Indemnification: A Misunderstood Concept

[Updated 3:00 p.m. EST, January 10, 2011:

In this post, I propose revised indemnification language. I’ll chew over whether still to offer as an option in Koncision’s confidentiality agreement the alternative language I offer below.]

[Updated 9:50 a.m. EST January 7, 2011:

The question of how the elements of a claim for indemnification relate to the elements of a claim for breach of contract has created enough of a stir that I can see that it might be counterproductive to incorporate indemnification language in Koncision’s confidentiality agreements. If I could cite authority to the effect that indemnification can be benign, then I’d be more of a champion for indemnification. But the fact is that I haven’t seen any authority either way on this issue.

I could try to be cute and simply say “shall pay” or “shall reimburse” instead of “shall indemnify.” But that would have zero substantive effect—anyone concerned with indemnification language should find such a switch equally problematic.

So now the question is how to articulate that the recipient is liable for unauthorized disclosure by its representatives (if that’s what the drafter elects to provide for). I remain lukewarm on “shall cause” when the party in question doesn’t actually have control. So here’s an alternative I came up with, with the input of Koncision editorial board member Chris Lemens:

Unauthorized Disclosure or Use by Representatives. If the Disclosing Party incurs any loss or liability arising out of disclosure or use of any Confidential Information by any one or more Representatives of the Recipient other than as authorized in this agreement, that disclosure or use will be deemed to have been by the Recipient for purposes of determining whether the Recipient breached any of its obligations under this agreement.

I’d be happy to hear what you think. But the most important point I take away from this discussion is that when you have three alternative ways of articulating an idea and none of them is perfect, it might be best to let the drafter choose. That’s currently what I’m contemplating doing.]

In this recent blog post I explained why I’m not a fan of imposing on a contract party an obligation that it doesn’t have control over. Rather than engage in that sort of indirect and counterintuitive risk allocation, I’d rather make my risk allocation explicit. One way to do that is by providing for indemnification.

My previous blog post was prompted by a confidentiality-agreement issue—what’s the best way to protect the disclosing party against losses caused by unauthorized disclosure by representatives of the recipient? I had pretty much opted for indemnification, but among those that I’ve consulted that has proved an unpopular choice, to an extent that has surprised me. The notion of indemnification has somehow become fraught with baggage; let’s look into that further.

Indemnification can be helpful in two ways. First, it allows you to replace a regime of contract claims with something more customized. For example, indemnification can help a party more likely to be subject to a claim by allowing it to specify time limits for bringing claims and put caps on liability. Second, indemnification can help a party more likely to bring a claim by, among other things, allowing it to bring in deep pockets and allowing it to provide a remedy for losses caused by nonparties.

I discussed the role of indemnification in this August 2009 AdamsDrafting blog post. What prompted me to write that post was the sense that drafters are too quick to throw indemnification provisions into a contract without considering whether a contract cause of action would be adequate.

That excessive use of indemnification may be what has prompted the resistance to indemnification that I’m now encountering. But I think we’re at risk of throwing the baby out with the bathwater. For example, it’s perhaps unhelpful to say that indemnification provisions don’t belong in confidentiality agreements. That’s like saying that representations don’t belong in confidentiality agreements. Instead, you have to look at how indemnification provisions are used.

I propose including in Koncision’s confidentiality agreements indemnification language that provides a remedy for losses incurred by the disclosing party that are caused by any representative of the recipient, for example a lawyer at its law firm. (This remedy would be in addition to any other contract claims.)

Here’s how that indemnification might be articulated:

Indemnification for Actions of Representatives. In addition to any other remedies available to the Disclosing Party, the Recipient shall indemnify the Disclosing Party against any losses and liabilities arising out of disclosure or use of any Confidential Information by any Representatives of the Recipient other than as authorized in this agreement.

That seems simple enough to me. And there are no hidden subtleties, in that the word indemnification, and the verb indemnify, don’t convey any hidden magic-word meanings. Instead, they simply serve to flag that the parties are providing for alternative causes of action, ones not otherwise arising out of the contract. I’m aware of  no basis for suggesting that, for example, the class of losses subject to indemnification is broader than the class of losses subject to contract claims, or that defenses to a claim for indemnification are drastically different from defenses to a claim for breach of contract.

And when determining whether any given any given losses or liabilities are covered by indemnification, you’d have to consider the same issues of foreseeability that you’d have to consider when determining whether any given losses and liabilities fall within the scope of a claim for breach of contract.

Suspicion of indemnification can perhaps be attributed in part to the tendency to see all sorts of doctrinal significance in what are in fact straightforward contract notions. (Represents and warrants, anyone?) And the fact that drafters clog up indemnification provisions with all sorts of extras shouldn’t undermine their essential utility.

Demonizing indemnification comes at a cost: it forces you to be less efficient in how you articulate risk allocation. This sort of categories-of-contract language issue is at the heart of MSCD, and I’m not particularly inclined to compromise.

But I am entirely open to being proved wrong. Have at it.

About the author

Ken Adams is the leading authority on how to say clearly whatever you want to say in a contract. He’s author of A Manual of Style for Contract Drafting, and he offers online and in-person training around the world. He’s also chief content officer of LegalSifter, Inc., a company that combines artificial intelligence and expertise to assist with review of contracts.

19 thoughts on “Indemnification: A Misunderstood Concept”

  1. I find that many lawyers automatically strike out an indemnity clause. Many of them are unwilling (read unable) to explain why. The objection generally seems to be that a claim under an indemnity clause would not be subject to the rules of causation, remoteness etc. I think that, absent any express drafting to the contrary, you are correct when you say that this position is unjustified.

    This does, however, raise my one objection to the use of indemnities. If, as drafted, an indemnity creates no greater rights than a simple breach of contract claim, it should only be used where a simple breach of contract claim would not suffice. Phrases such as ‘X shall indemnify Y against any breach by X of this Agreement’ are unnecessary and poor drafting. Yet I see them all the time. In your confidentiality agreement, I would agree with the drafting if you believe that disclosure by a Representative would not constitute a breach by the Recipient. In this case, the indemnity serves a valid purpose.

  2. Mike: You raise interested issues that I look forward to chewing over. But bear in mind that this part of the provision, whatever it ends up looking like, will only appear because the drafter opts for it in the Koncision questionnaire. I’ll be providing alternatives, plus guidance as to the implications of certain of the alternatives, but it’s the drafter who will decide what goes in. Ken

  3. I wouldn’t call it “[d]emonizing indemnification”; it’s more like avoidance behavior. Ideally clients would always soberly and rationally plan how they’d deal with various potential troubles. But sometimes they’re like Scarlett O’Hara — “I’ll think about that tomorrow.” A lawyer who criticizes that mindset too forcefully can acquire, fairly or not, a reputation as a deal-killer. Sometimes it’s OK to let the client implicitly assume the risk (after documenting that you advised of the risk and proposed a way to mitigate it, of course).

  4. Why not just say something along the lines that each party is responsible for its Representatives’ breaches of the Agreement to the extent that party would be liable had it breached on its own? This gives a completely backwards meaning to indemnity, which typically means third-party claims against the parties, not party claims against third parties. It also has the benefit of avoiding knee-jerk reactions to indemnities in NDAs, and brings in the limitations on liability that are negotiated without worrying about variance (for example, length of obligations, limitations on amounts, exclusion of special damages, etc) that would cause a party to incur more liability for its Rep’s breach than its own.

  5. @Adrian: I think it’s dangerous to assume that indemnification doesn’t give any broader rights than a breach of contract claim.

    Damages for breach of contract are usually meted out to return the parties to their original position. They have to be foreseeable and almost always, have to be causally linked to the breach. In other words, the Limitation of Liability language (not discussed in this post) will temper the breach claim down to direct and actual damages.

    On the other hand, indemnification-related damages are broader and deeper… and, if I wrote the Limitation of Liability language) are excluded from such limits. As such, they cover all damage types (including tenuous ones like indirects and specials), and for all of the things that Hadley v Baxendale has, over time, prevented.

    This is especially true when we’re talking about REALLY confidential information (such as PHI or PII – your financial data or your health information); or when we’re talking about breach of intellectual property restrictions. A breach of contract claim simply isn’t going to be enough.

      • Hi Ken: No, I can’t. But I’m not sure it’s relevant. It’s not the elements of the claim, it’s the fact that the limitation of liability language blocks almost all of the “juicy” damages (both in terms of type and amount) that people want.

        So perhaps, as is often the case with you, it’s a matter of language in terms of how I’m describing the issue. ;)

        There are really three types of claims:

        Failure to perform (which most people call breach)
        Attempt to ignore a contractual provision (which is breach, but of a different kind)
        Failure to perform in a way covered by indemnification (which most don’t call breach, they call it an “indemnification claim”)

        In the first case, the parties want damages but are typically limited to foreseeable and actuals (ie: direct damages capped at the value of the agreement).

        In the second case, you look to the contract to see if there’s a contractually-stated remedy. An example would be anti-assignment language, where the contract lists the potential solutions and doesn’t leave any choice as to behavior (“any attempt to assign this agreement shall be null and void”). [Sorry Ken, ignore the “null and void” for purposes of this conversation.]

        In the third, it’s also a contractually-stated remedy: the specific steps and actions required under the indemnification clause.

        So it’s a matter of language, not elements of the claim. Saying “breach of contract” really covers everything. If A and B contract, A acts in a manner so as to give rise to their indemnification of B, and B needs to sue to force A to indemnify, it’s still a “breach of contract” claim. The court is bound to enforce the terms of the agreement. Only B is going to seek specific performance as to the indemnification clause.

        Where I think things get even MORE interesting is that some contracts go on to have termination language that becomes active (or has the potential to become active) in the event of a breach. Coupled with language regarding “survival” (the ability of a contractual term to survive termination of the agreement for some period of time), things get extremely fascinating.

          • Yes, I think so. I wonder if people get wrapped around the axle when they’re thinking about breach of contract claims as being separate from indemnification claims only because of contractual language that says that “in addition to the remedies available herein, the non-breaching party shall be entitled to all other rights and remedies at law or equity”?

            And, perhaps, they’re thinking that in the case of indemnification, there are other legal and/or equitable remedies that would attach?

        • Great post!

          Jeff: I practise in India. In almost all the contracts the Indemnification and LOL provisions consumes a lot of time and delays the closure of the contracts. For long I could not understand the specific reasons for building an indemnity provision in the contract when the non defaulting party could claim damages for breach of contract.
          My view is that the Indemnification provisions also get tested against the provisions of the contract law: causation, remoteness, actual damages only etc; except that in the case of Indemnification, the party claiming indemnification need not have to wait until he has actually suffered a loss to bring a claim. [ruling of an Indian court]

    • I agree with you Jeff. I have always wondered why a counterparty wants the indemnity language when he already has a breach of contract action. I always try to limit indemnity clauses. to claims of third parties for bodily injury or property damage. As for first-party claims, contract law supplies the remedy. I also fear that use of indemnity for a contract claim exposes the indemnitor for greater liability than might otherwise apply. It is bad enough that courts are developing “contort” legal theores to supply tort actions for contract breaches; we do not need to give them opportunities to mix tort and contract theory by use of indemnity clauses in this context.
      If I were forced to accept an indemnity, perphaps I would try to scale it back by saying “To the extent any damages are determined in accordance with an action for breach of contract, the party in breach will indemnify the nonbreaching party for such damages”. But that is a waste of words, akin to a restatement of contracts law.

  6. Some ways to simplify:

    – Isn’t “In addition to any other remedies available to the Disclosing Party,” redundant?

    – “use of any Confidential Information by any Representatives of the Recipient other than as authorized in this agreement” = unauthorized use of confidential information

    – What’s the difference between use and disclosure of confidential information? If you disclose something, haven’t you necessarily used it?

  7. Ken, my view is that CDAs should be well-drafted but should also usually be drafted so as to minimise discussion time, ie stick with something that the other side will be comfortable with (whether as discloser or recipient – I also tend to go for 2 way CDAs). For this reason I tend to avoid:

    1. Indemnities
    2. Representations
    3. Explicit obligations on a party to be liable for breach by its representatives
    4. Liquidated damages for breach of the CDA

    On indemnities, the fact that some people think that an indemnity is broader than a contractual liability (eg as to remoteness, duty to mitigate, etc) is enough for me to want to avoid the term (in CDAs) unless I have a good reason to include it. I usually find it is unnecessary to specify the remedies for breach. (That also raises the issue of whether to include those clauses about injunctions, but that is straying off topic.)

  8. Canadian lawyers when using indemnification language in contracts have had to consider and contend with an Alberta case (the results of which were given the stamp of approval by the Supreme Court of Canada) of Mobil Oil Canada Ltd. v. Beta Well Service Ltd. (from 1973-1974). The gist of this case is that the concept of indemnity is limited to third party claims not to direct damages. Thus on this theory of interpretation if “Party A agrees to indemnify Party B if [___ fill in trigger for obligation___]” this would only protect Party B from claims made by third parties against Party B. It has been suggested that if direct loss and claims by 3rd Parties are supposed to be covered then added language is required such as “Party A agrees to be liable to Party B for Party B’s losses and agrees to indemnify Party B against any claims against B by other Persons if [___ fill in trigger for obligation___]. Not sure how relevant you will find this in the context of your CDA discussion but it is something Canadian lawyers must consider whenever using indemnification language.

    • Gord: I just finished reading the Mobil case; thanks for bringing it to my attention. The language at issue in the case was a mess, and the lower court’s holding didn’t bowl me over. I think it’s untenable as a matter of semantics to say that indemnification pertains only to third-party claims, and it’s inconsistent with how the word is actually used. You can make the meaning clear by how you articulate what is being indemnified against. Ken

      • An indemnity is more a debt obligation and the indemnity gives a contractual right to be made whole. I try and maintain indemnities only for claims made by third parties against one of the two parties to the agreement. For any losses that may arise from the contract, that should be left to damages between the parties. Having those as part of the indemnity could create a windfall for one party.

        • Excuse me, my previous reply didn’t make sense and was borderline rude: I was scared off by your first sentence! Broadly, I agree with you, but MSCD and the 2009 post linked to above suggest context where indemnification for claims between the parties can make sense.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.