Ground one: whether the restraint deed was void for uncertainty
7 The primary judge dealt with the argument as to uncertainty at [38]-[67] of his reasons. The primary judge noted that (as on appeal) Mr Hanna did not argue that the clause was uncertain because it did not have a clear English meaning. Rather, he dealt with the argument (repeated on appeal) that, as a so-called "cascading" restraint clause, it was uncertain because the provision contained no mechanism for the selection of which clause was to operate. Thus, it was said, the contracting party, could not know, from the terms of the contract, what the operative obligation was.
8 After a review of the authorities (J.Q.A.T Pty Limited v Storm [1987] 2 Qd R 162; Austra Tanks Pty Limited v Running [1982] 2 NSWLR 840; Lloyd's Ships Holdings Pty Limited v Davros Pty Limited [1987] FCA 70; 17 FCR 505; Brendon Pty Limited v Russell (1994) 11 WAR 280; Schindler Lifts Australia Pty Limited v Debelak [1989] FCA 311; 89 ALR 275; Tyser Reinsurance Brokers Pty Limited v Cooper [1998] NSWSC 689; Hydron Pty Limited v Harous [2005] SASC 176; Sear v Invocare Australia Pty Limited [2007] WASC 30; (2007) ATPR 42-149; Extraman (NT) Pty Limited v Blenkinship [2008] NTSC 31; 23 NTLR 77; Run Corp Ltd v McGrath Ltd [2007] FCA 1669; (2007) ATPR 42-198; and Northern Tablelands Insurance Brokers Pty Limited v Howell [2009] NSWSC 426; 184 1R 307) the primary judge concluded that the restraint deed was not uncertain.
9 It is unnecessary to recite how his Honour dealt with the various judicial opinions in these cases. It suffices to dispose of this aspect of the appeal to say that I agree with the primary judge's conclusion that the restraint deed was not void for uncertainty and that I generally agree with his Honour's reasons for that conclusion, and to set out briefly my reasons in respect of the arguments put on appeal (which reflected the arguments put to the primary judge).
10 Two principal reasons were propounded for the asserted conclusion as to uncertainty. First, on its proper construction the restraint deed was said to contain a single covenant which contained mutually inconsistent obligations. Secondly, even if the restraint deed contained more than one covenant it was uncertain because it made no provision or mechanism to determine which one or more of the several restraints applied and in what order.
11 The first argument relied heavily on the existence of the definite article ("the") before each of the phrases "Restraint Period" and "Restraint Area". It ignored or failed to give adequate weight, however, to the terms and effect of cl 4. Clause 4 made clear that the various periods and areas in cl 2 were part of separate and independent provisions. Thus there were nine restraints, from the widest (15 months in Australia) to the narrowest (12 months, in Mr Hanna's case, in the metropolitan area of Sydney). All were binding. Taken as individual covenants, all capable of being understood by the use of clear words and all being capable of being complied with without breaching any of the others, the one covenant argument must fail.
12 The second argument has implicit in it a proposition that there is a legal requirement for a hierarchy of the clauses and a mechanism for their order of operation. Reduced to its essential element, there was said to be uncertainty in more than one clause covering by different terms the same ground of a party's obligation. I cannot agree with the width of these propositions. It may be that if multiple obligations on the same subject matter so conflict that a contracting party cannot know what it is to do, such clause, or the contract in which it is found, is uncertain and void. For instance, a clause that says that the party must perform by doing only act X and another clause that says that the party must perform by doing act Y (which is inconsistent with X) may lead to a conclusion of uncertainty. No such difficulty arises here. Compliance with any relevant clause will not lead to breach of any other clause. All bind, but at one level of practicality the most relevant is the widest. Nevertheless, all are binding. Neither their operation nor any principle of law concerned with certainty of contract requires a mechanism or hierarchy of order of operation.
13 This deed has nine several clauses, each clearly expressed. It may be that a complex and difficult clause with multiple permutations and combinations would be so impenetrable as to lack coherent meaning and be uncertain. That is not the case here.
14 For these reasons I do not accept the second argument.
15 Before examining whether any of the cases relied upon alter these conclusions, it is necessary to say something about an argument raised in dialogue in the appeal and which the appellant relied upon. The point was not the subject of a ground of appeal or of specific written submission. The respondent did not, however, object to the reliance by the appellant upon it. The argument was that even if the restraint deed was not void for uncertainty, provisions amounting to repetitive and overlapping restraints of ever widening reach and subject matter were against public policy for the purposes of the Restraints of Trade Act 1976 (NSW), s 4(1).
16 The argument was not fully developed, but its essence was that clauses between employer and employee should exhibit a reasonable attempt to identify a clear and agreed reach for any post-employment constraint. Clauses which seek to establish a multi-layered body of restraints that are complex (even if certain) are against public policy.
17 I would reserve any concluded views about this as a matter of principle to an occasion where the matter was fully argued. Wootten J touched on the argument in Austra Tanks v Running at 843F-844A and 846B-C. Assuming, however, complexity and repetition, if unreasonable, are against public policy, this restraint deed is not capable of being so characterised. The operation of the clauses is tolerably clear. They are apparently an attempt to ensure that some post-contractual restraint would avail OAMPS within the temporal and geographic ranges identified. Given the common law rules and, in particular, those concerning severance and the so-called "blue-pencil" test, it is understandable why commercial parties seek to employ multiple severable clauses. The restraint deed is not against public policy by reason of the multiple and several operation of cll 2 and 4.
18 I turn now to the cases relied upon by the appellant. For the sake of brevity, I do not propose to analyse each case and its facts in detail. Rather, I will express my conclusions as to the relevance of each in short form.
19 In Austra Tanks v Running, Wootten J found a clause that purported to be one clause to be uncertain. The content of the one clause depended upon whether a court found any one or more possible elements enforceable. There were a large number of possible elements to one clause, the binding nature of which was unclear, depending as it did upon the view of a court in any proceeding. As Wootten J said at 843: "… the contract seeks to define the obligation through a series of inquiries as to what is enforceable." There was only one covenant and there was uncertainty as to its content. It was, as Wootten J said at 845, a covenant in substance like that in Davies v Davies (1887) 36 Ch D 359: a covenant "so far as the law allows". For these reasons, the case is distinguishable from the present facts. It provides no principle upon which to reach a conclusion about the restraint deed here different to that which I have come.
20 In Tyser Reinsurance Brokers v Cooper, Young J (as his Honour then was) concluded that a restraint clause was void for uncertainty because of the absence of a crucial definition. By way of obiter dicta, Young J said in respect of a clause that had multiple restraint periods that there were "very real difficulties" with such a clause, partly because it was "not a reasonable way of letting an employee know what are the requirements that bind him or her" and partly because of difficulties in the event of an appeal. His Honour's remarks were not directed at the question of certainty. They may go to the possible public policy question to which I have adverted. The case provides no principle contrary to the conclusion that I have reached about the restraint deed.
21 In Northern Tablelands Insurance Brokers v Howell, Barrett J found a clause with multiple periods of restraint void for uncertainty. The ratio of the decision was that the clause was one covenant with the various periods. There was therefore no clarity as to what the (single) obligation meant. Barrett J then dealt with the clause on the hypothesis that it imposed "several alternative restraints". He asked himself the question (see 315 [47]): "How then does one make the selection among them?" This was in a context in which one clause provided that if any part or provision of the clause was held to be illegal or unenforceable it was severable, and the remaining provisions shall stand. His Honour said the following about this (at [48]):
"[48] … Assume that the part or provision concerning 36 months is the only part or provision held to be illegal or unenforceable. Which of the 12 months and the 24 months provisions then stands? The answer must be that both stand, with the result that there is still no available means of determining which should be observed and enforced."
He continued at [49]-[51]:
"[49] This difficulty stemming from clause 9.7 is of the same kind as was referred to by Young J in Tyser Insurance Brokers Pty Ltd v Cooper (unreported, NSWSC, Young J, 7 December 1998):
'The restraint period should not differ depending on what a court should hold. First, it is not a reasonable way of letting an employee know what are the requirements that bind him or her; secondly, there is great difficulty if there is an appeal against the holding, both pending the appeal and, perhaps, afterwards.