106 As the judgment of Gleeson CJ in Montgomery at 33 [18] makes clear, the non-delegable duty of care that was imposed on the appellant in Burnie Port Authority was imposed because of the extremely hazardous nature of the activities that the appellant caused to be carried on. It was not submitted in this case that the duties of repair and maintenance in respect of which the non-delegable duty of care was said to be imposed on Railcorp, were activities of such a hazardous nature.
107 Gleeson CJ pointed out in Montgomery at 29 [9]-[10] that the non-delegable duty of care in issue in that case was not said to arise because the responsibility (for repair of roads and footpaths) was imposed on councils in terms or in a context that required them to perform it personally. It was, instead, said to arise because the responsibility of councils was said to be not merely to exercise reasonable care if they undertook the work themselves, but also, if they engaged a subcontractor, to ensure that the subcontractor exercised reasonable care. As his Honour pointed out at 34 [22] - 35 [24], such a duty was one that "[i]n many circumstances… could not be fulfilled". It was in truth "not a… special kind of duty, but an imposition… of a special kind of vicarious responsibility". If such a duty were to be imposed, "[i]t would be necessary to identify and justify the exceptions to the general rule that a defendant is not vicariously responsible for the negligence of an independent contractor, and to provide a means by which other exceptions may be identified when they arise".
108 Kirby J made a similar point at 60 [104]. His Honour said that the recognised categories of non-delegable duties of care were "exceptional" and existed "in well-established categories that are recognised by the law". His Honour pointed to "the general trend of contemporary tort law, to limit exceptional categories, and to reject new ones except on the basis of a clear analogy to a recognised class and then only for compelling reasons of legal principle and policy".
109 If one were to apply the reasoning process identified by Gleeson CJ and Kirby J, it would be necessary, in the course of justifying the exception to the general rule to which Gleeson CJ referred, to identify some existing category of non-delegable duty that could be extended by "clear analogy" to the present case.
110 Performance of that task would not identify the category exemplified in Burnie Port Authority as an appropriate analogy. That is because, as I have said, the decision in that case rested fundamentally on the proposition that the appellant was carrying out, or causing to be carried out, extremely hazardous activities. Instead, one might think, an appropriate analogy could be drawn with the responsibilities of authorities to maintain and repair highways. In essence, a railway track is no more than a highway on which trains, rather than motor vehicles, travel. The consequences of non-repair are serious in each case; perhaps more serious in the case of railways, because one train customarily carries many more people than one motor vehicle. But the activities of maintenance and repair are not of themselves extremely hazardous.
111 To complete the analogy, attention should be paid to the statutory obligations and powers conferred on highway authorities, and those imposed on Railcorp. The parties did not undertake that exercise. Thus, the analogy with highway authorities cannot be said to be complete, or perfect. But on the limited material available, it seems to be reasonably close. If no non-delegable duty of care is imposed on highway authorities, why should one be imposed on Railcorp?
112 I have dealt with this not because I think that Railcorp in 2000 or 2003 should have foreseen, and considered, developments in the law relevant to the question of its potential liability to SRA, but because of the reliance that Mr Jackman placed on the decision of Gillard J in Twentieth Super Pace Nominees. For the reasons that I have given, I do not think that the case offers a great deal in support of Railcorp's position. But if it is to be taken into account, then subsequent developments in relation to the topic of non-delegable duties of care should not be ignored.
113 I conclude that Railcorp did not owe a non-delegable duty of care to SRA in relation to maintenance and repair of the railway track and infrastructure vested in Railcorp. Thus, I do not accept Railcorp's submissions that its settlement with SRA can be justified on the basis that it owed such a duty.
Second basis: clause 6.2(a)(i)
114 Clause 6.2(a)(i) of the Access Agreement obliged Railcorp to maintain the rail infrastructure to at least the minimum standard required for accreditation as an owner of rail infrastructure. Section 16 of the Rail Safety Act, as it was then in force, provided that an applicant for accreditation as an owner must demonstrate that it possessed the competency and capacity to safely construct and maintain, or construct or maintain, the infrastructure. Thus, Mr Jackman submitted, the obligation under cl 6.2(a)(i) was to construct and maintain, or construct or maintain, the railway infrastructure so that it would be safe for use.
115 Mr Donaldson submitted that cl 6.2(a)(i) of the Access Agreement was "directed towards maintenance process and not outcomes". So too, he submitted, was s 16(1) of the Rail Safety Act. In substance, Mr Donaldson submitted, s 16(1) required the applicant for accreditation to demonstrate that it had in place an appropriate regime for maintenance and repair. That obligation, he submitted, was the one that was picked up by cl 6.2(a)(i). He submitted that cl 6.2(b) of the Access Agreement, which identifies the circumstances in which Railcorp could cease to maintain particular lines, reinforced that conclusion.
116 Mr Robson in substance adopted the submissions put on this point by Mr Donaldson.
117 I do not accept that cl 6.2(a)(i) is concerned only with the preparation of an appropriate plan, or regime, for maintenance or repair. The obligation that it imposes on Railcorp is one to "maintain" the rail infrastructure with which it is concerned. An obligation to maintain is not satisfied merely by putting in place a plan of maintenance.
118 The implied reference to s 16 of the Rail Safety Act is not for the purpose of demonstrating that the obligation to maintain imposed by the clear words of cl 6.2(a)(i) is one to be satisfied merely by the preparation of an appropriate plan and or regime, but to demonstrate the standard to which maintenance is to be performed. That standard is one that would satisfy (or be capable of satisfying) the Director-General that Railcorp was competent to construct and (or) maintain the relevant infrastructure. True it is that part of that process of satisfaction would require the submission of an appropriate plan for maintenance and repair. But it would involve also demonstrating that maintenance and repair according to that plan would enable the infrastructure to be "safely" maintained. Of necessity, that involves as a fundamental objective that the infrastructure be safe for its intended use.
119 To construe cl 6.2(a)(i) as Mr Donaldson submits would require the Court to ignore the plain meaning of the verb "maintain" and to substitute for it some obligation such as "plan for the maintenance of". In circumstances where the first stated objective to be achieved by the agreement is the "efficient and safe management and control of the NSW rail network including Rail Infrastructure Facilities" (cl 2.2(a)), the obligation to maintain should not be read down or diluted in the way that Mr Donaldson submits should be done.
120 Thus, I conclude, cl 6.2(a)(i), properly construed, does impose on Railcorp an obligation safely to maintain the relevant infrastructure.
121 That is not the end of the question. For the reasons that I have given, that obligation is not (and in 2000 and 2003 was not) one that involved some non-delegable duty. It was one that could be satisfied by entry into an appropriate form of contract, for the proper performance of appropriate maintenance works, with an appropriately qualified and competent contractor. Perhaps not surprisingly, Mr Donaldson did not submit that Fluor was not such a contractor. Nor did he submit that the Alliance Contract was not an appropriate way for Railcorp to discharge, in respect of the East Hills line, its obligation of maintenance. In particular, Mr Donaldson did not submit that the express consent given to the employment of subcontractors involved any derogation from whatever duty to maintain might be imposed on Railcorp by the Access Agreement.
122 In my view, the Alliance Contract was an appropriate mechanism for the performance of Railcorp's obligations in respect of maintenance. It is, strictly speaking, unnecessary to express such a firm conclusion. It is sufficient to say that the conclusion is one to which a reasonable legal adviser (internal or external), considering the Access Agreement and the Alliance Contract in 2000 or 2003, could have reached.
123 Thus, in my view, an objective circumstance by which the reasonableness (or otherwise) of the settlement between Railcorp and SRA should be assessed is the obligation that, it is at least arguable, is imposed on Railcorp by cl 6.2(a)(i) of the Access Agreement. However, that obligation is not equivalent to a warranty that the track is and will remain in a condition safe for use by SRA. Further, if the obligation is one that, at least arguably, could be satisfied by entering into an appropriate contract, such as the Alliance Contract, then an isolated act of negligent repair on the part of a subcontractor such as Alpcross would not necessarily demonstrate breach unless the obligation were non-delegable. Those, too, are matters that would bear upon the assessment of the objective reasonableness of the settlement.
Third point: implied term
124 Mr Jackman submitted that if the Access Agreement did not impose on Railcorp an express obligation to maintain the infrastructure so that it was safe for SRA's use, such a term should be implied. He relied on the objective set out in cl 2.2(a) of the Access Agreement, and submitted that the objective of safe management and control of the network supported the implication in fact of the term for which he contended.
125 Mr Donaldson submitted that no such term should be implied, having regard to the obvious care that had been taken by Railcorp and SRA to specify their respective rights and obligations under the Access Agreement. He referred also to cl 20.5, by which the parties agreed that the agreement (including its schedules and other documents referred to in those schedules) "constitutes the entire understanding and agreement between [Railcorp] and [SRA] as to the subject matter of the Agreement".
126 I do not think that such a term should be implied. If cl 6.2(a)(i) has the effect for which Railcorp contends, implication is not necessary. If cl 6.2(a)(i) does not have that effect, it nonetheless represents the parties' bargain as to the existence and content of any duty on Railcorp to maintain. On that basis, the suggested implied term would be inconsistent, in that it would impose on Railcorp an obligation different (on the hypothesis presently under consideration) from that for which the parties had stipulated. On either approach, there is no basis for implication.
Decision
127 I conclude that Railcorp has not discharged the burden of showing that the settlement sum was reasonable. I accept, having regard to the decision in Unity Insurance Brokers and the majority judgments in that case, that the focus of attention is the negotiated sum, and not the attention (or inattention) subjectively given to its amount and justification. However, it would be somewhat surprising if a settlement were held objectively to be reasonable, by reference to (for example) an analysis of the competing arguments, and a prediction of the outcome of litigation, where the party seeking to uphold the settlement had not in fact taken those matters into its consideration.
128 Once the concept of a non-delegable duty is discarded (as in my view it should be) Railcorp's position really depends on its submission that cl 6.2(a)(i) imposed on it an obligation to maintain the track in a condition safe for SRA's use. Whilst I think that it is the better view of the clause - and in any event it is a view that could not have been said, in 2000 or 2003, to be obviously wrong - it does not resolve this point. The clause, so construed, is not a warranty as to the present or future safety of the track and other infrastructure. Further, it was open to Railcorp to discharge the obligation by entering into an appropriate contract, with someone such as Fluor, for the performance of proper maintenance and repair. It is not the case that an isolated act of negligence by a subcontractor engaged by Fluor for the purposes of the Alliance Contract necessarily demonstrates a breach by Railcorp of its obligations under cl 6.2(a)(i).
129 An objective assessment of the settlement requires attention to be given to the strength of the opposing positions. It also requires attention to be given to the amount of the settlement sum, compared to the total amount of the claim. Thus, it may be reasonable for a defendant to pay a small sum by way of settlement, to achieve certainty, even though its defences, viewed objectively, are strong.
130 In this case, the amount paid by Railcorp to SRA represented substantially the whole amount claimed by SRA. To the extent that a reduction was negotiated, it was negotiated between the loss adjusters, by reference to questions of valuation of the train set. There is no basis in the evidence for concluding that Railcorp sought, or SRA allowed, some further reduction in the amount of the settlement sum to reflect the possibility that SRA might not succeed in showing that Railcorp had any liability to it, or that Railcorp might succeed in showing that it had a good defence to any such claim. Of course, there will be cases where the "defences" are so weak that no discount, or no significant discount, would be allowed by reference to the issue of liability. But this is not such a case. On the contrary, for the reasons that I have given, I think that there were serious issues as to whether in law Railcorp had any liability to SRA.
131 In those circumstances, one would expect the negotiated sum to make some allowance for the risk that SRA might lose. No such allowance was made. On the contrary, Railcorp agreed in effect to pay everything that, on a proper approach to the assessment of loss, SRA would be entitled to receive if it made good its case on liability.
132 Thus, looking objectively at the circumstances as they appeared (or should have been appreciated) in October 2003, I am not persuaded that the settlement was reasonable.
133 If the question of the reasonableness of the settlement sum is to be judged by reference to the matters that in fact were taken into account, then the answer is even clearer. There is simply no evidence that Railcorp took into account any factor that might suggest that it was appropriate to settle, let alone that whatever factors were taken into account were sufficient, objectively, regarded, to lead to the conclusion that the settlement sum negotiated was a reasonable one.
134 I accept the proposition that one would hope that two state-owned corporations should not expend their resources - in effect, public resources - in fighting out in court a dispute between them. However, there is no evidence that SRA's position was so entrenched that it would have commenced proceedings against Railcorp had the negotiations for settlement failed to produce an outcome satisfactory to it. Nor is there any evidence that Railcorp feared (whether reasonably or otherwise) that litigation would ensue if it did not settle with SRA. Ms Nicola's evidence as to her reasons for thinking the settlement justified go to the quantification of the amount agreed to be paid, not to the antecedent or underlying question of liability.
135 Further, at the time, Railcorp and SRA had both been "corporatised". Each had its own budget and management structure. Each, no doubt, had its own objectives, and its own performance criteria. Accepting that, in effect, each was owned by the State, it remains the fact that each was entitled - indeed, perhaps, obliged - to maximise its performance both operationally and commercially: subject, of course, to the overriding control of the Minister and the interests of rail commuters and users. At a very broad level of analysis, one could regard the dispute as being between different emanations of "the State". However, I am by no means sure that such a broad analysis has any bearing on an objective assessment of the reasonableness of the settlement.
136 For these reasons, I conclude that Railcorp has not discharged its onus of showing that the amount of the settlement negotiated with SRA was, regarded objectively, reasonable. Having come to that conclusion, it is unnecessary to deal with the alternative arguments put by Mr Donaldson (and adopted by Mr Robson) as to that portion of the compromised sum - a little under $115,000.00 - that relates to indemnification for amounts paid by SRA to injured passengers.
137 In case my conclusion should be overturned on appeal, I will indicate that there was no challenge to the quantification of the elements of the claim. If, contrary to what I have said, Railcorp is in principle entitled to recover what it has paid to SRA (and the workers' compensation insurer), it should have the whole amount together with interest. The assessment and verification of the actual figures was performed on a careful and rational basis.
138 The fourth issue should be answered by saying that if Railcorp is otherwise entitled to succeed in its claims against Alpcross or Fluor, the amount that it, is entitled to recover does not include the sum of $5,207,287.42 paid to SRA. It is limited to $579,558.00 together with the amounts paid in respect of a workers compensation claim (see at [30] above), a total of $590,940.00, together with interest.