Income Protection – managing the

cycles

Ashutosh Bhalerao, Luv Bhatnagar,

Phin Wern Ting

© < ClearView>

This presentation has been prepared for the Actuaries Institute 2018 Financial Services Forum.

The Institute Council wishes it to be understood that opinions put forward herein are not necessarily those of the

Institute and the Council is not responsible for those opinions.

Agenda

1. Current state of play

2. Are economic factors an ‘underestimated’ driver of IP claims

cost?

3. How can we better manage the cycle?

Setting the scene

0

500

1,000

1,500

2,000

2,500

3,000

30-Sep-07 30-Sep-12 30-Sep-16 30-Sep-17

Retail Income Protection - Inforce Premium ($'m) **

CAGR ~8.5%

** Strategic Insight 10 Year Review, 2007-2017

Is the Retail IP

segment growing?

Setting the scene

➢ Retail IP profit margins (pre tax) / Net premium rolling

-40.0%

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

Jun 2009 Dec 2009 Jun 2010 Dec 2010 Jun 2011 Dec 2011 Jun 2012 Dec 2012 Jun 2013 Dec 2013 Jun 2014 Dec 2014 Jun 2015 Dec 2015 Jun 2016 Dec 2016 Jun 2017 Dec 2017

Profit margins (pre-tax) / Net Premium Rolling 12 months

What’s happening

with profitability?

** APRA quarterly statistics

1. Volatility

2. Systemically

under-priced

3. Potential long

term trend?

Only 2 periods

with profit

margins > 10% !

Setting the scene

➢ Industry has attributed this to a ‘sustainability issue’

➢ Benefit design issues (e.g. one-duty, 10 hour rule, able to work in WP, generous built in terms which could be optional etc.), no doubt have a claims cost effect. But, why are they causing a deterioration in the last 10 years?

➢ Growing number of mental illness claims.

➢ In our view the adverse experience is more driven by economic conditions (e.g. Unemployment, under-employment, slow wage growth) and the extent to which these have had an impact on Income Replacement Ratios.

A model of drivers of IP experience

IPClaims

Cost

Un

em

plo

y-m

en

t

Un

de

rem

-p

loym

en

tSociety

Evolution(C21)

Nature of work(Gigs,

knowledge, etc)

MentalStress/

strength

TrendCyclical

Unemployment rate

decreasing but

underemployment

trending upwards

Wage growth

across all

industries tracking

CPI (in some

periods below,

including recent

periods)

What’s the current economic landscape?

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

Mar 2009 Aug 2010 Dec 2011 May 2013 Sep 2014 Jan 2016 Jun 2017

Economic factors

Unemployment rate Underemployment rate Wage Growth CPI

What does this mean for the IP product?

➢ Does IP claim cost have an economic linkage?

➢ Pro-cyclical vs anti-cyclical?

➢ Performance will vary by sector – can be a 2 speed economy!

➢ Is there an occupation linkage? What about Self-employed individuals?

➢ Benefit indexation has over the recent periods outpaced wage growth.

➢ What do these factors do to Income Replacement Ratios at policy inception vs point of claim?

But first, what can we do statistically?

70%

80%

90%

100%

110%

120%

130%

140%

88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13

Claims cost % FSC-KPMG ADI 2007-2011

ADI 2007-2011

Source: 2002 Report of the Disability Committee, FSC-KPMG Disability Income Experience Investigation 2009-2013

IAD 89-93

(re-based to ADI)

No industry data

Multivariate regression

Regression equation:

Claims costs AvE = - 0.05 + 22.8 x Underemployment – 1.4 x Unemployment – 8.6 x % ∆ AWE

Adjusted R-squared = 68%Significance of model F-statistic = 0.03 (< 0.05)

Correlation with Claims cost AvE

Underemployment rate 79%

Unemployment rate 37%

% change in AWE -10%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Wa

ge

Gro

wth

Un

em

plo

ym

en

t a

nd

un

de

rem

plo

ym

en

t

Economic indicators

Underemployment rate Unemployment rate % change in AWE

Source: ABS

Slow wage growth

Rising underemployment

80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Actual vs Predicted claims cost AvE

Predicted Actual

Observations

➢ What does this mean for claims cost?

➢ Declining NPAT, rising loss ratios from APRA statistics

➢ The statistics suggest an anti-cyclical trend?

➢ The exposure to sector and occupation will vary by insurer

➢ Lag or lead?

Incidence Termination

Employed Lag? Lag?

Self-employed Lead? Lag?

Further considerations

➢ Model limitations

➢ Lags in economic indicators

➢ Incidence versus termination experience

➢ Other studies

– Australia & South Africa

• Increase in claims incidence when unemployment rises

How have income replacement ratios moved over time?

➢ Lack of data to get this information directly.

➢ To answer this, we built a model with the following simplifying assumptions:

➢ Incomes for new policies have increased by AWE each year

➢ Anti-selective wage growth: -1% drift on inforce wage growth

➢ Initial replacement ratio is 75% for each year of sale

➢ Portfolio lapse rate of 12.5% p.a.

➢ Sum insured indexation take-up rate is 85%

➢ Average sum insured indexation rate is 4% p.a.

➢ Partial lapse rate (i.e. sum insured decrease) is 0.5% p.a.

How have income replacement ratios moved over time? Key takeaways

➢ IRRs have increased in recent times due to slow wage growth

➢ Claims will be linked to IRRs , which would explain some of recent deterioration

70.0%

71.0%

72.0%

73.0%

74.0%

75.0%

76.0%

77.0%

78.0%

79.0%

80.0%

Inco

me

Rep

lace

men

t R

atio

Year

Income Replacement Ratio - for portfolio

Portfolio average

How do we better manage the cycles?

Key aspects are…..

➢ Data

➢ Reinsurance Strategy

➢ Reserving

Data

➢ Ideally would track income replacement ratios (IRRs) including:

➢ IRRs at policy commencement versus claim time

➢ Estimated IRR for portfolio currently (broken by sector) and how this may change in

future

➢ However, currently difficult to get this as:

➢ Policy commencement - Financial evidence only always requested for guaranteed

agreed value (above limits for agreed value and indemnity)

➢ Renewal – no data collected

➢ Claim Time – income information only collected for indemnity (highest average in

last 3 years) and agreed value (highest average from one year pre policy

inception).

Would a bank not know the LVR on its loan portfolio?

Reinsurance

Economy improves, profits

made, insurer and reinsurer

prices drop

Traditional Pricing

and Reinsurance

makes the system

more volatile!

Economy worsens, losses made,

insurer and reinsurer prices

increase

Alternative Reinsurance Strategy

➢ Build automatic repricing in the treaty:

➢ Profit in the current year results in reinsurance rate dropping (like an automatic

credibility adjustment)

➢ Loss in the current year results in reinsurance rate increasing

➢ For example:

➢ Gross Premium p.a. = $200

➢ Reinsurance premium = $100 p.a.

➢ In Year 1, say actual reinsured claims are $60, compared to $85 expected (i.e. $25

profit for the reinsurer)

➢ Average duration is 10 years

Impact of $25 Profit on Reinsurance Premiums – Year 1

New Reinsurance Premium (Year 3) = $100 -$2.5= $97.5

Change in Reinsurance Premium = -$25/10=-$2.5

Original Reinsurance

Premium (Year 1) = $100

Years

$25

Profit

Impact of $10 Loss on Reinsurance Premiums – Year 2

$10

Claims

Loss

Original Reinsurance Premium (Year 2) = $97.5

Change in Reinsurance Premium = $10/10=-$1

New Reinsurance Premium

(Year 2) = $97.5 +$1 = $98.5

Years

$25

Profit

• Direct insurer pricing based on long term expected claims and reinsurance premiums

• Reduced reinsurance premiums from good times help pay for adverse experience in economic downturn => helps reduce pricing volatility for direct insurers

Reserving and Pricing➢ Assumptions should be more dynamic

➢ Allow for mean reversion:

➢ Incidence and Termination

➢ Allow for expected changes in IRRs based on forecast wage growth, underemployment

etc.

➢ Industry has failed to do this in previous cycles

A model of drivers of IP experience

Cyclical

Assumptions

Mean

reversion

Reserving

Reinsurance

Game theory

IP

Claims

Cost

Un

em

plo

y

-me

nt

Un

de

rem

-

plo

ym

en

t

Long term

Trends

Product

positioning

Claims mngt

RTW/Health

Early

intervention

Price changes

(up and down)

inevitable

Product

management

over time

So what??➢ IP experience impacted by:

➢ Cyclical factors– economic drivers such as wage growth, underemployment =>

resulting impact on IRRs

➢ Long term trends – mental illness, nature of work

➢ Benefit definitions have not changed, yet experience has deteriorated

➢ There are opportunities for better cycle management

➢ Allowing for mean reversion in pricing and reserving

➢ Reinsurance structuring

➢ We need to collect better data as an industry!!