The Case for a Risk - First Approach to Stock Selection - - PowerPoint PPT Presentation

▶
the case for a risk first approach
SMART_READER_LITE
LIVE PREVIEW

The Case for a Risk - First Approach to Stock Selection - - PowerPoint PPT Presentation

Dec 12, 2022 214 likes •465 views

The Case for a Risk - First Approach to Stock Selection Presentation to CFA Society of Dayton April 13, 2016 www.RevelationIR.com 219-213-2531 Time for a Change in Research Approach? Active management is losing its appeal to

slide-1
SLIDE 1

The Case for a ‘Risk-First’ Approach to Stock Selection

Presentation to CFA Society of Dayton April 13, 2016

www.RevelationIR.com 219-213-2531

slide-2
SLIDE 2

Time for a Change in Research Approach?

  • Active management is losing its appeal to investors
  • Past performance hasn’t justified higher fees
  • Traditional stock selection research still focuses on earnings
  • Earnings expectation game is entrenched and played by all parties -

companies, analysts, portfolio managers

  • Quantitative factor models, once an alternative approach, have also

become commonplace

  • Stock mispricing ‘anomalies’ are well known
  • Modern Portfolio Theory is the accepted framework for managing the

trade-off between expected returns and expected risk

  • Investors and their agents are more short-term and risk averse
  • Key Takeaway: How might an active equity manager evolve in

response to these trends, and maybe even take advantage of them?

2

slide-3
SLIDE 3

Purpose of this Presentation

  • Revelation Investment Research believes that active equity

managers must evolve to thrive

  • Many possibilities, but there’s a particular approach we believe offers

tremendous promise.

  • Consider Benjamin Graham’s statement: “The essence of portfolio

management is the management of risks, not the management of

  • returns. Well-managed portfolios start with this precept.”
  • In this presentation, we provide nine reasons why adding a

risk-first, ‘what could go wrong?’ research perspective can:

  • Enhance equity performance
  • Enhance client satisfaction
  • Enhance business results
  • We also illustrate one method for adding a risk avoidance

perspective to your equity investment process

3

slide-4
SLIDE 4

Let’s Frame the Stock Valuation Challenge

  • Consider the textbook Fair Value equation:

𝑄𝑠𝑗𝑑𝑓 = ෍

𝑢=1 ∞

[𝐷𝐺

𝑢 / 1 + 𝑠 𝑢]

  • Most researchers focus on the numerator: estimating future cash

flow or earnings levels, growth rates, and ‘surprises’

  • Few researchers pay much attention to the denominator: estimating

a discount rate that reflects the uncertainty of future cash flows

  • Yet changes in risk perception move stock prices just as directly as

changes in growth expectations

  • Key Takeaway: Could a risk-first approach to stock selection

be beneficial?

4

slide-5
SLIDE 5

Risk-First Stock Selection: Why?

  • 1. Most stocks underperform – “Creative destruction” at work
  • According to a JP Morgan study (“Eye on the Market”, Sept 2014) of

Russell 3000 members 1980 – 2014

  • 64% of individual stocks underperformed the index
  • 40% of all stocks had negative absolute returns
  • 40% of all stocks experienced “catastrophic losses” (defined as a 70% drop from

peak price with no recovery above 60% blow peak price)

  • According to recent studies of S&P 500 Index members
  • From 2000-2014, an average of 180 stocks each year had negative absolute

returns (source: S&P Dow Jones Indices)

  • Since 1980, 320 S&P 500 members were removed from the index for business

distress reasons (source: JP Morgan)

  • Key Takeaway: Avoiding big losers can improve an active

manager’s performance

5

slide-6
SLIDE 6

Risk-First Stock Selection: Why?

  • 2. Return volatility hurts wealth compounding
  • ‘Volatility drag’ can be estimated using: G = 𝐵 − (𝑇2/200)
  • Where G = Annual Geometric (compounded) Return, A= Annual Arithmetic

Return, S = Standard Deviation of Annual Return

  • For strategies with the same average return, lower volatility

produces higher compounded return

  • Key Takeaway: Reducing return variability alone can improve

long-term portfolio returns

6

Impact of Volatility on Return Compounding

Investment Strategy Avg Return Stdev of Returns Geometric Return 20 Year Growth of $1000 1 (high volatility) 10% 22% 7.58% $4310 2 10% 18% 8.38% $5000 3 10% 14% 9.02% $5630 4 (low volatility) 10% 10% 9.50% $6140

slide-7
SLIDE 7

Risk-First Stock Selection: Why?

  • 3. Even professional investors are ‘loss averse’, not just risk

averse as finance theory dictates

  • Loss aversion can trigger two related and damaging behavioral

tendencies

  • Anchoring – tendency to add cost basis (something unknown to the market and

irrelevant to a stock’s prospects) as an input to hold vs sell decisions

  • Disposition effect – tendency to sell winners too soon (so they don’t become

losers) and hold losers too long (in hope they recover and become winners)

  • Key Takeaway: Investing in lower risk stocks can reduce the

probability of losses on individual positions and at the portfolio level, reducing a common cause of bad decisions

7

slide-8
SLIDE 8

Risk-First Stock Selection: Why?

  • 4. Loss aversion can lead employers and clients to make ill-

timed business decisions that are damaging to you and/or your firm

  • Riskier strategies (by chance alone) are more likely to produce larger

short-term losses or longer runs of underperformance

  • Riskier strategies tend to perform worst in down markets, when

investor/decision-maker loss aversion is highest

  • Key Takeaway: Reducing investment strategy risk, especially

in down markets, can reduce business risk and career risk

8

slide-9
SLIDE 9

Risk-First Stock Selection: Why?

  • 5. Research that focuses on stocks’ growth potential can lead

to excessive optimism and overconfidence, which can trigger more damaging behavioral tendencies:

  • Lottery effect – extreme payoffs tend to influence decisions more

than their low probability of occurrence

  • Representativeness bias – tendency to see unwarranted familiarities

(this stock is the next _________)

  • Confirmation bias – tendency to accept information that supports

the original decision to buy and to discount conflicting information

  • Endowment effect – tendency for investors to place a higher value on

what they own than non-owners do

  • Key Takeaway: Greater focus on ‘what could go wrong?’ can

add new perspective and help prevent overvaluing upside potential

9

slide-10
SLIDE 10

Risk-First Stock Selection: Why?

  • 6. Forecasting earnings levels and growth rates is difficult and

largely ineffective

  • Analyst EPS forecasts are highly inaccurate and biased
  • 45% of reported quarterly EPS deviate by more than 5% from consensus

forecasts from 2001-2015

  • Consensus 5Y EPS Growth Rate forecasts typically average 10-15% annually,

while stocks’ actual EPS growth has averaged 6-8% annually

  • Analyst EPS growth forecasts have little stock selection usefulness
  • The 20% of stocks with the highest forecasted 1Yr EPS Growth have lagged by

2.0% annually from 2001 - 2015

  • The 20% of stocks with the highest forecasted 5Yr EPS Growth have lagged by

2.8% annually from 2001 – 2015

  • Key Takeaway: Focusing on the numerator of the Fair Value

equation is a challenging approach to finding mispriced stocks

10

slide-11
SLIDE 11

Risk-First Stock Selection: Why?

  • 7. Investing based on earnings-related ‘factors’ – another

approach linked to the Fair Value equation numerator – produces mixed results

  • Key Takeaway: earnings-related metrics are insufficient for

consistently finding mispriced stocks

11 Best 20% of Stocks Based on Screening Variable Below

Avg 12M Return vs Universe Avg 12M Volatility vs Universe

FY1 EPS / Price

2.1% 0.1%

PE / Estd 5Y EPS Growth

1.1% 1.5%

Earnings / Sales

0.5%

  • 1.6%

Earnings Quality (ie, accruals)

  • 2.0%

4.4%

Last 3M EPS Estimate Revisions

1.4% 0.5%

Last Qtr EPS Surprise

  • 0.1%

2.1%

Last 4Q EPS Growth

  • 0.7%

3.2%

Average

0.3% 1.5%

(Top 2300 Mktcap Universe, 2001-2015)

Earnings Screen Historical Performance

slide-12
SLIDE 12

Risk-First Stock Selection: Why?

  • 8. By contrast, even some simplistic risk-related factors have

been highly effective stock selection tools

  • For example, low volatility stocks have consistently outperformed,

while high volatility stocks have underperformed

  • Key Takeaway: Focusing on the denominator of the Fair Value

equation may be a more productive path to excess returns12

  • 40%
  • 30%
  • 20%
  • 10%

0% 10% 20% 30% 40% 1/01/1987 1/01/1989 1/01/1991 1/01/1993 1/01/1995 1/01/1997 1/01/1999 1/01/2001 1/01/2003 1/01/2005 1/01/2007 1/01/2009 1/01/2011 1/01/2013 1/01/2015

Avg 24month Buy & Hold Excess Return% 20% Lowest Volatility Stock Portfolio 20% Highest Volatility Stock Portfolio

slide-13
SLIDE 13

Risk-First Stock Selection: Why?

  • 9. If many equity investors have a similar research focus, e.g.,
  • Looking for reasons to buy a stock, not for reasons to avoid a stock
  • Forecasting EPS growth, not EPS uncertainty
  • Actively participating in the quarterly EPS reporting game
  • Managing portfolio risk, not stock-specific risk
  • Constructing portfolios to match the benchmark risk level
  • Are opportunities being created for users of a different approach?
  • Could earnings-related stock selection metrics be increasingly overused?
  • Could stocks held in benchmark-tracking portfolios be a ‘crowded trade’?
  • Could stocks’ absolute risk level be mispriced?
  • Could a research focus on downside risk prediction reveal new alpha factors?
  • Key Takeaway: A ‘risk-first’ research focus may enhance returns

and increase your strategy’s differentiation in the marketplace

13

slide-14
SLIDE 14

Risk-First Stock Selection: How?

  • Let’s now move on from the ‘why’ to the ‘how’
  • Every equity manager would likely approach this challenge in

a different way, asking questions such as:

  • What risk(s) am I trying to avoid? Over what horizon?
  • Should I use a different research methodology?
  • Do I plan to integrate this risk viewpoint into my current

process/model or use it as an overlay?

  • How does limiting stock-specific risk impact portfolio risk?
  • Today, I’ll share seven RIR suggestions and their rationale

14

slide-15
SLIDE 15

Risk-First Stock Selection: How?

  • 1. Goal: avoid ‘negative outcomes’
  • Riskier stocks must appear to offer higher returns or no one would buy

them, but as risk increases so does the probability of bad outcomes

  • What are ‘negative outcomes’ we wish to avoid?
  • Stocks that underperform or actually lose money
  • Portfolios that underperform, actually lose money, or perform inconsistently
  • Disappointed clients that withdraw assets
  • Disappointed bosses who demote/fire subordinates

15

Bad outcomes

Chart Source: Howard Marks, Oaktree Capital Management, LP

slide-16
SLIDE 16

Risk-First Stock Selection: How?

  • 2. Research objective: build a tool that predicts negative

investment outcomes

  • Negative investment outcomes => 𝑔 𝑞𝑠𝑝𝑐𝑏𝑐𝑗𝑚𝑗𝑢𝑧 ∗ 𝑞𝑏𝑧𝑝𝑔𝑔 ∗ 𝑢𝑗𝑛𝑓
  • Probability - identify a high proportion of stocks that underperform in a defined

time period

  • Payoff - identify stocks that experience the largest losses in a defined time period
  • Time - identify poorly performing stocks period-after-period through time
  • Secondary objective: build a tool works best in down markets when

loss aversion peaks

16

slide-17
SLIDE 17

Risk-First Stock Selection: How?

  • 3. Time horizon: focus on predicting negative outcomes over a

long time horizon of 1-2 years

  • For many stock selection criteria, predictive power, consistency, and

significance are greater at longer horizons

17

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 1M 3M 12M 24M Average IC Holding Period

Predictive Power Increases with Holding Period

(Generic Style Models from Oct 2014 RIR Product Usage Brief, MSCI IMI members, 2001-2014)

Risk Aversity Model Momentum Model Growth Model Value Model Quality Model

slide-18
SLIDE 18

Risk-First Stock Selection: How?

  • 4. Criteria selection: utilize diverse criteria that address the

Fair Value equation’s numerator and denominator

  • Risk level indicators – measures of overall risk perception and

vulnerability to bad news

  • Financial statement metrics that gauge financial leverage, earnings quality, and

fundamental stability

  • Investor sentiment metrics that gauge valuation, expected EPS growth, and stock

price volatility

  • Risk change indicators – measures of real and perceived risk changes

(i.e., ‘catalysts’)

  • Changes in financial statement metrics such as EPS, cash flow generation,

external financing, asset utilization, and capital intensity

  • Changes in investor sentiment metrics such as stock price, analyst forecasts, short

interest, and trading volume

18

𝑄𝑠𝑗𝑑𝑓 = ෍

𝑢=1 ∞

[𝐷𝐺

𝑢 / 1 + 𝑠 𝑢]

slide-19
SLIDE 19

Risk-First Stock Selection: How?

  • 5. Criteria formulation: consider using Boolean scoring as

alternative to cross-sectional factors

  • Boolean variables can be expressed as simple questions
  • Is P/E > 20?
  • Is Operating Margin < 1 yr ago?
  • Is the most recent EPS surprise < 0?
  • Boolean variables have many desirable qualities
  • Intuitive; transparent
  • Flexible; avail new data items
  • Surprisingly powerful (see Jan 2015 RIR Research Brief)
  • Can be combined into a checklist – a proven decision-making tool in other fields
  • Cross-sectional factor variables capture more information, but are

more complex to formulate and interpret

19

slide-20
SLIDE 20

Risk-First Stock Selection: How?

  • 6. Model formulation: create a composite score for each stock

by aggregating the Boolean 0 or 1 scores across the input criteria

  • Composite score interpretation
  • The more strikes against a stock, the more likely that stock will underperform

(have greater more downside risk)

  • Forecasting power comes from breadth of the input criteria
  • Advantages of scoring models
  • Provide a transparent, easy-to-use checklist appraisal of a stock’s prospects
  • Tend to be statistically robust

20

slide-21
SLIDE 21

Risk-First Stock Selection: How?

  • Example of a Downside Risk Prediction checklist

1. FY1 EPS / Price < 4%? 2. Free Cash Flow / Enterprise Value < 2%? 3. Mean 5Y EPS Growth Forecast > 15%? 4. ROE < 5%? 5. Short Interest / Shares Out > 5%? 6. 4Q Change in Shares Outstanding > 1%? 7. 4Q Sales Growth < 5%? 8. 4Q EPS Growth < 0%? 9. Last 3M EPS Forecast Revisions #Up - #Down < 0? 10. 6M Stock Price Chg < Market? 11. Annualized Stock Price Volatility > 30% 12. 6M Change in Short Interest Ratio > 1%?

  • Consider each ‘Yes’ answer to be an ‘Alert’
  • Summing the Alerts creates a downside risk score for each stock

21

slide-22
SLIDE 22

Risk-First Stock Selection: How?

  • Downside Risk Prediction checklist – Simple and Effective!
  • Stocks with 6 or more “Alerts” underperform
  • Stocks with 5 or fewer “Alerts” outperform – the absence of

downside risk alerts is a positive!

  • Note also that excess returns expand as the predictive horizon lengthens

22

  • 5.0%
  • 4.0%
  • 3.0%
  • 2.0%
  • 1.0%

0.0% 1.0% 2.0% 1 2 3 4 5 6 7 8 9 >=10

Number of Downside Risk 'Alerts'

Downside Risk Checklist Performance

(3M excess returns, largest 2000 stks, 2002-2016)

  • 14.0%
  • 12.0%
  • 10.0%
  • 8.0%
  • 6.0%
  • 4.0%
  • 2.0%

0.0% 2.0% 4.0% 6.0% 1 2 3 4 5 6 7 8 9 >=10

Number of Downside Risk 'Alerts'

Downside Risk Checklist Performance

(12M excess returns, largest 2000 stks, 2002-2016)

slide-23
SLIDE 23

Risk-First Stock Selection: How?

  • 7. Implementation: consider using the Downside Risk checklist

as a complimentary overlay to your existing process

  • As an “analyst”
  • Highlight potential weak performers in portfolios and their specific warning signs

for further research

  • Review/veto/break ties among stocks liked by your current investment process

from a ‘devil’s advocate’ perspective

  • As a screening tool
  • Exclude risky stocks from your selection universe
  • Highlight ‘high return/low risk’ stocks for further research
  • As a sell discipline
  • Eliminate stocks most likely to underperform from portfolios
  • Be careful that your stock-specific risk avoidance doesn’t introduce

undesirable relative risk exposures to your portfolio

23

slide-24
SLIDE 24

Risk-First Stock Selection: Summary

  • Avoiding ‘losers’ is as important to equity portfolio

performance as finding ‘winners’

  • Changes in risk perception move stock prices just as directly

as changes in growth expectations

  • A risk-first approach to stock selection can help investors:
  • Identify mispriced stocks
  • Identify stocks most vulnerable to risk reassessments
  • Avoid negative investment, business, and career outcomes
  • Next step: If you agree that a risk-first perspective might

enhance your stock selection process, we invite you to learn more by visiting www.RevelationIR.com or contacting us

24

www.RevelationIR.com 219-213-2531