Robert M. Hall

[Mr. Hall is of counsel in the Washington D.C. office of Piper Marbury Rudnick & Wolfe.  He is a former insurance and reinsurance executive and acts as an insurance consultant as well as an arbitrator and mediator of insurance and reinsurance disputes.  The views expressed in this article are those of the author and do not reflect the view of Piper Marbury Rudnick & Wolfe or its clients.  Copyright 2000 by the author.  Questions or comments may be addressed to the author at robertmhall@erols.com.]


Reinsurance underwriters seldom have the time to duplicate the underwriting efforts of their clients and for this reason rely, in part, on the cedent’s self-interest in seeking an underwriting profit on an individual risk or on a book of business.  To assure that the cedent actively seeks an underwriting profit, reinsurance contracts sometime require that the cedent maintain a specific “net retention” which is usually quantified but is sometimes not completely defined.  The purpose of this article is to explore various issues relating to net retentions and the case law related thereto.

Does an Unrevealed “Front” Violate the Expectation of a Net Retention?

While it is common practice that the policy issuing company will retain some net risk, it is not always contractually required to do so.  Some cedents and reinsurers engage in”fronting” relationships wherein a licensed company will issue policies and cede all or virtually all of the risk to the reinsurer which is often unlicensed in the relevant jurisdictions.  In general, the respective retentions, or lack thereof, in such situations are well understood by the parties.  However, there are some situations in which the cedent’s lack of a retention is not understood by the reinsurer or is concealed by the cedent.  This raises the issue of whether it is custom and practice in the insurance industry for a cedent to keep a retention such that the lack of a retention must be revealed to the insurer. 

An early case on point is Charlaron v. Ins. Co. of North American, 21 So. 267 (La.1896) wherein the cedent represented that it had a retention but did not quantify it.  The reinsurance related to cargo risks which were anticipated to have certain values but, in fact had lesser values.  The reinsurers argued that the cedent’s retention should be that remaining after subtracting the dollar value of reinsurance from the anticipated value of the cargo.  The court rejected the reinsurer’s argument and, in the course of its opinion, commented on custom and practice:

Reinsurances (sic) while they may be of the entire risk, are ordinarily of such portion as the insurer deems proper to insure. (citation omitted) Hence, it may be said that in reinsurances (sic) the general rule is the original insurer retains a part of the risk, and in the event of total loss, to the extent of the amount retained, shares the loss with the reinsurers.  If this is not the reinsurance desired, but the reinsurance is of the entire risk, for obvious reasons good faith exacts that purpose should be stated to the reinsurer. . . . All that can be said of any usage bearing on the controversy is that in reinsurances (sic) the original insurer, without express statement to the contrary, retains part of the risk, and that amount no custom fixes.[1]

Reliance Ins. Co. v. Certain Member Companies, 885 F.Supp. 1147 (S.D.N.Y.1995) involved a cession wherein the placement materials obscured the fact that the primary company was ceding 100% of the risk it assumed.  The court found that the cedent’s agents had concealed the 100% cession and that the reinsurer would not have assumed the risk under such circumstances.  The court commented: “The refusal of a reassured to retain any portion of a risk is not customary, at least outside the context of ‘blue water hull’ insurance or situations involving ‘fronting’, and shows a lack of confidence in the risk.”[2]  The court went on to find a custom and practice to reveal a 100% cession, citing Charlaron v. Ins. Co. of North America, and allowed rescission of the reinsurance agreement as a result.  But see Ins. Co. of North America v. Hibernia Ins. Co., 11 S.Ct. 909, 911 (1891) in which the court found only a local usage not to reinsure the whole risk. 

Several cases involving retrospective premium plans provide a different factual twist on the unrevealed fronting issue.  In Carlingford Australia Gen. Ins. Ltd. v. St. Paul Fire & Marine Ins. Co., 722 F.Supp. 48 (S.D.N.Y.1989), the cedent had a retrospective rating plan with the insured which reimbursed the cedent dollar for dollar up to $1,550,000 and by which the cedent received an administrative fee of $275,000.  The reinsurer assumed a layer of $4 million in excess of $1 million.  The reinsurer argued that the retrospective rating plan should have been revealed since it provided the cedent with double coverage and guaranteed the cedent a profit, thus providing no incentive to investigate and minimize claims.  The court granted a motion to amend the reinsurer’s answer to assert an additional defense and a counterclaim for rescission.

Travelers Ins. Co. v. Buffalo Reins. Co., 1990 WL 116741 (S.D.N.Y.) involved a facultative certificate in which the cedent “warrants to retain for its own account or that of its treaty and/or facultative reinsurers the amount of the liability specified in Item 3.”[3] The underlying policy was subject to a retrospective rating plan which is not described in any detail in the opinion.  The cedent argued that retrospective rating plan adjusted premium but did not alter its risk of loss.  The reinsurer argued that the retrospective rating plan reduced the cedent’s net risk of loss in the same fashion as unauthorized reinsurance of the net retention.  The court found the net retention clause to be ambiguous, allowed the reinsurer to amend its answer to include the retrospective rating plan defense but denied the reinsurer’s motion for summary judgment.

This case law suggests that a cedent conceals its lack of a retention at its peril.  However, the impact of retrospective rating on the cedent’s net retention suffers from lack of conclusive case as well as a variety of factual circumstances.  For instance one result may pertain to a rating plan which compensates the cedent dollar for dollar for each loss paid.  But another result may pertain to a more standard retrospective rating plans in which the insured and cedent share risk within a certain range of results and the insurer pays losses outside that range. 

What is the Result When a Net Retention Warranty is Breached?

Fortress Re Inc. v. Jefferson Ins. Co., 465 F.Supp. 333 (E.D.N.C.1978) aff’d 628 F.2d 860 (4th Cir.1980) involved a retention warranty in a facultative certificate.[4] The cedent failed to comply with its warranted retention and the court found that compliance was a condition precedent to the reinsurer’s liability:

The parties’ intent is explicated further by the retention provision’s very nature. (The cedent’s) agreement that it would retain a certain degree of risk vis-à-vis the (reinsurer) contemplates maintenance of these limits throughout the period prior to a reimbursement on the policy. The reasons for this are obvious and lie at the heart of the reinsurance law and finance. The greater a reinsured company’s stake in the outcome of litigation, or a claim based on the original policy, the more attentive that company will be to the investigation of the claim and minimization of the claimant’s recovery.[5]

The issue of net retention as a condition precedent arose also in Stonewall Ins. Co. v. Fortress Re Managers, Inc., 350 S.E.2d 131 (N.C. App.1986). The court upheld the finding of the lower court that the warranty provision was a condition precedent:

The rule has long been established in this jurisdiction that one party’s failure to comply with a condition precedent to a contract relieves the other party of its duty to perform under the contract irrespective of the prejudicial effect. (citations omitted) The representative of (the reinsurer) who testified explained that having the ceding company actually liable on the risk was significant to (the reinsurer) in terms of management and handling of the claims. We hold that (the cedent’s) compliance could reasonably be expected to influence the decision of the insurance company and that the trial court did not err in concluding that the (cedent’s) breach of the condition precedent was material. [6]

In a subsequent decision, the court ruled that the victor in the prior decision was collaterally estopped from taking the opposite position on another facultative certificate on which the ceding and assuming companies reversed positions. Penn Re, Inc. v. Stonewall Ins. Co., 708 F.Supp. 123 (E.D.N.C.1988).

Violation of a retention warranty was a factor in a series of related decisions which reached essentially the same result on this issue: Compagnie de Reassurance D’ile de France v. New England Reins. Corp., 825 F.Supp. 370 (D.Mass.1993); Compagnie de Reassurance D’ile de France v. New England Reins. Corp., 57 F.3d 56 (1st Cir.1995) and Compagnie de Reassurance D’ile de France v. New England Reins. Corp., 944 F.Supp. 986 (D.Mass.1996). The reinsurance contract contained a series of convoluted provisions related to the cedent’s retention which allowed some treaty business to reduce the cedent’s $250,000 net retention on the facultative risk.[7] To satisfy the concerns of the reinsurers, however, the contract contained a separate provision requiring the cedent to "co-reinsure for 10% participation on" one class of business ceded.[8]. The district court found no breach of the $250,000 net retention but found that the cedent had materially breached the warranty to maintain a 10% participation.[9] Ultimately, the court found that the breach of this warranty was intentional and constituted fraudulent concealment.[10]

The cedent contractually agreed to maximum cessions to be a multiple of the line it retained in Columbian Nat. Fire Ins. Co. v. Pittsburgh Fire Ins. Co., 210 N.W. 258 (Mich.1926). In the event that this was not adhered to, the contract called for the cession to be reduced to the appropriate amount. The court ruled that the issue of fraud should go to the jury stating:

(The cedent’s) records show repeated violations of the reinsurance contract by way of cessions to (the reinsurer), of portions of risks in excess of contract limitation, and failure of (the cedent) to retain stipulated parts of risks. The violation of the contract, by way of excess cessions to (the reinsurer), should have been readily apparent to both parties from their own records, and we may well leave them to their own remedy of readjustment in accordance with the terms of the contract. (citation omitted) Violations, by way of cessions to (the reinsurer) and other reinsurers, in breach of the contract requirement of retention of proportionate risk by (the cedent), could not have been discerned from information furnished (the reinsurer), and if done with intent to defraud rather than mere inadvertence, then reinsurance so procured from (the reinsurer) was in each instance void, and payments of losses thereon, without knowledge of the fraud, could, under proper pleading, could be recovered.[11]

There are several cases in which literal non-compliance with a net retention clause was excused by the courts. One is Scottish Fire Ins. Co. v. Stuyvesant Ins. Co., 76 S.E. 728 (N.C.1912) in which the court found that the reinsurer knew that the cedent had not complied with the net retention requirement and, therefore, waived it. Another is General Reins. Corp. v. Southern Surety Co., 27 F.2d 265 (8th Cir.1928) in which the cedent’s retention was less than that required on the contract in question but cedent had a retention which exceeded the gap on another contract. The court ruled:

The contention of the plaintiff in error that risk of loss under the $75,000 bond was reduced below the requisite amount by reason of the two provisions above mentioned in the bond is, we think, without merit. . . . The provision of the reinsurance contract relative to the retention by the surety company of risk of loss did not increase or diminish the risk of loss of the reinsuring companies. That provision was in reality simply to provide evidence of good faith.[12]

See also Ins. Co. of North America v. Hibernia Ins. Co., 11 S.Ct. 909 (1891) in which the court found that the cedent had not warranted any particular retention.

The case law suggests that the failure by a cedent to maintain an agreed, specified retention is likely to be regarded as a material breach of the contract which may result in rescission. If, however, the reinsurance contract calls for the reinsurer’s net line to be reduced proportionately to the cedent’s actual retention, this may be the appropriate remedy.

How Net is Net?

A number of cases raise the issue of whether "net retention" means net of all reinsurance, such as catastrophe or treaty reinsurance. One such case is Northwestern Mut. Fire Ass’n v. Union Mut. Fire. Co., 50 F.Supp. 785 (W.D.Wash.1943) aff’d144 F.2d 274 (9th Cir.1944). Pursuant to a treaty, the reinsurer agreed to retain $50,000 per risk, which was equal to the cedent’s retention. In the event that the cedent failed to retain its agreed risk, the reinsurer’s portion would be reduced accordingly. The cedent argued that by custom and practice, the term "net retention" did not include catastrophe excess of loss reinsurance, however this argument was rejected:

[T]he court finds that the terms of Article VIII (related to net retention) of said treaty of January 1, 1940 are plain, clear and unambiguous and do not permit of modification, amendment or interpretation by extrinsic evidence. That the court further finds that in any event, under the customs and usages of the insurance business and in the insurance world, the term ‘amount retained net without reinsurance by the reinsured company at its own risk and liability’ . . . does include and does apply to excess of loss reinsurance . . . and means what it says, namely: the amount retained net by the reinsured company . . . after deducting all reinsurance, including excess of loss reinsurance . . . . [13]

An argument that treaty reinsurance could be used to reduce a cedent’s net retention under a facultative certificate was made in Stonewall Ins. Co. v. Fortress Reins. Managers, 350 S.E.2d 131 (N.C.App.1986). The lower court found that the net retention clause was unambiguous and that there was no custom and practice in the insurance industry which would allow a net retention on facultative business to be reduced by treaty reinsurance. The appellate court affirmed stating:

The two people who negotiated the contract on behalf of the plaintiff and on behalf of defendants testified. Both testified that there was no discussion as to whether a portion would be ceded to treaty insurance or not. Hence there was no evidence of an intention by the parties to include treaty reinsurance as part of the amount retained by the company of its own account. We hold that the trial court did not err in finding the term ‘for its own account’ unambiguous and in ruling as a matter of law that the term did not include both net retention and treaty reinsurance.[14]

A subsequent court found that the reinsurer in Stonewall was collaterally estopped from making the same argument in the role of cedent in Penn Re, Inc. v. Stonewall Ins. Co., 708 F.Supp. 123 (E.D.N.C.1988 aff’d 894 F.2d 402 (4th Cir.1989). See also Fortress Re, Inc. v. Jefferson Ins. Co., 465 F.Supp. 333 (E.D.N.C.1978) aff’d 628 F.2d 860 (4th Cir.1980) in which treaty reinsurance was not allowed to net down a facultative retention.

A case allowing treaty reinsurance to net down a facultative retention is Commercial Union Ins. Co. v. Seven Provinces Ins. Co., 9 F.Supp. 49 (D.Mass.1998) aff’d 217 F.3d 33 (1st Cir.2000). The facultative agreement contained a convoluted net retention clause:

(1) Being a reinsurance of and warranted same NETT rate, terms and conditions as and to follow the settlements of (the cedent) and that the local office of the said Company retains during the currency of this insurance at least $225,000 BEING 50% of $450,000 EXCESS $50,000.00 COMBINED SINGLE LIMIT (2) subject to reduction by any general excess of loss catastrophe reinsurance whether effected by the head office or local office of the Company (3) on the identical subject matter and risk and in identically the same proportion on each separate part thereof, but (4) in the event of the retained line being less than as above, Underwriter’s line to be proportionately reduced.[15]

Not surprisingly, the court found this language to be ambiguous and allowed expert testimony on point. The expert for the cedent offered a variety of linguistic explanations for his position but also testified that it was custom and practice to allow treaty reinsurance to net down facultative exposure. An intent to disallow treaty reinsurance would have been indicated by unequivocal language. Further, he testified that the incentives which cause the use of a net retention clause "simply do not apply, however, when the reinsurance in question is treaty reinsurance. In that case, the reinsurance covers an entire class of business and cannot affect the incentives of evaluate any particular risk carefully." [16]The court found the testimony of the cedent’s expert witness more persuasive than that of the reinsurer and ruled accordingly on this point.

Case law suggests that the courts will enforce a straight-forward provision or warranty of net retention as written. However, convoluted language may be read as ambiguous generating a battle of the experts over custom and practice.

What is the Impact on Net Retention if the Cedent Drops Down?

Several cases explore the obligations of the reinsurer of an excess insurer when the excess insurer/cedent is obligated to or chooses to drop down into layer which is self-insured. The facultative certificate in Calvert Fire Ins. Co. v. Yosemite Ins. Co., 573 F.Supp. 27 (E.D.N.C.1983) stated that "(cedent) warrants to retain for its own account the amount of liability specified in Item 3 unless otherwise provided herein and the liability of (reinsurer) specified in Item 4 shall follow that of the (cedent) . . . and shall be subject in all respects to all the terms and conditions of (cedent’s) policy." [17]The policy contained a self-insured retention but provided that the cedent would drop down into the self-insured retention if the insured became insolvent. The court found that the reinsurer was not obligated to drop down with the cedent:

[T]he court finds that the first provision of the reinsurance certificate as it related to Item 4 is not ambiguous and that (the reinsurer is) entitled to judgment as a matter of law. By referring to the limits specified in Item 4 before providing that liability is to follow that of (cedent), the first provision of the reinsurance certificate must be construed to mean that (the reinsurer’s) liability follows that of (the cedent) with the limits specified in Item 4. To construe otherwise would effectively eliminate the provision in Item 4 that (the reinsurer’s) liability begins after the self-insured retention has been exhausted. [18]

An umbrella insurer, which was in excess of a $300,000 underlying layer reinsured 95% of its risk in Michigan Millers Mut. Ins. Co. v. North American Reins. Corp., 452 N.W. 2d 841 (Mich.Ct.App.1990). The insured only maintained a $100,000 underlying layer. Responsibility for the gap was unclear and the umbrella insurer contributed $25,000 toward a funding of the gap. When the umbrella insurer sought reimbursement from the reinsurer for a loss below the $300,000 layer, the reinsurer declined and suit followed. The court found that the obligation of the reinsurer to indemnify the cedent did not commence until the $300,000 layer had been breached. [19]

Case law suggests that a reinsurer will not be required to drop into a layer below that reinsured simply because the cedent is required or chooses to do so.


While there is not a plethora of case law dealing with net retention clauses in reinsurance contracts, that which exists indicates that warranties or representations of net retention will be enforced as written but that convoluted clauses may require expert testimony which can produce a different result.  The lack of a retention should be revealed by a cedent to its reinsurer.  In addition, reinsurers will not be required to drop down below reinsured layers simply because the cedent does so.


1.   21 So. 267 at 269.

2.   886 F.Supp. 1147 at 1150.

3.   Id. at *1.

4.   The retention provision reads:

The Company warrants to retain for its own account the amount of liability specified in Item 3 unless otherwise provided herein, and the liability of the Reinsurer specified in Item 4 shall follow that of the Company, except as otherwise specifically provided herein, and shall be subject in all respects to all the terms and conditions of the Company’s policy.

       465 F.Supp. 1147 at 1150.

5.   Id. at 339.

6.   350 S.E.2d 131 at 135.

7.   The contract provided:

The reassured’s minimum retention hereunder shall be $250,00 as respects any one risk. . . .
General Conditions: As respects so called “System Business” ceded to this Treaty, the net retention borne by this Company within the System, other than the Reassured, shall constitute compliance with the net retention requirement hereon.
. . .

For purposes of the net requirement hereon as regards “System Business,” quota share and excess of loss reinsurance maintained by a company within the System shall be disregarded.  For purposes of the net retention requirement hereon as regards “Non-System Business.”  Only excess of loss reinsurances (sic) maintained by the Reassured shall be disregarded.

825 F.Supp. 370 at 375 - 6.

8.   Id.

9.   825 F.Supp. 370 at 382.

10. 944 F.Supp. 986 at 1000.

11.  210 N.W. 258 at 259.

12.  27 F.2d 265 at 271.

13.  50 F.Supp. 785 at 789.

14.  359 S.E.2d 131 at 134.

15.  9 F.Supp. 49 at 53.

16.  Id. at 55.

17.  573 F.Supp. at 29.

18.  Id.

19.  452 N.W.2d at 843.