GSH Financial mark GSH Financial
(431) 305-0352 Book a Consultation
GSH FinancialGSH Financial

Home / Financial Tools / Variable Annuity GMWB Guarantee Monte Carlo Pricing Engine

Graduate-Level Modeling · Actuarial Science & Insurance Product Valuation

Variable Annuity GMWB Guarantee Monte Carlo Pricing Engine

A Guaranteed Minimum Withdrawal Benefit promises a fixed annual withdrawal for life, or for a set period, even after market losses have driven the underlying account value to zero. That promise is a long-dated, path-dependent put option written by the insurer, and this engine prices exactly that option by simulation.

How To Use This Model

Reading This Tool

Set the initial account value, the guaranteed annual withdrawal rate, the account's expected return and volatility, and the guarantee period. Each simulated path grows the account under geometric Brownian motion and subtracts the guaranteed withdrawal every year; whenever the account can no longer fully fund that withdrawal, the insurer's guarantee makes up the shortfall, and continues funding the full withdrawal for every year remaining.

The guarantee's present value is the risk-neutral expected present value of every dollar the insurer ends up paying that the account itself could not, averaged across thousands of simulated paths. Push the withdrawal rate up, or volatility up, and watch this liability grow, exactly the sensitivity that has made GMWB riders one of the more consequential tail risks on insurer balance sheets since the 2008 financial crisis.

Contract & Market Parameters

Pricing is risk-neutral: the account is projected at the risk-free rate for valuation purposes, while "expected return" above lets you separately explore how the account's real-world growth assumption affects depletion probability. Guaranteed withdrawals are funded first from the account, then from the guarantee once the account is exhausted.

What The Guarantee Is Actually Worth

Guarantee Present Value

-

As % Of Initial Account Value

-

Probability Of Account Depletion

-

95% Confidence Half-Width

-

Guarantee Value Sensitivity To Withdrawal Rate

-

-

When And How Accounts Run Dry

Average Depletion Year (If Depleted)

-

Average Terminal Account Value

-

Guaranteed Annual Withdrawal ($)

-

Years Of Withdrawal If Never Depleted

-

Distribution Of Account Depletion Year (Depleted Paths Only)

Why This Is A Put Option In Disguise

-

Reading The Withdrawal-Rate Sensitivity

-

Why Insurers Got This Wrong Before 2008

Many GMWB riders written in the mid-2000s were priced using historical equity return assumptions that dramatically understated tail risk, and often without dynamically hedging the resulting embedded option at all. When markets fell sharply in 2008, guarantee liabilities exploded in exactly the mechanical way this tool demonstrates, driving several insurers to withdraw the product entirely or reprice it far more conservatively, and turning GMWB hedging into one of the more sophisticated corners of insurance risk management.

Actuarial Science & Insurance Product Valuation

The Method

AVt = AVt-1·exp[(r−σ²/2) + σZt] − W,  W = rate × AV₀
Shortfallt = max(W − AVtavailable, 0),  triggered once AV first cannot fund W in full
Guarantee PV = EQ[Σt Shortfallt·e−rt]

The account value is simulated under the risk-neutral measure (drift equal to the discount rate, standard for guarantee valuation, since the account is a traded, hedgeable asset even though the guarantee payout is not). This implementation was validated against exact deterministic edge cases: with zero volatility and a growth rate equal to the discount rate, the guarantee value converges to zero (the account, growing at exactly the pricing rate, can always self-fund at low enough withdrawal rates); and with zero volatility and no growth, the exact year and size of account depletion is computable by hand, and the simulated guarantee value matched that hand calculation to the cent.

When To Actually Use This Model

  • Pricing new GMWB, GMIB or GLWB rider designs before they go to market, or re-pricing existing blocks as market assumptions shift.
  • Quantifying the tail sensitivity of a guarantee book to equity market drawdowns, essential input to an insurer's own hedging and capital adequacy decisions.
  • Explaining, to a non-specialist audience, why guarantee riders became so much more expensive (or were withdrawn) after periods of realized market volatility.

Key Assumptions & Limitations

  • No mortality or lapse assumptions are modeled here, a full GMWB/GLWB valuation incorporates policyholder mortality, lapse rates (which vary with moneyness of the guarantee), and often policyholder behavior modeling far more complex than pure financial option pricing.
  • Account returns follow simple geometric Brownian motion; real variable annuity sub-accounts often have fund-of-funds structures, fees, and asset allocation rules that affect the actual return distribution.
  • This engine prices the guarantee alone; the insurer's net economics also include rider fees collected from the policyholder, which are not included in the value shown here.

Foundational References

Milevsky, M. A., & Salisbury, T. S. (2006). Financial Valuation of Guaranteed Minimum Withdrawal Benefits. Insurance: Mathematics and Economics, 38(1), 21-38.

Bacinello, A. R., Millossovich, P., Olivieri, A., & Pitacco, E. (2011). Variable Annuities: A Unifying Valuation Approach. Insurance: Mathematics and Economics, 49(3), 285-297.

Ledlie, M. C., et al. (2008). Variable Annuities. British Actuarial Journal, 14(2), 327-389.

Need the underlying mortality assumptions this guarantee depends on?

Open The Lee-Carter Mortality Projection Engine →