Calculate premiums for insurance and annuities on joint lives using the equivalence principle.

A husband age 65 and wife age 62 want a policy that pays $500,000 at the second death and a reversionary annuity to whichever spouse survives. Pricing it requires two different "lives," each built from the same two underlying mortalities.

A status is a life-like entity that is "alive" while a condition holds and "dies" when it fails. Two statuses dominate joint pricing:

• Joint life status $(xy)$: alive while both $x$ and $y$ are alive. Time to failure $T_{xy} = \min(T_x, T_y)$.
• Last survivor status $(\overline{xy})$: alive while at least one is alive. Time to failure $T_{\overline{xy}} = \max(T_x, T_y)$.

Because the min plus the max equals the sum, you get the most useful identity in the topic:

$$A_{xy} + A_{\overline{xy}} = A_x + A_y, \qquad \ddot a_{xy} + \ddot a_{\overline{xy}} = \ddot a_x + \ddot a_y$$

KEY: Compute the easier of each pair (usually joint-life under independence), then back out the other via the identity.

When $T_x$ and $T_y$ are independent:

Common mistakes

• Using ${}_tp_x \cdot {}_tp_y$ under dependence. The product rule needs independence. With common shock rate $\lambda$, multiply by $e^{-\lambda t}$ or your ${}_tp_{xy}$ overstates joint survival and the premium is too low. Common lifestyle and broken-heart effects also violate independence and require explicit modeling.
• Wrong premium-paying status. Plugging $\ddot a_{\overline{xy}}$ into the denominator when premiums actually stop at the first death undercharges by 20 to 40 percent. Read the contract twice.
• Sign slip in the additive identity. $A_{\overline{xy}} = A_x + A_y - A_{xy}$, not plus. Forgetting the minus inflates the last-survivor insurance by the full $A_{xy}$. The same sign error breaks the term-version identity.

Bottom line

• Joint life status $(xy)$ fails at the first death, i.e. $\min(T_x,T_y)$; last survivor $(\overline{xy})$ fails at the second.
• Additive identity: $T_{xy}+T_{\overline{xy}}=T_x+T_y$ gives $A_{\overline{xy}}=A_x+A_y-A_{xy}$ (minus, not plus) and $\ddot a_{xy}+\ddot a_{\overline{xy}}=\ddot a_x+\ddot a_y$; holds for term versions too.
• Independence: ${}_t p_{xy}={}_t p_x\cdot{}_t p_y$ and $\mu_{xy}=\mu_x+\mu_y$; common shock adds $\lambda$ (multiply by $e^{-\lambda t}$); lifestyle and broken-heart also break independence.
• Recursions: ${}_{n+m}E_{xy}={}_{n}E_{xy}\cdot{}_{m}E_{x+n:y+n}$, ${}_{n}E_{xy}=v^n\,{}_{n}p_x\,{}_{n}p_y$, and $\ddot a_{xy:\overline{n+m}|}=\ddot a_{xy:\overline{n}|}+{}_{n}E_{xy}\,\ddot a_{x+n:y+n:\overline{m}|}$; use ${}_{n}E_{xy}$, not ${}_{n}p_{xy}$.

Exam shortcut

Build a 2×2 table of single-life and joint-life EPVs first. Last-survivor values fall out by subtraction; you almost never integrate twice. DECISION: The premium denominator follows the premium-paying status, not the benefit status. Read "premiums payable until..." and match the annuity to it. When a problem gives $\ddot a_x$, $\ddot a_y$, and $\ddot a_{xy}$, you already know $\ddot a_{\overline{xy}}$, the reversionary annuity, and (via $A = 1 - d\ddot a$) all three insurances. Convert once, then assemble.

The full lesson (about 11,634 words, 78 min read) adds 10 worked examples, all 17 common mistakes, a self-check, free in the app.

Learning objectives

• 3c

Browse all free Exam ALTAM lessons or jump into free Exam ALTAM practice questions.