Case Study 2: The Parametric Promise — and the Basis Risk That Comes With It

This case combines a real, public market reality — the genuine and growing use of parametric structures for catastrophe and climate risk — with a clearly-labeled constructed teaching example to make the central trade-off concrete. The real phenomenon is described qualitatively with no fabricated statistics; the worked numbers are illustrative, exactly as the book's other teaching examples are. It is paired with Case Study 1 as the complementary angle: where CS1 showed traditional indemnity coverage retreating under climate stress, this one examines the product many hope will help fill the gap — and the limit built into it.

Background

As traditional indemnity coverage has come under climate stress (Case Study 1), attention has turned to a different product shape as part of the answer: parametric insurance (owned by Chapter 26), which pays a fixed amount when a measured trigger is met rather than paying a proven, adjusted loss. Parametric structures are not new — they have a real, public history in catastrophe risk transfer, including catastrophe bonds and parametric covers used by governments, public entities, and large corporations to get fast liquidity after hurricanes, earthquakes, and other natural catastrophes. What is new is the push to extend the idea: to smaller commercial insureds, to climate perils like flood and extreme heat, and to products that pair a parametric trigger with the real-time sensors of continuous underwriting (§36.1).

The appeal is genuine and important. A business hit by a hurricane does not need a six-month claims adjustment; it needs cash this week to keep the lights on, make payroll, and reopen before its customers go elsewhere. A parametric cover delivers exactly that: because there is no loss to prove and adjust, the money arrives in days, and both sides knew the terms in advance. For the climate-stressed coastal economy of Case Study 1, that speed is not a luxury; it is often the difference between a business surviving the storm and not. Parametric is, rightly, one of the most discussed tools on the new product frontier.

The insurance / underwriting issue

The issue is the one the chapter names plainly: parametric trades indemnity's slow precision for speed and certainty, and the price of that trade is basis risk (§36.5) — the gap between the parametric payout and the policyholder's actual loss. To make it concrete, here is a constructed example.

📄 Read the Submission

text FIGURE 36.2 — "When the trigger and the loss disagree" [constructed teaching example] THE SUBMISSION A mid-size coastal business buys a parametric hurricane cover: it pays a fixed $500,000 if a Category 3-or-stronger storm passes within 25 miles of the insured location. THE CONTEXT The business wants fast post-storm liquidity to bridge the gap above its traditional property policy's 5% named-windstorm deductible (Chapter 12) — money in days, not months. SCENARIO A A Category 3 passes 26 miles away and causes $900,000 of damage. The trigger does NOT fire. The business collects nothing from the parametric cover, despite a severe loss. SCENARIO B A Category 3 passes 20 miles away but the business, by luck of local terrain, suffers only $50,000 of damage. The trigger fires; the business collects the full $500,000 — a windfall far above its actual loss. WHAT IT SHOWS The payout is tied to the TRIGGER, not the LOSS. Sometimes that cuts brutally against the insured (A); sometimes generously in their favor (B). That two-sided gap is basis risk, and it is the structural cost of parametric's speed. WHAT IT DOESN'T It does not mean parametric is "bad." It means parametric is a SUPPLEMENT to indemnity coverage — bought with eyes open to basis risk — not usually a replacement for it. THE LESSON Design the trigger to track the real loss as closely as the available data allows, size the parametric layer to the liquidity need, and never let a client believe a parametric cover makes them "fully covered." Basis risk is the fine print that is the whole point.

The underwriting judgment in a parametric product, as §36.5 argues, lives upstream in the design. There is no claim to adjust account by account; the entire risk-selection and fairness decision is baked into the choice of trigger, threshold, and geography, made once and then executed many times. Choosing the trigger is the underwriting. A trigger too loose (fires easily) is generous to insureds but expensive and prone to windfalls; a trigger too tight (fires rarely) is cheap but leaves insureds collecting nothing after real losses — the worst possible outcome for the product's reputation and for the people who bought it believing they were protected.

🤖 Model vs. Judgment Parametric is a vivid illustration of the chapter's claim that the new products relocate judgment rather than removing it. In an indemnity policy, an adjuster (and an underwriter at renewal) exercises judgment after the event, matching payment to proven loss. In a parametric policy, all of that judgment is moved before the event, into the trigger design — and then there is no human discretion left at claim time at all, by design (that is what makes it fast). The model and the data choose the trigger; the human chooses which trigger to accept and how much basis risk to live with. The skill is the same underwriting skill — read the risk, structure the deal, accept a calculated exposure, and be honest about what is and is not covered — exercised at the altitude of the product instead of the policy.

Outcome

Parametric insurance is, in the real market, a growing and genuinely useful part of catastrophe risk transfer — not a fad and not a panacea. Where it has worked best, two conditions have held: the trigger tracks the real loss reasonably well (good data, well-chosen parameters), and the buyer understood from the start that they were buying fast liquidity with basis risk, not full indemnity. Where parametric has disappointed, it has usually been because of basis risk that the buyer did not fully grasp — a trigger that failed to fire after a damaging event, leaving a policyholder who believed they were protected with nothing. The product's future on the climate frontier (§36.5) depends heavily on improving the data and the trigger design so the basis risk shrinks, and on selling it honestly as the supplement it is.

Lesson

The lesson generalizes well beyond parametric, and it is a fitting near-final lesson for the book. A new product is a new place to put underwriting judgment, not a way to escape it — and honesty about what a product does not cover is part of the underwriting. Parametric's speed is real and valuable; its basis risk is real and unavoidable; and the professional's job is to design the trigger to minimize the gap, size the cover to the need, and make absolutely sure the insured understands the trade. The same discipline the book has taught since Chapter 1 — never sell coverage you cannot stand behind, and never let the insured misunderstand what they bought (utmost good faith runs both ways, Chapter 4) — applies with full force to the cleverest products on the 2035 frontier. The tools change; the craft, and the honesty it demands, do not.

Discussion questions

  1. In Scenario A of Figure 36.2 the insured suffers \$900,000 in damage and collects nothing. Did the insurer do anything wrong? Distinguish a product working as designed from a product mis-sold, and say what would make it the latter. (§36.5)
  2. Explain why the chapter says choosing the trigger "is the underwriting." Where has the risk-selection judgment gone, compared with a traditional indemnity policy? (§36.5; Ch. 20)
  3. A client says, "I have a parametric hurricane cover, so I'm fully protected against storms." Write the two-sentence correction an honest underwriter or broker (Chapter 39) must give, naming basis risk.
  4. For the Harbor Steel forward look (The Underwriting File, this chapter), a parametric wind supplement sits on top of the 5% named-windstorm deductible (Chapter 12). Explain why a parametric layer is a sensible complement to — not a replacement for — the traditional property policy on that account.
  5. Parametric pairs naturally with the IoT sensors of §36.1, which can verify a trigger automatically. What does that pairing improve, and what classic problem (Chapter 1) would you worry about if the insured controlled the sensor that verifies their own trigger?