The South African

insurance gap (2022)

Quantifying the insurance gap by considering the
financial impact on South African households of the
death or disability of an earner in the household

Final Report

19 October 2022

A study by True South Actuaries & Consultants

 TABLE OF CONTENTS

1 INTRODUCTION ................................................................................................................ 7

1.1 DEFINITIONS .................................................................................................................................................... 7

1.2 BACKGROUND AND BRIEF ............................................................................................................................... 7

1.3 ACKNOWLEDGEMENTS ................................................................................................................................... 7

1.4 WHAT IS INCLUDED AND WHAT IS EXCLUDED .............................................................................................. 7

2 EXECUTIVE SUMMARY ..................................................................................................... 9

2.1 NUMBER OF DEATH EVENTS / DISABILITY EVENTS EXPECTED ...................................................................... 9

2.2 THE INSURANCE GAP ....................................................................................................................................... 9

2.3 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 10

2.4 THE INSURANCE GAP BROKEN DOWN INTO SEGMENTS ........................................................................... 10

3 INCREASE IN THE INSURANCE GAP SINCE THE PREVIOUS STUDY ............................ 11

3.1 THE INSURANCE NEED GREW BY 1.5% PA ................................................................................................. 11

3.2 ACTUAL COVER GREW BY 3.2% PA ............................................................................................................ 11

3.3 THE TOTAL INSURANCE GAP INCREASED BY 0.2% PA............................................................................... 13

4 PERSONALISING THE INSURANCE GAP........................................................................ 14

4.1 THE INSURANCE GAP PER EARNER .............................................................................................................. 14

4.2 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 15

4.3 FURTHER PERSONALISING THE INSURANCE GAP ....................................................................................... 15

5 THE INSURANCE GAP PER EARNINGS GROUP............................................................. 16

5.1 SEGMENTS..................................................................................................................................................... 16

5.2 FINDINGS....................................................................................................................................................... 16

www.asisa.org.za Page 2 of 44
 5.3 THE SIZE OF THE INSURANCE GAP PER SEGMENT ...................................................................................... 17

5.4 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 18

6 THE INSURANCE GAP PER EDUCATION LEVEL ............................................................ 19

6.1 SEGMENTS..................................................................................................................................................... 19

6.2 FINDINGS....................................................................................................................................................... 19

6.3 THE SIZE OF THE INSURANCE GAP PER SEGMENT ...................................................................................... 20

6.4 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 21

7 THE INSURANCE GAP PER AGE GROUP ........................................................................ 22

7.1 SEGMENTS..................................................................................................................................................... 22

7.2 FINDINGS....................................................................................................................................................... 22

7.3 THE SIZE OF THE INSURANCE GAP PER SEGMENT ...................................................................................... 23

7.4 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 24

8 THE INSURANCE GAP PER PROVINCE .......................................................................... 25

8.1 SEGMENTS..................................................................................................................................................... 25

8.2 FINDINGS....................................................................................................................................................... 25

8.3 THE SIZE OF THE INSURANCE GAP PER SEGMENT ...................................................................................... 26

8.4 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 28

9 THE INSURANCE GAP PER GENDER .............................................................................. 30

9.1 SEGMENTS..................................................................................................................................................... 30

9.2 FINDINGS....................................................................................................................................................... 30

9.3 THE SIZE OF THE INSURANCE GAP PER SEGMENT ...................................................................................... 30

9.4 RESPONSES TO THE INSURANCE GAP ......................................................................................................... 31

APPENDIX A: DEFINITIONS ................................................................................................... 33

“ACTUAL COVER” ....................................................................................................................................................... 33

www.asisa.org.za Page 3 of 44
 “COVER ADEQUACY” .................................................................................................................................................. 33

“DEATH EVENT” .......................................................................................................................................................... 33

“DISABILITY EVENT” .................................................................................................................................................... 34

“EARNER” .................................................................................................................................................................... 34

“EARNINGS” ................................................................................................................................................................ 34

“INSURANCE GAP” ...................................................................................................................................................... 34

“INSURANCE NEED” .................................................................................................................................................... 35

“RETIREMENT AGE” ..................................................................................................................................................... 35

APPENDIX B: MODELLING NOTES - EARNER POPULATION .............................................. 36

APPENDIX C: MODELLING NOTES - INSURANCE NEED ..................................................... 37

APPENDIX C1: TOTAL INCOME AT RISK................................................................................................................... 37

APPENDIX C2: REPLACEMENT RATIO....................................................................................................................... 38

APPENDIX C3: CAPITALISATION MULTIPLE ............................................................................................................. 40

APPENDIX D: MODELLING NOTES - ACTUAL COVER ......................................................... 41

APPENDIX D1: ASISA QUESTIONNAIRES ................................................................................................................ 41

APPENDIX D2: OTHER DATA SOURCES ................................................................................................................... 42

APPENDIX D3: ALLOCATION OF ACTUAL COVER TO EACH MODELLED EARNER - TRUE SOUTH MODELS ......... 42

www.asisa.org.za Page 4 of 44
 LIST OF TABLES

TABLE 1 - NUMBER OF EARNERS SUFFERING A DEATH AND/OR DISABILITY EVENT IN 2022 .... 9

TABLE 2 - POSSIBLE RESPONSES TO THE INSURANCE GAP.................................................................... 10

TABLE 3 – THREE COMPONENTS COMBINE TO DEFINE THE INSURANCE NEED ........................... 11

TABLE 4 - TOTAL ACTUAL COVER HELD RELATIVE TO PREVIOUS STUDY ......................................... 12

TABLE 5 - GROWTH IN LIFE COVER, SPLIT BETWEEN RETAIL AND GROUP INSURANCE ............ 12

TABLE 6 - GROWTH IN DISABILITY COVER, SPLIT BETWEEN RETAIL AND GROUP INSURANCE
................................................................................................................................................................................. 12

TABLE 7 - INSURANCE GAP: COMPARISON AGAINST PREVIOUS STUDY ......................................... 13

TABLE 8 - INSURANCE GAP IN TOTAL FOR ALL EARNERS AND PER-EARNER ................................ 14

TABLE 9 - POSSIBLE RESPONSES TO THE INSURANCE GAP.................................................................... 15

TABLE 10 – DEMOGRAPHICS OF EARNERS IN EACH OF THE EARNING SEGMENTS .................... 16

TABLE 11 - DERIVATION OF THE PER-EARNER DEATH INSURANCE GAP BY EARNINGS ........... 17

TABLE 12 - DERIVATION OF THE PER-EARNER DISABILITY INSURANCE GAP BY EARNINGS .... 17

TABLE 13 - RESPONSES TO THE DEATH INSURANCE GAP BY EARNINGS......................................... 18

TABLE 14 - RESPONSES TO THE DISABILITY INSURANCE GAP BY EARNINGS ................................. 18

TABLE 15 - DEMOGRAPHICS OF EARNERS IN EACH OF THE EDUCATION SEGMENTS ............... 19

TABLE 16 - DERIVATION OF THE PER EARNER DEATH INSURANCE GAP BY EDUCATION ......... 20

TABLE 17 - DERIVATION OF THE PER-EARNER DISABILITY INSURANCE GAP BY EDUCATION 20

TABLE 18 - RESPONSES TO THE PER-EARNER DEATH INSURANCE GAP BY EDUCATION .......... 21

TABLE 19 - RESPONSES TO THE PER-EARNER DISABILITY INSURANCE GAP BY EDUCATION .. 21

TABLE 20 – DEMOGRAPHICS OF EARNERS IN EACH OF THE CONSIDERED AGE SEGMENTS ... 22

TABLE 21 - DERIVATION OF THE PER-EARNER DEATH INSURANCE GAP BY AGE ......................... 23

www.asisa.org.za Page 5 of 44
TABLE 22 - DERIVATION OF THE PER-EARNER DISABILITY INSURANCE GAP BY AGE ................. 23

TABLE 23 - RESPONSES TO THE DEATH INSURANCE GAP BY AGE SEGMENTS .............................. 24

TABLE 24 - RESPONSES TO THE DISABILITY INSURANCE GAP BY AGE .............................................. 24

TABLE 25 - DEMOGRAPHICS OF EARNERS IN EACH OF THE PROVINCES ........................................ 25

TABLE 26 - DERIVATION OF THE PER-EARNER DEATH INSURANCE GAP BY PROVINCE ............ 26

TABLE 27 - DERIVATION OF THE PER-EARNER DISABILITY INSURANCE GAP BY PROVINCE .... 27

TABLE 28 - RESPONSES TO THE DEATH INSURANCE GAP BY PROVINCE ......................................... 28

TABLE 29 - RESPONSES TO THE DISABILITY INSURANCE GAP BY PROVINCE ................................. 29

TABLE 30 - DEMOGRAPHICS OF EARNERS SEPARATELY FOR MALES AND FEMALES.................. 30

TABLE 31 - DERIVATION OF THE PER-EARNER DEATH INSURANCE GAP BY GENDER ................ 30

TABLE 32 - DERIVATION OF THE PER-EARNER DISABILITY INSURANCE GAP BY GENDER ........ 31

TABLE 33 - RESPONSES TO THE DEATH INSURANCE GAP BY GENDER ............................................. 31

TABLE 34 - RESPONSES TO THE DISABILITY INSURANCE GAP BY GENDER ..................................... 32

TABLE 35 - INSURANCE NEED REPRESENTED AS PRODUCT OF THREE NUMBERS ....................... 37

TABLE 36 - TOTAL EARNINGS BY EARNERS INCREASED BY 2.0% PA .................................................. 38

www.asisa.org.za Page 6 of 44
1 INTRODUCTION

1.1 Definitions

Terms for which a definition is supplied in Appendix A, are denoted in italics.

1.2 Background and brief

ASISA retained True South Actuaries and Consultants to update previous studies conducted to

determine the extent of under-insurance in South Africa. This study is referred to as the “2022
Insurance gap Study” and reflects the situation as at the end of 2021. Previous studies were
conducted in 2007, 2010, 2013, 2016, and 2019 which reflected the extent of under-insurance
in South Africa as at the end of 2006, 2009, 2012, 2015, and 2018 respectively.

1.3 Acknowledgements

We express our thanks to the life insurers who participated in the study through the provision

of data on fairly tight timelines. We continued to experience improvement in both the quality

and quantity of data compared to the previous studies. We are also particularly grateful for the
many instances where senior officials in the respective organizations got personally involved to

assist us in securing the necessary data.

1.4 What is included and what is excluded

In Appendix A (defining terms that are denoted in italics throughout this document), the

insurance gap (see modelling notes in Appendix B) is defined as the difference between the
insurance need and actual cover:

• The insurance need is defined as the amount of cover required to meet the need that is
created by the death event and/or the disability event (see modelling notes in Appendix C):

o It assumes the household would want to maintain the pre-event standard of living.
o It further assumes that the need extends to the deceased / disabled household

member’s retirement age only as this study doesn’t express any view on post-
retirement provision adequacy.
o It excludes any additional short-term expenses related to the risk event, such as

funeral costs, medical costs and/or cost of structural changes to one’s home in the

case of a disability event.

www.asisa.org.za Page 7 of 44
• Actual cover considers benefits received post-event from insurers (retail and group-type

cover), self-insurance pension schemes (like the GEPF), and government disability grants.

(See modelling notes in Appendix D.)

o It excludes funeral cover (as it is pragmatically assumed that the objective of such
cover does not include income replacement).
o It excludes cover that provides for certain selected situations only, such as accident

only cover and cover from the Road Accident Fund, the Compensation Fund, and
short-term insurance.
o It ignores any shortfall that may result due to any waiting periods that may be

enforced by the disability cover product design.

www.asisa.org.za Page 8 of 44
2 EXECUTIVE SUMMARY

2.1 Number of death events / disability events expected

The table below shows the number of earners expected to suffer a death event 1 and / or a
disability event 2 during 2022:

Table 1 - Number of earners suffering a death and/or disability event in 2022

Death event 3 Disability event

Number of earners 14.3m 14.3m
Number of events expected per year 142 723 47 099
Number of events expected per day 391 129

2.2 The insurance gap

At a macro level

The insurance gap was determined using the same principles as used for previous studies. (See
appendices B, C, and D.) The insurance gap as at the end of 2021 was calculated to be 34.3

trillion (1 trillion = 1 000 billion = 1 000 000 million =1012 ):

• If South African households wanted to maintain their standards of living after a death event,

the insurance need for all earners combined is in the region of R25.7 trillion (see section
3.1). The extent of actual cover in force in the economy only amounts to R11.3 trillion (see

section 3.2). This leaves a death insurance gap of around R14.4 trillion (see section 3.3).

• If South African households wanted to maintain their standards of living after a disability
event, the insurance need for all earners combined is in the region of R37.2 trillion (see

section 3.1). The extent of actual cover in force in the economy only amounts to R17.3
trillion (see section 3.2). This leaves a disability insurance gap of around R19.9 trillion (see

section 3.3).

(Numbers are rounded.)

1
By reference to the demographics of the earners in each segment and application of the AIDS model of the Actuarial
Society of South Africa.

2
By reference to a disability investigation of the Actuarial Society of South Africa calibrated so that ratio of disabilities
to deaths is consistent with group premium rates obtained.

3
The number of death events expected aims to illustrate a long-term stable view of mortality and would therefore
exclude increases in deaths experience over recent years.

www.asisa.org.za Page 9 of 44
At a micro level

The insurance gap as at the end of 2021 was calculated to be R2.4m for the average South
African earner:

• If the average South African earner wanted to ensure that her/his family can maintain their

standard of living in the event of her/his death, provision would need to be made for R1.8m
of cover. However, the average South African earner has life cover of just less than R0.8m.
This leaves an average death insurance gap of about R1.0m.

• If the average South African earner wanted to ensure that her/his family can maintain their

standard of living in the event of her/his being subject to a disability event, provision would
need to be made for R2.6m of cover. As the average South African earner has disability

cover of just less than R1.2m, this leaves an average disability insurance gap of about R1.4m.

2.3 Responses to the insurance gap

A response to the insurance gap, could be to pro-actively purchase additional death and
disability cover.

Reactive responses (post the death event / disability event) include [1] curtailing household

expenditure and [2] shifting the burden of under-insurance to the remaining household
members of working-age by requiring increased contributions from them to total household

income. The extent required by each of the responses is summarised in the table below:

Table 2 - Possible responses to the insurance gap

Pro-active Reactive post death event / disability event

Cost to close gap (% of % reduction in Generating additional
earnings) household expenditure income per month
Death event 4.5% 30% 4 5 630
Disability event 2.6% 33% 7 443
Total 7.1%

2.4 The insurance gap broken down into segments

In the main body of this document, the insurance gap is reported for different segments of the
population in terms of income, education, age, province, and gender.

4
Expressed as a percentage of post event household expenditure (i.e. reduced due to smaller family size).

www.asisa.org.za Page 10 of 44
3 INCREASE IN THE INSURANCE GAP SINCE THE PREVIOUS STUDY

3.1 The insurance need grew by 1.5% pa

In the table below, the insurance need is expressed as the product of three numbers:

Table 3 – Three components combine to define the insurance need

Total insurance need (in R’bn) Total insurance need (in R’bn)
Note 2019 study 2022 study
Death Disability Death Disability
Total income at risk 1 2 863 2 863 3 042 3 042
* Replacement ratio 2 57% 78% 57% 78%
* Capitalisation factor 3 15.1 15.9 14.9 15.6
Insurance need 1*2*3 24 488 35 654 25 712 37 209

Note 1 - Income at risk: This includes all income as per the definition (Appendix A) of earnings.

The slow growth in the income at risk was due to a decrease in the workforce (by about 1.3m
workers), coupled by earnings per earner growing at less than CPI pa. (For more information,

see Appendix C1.)

Note 2 - Replacement ratio: This represents the proportion of household members’ personal

income that “will be missed” after the death event or the disability event. The replacement ratio

is lower for the death event reflecting the fact that the deceased earner’s portion of expenses

will no longer be part of the household budget. (For more information, see Appendix C2.)

Note 3 - Capitalisation factor: This factor is related to the number of years that the earner would

still have contributed to the household. It reflects the period from current day up to retirement

age. The capitalisation factors are slightly lower than that of the 2019 study, as the age of the
average earner increased slightly over the 3-year period. This was caused by rising

unemployment amongst the younger age groups. (For more information, see Appendix C3.)

3.2 Actual cover grew by 3.2% pa

As for the 2019 study, regulatory returns form the basis of the actual cover information

requirement, which ensures credibility of the results in at least two ways:

www.asisa.org.za Page 11 of 44
• High response rates: As the requested information is readily available, (the bulk had already

been provided to the regulator), most insurers respond to the information request resulting

in a lesser need to ratio data to account for non-respondents.

• Quality and comparability of information: Returns are audited and the specification for
what is required is uniform for all insurers alike. There is therefore less possibility for
interpretation / data extraction errors.

See Appendix D for more information on the sources used for determining the total level of
actual cover in the market and allocating this to different segments of the earner population.

Table 4 - Total actual cover held relative to previous study

Actual cover in R’bn Note 2019 study 5 2022 study Increase pa

Life cover 1 9 674 11 348 5.5%
Disability insurance cover 2 12 500 13 252 2.0%
Disability grants 3 3 895 4 014 1.0%
Total 26 069 28 615 3.2%

Note 1: Growth in life cover since previous study

Table 5 - Growth in life cover, split between retail and group insurance

Actual life cover in R’bn 2019 study 5 2022 study Increase pa

Retail insurance 5 689 6 681 5.5%
Group insurance 3 985 4 667 5.4%
Total life insurance cover 9 674 11 348 5.5%

Note 2: Growth in disability cover since previous study

Table 6 - Growth in disability cover, split between retail and group insurance

Actual disability cover in R’bn 2019 study 2022 study Increase pa

Retail insurance 4 722 4 974 1.8%
Group insurance 7 779 8 279 2.1%
Total disability insurance cover 12 500 13 252 2.0%

5
The 2019 study’s actual life cover figures were restated. The actual group life cover in place at the end of 2018 was
understated (and the life insurance gap thus overstated) in the 2019 study. The restatement was necessary to ensure
that both studies are on a like-for-like basis, enabling sensible comparisons to be made and conclusions to be drawn.

www.asisa.org.za Page 12 of 44
Note 3: The amount of the maximum social security disability grant increased by about 3.6% pa

since the 2019 study. However, the number of low-income earners eligible for the grant

decreased and thus the total cover available in the scenario where all earners become disabled

has increased by only 1.0% pa since the 2019 study.

3.3 The total insurance gap increased by 0.2% pa

Table 7 - Insurance gap: Comparison against previous study

R’bn 2019 study 5 2022 study

Death Disability Total Death Disability Total
Insurance need 24 488 35 654 60 142 25 712 37 209 62 921
Actual cover from insurance -9 674 -12 500 -22 174 -11 348 -13 252 -24 601
Disability grant cover - -3 895 -3 895 - -4 014 -4 014
Insurance gap 14 814 19 259 34 073 14 364 19 943 34 307
Cover adequacy 40% 46% 43% 44% 46% 45%

From the end of 2018 to the end of 2021, the insurance gap increased by 0.2% pa. The insurance

gap per earner increased at a steady rate of 3.2% pa; the decrease in the insurance gap was

therefore driven by the reduction in the number of earners over the 3-year period.

The actual cover as percentage of the insurance need (referred to as cover adequacy) indicates

that only 45% of the insurance need is currently met by actual cover. This percentage improved

slightly since the previous study due to the insurance need growing at a slower rate than the

actual cover.

From the end of 2018 to the end of 2021, the death insurance gap decreased slightly (1.0% pa),

while the disability insurance gap increased slightly (1.2% pa).

www.asisa.org.za Page 13 of 44
4 PERSONALISING THE INSURANCE GAP

4.1 The insurance gap per earner

The table below shows how the insurance gap of R34.3 trillion can be expressed as an average
gap of R2.4m per South African earner (R1.0m for death events and R1.4m for disability events):

Table 8 - Insurance gap in total for all earners and per-earner

Total insurance gap (in R’bn) Insurance gap per earner (in Rand)
Death Disability Death Disability
Insurance need 25 712 37 209 1 801 422 2 606 929
Total income at risk 3 042 3 042 213 093 213 093
* Replacement ratio 57% 78% 57% 78%
* Capitalisation factor 14.9 15.6 14.9 15.6
Actual cover -11 348 -17 266 -795 074 -1 209 704
Retail insurance -6 681 -4 974 -468 084 -348 472
Group insurance -4 667 -8 279 -326 990 -580 008
Government grants 0 -4 014 - -281 224
Insurance gap 14 364 19 943 1 006 348 1 397 225

In the case of a death event

If the average South African earner wanted to ensure that her/his family can maintain their

standard of living in the event of her/his death, provision would need to be made for R1.8m of
cover. However, the average South African earner has life cover of just less than R0.8m. This
leaves an average death insurance gap of about R1.0m.

In the case of a disability event

If the average South African earner wanted to ensure that her/his family can maintain their

standard of living in the event of her/his being subject to a disability event, provision would

need to be made for R2.6m of cover. However, the average South African earner has disability
cover of just more than R1.2m. This leaves an average disability insurance gap of about R1.4m.

www.asisa.org.za Page 14 of 44
4.2 Responses to the insurance gap

A response to the insurance gap, could be to pro-actively purchase additional death and

disability cover.

Reactive responses (post the death event / disability event) include [1] curtailing household

expenditure and [2] shifting the burden of under-insurance to the remaining household

members of working-age by requiring increased contributions from them to total household

income.

The table below indicates the extent required by each of these responses:

Table 9 - Possible responses to the insurance gap

Pro-active Reactive post death event / disability event

Cost to close gap (% of % reduction in Generating additional
earnings) household expenditure income per month
Death event 4.5% 30% 5 630
Disability event 2.6% 33% 7 443
Total 7.1%

4.3 Further personalising the insurance gap

Due to the diversity of the South African socio-economic landscape, the concept of the “average
South African earner” is less clear than (say) the “average Australian earner”. For this reason, it

makes sense to consider the insurance gap for different segments of the South African earner

population.

In the sections below we show the insurance gap and possible responses to it for the following

segments of the South African earner population:

• Section 5: Earnings groups

• Section 6: Level of education

• Section 7: Age groups
• Section 8: Province

• Section 9: Gender

www.asisa.org.za Page 15 of 44
5 THE INSURANCE GAP PER EARNINGS GROUP

5.1 Segments

The 14.3 million earners were divided into 5 groups with equal representation by number. The
first group represented the 20% poorest individuals within the universe of earners. The next

group represented the next 20% poorest individuals, etc.:

Table 10 – Demographics of earners in each of the earning segments

Average annual
Number of
Segment Segment bounds (net earnings) net earnings Average Age
earners (million)
(Rand)
Poorest 20% up to R33 154 p.a. 2.9 17 056 36
2 Quantile
nd
R33 155 to R67 005 p.a. 2.9 49 665 37
3rd Quantile R67 006 to R118 195 p.a. 2.9 90 645 37
4 Quantile
th
R118 196 to R246 922 p.a. 2.9 194 607 39
Richest 20% more than R246 923 p.a. 2.9 713 428 42
All 14.3 213 093 38

5.2 Findings

The numbers in the remainder of this section reveal that:

• For a death event, the cover adequacy shows a strong positive correlation with personal

income - i.e. the higher the income, the bigger proportion of the insurance need is met by
actual cover.
• This would have been the case for the disability event as well had it not been for government

disability grants. The level of the grant is such that it covers the full disability insurance
need of the poorest 20% earners. Most earners within this (poorest 20%) group would

therefore not have any need for additional (or any for that matter) disability insurance cover.

www.asisa.org.za Page 16 of 44
5.3 The size of the insurance gap per segment

Death insurance gap per earner

Table 11 - Derivation of the per-earner death insurance gap by earnings

Poorest 20% 2nd Quantile 3rd Quantile 4th Quantile Richest 20%
Insurance need 133 112 553 252 1 053 472 1 987 061 5 279 739
Total income at risk 17 056 49 665 90 645 194 607 713 428
* Replacement ratio 43% 61% 66% 61% 54%
* Capitalisation factor 18.2 18.1 17.7 16.6 13.6
Actual cover -7 424 -47 941 -163 868 -581 114 -3 174 736
Retail insurance -3 150 -11 369 -27 834 -270 448 -2 027 434
Group insurance -4 274 -36 571 -136 034 -310 666 -1 147 302
Government grants 0 0 0 0 0
Insurance gap 125 688 505 312 889 603 1 405 948 2 105 003
Cover adequacy (Cover/Need) 6% 9% 16% 29% 60%
Number of earners (millions) 2.9 2.9 2.9 2.9 2.9
Total insurance gap (R’bn) 359 1 443 2 539 4 014 6 010

For example, an earner that finds her/himself in the top 20% of South African earners, would
typically need life cover of almost R5.3m. Typically such an earner would only have life cover of

R3.2m, leaving an average insurance gap of around R2.1m.

Disability insurance gap per earner

Table 12 - Derivation of the per-earner disability insurance gap by earnings

Poorest 20% 2nd Quantile 3rd Quantile 4th Quantile Richest 20%
Insurance need 319 837 926 205 1 634 170 2 991 914 7 161 882
Total income at risk 17 056 49 665 90 645 194 607 713 428
* Replacement ratio 100% 100% 98% 89% 71%
* Capitalisation factor 18.8 18.6 18.4 17.3 14.2
Actual cover -517 999 -516 978 -626 701 -937 172 -3 449 411
Retail insurance -478 -5 406 -43 693 -204 714 -1 487 935
Group insurance -9 309 -77 994 -287 121 -613 056 -1 912 393
Government grants -508 211 -433 579 -295 886 -119 401 -49 083
Insurance gap -198 161 409 227 1 007 469 2 054 742 3 712 471
Cover adequacy (Cover/Need) > 100% 56% 38% 31% 48%
Number of earners (millions) 2.9 2.9 2.9 2.9 2.9
Total insurance gap (R’bn) -566 1 168 2 876 5 866 10 599

For example, the 20% poorest South African earners would typically need disability cover of

about R0.3m. Typically such an earner would have disability cover far exceeding this need,
mostly due to government grants leaving no insurance gap.

www.asisa.org.za Page 17 of 44
5.4 Responses to the insurance gap

Death insurance gap

The table below gives (for each of the segments) the degree of intervention required for each

of three possible responses to the insurance gap in the case of a death event:

Table 13 - Responses to the death insurance gap by earnings

Poorest 2nd 3rd 4th Richest

All
20% Quantile Quantile Quantile 20%
Purchase Cost of insurance (as % of
7.1% 9.7% 9.4% 6.9% 2.7% 4.5%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 16.9 10.5 5.4 2.4 0.7 1.3
equal to 1.5 times current
level)
Reduce
Required reduction in
household 11% 32% 39% 35% 27% 30%
household expenditure
expenditure
Additional Extra income required per
703 2 827 4 977 7 865 11 776 5 630
income month (net of tax)
required Extra income as % of
49% 68% 66% 48% 20% 32%
earnings pre-event

Disability insurance gap

For a disability event, the figures are as follows:

Table 14 - Responses to the disability insurance gap by earnings

Poorest 2nd 3rd 4th Richest

All
20% Quantile Quantile Quantile 20%
Purchase Cost of insurance (as % of
-4.6% 3.2% 4.4% 4.2% 2.1% 2.6%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required -0.4 0.8 1.6 2.2 1.1 1.2
equal to 1.5 times current
level)
Reduce
Required reduction in
household -13% 21% 35% 41% 38% 33%
household expenditure
expenditure
Additional Extra income required per
-1 056 2 180 5 366 10 945 19 775 7 443
income month (net of tax)
required Extra income as % of
-74% 53% 71% 67% 33% 42%
earnings pre-event

www.asisa.org.za Page 18 of 44
6 THE INSURANCE GAP PER EDUCATION LEVEL

6.1 Segments

Each earner is allocated to one of five education categories depending on the highest level of
education achieved:

• PS or lower: Primary school not completed.

• Some HS: Completed primary school, but not high school.

• Matric: Completed high school, but no diploma or degree.

• Diploma: Diploma but no degree.

• Degree: Degree or more.

Table 15 - Demographics of earners in each of the education segments

Number of earners Average annual

Segment Average Age
(million) earnings (Rand)
PS or lower 1.2 63 434 44
Some HS 4.9 113 354 39
Matric 5.6 173 884 35
Diploma 0.9 369 622 39
Degree 1.7 666 270 41
All 14.3 213 093 38

6.2 Findings

The numbers in the remainder of this section reveal that:

• For the death event, the cover adequacy shows a strong positive correlation with highest
level of education achieved.

• This would have been the case for the disability event as well had it not been for government

grants. These grants are targeted at the poor where there is a bias to lower levels of

education.

www.asisa.org.za Page 19 of 44
6.3 The size of the insurance gap per segment

The tables below highlight how different the insurance gap is for the different segments. We

show figures for the “average earner” within each segment.

Death insurance gap per earner

Table 16 - Derivation of the per earner death insurance gap by education

PS or lower Some HS Matric Diploma Degree

Insurance need 495 581 988 477 1 793 816 2 978 781 4 567 780
Total income at risk 63 434 113 354 173 884 369 622 666 270
* Replacement ratio 65% 62% 60% 55% 51%
* Capitalisation factor 12.1 14.0 17.2 14.6 13.4
Actual cover -73 808 -237 304 -521 747 -1 618 717 -3 452 131
Retail insurance -27 032 -103 579 -283 716 -940 721 -2 235 195
Group insurance -46 777 -133 725 -238 031 -677 996 -1 216 936
Government grants 0 0 0 0 0
Insurance gap 421 773 751 173 1 272 069 1 360 064 1 115 650
Cover adequacy (Cover/Need) 15% 24% 29% 54% 76%
Number of earners (millions) 1.2 4.9 5.6 0.9 1.7
Total insurance gap (R’bn) 513 3 710 7 107 1 164 1 870

For example, earners with matric as highest qualification would typically need life cover of

almost R1.8m. Typically such an earner would only have cover of R0.5m, leaving an insurance

gap of R1.3m – implying that only 29% of the life insurance need is protected by actual cover.

Table 17 - Derivation of the per-earner disability insurance gap by education

PS or lower Some HS Matric Diploma Degree

Insurance need 747 544 1 454 137 2 590 227 4 363 674 6 510 463
Total income at risk 63 434 113 354 173 884 369 622 666 270
* Replacement ratio 93% 87% 83% 77% 69%
* Capitalisation factor 12.7 14.8 18.0 15.4 14.1
Actual cover -383 918 -615 133 -1 009 416 -2 082 418 -3 782 433
Retail insurance -18 057 -70 321 -225 129 -747 308 -1 615 152
Group insurance -67 500 -211 582 -474 042 -1 192 376 -2 077 734
Government grants -298 360 -333 230 -310 246 -142 734 -89 547
Insurance gap 363 626 839 004 1 580 811 2 281 257 2 728 030
Cover adequacy (Cover/Need) 51% 42% 39% 48% 58%
Number of earners (millions) 1.2 4.9 5.6 0.9 1.7
Total insurance gap (R’bn) 442 4 143 8 832 1 953 4 572

For example, an earner with at least a degree, would typically need disability cover of about
R6.5m. Typically, such an earner would only have cover of R3.3m, leaving a substantial insurance
gap of R3.8m – implying that 58% of the disability insurance need is protected by actual cover.

www.asisa.org.za Page 20 of 44
6.4 Responses to the insurance gap

Death insurance gap

The table below gives (for each of the segments), the degree of intervention required for each

of three possible responses to the insurance gap in the case of a death event:

Table 18 - Responses to the per-earner death insurance gap by education

PS or
Some HS Matric Diploma Degree All
lower
Purchase Cost of insurance (as % of
6.4% 6.4% 7.0% 3.4% 1.5% 4.5%
additional current earnings)
insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 5.7 3.2 2.4 0.8 0.3 1.3
equal to 1.5 times current
level)
Reduce
Required reduction in
household 30% 35% 39% 26% 14% 30%
household expenditure
expenditure
Additional Extra income required per
2 359 4 202 7 116 7 608 6 241 5 630
income month (net of tax)
Required Extra income as % of
45% 44% 49% 25% 11% 32%
earnings pre-event

Disability insurance gap

For the disability event, the figures are as follows:

Table 19 - Responses to the per-earner disability insurance gap by education

PS or
Some HS Matric Diploma Degree All
lower
Purchase Cost of insurance (as % of
2.4% 3.0% 3.6% 2.4% 1.6% 2.6%
additional current earnings)
insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 0.9 1.4 1.6 1.1 0.7 1.2
equal to 1.5 times current
level)
Reduce
Required reduction in
household 21% 31% 39% 34% 27% 33%
household expenditure
expenditure
Additional Extra income required per
1 937 4 469 8 420 12 151 14 531 7 443
income month (net of tax)
required Extra income as % of
37% 47% 58% 39% 26% 42%
earnings pre-event

www.asisa.org.za Page 21 of 44
7 THE INSURANCE GAP PER AGE GROUP

7.1 Segments

Earners were categorised based on their age last birthday:

Table 20 – Demographics of earners in each of the considered age segments

Number of earners Average annual

Segment Average Age
(million) earnings (Rand)
Under 30 3.7 123 350 25
30-39 4.3 191 012 34
40-49 3.7 270 876 44
50-54 1.4 263 546 52
55 and over 1.3 324 838 58
All 14.3 213 093 38

7.2 Findings

The numbers in the remainder of this section reveal that cover adequacy is lowest at the younger
ages. Earners in the older age categories (50-54 and 55 and over) tend to have adequate

insurance in place, i.e. the insurance need is covered by actual cover in place:

• Insurance need: As mentioned earlier, calculations assume that an insurance need only

exists up to intended retirement age. As such, the insurance need for older earners is a
much smaller multiple to current earnings compared to younger earners. This is evidenced

in the capitalisation factors in the table below.

• Actual cover: Lump sum benefits do not take into account the diminishing need for
insurance with increasing age.

www.asisa.org.za Page 22 of 44
7.3 The size of the insurance gap per segment

The tables below highlight how different the insurance gap is for the different age-group

segments. We show figures for the “average earner” within each segment:

Death insurance gap per earner

Table 21 - Derivation of the per-earner death insurance gap by age

Under 30 30-39 40-49 50-54 55 and over

Insurance need 1 820 505 2 178 550 2 082 227 1 076 306 468 831
Total income at risk 123 350 191 012 270 876 263 546 324 838
* Replacement ratio 58% 57% 57% 55% 55%
* Capitalisation factor 25.6 20.0 13.4 7.4 2.6
Actual cover -209 265 -763 791 -1 052 582 -1 261 796 -1 350 122
Retail insurance -84 799 -401 595 -652 151 -856 381 -851 255
Group insurance -124 466 -362 196 -400 430 -405 415 -498 867
Government grants 0 0 0 0 0
Insurance gap 1 611 240 1 414 760 1 029 646 -185 491 -881 291
Cover adequacy (Cover/Need) 11% 35% 51% > 100% > 100%
Number of earners (millions) 3.7 4.3 3.7 1.4 1.3
Total insurance gap (R’bn) 5 954 6 013 3 788 -251 -1 140

For example, an earner aged between 30 and 39 typically needs R2.2m of life cover to ensure

the household can maintain its standard of living after her/his death. Typically, such an earner

would have life cover of less than R0.8m, leaving an average insurance gap of more than R1.4m.

Disability insurance gap per earner

Table 22 - Derivation of the per-earner disability insurance gap by age

Under 30 30-39 40-49 50-54 55 and over

Insurance need 2 768 041 3 167 075 2 831 317 1 561 705 763 039
Total income at risk 123 350 191 012 270 876 263 546 324 838
* Replacement ratio 86% 81% 75% 75% 74%
* Capitalisation factor 26.0 20.6 13.9 7.9 3.2
Actual cover -1 038 817 -1 365 184 -1 494 630 -953 203 -645 440
Retail insurance -111 976 -328 969 -569 051 -449 754 -354 757
Group insurance -426 174 -717 588 -749 250 -410 483 -263 724
Government grants -500 667 -318 627 -176 329 -92 967 -26 958
Insurance gap 1 729 224 1 801 891 1 336 687 608 501 117 599
Cover adequacy (Cover/Need) 38% 43% 53% 61% 85%
Number of earners (millions) 3.7 4.3 3.7 1.4 1.3
Total insurance gap (R’bn) 6 390 7 658 4 917 825 152

For example, an earner that is younger than 30 typically needs R2.8m of disability cover to
ensure the household can maintain its standard of living after her/his disability. On average,

such an earner has life cover of around R1.0m, leaving an insurance gap of R1.7m.

www.asisa.org.za Page 23 of 44
7.4 Responses to the insurance gap

Death insurance gap

The table below gives (for each of the segments) the degree of intervention required for each

of three possible responses to the insurance gap in the case of a death event:

Table 23 - Responses to the death insurance gap by age segments

55 and
Under 30 30-39 40-49 50-54 All
over
Purchase Cost of insurance (as % of
13.2% 6.8% 3.5% -0.7% -2.8% 4.5%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 7.7 1.9 1.0 -0.1 -0.7 1.3
equal to 1.5 times current
level)
Reduce
Required reduction in
household 53% 47% 29% -5% -20% 30%
household expenditure
expenditure
Additional Extra income required per
9 013 7 914 5 760 -1 038 -4 930 5 630
income month (net of tax)
required Extra income as % of
88% 50% 26% -5% -18% 32%
earnings pre-event

Disability insurance gap

For the disability event, the figures are as follows:

Table 24 - Responses to the disability insurance gap by age

55 and
Under 30 30-39 40-49 50-54 All
over
Purchase Cost of insurance (as % of
5.4% 3.7% 2.0% 1.1% 0.2% 2.6%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 1.7 1.3 0.9 0.6 0.2 1.2
equal to 1.5 times current
level)
Reduce
Required reduction in
household 46% 46% 30% 13% 2% 33%
household expenditure
expenditure
Additional Extra income required per
9 211 9 598 7 120 3 241 626 7 443
income month (net of tax)
required Extra income as % of
90% 60% 32% 15% 2% 42%
earnings pre-event

www.asisa.org.za Page 24 of 44
8 THE INSURANCE GAP PER PROVINCE

8.1 Segments

Earners were categorised based on the province they reside in:

Table 25 - Demographics of earners in each of the provinces

Number of earners Average annual

Segment Average Age
(million) earnings (Rand)
Western Cape 2.0 276 725 39
Eastern Cape 1.2 167 441 38
KwaZulu-Natal 2.3 161 326 37
Northern Cape 0.3 174 091 37
Free State 0.8 152 364 38
North West 0.8 160 469 40
Gauteng 4.9 260 679 39
Mpumalanga 1.0 181 948 38
Limpopo 1.0 157 772 38
All 14.3 213 093 38

8.2 Findings

The numbers in the remainder of this section reveal that:

• Cover adequacy (actual cover / insurance need) for the death event, ranges quite a lot from

26% (Limpopo) to 56% (Western Cape).

• For the disability event, the cover adequacy (actual cover / insurance need) ranges from 32%

(Limpopo) to 54% (Western Cape).

www.asisa.org.za Page 25 of 44
8.3 The size of the insurance gap per segment

We show figures for the “average earner” within each segment:

Death insurance gap per earner

Table 26 - Derivation of the per-earner death insurance gap by province

Western Cape Eastern Cape KwaZulu-Natal Northern Cape Free State North West Gauteng Mpumalanga Limpopo
Insurance need 2 197 176 1 535 175 1 548 999 1 625 508 1 244 308 1 319 579 2 094 109 1 720 364 1 428 976
Total income at risk 276 725 167 441 161 326 174 091 152 364 160 469 260 679 181 948 157 772
* Replacement ratio 57% 61% 59% 58% 54% 56% 55% 58% 58%
* Capitalisation factor 14.0 15.1 16.2 16.0 15.0 14.6 14.5 16.4 15.6
Actual cover -1 236 701 -647 365 -483 523 -563 343 -411 391 -452 223 -1 077 517 -468 824 -374 035
Retail insurance -684 058 -366 370 -306 618 -317 953 -300 699 -294 466 -599 327 -347 882 -316 080
Group insurance -552 643 -280 995 -176 905 -245 390 -110 693 -157 757 -478 190 -120 943 -57 956
Government grants 0 0 0 0 0 0 0 0 0
Insurance gap 960 475 887 810 1 065 476 1 062 166 832 917 867 356 1 016 592 1 251 539 1 054 940
Cover adequacy 56% 42% 31% 35% 33% 34% 51% 27% 26%
Number of earners (m) 2.0 1.2 2.3 0.3 0.8 0.8 4.9 1.0 1.0
Total insurance gap (R’bn) 1 955 1 095 2 413 334 659 709 4 945 1 236 1 018

For example, the average Limpopo earner typically needs R1.4m of life cover to ensure the household can maintain its standard of living after her/his death.
Typically, such an earner would have life cover of less than R0.4m, implying that only 26% of the life insurance need is protected by actual cover.
Disability insurance gap per earner

Table 27 - Derivation of the per-earner disability insurance gap by province

Western Cape Eastern Cape KwaZulu-Natal Northern Cape Free State North West Gauteng Mpumalanga Limpopo
Insurance need 3 122 192 2 173 993 2 230 900 2 463 720 1 953 496 2 025 161 3 008 167 2 508 198 2 109 393
Total income at risk 276 725 167 441 161 326 174 091 152 364 160 469 260 679 181 948 157 772
* Replacement ratio 76% 81% 82% 83% 81% 82% 76% 81% 82%
* Capitalisation factor 14.9 16.0 17.0 17.1 15.8 15.3 15.1 17.0 16.3
Actual cover -1 692 893 -1 094 415 -932 564 -1 029 872 -741 521 -819 120 -1 503 771 -819 843 -678 093
Retail insurance -485 470 -264 063 -242 399 -249 364 -225 105 -231 326 -444 500 -274 303 -240 799
Group insurance -979 983 -498 487 -339 074 -467 799 -197 395 -282 794 -830 474 -230 889 -102 640
Government grants -227 440 -331 864 -351 091 -312 710 -319 021 -305 001 -228 797 -314 651 -334 653
Insurance gap 1 429 299 1 079 578 1 298 336 1 433 848 1 211 975 1 206 040 1 504 396 1 688 355 1 431 301
Cover adequacy 54% 50% 42% 42% 38% 40% 50% 33% 32%
Number of earners (m) 2.0 1.2 2.3 0.3 0.8 0.8 4.9 1.0 1.0
Total insurance gap (R’bn) 2 909 1 332 2 940 451 959 985 7 318 1 667 1 381

For example, the average Gauteng earner typically needs approximately R3.0m of disability cover to ensure the household can maintain its standard of living

after her/his disability. (This is higher than the average earner in South Africa due mostly to the superior average earnings of earners in Gauteng.) Typically,

such an earner would have disability cover around R1.5m, implying that additional disability cover of almost R1.5m is required for the average Gauteng earner.

www.asisa.org.za Page 27 of 44
8.4 Responses to the insurance gap

Death insurance gap

The table below gives (for each of the segments) the degree of intervention required for each of three possible responses to the insurance gap in the case of a

death event:

Table 28 - Responses to the death insurance gap by province

Western KwaZulu- Northern

Eastern Cape Free State North West Gauteng Mpumalanga Limpopo All
Cape Natal Cape
Purchase Cost of insurance (as % of
3.3% 5.0% 6.3% 5.9% 5.2% 5.2% 3.7% 6.6% 6.4% 4.5%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 0.8 1.4 2.2 1.9 2.0 1.9 0.9 2.7 2.8 1.3
equal to 1.5 times current
level)
Reduce
Required reduction in
household 23% 34% 40% 38% 35% 38% 24% 42% 44% 30%
household expenditure
expenditure
Additional Extra income required per
5 373 4 966 5 960 5 942 4 659 4 852 5 687 7 001 5 901 5 630
income month (net of tax)
required Extra income as % of
23% 36% 44% 41% 37% 36% 26% 46% 45% 32%
earnings pre-event

www.asisa.org.za Page 28 of 44
Disability insurance gap

For the disability event, the figures are as follows:

Table 29 - Responses to the disability insurance gap by province

Western KwaZulu- Northern

Eastern Cape Free State North West Gauteng Mpumalanga Limpopo All
Cape Natal Cape
Purchase Cost of insurance (as % of
2.1% 2.6% 3.2% 3.3% 3.2% 3.1% 2.3% 3.7% 3.6% 2.6%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 0.8 1.0 1.4 1.4 1.6 1.5 1.0 2.1 2.1 1.2
equal to 1.5 times current
level)
Reduce
Required reduction in
household 28% 33% 38% 40% 38% 40% 28% 45% 47% 33%
household expenditure
expenditure
Additional Extra income required per
7 613 5 751 6 916 7 638 6 456 6 424 8 013 8 993 7 624 7 443
income month (net of tax)
required Extra income as % of
33% 41% 51% 53% 51% 48% 37% 59% 58% 42%
earnings pre-event

www.asisa.org.za Page 29 of 44
9 THE INSURANCE GAP PER GENDER

9.1 Segments

Earners are split between male and female as following:

Table 30 - Demographics of earners separately for males and females

Number of earners Average annual

Segment Average Age
(million) earnings (Rand)
Male 8.1 235 922 38
Female 6.1 182 928 38
All 14.3 213 093 38

9.2 Findings

The numbers in the remainder of this section reveal that:

• Females are marginally less adequately covered than males for both death and disability
events.

• Males make up roughly 58% of the total insurance gap, and females the remaining 42%.

• The cost of purchasing additional insurance is cheaper for females compared to males.

9.3 The size of the insurance gap per segment

Death insurance gap per earner

Table 31 - Derivation of the per-earner death insurance gap by gender

Male Female
Insurance need 1 896 741 1 675 472
Total income at risk 235 922 182 928
* Replacement ratio 56% 57%
* Capitalisation factor 14.2 16.0
Actual cover -888 221 -671 993
Retail insurance -536 505 -377 675
Group insurance -351 716 -294 318
Government grants - -
Insurance gap 1 008 520 1 003 479
Cover adequacy (Cover/Need) 47% 40%
Number of earners (millions) 8.1 6.1
Total insurance gap (R’bn) 8 194 6 170
For example, a male earner with would typically need death cover of about R1.9m. Typically

such an earner would only have cover of R0.9m, leaving an insurance gap of R1.0m – implying

that only 47% of the life insurance need is protected by actual cover.

Disability insurance gap per earner

Table 32 - Derivation of the per-earner disability insurance gap by gender

Male Female
Insurance need 2 738 847 2 432 617
Total income at risk 235 922 182 928
* Replacement ratio 77% 81%
* Capitalisation factor 15.1 16.5
Actual cover -1 291 813 -1 101 208
Retail insurance -398 105 -282 889
Group insurance -602 359 -550 474
Government grants -291 349 -267 845
Insurance gap 1 447 034 1 331 409
Cover adequacy (Cover/Need) 47% 45%
Number of earners (millions) 8.1 6.1
Total insurance gap (R’bn) 11 757 8 186

For example, a female earner with would typically need disability cover of about R2.7m. Typically

such an earner would only have cover of R1.3m, leaving an insurance gap of R1.4m – implying

that 45% of the disability insurance need is protected by actual cover.

9.4 Responses to the insurance gap

Death insurance gap

The table below gives (for each of the segments) the degree of intervention required for each

of three possible responses to the insurance gap in the case of a death event:

Table 33 - Responses to the death insurance gap by gender

Male Female All

Purchase Cost of insurance (as % of
4.2% 4.9% 4.5%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 1.1 1.5 1.3
equal to 1.5 times current
level)
Reduce
Required reduction in
household 31% 29% 30%
household expenditure
expenditure
Additional Extra income required per
5 642 5 614 5 630
income month (net of tax)
required Extra income as % of
29% 37% 32%
earnings pre-event

www.asisa.org.za Page 31 of 44
Disability insurance gap

For the disability event, the figures are as follows:

Table 34 - Responses to the disability insurance gap by gender

Male Female All

Purchase Cost of insurance (as % of
2.6% 2.7% 2.6%
additional current earnings)
Insurance Increase in current level of
cover (e.g. 1.5x implies
additional cover required 1.1 1.2 1.2
equal to 1.5 times current
level)
Reduce
Required reduction in
household 35% 30% 33%
household expenditure
expenditure
Additional Extra income required per
7 708 7 092 7 443
income month (net of tax)
required Extra income as % of
39% 47% 42%
earnings pre-event

www.asisa.org.za Page 32 of 44
APPENDIX A: DEFINITIONS

“Actual cover”

The actual cover is the total amount of existing insurance cover of various kinds. It includes
cover from insurers (retail and group-type cover), self-insurance pension schemes (like the

GEPF), and government disability grants.

Cover types designed to provide for shorter term expenses are excluded (consistent with the

definition for insurance need). Cover disregarded in actual cover include:

• Funeral Cover: It was pragmatically assumed that funeral cover provides for short-term
expenditure related to the death-event and is therefore not typically earmarked for post-
event provision.

• Road Accident Fund: The Road Accident Fund provides cover for expenses incurred

(medical and legal) as well as loss of support that are the result of certain types of road

accidents. It would not be suitable for an individual will take this into consideration when

doing a financial needs analysis.

• Compensation Fund: Workman’s compensation cover is of a short-term, immediate nature.

Since the definition of insurance need excludes such short-term costs, this source of cover
was excluded from the study.

• Short-term insurers: Short-term insurers also provide a degree of life and disability cover.

Given [a] the fairly modest quantum and [b] the complexities involved in obtaining detailed

data from providers this source was pragmatically excluded.

“Cover adequacy”

Cover adequacy is calculated as the actual cover as percentage of the insurance need. It

therefore reflects the extent to which the insurance need is covered by actual cover. A number
of 100% indicates no need for additional insurance.

“Death event”

For purposes of this report, a death event is defined as the death of an active earner. To

determine the death insurance gap, we essentially consider the separate death of each of the

earners in South Africa (assuming that all other members of the household survive) and then

aggregate the result over the 14.3 million people.

www.asisa.org.za Page 33 of 44
“Disability event”

For purposes of this report, a disability event is defined as total and permanent disability of an

active earner, i.e. where it is unlikely for the disabled person to return to work any time after the
event.

“Earner”

These are the individuals for which an insurance gap was calculated and aggregated to arrive at
the total gap for purposes of this study. To be included, an individual had to be [a] South

African, [b] earning a regular income (i.e., be an “active earner”) and [c] between the ages of 18

and 65.

Only individuals with the following main sources of incomes were counted as “active earners”:

salaries or wages, net profit from business activities, net profit from professional practices or

net profit from commercial farming. Individuals with other main sources of incomes were not

considered “active earners”, these sources include: social welfare grants, regular

allowances/remittances received from non-household members, regular receipts from pension

from previous employment, income from letting fixed property, interest received on deposits,

etc.

“Earnings”

Throughout this document reference to earnings implies annual payments for ordinary-time,

standard or agreed hours for all active earners before taxation and other deductions. It includes

salaries, wages, commissions, fees, and employer’s contributions (e.g. to pension, provident,
medical aid, sick pay and other funds). The definition includes bonuses (performance or

otherwise) and overtime payments.

“Insurance gap”

The insurance gap is defined as the difference between the insurance need and actual cover. It
therefore represents the total net additional cover that will be purchased by South African active

earners in the following situation:

• Those that are under-insured purchase additional cover, so that their actual cover equals
their Insurance need.

www.asisa.org.za Page 34 of 44
• Those that have adequate insurance cover in place, reduce their current actual cover to

reflect their respective insurance need. (This includes those that do not have an insurance

need, but do have actual cover, terminating their policies.)

“Insurance need”

This is the amount of cover required to meet the need that is created by the death event and/or

disability event. It excludes any short-term expenses related to the risk event. E.g. for the death
event, funeral costs were not considered. Neither was additional medical or equipment

expenditure that may be required as a result of the disability event.

It was assumed that the household maintains its current living standards after the death event

/ disability event. Expenditure post event changes only insofar as this event would lead to a
reduction or elimination of certain household expenses from that point forward.

It was assumed that an insurance need only exists up to intended retirement age. From this point

onwards it was assumed that prior retirement provision would fund the household’s

expenditure. This study therefore ignores the extent to which insufficient allowance may
currently be made by earners for post-retirement expenses.

“Retirement age”

Retirement age was taken to be between 60 and 65, depending on the earner’s current age: For
those younger than 58, it was assumed that retirement would take place at age 60. Earners

older than 63 were assumed to have intended retirement age of 65. The assumption for the
intended retirement age of earners aged between 58 and 63 was phased in between 60 and 65.

www.asisa.org.za Page 35 of 44
APPENDIX B: MODELLING NOTES - EARNER POPULATION

Where assumptions were required, we generally aimed to set these at objective “best estimate”
levels. However, where this proved difficult, our approach was to rather err in the direction that

would provide a lower insurance gap.

The base source of information in terms of the composition of the South African earner
population was the metadata from StatsSA’s Living Conditions Survey 2014/2015. The 14.3m
earners are represented by about almost 21,000 model points with suitable weights to ensure

objectively weighted representation.

Information was updated using a combination of the following sources:

• The latest General Household survey published by StatsSA

• Quarterly Labour Force Surveys (QLFS) published by StatsSA

• Quarterly Employment Statistics (QES) published by StatsSA

• Consumer Price Indices (CPI) published by StatsSA

For each of the model points representing a number of South African earners, the insurance gap

was determined as the difference between the insurance need and actual cover (see modelling

notes in Appendix C and Appendix D).

www.asisa.org.za Page 36 of 44
APPENDIX C: MODELLING NOTES - INSURANCE NEED

The modelling of the insurance need is best explained by considering its breakdown into three
components:

Table 35 - Insurance need represented as product of three numbers

More Total (R’bn) Average per earner (Rand)

information Death Disability Death Disability
Total income at risk Appendix C1 3 042 3 042 213 093 213 093
* Replacement ratio Appendix C2 57% 78% 57% 78%
* Capitalisation factor Appendix C3 14.9 15.6 14.9 15.6
Total 25 712 37 209 1 801 422 2 606 929

Appendix C1: Total Income at Risk

Definition

Total income at risk allowing for all income as per the definition (Appendix A) of earnings.

Decline in workforce since previous study = 2.9% pa

The Living Conditions Survey 2014/2015 reported 14.8m earners. For the 2019 study, to derive

the number of earners at the end of 2018, we allowed for growth of 5.4% (in total, not pa). For

the 2022 study, we allowed for a decline in the workforce of 8.4% (in total, not pa) from the
2019 study, as per the QLFS publication. Total number of earners end 2021 (14.3m) has

therefore decreased by 2.9% per annum from the previous study.

The sharp decrease in the workforce was a result of substantial rises in unemployment during

2020, because of the Covid-19 pandemic and related lockdowns. This decrease in the workforce

is substantiated by the QES publication and by information on the Trading Economics website.

The segments of the workforce that saw the greatest decline from the end of 2018 to the end
of 2021 were younger age groups and less educated groups. The following illustrates the point,

showing “1-in-how many” earners that were in the workforce at the end of 2018 were no longer
in the workforce by the end of 2021:

• Overall: 1-in-12

• Per age group:

o Under 30: 1-in-7
o 30-39: 1-in-11

o Over 40: 1-in-25

www.asisa.org.za Page 37 of 44
 • Per education group:

o Primary school not completed: 1-in-3

o Completed primary school, but not high school: 1-in-8

o Completed high school (and/or some diploma/degree): largely unchanged

Growth in average earnings since previous study = 3.8% pa

This study assumes that the average level of earnings increased by 3.8% pa since the previous
study. The main data source for arriving at this parameter was the QES publication.

This increase in average earnings is substantiated by BankservAfrica “take-home pay report”

and by information on the Trading Economics website.

Total level of earnings modelled

Combining the decline in earners (2.9% pa) and the growth in earnings (3.8% pa) results in a
modelled increase in total earnings of 2.0% pa:

Table 36 - Total earnings by earners increased by 2.0% pa

2019 study 2022 study Increase per annum

Number of earners (million) 15.6m 14.3m -2.9%
Average annual earnings (Rand) 183 649 213 093 3.8%
Total annual earnings (R'm) 2 862 903 3 041 512 2.0%

The 2.0% increase pa in total earnings modelled compares positively with the increase derived

from the QES publications, where the increase over the same three-year period was reported

as 2.3% pa.

Appendix C2: Replacement Ratio

The replacement ratio represents the proportion of the household earner member’s personal

income that “will be missed” after the death event or the disability event. It is calculated as the

“household budget deficit post-event” divided by personal income at risk. The “household
budget deficit post-event” is calculated (for each of the model points) as the difference

between:

• Household expenses post-event (an annual figure): This takes into account the fact that, in
a death event, expenses directly related to the earner considered will disappear from the

household expense budget.

www.asisa.org.za Page 38 of 44
• Household income post-event (an annual figure): This takes into account income that will

continue after the death event / disability event mostly from other earners and retired

household members.

Insurance proceeds are not taxed (the taxation on both life and disability insurance products is
now mostly based on ”tax-free benefits through post-tax premiums”). The portion of personal
income that was directed towards income tax is therefore excluded when the insurance need is

calculated.

For the death insurance need, the replacement ratio additionally takes the following into

account:

• Insurance cover for single-member families is only required to the extent that support

payments to other households formed part of income.

• When an earner dies, the household expenditure will be lower post-event. For modelling

purposes, we allocated total household expenses into different categories:

o Fixed expenses: Expenses that cannot sensibly be assigned / allocated to any

specific member in the household and would therefore not change materially much

should the family become smaller. Examples include expenditure on housing,

washing and cleaning expenditure, and domestic worker wages.

o Adult expenses: Post-event expenditure is adjusted by taking into account the

number of adults in the household before and after the event. Examples include

alcoholic beverages.
o People expenses: Post-event expenditure is adjusted by taking into account the

number of people in the household before and after the event. Examples include

food, clothing, reading matter, and stationary.

• To the extent that savings represent provision for retirement, it needs to remain in the

expense base as we are relying on these contributions to provide the household with an
income from the intended retirement age. To the extent that it represents wealth creation

though, it should be excluded from the expense base in line with definitions of insurance
need (maintenance of current standard of living).

For the disability insurance need, the replacement ratio also takes into account the savings-

element with wealth creation motive that will not be required in the post-event situation.

www.asisa.org.za Page 39 of 44
Appendix C3: Capitalisation Multiple

A capitalisation factor is calculated by determining the number of years that the earner would

still have contributed to the household up to retirement age only.

Generally speaking, the term is the period that household members would have remained

dependent on income at risk. As current retirement provision expenditure was retained in the

expense base, it is appropriate to allow the dependency duration to cease at what would have
been the retirement date.

In determining these factors, an interest rate that exceeds living expense inflation by 1.9% was
assumed.

www.asisa.org.za Page 40 of 44
APPENDIX D: MODELLING NOTES - ACTUAL COVER

Accurate calculation of the insurance need for each of the sample / model points on the
representative dataset (populated metadata from Statistics South Africa) is possible as all

information affecting it is available at this level. This, however, is not the case for actual cover.
The StatsSA datasets do not contain any information on product ownership or insurance cover.
Hence, additional resources had to be used to [1] determine the total level of actual cover as at
end 2021 and [2] assign this cover to each of the sample points in the dataset.

Appendix D1: ASISA questionnaires

The long-term insurance industry is the primary source of life and disability cover. A
questionnaire was sent to all relevant insurers who were requested to provide

• Information on total cover amounts in the format that this information is reported annually

to the regulator (the so-called TP2.1 and TP2.2 statements). These statements contain

information on the total payments (separately for Retail and Group cover and per cover

type) that would be made in the hypothetical scenario where all their policyholders were

subject to a separate and independent death event and disability event.

• Additional information that would allow adjustments to these figures for purposes of
determining the actual cover for this study. This includes information that would allow

approximate allocation of total cover amounts between different age groups, genders, and

socio-economic groups.

The following adjustments were made to the insurer-provided data:

• Capitalising disability income cover: Income disability cover was capitalised by discounting

regular payments. Payments were multiplied by annuity factors allowing for the term to

retirement (dependant on current age) as well as whether payments would escalate or not

and at which rate (supplied by most insurers).

• Translating socio-economic groups to income groups: For retail cover, insurers were

requested to provide information split per socio-economic group as per their own
definition / categorisation.

• Allowing for non-respondents: The response factor was quite high with all the major
insurers (by market share) providing information. Approximate allowance was made for the

(around 9% of market share) non-respondents.

www.asisa.org.za Page 41 of 44
• Excluding out-of-scope cover: As the study is concerned with the actual cover of active

earners only, we had to (approximately) exclude such (retail) cover held by retired and

unemployed individuals.

Appendix D2: Other data sources

Self-insured pension schemes provide a material proportion of total group risk cover.

Allowance was made for such cover based on approximations from publicly available
information (e.g. annual reports) as well as discussions with advisors to these schemes.

Government is a major source of disability cover through its disability income grant and the
study paid due consideration to the conditions for payment of these grants. Current

qualification criteria and levels were allowed for.

Appendix D3: Allocation of actual cover to each modelled earner - True South
models

This total level of actual cover (derived from the above sources) then had to be allocated to

each of the “model points” representing the South African earner population. This was done

separately for individual life cover, individual disability cover, group life cover, and group

disability cover using two True-South developed models which were calibrated using the data

sources described above:

• TSPO-model: The True South Product Ownership model returns the probability of a South

African earner having life or disability cover based on supplied information such as

education, age, income, marital status, family composition, and geography (per province).
• TSCL-model: The True South Cover Level models return the level of cover given that cover

does exist based on similar information required by the TSPO-model.

Genetic algorithm technology6 was used to solve the optimisation problem of fitting the model

parameters. As mentioned elsewhere, input for deriving the parameters were obtained from a
wide variety of sources.

6 A genetic algorithm is an experience-based technique for problem solving that mimics the process of natural evolution
(i.e. using concepts inspired by natural evolution, such as inheritance, mutation, selection, and crossover). This approach
is routinely used to generate useful solutions to optimization and search problems, including previously unsolvable,
complex non-linear problems.

www.asisa.org.za Page 42 of 44
The total actual cover for each of the sample points was derived by multiplication of the

probability of being insured (from the TSPO-model) with the average level of cover (from the

TSCL-model) for each of the four cover types (life vs. disability and retail vs. group).

Some interesting findings derived from the TSPO model are provided below:

• Income:

A person with earnings in the top 20% is about 2.4 times more likely to have life insurance (2.3

times for a disability insurance policy) compared to an otherwise identical person (including

education level, age, etc.), but with earnings in the second highest 20%. The following graph
shows the relative probability that earners have a retail life or disability insurance policy, per
income level (using the poorest 20% income group as a base, =1):

Figure 1 - Relative probability of having a retail life or disability policy, per income level

120
Relative prob of owning a policy

100

80

60

40

20

0
Poorest 20% Q2 Q3 Q4 Richest 20%

Income level
Retail death Retail disability

• Level of education

There is a very strong correlation between insurance policy ownership and highest level of
education attained. The likelihood of a matric graduate having a life insurance policy is 4.8
times higher than an otherwise identical person (with regard to income, age, gender, family

composition, marital status), but with only a primary school education. The relativity is more
pronounced for disability policy ownership, with the factor being 6.7. The following graph
shows the relative probability that earners have a retail life or disability insurance policy, per

education level (using the primary school or lower group as a base, =1):

www.asisa.org.za Page 43 of 44
Figure 2 - Relative probability of having a retail life or disability policy, per education

Relative prob of owning a policy 25

20

15

10

0
PS or lower Some HS Matric Dipl Degree

Education
Retail death Retail disability

• Age:

A 50 to 54-year-old person is 5.2 times more likely to own a life insurance policy (2.1 times for

a disability insurance policy) compared to an otherwise identical person under the age of 30.

The following graph shows the relative probability that earners have a retail life or disability

insurance policy, per age group (using the under 30 group as a base, =1):

Figure 3 - Relative probability of having a retail life or disability policy, per income level

6
Relative prob of owning a policy

0
under 30 30-39 40-49 50-54 55 and over

Age
Retail death Retail disability

• Province: The above factors adequately explain the extent of likely insurance cover, except

for group insurance where there is a much-enhanced probability of group life cover and /
or disability cover if an earner resides in Gauteng or the Western Cape.
• Group cover: The main determinants for the level of group cover are salaries and wages,

age, and education level.

www.asisa.org.za Page 44 of 44