FNC research

Controlling Forum-Shopping: The Impact of

MacShannon v. Rockware Glass Ltd1
By Rhona Schuz

- “Forum-shopping” has been defined3 as a “plaintiff by-passing his natural forum and
bringing his action in some alien forum which would give him relief or benefits which
would not be available to him in the natural forum”. Forum-shopping may be seen as an
evil4 either because it often results in unfairness to the defendant or because the choice of
a forum which has little or no connection with the parties or the subject-matter is thought
to be objectionable per se. The primary concern will be with regulation of forum-shopping in
England. As recently as 1973, Lord Denning MR refused to disapprove of forum-shopping
in England since “it is a good place to shop in, both for quality of goods and the speed of
service”.5 Lord Denning's words typify the attitude, now generally outmoded, that English
justice is superior to that administered in other countries and so ought not to be denied
even to foreigners. Although Lord Denning's view is still shared by some,6 it is no longer
prevalent among the judiciary as a whole.

- The House of Lords8 has on several occasions expressed the need to combat forum-
shopping. There are several devices by which forum-shopping can be controlled: choice-of-
law rules, jurisdictional rules and staying proceedings.
A. Choice-of-Law Rules
A plaintiff will only acquire a substantive advantage from forum-shopping if the law chosen
by the forum to govern the substantive issues in the case is different from that which would
be applied by the natural forum. Thus, the incentive to forum-shop for substantive
advantages will be greatly reduced if
(1) the lex fori never governs substantive issues;
(2) as few issues as possible are categorised as procedural; and
(3) there is greater harmonisation between the choice-of-law rules of different
countries.9However, even if these principles were adopted, plaintiffs might still be tempted
to forum-shop for procedural advantages.
B. Jurisdictional Rules
The most effective way to prevent a plaintiff from forum-shopping in search of either
procedural or substantive advantages is to refuse jurisdiction in cases which are not closely
connected with England. To introduce such an approach, however, would be a radical
departure from the common law rule that an English court has jurisdiction in any case where
the defendant has been properly served with a writ while he is present in England or Wales,
however temporary that presence may be.This rule is objectionable chiefly because whether
or not the defendant is present in this country may be totally fortuitous. Interestingly, the
English courts will not recognise the judgment of a foreign court if the only basis for the
jurisdiction of that court is the presence of the defendant; residence is required. 10

C. Staying Proceedings
The need for the discretion to grant a stay to serve this purpose has been recognised by the
House of Lords in The Atlantic Star16 and MacShannon v. Rockware Glass Ltd.17 These cases
radically altered the approach of the courts towards staying proceedings in cases where
England is not the most appropriate forum. In particular, the test to be applied in these cases
was reformulated in MacShannon.

- Before MacShannon the test to be applied on motions for a stay, whether or not there was
a lis alibi pendens, was that enunciated by Scott LJ in St. Pierre v. South American Stores (Gath &
Chaves Ltd.):(1) A mere balance of convenience is not a sufficient ground for depriving a plaintiff of the
advantages of prosecuting his action in an English court if it is otherwise properly brought … (2) In order to justify
a stay two conditions must be satisfied, one positive and the other negative: (a) the defendant must satisfy the court
that the continuance of the action would work an injustice because it would be oppressive or vexatious to him or
would be an abuse of the process of the court in some other way; and (b) the stay must not cause an injustice to the
plaintiff. On both the burden of proof is on the defendant.21

This test was not effective to control forum-shopping. Firstly, it was not sufficient for the
defendant to show that trial in England would cause him injustice because it was not the
natural forum; he also had to show that it was “oppressive and vexatious” to him for the
action to continue. Thus, a defendant was rarely able to bring the plaintiffs conduct “within the
scope of one of these grave allegations”22 even if there was blatant forum-shopping. It has
been said that “the absence of any real connection between the proceedings and England does
not itself make it vexatious and oppressive for the action to be continued in
England”.23 Secondly, even if the defendant satisfied the first condition of Scott LJ's test, he
then had to show that a stay would cause “no injustice to the plaintiff'. It would seem that this
test was an absolute one and not a relative one. It was not sufficient for the defendant to show
that the injustice caused by a stay was less than the injustice which would be caused to him
by the refusal to grant a stay: he had to show that no injustice would be caused to the
plaintiff. The test propounded by Scott LJ was accepted as an authoritative statement of the
law by all the Law Lords in The Atlantic Star,24 but a majority held that the words “oppressive and
vexatious” ought to be interpreted broadly and understood in a morally neutral sense. The
majority in the House of Lords in The Atlantic Star thus began the “liberalising” process which
was continued in MacShannon.

MacShannon test

1st limb: The defendant must satisfy the court that there is another forum to whose
jurisdiction he is amenable in which justice can be done between the parties at substantially
less inconvenience or expense.

Clearly, convenience and expense48 are important in determining whether there is a more
natural forum, but they are by no means the only factors . For example, in Castanho49 the
House of Lords and Brandon LJ in the Court of Appeal were prepared to assume that
justice could be done in England at substantially less inconvenience and expense but
 nevertheless felt that, since the defendants were part of a group of companies based in
Texas, Texas was just as natural and proper a forum.

There has been some confusion over whether the search is for the natural forum or
a more natural forum.50 it now seems settled51 that the first limb may be satisfied by showing that
the foreign forum is more natural or appropriate than the English forum, even though it cannot
properly be referred to as the natural forum.
In most of the cases there has been no attempt to define naturalness or appropriateness in
relation to possible forums. In MacShannon, Lord Keith refers to the natural forum as being “that
with which the action has the most real and substantial connection”.52 This definition or some
variation of it has been adopted in some recent cases.53 It will be noted that in this context it is
connection with the action rather than the transaction which is relevant. The factors to be considered
and the weight to be given to each factor will differ from those under the proper law of the
contract test. Provided these differences are borne in mind, the analogy with the proper law test is
a useful one. However, the reference here should be to a forum with “a more real and substantial
connection” in order to correspond to the preferable “more natural forum” test.
It has been suggested56 that the real and substantial connection test should totally replace the first
limb of Lord Diplock's test. The advantage of this formulation is that its theoretical basis is more
comprehensible because lawyers and the courts are familiar with the concept.
In deciding whether there is a more natural forum, the courts have taken into account inter alia: the
law which governs the dispute;57 the place where the evidence is situated and in particular the
availability and convenience of witnesses giving oral evidence;58 the place where the incident
occurred (if different);59 the language of the evidence;60 and the residence (and nationality) of the
parties.61 Where the factors point in different directions, the court must put them in the balance in
order to make its decision. It is difficult to establish the criteria used by the court in performing
this balancing exercise. It is unclear whether any more weight should be given to some factors
than to others. Three factors62 have emerged from the cases as potentially decisive:
lis alibi pendens,
- Abidin Daver case
the existence of related proceedings,
- The Messiniaki Tolmi
lex causae.
-The Hida Maru

2nd limb The Second Limb

The stay must not deprive the plaintiff of a legitimate personal or juridical advantage which would be available to him if he
invoked the jurisdiction of the English court.
The second limb of the test retains the concept of the plaintiffs inherent right to English justice. To
what extent the forum-shopping plaintiff will be protected by this second limb depends on two
factors. First, what type of disadvantages are the courts prepared to recognise as justifying the
plaintiffs continuation of proceedings in a prima facie non-natural forum? Second, is a finding that
the plaintiff will be deprived of a legitimate advantage decisive in preventing a stay or merely a
precondition to the application of the balance of justice? These alternatives will be examined below
under the heading, “The Balance of Justice”,77 but should be borne in mind during discussion of the
various advantages.
The courts control the relevance of advantages by two means. First, the advantage must be
real and not speculative or fanciful. Second, the advantage must be legitimate. No doubt
the court's view of the plaintiffs motive may also indirectly influence its finding of whether
the advantage is real. Some of the advantages on which plaintiffs have sought to rely in the
cases will now be examined.
1. Procedural advantage
-The El Amria and El Mina, opportunity to cross examinate witness is a legitimate
advantage
- Smith Kline & French Laboratories Ltd v. Bloch,93 Lord Denning MR stated that
“lawyers on contingency fees” were an illegitimate advantage. + jury trial is not a
legitimate advantage
 2. Prospect of higher damages
Castanho= House of Lords held that the prospect of obtaining higher damages in the USA
was a legitimate advantage which would justify the refusal to grant an injunction. Similarly
in The Jalakrishna,117 on an application for a stay, the fact that the plaintiffs damages would
be at least £17,500 greater in England than in India was held to be legitimate.

Smith Kline118 = the plaintiffs claim that he would obtain higher damages in the USA was
considered highly speculative because there was expert evidence to the effect that the
Pennsylvanian court would probably apply English law to the issue of damages, and thus
there was held to be no real advantage. Lord Denning MR claimed obiter that such an
advantage was anyway an illegitimate one. Such a view appears to be irreconcilable with the
House of Lords' view in Castanho.119 However, the two cases can be distinguished on the
same basis as that postulated for the cases on “delay”. In Castanho120 it was not unfair to
the defendant to sue in Texas, because that forum was just as natural as the
English forum and so the prospect of higher damages in Texas was a legitimate advantage.
Conversely, in Smith Kline,121 it was unfair to sue in Philadelphia since the real defendant
was English and England was “undoubtedly the natural forum” for litigating the dispute.

3. Substantive advantages
A plaintiff will only be deprived of a substantive advantage where the English court will
choose a law to govern the case different from what would be chosen by the foreign court
(or apply the same law differently). The only case where a substantive advantage has been
relied on in refusing a stay is The Hida Maru.123
It is suggested that in general the court ought not to compare the way the case would be
decided in the competing forums.124 Such comparisons are as invidious as comparisons of
procedure and more likely to be based on conflicting foreign evidence. Further, the lex
causae is given great weight in deciding whether there is a more natural forum and should
not be resurrected under the second limb.
However, public policy may demand that where English law is the lex causae and there is
clear evidence that the foreign court will not give proper effect to English law, then a stay
should be refused. The Hida Maru probably falls within such an exception.

The Balance of Justice127

What is the position where the two limbs of the test give conflicting
results? Presumably, under Lord Diplock's test,128 the application for a stay
would fail unless both limbs were satisfied. Lord Keith, however,
in MacShannon discussed the position expressly and envisaged a different
approach. He said that it is
necessary to weigh against each other the advantages to the plaintiff and the disadvantages
to the defendant and a stay will not be granted unless the court concludes that to refuse it
would involve injustice to the defendant and [the grant would involve] no injustice to the
plaintiff.129

The concept of justice as the deciding factor can also be found in Lord
Salmon's judgment. He stated that
the question as to whether [an action] should be stayed depends on whether the defendants
can establish that to refuse a stay would produce injustice … To my mind, the real test of a
stay depends on what the court in its discretion considers that justice demands.130

Is the MacShannon Test an Effective Obstacle to Forum-Shoppers?

Stays in foreign jurisdiction/arbitration clause cases (distinction bet FNC & ASI)
Before The Atlantic Star,194 cases in which it was sought to stay an action
brought in breach of a foreign jurisdiction clause were decided on entirely
different principles from those used in forum non conveniens and lis alibi
pendens cases. However, following the liberalisation of
the forum non conveniens test, the two lines of cases have moved closer
together. Many factors are common to both195 and, in applying the Mac-
Shannon test, the courts have clearly been considerably influenced by the
foreign jurisdiction clause cases.196 Similarly, the influence of Mac-
Shannon can be seen in recent foreign jurisdiction clause cases, but
ironically this influence seems to have led to a relaxation rather than an
increase in the control of forum-shopping in those cases.
The explanation for this curious phenomenon lies in the fundamental
differences between the two types of case. Ex hypothesi, In other words,
the role of the stay in the foreign jurisdiction clause case is not simply to
control forum-shopping, but also to promote adherence to the
bargain.198 All factors relevant to the decision whether to grant a stay
should be viewed in this light. For example, when considering where
evidence is situated or the availability of witnesses it should be
remembered that the parties may well have been able to foresee these
facts when choosing a forum.199 Far less weight should be given to these
matters than in forum non conveniens cases unless the circumstances of the
dispute are entirely different from those which could reasonably have
been anticipated.
Grant of stay case: The El Amria,200 Brandon LJ increased the scope for
taking into account procedural advantages in foreign jurisdiction clause
cases. His comments seem to have been motivated by
the MacShannon cases but appear to be misconceived. Any attempt to
balance advantages and disadvantages to the parties of trial in the
alternative forums detracts from the precedence which should be given to
the forum to which the parties have contracted to submit their disputes.
In a number of recent first instance cases,201 whilst lip service has been
paid to the principles applicable to foreign jurisdiction clause cases,
considerable weight has been attached to questions of convenience and
appropriateness of forum and very little importance has been given to the
agreement of the parties. It is suggested that this tendency results from a
blurring of the differences between foreign jurisdiction clause
and forum non conveniens cases.202 Although the courts make a formal
distinction between the two tests, the actual way in which the discretion is
exercised may not be very different.203 In this context, the influence
from MacShannon is unhelpful since emphasis on convenience and
naturalness of forum is likely to result in more stays being refused and
hence more forum-shopping.
Conclusions and alternatives to MacShannon
The analysis of the MacShannon test and the way in which it has been applied reveal a number of
defects. First, the test is not a totally effective obstacle to forum-shopping, although it is more
successful than its predecessor. Second, great difficulty and uncertainty have arisen over the
identification of legitimate advantages. Third, because of the widely different nature of the factors
to be weighed, the balance of justice cannot be performed on any more practicable basis than
judicial instinct. Fourth, proper consideration has not always been given to the application and
influence of the MacShannon principles in other types of cases.
The second and third flaws are probably the cause of the first. By allowing the plaintiff to plead
legitimate advantages on the second limb, the implication is that the corresponding disadvantages
to the defendant are of less weight, and this affects the way in which the balance of justice is seen
to lie.219 It is noteworthy that, in the two admiralty cases where forum-shopping was
condoned,220 the legitimate advantages to the plaintiff of trial in England were the determining
factors in the refusal to grant the stay. The pro-plaintiff bias of the test is, of course, exacerbated
International and Comparative Law Quarterly 35, 1986 pp 374–412 at 409
if the burden of proof is placed on the defendant for the second as well as the first limb. The fourth
defect will only be rectified if there is clearer recognition by the courts that, even though the
relevant factors in injunction and foreign jurisdiction clause cases are similar to those
in forum non conveniens cases, the policy considerations are quite different. The reformulations of
the MacShannon test considered below would not solve this problem.221
Various alternatives to the MacShannon formulation have been suggested. Some dicta and
commentators222 support a single “balance of justice” or “critical equation” test based on Lord
Wilberforce's dictum in The Atlantic Star.223 Although such an approach does eliminate the need to
identify legitimate advantages, it would surely lead to much greater uncertainty than
the MacShannon test because of the wide flexibility and lack of guidelines. Such a discretion has
been described as “unstructured and unstructurable”224 and would have to be based entirely225 on
judicial instinct.
In contrast, a formulation which is much more precise than MacShannon has been advocated by
Briggs:226

An action brought in the English courts shall be stayed if either (a) it is brought in breach of an agreement to submit the
dispute to the jurisdiction of a foreign court, or (b) the courts of another country have the closest and most real connection
with the dispute; and that foreign court has jurisdiction to hear the dispute. Provided always that (i) notwithstanding the
above, an action shall be stayed if the sole reason for the foreign court not having jurisdiction to hear the dispute is that, by
the law administered in that court, the claim of the plaintiff is barred by lapse of time; (ii) no action shall be stayed if the
judgment of the foreign court would not be entitled to be recognised in England; and (iii) no action shall be stayed where to
do so would be contrary to express statutory provision.

Even the wording of this formula contains a number of uncertainties. In the first place, no mention
is made of where the burden of proof lies, but it is presumed that, since this formula is to replace
Dicey and Morris's

International and Comparative Law Quarterly 35, 1986 pp 374–412 at 410

Rules 30 and 31, the onus will be on the defendant to establish that there is another forum which
has the closest and most real connection with the dispute or that there is a valid foreign
jurisdiction clause. Second, the reference is to the connection with the dispute rather than to the
action.227 Does this mean that factors such as convenience of witnesses and availability of
evidence, which have nothing to do with the actual dispute but are very important when
considering trial, cannot be considered? Third, the superlative form “closest and most real” rather
than the comparative “closer and more real” is used. This would suggest that, in the unlikely event
of the closest and most real forum not accepting jurisdiction, a stay could not be granted even if
there was a third forum which had a closer and more real connection with the dispute than
England and which was prepared to take jurisdiction.
However, Briggs's formula does have a number of advantages over the present MacShannon test. In
particular, it does not suffer from the defects of that test outlined above. Moreover, a test based
on degree of connection is preferable to one based on naturalness of forum. Furthermore, Briggs's
formula does ensure that stays will only be granted in foreign jurisdiction clause cases in the
narrow circumstances covered by his second and third provisos.
Nevertheless, Briggs's proposals goes too far in the direction of precision and certainty: the
elimination of all discretion from the decision whether to grant a stay in
both forum non conveniens and foreign jurisdiction clause cases might lead to injustice. Briggs's
reliance on the non-enforceability of judgments obtained in breach of natural justice as a
safeguard against stays causing injustice is an over-simplification. First, as he points out, this
ground of non-enforceability is very rarely pleaded successfully. Second, other factors besides the
denial of natural justice may cause injustice to the plaintiff. In particular, in the common situation
where the degree of connection with one country is only marginally closer than with another, it
may be unfair to determine the case by reference to factors which affect the degree of connection
to the exclusion of other factors. For example, a key witness may be too ill to travel228 to that
country or may be unable to obtain a visa. It is doubtful whether these facts would be relevant in
assessing the degree of connection and yet they could be fundamental to the justice of the case.
Even pure procedural advantages such as the possibility of subpoenaing a witness or obtaining
proper discovery should be relevant when one forum is not clearly more closely connected with
another. If the Briggs formula were introduced the courts would, in order to avoid injustice, take
these sorts of factor into account sub silentio in determining whether there is
International and Comparative Law Quarterly 35, 1986 pp 374–412 at 411
another forum with the closest and most real connection, because judicial discretion cannot be
removed completely from the decision whether to grant a stay. It is therefore better that any
formulation expressly provides for the consideration of the justice of the case so that the discretion
is, as far as possible, exercised openly.
It is suggested that Briggs's formulation be amended to take account of this residuary
discretion229 and to settle the ambiguities discussed above. It would read as follows:
Any action brought in the English courts shall be stayed if the defendant shows that either (a) it is brought in breach of an
agreement to submit the dispute to the jurisdiction of a foreign court230 or (b) the courts of another country have a closer and
more real connection with the action and that the foreign court has jurisdiction to hear the dispute unless the plaintiff proves
that in all the circumstances of the case it would be unjust for the action to be tried in that foreign court. Provided always
that (i) no action shall be stayed if the judgment of the foreign court would not be entitled to be recognised in England; and
(ii) no action shall be stayed where to do so would be contrary to express statutory provision.231
The incorporation of an “escape clause” based on justice might seem to make the formulation as
uncertain as a simple “critical equation” test. But this proposal is a more structured, and therefore
more satisfactory, version of the “critical equation” test. The key difference is the presumption that
it is just for the case to be heard in the forum with the closer and more real connection. There is a
very heavy burden on the plaintiff to show that it would be unjust for the case to be tried there,
despite the degree of connection between that forum and the action.
Clearly decisions based on justice232 must depend to a large extent on judicial instinct, although no
doubt if the formulation were to be adopted a body of case law would soon build up as to which
factors were likely to affect justice and which were not. This might seem to be similar to the
jurisprudence which has developed in relation to the identification of legitimate advantages. No
doubt issues would arise very similar to those now pleaded under the second limb of
the MacShannon test. However, because these factors would be pleaded in relation to the question of
justice rather than as an advantage to the plaintiff, the court would be more likely to view them
objectively. Where the benefit to one
International and Comparative Law Quarterly 35, 1986 pp 374–412 at 412
party is cancelled out by the detriment to the other party,233 the factor would not be taken into
account unless it would be unjust for one party to rely on that factor. For example, the availability
of higher damages would not be relevant because this is as much a detriment to the defendant as
it is a benefit to the plaintiff. However, to return to an earlier example, although a defendant might
benefit from the non-availability of a key witness who was too ill to travel, it would be unjust for
him to rely on this advantage.
Since the court would have to look at justice in all the circumstances of the case, the more closely the
action was connected with the foreign forum, the more difficult it would be to show that trial there
would be unjust. A fortiori it would be even more difficult to show that trial in a previously
chosen forum would be unjust. Factors which would make it unjust for a case to be heard in a
marginally more closely connected forum would not make it unjust for the case to be heard in a
previously agreed forum. Within the formula, the courts would be able to keep clear the important
differences between foreign jurisdiction clause cases and forum non conveniens cases. The court's
discretion not to grant a stay in favour of a forum with a closer connection would most often be
exercised in cases where there was only a marginally greater connection with the foreign forum. It
is in these cases that it is important that the court should be able to look at a wider range of
factors; more flexibility is therefore required.
The proposed formula seeks to combine the increased precision of Briggs's formula with the
necessary flexibility of the MacShannon formula. It represents a further evolution
of MacShannon principles rather than a complete change of direction. The importance of the degree
of connection is increased and that of the advantages to the plaintiff minimised. This, together
with the clarification of the onus of proof, would reduce uncertainty and facilitate control of forum-
shopping.