the scarcity value theme of consensus sell ratings

playing out well, we put to test the street’s capability

to predict the winners.

Consensus recommendation change from “Net sell”

to “Net buy”

Valuation

P/E

P/B

June 2017

Thematic | Contrarian Investing

What is stock popularity?

In this report, we have tried to analyze whether a stock’s popularity with the

consensus can tell us about its market price performance. Bloomberg collects

analyst recommendations on each stock and assigns a consensus rating based on

these recommendations. It assigns 5 points for every buy recommendation, 3 points

for every hold recommendation and 1 point for every sell recommendation. A

consensus rating is arrived by taking the average of these scores.



Consensus

Recommendations

A stock with a consensus rating of 5 would have all buy recommendations.

A stock with a consensus rating of 3 would have equal number of sell and buy

recommendations, apart from hold/neutral recommendations.

A stock with a rating change from <3 to >3 has a recommendation change from

a net sell to a net buy.

A stock with a consensus rating of 1 would have all sell recommendations.

Consensus sell rating (%) of a stock is number of Sell recommendations/Total

recommendations.

Consensus buy rating (%) of a stock is number of Buy recommendations/Total

recommendations

We have used the consensus rating of a stock as a measure of its popularity.

stock with a consensus rating of 5 is very popular with analysts, whereas a stock

with a consensus rating of 1 is very unpopular. A stock with a consensus rating of 3

is termed as neutral.

Exhibit 7: Stock popularity of 30 largest companies ordered by popularity with analysts

Consensus rating of 30 largest stocks of BSE100

100%

75%

50%

25%

Sell(%)

Hold(%)

Buy(%)

Consensus rating(rhs)

5.0

3.8

2.5

1.3

0.0

Source: Bloomberg

Amongst these largest 30 companies,

ITC

is trading at relatively attractive valuations

(32x FY18E EPS; at a significant discount to its FMCG peers). Cigarette volume

growth has been resilient despite demonetization while GST has been a positive

event for ITC. Earnings predictability and cigarette volume resilience coupled with

valuation discount to the sector has resulted in the stock being the most popular

large cap name on the street as per consensus popularity score.

HDFC Bank

remains one of the most popular stocks, given its solid earnings growth

led by consistent market share gains (value migration from public to private sector

June 2017

Thematic | Contrarian Investing

banks), pristine asset quality and unparalleled track record of multiple years of 20%+

earnings growth. HDFC Bank is one of our long-term picks in the BFSI space.

M&M

has seen improvement in its popularity, as it is widely expected to be the key

beneficiary of improving rural consumption prospects on the back of second

consecutive year of normal monsoon. The management’s upbeat guidance on

tractor volume growth coupled with potential recovery in UV market share is also

driving the overall bullish sentiment on the stock. M&M, Tata Motors and Maruti

are our preferred bets in the Auto sector.

ICICI Bank

has seen improvement in its consensus popularity scores post 4QFY17

results.

Overall pool of stressed loans is showing signs of stability, and bulk of NPA

recognition is happening from the watch list and OSRL. The government and RBI’s

sense of urgency

for resolution of lumpy stressed loans is also providing comfort at

the margin.

Among the corporate-oriented private sector banks, the street’s

preference has tilted towards ICICI Bank.

Asian Paints

has seen erosion in its consensus popularity scores largely owing to a

combination of lifetime high valuations (42x FY19E EPS) coupled with moderating

Decorative Paints volume growth. Implementation of GST could also impact

performance in 1HFY18.

The IT sector is facing multiple headwinds: margin pressure in traditional services

along with protectionist impetus in the US, which makes

TCS

and

Wipro

one of the

least popular stocks, with consensus popularity score of 2.6-3.

Macroeconomic issues (sharp INR depreciation) in key export markets including

Nigeria and Egypt, absence from the scooter segment (higher growth than

motorcycles), and continued impact of demonetization on two-wheeler volumes

have rendered a neutral consensus recommendation for

Bajaj Auto.

Elevated slippages, weak business outlook and expected pressure on core income

have led to

Axis Bank’s

drop in popularity.

June 2017

Thematic | Contrarian Investing

Analyzing themes based on consensus recommendations

For our study, we have divided the BSE-100 into five quintiles (groups of 20), based

on different selection criterion. We have begun with the December 2006 quarter,

and have re-balanced the quintiles quarterly, based on criterion/theme.

As a next step, we have evaluated the absolute total returns (including dividends) of

each quintile and compared these with the returns from the BSE-100. This exercise

has given us important insights for stock selection.

Central theme: Popularity as the selection criterion

This is the central theme of our study. The BSE-100 Index constituents have been

divided into quintiles based on popularity score, with Quintile-1 being composed of

the highest popularity stocks and so on. This has been done for each quarter starting

December 2006 and the total returns from each quintile have been evaluated.

Neutral to moderately popular stocks (Quintile-4) performed the best, followed by

the most popular stocks (Quintile-1, and then, Quintile-2). The least popular stocks

(Quintile-5) and those in Quintile-3 were perennial underperformers, rarely beating

the benchmark.

As a part of our analysis, we

have divided BSE100 in 5

groups of 20-Quintile1 (Q1),

Quintile2 (Q2), Quintile3

(Q3), Quintile4 (Q4) and

Quintile5 (Q5) respectively

with Q1 being composed of

most popular Quintile have

also beaten the BSE 100 benchmark returns, which allows us to categorize BSE 100

constituents in 60 winners and 40 Losers.

Exhibit 14: Existing BSE100 constituents divided in five quintiles as per popularity

SUNP

GNP

BHFC

CLGT

UNSP

BOB

AXSB

BOS

HUVR

SIEM

BHIN

NEST

BJAUT

DABUR

BHEL

PNB

ACEM

RPWR

IDEA

RCAPT

HDIL

ARBP

ITC

VEDL

HDFCB

HNDL

NTPC

ICICIBC

PWGR

UPLL

IIB

TTMT

SBIN

EIM

INFO

MSIL

HCLT

YES

RELI

TTCH

GMRI

ONGC

HDFC

IOCL

BRIT

COAL

TECHM

IHFL

POWF

JSP

TATA

BHARTI

EXID

ADSEZ

RECL

SHTF

GAIL

BPCL

HPCL

RIL

BAF

MSS

LPC

CDH

TGBL

JSTL

KMB

UTCEM

GRASIM

MMFS

KKC

TTAN

LICHF

HMCL

NMDC

APNT

CIPLA

TPWR

SAIL

RCOM

UNBK

BOI

CBK

ABB

ACC

GCPL

TCS

MRCO

WPRO

UBBL

DIVI

DRRD

DLFU

IDBI

Source: MOSL, Bloomberg

Amongst the current constituents of the BSE 100, we note that Quintile-4 (Q4) has

popularity scores of 3-3.45. We are highlighting Sun Pharma, Axis Bank, Bajaj Auto,

Colgate and PNB as our contrarian Buy ideas based on popularity metrics.

June 2017

Thematic | Contrarian Investing

An important question remains unanswered!

Why does the quintile composed of

stocks that have low consensus ratings/popularity (Quintile-4) outperform all the

other quintiles, even bettering the performance of the most popular stocks?

This is because the investment thesis of the most popular stocks is quite popular,

making them more likely to be discussed in the market. And, as such, there are

fewer marginal buyers with time. On the other hand, neutral to moderately popular

stocks are usually neglected by the street and their drivers are evolving. As the

drivers evolve, there is a sudden surge in investor interest in such stocks, leading to

a spike in returns. Thus, the stocks that were previously neglected by investors

deliver significant outperformance.

On the other hand, the least popular stocks underperform. This is because the least

popular stocks have a lot of sell recommendations. Sell recommendations are not

popular on the street – only about 19% of recommendations are sell as against

556% buy and 24% hold. So, they have scarcity value and are more likely to be acted

on (exhibit 21). We delve on this aspect in greater detail later in this report.

Quintile-4 offers higher dividend yield than market, which provides a downside

protection, but the difference is only marginal, which prompts us to look at the

beta

of the quintiles to explain the outperformance of Quintile-4. This is of particular

interest for Qunitile-1 and Quintile-4.

Exhibit 15: Quintile 1, 2 and 4 offer higher yields than market, but only marginally

14.6

Performance of Quintiles

13.7

Total Returns

16.0

12.0

5.6

1.7

4.0

1.6

14.3

10.2

6.9

1.7

5.5

1.4

Market

Source: Bloomberg

1.5

8.7

Price Returns

Dividend Returns

12.9

Quintile-1 seems to have higher beta than Quintile-4 but the returns delivered by

Quintile-1 stocks are lower. This tests the efficacy of EMH, which suggests that in

order to deliver higher returns, one must take higher risk. However, the difference in

beta is not very high, which prompts us to look at the up-capture and down-capture

ratios. These capture up-market and down-market separately, overcoming the

shortcomings of beta.

Exhibit 16: Quintile-1 seems to have higher beta than Quintile-4, but lags on returns

1.8

1.6

1.4

1.2

1.0

0.8

Quintile 1

Quintile 4

Beta of Quintiles

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Up-capture and down-capture

Up-capture measures the percentage of market gains captured by a manager when

the markets are up. Down-capture measures the percentage of market losses

endured by a manager when the markets are down.

The two elements that drive the market are “greed” and “fear”. The up-capture and

down-capture ratios are a useful way of separating the two, so they can be analyzed

independently. Also, up-capture and down-capture address the shortcomings of

beta, which fails to distinguish between up and down markets.

The

upside capture ratio

should be greater than 100, indicating that when the

market is up, the portfolio on an average, has done even better. The higher the up-

capture, the better is the performance of the portfolio.

Alternatively, the

downside capture ratio

should be less than 100, meaning that

when the market is down, the manager has caught only a fraction of the losses. The

lower the down-capture, the better is the performance of the portfolio.

Computation of up-capture and down-capture

For the up-capture, the first step is the identification of all periods in which the

market was up. For those up-market periods, the returns of both the portfolio and

the benchmark are geometrically compounded and then annualized.

Up Capture

= number of positive benchmark returns

= k-th positive benchmark return

= manager return for the same period as the i-th positive benchmark return

= number of years, counting periods of positive benchmark returns only

For the down-capture, the same process is repeated, but for the periods during

which the market was down.

June 2017

Thematic | Contrarian Investing

Exhibit 17: Up-capture and down-capture of quintiles (on quarterly return basis)

160

140

120

Down capture

Source: Bloomberg

100

120

140

Up capture

The neutral stocks (Quintile-4) have high upside capture ratio, but what separates

them from the rest is their superior downside protection. Quintile-4 has the highest

up-capture ratio and the lowest down-capture ratio, which explains the

outperformance of Quintile-4 over all the other quintiles.

The quintiles Q3 and Q5 have the lowest up capture ratios( 87 and 102) respectively

which implies that they rise less/ equal to the market in periods when benchmark

has risen but have substantially higher down capture ratios(111 and 122)

respectively, which implies that these groups perform worse than market in case of

market downturn.

The characteristics exhibited by these quintiles are exactly opposite to

characteristics exhibited by neutral to moderately popular stocks (Q4) having up

capture of 134 and down capture of 108. What’s also quite apparent that, none of

the quintile had a desirable down capture of less than 100!

Exhibit 18: Neutral stocks have superior downside protection

Upside capture ratio

Downside capture ratio

139

121

132

119

111

134

108

102

122

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Quintile-4 is best on risk-reward metric

Additionally, we have calculated average return per standard deviation of returns of

members comprising the quintile. Standard deviation is a measure of risk and

average return/standard deviation can be seen as a measure of return per unit of

risk.

Exhibit 19: Average Return/Standard Deviation for Quintile-1 and Quintile-4

Quintile 1

All Returns

Positive Returns

Negative Returns

0.13

0.65

-0.69

Quintile 4

0.20

0.85

-0.56

Source: Bloomberg

The findings are similar to what we observed in up-capture and down-capture

ratios. Quintile-4 beats all other quintiles on risk-adjusted returns. Its higher return

per standard deviation implies that the risk-reward ratio is better for Quintile-4.

Exhibit 20: Quintile-4 is best positioned on risk adjusted returns basis

2.5

1.5

0.5

-0.5

-1.5

-2.5

Average Returns/Standard deviation

Quintile 1

Quintile 4

Source: Bloomberg

The relative outperformance of the neutral stocks can be explained by the robust

outperformance of these stocks both during up markets and down markets. When

the market is up, they gain the most compared with other quintiles. When the

market is down, they lose the least. The probable reason is that expectations from

these stocks are muted; so, they deliver significant returns in case of an earnings

surprise.

We attribute the success of Quintile-4 (neutral to moderately-popular stocks) to the

fact that this quintile captures stocks just upgraded by the street (from a sell/neutral

consensus rating).

This prompted us to look at all those stocks upgraded by the street, and particularly,

those upgraded

from a “net sell to a net buy” (consensus rating change from <3 to

>3),

as these are the stocks forming Quintile-4 and would be part of Qunitile-4 in the

future as well. Among all the strategies outlined on the report, this strategy has

delivered the best performance.

June 2017

Thematic | Contrarian Investing

Final thoughts: Is consensus change significant?

We looked at instances of change in consensus rating from ‘net sell’ to ‘net buy’ over

2006-2017 across BSE-100 constituents. A change in consensus rating from ‘net sell’

to ‘net buy’ means a change in consensus rating from <3 to >3. We found that a

strategy of investing in stocks with consensus rating change from <3 (net sell) to >3

(net buy) yields better results than any other strategy discussed in this study.

Essentially, these are stocks that (a) have been ignored, (b) have negative news flow

associated with them, or (c) are battling short-to-medium-term challenges. Thus,

typically they have consensus popularity rating <3. However, marginal earnings

surprise, improvement in news flow, and beaten-down valuations are some of the

factors that result in popularity scores moving closer to and then over 3 (Net Buy).

These stocks are yet to become over-owned/most popular names on the street

(Quintile-1 of the popularity thesis discussed in earlier sections).

By investing in stocks for which consensus recommendation has changed from ‘net

sell’ to ‘net buy’ as soon as they have been upgraded, and then holding them for 12

months, an investor would have earned an average return of 24.1%.

Exhibit 21: Indexed return over next 12 months from “Net sell to Net buy” Change

132

124

116

108

100

t+1

t+2

t+3

t+4

t+5

t+6

t+7

t+8

t+9

t+10

t+11

t+12

Source: MOSL, Bloomberg

Avg. Return of stocks(consensus change from Net sell to Net Buy)

124.1

We found this strategy to be effective across the entire time horizon. The lowest

return delivered by the strategy across the time period was -5.7% in 2007, impacted

by 2008 global financial crisis (GFC).

June 2017

Thematic | Contrarian Investing

Exhibit 22: Returns have been significant across all years (2003-2016)

Average Returns

No. of companies

76.3

52.1

32.5

25.6

42.0

2.5

10.9

4.7

18.0

25.9

12.0

-5.7

Source: MOSL, Bloomberg

We highlight that the investment was done after the stock’s popularity rating

changed from <3 to >3, that is, from Net Sell to Net Buy (and not Consensus Buy).

Qualitatively, it suggests that when consensus recognizes the changing fortunes of

underlying business and upgrades ratings, it pays to be among the early birds in

investing in such a business. Our study shows that returns earned from this strategy

are healthy even in absolute terms. However, one needs to move quickly, certainly

before the stock becomes a Consensus Buy and crosses a popularity score of 3.5.

We applied this framework on BSE 100 constituents. Below is a snapshot of stocks

where the consensus popularity score has changed from <3 to >3.

Exhibit 23: Stocks where popularity has improved from <3 to >3, recently

Company

BHEL

Nestle India

Punjab Natl. Bank

Siemens

Idea Cellular

Ambuja Cement

Mar'17

2.4

2.7

2.6

2.2

2.9

2.8

Apr'17

2.4

2.7

2.6

2.4

2.9

May'17

2.2

2.5

2.6

2.8

3.0

June'17

3.0

3.1

3.0

3.2

3.0

Needless to say, this metric has thrown names where a healthy degree of skepticism

exists, and therefore, they do not appear in the list of stocks with fundamentally

positive investment thesis. For example:



BHEL:

Future growth potential of thermal energy in India is being debated even

as BHEL’s slow-moving order book constricts its earnings performance.



Nestle:

Five years of low volume growth and loss in market share in key

categories has hindered the long-term positive narrative around the stock.



PNB:

Asset quality issues hampering the entire PSU Banks pack.



Siemens:

Delay in recovery of private capex cycle.



Idea Cellular:

Internecine price competition in the Telecom sector, which is

witnessing heightened competitive intensity.



Ambuja Cements:

Consistent earnings underperformance vis-à-vis large cap

peers in the sector.

June 2017

Thematic | Contrarian Investing

Cases based on

popularity theme

We looked at the

performance of Quintile-4,

the quintile with neutral to

moderately popular stocks.

Quintile-4 has delivered

significant returns

accompanied by consensus

upgrade. Based on recent

history, we present

Ashok

Leyland, Kotak Mahindra

Bank, Dabur, Exide

Industries, Asian Paints

and Hindalco

as case

studies.

June 2017

Thematic | Contrarian Investing

Case study: 1

Ashok Leyland

Exhibit 24: Popularity of Ashok Leyland

4.5

Exit from 4Q

Entry in 4Q

As the business outlook for Ashok Leyland

improved, so did the consensus ratings. From

a ‘net sell’ in the previous months, the stock

was upgraded to a ‘net buy’ in May 2014. It

entered Quintile-4 (neutral to moderately

popular stocks) in the quarterly rebalancing in

June 2014 and exited the quintile at the end of

September 2015. While it was in Quintile-4, it

significantly outperformed the BSE-100.

4.0

3.5

3.0

2.5

2.0

3.2

3.8

Source: Bloomberg

Exhibit 25: Ashok Leyland – upgrade from ‘net sell’ to ‘net buy’ delivered

Post the downturn in the CV cycle, Ashok

strong returns

Leyland emerged leaner and stronger. It

Net buy(%)

Buy(%)

Ashok Leyland

100

focused on generating cash and curtailing

Sell(%)

Hold(%)

debt by cutting working capital, controlling

capex, and monetizing non-core assets.

Leverage reduced and so did interest costs.

With the CV cycle turning up in 2HFY15, Ashok

Leyland, being a pure play CV player,

-20

Strong

performed well. Its M&HCV market share

-40

increased to a decadal high supported by

operating leverage, lower RM costs and

favorable product mix, Ashok Leyland’s

Source: Bloomberg

EBITDA margin expanded to a 10-quarter high

in 4QFY15. The stock significantly

Exhibit 26: Ashok Leyland – indexed returns

outperformed market indices.

Ashok Leyland

BSE100

290

Indexed Returns

254.3

Stock performance:

Over the 15-month

period, June 2014 to September 2015, while

the BSE-100 delivered 6.3% return, Ashok

Leyland delivered 154.3% return, generating

significant alpha. Even with 5% allocation in an

equal-weighted portfolio of 20 stocks

(Quintile-4 is a group of 20 stocks), it alone

delivered 7.4% alpha.

240

190

140

106.3

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Case study: 2

Kotak Mahindra Bank (KMB)

Exhibit 27: Popularity of KMB

4.5

Kotak Mahindra Bank (KMB) entered Quintile-

4 (neutral to moderately popular stocks) in

September 2013 and remained in this quintile

till June 2016, barring the quarter ended

March 2015. The stock had been a ‘net sell’

for a number of years, barring sporadic

instances. Once it became a ‘net buy’ in

September 2013, it delivered significant

outperformance over BSE-100.

With healthy capitalization (CET1: 17-18%),

improved liability profile, robust risk

management system (stress loans at just 2%),

and strong presence across loan products,

KMB was well placed for the growth cycle. As

macroeconomic conditions improved, KMB

did well – it’s NPAs reduced and Banking

business profit increased, driven by loans and

fee. Additionally, the merger of ING Vysya

Bank (VYSB) with KMB in FY15-16 delivered

strategic benefits. While KMB had a strong

presence in the North and the West (80% of

KMB branches) at that time, it made deep

inroads into the South with the merger (64%

of VYSB branches).

Stock performance:

The stock significantly

outperformed the broader indices during its

consensus upgrade cycle. Over the 21 months,

September 2013 to June 2016, the BSE-100

delivered 53% return; KMB delivered 125%

return. In an equal-weighted portfolio of 20

stocks in Quintile-4, KMB alone contributed an

alpha of 3.6%.

4.0

3.5

3.0

2.5

Entry in Q4

Exit from Q4

3.9

Source: Bloomberg

Exhibit 28: KMB upgrade from ‘net sell’ to ‘net buy’ delivered strong

returns

-10

-30

Net buy(%)

Sell(%)

Buy(%)

Hold(%)

Kotak Mahindra Bank

Strong upgrade

Source: Bloomberg

Exhibit 29: KMB – indexed returns; outperformed BSE-100 by 72% over

September 2013-June 2016

250

210

170

130

Indexed Returns

KMB

BSE100

225.5

153.1

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Case study: 3

Dabur India

Exhibit 30: Popularity of Dabur

4.2

4.0

Exit from Q4

Dabur became a part of Quintile-4 in

December 2014 and exited at the end of June

2016. The stock saw moderate upgrades

during this period and outperformed the

broader market.

‘Net buys’ on the stock reduced significantly

before December 2014 due to slowdown in

consumer demand and then saw moderate

recovery in ‘net buys’, with a couple of ‘holds’

changing to ‘buys’.

During the period December 2014 to June

2016, Dabur’s consistent volume delivery in an

overall weak macro scenario was impressive,

underscoring its portfolio strength and carry-

over benefits of distribution expansion. New

initiatives on distribution ramp-up in the

urban chemist channel and recovery in

Namaste helped earnings grow. This led to

some rating upgrades on the street and the

stock price moved up in tandem with

earnings.

Stock performance:

While the BSE-100 moved

up 3% during this 18-month period, Dabur

moved up 33.4%, significantly outperforming

the benchmark. In an equal-weighted

portfolio of 20 stocks in Quintile-4, Dabur

alone contributed an alpha of 1.5%.

3.8

3.6

3.4

3.2

3.0

3.7

3.5

Entry into Q4

Source: Bloomberg

Exhibit 31: ‘Net buys’ on Dabur increased during the period December

2014 to June 2016

100

-20

Source: Bloomberg

Dabur

Net buy(%)

Buy(%)

Sell(%)

Hold(%)

Exhibit 32: Dabur – indexed returns; outperformed BSE-100 by 30% over

December 2014 – June 2016

140

128

116

104

Indexed Returns

DABUR

BSE100

133.4

103.0

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Exhibit 33: Snapshot of Quintile-4 (neutral to moderately popular stocks)

CMP

Company

Hind. Unilever

Hero Motocorp

Titan Company

M & M Fin. Serv.

Bosch

Bank of Baroda

Bharti Infra.

Dabur India

Colgate-Palm.

Axis Bank

Bajaj Auto

Hind.Zinc

United Spirits

Cipla

Asian Paints

Ambuja Cem.

Tata Power Co.

ACC

Godrej Consumer

Popularity

score

3.18

3.46

3.50

3.51

3.18

3.24

3.13

3.05

3.26

3.19

3.08

3.00

3.25

2.86

2.88

3.00

2.85

2.65

(INR)

1094

3733

520

341

24277

165

375

290

1101

508

2840

253

2294

539

1164

244

1652

960

MCap

(INR B)

2367

746

462

193

762

381

711

511

299

1210

822

1067

333

434

1116

485

223

310

654

Trailing PE (x)

5-yr Avg. Current

40.1

21.2

40.6

18.0

47.7

7.9

29.9

39.3

41.7

14.4

19.6

9.6

110.2

31.0

49.6

29.3

93.2

29.2

40.4

55.7

22.1

57.6

48.1

51.3

27.6

25.2

40.0

51.9

33.1

21.5

12.8

85.8

33.9

55.4

50.0

16.0

49.0

50.8

Prem/Disc

(%)

39.0

4.3

41.8

167.4

7.6

250.0

-15.4

1.7

24.5

129.0

9.8

34.0

-22.1

9.2

11.7

70.8

-82.9

67.6

25.7

FY17

42490

33771

8017

4002

14441

13831

27470

12769

5774

36793

38276

83162

3885

12766

20162

9703

13969

6343

12876

PAT (INR M)

FY18E

49312

39559

9123

7309

19859

43989

32343

13623

6977

56151

43513

91033

5428

16022

22184

13163

18253

8772

14842

FY19E

58367

40188

10784

9269

23692

57488

36882

16110

8450

98609

51569

100156

7530

20071

26318

14298

18951

11021

17001

PAT

CAGR (%)

FY17-19

11.2

6.0

10.4

32.3

17.9

60.8

10.3

8.1

13.5

38.9

10.4

6.4

24.7

16.3

9.3

13.8

10.7

20.2

9.7

ROE

(%)

FY17

65.6

35.7

20.6

6.4

15.8

4.1

16.2

28.4

50.4

6.9

25.3

24.4

21.3

10.2

28.5

5.0

11.2

7.5

24.6

Source: MOSL, Bloomberg

June 2017

Thematic | Contrarian Investing

Consensus sells are significant underperformers

The underperformance of Quintile-5 (the least popular stocks) in the popularity

theme proves that owing to their scarcity value, sell recommendations can be quite

efficient in predicting stock performance. This scarcity value prompted us to look at

consensus sell ratings in isolation.

Once again, we constructed quintiles, this time based on sell recommendations as a

percentage of total recommendations. The intent was to explore how good the

consensus sell recommendations are. We found consensus sell ratings to be quite

right, if we take a longer term horizon!

Quintile-1, composed of stocks with the highest consensus sell ratings,

underperformed the market significantly. So, the consensus has been quite right in

predicting sells.

This can have meaningful implications for investors who can short-

sell stocks.

Exhibit 34: Quintile-1, with highest consensus sell ratings, is the biggest underperformer

500

400

300

200

100

Absolute Total returns(Average

)

Performance of different quintiles of BSE100 by consensus sell rating ( quarterly rebalance)

Q1(CAGR-7.1%)

Q4(CAGR-10.7%)

Q2(CAGR-12.3%)

Q5(CAGR-14.8%)

Q3(CAGR-9.4%)

BSE100 CAGR(10.2%)

409.8

327.1

282.9

269.6

251.7

201.7

Source: Bloomberg

On the other hand, Quintile-5, composed of stocks with the least consensus sell

ratings, performed the best. It is important to note that a portfolio of stocks with

minimum sell ratings may not have many buys – the consensus could be crowded

with hold ratings, making it a portfolio of

neutral to moderately popular stocks.

Additionally, we observed that sell ratings were particularly scarce in 2004-2008.

Both Quintile-4 and Quintile-5 had zero sell ratings. So, we repeated the same

analysis from 2009, when Quintile-4 and Quintile-5 became materially different. The

thesis stayed unchanged!

June 2017

Thematic | Contrarian Investing

With the median measure as the basis of the performance, the results become even

clearer, with Quintile-1 underperforming and Quintile-5 outperforming!

Exhibit 35: Quintile-1 underperformed, while Quintile-5 was the best performer!

300

250

200

150

100

Performance of different quintiles of BSE100 by consensus sell rating ( quarterly rebalance)

Quintile 1

Quintile 3

Quintile 5

Quintile 2

Quintile 4

Absolute Total returns (Median

)

258.4

227.0

232.5

167.7

103.7

Source: Bloomberg

Consensus sell recommendations have reasonable efficacy! We present an example

based on the consensus sell rating theme, and then, put to test the efficacy of

consensus buy ratings.

We present below the snapshot of Quintile 1 of current BSE 100 constituents with

highest SELL Ratings and low popularity score.

Exhibit 36: Ten stocks with highest SELL ratings in BSE 100 – Quintile-1 of Consensus Sell

theme

Company

Divi's Lab.

Dr Reddy's Labs

DLF

TCS

ACC

Godrej Consumer

Marico

Cipla

Asian Paints

Wipro

Buy

Hold

Sell

% of Sell

ratings

56%

51%

50%

48%

44%

41%

40%

Popularity

score

2.44

2.42

2.14

2.63

2.65

2.60

2.86

2.88

2.58

The list is dominated by companies from IT, Pharma and Consumer sectors. While IT

and Pharma are facing regulatory as well as business model evolution challenges,

the Consumer sector is characterized by peak valuations.

June 2017

Thematic | Contrarian Investing

Case study: 4

Steel Authority of India (SAIL)

Exhibit 37: SAIL has been consistently downgraded over the last five years

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

Entry in Q1

Member till date

SAIL is a member of Quintile-1 from July 2011

till date. The stock has consistently been a net

sell and the consensus’ prediction of stock

performance has been correct.

Over 2003-13, a significant portion of SAIL’s

earnings was driven by its captive iron ore

advantage. As iron ore prices (and resultantly

steel prices) have since declined globally,

SAIL's competitive advantage has been

eroded. Its employee cost per ton and

operating cost are higher. Slow ramp-up of

capacities has also impacted SAIL's

performance. Eroding competitive advantage,

higher operating cost and weak project

execution are driving SAIL’s

underperformance.

Stock performance:

As apparent, SAIL has

significantly underperformed the benchmark

BSE100 Index (-49.3% vs. 81.7%).

Source: Bloomberg

Exhibit 38: SAIL – consensus rating has been “strong sell” for recent and

distant past

100

-50

-100

Net buy(%)

Sell(%)

Buy(%)

Hold(%)

SAIL

Source: Bloomberg

Exhibit 39: Sail has consistently underperformed indices

220

170

120

Indexed Returns

SAIL

BSE 100

181.7

50.7

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Consesus buys are not always great investments

To test the efficacy of consensus buy ratings, we constructed quintiles based on

buys as a percentage of total recommendations. We looked at the performance of

the quintiles and found that consensus buys are not necessarily great investments.

Quintile-4, the quintile with the second-lowest (in terms of percentage) consensus

buy ratings, delivered the maximum returns! Quintile-4 was followed by Quintile-2,

which was followed by Quintile-1. Quintile-4 has been the front runner, almost

through the entire decade!

Exhibit 40: Quintile-1 does well, but there is not much difference across top-4 quintiles

450

350

250

150

Absolute Total returns(Average

)

Performance of different quintiles of BSE100 by consensus sell rating ( quarterly rebalance)

Q1(CAGR-11.7%)

Q3(CAGR-10.5%)

Q5(CAGR-7.1%)

Q2(CAGR-11.8%)

Q4(CAGR-14.6%)

BSE100(CAGR-10.2%)

404.9

314.1

309.5

278.1

269.6

201.6

Source: Bloomberg

So, contrarian investing does works with consensus buy recommendations. With

Quintile-4 emerging as the winner both in the popularity theme and the consensus

buy theme, we checked for how many stocks they have in common. We found the

number of common members in both these portfolios to be significantly high.

Exhibit 41: Number of common members across Quintile-4 (popularity) and Quintile-4

(consensus buy) is significantly high (65-75%)

Common members across Q4 of popularity and consensus buy theme

17.5

16.0

14.5

13.0

11.5

10.0

Source: Bloomberg

In a market characterized by preponderance of BUY calls, it is not difficult to find

names enjoying tremendous popularity on the street.

June 2017

Thematic | Contrarian Investing

We present below the snapshot of BSE 100 constituents that are top-most

Consensus BUY calls today.

Exhibit 42: Ten stocks with highest BUY ratings in BSE 100 – Quintile-1 of Consensus Buy

theme

Company

ITC

Aurobindo Pharma

Vedanta

HDFC Bank

Federal Bank

M&M

NTPC

Power Grid Corpn

ICICI Bank

Hindalco Inds.

Buy

Hold

Sell

% of Buy

ratings

93%

92%

91%

89%

87%

86%

84%

Popularity

score

4.76

4.79

4.64

4.63

4.60

4.57

4.53

4.48

4.61

June 2017

Thematic | Contrarian Investing

Summary: Performance of the quintiles

A review of the performance of the quintiles highlights that the central theme based

on popularity is the most effective way of creating quintiles. It meaningfully

separates the winners from the underperformers.

As evident, popularity is the best selection criterion amongst the three themes

(popularity, consensus sell and consensus buy) for creation of quintiles. It separates

the index members amongst the quintiles efficiently, which can be seen in

significant return differential across quintiles. If the selection criterion is efficient,

quintiles will show variation in returns otherwise the quintile returns will show little

difference.

The consensus sell theme comes second, owing to scarcity value of sell ratings,

whereas the consensus buy theme comes last.

The results seem fit, as popularity is an aggregate measure of sell, buy and hold

recommendations and should convey more information than sell/buy

recommendations in isolation.

Exhibit 43: Average returns (CAGR; 2006-16) across quintiles by strategy

Strategy

Popularity

Consensus Sell

Consensus Buy

Quintile 1

14.6

7.1

11.7

Quintile 2

13.7

12.3

11.8

Quintile 3

5.6

9.4

10.5

Quintile 4

16.0

10.7

14.6

Quintile 5

6.9

14.8

7.1

Source: Bloomberg

What’s particularly interesting is that in the long term, if we decide to select an

entire basket of stocks (quintile) based on any strategy, even the worst performing

quintile would beat the returns from a savings bank account.

June 2017

Thematic | Contrarian Investing

Annexure: Focus stocks – Quintile-4 (popularity theme winners)

Consensus BUY (%)

Consensus SELL (%)

Consensus HOLD (%)

Net Buy (%)

Exhibit 44: Axis Bank

Net buy(%)

100

Buy(%)

Sell(%)

Hold(%)

Exhibit 45: Bajaj Auto

Net buy(%)

100

Buy(%)

Sell(%)

Hold(%)

Source: Bloomberg

Exhibit 46: Colgate

Net buy(%)

140

100

-20

-60

Buy(%)

Sell(%)

Hold(%)

Exhibit 47: Dabur

100

Net buy(%)

Buy(%)

Sell(%)

Hold(%)

Source: Bloomberg

Exhibit 48: Godrej Consumer

Net buy(%)

100

Buy(%)

Sell(%)

Hold(%)

Exhibit 49: Hero Motors

Net buy(%)

-10

-40

Buy(%)

Sell(%)

Hold(%)

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Contrarian strategies based on valuation multiples

While the first section of this note dealt with contrarian strategies based on analyst

recommendations, this part deals with the contrarian strategies based on valuation

multiples.

In his classical book,

Contrarian Investment Strategies: The Psychological Edge,

David Dreman

suggested, “Favored stocks underperform the market while out-of

favor companies outperform the market, but the reappraisal often happens slowly,

even glacially”. Empirical evidence from the US markets suggests that the companies

the market expects the best future for – as measured by

price-to earnings, price-to

cash-flow, price-to-book-value –

have consistently done the worst, while the stocks

believed to have the most dismal future have done the best. We will see whether

this holds true in an emerging market like India.

June 2017

Thematic | Contrarian Investing

Theme 1: Based on price-to-earnings (P/E) ratio

For every quarter starting with the quarter ended December 2006, we’ve divided

the BSE-100 constituents into quintiles based on P/E ratio and evaluated the returns

delivered by each quintile. Quintile-1 is composed of the highest P/E stocks, while

Quintile-5 has the lowest P/E stocks.

Conclusion:

We found that low P/E stocks outperformed the high P/E stocks by a

huge margin. The margin of outperformance increased with trailing P/E as the

criterion for forming the quintiles instead of 12-month forward P/E.

P/E

Exhibit 50: Low P/E stocks outperformed, whereas high P/E stocks failed to beat the market

600

450

300

150

High PE(CAGR-7.5%)

Q3(CAGR-13.4%)

Low PE(CAGR-16.6%)

Performance of different quintiles of BSE100

Q2(CAGR-10.5%)

Q4(CAGR-6.3%)

BSE100(CAGR-10.2%)

483.8

364.2

269.6

277.4

210.8

187.4

Source: Bloomberg

Theme 2: Based on price-to-book-value (P/B) ratio

For every quarter starting with the quarter ended December 2006, we’ve divided

the BSE-100 constituents into quintiles based on P/B ratio and evaluated the returns

delivered by each quintile. Quintile-1 is composed of the highest P/B stocks, while

Quintile-5 has the lowest P/B stocks.

Conclusion:

Over the decade, low P/B stocks have underperformed the high P/B

stocks on forward multiples and outperformed high P/B stocks on trailing basis. The

alpha generated by low P/B stocks was much lower than other stocks. However, low

P/B stocks led all the other quintiles on the returns chart for the majority of the

decade.

P/B

Exhibit 51: Returns from BSE-100 and all quintiles barring Quintile-4 are quite close

450

350

250

150

Absolute Total returns(Average

)

High PB(CAGR-13.8%)

Q3(CAGR-10.8%)

Low PB(CAGR-12.5%)

Q2(CAGR-12.4%)

Q4(CAGR-6.9%)

BSE 100(CAGR-10.2%)

Performance of different quintiles of BSE100

376.8

330.6

333.2

287.3

269.6

198.6

Source: MOSL, Bloomberg

June 2017

Thematic | Contrarian Investing

Low P/E stocks outperform high P/E stocks

Benjamin Graham’s “Intelligent Investor”

cites a study involving the constituents of

Dow Jones Industrial Average.

The performance of the 10 lowest P/E stocks, 10

highest P/E stocks, and of all the 30 stocks in the index was measured over set

periods between 1937 and 1969. In each time span, the low P/E did better than the

market and the high P/E stocks did worse.

In their legendary paper,

“Cross-section of Expected Stock Returns”,

Journal of

Finance, 1992,

Eugene Fama and Kenneth French

reported results similar to

Benjamin Graham’s findings. Several studies before and after the Fama-French study

also point towards the outperformance of out-of-favor stocks.

The findings of our study are similar. For every quarter starting with the quarter

ended December 2006, we divided the BSE-100 constituents into quintiles based on

one-year forward P/E ratio and evaluated the returns delivered by each quintile.

Quintile-1 has the highest P/E stocks, while Quintile-5 has the lowest P/E stocks.

Here’s what we found:



INR1k invested in the low P/E quintile in December 2006 would be worth

INR4.84k today.



INR1k invested in the high P/E quintile in December 2006 would be worth

INR1.87k today.



INR1k invested in the BSE-100 in December 2006 would be worth INR2.77k

today.

With remarkable consistency, investors misjudged subsequent performance. The

results are completely uniform. The most favored stocks by the market with highest

PE ratio continuously shows dismal returns, whereas the least favored stocks with

lowest PE ratio were rewarded by the market.

Exhibit 52: Low P/E stocks outperform market, whereas high P/E stocks fail to beat the market

600

450

300

150

High PE(CAGR-7.5%)

Q3(CAGR-13.4%)

Low PE(CAGR-16.6%)

Q2(CAGR-10.5%)

Q4(CAGR-6.3%)

BSE100(CAGR-10.2%)

Performance of different quintiles of BSE100

483.8

364.2

269.6

277.4

210.8

187.4

P/E

Source: Bloomberg

Investors/analysts have consistently overvalued the favorite stocks and undervalued

the out-of-favor stocks! This has worked the same way across a long time period

(Benjamin Graham’s study focused on the period, 1937-1969, Fama-French’s study

June 2017

Thematic | Contrarian Investing

focused on the period, 1962-1989, and our study covers 2006-2016) and across

geographies (the first two studies focused on the US and we focused on India).

Exhibit 53: The results are better with trailing P/E ratio

650

500

350

200

Absolute Total returns(Average

)

Performance of different quintiles of BSE100

High PE(CAGR-4.2%)

Q3(CAGR-10.9%)

Low PE(18.5%)

Q2(CAGR-9.9%)

Q4(CAGR-6.8%)

BSE100(CAGR-10.2%)

571.1

288.5

269.6

263.7

195.3

152.3

Source: Bloomberg

We repeated the same exercise with trailing P/E ratios. The results were even

better, bringing to the fore two interesting observations:

1. Trailing valuation multiple better predictor:

Low P/E stocks based on trailing

earnings returned 18.5%, generating an alpha of 8.3% over the BSE-100

benchmark on annual basis, almost 1.7% more than low P/E stocks based on

forward estimates. Returns from high P/E stocks, which failed to even beat the

benchmark, were dismal on both trailing and forward earnings.

The market is slow in re-appraisal of low P/E stocks.

Over the last 10 years, on

an average, almost 63% of the members across low P/E quintiles based on

trailing and forward earnings are common. On the other hand, just 20% of the

stocks are common across high P/E quintiles on trailing and forward earnings,

suggesting that re-appraisal happens much faster but erroneously.

Exhibit 54: Market is slow in re-appraisal of low P/E stocks, but quick for high P/E stocks

-1

Common Low PE Stocks

Common High PE Stocks

53.3

26.7

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Case study: 5

Indian Oil Corporation (IOC)

Exhibit 55: IOCL beats BSE-100 by a significant 59.1% over March-

December 2016

IOCL became a member of the low P/E quintile

in March 2016 and exited at the end of

December 2016, delivering a return of 67.6%

against the BSE-100’s 8.1% return during the

nine months.

IOCL is the largest refining company in India,

with ~43% market share in petrol and ~48%

market share in diesel. It has the largest

network of product pipelines, terminals and

depots, giving it a pan-India reach.

While HPCL and BPCL got re-rated in 2014

post deregulation, IOCL has gone through a

de-rating because of the concerns on the large

capex of INR350b (almost one-third of the

market cap at that time). Now that there is

more visibility on the Paradip project, with

85% utilization in March 2017, the stock has

started to re-rate. It still trades at 5.6x FY19E

EV/EBITDA (MOSL estimates) and 9.4x FY19E

EPS. Global peers trade at 10.8x one-year

forward EPS and 6.8x one-year forward

EV/EBITDA, with lower RoE of 13% (IOCL’s

RoE: ~20%).It has however exited from the

low P/E quintile.

The 15mmtpa Paradip refinery, with a Nelson

Complexity of 12.2 would have higher GRM

than IOCL’s vintage refineries and is expected

to contribute to the company’s profits. With

successful commissioning of the Paradip

refinery, IOCL would also have larger

availability of products in the South, which it

traditionally lacked.

180

160

140

120

100

Indexed Returns

IOCL

BSE 100

167.6

108.1

Source: Bloomberg

Exhibit 56: Trailing P/E of IOCL

IOCL PE(x)

10.2

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Case study: 6

Rural Electrification Corporation

Exhibit 57: RECL has stayed ahead of BSE-100 for the majority of the

timeframe

270

220

170

120

RECL is a member of the low P/E quintile

from June 2011 and has delivered 154%

returns, beating the BSE-100 by 73%. It has

been consistently valued at lower multiple

by the market and it has consistently

outperformed the market. There have been

instances across these 5-6 years when

returns from the BSE-100 were higher, but

for most of the time, RECL outperformed.

POWF is another stock with the same story;

it too has been outperforming the market.

RECL is the government’s nodal agency for

financing rural electrification programs. It

has strategic importance in funding power

infrastructure in India. It has good capital

adequacy, with healthy capitalization,

diversified resources, and stable

profitability.

RECL has delivered decent PAT CAGR of

17.1% over FY11-16, but the stock has

always been a member of the low P/E

quintile. Additionally, RECL offers good

dividend yield, which provides protection to

investor returns.

There have been concerns, too.

Implementation of the UDAY scheme would

result in loan book/NIM compression for

RECL. Loan book has stayed flat due to

prepayment under UDAY. However,

disbursements were up 26% YoY in FY17.

RECL trades at a huge discount to its peers.

Indexed Returns

RECL

BSE 100

254.0

181.7

Source: Bloomberg

Exhibit 58: Trailing P/E of RECLRs

RECL PE(x)

6.1

Source: MOSL,

June 2017

Thematic | Contrarian Investing

Exhibit 59: Low P/E stocks have generated significant alpha across all time horizons

200

150

100

High PE

Low PE

BSE100

Source: Bloomberg

Average Price appreciation of quintiles across different time horizons

1-year

2-year

3-year

5-year

7-year

The low P/E stocks offer higher dividend yield (2.8%) than the market (1.5%), which

acts as a support for stock performance in a downturn. On the other hand, stocks

with high P/E ratios offer lower dividend yield (0.68%) than the market. Worse, they

fail to deliver the growth they promise based on higher PE ratios!.

Exhibit 60: High dividend yield supports performance of low P/E stocks in bear phases

Perfomance of Quintiles

Total Return

Price Return

Dividend Return

18.5

15.7

9.9

8.9

10.9

9.1

2006-2017(curr.)

10.2

4.2

6.8

1.7

8.7

1.5

3.5

0.7

1.1

4.8

1.9

2.8

BSE 100

Source: Bloomberg

Low P/E quintile has the highest up-capture ratio (156), driven by low valuations and

it beats the other quintiles by a significant margin (next best is Quintile-3: 106).

However, what aids the performance of these stocks is that along with the

stupendous up-capture, they also have a satisfactory down-capture!

Exhibit 61: Low P/E stocks have fantastic up-capture, with a decent down-capture

Up capture

180

160

140

Down capture

120

100

Source: MOSL, Bloomberg

100

120

140

Winner

Up Capture 156

Down Capture 119

June 2017

Thematic | Contrarian Investing

The high P/E stocks perform miserably, with the highest down-capture (124) during

a downturn and the lowest up-capture (91) during an upturn.

Exhibit 62: Up-capture and down-capture of quintiles

Upside capture ratio

Downside capture ratio

124

100

106

104

119

156

119

Source: MOSL, Bloomberg

Exhibit 63: Fifteen stocks from our universe with lowest Trailing P/E

Company

Power Fin.Corpn.

Rural Elec.Corp.

HPCL

BPCL

IOCL

ONGC

NMDC

NTPC

Hindalco Inds.

Tech Mahindra

Reliance Inds.

JSW Steel

Tata Steel

HCL Technologies

Power Grid Corpn

CMP

(INR)

129

182

512

642

394

161

109

159

192

387

1433

197

515

849

202

MCap

(INR B)

340

359

521

842

1911

2060

346

1310

396

381

4643

477

500

1199

1055

Trailing PE (x)

5-yr Avg.

6.2

5.5

21.2

15.5

14.1

11.8

12.0

11.3

39.3

14.8

11.9

28.2

8.5

18.4

14.0

Current

5.0

5.8

8.4

8.9

9.6

9.8

11.0

12.2

11.8

12.5

13.4

13.3

13.9

14.2

Prem/Disc (%)

-19.5

6.2

-60.5

-43.0

-31.8

-17.3

-8.4

8.3

-69.9

-15.1

12.8

-52.7

63.6

-22.9

0.8

PAT CAGR (%)

FY17-19

5.6

8.8

-9.2

-1.0

1.9

13.4

8.6

7.6

16.9

6.1

6.2

15.1

21.6

3.4

13.0

Popularity

RoE (%)

Score

FY17

3.5

17.9

3.4

19.9

3.4

32.4

3.7

32.4

4.0

22.3

4.4

10.4

2.8

12.4

4.7

11.5

4.4

14.0

3.9

18.4

4.1

11.9

3.8

17.3

3.8

15.4

4.1

27.5

4.7

16.2

Source: MOSL, Bloomberg

Exhibit 64: Fifteen stocks from our universe with highest trailing P/E

CMP

Company

Ambuja Cem.

M & M Fin. Serv.

Marico

Havells India

Colgate-Palm.

Godrej Consumer

Bosch

Asian Paints

Hind. Unilever

Nestle India

Titan Company

BHEL

Siemens

ABB

United Breweries

(INR)

244

341

314

474

1101

960

24277

1164

1094

6773

520

137

1353

1486

781

MCap

(INR B)

485

193

404

296

299

654

762

1116

2367

653

462

335

482

315

207

Trailing PE (x)

5-yr Avg.

29.3

18.0

41.7

28.6

41.7

40.4

47.7

49.6

40.1

92.2

40.6

20.6

90.9

99.1

102.9

Current

50.0

48.1

49.9

49.6

51.9

50.8

51.3

55.4

55.7

57.4

57.6

67.6

75.9

75.4

89.9

PAT CAGR (%) POPULARITY

ROE (%)

Prem/Disc (%) FY17-FY19

70.8

13.8

167.4

32.3

19.6

10.2

73.1

14.8

24.5

13.5

25.7

9.7

7.6

17.9

11.7

9.3

39.0

11.2

-37.7

5.8

41.8

10.4

228.4

34.7

-16.5

23.2

-23.8

17.7

-12.7

19.2

SCORE

FY17

3.0

5.0

3.4

6.4

2.9

36.7

3.0

18.2

3.3

50.4

3.2

24.6

3.5

15.8

3.2

28.5

3.9

65.6

2.5

39.0

3.5

20.6

2.2

1.5

2.6

9.3

2.1

12.7

2.8

10.4

Source: MOSL, Bloomberg

June 2017

Thematic | Contrarian Investing

Analyzing stocks based on premium/discount w.r.t history



After establishing the premise that low PE stocks outperform high PE stocks, we

went a step forward of analyzing stocks based on their PE discount/premium

w.r.t history. This is because we understand that there are certain sectors like

consumer stocks that have outperformed index even when they have traded on

higher valuations since 2008 or certain sectors always trade at lower multiples

like OMC’s, so looking at stock’s valuation discount/premium w.r.t history can

also be quite meaningful indicator.

For every quarter starting with the quarter ended December 2010, we divided

the BSE-100 constituents into quintiles based on discounts/premiums to their

trailing ratios and evaluated the returns delivered by each quintile. Quintile-1

has the highest P/E stocks, while Quintile-5 has the lowest P/E stocks.

Exhibit 65: Low P/E stocks outperform market, whereas high P/E stocks fail to beat market

250

200

150

100

Absolute Total returns(Average

)

High PE(CAGR-4.2%)

Q3(CAGR-10.9%)

Low PE(18.5%)

Q2(CAGR-9.9%)

Q4(CAGR-6.8%)

BSE100(CAGR-10.2%)

Performance of different quintiles of BSE100

218.4

177.7

170.2

168.0

153.5

126.8

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Annexure: Focus stocks - Low P/E quintile

Exhibit 66: Wipro trailing PE

WIpro PE (x)

Exhibit 67: Coal India trailing PE

Coal India PE (x)

15.3

17.8

Source: Bloomberg

Exhibit 68: Divi’s Laboratories trailing PE

DIVI PE (x)

Exhibit 69: Tata motor trailing PE

Tata motors PE (x)

21.7

14.6

Source: Bloomberg

Exhibit 70: Tech Mahindra trailing PE

Tech Mahindra PE (x)

Exhibit 71: RECL trailing PE

RECL PE(x)

6.1

12.0

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Low P/B stocks dominate, but overall results mixed

For every quarter starting with the quarter ended December 2006, we divided the

BSE-100 constituents into quintiles based on P/B ratio and evaluated the returns

delivered by each quintile. Quintile-1 is composed of the highest P/B stocks, while

Quintile-5 has the lowest P/B stocks.

The most favored stocks, with the highest P/B ratios (Quintile-1) delivered highest

returns (CAGR of 13.8%), followed by the low P/B quintile (CAGR of 12.5%) followed

by Quintile-2 (CAGR of 12.4%) and Quintile-3 (CAGR of 10.8%). Among all the

quintiles, only Quintile-4 failed to beat the benchmark.

The winner quintile (high P/B) did not generate as much alpha as the winner quintile

of the previous theme (low P/E stocks). So, over the last 10 years, forward P/B ratio

has not been as effective as P/E ratio in segregating the winners from losers.

Cross-section of Expected Stock Returns,

Journal of Finance, 1992,

Eugene Fama

and

Kenneth French

highlighted the success of low P/B stocks over high P/B stocks.

In India, this has not worked as well as themes based on P/E ratios. However, low

P/B stocks have dominated other quintiles for most of the decade under

consideration.

Exhibit 72: Returns of BSE-100 and all quintiles barring Quintile-4 are quite close

450

350

250

150

Absolute Total returns (Average

)

High PB(CAGR-13.8%)

Q3(CAGR-10.8%)

Low PB(CAGR-12.5%)

Q2(CAGR-12.4%)

Q4(CAGR-6.9%)

BSE 100(CAGR-10.2%)

Performance of different quintiles of BSE100

376.8

330.6

333.2

287.3

269.6

198.6

P/B

Source: Bloomberg

We have repeated the same analysis with trailing P/B ratio. The low P/B stocks

based on trailing BV did better than low P/B stocks based on forward BV, whereas

the high P/B stocks did somewhat worse. Again, low P/B stocks led the other groups

for the majority of the decade and are the winners.

June 2017

Thematic | Contrarian Investing

Exhibit 73: Trailing P/B ratio also presents the same picture

450

350

250

150

Absolute Total returns(Average

)

High PB(CAGR-11.7%)

Q3(CAGR-7.8%)

Low PB(CAGR-13.7%)

Performance of different quintiles of BSE100

Q2(CAGR-12.8%)

Q4(CAGR-8.8%)

BSE 100(CAGR-10.2%)

371.4

342.3

311.7

269.6

215.8

238.2

Source: Bloomberg

However, the results are again much less appealing as compared to the P/E theme.

In our previous theme, we have done our analysis on BSE-100 excluding banks, as

the price-earnings ratio is less relevant for banks. Also, private banks in general have

been one of the biggest wealth creators, even at relatively high valuations. We want

to be sure that the above result has not been affected by this exclusion of banks.

So, we repeated the same analysis on BSE-100 excluding banks. The results were

similar, with low P/B stocks dominating, but high P/B stocks have also delivered

significant returns. Similar results were achieved with forward P/B ratios.

June 2017

Thematic | Contrarian Investing

Summary: Performance of contrarian themes

A review of the performance of contrarian themes based on relative valuation

metrics suggests that the P/E ratio is a better predictor of returns as compared to

P/B ratio.

Exhibit 74: Price to Earnings> Price to Book in stock selection

625

500

375

250

125

Indexed Returns

Low PE

Low PB

Perfromance of different themes

BSE 100

571.1

324.8

269.6

Source: Bloomberg

It separates the index members among the quintiles efficiently, which can be seen in

the significant return differential across quintiles.

If the selection criterion is efficient, the quintiles show variation in returns.

Otherwise, there is little difference in the quintile returns, which seems to be the

case with P/B. Hence, the P/B theme ranks the last in our themes.

Exhibit 75: Returns (CAGR 2006-17(current)) by quintile under each strategy

Strategy

Trailing PE

Trailing PB

Quintile 1

4.19

11.73

Quintile 2

9.92

12.76

Quintile 3

10.89

7.79

Quintile 4

6.75

8.83

Quintile 5

18.53

13.66

Source: Bloomberg

What’s particularly interesting is that in the long term, if we decide to select an

entire basket of stocks (quintile) based on any strategy, even the worst-performing

quintile would beat the savings account rate.

June 2017

Thematic | Contrarian Investing

Analysis

sectors on our

framework

June 2017

Thematic | Contrarian Investing

Metals: A contra bet of CY16



Metals & mining was one of the best performing sectors in FY17, delivering

returns (Nifty Metals index) of 61.9% and an alpha over the benchmark Nifty of

18.5%.

The strong performance was primarily driven by a recovery in earnings (off the

bottom) and aided by re-rating of multiples. Earnings for our metals pack (in

EBITDA terms) grew 68% YoY in FY17, driven by higher prices across

commodities and an increase in volumes.

The sector is a living demonstration of the Contrarian Buy hypothesis

highlighted in this report. The sector traded at a discount and its popularity hit

the bottom in early-2016 when metal prices were at historical lows.

Valuations profile

Metals are trading at 6.7x EV/EBITDA, almost at a 10% discount to their long-term

averages. Metals are generally counter-cyclical – they trade at a premium at the

bottom of the earnings cycle and at some discount when the cycle is at its peak.

This, however, was not the case in early-2016, when commodity prices were at

their historical lows and the sector was available at a discount to its long-term

average.

The de-rating was due to concerns about sustainable demand growth for

commodities in China and deflation, in our view.

Exhibit 76: Metals trade at 7% discount to historical average on EV/EBITDA basis

Metals Sector EV/EBDITA (x)

13.5

10.5

7.5

4.5

1.5

4.1

7.3

6.8

LPA (x)

11.2

Source: Bloomberg

The outlook has since improved, as commodity prices have recovered from their

lows, demand is growing at a steady pace in China and also led by a few supply-side

initiatives. This is evident from the consensus rating profile of the sector.

The sector is now available at close to its long-term average multiple. In our view,

the future performance will continue to be driven primarily by growth in earnings –

partly on higher prices and also on improving utilization. A re-rating will be a factor

of stern actions to curb overcapacity in China and developments around

infrastructure investments in the US.

June 2017

Thematic | Contrarian Investing

Sector consensus rating profile

Exhibit 84-85 shows how the popularity of the metals sector has improved gradually

after bottoming out in December 2015/January 2016, while Exhibit 145 shows the

change in the net buy position.

Exhibit 77: Metals – popularity

100

Metals

Net buy(%)

Buy(%)

Sell(%)

Hold(%)

Source: Bloomberg

Exhibit 78: Metals – popularity with analysts has increased, currently close to LTA

Popularity rel. to history

4.0

3.8

3.6

3.4

3.2

3.0

Metals popularity

Popularity

Average

0.5

0.3

0.0

-0.3

-0.5

Source: Bloomberg

As apparent, metals’ popularity was at bottom in Dec 15-Mar 16, and it was a

contrarian call (in hindsight), which would have worked out very well.

Our top picks

Hindalco

is our top pick in the sector, as (a) its India business has the competitive

advantage of low-cost raw materials and a growing demand market, (b) Novelis is

generating healthy cash and has attractive growth opportunities with the rising

adoption of aluminum in automobiles (Novelis, being a first mover and with strong

customer relations, has an edge) and (c) capex cycle is behind, and with strong FCF

generation, the focus is on deleveraging. We value the stock on 6.5x FY18E

EV/EBITDA, at INR242/share.

June 2017

Thematic | Contrarian Investing

Pharma: Ripe for a contrarian bet

FY17 was a tough year for the pharmaceuticals and healthcare sector, delivering

returns of -5.26% (Nifty Pharma & Healthcare index), as against the benchmark

Nifty’s 18.5% returns. The underperformance has continued in FY18. The Indian

pharma industry finds itself engulfed in challenges posed by:

Pricing pressure in the US generics market

Investment in R&D in generic and specialty businesses at all-time high – returns

to be visible in 2-3 years

Issues pertaining to US FDA inspection and resolution

Promotion of generic drugs in India – implementation will not be easy

We analyzed the sector through our Contrarian Investing framework, looking at

analyst ratings, valuation profile, etc.

Sector consensus rating profile

Pharma sector popularity with consensus is at its lowest level in several years

Exhibit 79: Pharma sector popularity – Buy ratings lowest since 2010

Pharmaceuticals

Net buy(%)

Sell(%)

Buy(%)

Hold(%)

Source: Bloomberg

Exhibit 80: Pharma sector popularity with analysts is below long-term average

5.0

4.0

3.0

2.0

1.0

Popularity rel. to history(rhs)

Popularity

Average

Pharmaceutical popularity

0.4

0.2

0.0

-0.2

-0.4

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

Valuations profile

These factors weighed on valuations, which have corrected significantly over the

past two years. The sector trades at 20.2x, at a 10% discount to its long-term

average.

Exhibit 81: Pharma sector trading at 10% discount to LPA (on P/E basis)

13.0

22.3

20.2

Healthcare Sector PE (x)

LPA (x)

34.8

Source: Bloomberg

Earnings growth took a hit in FY17, coming in below long-term average growth. We

expect earnings growth to remain subdued in FY18 as well.

Exhibit 82: Earnings growth likely to remain subdued…

Pharma sector EPS Growth (%)

42.2

28.2

30.0

15.7

19.3

4.0

14.1

10.5

8.0

Source: Bloomberg

However, several pharma companies now appear cheap on PEG basis (FY19).

Exhibit 83: …however several pharma companies appear cheap on PEG basis

1.6

0.9

1.2

0.9

1.1

0.8

PEG

0.9

0.4

0.5

1.2

0.4

1.2

2.0

1.5

1.4

0.9

Source: Bloomberg

June 2017

Thematic | Contrarian Investing

In our view, the Indian pharma sector fits well with the tenets of Contrarian

Investment. Performance of the sector over the past two years has been subdued

and news flow remains adverse. Amid this tumultuous phase, valuations have come

off and the sector now trades at a discount to the long-period average. Popularity of

the sector is also at its lowest in many years. The sector has seen a series of earnings

cuts in the last 18 months and its weightage in the key indices has shrunk given the

underperformance. Even from an institutional ownership perspective, the

healthcare sector weight in the Indian Mutual Funds portfolio has reached a new

low of 6.1% (-50bp MoM). As a result, the sector has fallen to the ninth position in

sector allocation of mutual funds – it was at the fifth position 12 months ago.

Fundamentally, while we do not expect the sector’s fortunes to change overnight,

we believe it no longer suffers from an adverse combination of ‘high expectations

and premium valuations’. Valuations now offer an attractive entry opportunity from

a medium- to long-term perspective, in our view, even as expectations have

moderated.

Our top picks:

Sun Pharma, Aurobindo Pharma, Fortis Healthcare

June 2017