MORTALITY ASSUMPTIONS:

THE IMPORTANCE OF GETTING

THEM RIGHT, AND THE

CHALLENGES TO DOING SO BUDAPEST PENSION SEMINAR 2017 Longevity Risk and Social Security Budapest, 21 September 2017

Jessica Mosher, Policy Analyst, Financial Affairs Division

Mortality Assumptions and Longevity Risk: Implications for pension funds and annuity providers (2014)

Technical assistance for the development of the regulatory mortality tables for Chile (2016) and Peru (2017)

The fragmentation of retirement markets due to differences in life expectancy, Business and Finance Outlook (2016)

Ongoing work on the implications of differences in longevity across socioeconomic groups and different arrangements for the management of longevity risk

2

Recent OECD work on longevity and

pensions

Why are accurate mortality assumptions important for pensions and annuities?

What are the findings and policy recommendations from an assessment on the mortality assumptions used in practice internationally?

What are some of the challenges to setting accurate mortality assumptions and effectively managing longevity risk?

What are some implications for pension policy?

Questions addressed today

3

WHY ARE ACCURATE MORTALITY ASSUMPTIONS

IMPORTANT?

4

Mortality Assumptions • Needed to determine how long pension/annuity

payments are expected to be made – The level of mortality today

– The improvement in mortality tomorrow (i.e. the expected increase in life expectancy)

Longevity Risk • The risk that individuals will live longer than

assumed, and therefore that pension/annuity payments will have to be made for a longer period than planned for

5

Some definitions…

Idiosyncratic risk (aka volatility)

• Fluctuations around the actual average mortality assumed

• Managed through pooling/law of large numbers

Level risk

• The risk that mortality today is lower than expected

• Managed through assumption setting and actual-to-expected analysis

Trend risk

• The risk that mortality improvements will be higher than expected

• Systemic risk much more difficult to manage

• Expected risk – The extent to which mortality assumptions are in line with

expectations to address expected improvements in life expectancy

• Unexpected risk – The impact of additional unexpected increases in life expectancy

6

Longevity risk

Must first have a reasonable estimation of expected pension/annuity liabilities to be able to assess the impact of unexpected increases in longevity to measure risk exposure

• The financial impact of a 25% decrease in mortality will not be the same if assumptions include no improvements (Scenario 2)

7

Mortality assumptions and quantifying

unexpected longevity risk

Future pension payments using different mortality assumptions

8

Underlying population and mortality

improvements matter

16 18 20 22 24

Canada (CPM)

France (TGH/F 05)

Germany (DAV 04, 2nd orderAgg Target)

Israel (Pension BE)

Mexico (EMMSA 09)

Netherlands (AG-Prognosetafel)

Spain (PERM/F P)

Switzerland (BVG 2010)

UK (SAPS 2)

US (RP2014)

Chile (2014)

Male Life Expectancy Age 65

Population

Pensioner/Insured - Period

Pensioner/Insured - Cohort

16 18 20 22 24 26 28

Canada (CPM)

France (TGH/F 05)

Germany (DAV 04, 2nd orderAgg Target)

Israel (Pension BE)

Mexico (EMMSA 09)

Netherlands (AG-Prognosetafel)

Spain (PERM/F P)

Switzerland (BVG 2010)

UK (SAPS 2)

US (RP2014)

Chile (2014)

Female Life Expectancy Age 65

9

Lower interest rates increase exposure to

longevity risk

FINDINGS AND POLICY IMPLICATIONS OF AN

ASSESSMENT OF MORTALITY ASSUMPTIONS USED BY PENSION FUNDS AND ANNUITY PROVIDERS

10

Metric: compare annuity values based on assumed and projected mortality

• Expected differences in the current provisioning needed to meet future payments

Forming expectations: quantitative outputs and qualitative judgement

• Projection models

– Lee Carter, Cairns-Blake-Dowd, P-spline and CMI models

• Model Calibration

– General population mortality

• Interpretation of results

– Reasonableness of projections given historical context

11

Quantifying the expected longevity risk

Classification

Expected Longevity

Risk

Pension Plans Annuity Providers

Serious 10-20%

Brazil (US 1983IAM), China (CL2000-2003), Switzerland (EVK2005)

Brazil (US Annuity 2000), China (CL2000-2003)

Significant 5-10%

Canada (UP94-ScaleAA), Japan (EPI2005), US (RP2000-ScaleAA)

Moderate 2-5% Chile (RV2009), Spain (PERM/F C 2000)

Brazil (BR-EMS 2010), Canada (GAM94-CIA), Chile (RV2009), Spain (PERM/F C 2000) US (GAM94-ScaleAA)

Monitor

<2%; specific issues to address

Canada (CPM), France (TGH/F 2005), Israel, Mexico (EMSSA 1997), Spain (PERM/F P 2000) Switzerland (BVG 2010, VZ 2010), US (RP2000-ScaleBB)

France (TGH/F 2005), Israel, Mexico (EMSSA 2009), Japan (SMT 2007), Spain (PERM/F P 2000)

OK

little to no expected shortfall

Netherlands (AG-Prognosetael 2010), UK (SAPS1-CMI), UK (SAPS2-CMI), US (RP2014-MP2014)

Germany (DAV 2004 R), Netherlands (AG-Prognosetael 2010), Switzerland (ERM/F 2000), UK (PCMA/PCFA 2000-CMI)

12

Expected longevity risk of standard mortality

tables

1. The regulatory framework should ensure that pension funds and annuity providers use appropriate mortality tables to account and provision for expected future improvements by establishing clear guidelines for the development of mortality tables used for reserving for annuity and pension liabilities.

2. Governments should facilitate the measurement of mortality for the purposes of assumption setting and the evaluation of basis risk of index-based hedging instruments.

3. The regulatory framework should provide incentives for the management and mitigation of longevity risk.

4. Governments should encourage the development of a market for instruments to hedge longevity, particularly index-based instruments, by facilitating transparency and standardization of longevity hedges in order to ensure the capacity for pension plans and annuity providers to continue to provide longevity protection to individuals.

13

Summary of Policy Implications E

xp

ecte

d r

isk

U

nex

pec

ted

ris

k

Tables should account for the expected future improvements in mortality

• Analysis showed that tables which do not account for improvement risk having a shortfall of provisions of over 10%

• For countries assessed, accounting for mortality improvements add 2-2.5 years of life expectancy at age 65 on average

Tables should be regularly updated

• This will ensure tables are in line with recent mortality experience and limit the impact of reserve increases

Tables should be based on the relevant population

• Life expectancy and pensioner/annuitant mortality can vary significantly from one country to the next and across various sub-groups of the population

14

1) Establishing guidelines for mortality tables

Accurate and timely mortality data should be available

• Could be used to inform mortality assumptions and keep them up-to-date

Mortality data should be provided with a socio-economic indicator

• Level of education and income is significantly correlated with life expectancy

15

2) Facilitating the measurement of mortality

CHALLENGES TO GETTING MORTALITY ASSUMPTIONS

RIGHT

16

Differences in the level of mortality across sub-populations • Significant difference in mortality across

socioeconomic groups

Lack of sufficient data for mortality improvement assumptions • Base on general population

External shocks may change underlying demographics • 2008 pension reform in Chile

• 2014 pension freedoms in the UK

• 2016 lump-sum allowance in Peru

17

Challenges to establishing mortality

assumptions

Pensioner mortality had improved at a much lower rate than the population

• Pensioners have experienced much lower ‘measured’ mortality improvements on average

18

Chilean example: demographic challenges

-1,0%

0,0%

1,0%

2,0%

3,0%

4,0%

5,0%

45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85

Age

Male

Implied Projected

Disabled population had no measured mortality improvement

• Mortality has increased significantly over last several years

19

Chilean example: demographic challenges

0,00

0,01

0,02

0,03

0,04

0,05

0,06

55 56 57 58 59 60 61 62 63 64

Age

1999 2000 2001 2002 2003 2004 2005

2006 2007 2008 2009 2010 2011 2012

Pensioner mortality had improved at a much lower rate than the population

• This was due to the increased coverage for lower income individuals

Disabled population had no measured mortality improvement

• This was due to a change in the definition of disability, removing the 3 year waiting period to qualify for permanent disability

20

Chilean example: 2008 pension reforms

15%

20%

25%

30%

35%

199

9

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

1st quartile

2nd quartile

3rd quartile

4th quartile

United Kingdom

• In 2012, 25% of annuities sold to existing customers were for pension pots less than 5,000 GBP

• In July-September 2015, 90% of individuals not taking a guaranteed annuity rate had pension pots less than 10,000 GBP

Peru

• In 2016, pensioners now allowed to withdraw 95.5% of their accumulated pension assets as a lump sum

• Annuity sales fell by 30% in 2016 compared to 2015

21

Another demographic shock:

change in payout rules

Demographic changes

• Changes in coverage for the public pension scheme

• Selection effect for those who remained in the DC system

• Skewed towards which socioeconomic group?

22

Potential challenges given the design of the

Hungarian pension system

IMPLICATIONS FOR PENSION POLICY

23

Facilitate the measurement and management of longevity risk

• Timely and accurate data by socioeconomic indicator should be made available

Ensure that rules governing access to pensions do not penalize lower socioeconomic groups

• Maximum withdrawal limits

• Annuitization requirements

• Minimum age to gain full access

24

Challenges for policy makers

The additional years beyond age 65 to maintain a ratio of years in retirement to years working at 0.33 in France

-1 0 1 2 3 4 5 6

Manual

Non-manual

Agriculture

Intermediate

Small employers

Higher managerial & professional

1987

1995

2004

2011

Differences in age of entry into labour force • Ratio of years working per year in retirement for Hungarian

males:

Differences in disability rates

Differences in unemployment rates

Progressivity of public pension

Progressivity of tax rates

25

Additional variables to consider

Education

Life Expectancy age 65

Entry age 20

Entry age varies

Low 10.19 4.4 4.6

Medium 14.16 3.2 3.2

High 17.09 2.6 2.5

6.9 (1.8) (2.1)

1. Retirement ratio – value of time

• the number of years spent contributing to pensions divided by the life expectancy at the age of retirement

2. Asset payout ratio – value of savings

• the present value of pension income over the expected time spent in retirement divided by the assets accumulated at retirement

3. Pension wealth ratio – value of pension

• the present value of pension income relative to annual earnings just before retirement

– Compare private pensions and total pensions (including public)

– Compare gross and net of tax

4. Purchasing power ratio – value of maintaining a certain standard of living

• average pension income in real terms over the expected time spent in retirement divided by the last salary received before retirement

– Also consider relative to consumption

26

How can we assess the impact on pensions?

Immediate annuity offers an attractive income for all income groups compared to other payout designs

• But higher dispersion of outcomes across income groups

• High income groups benefit more from indexation to inflation

Progressivity of public pension and tax rules completely offset inequality of pension wealth ratio

• Is this the right way to address pension inequalities in life expectancy?

All pension income ratios show similar pattern across payout designs for a given income class

• But net measures show potential tax incentives to take a lump-sum vs. purchase an annuity

27

Early insights from Korea

Asset payout ratio • Cannot consider the impact of public pensions

Pension wealth ratio • Doesn’t adequately capture in equality

– Value of early contributions by low income groups is greater than the lost value of pension income due to shorter life expectancies

Total pension and net pension • Is this the right measure to consider offsetting impact of

progressive policies?

Impact of disability and unemployment • Lack of statistics by socioeconomic group

• Challenging to model

28

Challenges for the analysis

Expected longevity risk is unavoidable and must be accounted for in mortality assumptions used to value pension and annuity liabilities

• Current Level: Mortality tables should be regulatory updated based on relevant data

• Trend: Mortality improvements should be accounted for

Challenges for getting assumptions right

• Data availability

• Coverage of pension system

• Demographic shocks due to changes in pension rules

Challenges for pension policy

• Creating sustainable pension policies to address longer life expectancies which consider the heterogeneity of the population’s mortality

29

Key Takeaways

THANK YOU!

Jessica Mosher, Policy Analyst Email: jessica.mosher@oecd.org