Inferring Coverage Terms From Payment Graphs

Reading a Payment Graph

A payment graph shows how the insurer’s payment changes as the loss amount changes. Its horizontal coordinate is the loss $X$, and its vertical coordinate is the insurer’s payment per loss, $Y^L$. Both axes must be read in the units shown: for example, a coordinate of 250 may mean $250 or $250,000 if the axes are stated in thousands.

The graph’s shape gives clues about coverage terms. A flat segment at zero indicates that the insurer makes no payment over that range. A rising straight segment indicates that payment increases with the loss; its slope measures how much the payment increases for each additional dollar of loss. A flat segment above zero indicates that payment has reached a cap. Where the slope changes, a coverage rule or component may have changed.

Definition: For a graph of insurer payment $Y^L$ against loss $X$, the starting point of positive payment is the loss amount where $Y^L$ first becomes positive. A segment slope is the change in payment divided by the change in loss over that segment. A plateau is a horizontal segment, indicating a constant payment as loss increases.

These clues are most useful when the relevant contract structure is specified or reasonably constrained. For instance, in a simple deductible-and-coinsurance design with a payment cap, the graph can reveal the deductible, the insurer’s payment percentage, and the maximum payment. But a graph describes payment outcomes, not necessarily the exact wording that produced them. Different contracts can generate the same curve.

What the Shape Reveals

Start with the axes and units, then inspect the graph from left to right. A zero-payment interval followed by a rising segment is consistent with a deductible: the insurer does not pay until the loss exceeds a threshold. The first rising segment’s slope indicates the payment percentage when the contract applies a constant share of covered loss. A later plateau identifies a maximum payment over the displayed range.

For a simple contract that pays a percentage $c$ of the loss above a deductible $d$, subject to a maximum payment $M$, the payment function is:

Y^L=\min\!\left(c(X-d)_+,\,M\right), \qquad (z)_+=\max(z,0).

The zero-payment region ends at $X=d$. Before the cap is reached, a rise of $\Delta X$ produces a payment increase of $c\Delta X$. Thus the slope is $c$. The plateau begins when $c(X-d)=M$, or $X=d+M/c$. This threshold is the loss at which the payment cap is reached; it is not necessarily the amount of a policy limit as worded in a contract.

Interpretation guide: Under a simple deductible, percentage, and payment-cap model, the zero-to-rising breakpoint identifies $d$, the rising-segment slope identifies $c$, and the plateau height identifies $M$. A straight rising segment that starts at a nonzero loss $a$ and ends at $b$ can also describe a coverage layer: $a$ is its attachment point, and $b$ is its exhaustion point.

A graph can contain several rising segments. If the slope changes at a breakpoint, read the payment immediately before and after that point. The change may indicate a switch between coverage components, a change in payment percentage, or a layer becoming active or exhausted. The payment value does not usually reset at a breakpoint: for a continuous payment function, the curve joins at the same vertical coordinate.

Why Slope Identifies a Payment Percentage

The slope interpretation follows directly from the payment rule. Suppose that, over an interval, each additional dollar of loss in excess of a deductible is covered at percentage $c$, with no cap reached in that interval. For two losses $X_1$ and $X_2$ in that interval, where $X_2>X_1>d$, payments are $c(X_1-d)$ and $c(X_2-d)$. Their difference is $c(X_2-X_1)$. Dividing the payment change by the loss change gives $c$.

\frac{Y^L(X_2)-Y^L(X_1)}{X_2-X_1} = \frac{c(X_2-d)-c(X_1-d)}{X_2-X_1} =c.

This calculation also shows why the slope does not identify the deductible by itself: $d$ cancels from the difference. The deductible comes from the loss coordinate where positive payment begins. Both pieces of information are needed to interpret the graph.

Worked Examples

Worked Example: Inferring a Deductible, Percentage, and Payment Cap

A hypothetical policy has a payment graph with loss $X$ in dollars and insurer payment $Y^L$ in dollars. The graph shows no payment through $X=120$, a straight rise from $(120,0)$ to $(720,480)$, and a horizontal plateau at $Y^L=480$ for losses above $720. Under the simple deductible, percentage, and payment-cap model, infer the terms and calculate the payment for a loss of $470.

Method. Read the first positive-payment loss for the deductible, calculate the rising-segment slope for the payment percentage, and read the plateau height for the payment cap.

Deductible. The graph remains at zero through a loss of $120, and the positive-payment segment begins at that loss. Thus $d=\$120$.

Payment percentage. The slope between the two stated coordinates is $480/(720-120)=480/600=0.8$. The insurer pays 80% of the loss above the deductible before the cap applies.

Payment cap. The plateau is at $480, so $M=\$480$. The graph is consistent with reaching the cap at a loss of $120+480/0.8=120+600=\$720$, matching the stated breakpoint.

Payment at $X=\$470$. The loss above the deductible is $470-120=\$350$. Applying the 80% payment percentage gives $0.8(350)=\$280$. This is below the $480 cap, so the payment is $280.

Check. The point $(470,280)$ lies on the rising segment: $280/0.8+120=350+120=470$. A common mistake is to apply 80% to the full $470, obtaining $376; the deductible must be removed first.

Worked Example: Finding the Terms of a Coverage Layer

A hypothetical layer’s payment graph is zero up to a loss of $300, rises linearly to a payment of $300 at a loss of $900, and remains at $300 for larger losses. On the rising segment, the payment at a loss of $500 is $100. Infer the attachment point, exhaustion point, payment percentage, and maximum payment. Then calculate the payment at a loss of $760.

Method. For a single layer, the attachment point is where payment begins, the exhaustion point is where the layer’s payment stops increasing, and the slope gives the payment percentage within the layer.

Attachment point. Payment first becomes positive at $X=\$300$, so the attachment point is $300.

Payment percentage. Using the point at $X=\$500$, the loss above attachment is $500-300=\$200$. The payment is $100, so the slope is $100/200=0.5$, or 50%.

Exhaustion point and maximum payment. The graph stops rising at $X=\$900$, where payment reaches $300. The layer therefore exhausts at a loss of $900 and has a maximum payment of $300. As a consistency check, the loss covered within the layer is $900-300=\$600$, and $0.5(600)=\$300$.

Payment at $X=\$760$. The loss within the layer is $760-300=\$460$, giving a payment of $0.5(460)=\$230$. Since $230 is below the $300 maximum, the payment is $230.

Check. The point $(760,230)$ falls on the rising segment, and the remaining payment capacity is $300-230=\$70$. The common mistake is to treat the $900 exhaustion point as the layer’s maximum payment; it is a loss coordinate, whereas the maximum payment is the vertical coordinate, $300.

Worked Example: Reading Successive Primary and Excess Segments

A hypothetical payment graph is zero through a loss of $100. It then rises with slope 1 to $(500,400)$. From $X=500$ to $X=900$, it rises with slope 0.5, reaching $(900,600)$; after $900, it is flat at $600. Under a model with a primary component followed by an excess component, describe the terms indicated by the graph and calculate the payment for a loss of $700.

Method. Read each segment separately, keeping the payment at the end of one segment as the starting payment for the next. A slope change does not erase payment already accumulated.

Initial zero segment. Positive payment begins at $X=\$100$, indicating a deductible of $100 under this model.

Primary segment. The slope from $100 to $500 is $(400-0)/(500-100)=400/400=1$. The primary component pays 100% of loss above $100 over this interval. It has accumulated $400 by $X=\$500$.

Excess segment. Its attachment point is $500, where the primary segment stops rising. The slope from $500 to $900 is $(600-400)/(900-500)=200/400=0.5$, so this segment adds payment at 50% of the loss. It adds at most $200, because the total payment rises from $400 to $600. The excess segment exhausts at $X=\$900$.

Payment at $X=\$700$. The excess segment has encountered $700-500=\$200$ of additional loss, so it adds $0.5(200)=\$100$. Add the $400 already paid by the primary component: $400+100=\$500$.

Check. The graph’s second segment is halfway from $500 to $900 at $X=\$700$, so its added payment is half of $200, or $100; the total is $500. A common mistake is to report only the excess-segment addition of $100 instead of the total insurer payment.

Limits of Graph-Based Inference

A graph supports conclusions about payments under the stated axes and model, but it does not always identify a unique policy description. For example, a maximum payment of $480 is directly visible as a plateau height. Whether contract wording calls that amount a payment limit, a layer limit after coinsurance, or something else depends on how the policy defines its covered amount. Similarly, successive segments may be described as separate coverage components or as one rule with a changing payment percentage.

When interpreting a graph, report what the graph establishes before attaching a contract label. For instance, “payment begins at a loss of $120, rises at 80% per dollar, and plateaus at $480” is a direct description. “The policy has a $480 limit on covered loss” may not follow unless the relationship between covered loss and payment has been specified.

Always check continuity at breakpoints. If one segment ends at $(x,y)$, the next segment for the same payment function should start at that same point unless the problem explicitly describes a discontinuity. Also check that an inferred cap is reached at the stated loss: under the simple model, the loss at the start of the rising segment plus the required loss increase to reach the cap must match the plateau’s starting loss.

Key takeaway: Read horizontal coordinates as loss amounts and vertical coordinates as payments. Use the first positive-payment point for the starting threshold, the rise per dollar for a payment percentage, and the plateau height for a payment cap. Treat exact policy wording as an inference only when the model supports it.

Check Your Understanding

For each graph description, infer only the terms supported by the stated simple model, and check that the breakpoints and payment amounts agree.

A graph is zero through a loss of $80, rises with slope 0.75, and plateaus at $300. At what loss does the plateau begin, and what are the deductible, payment percentage, and maximum payment? (Answer: deductible $80; payment percentage 75%; maximum payment $300; plateau begins at loss $480.)
A layer begins paying at loss $250, has slope 0.4, and plateaus at payment $180. At what loss does it exhaust, and what is its maximum payment? (Answer: exhaustion at loss $700; maximum payment $180; the layer covers $450 of loss between attachment and exhaustion.)
A graph is zero through $50, rises with slope 1 to payment $300 at loss $350, then rises with slope 0.5 until it plateaus at loss $750. What is the maximum total payment? (Answer: $500; the second segment adds $0.5(750-350)=\$200$ to the first segment’s $300.)
A simple deductible-and-percentage graph begins rising at $150, has slope 0.6, and plateaus at payment $300. At what loss does the cap apply? (Answer: loss $650, since $150+300/0.6=650$.)
A graph is zero through loss $100, rises with slope 0.8 to loss $500, then rises with slope 0.4 until loss $1,000 and becomes flat. What is the plateau payment, and how much does the second segment add? (Answer: plateau payment $520; the second segment adds $200 to the $320 accumulated in the first segment.)

What You'll Learn