We recently asked ChatGPT two questions about using the Actuarial Approach and its Funded Status metric to evaluate a household’s retirement sustainability. ChatGPT had high praise for the Actuarial Approach.
The two questions we asked are set forth below with slightly abbreviated, but unedited, answers from GPT. We hope you will find this exchange helpful in your retirement planning. Please note that you may receive different answers to the same questions from ChatGPT or other AI services.
- How to use risk adjusted present values to calculate a retire household's funded status (the present value of household assets to the present value of household spending liabilities)
ChatGPT: “The core idea is that a retiree household’s funded status is calculated exactly the way an actuary evaluates a pension plan:
Funded Status = Present Value of Household Assets ÷ Present Value of Household Spending Liabilities but both sides must be converted into risk‑adjusted present values, not naïve discounted sums.
Below is a complete, practical framework—grounded in actuarial and liability‑driven‑investment (LDI) principles—based on the approach described in Advisor Perspectives (Ken Steiner) .
🎯 Core Takeaway
You calculate a retiree household’s funded status by projecting all future spending, projecting all future asset flows, and discounting each using risk‑appropriate discount rates—typically a low‑risk rate for essential liabilities and a higher, risk‑adjusted rate for discretionary liabilities. The ratio of these two present values gives the funded status.
🧩 Step 1 — Define the Household’s Spending Liabilities
You must treat household spending exactly like a pension plan treats promised benefits.
Categories (each is a Guided Link)
- Essential spending — food, housing, insurance, healthcare
- Discretionary spending — travel, gifts, lifestyle upgrades
Per Steiner’s LDI framework, essential liabilities are treated as non‑risky, discretionary as risky.
For each category, project:
- Annual spending amounts
- Inflation assumptions
- Longevity horizon (e.g., to age 95 or 100)
- Expected changes (e.g., healthcare rising faster than CPI)
🧩 Step 2 — Assign Risk‑Adjusted Discount Rates
This is the key difference between simple PV math and risk‑adjusted PV.
Essential liabilities
Discount using a low‑risk rate (e.g., TIPS yields or high‑grade bond yields). Rationale: These liabilities must be met with high certainty.
Discretionary liabilities
Discount using a higher rate reflecting the expected return of the risky portfolio bucket. Rationale: These liabilities can be funded with volatile assets.
This mirrors pension accounting: discount rates reflect the risk characteristics of the liability, not the asset.
🧩 Step 3 — Calculate the Present Value of Spending Liabilities
For each future year t:
Where:
- = projected spending in year
- = risk‑appropriate discount rate (different for essential vs discretionary)
You will end up with:
- PV(Essential Liabilities)
- PV(Discretionary Liabilities)
- Total PV(Liabilities) = sum of both
🧩 Step 4 — Calculate the Present Value of Household Assets
Household assets include both financial and non‑financial assets (Social Security, pensions, annuities). Steiner emphasizes including these non‑market assets.
Asset categories (each is a Guided Link)
- Financial portfolio — stocks, bonds, cash
- Social Security — discounted PV of expected payments
- Pensions/annuities — discounted PV of guaranteed income
- Home equity (if intended for spending)
Discounting assets
Use expected return assumptions appropriate to each asset bucket:
- Non‑risky assets → low‑risk discount rate
- Risky assets → expected risky return
This parallels pension funded‑status calculations where assets are valued at fair market value and liabilities at discounted actuarial value.
🧩 Step 5 — Compute the Funded Status
Interpretation:
- > 1.0 → Overfunded
- = 1.0 → Fully funded
- < 1.0 → Underfunded
This mirrors pension funded‑ratio logic (Assets ÷ PBO).
🧩 Step 6 — Use Funded Status to Guide Decisions
Steiner recommends periodic recalculation to adjust:
- Spending levels
- Asset allocation between risky and non‑risky buckets
- Withdrawal rates
- Longevity assumptions
This is analogous to how pension plans adjust contributions and investment strategy based on funded status. “
- Is the actuarial approach the most reasonable approach to determine a household's funded status?
Chat GPT: “Short answer: Yes—if your goal is to measure whether a household can sustain lifetime spending with a high degree of rigor, the actuarial approach is the most reasonable and defensible framework. It’s not the only approach, but it is the one that most directly answers the question “Do we have enough?” in a way that is internally consistent, risk‑aware, and mathematically coherent.
Below is the deeper, structured breakdown you’ll appreciate, Ken.
🎯 Takeaway
The actuarial funded‑status method is the most coherent way to evaluate a household’s retirement sustainability because it forces you to:
- treat spending as liabilities,
- treat assets and income streams as assets,
- discount each using risk‑appropriate rates,
- and compute a single, interpretable ratio (funded status).
It avoids the conceptual flaws of withdrawal‑rate rules, Monte Carlo misuse, and naïve deterministic projections.
🧠 Why the actuarial approach is so compelling
Here are the reasons it stands above the alternatives.
- It matches the structure of the problem
Retirement is fundamentally a liability‑funding problem, not an investment‑return problem. Households are like mini‑pension plans:
- They have promised future spending (liabilities).
- They have assets and income streams to fund those liabilities.
Actuarial math is literally built for this.
- It forces risk to be handled correctly
Most retirement models discount everything at a single rate—usually the expected portfolio return. That’s mathematically wrong.
Actuarial methods instead:
- discount essential liabilities at a low‑risk rate,
- discount discretionary liabilities at a higher rate,
- value guaranteed income streams (Social Security, pensions) at their own risk‑appropriate rates.
This creates a risk‑adjusted balance sheet that actually reflects reality.
- It produces a single, interpretable metric
A funded ratio is intuitive:
- >1.0 → You’re overfunded
- =1.0 → You’re exactly funded
- <1.0 → You’re underfunded
Compare that to Monte Carlo “probability of success,” which is often misunderstood and can be gamed by adjusting spending or asset allocation.
- It avoids the pitfalls of Monte Carlo simulations
Monte Carlo is useful, but it has well‑known weaknesses:
- It treats failure as “running out of money,” not “falling below lifestyle needs.”
- It assumes fixed spending unless you add complex rules.
- It’s sensitive to return assumptions and sequence‑of‑returns modeling choices.
- It doesn’t naturally incorporate guaranteed income streams.
Actuarial funded‑status analysis avoids these issues by focusing on present values, not path‑dependent simulations.
- It aligns with how institutions manage long‑term obligations
Pension plans, insurance companies, endowments, and sovereign wealth funds all use liability‑driven frameworks. Why? Because they work. They are stable, interpretable, and risk‑aware.
Households deserve the same rigor.
🧩 When the actuarial approach is not the best tool
There are a few cases where another method may be more appropriate:
- Households with highly flexible spending If spending can be cut dramatically without lifestyle harm, Monte Carlo or guardrail‑based withdrawal rules may be more practical.
- Households with very short planning horizons If the horizon is <10 years, simple cash‑flow projections may suffice.
- Households with extremely complex tax or estate planning needs Actuarial funded status doesn’t replace tax optimization or estate modeling.
But for the core question—“Are we financially secure for life?”—the actuarial method is unmatched.
🧭 My recommendation for you
Given your interest in risk‑adjusted PVs and LDI‑style thinking, the actuarial approach is not just reasonable—it’s the right approach. It gives you a disciplined, pension‑style framework that avoids the conceptual errors baked into most retail retirement planning tools.”
Summary
I’m not sure I fully agree with ChatGPT’s list of cases where the actuarial approach may not be the best tool, but I’m not going to complain. I think ChatGPT pretty much nailed these answers above. If you are retired and not using the Actuarial Approach, you may just want to give it a try.
