Summary For Joint Life and Last Survivor Benefits

Principal of actuarial Model
IRERA Regine 220008920

CBE-APPLIED STATISTICS
July 2022
Tables of contents
Joint life and last survivor benefits

Joint life notation
independent future lifetimes
A multiple state model for independent future
lifetimes
A model with dependent future lifetimes
The common shock model
Notes and further reading
IRERA Regine 220008920 (CBE-APPLIED STATISTICS)
Principal of actuarial Model July 2022 2 / 49
Basic Distribution I
Insurance benefits which are dependent on the joint

mortality of two lives, typically a married couple,
form an important part of the insurance portfolio.
t = 0, (x) and (y) are alive, are aged x and y,

respectively, at t = 0, and are partners in some joint
life contingent benefit context.

Basic Distribution II
We two most common types of benefit which are

contingent on two lives which are
1.A joint life annuity:Its annuity that payable until

the first death of (x) and (y).
2.A last survivor annuity:Its covers two lives

under one policy. And an annuity payable until the
second death of (x) and (y).

Basic Distribution III
which means the death benefit is paid after the
second person covered under the policy
dies.Actually, premiums continue to be paid after
the first insured dies.
Other definition we have know
A reversionary annuity: is a life annuity that

starts payment on the death of a specified life, say
(x), if (y) is alive at that time, and continues
through the lifetime of (y).

Basic Distribution IV
Moreover, A pension plan may offer a reversionary

annuity benefit as part of the pension package,
where (x) would be the pension plan member, and
(y) would be a partner eligible for spousal benefits.
A joint life insurance: it’s a life insurance policy

for two people pays a death benefit on the first
death of (x) and (y).

Basic Distribution V
A last survivor insurance: Last-survivor or

second-to-die life insurance covers two lives under
one policy. The death benefit is paid after the
second person covered under the policy dies.
Generally, premiums continue to be paid after the

first insured dies. There pays a death benefit on the
second death of (x) and (y).

Basic Distribution VI
contingent insurance:it refers to a policy that is
contingent on the absence of other insurance. pays
a death benefit contingent on multiple events. The
most common is a benefit paid on the death of (x),
say, but only if (y) is still living at that time.
Joint lie notation and formula
Actually, we express the present values of the joint

life benefits in terms of random variables, so that we
can value them as the expected value of the present
value.
Basic Distribution VII
Then T is the future lifetimes of (x) and (y) are

represented by TX and TY Where its represent the
times until the first person and the second to die x
and y.
Time to first death TXy = min (Tx , Ty )
Time to last death Txy − = max (Tx , Ty )

Basic Distribution VIII
Moreover, we refer to the subscript, xy and xy − as a

status; xy (also written as {xy } is the joint life
status and xy is the last survivor status. Hence, Txy
and Txy − are random variables representing the time
until the failure of the joint life status and the last
survivor status, respectively.
(x) dies first so that the realized values of Tx and

Txy are the same, and consequently the realized
values of Ty and Txy are the same.

Basic Distribution IX
And (y) dies first so that the realized values of Ty

and Txy are the same, and consequently the realized
values of Tx and Txy are the same.
Other terms we should know
tPxy = Pr[(x) and (y) are both alive in t years]=

Pr [Txy > t]

Basic Distribution X
t qxy = Pr[(x) and (y) are not both alive in t
years]= Pr [Txy ≤ t]
µ|t qxy = Pr[(x) and (y) are both alive in u years,

but not in u + t years]= Pr [u < Txy ≤ u + t]
1
t qxy= Pr[(x) dies before (y) and within t years]=
Pr [Tx ≤ t and Tx < Ty ]
2
t qxy= Pr[(x) dies after (y) and within t years]=
Pr [Ty < Tx ≤ t]
Basic Distribution XI
¯ = Pr[at least one of (x) and (y) is alive in t
tP xy
years] = Pr [Txy¯ > t]
t qxy
¯= Pr[(x) and (y) are both dead in t years]=
Pr [Txy¯ ≤ t]
The following are the relationship among those

observation
Tx + TY = Txy + Txy −

Basic Distribution XII
ν Tx + ν Ty = ν Txy + ν Txy −
Ax :n⌉ is the probability of (x) dying first where the

benefit is paid only if x dies first, before the term, n
years, expires.
Joint life annuity: a continuous payment at unit rate

per year while both (x) and (y) are still alive:

−
axy =E aT−xy ⌉

Basic Distribution XIII
And, If there is a maximum period, n years, for the
annuity, then we refer to a ’temporary’ or ’term’
joint life annuity. The notation for the EPV is axy
−
:n⌉
with the formula of

−
axy :n⌉ =E −
amin(Txy )
⌉
ax¯|y Reversionary annuity a continuous payment

at unit rate per year, starting on the death of (x) if
(y) is still alive then, and continuing until the death
of (y):

Basic Distribution XIV
¯ ⌉ I (Tx < Ty )

ax¯|y = E Tx
ν aTy −Tx
A¯xy Contingent insurance:a unit payment

immediately on the death of (x) provided (x) dies
before (y):
Axy = E [ν Tx¯I (Tx < Ty )]
where I is the indicator function.

Basic Distribution XV
The unction for whole life annuity and insurance is
a¯xy = 1−A
δ
x
Independent future lifetimes
The first approach to modelling the survival of two

lives assumes that the future lifetime for each
individual is not affected in any way by the other life.

Basic Distribution XVI
Firstly, we have to know that the random variables
Tx and Ty re independent. So e have the survival
function which is tpx and tpy and we have they
forces o mortality which is µx +t and µy +t .
From our assumption of independence, we can write

for t ≤ 0
tpxy = Pr [Tx > t and Ty > t] = tPx tPy
¯ =Pr [Tx > t or Ty > t]=1 − tqx tqy

and tP xy

Basic Distribution XVII
tPx + tPy − tPx tPy
Husband Wife
x lx y ly
65 43302 60 47260
66 42854 61 47040
67 42081 62 46755
68 41351 63 46500
69 40050 64 46227
By using this example for The table above which

shows extracts from two life tables appropriate for a
Basic Distribution XVIII
husband and wife, who are assumed independent
with respect to mortality.
a. Calculate 3Pxy for a husband aged x = 66 and a

wife aged y = 60.
¯ for a husband aged x = 65 and a

b. Calculate 2Pxy
wife aged y = 62.
c. Calculate the probability that a husband,

currently aged 65, dies within two years and that his
wife, currently aged 61, survives at least two years.
Basic Distribution XIX
d. Explain the meaning o the symbol äxy :n⌉
e. Explain the meaning of the symbol äxy¯ :n⌉
f. Calculate äxy :5⌉ äxy¯ :5⌉ for a husband aged x = 65

and a wife aged y = 60 at a rate of interest 5% per
year.

Basic Distribution XX
Solution:
a. 3Pxy =3 px 3 py = lx (69)
lx (66) × ly (63)
ly (65) = 40050
42854 × 46500
47260 =
0.9195
¯ = lxlx (67)
b. 2P xy (65) +
ly (69)
ly (67) − lx (67)
lx (65) × ly (69)
ly (67)
= 42081
43302 + 46227
46755 − 42081
43302 × 46227
46755 = 0.9997
c. Since the two lives are assumed to be independent

with respect to mortality, the required probability is

Basic Distribution XXI

1− lx (67)
lx (65) × ly (68)
ly (66)
1− = 0.0279

42081

46500
43302 × 47040
d. The symbol äxy¯ :n⌉ represents the EPV, at a given

constant rate of interest, of a series of at most n
annual payments , each of unit amount with the
first payment due now, with each payment being
made only if the lives (x) and (y) are both alive at
the time the payment is due.

Basic Distribution XXII
e. The symbol äxy¯ :n⌉ represents the EPV, at a given
constant rate of interest, of a series of at most n
annual payments, each of unit amount with the first
payment due now, with each payment being made
only if at least one of (x) and (y) is alive at the time
the payment is due.
f. From the definitions in parts (d) and (e), we can

write down the following formulae
äxy¯ :5⌉ =
P4 t
t=0 ν tPxy

Basic Distribution XXIII
äxy¯ :5⌉ = ¯
P4 t
t=0 ν tP xy
Where ν = 1.05
1
x = 65 andy = 60. These are
derived in exactly the same way as formula (5.10).
The numerical values of the annuities are
äxy :5⌉ = 4.3661 and äxy¯ :5⌉ = 4.5437
A multiple state model for independent future

lifetimes

Basic Distribution XXIV
In this, we described how the single life future

lifetime random variable Tx is related to a two state
multiple state model, which is called the alive-dead
model.
Where Loosely,’ for a life aged x, where Tx is the

time to the transition from state 0 "alive" to state 1
"dead".

Basic Distribution XXV
00
tPxy =Pr[(x) and (y) both alive at time t | both alive
at time 0]
01
tPxy =Pr[(x) and (y) both alive at time t | both alive
at time 0]
and for 0 < s < t,

22
= Pr [(y )alive at time t+s | (y) alive and (x)
t Py +s
dead at time S]

Basic Distribution XXVI
From the assumption, the force of mortality for (x)
at age x + t is µm
x +t , whether (y) is then alive or not.
Hence, Tx has a distribution determined by

x +t }t≥0 . Same as, Ty has a distribution
{µm
determined by {µfx +t }t≥0
The joint survival probability for subsequently, (x)

dies and (y) survives in the interval from time s to
time t.

Basic Distribution XXVII
Pr[Tx > s and Ty > t]=t Pxy

00
+s Pxy
00 02
t−s Px +s:y +s ,
Moreover, Pr[Tx > s and Ty > t]=t Pxy

00
+s Pxy
00
02
t−s Px +s:y +s ,
= tPx tPy + tPy (sPx − tPx )
= sPx tPy
=Pr [Tx > s]Pr [Ty > t

Basic Distribution XXVIII
Example: Using the multiple state model, with the

independence for assumption 1 write down integral
equations for each of the following, and describe the
benefit being valued.
a. A¯xy
01
b. A¯xy + A¯02
01
xy
c. a¯xy
01

Basic Distribution XXIX
Solution:
a. A¯xy = 0∞ e −δt tPxy µy +t dt = A¯x :y

01 R 00 f
Where µ01
x +t:y +t = µy +t
f
This is the EPV of a unit benefit paid on the death

of (y) provided (x) is still alive at that time.
b. Ā01
xy + Āxy =
R ∞ −δt
+ µm ¯
x +t dt = Axy

02 00 f
0 e tPxy µy +t

Basic Distribution XXX
This is the EPV of a unit benefit paid on the first

death of (x) and (y).
c. āxy
01
= 0∞ e −δt tPxy
01
dt =
R
tPx (1 − tPy ) dt = āx − āxy = āy |x

R ∞ −δt
0 e
This is the EPV of a unit reversionary annuity

payable to (x) after the death of (y).
For A model with dependent future lifetimes

Basic Distribution XXXI
The following three factors are often cited as
sources of dependence between married partners. for
couples who purchase joint life insurance or annuity
products.
1. The death of the first to die could adversely

affect the mortality of the survivor. This is
sometimes called the ’broken heart syndrome’.
2. The two lives could die together in an accident.

This is called the ’common shock’ risk.

Basic Distribution XXXII
3. The two lives are likely to share a common

lifestyle. like mortality is related to wealth and levels
of education.
Actually, Married couples tend to have similar levels

of wealth and similar levels of education. They may
also share interests, for example, related to health
and fitness.

Basic Distribution XXXIII
State Dependent Mortality Assumption. The force

of mortality for each life depends on whether the
other partner is still alive. If the partner is alive, the
intensity may depend on the exact age of the
partner, as well as the age of the life being
considered.
And If one partner has died, the transition intensity

for the survivor depends only on the survivor’s age
and state.

Basic Distribution XXXIV
x +t:y +t is the intensity of mortality for (y) at age y

µ01
+ t given that (x) is still alive and aged x + t.
Moreover, if one partner, let say (x), has died, the

intensity of mortality for (y) depends on her current
age, and the fact that (x) has died, but not on how
long he has been dead.
So This model allows for some dependence between

the lives; the death of, say, (x) affects the transition

Basic Distribution XXXV
intensity of (y). It does not allow for both lives

dying simultaneously.
ii
Formula: tPxy = tPxy
ii
or i=0,1,2,3,..
= exp − + µ02
n R t 01 o
00
tPxy 0 µx +s:y +s x +s:y +s ds
= exp −
n Rt o
11
t Px 0 µ01
x +s ds
tPy22 = exp −
n Rt o
0 µ23
y +s ds

Basic Distribution XXXVI
=
01 Rt 00 01
Example:tPxy 0 sPxy µx +s:y +s t − sPx11+s ds
You can ask the difference between t qxy

1
and t Pxy
02
.
So, The probability t qxy

1
is the probability that (x)
dies within t years, and that (y) is alive at the time
of (x) ’s death. And it allows for the possibility that
(y) dies after (x), within t years.

Basic Distribution XXXVII
While, The probability t Pxy

02
is the probability that
(x) dies within t years, and that (y) is alive at time t
years.And does not allows or the possibility that (y)
dies after (x) within t years.
1
t qxy= 0∞ rPxy µx +r :y +r dr This is the probability that
00 02
R
(x) dies within t years, and that (y) is alive at the

time of the death of (x).

Basic Distribution XXXVIII
2
tqxy This is the probability that (x) dies within t
years and that (y) is already dead when (x) dies,
conditional on (x) and (y) both being alive at time
0.
Example: Derive the following expression for the

probability that (x) has died before reaching age x +
t, given that (x) is married to (y) at time 0:
x +s:y +s ds +
Rt 00 Rt Rs
0 sPxy µ02 00 01 11
0 0 µPxy µx +u:y s−u Px +u du
µ13
x +s ds

Basic Distribution XXXIX
Solution:For (x) to die before time t we require the
process either to
The total probability of these events, integrating

over the time of death of (x), is
µx +s:y +s ds +
Rt 00 02 R∞ 01 13
0 sPxy 0 sPxy µx +s ds
= exp − + µ02
n R s 01 o
00
sPxy 0 µx +u:y +u x +u:y +u du
=
01 Rs 00 01
sPxy 0 uPxy µx +u:y +u s − uPx11+u du

Basic Distribution XL
Āy = 0∞ e −δt (tPxy µx +t:y +t + tPxy

00 01
µy +t )dt this is the
02 23
R
formula For the EPVs of the lump sum payments.
For The common shock model
We can extend this dependence by allowing for (x)

and (y) to die at the same time, for example as the
result of a car accident.

Basic Distribution XLI
Let ues example, In the context of the common
shock model, calculate 10 P70:65
00
, 10 P70:65
01
and 10 P70:65
02
let use the following transition intensities

y +t
x +t:y +t = Bf Cf
µ01
x +t:y +t = Bm Cm
µ02 x +t
x +t = Cm dm
µ13 x +t
y +t = Cf d
µ23 y +t

Basic Distribution XLII
Bf = 9.741 × 10−7 cm = 1.0989
Bm = 2.622 × 10−5 dm = 1.0725
Cm = 3.899 × 10−4 df = 1.1020
Example: A husband and wife, aged 63 and 61,

respectively, have just purchased a joint life 15-year
term insurance with sum insured $200 000 payable
immediately on the death of the first to die, ·or on
their simultaneous deaths, within 15 years. Level

Basic Distribution XLIII
premiums are payable monthly for at most 15 years
while both are still alive.
Calculate
a. the monthly premium
b. the gross premium policy value after 10 years,

assuming both husband and wife are still alive.
Use the common shock model with the following

basis.
Basic Distribution XLIV
Interest: 4% per year
Survival model: As in Example 9.10.
Expenses: Initial expenses of

Renewal expenses of 10% of all premiums
$200 on payment of the sum insured

Basic Distribution XLV
Solution:
a. The EPV of a unit sum insured is
0 ν tP63+t:61+t (µ63+t:61+t + µ02

63+t:61+t + µ63+t:61+t )
R 15 t 00 01 03
and the EPV of a unit premium per month is

(12)
12ä63:61:15⌉ =
P179 t/12 00
t=0 ν t/12P63:61

Basic Distribution XLVI
And let µ00
x +t:y +t = R=1 µx +t:y +t then the formula
3 0K P
for the monthly premium, P is
p = (200200 X 0.25574 + 500)/(0.9 X 12 X

9.87144) = = $484.94
B. The gross premium policy value at duration 10

years, just before the premium then due is paid, is
10 V=
20020 05 ν t tP73:71 + +

00
µ01 µ02
µ03
73+t:71+t d
R
73+t:71+t 73+t:71+t

Basic Distribution XLVII
P59 t/12 00
−0.9P t=0 ν t/12P73:71
= 200200 X 0.17776-0.9 X 484.94 X 12 X 4.14277

= $13890


Summary For Joint Life and Last Survivor Benefits

Uploaded by

Copyright:

Available Formats

Summary For Joint Life and Last Survivor Benefits

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Summary For Joint Life and Last Survivor Benefits

Uploaded by

Copyright:

Available Formats

Principal of actuarial Model

IRERA Regine 220008920

Joint life and last survivor benefits

Insurance benefits which are dependent on the joint

t = 0, (x) and (y) are alive, are aged x and y,

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

We two most common types of benefit which are

1.A joint life annuity:Its annuity that payable until

2.A last survivor annuity:Its covers two lives

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

Other definition we have know

A reversionary annuity: is a life annuity that

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

Moreover, A pension plan may offer a reversionary

A joint life insurance: it’s a life insurance policy

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

A last survivor insurance: Last-survivor or

Generally, premiums continue to be paid after the

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

Actually, we express the present values of the joint

Then T is the future lifetimes of (x) and (y) are

Time to first death TXy = min (Tx , Ty )

Time to last death Txy − = max (Tx , Ty )

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

Moreover, we refer to the subscript, xy and xy − as a

(x) dies first so that the realized values of Tx and

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

And (y) dies first so that the realized values of Ty

Other terms we should know

tPxy = Pr[(x) and (y) are both alive in t years]=

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

µ|t qxy = Pr[(x) and (y) are both alive in u years,

The following are the relationship among those

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

Ax :n⌉ is the probability of (x) dying first where the

Joint life annuity: a continuous payment at unit rate

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

ax¯|y Reversionary annuity a continuous payment

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

A¯xy Contingent insurance:a unit payment

Axy = E [ν Tx¯I (Tx < Ty )]

where I is the indicator function.

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

The unction for whole life annuity and insurance is

Independent future lifetimes

The first approach to modelling the survival of two

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

From our assumption of independence, we can write

tpxy = Pr [Tx > t and Ty > t] = tPx tPy

¯ =Pr [Tx > t or Ty > t]=1 − tqx tqy

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)

By using this example for The table above which

a. Calculate 3Pxy for a husband aged x = 66 and a

¯ for a husband aged x = 65 and a

c. Calculate the probability that a husband,

d. Explain the meaning o the symbol äxy :n⌉

e. Explain the meaning of the symbol äxy¯ :n⌉

f. Calculate äxy :5⌉ äxy¯ :5⌉ for a husband aged x = 65

IRERA Regine 220008920 (CBE-APPLIED STATISTICS)