Australian Energy Regulator v Snowy Hydro Limited

National Electricity Market 5 The AER is a body corporate established pursuant to s 44AE of the Competition and Consumer Act 2010 (Cth) (CCA). It is able to act as a receptacle for, inter alia, a function or power under State energy law (s 44AI). Relevantly, "a State energy law" includes "a uniform energy law that applies as a law of a State…" (s 4(1)). In particular, a "uniform energy law" includes the "South Australian Electricity legislation", which is in turn defined to mean "the National Electricity Law set out in the Schedule to the National Electricity (South Australia) Act 1996 of South Australia…and…any regulations…made under Part 4 of that Act" (s 4(1)). 6 The National Electricity (South Australia) Act 1996 (SA) (SA Act) provides that the National Electricity Law set out in the Schedule to that Act applies as a law of South Australia (s 6). The National Electricity (Victoria) Act 2005 (Vic) (Victorian Act) provides that the National Electricity Law set out in the SA Act applies as a law of Victoria, and may be referred to as the National Electricity (Victoria) Law (s 6). Thus, the National Electricity (Victoria) Law is itself a State energy law as contemplated by the CCA. The National Electricity (New South Wales) Act 1997 (NSW) (NSW Act) also provides that the National Electricity Law set out in the SA Act applies as a law of NSW, and may be referred to as the National Electricity (NSW) Law (s 6); likewise, it is a State energy law as contemplated by the CCA. For present purposes, the AER is exercising and has exercised powers and functions under the National Electricity (Victoria) Law and the National Electricity (NSW) Law; a paradigm case of "cooperative federalism" at work. For convenience I will refer to these Laws collectively as the "National Electricity Law". 7 Under the National Electricity Law, rules governing the operation of the National Electricity Market and the participants' conduct therein can be and have been made, known as the "National Electricity Rules" (see Part 7 of the National Electricity Law). 8 The National Electricity Market (NEM) is a wholesale electricity market, which electrically connects Victoria, South Australia, Queensland, New South Wales, Tasmania and the Australian Capital Territory. It operates to facilitate the exchange of electricity between generators and retailers, who resell electricity to businesses and householders. The Australian Energy Market Operator (AEMO) operates the NEM. The Australian Energy Market Commission (AEMC) makes and amends the National Electricity Rules (s 34 and Sch 1 of the National Electricity Law), except those produced at their inception to replace the National Electricity Code that were made by the relevant minister (s 90(1) of the National Electricity Law). The AER, as provided for under s 15 of the National Electricity Law, inter alia, monitors compliance by registered participants in the NEM with the National Electricity Law and the National Electricity Rules, investigates breaches of the National Electricity Law and the National Electricity Rules and institutes and conducts proceedings against registered participants for contraventions of the National Electricity Law and the National Electricity Rules. That is the capacity in which the AER acts in the present proceeding. 9 The NEM is known as an "energy-only gross pool". Mandatory participation is imposed. All wholesale electricity sales must occur through the NEM. Payments to generators are based upon energy supplied rather than generator capacity or availability. Electricity generators sell their generated electricity into the NEM. Physical supply and demand is sought to be matched "instantaneously". Generators make offers for supply, with such offers "matched" to meet demand in a cost efficient way. The "matching" is implemented by the AEMO issuing dispatch instructions to the relevant generators. As part of the NEM mechanism, the NEM determines a spot price every half-hour in each geographic region of the NEM of which there are five regions (the ACT is part of the NSW region). Financial transactions for all electricity traded are based on these spot prices, although there is a secondary derivatives market which is used to hedge against and to minimise risk in relation to spot price movements, actual or forecast. 10 To elaborate, supply and demand in the NEM is "matched" at every point in time by a centrally coordinated dispatch process. The mechanism for setting the spot price has the following key elements. Generators specify the quantities of electricity that they are willing to produce at various prices; they do so for each of the five minute dispatch intervals in a day. They make offers/bids to the AEMO prior to each trading day specifying such quantities and price bands for each of the dispatch intervals. Bids can be varied under specified circumstances and re-bidding can occur; this becomes more restricted the closer one gets to physical dispatch. The AEMO collates all such offers/bids and produces a plan setting out the proposed energy output level of each generator. The offers of generators are ranked from lowest to highest price for the particular dispatch interval. The generator with the highest price offer is treated as being to be dispatched last or not at all if the electricity supplied from the lower price offers is sufficient to meet demand. The order is known as the "merit-order"; another colloquialism is "bid stack". The AEMO determines a dispatch price for each dispatch interval (each five minutes) through a central dispatch process. This is done for each region of the NEM, including the Victorian region. The dispatch prices for the dispatch intervals are determined through the aggregation of dispatch offers/bids "matched" against the demand within the particular region, so as to represent the marginal value of supply. What that means is that the offer price for the last generator to be dispatched sets the price for all. So, for example, if the forecast demand is for, say, 1000 MW for a particular five minute interval in a particular region and you have generator bids of $50/MWh for Generator A for 500 MW, $100/MWh for Generator B for 300 MW, $150/MWh for Generator C for 200 MW and $200/MWh for Generator D for 300 MW, Generators A, B and C would be dispatched (in that order) but not Generator D. Moreover, as Generator C would be the "last" dispatched, its price would represent the marginal value of supply. Accordingly, although Generators A and B bid lower, they would, simplistically put, receive the higher price for their combined 800 MW. Generally, the wholesale spot price for a particular region paid to generators is the average dispatch price over 30 minutes (trading interval); all generators who are dispatched during this period are paid at this price regardless of their bid price. The AEMO gives dispatch instructions to the generators that accommodate the merit order and permit the NEM to operate, including instructions that may be necessary where there are regional constraints or system security or safety issues.

The Present Proceedings 11 The AER has brought the present proceedings against Snowy Hydro alleging nine contraventions of cl 4.9.8(a) of the National Electricity Rules (versions 53, 54 and 55 at the relevant times). Snowy Hydro contends that cl 4.9.8(a) is invalid. Clause 4.9.8(a) provides: (a) A Registered Participant must comply with a dispatch instruction given to it by AEMO unless to do so would, in the Registered Participant's reasonable opinion, be a hazard to public safety or materially risk damaging equipment. 12 Snowy Hydro is a registered participant in the NEM. It is an electricity generator and operates and owns hydro-electric scheduled generating units which are situated in NSW (but in the Victorian region of the NEM) and have registered generation capacities of 1,500 MW (Murray Units). The Murray Units are connected to the Murray to Dederang transmission line, allowing electricity generated by the Murray Units to be transmitted southwards for distribution and consumption within the Victorian region of the NEM. Snowy Hydro is a Scheduled Generator in relation to the Murray Units. It is also a Scheduled Generator in relation to a group of gas fired scheduled generating units, which are situated in Victoria and have registered generation capacities of 300 MW (Valley Power Units). The electricity generated from the Valley Power Units is supplied into the Victorian region of the NEM. 13 The AER alleges that on nine separate occasions Snowy Hydro contravened cl 4.9.8(a). Paragraph 1 of the originating application summarises the breaches in the following terms: …the Respondent is in breach of clause 4.9.8(a) of the Rules by failing to comply with a dispatch instruction given to it by the Australian Energy Market Operator (AEMO) when it did not hold the reasonable opinion that to comply with the dispatch instruction would be a hazard to public safety or materially risk damaging equipment in respect of each of the following dispatch intervals (the Contraventions): Generating Unit Date Dispatch interval Murray 29 November 2012 15:50:00 Murray 29 November 2012 16:15:00 Murray 21 June 2013 09:45:00 Valley Power 18 February 2013 16:35:00 Valley Power 18 February 2013 16:40:00 Valley Power 18 February 2013 16:45:00 Valley Power 18 February 2013 17:05:00 Valley Power 18 February 2013 17:10:00 Valley Power 18 February 2013 17:15:00

14 A typical example of the allegations made in relation to a particular contravention can be gleaned from the statement of claim for the first contravention (omitting particulars) which alleges: Contravention 1 - 29 November 2012 - dispatch interval ending 15:50:00 18. At 15:07 and 15:12 on 29 November 2012, in expectation of extremely high dispatch prices and an extremely high spot price for TI (trading interval) 16:00, and with the intention of ensuring that all of the available capacity of the Murray Units would be dispatched during TI 16:00, the Respondent submitted two rebids of the available capacity of the Murray Units for TI 16:00, the result of which was that the whole of the available capacity was reallocated to the $53 price band. 19. In each of DI (dispatch interval) 15:35, DI 15:40 and DI 15:45, the Respondent received dispatch instructions providing a target output of 1,348.00 MW for the Murray Units. 20. At 15:45:00, the output of the Murray Units was approximately 1,342.44 MW. 21. At 15:45:04: 21.1 AEMO determined and published a dispatch price of $12,497.76 for DI 15:50; and 21.2 AEMO gave, and the Respondent received by around 15:45:30, a dispatch instruction for DI 15:50, to reduce the output of the Murray Units to 1,164.74 MW by 15:50:00 (the dispatch instruction for DI 15:50). 22. The reduction in target output given in the dispatch instruction for DI 15:50, relative to the target output of 1,348 MW received for the preceding dispatch intervals, occurred because a network constraint was binding to avoid overload on the Murray to Dederang transmission line. 23. The Respondent knew that the reduction in target output had occurred because a network constraint was binding to avoid overload on the Murray to Dederang transmission line. 24. At or around 15:45, the Respondent: 24.1 knew that an extremely high dispatch price had been determined for DI 15:50; and 24.2 expected that an extremely high spot price would be determined for TI 16:00. … 26. At 15:47:22, the Respondent submitted a rebid (the 15:47 ramp rate rebid) to be effective from DI 15:55, by which it reduced the down ramp rate for the Murray Units from 50 MW per minute to the minimum permitted ramp rate of 3 MW per minute. 27. Between 15:45 and around 15:48, the output of the Murray Units was initially reduced from 1,342.44 MW to approximately 1,294 MW. … 29. The output from the Murray Units at 15:50:00 was 1,328.47 MW, namely 163.73 MW in excess of the level of output specified in the dispatch instruction for DI 15:50. 30. At no time between 15:45:00 and 15:50:00 did the output of the Murray Units reduce to 1,164.74 MW. 31. By reason of the matters alleged in paragraphs 21.2 and 29 above, alternatively paragraphs 21.2, 29 and 30 above, the Respondent failed to comply with the dispatch instruction for DI 15:50. 32. The Respondent did not hold the reasonable opinion that compliance with the dispatch instruction for DI 15:50 would either be a hazard to public safety, or would materially risk damaging equipment at any time during DI 15:50. 33. By reason of the matters alleged in paragraphs 31 and 32 above, the Respondent contravened clause 4.9.8(a) (Contravention 1). 34. Further, the Respondent deliberately maintained its generation output from the Murray Units above the level of output specified in the dispatch instruction for DI 15:50: 34.1 with the intention of benefiting from maintaining high levels of output during DI 15:50 and the very high spot price expected for TI 16:00; and 34.2 despite knowing that the Murray Units had received a reduced target output for DI 15:50 because a network constraint was binding to avoid overload on the Murray to Dederang transmission line. … 38. By reason of: 38.1 its non-compliance with the dispatch instruction for DI 15:50 referred to in paragraph 31 above; and 38.2 the effect of its non-compliance with the dispatch instruction for DI 15:50 combined with the 15:47 ramp rate rebid, the Respondent: 38.3 generated a greater quantity of energy from the Murray Units during DI 15:50, DI 15:55 and DI 16:00 than it would have generated; and 38.4 earned a greater trading amount from the Murray Units for TI 16:00 than it would have earned, had it complied with the dispatch instruction for DI 15:50. 39. By reason of the effect of the Respondent's non-compliance with the dispatch instruction for DI 15:50 referred to in paragraph 31 above combined with the 15:47 ramp rate rebid: 39.1 a lower quantity of electricity was able to be imported from the New South Wales region into the Victorian region of the NEM during DI 15:55 and DI 16:00 than would have been imported had the Respondent complied with the dispatch instruction for DI 15:50; 39.2 generating units in the New South Wales and Queensland regions of the NEM were dispatched at lower levels during DI 15:55 and DI 16:00 than they would have been had the Respondent complied with the dispatch instruction for DI 15:50; 39.3 the dispatch price for DI 15:55 was higher than it would have been had the Respondent complied with the dispatch instruction for DI 15:50; and 39.4 the spot price for TI 16:00 was higher than it would have been had the Respondent complied with the dispatch instruction for DI 15:50. 15 As for the defence to such allegations, many of the objective framework facts are not in dispute. But pleas of knowledge, expectation or other states of mind of Snowy Hydro are denied. 16 In terms of Snowy Hydro's defence to the alleged first contravention, it is pleaded: 33. The respondent denies paragraph 33. In further answer to paragraphs 18 to 33 of the statement of claim, the respondent relies upon the following matters: (a) the respondent's failure to meet the output target in the dispatch instruction for DI 15:50 occurred in the following circumstances: (i) in each of DI 15:35, DI 15:40 and DI 15:45, Mr Whish had targeted output of 1,348.00 MW, in accordance with the dispatch instructions received for each of those dispatch intervals; (ii) Mr Whish received a target output for DI 15:50 of 1,164.74MW; (iii) although Mr Bradburn knew that there was a network constraint binding for DI 15:50, Mr Bradburn believed that this constraint was only transitory, and thus was unlikely to constrain the output of the Murray units in DI 15:55; (iv) furthermore, Mr Bradburn understood that the capacity of the Murray units had been, or would be, bid into the market in price band 1 (minus $1,000), that being the appropriate bid to make in the circumstances, thus making it very likely that the respondent would receive an instruction to dispatch for DI 15:55 1,348.00 MW, that is, all of its available capacity; (v) as a result of (iii) and (iv) above, at approximately 15:46, Mr Bradburn telephoned Mr Whish and informed him that the target output for DI 15:55 was likely to be close to 1,348.00 MW; (vi) having been informed by Mr Bradburn of the likely target output for DI 15:55, Mr Whish concluded that it would potentially jeopardise the safe operation of the plant if he tried to reduce the output of the unit from 1348.00 MW to 1,164.74 MW in one interval, and then increase it back to 1348.00 MW in the next, in order to comply with output targets in successive dispatch instructions. In particular, Mr Whish was concerned that a rapid decrease and increase of this sort would create a risk of 'air entrainment', which had the potential to (a) cause damage to the Murray Units including catastrophic destruction to the Murray Units and surrounding infrastructure, and (b) imperil the personal safety and even the lives of personnel and the public in the vicinity of the Murray Units; (vii) accordingly, Mr Whish decided to maintain the output of the Murray units during DI 15:50 at approximately 1,348.00 MW. (b) the respondent was not required to comply with the dispatch instruction for Dl 15:50 because: (i) for the reasons pleaded in the respondent's cross claim, clause 4.9.8(a) is invalid; (ii) alternatively, (if clause 4.9.8(a) is valid), in the circumstances pleaded in paragraph (a) above, the respondent held a reasonable belief that compliance with the dispatch instruction for Dl 15:50 would have been a hazard to public safety; (iii) further or alternatively, in the circumstances pleaded in paragraph (a) above, the respondent held a reasonable belief that compliance with the dispatch instruction for Dl 15:50 materially risked damaging equipment; (iv) further or alternatively, in the circumstances pleaded in paragraph (a) above, the respondent took reasonable steps to comply with the dispatch instruction for Dl 15:50; (v) further or alternatively, in the circumstances pleaded in paragraph (a) above, compliance with the dispatch instruction for Dl 15:50 was impracticable; (vi) further or alternatively, in the circumstances pleaded in paragraph (a) above, the respondent's failure to meet the output target in the dispatch instruction for Dl 15:50 was the result of a reasonable assumption, or alternatively reasonable mistake, being that the constraint would be temporary and the full output of the Murray Units would be requested to be dispatched in Dl 15.55. 17 For present purposes, although nine contraventions are pleaded, the allegations and defences are in similar form. It is not necessary to refer to the AER's reply for present purposes. 18 Snowy Hydro's cross-claim pleads various matters in support of its invalidity contention. The plea of invalidity of cl 4.9.8(a) (at [46] et seq) is based upon many factual allegations that are said to lead to the conclusion that cl 4.9.8(a) is unreasonable, lacks proportionality and is uncertain. 19 First, "(e)xact compliance with dispatch instructions in every dispatch interval is a physical impossibility" (at [17]) (the impossibility scenario). Various factual matters are then listed (at [18]). 20 Second, most of the scheduled generating units in Australia operate by burning fossil fuel (at [19]), that there is variability as to the precise MW quantity of electricity produced by such a unit at any point in time due to various factual matters (at [20]), and that accordingly it is difficult for an operator to control the precise MW quantity of electricity produced (at [21]). There were also variability factors for hydro-electric generating plant that gave rise to similar problems (at [22]-[24]). 21 Third, because of the use of governor control systems that automatically adjust the output of a unit in response to frequency fluctuations, that a Scheduled Generator's output necessarily deviates from the target specified in a dispatch instruction (at [25]-[28]). 22 Fourth, the failure of scheduled generating units, and associated control systems, a common occurrence apparently, results in such units not complying with dispatch instructions (at [29]). 23 Fifth, because the metering equipment for scheduled generating units has an error accuracy tolerance of between 0.5% and 3%, this may lead to non-compliance with a dispatch instruction (at [30]). 24 Sixth, because dispatch instructions may change the required output from one five minute dispatch interval to the next, that when moving from one level of output to another, it is often impossible to reach the precise output level in the short time available (at [31]-[32]). 25 Seventh, because the AEMO may at any time issue a dispatch instruction to provide frequency control ancillary services (to help maintain frequency within the power system), it may then be simultaneously impossible to implement exactly a dispatch instruction (at [33]-[35]). 26 It is then said that "(m)ost scheduled generators fail to comply with dispatch instructions every day that they are generating" (at [36]). But as I put to Mr Cameron Moore SC, senior counsel for Snowy Hydro, this de facto position is only a jury point in the absence of a forensic causal analysis regarding each specific non-compliance. 27 It is then said that the design and operation of the NEM contemplates non-compliance (at [37]-[45]). First, the system of the use and procurement of frequency control ancillary services implies that the National Electricity Rules contemplate discrepancies between dispatch instructions and the electricity generated by generators; discrepancies are managed using such services (at [38]-[40]). Second, because of the power to declare units as non-conforming, with the AEMO permitted to publish non-conformance thresholds, that such power suggests that strict conformance with dispatch instructions is not necessary (at [41]-[42]). Third, because of the AEMO's power to give directions generally "if AEMO is satisfied that it is necessary to maintain system security" that this power is somehow inconsistent with strict compliance because such directions could override (at [43]-[45]). 28 Snowy Hydro then draws together these factual threads and pleads: Invalidity of clause 4.9.8(a) 46. By reason of the matters pleaded in paragraphs 17 to 34, all Scheduled Generators will: (a) regularly fail to achieve the particular output specified in their dispatch instructions; and (b) as a result, if clause 4.9.8(a) applies in accordance with its stated terms, regularly contravene a civil penalty provision in their ordinary day-to-day operations. 47. By reason of the matters pleaded in paragraphs 17 to 45, clause 4.9.8(a) is invalid on one or more of the following grounds: (a) unreasonableness; and/or (b) lack of reasonable proportionality to the subject matter the grant of the rule­ making power under the National Electricity Law. 48. In the alternative, if clause 4.9.8(a) does not require a registered participant to achieve the exact output specified in a dispatch instruction, it fails to communicate the obligation actually imposed by clause 4.9.8(a), and is accordingly invalid on the ground that it is uncertain. 49. As clause 4.9.8(a) is invalid, it is not picked up and applied as part of the law of Victoria or of New South Wales pursuant to the National Electricity (Victoria) Act 2005 or the National Electricity (New South Wales) Act 1997. Alternative construction of clause 4.9.8(a) 50. In the alternative, if clause 4.9.8(a) of the Rules is valid, on its proper construction it does not require strict compliance with a dispatch instruction or compliance in all circumstances, including but not limited to: (a) circumstances in which the Registered Participant has taken reasonable steps to comply; (b) further or alternatively, circumstances in which compliance is not practicable; and (c) further or alternatively, circumstances of inadvertence or misunderstanding; Relief claimed 51. The Cross-Claimant seeks: (a) A declaration that clause 4.9.8(a) of the National Electricity Rules is invalid, on the basis that the rule is not authorised by the rule making power in s 90 of the National Electricity Law… 29 In substance, it is said that cl 4.9.8(a) is invalid for unreasonableness. Second, it is said that it is invalid due to a lack of reasonable proportionality. Third, it is said that it is invalid for uncertainty. Now just stopping there, it is apparent that detailed evidence addressing how the NEM works in practice and addressing each of the matters referred to in [19]-[27] above will need to be led and assessed in order to address these conclusions. A significant number of these factual allegations are in contest - see [17]-[45A] of the AER's defence to cross-claim. As is summarised in the affidavit of Leanne Hanna, solicitor for the AER, affirmed on 9 September 2014, responding to the affidavit of Justin Oliver, solicitor for Snowy Hydro, sworn on 25 August 2014, which I have also considered, contentious factual issues will need to be explored concerning: whether the technical matters listed at [20] and [23] of the cross-claim make it impossible for the operator of a fossil-fuel burning generating unit or a hydro-electric generating unit to comply with a dispatch instruction ([18(a)] of the cross-claim which is denied, at least in part); the implications of the operation of a governor control system attached to a generating unit ([28] of the cross-claim which is denied, at least in part); the regularity with which scheduled generating units or their associated control systems in the NEM fail, resulting in those generating units being unable to comply with a dispatch instruction ([29] of the cross-claim which is denied, at least in part); the significance of the error accuracy tolerance of metering equipment ([30] of the cross-claim which is denied, at least in part); whether it is impossible to reach the precise output level specified in the dispatch instruction within the time period between the giving of the dispatch instruction and the end of the dispatch interval ([32] of the cross-claim which is denied, at least in part); the implications of a Scheduled Generator receiving a dispatch instruction in relation to frequency control ancillary services ([34] of the cross-claim which is denied, at least in part); whether the National Electricity Rules contemplate regular non-compliance with dispatch instructions ([37] of the cross-claim which is denied, at least in part); and whether cl 4.9.8(a) is a disproportionate exercise of the rule-making power under s 90(1) of the National Electricity Law ([47(b)] of the cross-claim which is denied); to this may be added reference to s 34. 30 In the alternative to its invalidity arguments, Snowy Hydro asserts that cl 4.9.8(a), if valid, on its proper construction does not require strict compliance with a dispatch instruction where a registered participant has taken reasonable steps to comply, compliance is not practicable or where there is inadvertence or a misunderstanding. But these alternative arguments, if directly or indirectly dealt with as part of any separate determination, move into the offending territory of potentially hypothetical questions and interlocutory declarations which I will explain later.

Validity of Clause 4.9.8(a) - Practical Operation 31 In terms of determining the "unreasonableness" of cl 4.9.8(a) such as to invalidate it, a high threshold is applied. The criterion does not "invite judicial merits review of delegated legislation" (A-G (SA) v Adelaide City Corporation (2013) 249 CLR 1 at [48] per French CJ). Rather, one is concerned with the contemplated ambit of power. Delegated legislation that has no rational connection to the statutory purpose or is on its face capricious or oppressive may be found to be unreasonable. Further "(t)o determine whether [the rule] is an exercise of a power, it is not always enough to ascertain the subject matter of the power and consider whether the [rule] appears on its face to relate to that subject"(Williams v Melbourne Corporation (1933) 49 CLR 142 at 155 per Dixon J). One needs to examine the legal and practical operation of the rule, in this case cl 4.9.8(a), to determine its character and connection with the relevant rule making power and the object thereof (A-G (SA) v Adelaide City Corporation (supra) at [117]-[123] per Hayne J). 32 Necessarily then, significant factual issues will need to be addressed, as both the AER and Snowy Hydro accepted would be the case. Snowy Hydro's argument was more that the factual issues on this invalidity aspect would involve different factual issues to those specifically addressing the alleged nine contraventions. The AER accepted this to some extent, but not completely, particularly in light of one of Snowy Hydro's alternative construction arguments addressing a construction of cl 4.9.8(a) to the effect that compliance was not necessary where compliance was not practicable (at [50(b)] of the statement of cross-claim); this overlapped with defences to seven of the nine contraventions; in considering issues of invalidity, issues of construction were necessarily interwoven. And more broadly, there is a potential conceptual overlap between the concept of impossibility, one basis for the invalidity contention, and impracticability. 33 Relatedly, in terms of determining whether cl 4.9.8(a) lacks proportionality, if "it could not reasonably have been adopted as a means of attaining the ends of the power" then it may be invalid (Williams v Melbourne Corporation (supra) at 155 per Dixon J). Similarly expressed, if cl 4.9.8(a) is not "capable of being considered to be reasonably proportionate to the pursuit of the enabling purpose", then it may be invalid (South Australia v Tanner (1989) 166 CLR 161 at 165 per Wilson, Dawson, Toohey and Gaudron JJ). Again, the Court must consider the legal and practical effect of cl 4.9.8(a) and must make its "own assessment of the directness and substantiality of the connexion between the likely operation of the [rule] and the statutory object to be served", in order to determine whether the substantiality of the connection "is so exiguous that [cl 4.9.8(a)] could not reasonably have been adopted as a means of fulfilling the statutory object" (at 179 per Brennan J, dissenting in the result but not the principle). 34 Again, both parties accepted that evidence would be required to address such matters. The AER also contended, which I accept, that evidence on the cross-claim going to the proportionality of cl 4.9.8(a) may involve factually investigating the economic efficiency and system security objectives that are advanced by strict compliance with that clause. But such evidence may also overlap with evidence that may be adduced on the penalty phase. Again, there is not a definitive separation. 35 Generally, Snowy Hydro seeks to establish various factual matters said to support its invalidity argument. It is said that these are separate to the facts and evidence relevant to the contraventions. But various points. First, these are not said to be exhaustive factual categories supporting invalidity. Second, in terms of assessing these points, I am likely to be considerably assisted by how such technical issues are explained with reference to Snowy Hydro's units. But detailed evidence concerning Snowy Hydro's units is only likely to be called when dealing with the alleged contraventions themselves. True it is that the alleged contraventions and the defences thereto do not raise or point to these technical issues specifically, but in enquiring on these more general topics there is advantage to having them specifically contextualised to the Snowy Hydro units. Further, Snowy Hydro is likely to adduce expert engineering and operational evidence to address the factual allegations going to invalidity and causal connections necessary to establish the impossibility scenario. But the same expert witnesses called to address such topics in the invalidity context are likely to be called when dealing with the Snowy Hydro units in any event in relation to the specific contraventions. There is no precise demarcation as to both the witnesses and some of the concepts involved between the invalidity case on the one hand and the contravention case on the other hand.

Separate Determinations- Relevant Principles 36 The default position is that in a proceeding, all issues of fact and law should be determined at the one time following a trial. 37 A further default position is that the hearing and determination of a cross-claim should take place at the same time as the principal proceeding (r 15.10(1)(c) of the Federal Court Rules 2011 (Cth) (FCR), but cf r 15.13 of the FCR). Some authorities suggest that it is only in an unusual case that a cross-claim should be heard and determined separately (for example Barclays Bank v Tom [1923] 1 KB 221 and Shrimp v Landmark Operations Ltd (2007) 163 FCR 510). Generally, any approach must be informed by the broader and more flexible elements of s 37M and s 37P of the FCA. Moreover, cases dealing with the separate determination of cross-claims involving contribution or indemnity claims against concurrent wrongdoers, insurers and the like often involve different considerations to the cross-claim in the present case, which is in essence a formalisation of a point also taken as a defence. The present context is more comparable to the context of dealing with a separate question under r 30.01. Accordingly, the principles that guide the exercise of any discretionary power under r 30.01 also guide the exercise of any discretionary power under r 15.13 in the present case. 38 Rule 15.13 is expressed broadly. Likewise r 30.01 gives the Court a broad power to make an order for the determination of a separate question in a proceeding. Sometimes this has been described as a power to be exercised where it is "just and convenient" to do so, although this is not the precise language of r 30.01. 39 An important, if not overarching, consideration informing the exercise of any power under r 15.13 or r 30.01 is the need to facilitate a just resolution involving, inter alia, efficient case management (s 37M of the FCA). That is, will this proceeding be managed more efficiently if there is a separate determination of the invalidity question or cross-claim? 40 An issue may not be appropriate for separate and preliminary determination where: (a) It is one of two or more alternative ways in which a case is framed and determination of the separate issue would still leave significant other issues unresolved; (b) It gives rise to significant contested factual issues both at the time of the hearing of the separate determination and at trial; (c) It results in significant overlap between the evidence adduced on the hearing of the separate determination and at trial (GMB Research & Development Pty Ltd v Commonwealth of Australia [1997] FCA 934 (GMB) and Reading Australia Pty Ltd v Australia Mutual Provident Society (1999) 217 ALR 495 (Reading Australia) at [8]); (d) Related to point (c), it results in the calling of the same witnesses at both stages of the hearing (GMB and Reading Australia). Such a factor may be significant if not exacerbated if credit issues are involved in relation to lay witnesses or reliability issues are involved in relation to expert witnesses, although the latter problem has less significance; (e) It would or is likely to prolong rather than shorten the litigation. 41 Generally, the attractions of trials of issues rather than of cases in their totality, are "often more chimerical than real". Single issue trials should "only be embarked upon when their utility, economy and fairness to the parties are beyond question" (Tepko Pty Ltd v Water Board (2001) 206 CLR 1 at [168]-[170] per Kirby and Callinan JJ). That could hardly be said in the present case. 42 Further, overarching these considerations is the nature of the question or issue, and whether it is hypothetical. Its factual underpinning is highly significant in determining the appropriateness of any separation. 43 A question can be the subject of an order for separate determination even though it will not finally determine any of the parties' rights, but it must be based on concrete and established or agreed facts. 44 Advisory opinions are to be avoided. However apparently helpful an advisory opinion may be to a party "(t)he law is not judicially administered by judicial declarations of its content 'divorced from any attempt to administer [the applicable] law'" (North Ganalanja Aboriginal Corporation v Queensland (1996) 185 CLR 595 at 612 per Brennan CJ, Dawson, Toohey, Gaudron and Gummow JJ). 45 A declaration should not be made or an answer given which amounts to no more than a declaration that the law dictates a particular result when certain facts in the material or pleadings are established. The facts must be stated and identified with precision. And the facts should not be left open. They should be established or agreed. The efficient administration of courts "is incompatible with answering hypothetical questions which frequently require considerable time and cause considerable expense to the parties, expense which may eventually be seen to be unnecessarily incurred" (Bass v Permanent Trustee Co Ltd (1999) 198 CLR 334 at [49] per Gleeson CJ Gaudron, McHugh, Gummow, Hayne and Callinan JJ). Further, it is "no part of judicial power to effect a determination of rights by applying the law to facts which are neither agreed nor determined by reference to the evidence in the case" (at [56]); the demurrer procedure can be put to one side for present purposes. 46 Further, great caution needs to be exercised in formulating a separate question for determination on the parties' agreement to proceed on the basis of assumed facts. Assumed facts may be incomplete or insufficiently precise and "parties may also have different views concerning the effect or duration of the assumptions" such that "they may consider that they can depart from the assumptions and reagitate the facts in another part of the case" (AWB Ltd v Cole (No 2) (2006) 233 ALR 453 at [35] per Young J). 47 Finally, providing that such guiding principles are not infringed, an issue may be appropriate for separate and preliminary determination if it contributes to the saving of time and cost by substantially narrowing the issues for trial, leads to disposal of the action or contributes to the settlement of the proceeding (Reading Australia).

Should there be a separate determination? 48 In my view there should not be a separate determination of the cross-claim or the invalidity question arising on the defence. The matters that point against such a separate determination are common to the context of both an order for separation under r 15.13 or under r 30.01. 49 First, on any view, the separate question or determination of the cross-claim will involve both legal and factual questions that both parties accept are likely to take a hearing time of three days. Contrastingly, the parties accept that a trial on all issues, other than the penalty phase, is likely to take a hearing time of ten days. Although this may on balance indicate a utility in determining the cross-claim or invalidity issue separately, the utility is not overwhelming. A substantial trial will still be necessary on the separate question or cross-claim. 50 Second, it would seem that witnesses giving expert evidence on factual issues relating to the cross-claim and invalidity issues will also be witnesses who will need to be called to give expert evidence on factual issues relating to the nine contraventions. It is undesirable to split the evidence of witnesses. This is particularly so with lay witnesses where credit findings may be made in relation to such witnesses on the trial of the separate question, who are then proposed to be called later. The problem is not so acute with expert witnesses where reliability issues rather than credit issues are likely to be in play. But such a split in the evidence of expert witnesses is still undesirable. 51 Third, a determination on the cross-claim is unlikely to dispose of the proceeding. Theoretically, if cl 4.9.8(a) was held to be invalid, then that would effectively dispose of the proceeding. But in the face of s 42(2) of Sch 2 to the National Electricity Law, there is at least a real prospect that the clause will be read down or given a construction that preserves validity. In such a situation then, the separate determination would not dispose of the proceeding. Snowy Hydro contended that in such a situation the Court would have ruled upon or given its preferred construction and that this would be of some utility to the parties. It might assist them to narrow the issues and the evidence led in the second stage trial. Alternatively, it might assist them to settle the case. But there are several difficulties with this approach. If one was to give a binding ruling on construction, then that is perilously close to a form of interlocutory declaration. Alternatively, if it was not binding, then both I and the parties would be free to depart from such a construction after I had heard all the evidence involved in the second stage. But if that were to be the case, the separate determination would have been unhelpful. Moreover, if the AER was not content with that construction, and assuming it perceived it to be binding, then it might proceed to launch an application for leave to appeal. There might be a consequent fragmentation of the proceeding. Mr Peter Gray QC, senior counsel for the AER, stated that the AER would not be prepared, if I proceeded with the separate determination, to give an undertaking not to challenge such an interlocutory determination at the time it was made on such construction questions, assuming I was to rule against invalidity. More helpfully, Snowy Hydro was prepared to give an undertaking not to challenge any interlocutory determination at the time it was made if I ruled against invalidity, so as to avoid the fragmentation problem. Of course, if I ruled in favour of the invalidity sought by Snowy Hydro, the proceeding would be over and the AER would have its ordinary appeal rights. 52 I should elaborate on my reasons for the conclusion that in the face of s 42(2) of Sch 2, a broad declaration of invalidity is likely to be problematic. Subsections 42(1) and (2) provide as follows: 42 - National Electricity Rules to be construed so as not to exceed the legislative power of the Legislature of this jurisdiction or the powers conferred by this Law (1) The National Electricity Rules are to be construed as operating to the full extent of, but so as not to exceed, the legislative power of the Legislature of this jurisdiction or the power conferred by this Law under which they are made. (2) If a provision of the National Electricity Rules, or the application of a provision of the National Electricity Rules to a person, subject matter or circumstance, would, but for this clause, be construed as being in excess of the legislative power of the Legislature of this jurisdiction or the power conferred by this Law under which it is made - (a) it is a valid provision to the extent to which it is not in excess of the power; and (b) the remainder of the National Electricity Rules, and the application of the provision to other persons, subject matters or circumstances, is not affected. Paragraph 42(2)(a) suggests that a construction of cl 4.9.8(a), read down to avoid the impossibility scenario, read down so as to permit of circumstances where reasonable steps were taken to comply or read down where compliance was not practical or non-compliance was due to circumstances of inadvertence or misunderstanding, is a realistic potential construction. Further, para (b) also suggests that even if such a result might theoretically apply at a general level, it might not apply to Snowy Hydro if Snowy Hydro was not in the position where the impossibility scenario or these other carve outs factually applied to it. Further, on any view, the breadth of s 42(2) would suggest that a decontextualised and broad declaration of invalidity of the type sought by Snowy Hydro may be inapposite. Further, there are also potential issues that might arise in relation to dealing with Snowy Hydro's conduct prior to any such declaration of invalidity, if one were made, by reason of the operation of s 43 of Sch 2 (although I have not heard any argument on this), which provides as follows: 43 - Invalid Rules (1) If the Court orders (by declaration or otherwise) that a Rule is invalid, the order of the Court does not - (a) revive anything not in force or existing at the time of the order of the Court; or (b) affect the previous operation of the Rule or anything suffered, done or begun under the Rule; or (c) affect a right, privilege or liability acquired, accrued or incurred under the Rule; or (d) affect a penalty arising because of a breach of the Rule; or (e) affect an investigation, proceeding or remedy in relation to such a right, privilege, liability or penalty. (2) A penalty may be imposed and enforced, and any such investigation, proceeding or remedy may be begun, continued or enforced as if the Rule had not been ordered by the Court as invalid. 53 Fourth and relatedly, there is an argument that if any declaration of invalidity was based upon factual circumstances not presented in the main proceedings in dealing with the facts applying to the nine contraventions, then it may be hypothetical and advisory. Now Snowy Hydro contends for invalidity based on a different set of general factual circumstances. And it says that if these are made good, then it is entitled to such a declaration of invalidity and it does not matter that such general circumstances may not appertain to or explain Snowy Hydro's conduct in relation to the nine contraventions. And it contends that such a declaration would not be hypothetical; it says that in its day to day operations, Snowy Hydro is subject to cl 4.9.8(a) and it has a real interest in determining its validity. But if the AER's contentions are correct in relation to the ambit and scope of s 42(2), then such a declaration may be hypothetical and may not avoid liability. If invalidity is based upon the impossibility scenario, and yet that impossibility scenario did not factually apply to justify Snowy Hydro's conduct in relation to the nine contraventions, then Snowy Hydro arguably may not under s 42(2)(b) take the benefit of such a declaration. That is, only registered participants who were faced with the impossibility scenario would take the benefit of a declaration of invalidity based upon the impossibility scenario. Such a declaration might be entirely hypothetical and advisory only in respect of the position of Snowy Hydro. One could not presently tell, and could only do so after considering the facts relating to the nine contraventions. I am not in a position to presently say whether the AER's argument will ultimately be sustained, but it is at present reasonably arguable. What the AER submits is that to avoid the possibility of such a hypothetical scenario from arising, the facts need to be found on the contravening conduct, and it then needs to be assessed as to whether these facts fit into a "subject matter or circumstance" (the language of s 42(2)) that would then produce invalidity of cl 4.9.8(a) in its application to Snowy Hydro. 54 Fifth and relatedly, hypothetical questions also arise in the event that rather than declaring cl 4.9.8(a) to be invalid, one was to read it down and give it one of Snowy Hydro's alternative constructions. The various alternative constructions may have hypothetical operation only in the sense that the "carve outs" that might be found of a registered participant taking reasonable steps to comply, or compliance not being reasonably practicable or the like, might not apply to Snowy Hydro's conduct in any event that might be found in relation to the nine alleged contraventions. Now Snowy Hydro contends that there would be utility in advance in "determining" the construction issue even if invalidity was not found. The problems associated with this have been adverted to above (see [51] above). But more generally, it is more appropriate to find all the facts first and then apply the present or read down form of cl 4.9.8(a) to such facts. Moreover, in terms of any question of invalidity of cl 4.9.8(a) as it applied to Snowy Hydro's conduct relating to the nine contraventions, I would be more appropriately informed after a full trial to tailor any invalidity declaration to Snowy Hydro's circumstances rather than to proceed now in a decontextualised fashion in the constrained framework of a separate determination, and in a fashion where true effect or emphasis to s 42(2) of Sch 2 could not be given. 55 Sixth, Snowy Hydro asserts that the perceived problems and risks that may arise as discussed in [51]-[54], if the separate determination proceeded, may not arise, and that I should proceed with a "wait and see" approach. This has some superficial allure. Nevertheless, as the risks are real, it is preferable to proceed in a manner that avoids such risks, particularly as the separate determination on any view is likely to be a substantial legal and forensic exercise of three days in any event. If the separate determination had involved a straight forward one day legal exercise as compared with a complex forensic trial, then I might have taken a different view. But as the separate determination will involve a substantial forensic contest between the parties, it is preferable to now proceed expeditiously to a trial on all issues, save as to penalty. 56 In summary, there are more risks than advantages in proceeding with a separate determination. It is unlikely to contribute to the saving of time and cost. It is unlikely to substantially narrow the issues or lead to disposal of the action given that, in light of a preliminary consideration of s 42(2) of Sch 2, a decontextualised declaration of invalidity is unlikely to be made - that is, decontextualised from the position and circumstances of Snowy Hydro. And even if such a decontextualised declaration was in prospect, this would only be after a three day trial involving witnesses who may also be called on the second stage if that was proceeded with. If a declaration was not made, the same expert witnesses who gave evidence on the separate determination would need to be recalled. Further, if a declaration was not made, there are risks of fragmentation if, for example, the AER sought to challenge any position on the construction of cl 4.9.8(a) that had been expressed as part of refusing the declaration of invalidity. Moreover, as to whether any views that were expressed on construction might assist to settle the proceeding, in the scenario where a broad declaration of invalidity was refused, that is possible but unlikely. As I say, if the AER's present position is maintained, it may seek to challenge such a view, alternatively assert that it was non-binding at that stage or assert that it should have diminished force because it was decontextualised from the actual conduct of Snowy Hydro in relation to the nine contraventions. For these and earlier reasons, and when considered in the overall context of the mandate under s 37M(3) of the FCA, Snowy Hydro's application should be refused. I certify that the preceding fifty-six (56) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Beach. Associate: Dated: 19 September 2014