Umbrella vs. Excess Liability: What Is the Difference?

Umbrella and excess liability policies both add limits above your primary liability and auto policies. The difference is what happens when no underlying policy covers a claim. An excess policy pays only when an underlying policy does. A commercial umbrella can be broader and "drop down" to pay a claim the underlying policies exclude, after you pay a self-insured retention (SIR). Every umbrella is a type of excess policy, but not every excess policy is an umbrella.

Buyers, and even some policies, use the two labels interchangeably, so the safe move is to read what the form actually does rather than what the cover page calls it.

What do umbrella and excess liability policies do?

Both policies exist because primary limits run out and jury awards can be very large. According to Marathon Strategies, U.S. juries returned 135 verdicts of $10 million or more against corporate defendants in 2024, totaling $31.3 billion.^[1] A standard commercial general liability policy or business auto policy often has a $1,000,000 per occurrence limit, and a serious injury verdict can pass that quickly.

An umbrella or excess policy adds a second layer, commonly $1,000,000 to $10,000,000 or more, that pays the part of a covered loss above the underlying limit. The excess layer also matters for defense. Once a primary policy's limit is used up, its duty to defend usually ends, and the excess insurer can take over the defense. Whether that defense reduces your limit depends on the form: the ISO umbrella pays defense costs in addition to its limits, while many non-ISO excess forms count defense as part of the loss, so a long lawsuit uses up the limit you bought.

Both policy types have a second job. Your aggregate limit caps what the underlying policy pays for all claims in a policy year. When claims reduce or use up that aggregate, the excess policy drops down and responds in the underlying policy's place.

What is the difference between an umbrella and an excess liability policy?

The difference is what happens when no underlying policy responds to a loss. A plain excess liability policy, whether follow form or written on its own terms, only adds limits. Even a stand-alone excess policy with broader wording than the underlying policies cannot drop down to provide primary coverage when no underlying policy responds.

An umbrella, such as one written on the ISO Commercial Liability Umbrella Coverage Form CU 00 01 04 13, has its own insuring agreements, which can be broader than the underlying ones in some areas. If a loss falls under the umbrella but under none of the underlying policies, the umbrella drops down and pays as the first layer. You pay the self-insured retention (SIR) shown on the umbrella's declarations first. That third ability is the whole distinction. The rest of the mechanics compare like this:

Some carriers sell both in one contract, a bifurcated policy with a follow-form part and an umbrella part, each with its own exclusions. On those, the answer can differ from one line of coverage to the next within the same policy.

What does follow form mean?

A follow-form excess policy adopts the terms, conditions and exclusions of the underlying policies on its schedule. If the underlying policy covers a claim, so does the follow-form excess, at a higher layer, and if the underlying policy excludes it, so does the excess. The form is short because it does not write its own terms, and predictable because the excess layer is never broader than what sits below it.

The catch is that true follow form is rare. Most policies described as "follow form" are conditional follow form: they follow the underlying wording except where the excess policy's own provisions differ. Where they conflict, the excess can pay less than the underlying policy would have, so a claim your CGL covers can still be denied by the layer above it.

For example, a pollution or drone exclusion added to the excess layer by endorsement applies whether or not the underlying policy would have paid. Read every exclusion in the excess policy and sort them into absolute exclusions, which always apply, and conditional ones, which give way when the underlying policy provides coverage.

How does the self-insured retention work?

The self-insured retention (SIR) is a dollar amount on a commercial umbrella's declarations, and it applies to only one kind of claim: one where the umbrella is broader than the underlying policies and drops down to pay as the primary insurer. You pay the SIR before the umbrella responds. An SIR is not a deductible. A deductible is taken from each loss the insurer pays from the first dollar, while the SIR is paid by you before the policy responds, and it never applies to ordinary excess claims.

If a $1,500,000 loss is covered by your $1,000,000 underlying policy and the umbrella pays the $500,000 above it, no SIR applies. No SIR applies either when the umbrella replaces an exhausted underlying aggregate, because under the CU 00 01 definition the retention does not apply to occurrences the underlying insurance would have covered if its limits had not been used up.

Plain excess policies have no SIR, because they never drop down for broader coverage. The SIR differs by policy, often $10,000 and sometimes $0, and some insureds take a higher retention to lower the premium. The retention applies per occurrence or per policy period, and the umbrella's limits sit above the retention rather than being reduced by it.

How do these policies stack with your aggregate limits?

The excess layer works together with your underlying aggregates, and that is where buyers lose money. Excess insurers usually require you to keep the underlying policies on the schedule, such as commercial general liability, business auto liability and employers liability, at specified limits. Most also require those aggregates to be unimpaired, meaning fully available, on the excess policy's effective date.

Only paid claims impair an aggregate. A large reserve on an open claim does not, so a policy with a $500,000 reserve under a $2,000,000 aggregate still counts as unimpaired until the money is actually paid.

Timing is another trap. When an excess policy's dates don't match the underlying policy's dates, the layers are "non-concurrent." Claims paid before the excess policy started can use up the aggregate the excess insurer assumed was full. Umbrellas are especially exposed on timing, because they drop down only for injury that happens during their own policy period.

The math can be harsh. A contractor's CGL runs January to January with a $1,000,000 occurrence limit and a $2,000,000 aggregate. In June the contractor adds a $3,000,000 umbrella with the same expiration date. Two accidents from before June settle for a total of $1,100,000, leaving $900,000 in the aggregate when a $2,000,000 crane collapse happens. The CGL pays its remaining $900,000. The umbrella responds as if the aggregate were unimpaired and pays only above the full $1,000,000 occurrence limit. The contractor pays the $100,000 difference.

The clean fixes are to write the layers on the same dates, cancel and rewrite so the dates line up, or have the insurer add non-concurrency wording. Ask your broker which one your placement uses.

When does each policy fit?

Choose follow-form or stand-alone excess when your underlying policies already cover your real exposures and the only problem is the size of the limit. That is common when contracts require $5,000,000 or $10,000,000 in total liability limits. It is the simpler and usually cheaper option.

How much limit to buy keeps rising, and the rise can be measured. According to an analysis by Triple-I and the Casualty Actuarial Society, legal system abuse increased U.S. liability insurance losses by $231.6 billion to $281.2 billion over the ten years ending in 2024, driven by claim severity, not claim frequency.^[2] If one number should size your layer, it is $51,000,000, the median nuclear verdict in 2024 in Marathon's data, up from $21,000,000 in 2020.^[1] A limit that covered your worst case five years ago may not cover a median verdict today.

Choose an umbrella when the extra layer also needs to fill gaps. Because its insuring agreement is broader than a primary policy's, it can pick up claims the underlying policies miss. In exchange you take on the SIR and a longer form to read. Contractors and businesses with contractual liability beyond what a standard CGL covers use umbrellas most, but anyone whose operations outgrow off-the-shelf primary forms can buy one.

Whichever you buy, match the policy period to your underlying terms and keep the required underlying limits in force. Check the schedule of underlying insurance at every renewal, because a lapsed underlying policy can leave the excess layer with nothing beneath it.

Keeping the underlying policies in place is a policy duty, not just good practice. CU 00 01 04 13 requires the insured to notify the umbrella insurer in writing, as soon as practicable, when an underlying policy is canceled, not renewed, replaced, or changed in limits or scope. In a brokered placement, your broker sends that notice. Broadening underlying coverage mid-term without telling the umbrella insurer also works against you: the umbrella pays no more than it would have before the change, so the new exposure has no excess layer above it.

This guide is for educational purposes and summarizes standard ISO policy language. Your policy's specific terms, conditions, and endorsements control. Talk to a licensed broker about your actual exposures.

What to check at your next umbrella or excess renewal

Read the form itself, not the cover page, to see which kind of policy you have. Both types add limits above your primary coverage, but only a commercial umbrella can be broader and drop down to pay claims the underlying policies never covered, after its self-insured retention. At renewal, confirm that the dates and required underlying limits line up across every layer, and compare your schedule of underlying insurance with each exclusion in the excess layer.