Coverage Modifications: Inflation, Deductible Changes
Understand how inflation and deductible changes affect loss distributions and premiums for Exam STAM.
Impact of Inflation on Losses
Inflation uniformly scales all losses by a factor (1+r), where r is the annual inflation rate. If X represents losses before inflation, then inflated losses are Y = X*(1+r). This scaling affects every aspect of the loss distribution: the mean, variance, and all higher moments change. For the mean, E[Y] = (1+r)*E[X]. For the variance, Var(Y) = (1+r)^2 * Var(X). The entire distribution shifts and stretches to the right.
For parametric distributions, inflation transforms the parameters in predictable ways. If losses follow an exponential distribution with mean theta, inflated losses follow an exponential with mean theta*(1+r). For a Pareto distribution with parameters alpha and theta, inflated losses have the same shape parameter alpha but the scale parameter becomes theta*(1+r). For a lognormal distribution with parameters mu and sigma, inflated losses have parameters mu + ln(1+r) and sigma (the log-scale shift preserves the shape while increasing the location). Knowing how inflation maps through each distribution family is essential for Exam STAM calculations.
Inflation and Deductibles
When a policy has a deductible d, inflation creates a dual effect. First, losses that were previously below the deductible may now exceed it after inflation, increasing claim frequency. Second, the amount paid on claims that already exceeded the deductible increases because the inflated loss minus the fixed deductible is larger. The insurer faces higher costs from both more claims and larger claim payments.
To quantify this, consider the expected cost per loss with deductible d. Before inflation, the expected cost per loss is E[(X-d)+] = E[X] - E[X wedge d]. After inflation at rate r, the ground-up loss becomes X*(1+r), and the expected cost per loss is E[(X*(1+r) - d)+] = (1+r)*E[X] - E[X*(1+r) wedge d]. Notice that the deductible d remains fixed in nominal terms, but in real terms it has effectively decreased to d/(1+r) relative to the inflated loss distribution.
For example, suppose losses follow a Pareto distribution with alpha = 3 and theta = 1000. The expected cost per loss with a deductible of 500 is E[X] - E[X wedge 500]. After 10% inflation, theta becomes 1100 and the expected cost per loss with the same 500 deductible changes. Both the limited expected value at 500 and the unlimited mean change, and the net effect is an increase in cost that exceeds 10% because the fixed deductible absorbs a smaller proportion of the inflated losses.
The Loss Elimination Ratio
The loss elimination ratio (LER) measures the proportion of expected losses eliminated by a deductible: LER(d) = E[X wedge d] / E[X]. This ratio tells you what fraction of total expected losses the policyholder absorbs through the deductible. A higher LER means the deductible is more effective at reducing insurer costs.
Inflation reduces the LER because the deductible stays fixed while losses grow. If the original LER is E[X wedge d] / E[X], after inflation at rate r, the new LER is E[X*(1+r) wedge d] / E[X*(1+r)]. The numerator increases more slowly than the denominator because the limited expected value is capped while the mean grows proportionally. Over time, persistent inflation erodes the effectiveness of a fixed deductible, which is why insurers periodically increase deductible levels.
Deductible Changes and Premium Impact
Changing the deductible from d1 to d2 affects the expected cost per loss and per payment. The expected cost per loss with deductible d equals E[X] - E[X wedge d]. When comparing two deductibles, the ratio of expected costs per loss is (E[X] - E[X wedge d2]) / (E[X] - E[X wedge d1]). This ratio, sometimes called the loss elimination ratio adjustment factor, is used to adjust premiums when deductible levels change.
The expected cost per payment differs from the expected cost per loss. The expected cost per payment conditions on a claim exceeding the deductible: E[X - d | X > d] = (E[X] - E[X wedge d]) / P(X > d) = (E[X] - E[X wedge d]) / S(d). This quantity is always larger than the expected cost per loss because it excludes the zero-payment cases.
For a concrete example, suppose losses follow an exponential distribution with mean theta = 2000 and you want to compare deductibles of 500 and 1000. The limited expected value for an exponential is E[X wedge d] = theta*(1 - exp(-d/theta)). So E[X wedge 500] = 2000*(1 - exp(-0.25)) = 2000*0.2212 = 442.4, and E[X wedge 1000] = 2000*(1 - exp(-0.5)) = 2000*0.3935 = 787.0. The expected costs per loss are 2000 - 442.4 = 1557.6 and 2000 - 787.0 = 1213.0, respectively. Increasing the deductible from 500 to 1000 reduces the expected cost per loss by about 22%.
Policy Limits and Inflation
Policy limits cap the insurer's payment at a maximum amount u. With both a deductible d and limit u, the insurer pays min(max(X - d, 0), u) per loss. The expected insured cost per loss is E[X wedge (d+u)] - E[X wedge d]. Inflation pushes more of the loss distribution against the policy limit, compressing payments at u and reducing the insurer's cost relative to the unlimited case.
When inflation raises losses but the limit remains fixed, the expected insured cost grows more slowly than the ground-up losses because more claims hit the ceiling. This means the insurer's share of losses decreases over time, which benefits the insurer but leaves the policyholder exposed to losses above the limit. This phenomenon is called "leveraging" of limits by inflation.
Combined Inflation and Coverage Changes
Exam STAM problems frequently combine inflation with coverage modifications. A typical problem might ask: given a loss distribution, compute the expected insured cost under the current deductible and limit, then recompute after applying inflation and changing the deductible. The solution requires computing limited expected values at multiple points on the inflated distribution.
The general approach is: (1) apply inflation to the distribution parameters, (2) compute E[X wedge d] and E[X wedge (d+u)] for the inflated distribution, (3) find the expected insured cost as the difference, and (4) compare to the original cost. Being systematic about which distribution parameters change and which coverage parameters remain fixed is the key to avoiding errors.
Study Tips
For Exam STAM, memorize how inflation transforms parameters for each major distribution family (exponential, Pareto, lognormal, gamma). Practice computing limited expected values before and after inflation, and be comfortable working with both the loss elimination ratio and the expected cost per payment. Pay attention to whether a problem asks for the cost per loss or cost per payment, as confusing these is a common source of errors. Work through problems that layer inflation on top of deductible and limit changes to build fluency with the full calculation pipeline.