Indemnity Wars

Indemnity Wars

August 7, 2013

By Dean B. Thomson

The battle in Minnesota over whether and how one party should indemnify another on a construction project has constantly been fought. After 30 years under one statutory scheme of allowable indemnity agreements and agreements to insure that indemnity, the Minnesota legislature has again stepped into the fray and re-aligned the rules of engagement. This month’s Briefing Paper examines the new amendments to the indemnity statute and provides insights on how to work within the new law.

Context and History

In their broadest form, construction indemnity agreements are essentially a promise in which one party (the promisor) agrees to indemnify or hold harmless another party (the promisee) for acts or omissions relating to a construction project. Sometimes the agreements require the promisor to indemnify the promisee even for the latter’s own negligence; this very broad indemnity is not based on fault and, not surprisingly, is called a “broad form” indemnity. By contrast, narrow or limited indemnity agreements are those in which the promisor agrees to indemnify the promisee, but only to the extent caused by the promisor’s negligence.

In 1983, the Minnesota legislature established the framework with which we all should be generally familiar: an indemnification agreement in a construction contract is unenforceable except to the extent the injury or damage in question is attributable to the negligence or breach of contract of the promisor. But, under the 1983 statute, the promisor can still agree to procure insurance for the benefit of others, including the promisee. In effect, the statute made an exception for the common practice whereby the promisee would require the promisor to agree in the construction contract to provide an insurance policy that would insure the parties’ contractual indemnity agreement. By this mechanism, the parties could agree to a broad form indemnity that by itself would be unenforceable, but if the promisor agreed to provide insurance that covered the obligations of the broad form indemnity agreement, then the agreement to provide insurance was still valid and enforceable.

Nevertheless, this was, and still is, a convoluted statutory scheme. The general justification offered for it, however, was

two-fold: first, proponents argue public policy doesn’t favor broad form indemnities because it’s considered bad policy for one party to absorb and relieve some other party for that other party’s negligence. Second, the statute nevertheless recognized that there are insurance products expressly designed to allow one party to provide insurance that covers the negligence of another party. The statute attempted to recognize that if the parties were willing to allocate payment of the premium for insurance that covered both their negligence, then the statute shouldn’t prohibit that contractual risk management technique.

The validity of this risk management technique was upheld by the Minnesota Supreme Court in Holmes v. Watson Forsberg Co., 488 N.W.2d 473, 475 (1992):

In our view, the legislature both anticipated and approved a long-standing practice in the construction industry by which the parties to a subcontract could agree that one party would purchase insurance that would protect “others” involved in the performance of the construction project. Such a risk allocation method is a practical response to problems inherent in the performance of a subcontract and…the parties are free to place the risk of loss upon an insurer by requiring one of the parties to insure against that risk.

A common example used to explain the benefits of this risk management technique is Builders Risk insurance. Owners, general contractors, and subcontractors can all subject the project to the risk of fire. If a fire occurs, the parties could all sue each other for their alleged respective fault for the blaze. After years of litigation, one or more of them will be held liable. In the meantime the building will remain unconstructed until the litigation proceeds are recovered to rebuild it. Fortunately, there is an insurance solution to avoid this morass – Builders Risk insurance can be purchased by one of the parties for the peril of fire, and all the interests of parties on the project will be covered by it. By agreement, all the covered parties will waive their claims against each other to the extent caused by that insurance (called a waiver of subrogation), and the risk of fire caused by the negligence of one or more of the parties will effectively be transferred to insurance. Of course, the insurance companies charge a substantial premium to accept this risk, but by intelligent underwriting the insurers will make a profit, and the owners, contractors and subcontractors who are the beneficiaries of the policy will receive a relatively quick recovery and avoid the cost and delay caused by litigating over fault and responsibility.

Problems with the Model

Insurance risk transfer sounds good in theory, and it can work efficiently and fairly in practice – in certain circumstances. Builders Risk insurance is an example of when theory and practice mesh well for one primary reason: the premium covering the insurance risk can be fully priced, is known up front, and can be transferred by contract with no later ramifications. In other words, if the General Contractor agrees to buy Builders Risk insurance for $10,000, it can charge the Owner $10,000 for that insurance and protect the project from fire caused by either Acts of God or the parties’ negligence. If a fire starts, the loss is adjusted and paid, and no additional charges are assessed to or absorbed by the parties.

But agreements to transfer risk by insurance do not always work this way. Sometimes, the cost consequence of transferring risk by insurance is not always known up front when the agreement to procure insurance is made. As a result, one party can promise to procure insurance for the negligence of another at a relatively low price, but when the loss strikes maybe years later, some fear that the insurer will raiseinsurance rates of the party providing the insurance as a means of recovering that loss, thereby making it more expensive for that party to do business.

In short, the full costs of transferring risk to certain types of insurance are not always fully known at the time of contracting and can’t be transferred for a complete price as part of the original bargain. In that case, the promisor is being asked to absorb costs in the form of potentially higher premiums, which cannot perfectly be transferred to insurance. As a result, the public policy favoring the transfer to insurance of the complete risk associated with indemnifying another party for the consequences of that other party’s negligence is not being fully respected.

An example is Additional Insured coverage. A subcontractor can promise to obtain Additional Insured coverage for a general contractor that covers the general contractor for its own negligence, and the price for that Additional Insured endorsement is quite low. If a claim is made against the general contractor years later, that contractor will look to the Additional Insured coverage provided by the subcontractor’s insurer for indemnity. The subcontractor’s insurance policy will respond, but the initial premium may not have been adequate to cover the cost of the future claim, and the subcontractor’s insurer may increase the cost of the subcontractor’s current premium to cover the extraordinary loss. Naturally, downstream promisors don’t like that result and felt that this aspect of the current law was unfair as the promisor, by being required to purchase Additional Insured insurance for the benefit of the promisee, was not being fully compensated for the risk it was taking.

Legislative Intervention

The legislature in 2013 tried to correct some of the practical problems in the current legislative scheme of allowing parties to insure the parties’ indemnity agreement. The new amendments can be found in Minn. Stat. §337.05, which is the provision allowing agreements to provide insurance for the benefit of others, and they take effect on August 1, 2013.

Some have described the amendment as a fundamental “game changer”; others believe it barely changes the status quo. As with most claims by advocates, the truth may lie somewhere in the spectrum between the poles, but it will likely take a few significant court cases to learn the exact impact of the new statute. What the amendment does expressly state is that agreements to provide insurance coverage to other parties for the negligence of any of those other parties is against public policy and void.

In the realm of commercial general liability (CGL) insurance, coverage provided to another party certainly means Additional Insured coverage because one party (e.g. the subcontractor) is agreeing to provide Additional Insured coverage to another party (e.g. the general contractor). Thus, the new amendment should significantly restrict the types of widely used blanket-type Additional Insured endorsements that an upstream promisee can demand from a downstream promisor. For example, the Insurance Services Organization (ISO) 2004 Additional Insured endorsements allowed Additional Insured coverage for the partial negligence of the additional insured; the new Minnesota amendment should restrict that coverage unless the endorsement is appropriately changed. The new ISO 2013 Additional Insured endorsements were drafted to respond to statutory amendments from many states restricting Additional Insured coverage to the promisor’s negligence. The 2013 endorsements limit that coverage “to the extent provided by law,” which in Minnesota would mean to the extent caused by the promisor’s negligence.

The new statutory amendments contain their own set of exceptions, however, that attempt to preserve many of the benefits intended by the original statute. For instance, the amendment expressly allows Builders Risk insurance because, even though it provides insurance to some other party for that party’s negligence, the premium is known up front and can be fully priced and allocated by contract. The exceptions also include performance and payment bonds and workers compensation insurance. The latter was granted an exception because the legislature has previously determined that a statutory program insuring the negligence of workers and employers in return for a quicker mechanism of payment to injured workers is good public policy.

Perhaps the most interesting exception is the one for project-specific insurance, including, without limitation, owner or contractor controlled insurance programs (known as OCIPs or CCIPs). On larger projects, the Owner can purchase an OCIP program of insurance including workers compensation and liability coverage for a certain price that will cover all on-site project participants. Because it is a one-time policy for a specific project, such an OCIP should not adversely affect the participants’ future insurance premiums for their regular, on-going insurance policies. Thus, OCIPs and CCIPs satisfy the amendment’s concern about forcing certain parties to assume unknown, future premium costs, which is why they received an exception from the scope of the amendment. There are other project specific policies that the parties can negotiate to provide that should also fall within this exemption.

Finally, the exceptions include promises to provide insurance coverage for the vicarious liability of the promisee, liability imposed by warranty, or work performed within 50 feet of a railroad. Often a promisee is sued by a third party because it is vicariously liable for the acts of the downstream promisor. For example, an Owner can be sued by an injured third party because the law requires that it be responsible for the work of the general contractor that it hired. This type of derivative, vicarious liability can drag a party into litigation simply because it hired the downstream contractor to perform the work. To avoid any potential confusion on the point, the amendment makes clear that the promisor can still agree to provide this type of liability insurance to the promisee.


There are still several questions surrounding indemnity agreements left unanswered by the statute. It often is argued that agreements to defend are different in kind and scope from indemnity agreements. Most contracts require downstream promisors to defend upstream indemnitees from even mere allegations of liability. The statute does not expressly address agreements to defend. Nor does the statute specifically address or modify agreements to indemnify. The original statute’s prohibition of broad form indemnity agreements was left untouched. What was modified was the allowable scope of agreements to provide insurance coverage to another party. The meaning of the phrase “to provide insurance coverage to one or more other parties, including third parties” will now be the focus of interpretation. Unfortunately, the original statute generated a significant amount of litigation to establish the meaning of that statute in the broad variety of circumstances that arise on a construction project. The new statute certainly intends to make a change, and hopefully not as much litigation will be needed to fully define its new scope.

As with any change, tried and tested approaches will have to be modified. Currently, existing contractual indemnity and insurance provisions have been literally tried and tested in many court battles. They will now have to be changed to respond to the new statute. Yet, contracts can still be negotiated and insurance purchased to achieve almost any desired result. The types of insurance specified and purchased, however, will have to be changed. Legal and insurance counsel should be consulted to help you draft and negotiate contractual agreements that will not, in the words of the new amendment, be found to be “against public policy” and “void and unenforceable”.


This discussion is generalized in nature and should not be considered a substitute for professional advice.© 2013 FWH&T