Investing with Ulysses · | 020 3941 2801 | [email protected] 15% MSCI World Equity Annualised Returns Sell date, quarterly

www.oxfordrisk.com | 020 3941 2801 | [email protected]

Greg B Davies, PhD@GregBDavies

Investing with UlyssesBetter Decisions Through Behavioural Design


Which investor is happiest at the end?

Port

folio

Ret

urn

Time


Double your money in 10 years … in three different ways


<-5%-5% to

0%

0 to 5%

5 to 10%

10 to 15% >15%

MSCI World Equity Annualised Returns

Sell date, quarterly

Buy date,quarterly

If you bought in 1975 and sold in 1995, your average annual return would have

been 12%



<-5%-5% to

0%

0 to 5%

5 to 10%

10 to 15% >15%

MSCI World Equity Annualised Returns

Sell date, quarterly

Buy date,quarterly

No nominal losses over any holding period of 12 years

or more

Zone of Anxiety



Exuberance

Reluctance

DenialExcitement

Optimism

Panic

Fear

Despondency

Desperation

Capitulation

Reluctance

Depression

Apathy

Indifference


Paying too much for emotional comfort


Po

rtfo

lio R

etu

rn

Time

Danger of selling lowDanger of buying high

Behaviour Gap

The costs of being human



Po

rtfo

lio R

etu

rn

Time

Behaviour Gap

Emotional Insurance

The costs of being human




Knowledge Intention Comfort Examples

Disclosure Little, or none None NoneCaveat emptor

Disclaimers

Traditional Education

Little, or none Little, or none Little, or noneSeminars

Classes

Nudges None, or negative None, or negative SomeAuto-enrolment

Defaults

Engaged Choice

Yes Yes Yes

Just in time education

Gamification

The road to decision prosthetics



Better decisions by design


Knowledge

Environment ControlDecision Prosthetics


Alternative X

• Lose $4,000

Alternative Y

• 1/3 chance nothing lost

• 2/3 chance $6,000 lost

67% go for X

Alternative A

• Recover $2,000

Alternative B

• 1/3 chance $6,000 recovered

• 2/3 nothing recovered

92% go for A

Would you choose A or B?

Source: Wang, 1996

Imagine you bought $6,000 worth of stock from a now bankrupt company

There are two alternatives to recover money…

Would you choose X or Y?

Why choice/gamble based “revealed preferences” don’t work



-60%

-40%

-20%

0%

20%

40%

60%

80%

Dec

-19

79

Dec

-19

84

Dec

-19

89

Dec

-19

94

Dec

-19

99

Dec

-20

04

Dec

-20

09

10 Year

Framing horizon

MSCI World Index, 1970 – Dec 2013


The information we should base decisions on


-60%

-40%

-20%

0%

20%

40%

60%

80%

Dec

-19

79

Dec

-19

84

Dec

-19

89

Dec

-19

94

Dec

-19

99

Dec

-20

04

Dec

-20

09

1 Year 10 Year

Framing horizon

MSCI World Index, 1970 – Dec 2013


The information we do base decisions on


Decision Constitution

1. Circle of competence

2. Automated actions

3. Contingency plans

4. Cooling off period

Make more decisions ahead of time



Initiate

Plan Deviate

Terminate

Plan Repeat

Reality


No strategy survives first contact with reality


Example investor financial personality recommendations

Low Composure

• Reduce reporting detail and frequency

• Portfolio construction to smooth journey

High Familiarity Bias

• Home biased portfolios or familiar assets

• Draw attention to specific assets in reporting

Low Involvement

• Provide lots of decision support and interaction

• Reduce reporting detail and frequency

High Impulsivity

• Focus narrative away from market conditions

• Increase focus on spending plans and goals

Composure

Familiarity Bias

Involvement

Impulsivity


March 1918


The sweet spot for better decisions

Data Digital

Design

Grant me:

The serenity to accept what computers do better than people,

The courage to let people do what they do better than computers, and…

The wisdom to know the difference.


www.oxfordrisk.com | 020 3941 2801 | [email protected] | 020 3941 2801 | [email protected]


Oxford Risk1 Paris GardenLondon, SE1 8ND, UK

+44 (0) 203 [email protected]



Ris

kP

rofi

le

Retirement

How risk capacity affects risk profile over time

Static risk tolerance

Dynamic risk capacityEarly in life: risk capacity justifies increase from RT

In retirement: risk capacity implies decrease from RT


Subjective assessment

• Adviser or client self-assessment

• Inaccurate and unstable

• Procyclical

• Lacks robust regulatory evidence

“Games”

• Engaging, but gimmicky

• No evidence of accurate assessment

• No basis in academic evidence

“Revealed” preferences

• Often tests numeracy, not Risk Tolerance

• Doesn’t measure RT, but in-the-moment attitudes

• Highly unstable

• Procyclical

Data-driven psychometric

• Strong academic and theoretical grounding

• Empirically validated and stable

• Regulatory comfort

There are many poor approaches to assessing risk preferences

Documents

Investing with Ulysses · | 020 3941 2801 | [email protected] 15% MSCI World Equity Annualised Returns Sell date, quarterly