Parametric Triggers And Physical Measurements

Introduction

A parametric catastrophe bond uses a defined physical measurement to determine whether, and how much, a contractual payment is due. The measurement might be wind speed at named stations, ground acceleration recorded by specified instruments, or rainfall accumulated over a stated period. The calculation is only as clear as the contract’s definitions: the measured quantity, units, locations, observation window, and treatment of threshold boundaries all matter.

This tutorial focuses on translating a specified measurement into the payment stated by a trigger schedule. It does not require estimating an insurer’s losses. A parametric payment follows the contractual measurement rule, whether or not that amount matches the actual loss from the event.

Definition: A parametric trigger is a contractual rule that uses one or more specified physical measurements to determine a payment. The contract must define the measurement and its source, the relevant event or observation period, any locations or aggregation rule, the threshold schedule, and how boundary values are treated. The resulting payment is the amount produced by applying that rule to the qualifying measurements.

Read the Measurement Rule Before Calculating

A number alone does not establish whether a trigger is met. For example, “wind speed of 150” is incomplete unless the contract says what kind of wind speed is measured, in what units, where it is observed, and over what period. A one-minute sustained speed and a short-duration gust are not interchangeable measurements.

Before evaluating a schedule, identify the following items from the contract. They are part of the calculation, not background detail.

Measured quantity and unit: such as maximum three-second gust in miles per hour, peak ground acceleration in units of gravity, or total rainfall in millimeters.
Source and location: the named station, instrument, or other specified data source.
Observation window: the event period, time interval, or accumulation period used to obtain the measurement.
Aggregation rule: for example, the maximum reading across listed stations or the sum of daily readings during a specified period.
Payout schedule: the payment associated with each measurement range, including which endpoint belongs to which range.

If the contract specifies the maximum reading from several stations, use the maximum; do not average the readings. If it specifies accumulated rainfall, use the total over the stated period; do not substitute a daily average. If a reading is unavailable or invalid, the contract’s data rules govern. A calculation should not silently introduce a substitute source or a new aggregation rule.

Continuous Payout Schedules

A common continuous schedule pays nothing at or below an attachment measurement $a$, increases proportionally between $a$ and an exhaustion measurement $b$, and reaches a maximum payment $M$ at or above $b$. Assume $b>a$, the qualifying measurement $x$ is defined by the contract, and all payments are in the same currency units as $M$. Then the payment is

P(x)= \begin{cases} 0, & x\le a,\\ M\displaystyle\frac{x-a}{b-a}, & a<x<b,\\ M, & x\ge b. \end{cases}

The ratio $(x-a)/(b-a)$ is the fraction of the measurement interval from attachment to exhaustion that has been traversed. It is not the fraction of the measurement $x$ itself. The lower and upper branches cap the result at zero and $M$, respectively.

Boundary rule: Under this schedule, a measurement exactly equal to $a$ produces zero, while a measurement exactly equal to $b$ produces $M$. A measurement above $b$ also produces $M$; the payment does not continue increasing beyond the stated maximum.

The formula follows from linear interpolation. Between the two specified points $(a,0)$ and $(b,M)$, a straight line has slope equal to change in payment divided by change in measurement:

\frac{M-0}{b-a}=\frac{M}{b-a}.

Starting at payment zero when $x=a$, moving from $a$ to $x$ therefore changes the payment by the slope multiplied by $x-a$:

P(x)=0+\frac{M}{b-a}(x-a).

This gives the middle branch. The contract’s zero floor and maximum cap give the other two branches. The interpolation is justified only if the schedule actually specifies a proportional increase between the two endpoints; it should not be assumed for a stepwise schedule.

Worked Example: Interpolating a Wind-Speed Payment

A hypothetical bond provides a payment based on the maximum three-second wind gust recorded at a named station during a defined storm window. The contract states that the payment is zero at or below 140 miles per hour, rises linearly to a maximum of $18 million at 170 miles per hour, and remains at that maximum above 170 miles per hour. The qualifying station records 158 miles per hour. Calculate the payment.

Method. This is a continuous schedule. The attachment is $a=140$, the exhaustion point is $b=170$, the maximum payment is $M=\$18$ million, and the qualifying measurement is $x=158$. All wind measurements use the contract’s stated unit and gust definition.

The measurement lies between the two thresholds because $140<158<170$, so use the interpolation branch. The interval from attachment to exhaustion is

b-a=170-140=30\text{ miles per hour}.

The measurement is $158-140=18$ miles per hour above attachment. Its fraction of the payout interval is

\frac{x-a}{b-a}=\frac{158-140}{170-140}=\frac{18}{30}=0.60.

The payment is therefore

P(158)=\$18\text{ million}\times 0.60=\$10.80\text{ million}.

Check. The measurement is 60% of the way from 140 to 170, so the payment is 60% of the maximum. The result is between zero and $18 million, as required. The common mistake is to calculate $158/170$ of the maximum, which would give about $16.73 million. That ratio measures the gust as a fraction of 170, not the distance traversed within the contractual payout interval.

Stepwise Schedules and Inclusive Boundaries

Not every contract pays proportionally. A stepwise schedule assigns a fixed payment to each measurement range. For instance, the contract might specify no payment below a first threshold, one fixed amount from that threshold up to a second, and a larger fixed amount from the second threshold upward. The written inequalities determine which payment applies at an exact boundary.

For a stepwise schedule, first determine the contractual qualifying measurement, then compare it with the ranges in order. Do not interpolate between fixed payment levels unless the contract says to do so. An equality sign matters: $x\ge c$ assigns the value $c$ to that range, while $x>c$ does not.

Worked Example: Applying an Earthquake Step Schedule

A hypothetical bond uses peak ground acceleration (PGA), measured in units of gravity, at the maximum of three specified stations during the contract’s event window. The payment is $0 if PGA is below 0.25; $4 million if PGA is at least 0.25 but below 0.40; $9 million if PGA is at least 0.40 but below 0.55; and $15 million if PGA is at least 0.55. The three qualifying station readings are 0.37, 0.40, and 0.33. Find the payment.

Step 1: Apply the aggregation rule. The contract specifies the maximum station reading, so

x=\max(0.37,0.40,0.33)=0.40.

Step 2: Apply the boundary definitions. The measured value is exactly 0.40. It does not fall in the $4 million band, which requires $x<0.40$. It does fall in the next band, which includes $x\ge0.40$ and requires $x<0.55$.

Step 3: Read the payment. The contractual payment is $9 million.

Check. The maximum reading is the middle station’s 0.40, and the inclusive lower boundary assigns that value to the $9 million band. Averaging the readings would give $(0.37+0.40+0.33)/3=0.3667$, but averaging is not the specified aggregation rule. That incorrect method would place the result in the $4 million band.

Inverse-Severity Measurements

Some physical triggers pay more when the measured quantity is lower. A drought-related schedule, for example, can increase as accumulated rainfall decreases. The same principles apply: use the contractual measurement, keep the stated units, and follow the specified endpoints. The direction of increasing severity changes the interpolation expression.

Suppose a schedule pays zero when rainfall $R$ is at or above $a$, and pays a maximum $M$ when rainfall is at or below $b$, where $b<a$. If the payment increases linearly as rainfall falls between these points, then

P(R)= \begin{cases} M, & R\le b,\\ M\displaystyle\frac{a-R}{a-b}, & b<R<a,\\ 0, & R\ge a. \end{cases}

Here $a-R$ measures how far rainfall has fallen from the no-payment threshold, and $a-b$ is the full rainfall interval over which the payment increases. The endpoint checks are $P(a)=0$ and $P(b)=M$. This formula applies only when the contract specifies a linear schedule across that interval.

Worked Example: Calculating a Rainfall-Based Payment

A hypothetical bond uses total rainfall at a named gauge over a specified 30-day period. It pays nothing if the total is at least 80 millimeters, increases linearly as rainfall falls, and pays a maximum of $10 million at 40 millimeters or less. The contract-defined 30-day total is 52 millimeters. Calculate the payment.

Method. This is an inverse-severity continuous schedule with $a=80$, $b=40$, $M=\$10$ million, and $R=52$ millimeters. Since $40<52<80$, use the linear branch.

The full rainfall interval is

a-b=80-40=40\text{ millimeters}.

Rainfall is $80-52=28$ millimeters below the no-payment threshold, so the fraction of the interval traversed is

\frac{a-R}{a-b}=\frac{80-52}{80-40}=\frac{28}{40}=0.70.

The payment is

P(52)=\$10\text{ million}\times0.70=\$7.00\text{ million}.

Check. The measurement lies between 40 and 80 millimeters, so the payment must be between zero and $10 million. At 52 millimeters, 70% of the interval from 80 down to 40 has been traversed, consistent with a 70% payment. Using $52/80$ as the payment fraction would ignore both the zero-payment threshold and the direction of the schedule.

What the Trigger Payment Does—and Does Not—Represent

The trigger calculation produces the contractual amount associated with the specified physical measurement. It is not, by itself, an estimate of the sponsor’s actual catastrophe loss or a statement that actual losses equal the payment. A storm may cause substantial losses even when the measured wind does not reach attachment; conversely, a severe qualifying measurement can produce a payment that exceeds the sponsor’s loss.

This difference is one source of basis risk: the parametric result and the exposure the sponsor seeks to protect need not move together exactly. For this calculation, however, the relevant question is narrower: what amount does the stated schedule assign to the valid measurement? Keep that answer separate from any later analysis of how a payment is applied within a bond’s capital structure.

A reliable calculation can be organized as a short sequence:

Establish the contractual measurement.
Confirm the physical quantity, units, source, locations, observation period, and aggregation rule.

Classify the schedule.
Determine whether the payment is stepwise or continuous, and whether greater or lower measurements increase payment.

Locate the measurement in the schedule.
Apply each inequality as written, paying particular attention to equality at a threshold.

Calculate and check.
For a continuous schedule, calculate the fraction of the threshold interval traversed. Verify that the result respects the stated minimum and maximum.

The most frequent errors are using the wrong observation or aggregation rule, treating a stepwise schedule as linear, and assigning an exact threshold to the wrong band. A final boundary check is especially useful: substitute each threshold into the stated rule and verify the payment implied by the contract.

Key takeaway: A parametric payment is calculated from the contract-defined physical measurement and payout schedule. Use the correct measurement rule, respect every inclusive or exclusive boundary, and interpolate only when the schedule specifies a continuous payout.

Check Your Understanding

For each hypothetical contract, apply the stated measurement rule and payout schedule. Treat amounts as payments in U.S. dollars.

A wind trigger pays zero at or below 145 miles per hour and increases linearly to $12 million at 175 miles per hour, remaining capped thereafter. The qualifying gust is 160 miles per hour. What is the payment? (Answer: $6.00 million)
An earthquake schedule pays $2 million for $0.18\le x<0.30$, $6 million for $0.30\le x<0.45$, and $10 million for $x\ge0.45$, with no payment below 0.18. The contract uses the maximum PGA across three named stations, which record 0.29, 0.30, and 0.27. What payment applies? (Answer: $6 million)
A rainfall trigger pays zero at or above 65 millimeters and a maximum of $8 million at or below 25 millimeters, with a linear schedule between those thresholds. The qualifying period total is 40 millimeters. Calculate the payment. (Answer: $5.00 million)
A wind schedule pays zero for $x\le100$ miles per hour, increases linearly to $6 million at 130 miles per hour, and is capped at $6 million above 130. The qualifying measurement is exactly 100 miles per hour. What is the payment, and which boundary rule controls? (Answer: $0; the attachment threshold is inclusive in the zero-payment range)
A hypothetical bond uses the maximum of two station readings and pays $0 below 0.20 PGA, $3 million for $0.20\le x<0.35$, and $7 million for $x\ge0.35$. The readings are 0.34 and 0.35. Identify the qualifying measurement and payment, and explain why averaging is not appropriate. (Answer: The qualifying measurement is 0.35 and the payment is $7 million; the contract specifies the maximum, not the average)

What You'll Learn