A closer look at the replacement costs clause

Author: Emma Krikke, attorney-at-law and partner in VanNiekerkCieremans Advocaten. Date of publication: 1 March 2017. This article is a translation of the Dutch article published in the law journal TAV. See: E.E. Krikke, "De vervangingskostenclausule nader beschouwd", TAV 2017, 2, p. 17 - 22.

Introduction

In its judgment of 4 October 2016, The Hague Court of Appeal considered the interpretation of a replacement costs clause in a business liability policy.[1] Unlike other exclusion clauses (for example, the intent clause), the replacement costs clause has generated little case law and literature. In practice, however, this exclusion for the costs of redelivering the property or service that has been delivered or provided plays an important role and frequently gives rise to discussion of how it should be interpreted.

The Court of Appeal’s judgment therefore provides an opportunity to take a closer look at the replacement costs clause and to provide practitioners with more guidance for cases in which they are confronted with this clause. First, I will outline the general context and discuss much of the limited case law that is available on the replacement costs clause. Afterwards, I will examine in more detail the application of this exclusion in the case of a living product by reference to the judgment of The Hague Court of Appeal of 4 October 2016.

The replacement costs clause

The replacement costs clause (also known as the redelivery clause) is one of the standard exclusions that appear in every business liability policy. By including this clause, the insurer aims to exclude its liability for business risk that should be borne by the insured.[2] After all, the insured must be responsible for correctly performing his obligations to his counterparty under the contract he has concluded. In other words, it is not the insurer’s role to pay compensation for bad business practice.

The wording of replacement costs clauses differs. The clause in the Dutch Insurance Exchange Liability Policy 2014 (‘NBA 2014’) reads as follows:

‘3.5 This insurance does not cover claims for compensation of:

3.5.1 Property damage to property delivered or completed by or under the responsibility of the insured.

3.5.2 The costs and expenses of the recall, rectification, replacement, remedy or repair of property delivered or completed by or under the responsibility of the insured, unless said costs and expenses qualify as loss mitigation costs.

3.5.3 The costs and expenses of the re-performance of activities carried out by or under the responsibility of the insured.

3.5.4 The exclusions described in articles 3.5.1 through 3.5.3 also apply to any loss or damage arising as a result of the property delivered or completed or the activities performed being unfit for (proper) use, irrespective of the party who sustained the loss or damage or incurred the costs and expenses.

3.5.5 In the event of property damage caused by property delivered or completed by or under the responsibility of the insured to other property that was delivered or completed earlier by or under the responsibility of the insured, the exclusions described in articles 3.51 through 3.5.3 do not apply to such other property.

3.5.6 In the event of property damage caused by activities performed by or under the responsibility of the insured to other property that was delivered or completed earlier by or under the responsibility of the insured, the exclusions described in articles 3.51 through 3.5.3 do not apply to such other property.

3.5.7 However, the exclusions described in articles 3.5.1 through 3.5.3 do apply if the property delivered or completed or the activities performed are the subject of one and the same contract.’

What are replacement costs?

The first question about the interpretation of the clause is exactly what damage is excluded under it. In other words, what falls under the ‘replacement costs’? Professor J.H. Wansink describes it as ‘all costs associated with the deployment of labour and the use of materials and means of transport in order to provide the other party with the agreed performance after all’.[3] What is excluded from coverage in a specific instance is a question of interpretation of the insurance contract. Under Dutch insurance law, insurance contracts should be interpreted in accordance with the criterion adopted by the Dutch Supreme Court in the Haviltex judgment. According to this criterion, the clause should be interpreted by reference to what the contracting parties reasonably understood by the clause in the given circumstances and what they could reasonably expect of each other.[4]

The application of the Haviltex criterion to insurance contracts was elaborated by the Supreme Court in its judgment in Chubb v Dagenstaed, where it held that in the case of insurance conditions that are not normally the subject of negotiations between parties (and have not actually been negotiated), the interpretation depends, in particular, on objective factors such as the wording of the clause in question when read in the light of the policy conditions as a whole and of any explanatory notes accompanying the policy conditions.[5]

It is apparent from the case law that the decisive factor in determining what costs qualify as replacement costs is what performance was contractually required. If the insured is contractually bound to deliver goods but not to install them, only the labour, material and transport costs associated with that delivery are excluded from coverage.[6] In the little literature that is available on the replacement costs clause, an example borrowed from Professor Wansink is often mentioned, namely the case of a plumber who has installed a pipe incorrectly and, in order to repair it, has to chisel out a section of the wall in which that pipe is located. The costs associated with chiselling out the pipe remain at the expense of the insured as they relate to the (proper) re-performance of the primary agreed deliverable, namely the installation of a functioning pipe.[7] If the repair of the material damage coincides with the (total) primary deliverable, the repair is therefore excluded under the replacement costs clause. If the repair does not coincide with the primary deliverable, the costs are covered by the insurance.

The judgment of Rotterdam District Court of 31 August 2011 is illustrative in this context.[8] In this case, so-called Weatherex nails were delivered to a contractor who had used them to attach wood cladding to houses. As the heads of the nails caused rust stains in the paint on the cladding, the contractor was obliged to remove the nails and the wood cladding so that new nails could be used. However, the removal of the rusted nails from the cladding caused such damage to the cladding that it too had to be replaced. To obtain compensation for his damage, the contractor sued the supplier of the nails, who in turn sued their manufacturer. The manufacturer claimed under its business liability insurance. The insurer refused to compensate the full damage, invoking the replacement costs clause. It was not in dispute between the parties that the delivery of the new nails was excluded from coverage. What was in dispute between them was whether the costs of installing new wood cladding and the costs of painting that new cladding were excluded by virtue of the replacement costs clause. The District Court answered this question in the negative. In so doing, it looked strictly at the nature of the primary deliverable which the insured was contractually bound to provide to its counterparty. This primary deliverable consisted solely of the delivery of the nails and did not involve their processing. According to the District Court, the damage caused by the removal of the defective nails did not come within the definition of the business risk, and was therefore not excluded under the replacement costs clause.

Property and/or activities belonging to the same transaction

Another interpretation problem occurs where property delivered by the insured is damaged by defective property delivered previously or a service provided previously. Most replacement costs clauses limit the exclusion to the property that has been delivered and has caused the damage; see also articles 3.5.5 and 3.5.6 of the NBA 2014 cited above. At first sight, the clause seems to be of limited value to the insured as article 3.5.7 subsequently provides that the exclusion remains applicable in its entirety if both items of property are part of one and the same transaction (the ‘single transaction provision’). However, Professor Wansink takes the view that it follows from the clause’s rationale that the insurer would act in a manner contrary to the dictates of reasonableness and fairness by invoking the clause in cases where the policyholder demonstrates that an item of property delivered by him has been damaged by another item belonging to the same transaction.[9]

The Supreme Court seems to have upheld this view in its judgment of 14 June 2002 (which is, incidentally, the only Supreme Court judgment that deals with the subject of the replacement costs clause).[10] In the case that resulted in that judgment, a contractor had improperly installed a fireplace as part of a more comprehensive construction contract; the fireplace was found to be insufficiently insulated. Use of the fireplace caused fire damage to the building. The contractor's business liability insurer refused to provide cover, arguing that the fireplace had been installed as part of the same transaction as the other work that had been carried out on the building. As a mistake had been made when installing the fireplace, the insurer took the view that the work of reinstalling the fireplace and carrying out once again all other work on the building was not covered by the policy. However, the Court of Appeal ruled otherwise (a decision which was later upheld by the Supreme Court on appeal in cassation). The Court of Appeal held that in the event of damage to delivered property that is part of one and the same transaction as the work that caused the damage, the costs of carrying out that other work are only excluded from coverage if that other work did not cause the damage but was also not properly carried out. It based this ruling on the clause’s rationale, namely that coverage of the direct consequences of defective performance by the entrepreneur is excluded, but not damage to all kinds of other items of property that were ‘coincidentally’ also delivered or manufactured by the entrepreneur and also suffered damage.[11] The rationale of the ‘single transaction provision’ also played a role in the Court of Appeal’s decision, namely to prevent problems in cases where it is hard to determine what should be considered the (relevant) cause of the damage. However, problems of that kind did not occur in the present case as the parties agreed that the damage was due to the faulty installation of the fireplace.

The costs of re-performing work that was part of the same transaction and was properly performed are therefore not covered by the replacement costs clause exclusion. However, the costs of repeating defective work that did not cause the damage are excluded from coverage. This is only logical because, according to the text, the replacement costs clause also relates to damage and costs incurred in connection with the rectification of property delivered under the responsibility of the insured.[12] This is also the case if the property to be rectified did not cause the damage. An example would be the situation in which a contractor had not only failed to insulate the chimney sufficiently but had also made mistakes (as part of a more wide-ranging construction contract) when installing the water mains. In such a case, the costs of installing proper water mains after the fire would not be covered.

Damage caused by property that has been integrated into other property

A different delineation problem occurs where delivered property has been used and processed by the other party and has thus lost its independence. This is problematic because the delivered property no longer exists, and the damage is thus caused to ‘new property’. Professor Wansink considers that in such a case the replacement costs clause only excludes the subsequent delivery of the sound property of the type in question.[13] However, in a specific situation it is not always clear exactly what costs are excluded. This can be illustrated by the following example. Suppose that the insured delivers milk powder, from which the purchaser makes and resells oatmeal porridge. It later transpires that the milk powder is contaminated with salmonella, which means that the oatmeal porridge too is no longer fit for consumption. In such a situation, the question arises of whether there is damage to a delivered product (the milk powder) or to another product (namely the oatmeal).[14] Depending on the answer to the question of what property was damaged, coverage exists under the policy.

Rotterdam District Court ruled on a similar issue in a case involving delaminating roof panels.[15] The central question in that case was whether the costs of replacing and repairing defective roof panels that had been mounted on and become part of roofs were excluded from coverage under the replacement costs clause. The insured argued that this was not the case, contending that the damage was not to the panels but to the roof.

According to the insured, the panels had been technically integrated into the entirety of the roof in such a way that they were no longer identifiable on their own and had in fact lost their own recognisable material autonomy, and the damage to the panels should be regarded as having affected the structural framework of the roof as a whole. According to the insured, the insurer could not invoke the replacement costs clause as the insured had delivered only the panels and therefore not the roof. The District Court took the view that it was indeed necessary to assess whether the panels had lost their material independence. If that were the case, the damage would not fall under the exclusion. According to the District Court, all circumstances of the case should be taken into account in making this assessment. The District Court considered that the roof panels had continued to exist ‘for the purposes of insurance law’, thereby explicitly deviating from an assessment criterion based to a greater extent on property rights law. Since, in its view, the panels were still identifiable as such even after they had been installed and they also did not form part of the main structural framework, the District Court held that the roof panels still existed independently and that their replacement was excluded under the replacement costs clause.

Replacement costs in the case of a living product

The question raised above as to whether the damage is to the delivered property or to new property also arises in the case of living products. To illustrate this, Professor Wansink gives the example of so-called mini tubers, which are delivered to a grower and then grown and propagated into seed potatoes. If the mini tubers are found to have been infected with ring rot, as a result of which the seed potatoes are no longer usable, the question arises of whether insurers can invoke a replacement costs clause. Wansink answers this question in the negative, because the product delivered by the insured was not the seed potatoes in which the infection was found but only their progenitor plants. He continues by stating that a different view of the question of coverage would be possible only if the seed potatoes in question were directly derived one-to-one from the delivered mini tubers.[16]

That ‘one-to-one’ approach was central to the judgment of The Hague Court of Appeal of 4 October 2016.[17] In that case, insurers refused coverage on the basis of the replacement costs clause, arguing that it could be inferred from Professor Wansink’s thesis that a so-called ‘one-to-one criterion’ existed and should be applied in cases where a living product is delivered by the insured. This defence is interesting because Professor Wansink himself does not mention a criterion, and no basis for such a criterion can be found in other literature either.

The facts of the case were as follows. Mol Agrocom (the insured) supplied first-year onion sets to growers. During cultivation, the onions (now in their second year) were found to be infected with downy mildew. The growers subsequently sued Mol Agrocom for compensation for their damage, whereupon Mol Agrocom claimed under its business liability insurance. However, the insurers refused to provide cover, invoking the replacement costs clause. The insurers argued that the first-year onion sets delivered by Mol Agrocom had grown one-to-one into onion plants and (subsequently) into second-year table onions. According to the insurers, this showed that the second-year onions that had been lost were the same products as the onion sets delivered by Mol Agrocom. The insurers also submitted that the application of the direct, one-to-one criterion cited by Professor Wansink in order to determine the scope of the replacement costs clause was not a matter of debate in the insurance industry. They went on to argue that there was no uncertainty at all about the application of this criterion. Applying the criterion should lead to exclusion of coverage.

The Court of Appeal discusses in detail the application of the ‘one-to-one criterion’ advocated by the insurers, but sees no reason why such a criterion should be applied. It starts by establishing that whether the damaged onions should be treated as ‘property delivered by or under the responsibility of the insured organisation’ is a question of interpretation of the insurance contract and not a question of biology. According to the Court of Appeal, the ‘one-to-one criterion’ plays no role in this interpretation, because it is not apparent from the text of the clause or from other circumstances that Mol Agrocom should have understood the clause to have that meaning. Moreover, as the Court of Appeal adds, the application of the one-to-one criterion leads to so much uncertainty that this alone makes it unsuitable for use as a guideline when it comes to interpreting the replacement costs clause.

Finally, the Court of Appeal holds as follows:

‘The Court of Appeal cannot accept the [insurers’] argument that within the insurance industry there is no uncertainty whatever about the application of the replacement costs clause to living objects (exclusively) on the basis of the one-to-one criterion.’

The Court of Appeal then explains how the clause should be interpreted. To begin with, it states that an exclusion should be interpreted restrictively. It goes on to hold that the size of the onion sets delivered by the insured, their suitability for human consumption and their fair market value differ to such an extent from those of the second-year onions that the latter must be regarded as new products. According to the Court of Appeal, this interpretation is also in accordance with the rationale of the replacement costs clause, namely that the insurer is not responsible for the quality of the delivered product and that there is consequently no coverage for damage directly caused by defective contractual performance. After all, the majority of the damage for which Mol Agrocom was held liable by its customers and for which it claimed coverage under its business liability policy consisted of damage other than the costs involved in delivering sound onion sets (‘the replacement costs’). The Court of Appeal also considers that a broader interpretation of the replacement costs clause would mean that the exclusion clause is and remains applicable to the entire process of growing and propagating, which would disproportionately limit the scope of the insurance.

The Court of Appeal concludes by observing that if insurers intend this clause to be interpreted more broadly, they must state this clearly and unambiguously in the policy conditions because the applicability of the one-to-one criterion advocated by the insurers is not apparent from the clause and the Court of Appeal also considers that this criterion does not have a sufficient capacity for differentiation when applied to living products.

The Court of Appeal therefore clearly considers that no one-to-one criterion is applicable to replacement costs clauses. The first consideration for the Court of Appeal is the appearance of the damaged property. If that differs (significantly) from that of the delivered property, any damage is considered to be damage to new property and not to the delivered property. The efforts which the customer has made to achieve this transformation are not taken into account by the Court of Appeal in assessing whether there is damage to delivered property. Accordingly, the Court of Appeal is seemingly in agreement with the view expressed by Rotterdam District Court in the case concerning delaminating roof panels The decisive question is whether – taking into account all the circumstances of the case – there is a new product. Here, the criterion applied in the case of a living product does not in fact differ from that applied to the intermingling of non-living products.

Finally, it is worth noting that, in the Court of Appeal’s opinion, it had been established that only 5% of the onion sets had been infected with mildew, and that this 5% had affected the remaining 95%. The insurers had invoked the provision in the replacement costs clause to the effect that the goods were part of one and the same transaction and therefore that the non-diseased onion sets too were therefore excluded from coverage. At first instance, the District Court had held that invoking that provision was unacceptable according to the standards of reasonableness and fairness. However, the Court of Appeal has ruled differently, considering it important that the case concerns a large batch of onion sets and that as mildew is a contagious plant disease it was only to be expected that the other onions from the same batch could also be infected with this disease. In these circumstances, the Court of Appeal saw no reason why it would be unreasonable or unfair to invoke the exclusion. This seems to differ from the Supreme Court judgment of 14 June 2002 discussed earlier in this article because in the present case the rationale of the ‘single-transaction provision’ does indeed necessitate exclusion, given that it is practically impossible to determine afterwards which onion sets were already infected upon delivery and which onion sets should therefore be regarded as the (relevant) cause of the damage.[18]

Conclusion

The replacement costs clause clearly aims to exclude the business risk from coverage. How the clause should be interpreted in specific cases regularly leads in practice to problems of interpretation. As there is little case law on the replacement costs clause, much depends on the facts of the case. The case law that is available is from lower courts and is highly dependent on the specific circumstances of the case. A constant factor is the decision by District Courts that exclusions should generally be interpreted restrictively, as should the replacement costs clause. Just as The Hague Court of Appeal concludes in its judgment of 4 October 2016, insurers who wish the replacement costs clause to be interpreted more broadly would be well advised to ensure that the clause is drafted in such a way that it is not open to two interpretations. In this context, the following tips may be of interest to practitioners.

• Insurers would be well advised to examine the replacement costs clause in the policy and assess whether – in the light of recent case law – it actually provides the desired coverage.
• A point deserving special attention in this connection is that processing may turn a delivered product into a new product. If it is the insurer’s intention to exclude damage to the ‘newly constituted’ product, a provision to this effect should be expressly included in the policy. In the absence of such an express provision, insurers run the risk that a court called upon to assess the clause may conclude from the facts of the case that the damage has been caused to a (new) product which differs from the delivered product and that coverage therefore exists under the policy.
• This aspect deserves extra attention where an insured delivers living property. The Hague Court of Appeal is clear in its ruling that it is not apparent from the replacement costs clause that a one-to-one criterion applies. If insurers wish to exclude damage to a product emanating from the follow-up process, this must be expressly stated in the policy conditions.

[1] The Hague Court of Appeal, 4 October 2016, ECLI:NL:GHDHA:2016:2811. This judgment is a follow-up to the earlier judgment of The Hague Court of Appeal of 16 February 2016, ECLI:NL:GHDHA:2016:282.

[2] J.H. Wansink, De algemene aansprakelijkheidsverzekering, Deventer: Kluwer 2006. p. 175.

[3] J.H. Wansink, op. cit., p. 178.

[4] Dutch Supreme Court, 13 March 1981, ECLI:NL:HR:1981:AG4158, NJ 1981/635 Haviltex judgment).

[5] Dutch Supreme Court, 16 May 2008, ECLI:NL:HR:2008:BC2793, NJ 2008/284 Chubb v Dagenstaed judgment). As it would be beyond the scope of this article to consider the interpretation of insurance contracts in depth, I refer the reader to J.H. Wansink, N. van Tiggele-van der Velde and F.R. Salomons, Mr. C. Assers Handleiding tot de beoefening van het Nederlands burgerlijk recht. 7. Specific contracts Part IX*. Insurance, Deventer: Kluwer 2012, no. 354 T. Riyazi this number of TAV: “Uitleg van verzekeringsovereenkomsten; waar staan we op de glijdende schaal?” TAV 2017/2.

[6] P.M. Leerink, ‘De aansprakelijkheidsverzekering’ in: M.L. Hendrikse, Ph.H.J.G. van Huizen and J.G.J. Rinkes (eds.), Verzekeringsrecht, serie Recht en praktijk, Deventer: Kluwer 2014. p. 672.

[7] J.H. Wansink, op. cit., p. 179.

[8] See Rotterdam District Court, 31 August 2011, ECLI:NL:RBROT:2011:BR7081, RAV 2011/111, legal ground 4.6.

[9] J.H. Wansink, op. cit., p. 191.

[10] Supreme Court, 14 June 2002, ECLI:NL:HR:2002:AE1536, NJ 2002/481.

[11]For a further commentary on this judgment, in which attention is paid to the derogating effect of article 6:248, paragraph 2, of the DCC, see T.J. Dorhout Mees, Annotation to judgment of The Hague Court of Appeal of 27 June 2000, no. 98/977 (not published at rechtspraak.nl) in AV&S 2001, p. 186 ff. For a detailed discussion of the Court of Appeal’s reasoning in that case, see also the opinion of Advocate General Keus, which accompanies the judgment of the Supreme Court in case ECLI:NL:PHR:2002:AE1536.

[12] In this context, see also the judgment of Rotterdam District Court of 12 September 2012 (Rotterdam District Court, 12 September 2012, ECLI:NL:RBROT:2012:BY1964) concerning a case in which insulation material had been delivered by the insured. The District Court held that the damage connected with the application of a double layer of insulation material and the costs incurred as a result of the reduction in the lifespan of the insulation material were covered by the exclusion since this damage was connected with the need to rectify or repair the property delivered by the insured.

[13] J.H. Wansink, op. cit., p. 179.

[14] This example is taken from an American case, namely: Netherlands Ins. Co v. Main Street Ingredients, LLC, Civil No. 11-533 (DSD/FLN), 2013, U.S. Dist. Lexis 2685. In that case, the judge determined that the damage had been caused to the oatmeal porridge and not to the milk powder delivered by the insured.

[15] See Rotterdam District Court, 17 April 2013, ECLI:NL:RBROT:2013:5837.

[16] J.H. Wansink, op. cit., p. 180.

[17] The Hague Court of Appeal, 4 October 2016, ECLI:NL:GHDHA:2016:2811. This judgment is a follow-up to the earlier judgment of The Hague Court of Appeal of 16 February 2016, ECLI:NL:GHDHA:2016:282.

[18] A detailed comparison of the two judgments would be beyond the scope of this article. However, in view of the limited literature on this subject, further consideration of the judgments is recommended.