Insurance fails to pay up.

First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial trip
payment." (Day one is the date the “initial trip payment” is received.)
This exclusion applies to all coverages. It applies to the Insured, all
the Immediate Family Members, and Traveling Companions, whether or not
they are traveling with the Insured. Please read it carefully.


$25,000: Medical Expenses
Covers necessary medical expenses up to one year after the sickness or
injury, provided you sought initial medical treatment while on your trip.

$300,000: Emergency Medical Transportation
Covers evacuation and transportation to the nearest adequate medical
facility.

These amounts appear adequate for those portions of medical expenses not
covered by my regular insurance when abroad. Feel free to research the
conditions for Travel Guard for details that I've missed. The coverage
for those over 70 does not appear impaired but I'm sure the fee for the
policy will increase considerably.

Roland Perry wrote:
In message , at 01:52:40 on Tue, 5
Oct 2004, Frank F. Matthews remarked:

Travel Guard is one such.

Roland Perry wrote:

In message , at 18:30:56 on
Sun, 3 Oct 2004, Frank F. Matthews
remarked:

Much simpler to get a policy that will cover preexisting conditions.

Available from the Porcine Aviator Insurance Co?


Only up to a point. Their conditions for pre-existing conditions a

Worldwide Health Extended - $2,500 limit;
Worldwide Health - For U.S. Citizens Traveling Abroad - $1,000 Limit;
Worldwide Health - For Non-U.S. Citizens Traveling Outside Their Home
Country - No Coverage

And in common with other American-based health insurance they have high
"excess" (deductible) [you get to choose from $100 to $2500] and then
pay only the first 80% of the next $5000 within the USA.

Europeans are more familiar with policies with zero excess for health,
and 100% payments.

And the premiums will reflect the choice of excess, so it might be very
expensive to have lower than $500, and so an American abroad with a
pre-existing condition will get a maximum payment of $500, or if
travelling inside the USA, 80% of $2,000.

These amounts won't cover much more than an initial consultation should
you fall ill on holiday, and are nothing like the $20M that a typical
European policy will pay.

The fees section also has stealth conditions. For example if you are
aged 70-80 US citizen travelling abroad, you can only buy up to $50K of
cover, not $1M. That's $480 for a $500 deductible, a huge premium to
European eyes, and for such small coverage.

There are other exclusions that sound odd to European ears, such as:
"Charges for use of Emergency Room for treatment of Illness unless the
patient is directly admitted to the Hospital as Inpatient for further
treatment of that Illness." So if the visit to the emergency room cures
the condition, you don't get paid! Perhaps this reflects an American
culture that the hospital will admit you as a lucrative in-patient if
you have more than a scratched fingernail, but it's surprising how many
minor injuries a British Emergency room will provide full treatment for
(ahead of you returning home for a follow-up).

Finally (and this is the really crucial one in the context of the
current discussion) there appears to be *no* coverage whatsoever for
cancelling a trip because one of the group, or a close relative, falls
ill. (Only for aborting a trip once it's started).

On 5/10/04 15:38, in article , "Frank
F. Matthews" wrote:

First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial trip
payment." (Day one is the date the ³initial trip payment² is received.)
This exclusion applies to all coverages. It applies to the Insured, all
the Immediate Family Members, and Traveling Companions, whether or not
they are traveling with the Insured. Please read it carefully.


$25,000: Medical Expenses
Covers necessary medical expenses up to one year after the sickness or
injury, provided you sought initial medical treatment while on your trip.

$300,000: Emergency Medical Transportation
Covers evacuation and transportation to the nearest adequate medical
facility.

These amounts appear adequate for those portions of medical expenses not
covered by my regular insurance when abroad. Feel free to research the
conditions for Travel Guard for details that I've missed. The coverage
for those over 70 does not appear impaired but I'm sure the fee for the
policy will increase considerably.

Roland Perry wrote:
In message , at 01:52:40 on Tue, 5
Oct 2004, Frank F. Matthews remarked:

Travel Guard is one such.

Roland Perry wrote:

In message , at 18:30:56 on
Sun, 3 Oct 2004, Frank F. Matthews
remarked:

Much simpler to get a policy that will cover preexisting conditions.

Available from the Porcine Aviator Insurance Co?


Only up to a point. Their conditions for pre-existing conditions a

Worldwide Health Extended - $2,500 limit;
Worldwide Health - For U.S. Citizens Traveling Abroad - $1,000 Limit;
Worldwide Health - For Non-U.S. Citizens Traveling Outside Their Home
Country - No Coverage

And in common with other American-based health insurance they have high
"excess" (deductible) [you get to choose from $100 to $2500] and then
pay only the first 80% of the next $5000 within the USA.

Europeans are more familiar with policies with zero excess for health,
and 100% payments.

And the premiums will reflect the choice of excess, so it might be very
expensive to have lower than $500, and so an American abroad with a
pre-existing condition will get a maximum payment of $500, or if
travelling inside the USA, 80% of $2,000.

These amounts won't cover much more than an initial consultation should
you fall ill on holiday, and are nothing like the $20M that a typical
European policy will pay.

The fees section also has stealth conditions. For example if you are
aged 70-80 US citizen travelling abroad, you can only buy up to $50K of
cover, not $1M. That's $480 for a $500 deductible, a huge premium to
European eyes, and for such small coverage.

There are other exclusions that sound odd to European ears, such as:
"Charges for use of Emergency Room for treatment of Illness unless the
patient is directly admitted to the Hospital as Inpatient for further
treatment of that Illness." So if the visit to the emergency room cures
the condition, you don't get paid! Perhaps this reflects an American
culture that the hospital will admit you as a lucrative in-patient if
you have more than a scratched fingernail, but it's surprising how many
minor injuries a British Emergency room will provide full treatment for
(ahead of you returning home for a follow-up).

Finally (and this is the really crucial one in the context of the
current discussion) there appears to be *no* coverage whatsoever for
cancelling a trip because one of the group, or a close relative, falls
ill. (Only for aborting a trip once it's started).


English insurance law is particularly insurer-friendly. Any
misrepresentation, material or not, may invalidate a policy. A material
misrepresentation certainly will invalidate it.

By contrast, in France they reduce the payment to the value of cover the
premium would have bought with all pre-existing conditions taken into
account.

In Germany, the pre-existing condition will invalidate the policy only if it
is relevant to the loss.

One might consider buying an offshore policy, such as AIG's, aimed at
expatriates.

I have often sent my insurers letters notifying them of happenings (such as
my unexpected employment assignment abroad). Sometimes they reply, sometimes
not. A few years ago I told my home fire insurer that although I had bought
minimum contents cover, I specifically wanted it to be secondary to my
household goods policy bought from an American insurer (because that was not
only new-for-old, much cheaper, and with much higher limits, but also had
far fewer exclusions). Of course they assented because it reduced their
risk; it also meant I didn't have double cover, the worst of all possible
situations since each insurer will split the loss and pay according to its
own terms.

On 5/10/04 15:38, in article , "Frank
F. Matthews" wrote:

First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial trip
payment." (Day one is the date the ³initial trip payment² is received.)
This exclusion applies to all coverages. It applies to the Insured, all
the Immediate Family Members, and Traveling Companions, whether or not
they are traveling with the Insured. Please read it carefully.


$25,000: Medical Expenses
Covers necessary medical expenses up to one year after the sickness or
injury, provided you sought initial medical treatment while on your trip.

$300,000: Emergency Medical Transportation
Covers evacuation and transportation to the nearest adequate medical
facility.

These amounts appear adequate for those portions of medical expenses not
covered by my regular insurance when abroad. Feel free to research the
conditions for Travel Guard for details that I've missed. The coverage
for those over 70 does not appear impaired but I'm sure the fee for the
policy will increase considerably.

Roland Perry wrote:
In message , at 01:52:40 on Tue, 5
Oct 2004, Frank F. Matthews remarked:

Travel Guard is one such.

Roland Perry wrote:

In message , at 18:30:56 on
Sun, 3 Oct 2004, Frank F. Matthews
remarked:

Much simpler to get a policy that will cover preexisting conditions.

Available from the Porcine Aviator Insurance Co?


Only up to a point. Their conditions for pre-existing conditions a

Worldwide Health Extended - $2,500 limit;
Worldwide Health - For U.S. Citizens Traveling Abroad - $1,000 Limit;
Worldwide Health - For Non-U.S. Citizens Traveling Outside Their Home
Country - No Coverage

And in common with other American-based health insurance they have high
"excess" (deductible) [you get to choose from $100 to $2500] and then
pay only the first 80% of the next $5000 within the USA.

Europeans are more familiar with policies with zero excess for health,
and 100% payments.

And the premiums will reflect the choice of excess, so it might be very
expensive to have lower than $500, and so an American abroad with a
pre-existing condition will get a maximum payment of $500, or if
travelling inside the USA, 80% of $2,000.

These amounts won't cover much more than an initial consultation should
you fall ill on holiday, and are nothing like the $20M that a typical
European policy will pay.

The fees section also has stealth conditions. For example if you are
aged 70-80 US citizen travelling abroad, you can only buy up to $50K of
cover, not $1M. That's $480 for a $500 deductible, a huge premium to
European eyes, and for such small coverage.

There are other exclusions that sound odd to European ears, such as:
"Charges for use of Emergency Room for treatment of Illness unless the
patient is directly admitted to the Hospital as Inpatient for further
treatment of that Illness." So if the visit to the emergency room cures
the condition, you don't get paid! Perhaps this reflects an American
culture that the hospital will admit you as a lucrative in-patient if
you have more than a scratched fingernail, but it's surprising how many
minor injuries a British Emergency room will provide full treatment for
(ahead of you returning home for a follow-up).

Finally (and this is the really crucial one in the context of the
current discussion) there appears to be *no* coverage whatsoever for
cancelling a trip because one of the group, or a close relative, falls
ill. (Only for aborting a trip once it's started).


English insurance law is particularly insurer-friendly. Any
misrepresentation, material or not, may invalidate a policy. A material
misrepresentation certainly will invalidate it.

By contrast, in France they reduce the payment to the value of cover the
premium would have bought with all pre-existing conditions taken into
account.

In Germany, the pre-existing condition will invalidate the policy only if it
is relevant to the loss.

One might consider buying an offshore policy, such as AIG's, aimed at
expatriates.

I have often sent my insurers letters notifying them of happenings (such as
my unexpected employment assignment abroad). Sometimes they reply, sometimes
not. A few years ago I told my home fire insurer that although I had bought
minimum contents cover, I specifically wanted it to be secondary to my
household goods policy bought from an American insurer (because that was not
only new-for-old, much cheaper, and with much higher limits, but also had
far fewer exclusions). Of course they assented because it reduced their
risk; it also meant I didn't have double cover, the worst of all possible
situations since each insurer will split the loss and pay according to its
own terms.

In message , at 14:38:22 on Tue,
5 Oct 2004, Frank F. Matthews remarked:
First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial
trip payment." (Day one is the date the “initial trip payment� is
received.) This exclusion applies to all coverages.

Fair enough. Their website does them few favours, when it comes to
finding information :-(
--
Roland Perry

"Sufaud" wrote in message
...
[snip]
English insurance law is particularly insurer-friendly. Any
misrepresentation, material or not, may invalidate a policy. A
material
misrepresentation certainly will invalidate it.
[snip]

The position in England is not quite that bad from the policyholder's
perspective.

First, the material misrep. has to have induced the decision to
insure, or
affected the terms. (Pan Atlantic v Pine Top)

Secondly, the Financial Ombudsman Service can impose a fair and
reasonable result and not infrequently does.

(uk.legal added)

Roland Perry wrote:
In message , at 14:38:22 on Tue,
5 Oct 2004, Frank F. Matthews remarked:

First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial
trip payment." (Day one is the date the “initial trip payment� is
received.) This exclusion applies to all coverages.

Fair enough. Their website does them few favours, when it comes to
finding information :-(

And, as with most US insurance, the policy is state specific. In my
case the price & conditions are specific to Texas.

I must admit that when I tried "International" for a residence things
cratered. Perhaps they only work with US & Canada although the
international choice is on the list.

"Frank F. Matthews" wrote in message
...
Roland Perry wrote:
In message , at 14:38:22 on Tue,
5 Oct 2004, Frank F. Matthews remarked:

First consideration of pre-existing conditions.

PRE-EXISTING MEDICAL CONDITION EXCLUSION WAIVER:
The Pre-Existing Medical Condition Exclusion will be waived if the
insurance is purchased within fifteen calendar days of the "initial trip
payment." (Day one is the date the “initial trip payment� is
received.) This exclusion applies to all coverages.

Fair enough. Their website does them few favours, when it comes to
finding information :-(

And, as with most US insurance, the policy is state specific. In my case
the price & conditions are specific to Texas.

I must admit that when I tried "International" for a residence things
cratered. Perhaps they only work with US & Canada although the
international choice is on the list.

I thought the US was international for Texans.

On 5/10/04 21:05, in article , "Rhoy the
Bhoy" wrote:

The position in England is not quite that bad from the policyholder's
perspective.

First, the material misrep. has to have induced the decision to
insure, or
affected the terms. (Pan Atlantic v Pine Top)

Secondly, the Financial Ombudsman Service can impose a fair and
reasonable result and not infrequently does.

Aside from the question of what is holding and what is dictum (and that the
appellant lost in your case) the fact remains that policyholders in the UK
are in far worse a position than their counterparts in the USA or on the
Continent.

Here's your case in full:

-----

PAN ATLANTIC INSURANCE CO. LTD. AND ANOTHER
APPELLANTS
AND
PINE TOP INSURANCE CO. LTD.

RESPONDENTS
[HOUSE OF LORDS]
[1995] 1 AC 501

HEARING-DATES: 14, 15, 16, 17, 21, 22, February 21, March 25 July 1994
25 July 1994

CATCHWORDS:
Insurance - Reinsurance - Non-disclosure - Reinsurance renewal - Plaintiffs
failing to disclose additional losses for previous year - Whether "material
circumstance" - Effect on mind of prudent insurer - Avoidance of contract -
Whether insurer induced by non-disclosure to enter into policy - Marine
Insurance Act 1906 (6 Edw. 7, c. 41), ss. 17, 18(1)(2), 20(1)(2)

HEADNOTE:
Between 1977 and 1982, the plaintiffs wrote a quantity of direct American
liability insurance, much of it being "long-tail" business in which a long
period of time might elapse before claims matured. In relation to the years
1977 to 1979, the plaintiffs' casualty account was reinsured with other
insurers for excess of loss above a certain figure. For the 1980 policy
year, the defendants became reinsurers in respect of that cover, and they
subsequently renewed the reinsurance for 1981 and 1982. Disputes arose in
respect of all three years, and the plaintiffs obtained judgment in
proceedings against the defendants in respect of the 1980 and 1981
reinsurance contracts. In 1987, they commenced proceedings on the 1982
contract to recover paid and outstanding claims as at 31 December 1986 and
for declarations that the contract was valid. By their points of defence,
the defendants sought to avoid the plaintiffs' claims on the ground of
non-disclosure of loss statistics for the underwriting years 1977 to 1979
and consequent misrepresentation of the risk. By subsequent amendment, they
also sought to avoid the plaintiffs' claims on the ground of non-disclosure
in respect of the 1980 and 1981 underwriting years. The judge rejected their
defence based on non-disclosure of the statistics for 1977 to 1979 and held
that an inaccuracy in the 1980 statistics was not material, but upheld the
defendants' defence based on non-disclosure of the losses for 1981. It was
common ground that the true losses for 1981 had been U.S.$468,168 as against
disclosed losses of U.S.$235,768; that the plaintiffs had had information
about those additional losses before the slip had been signed on 13 January
1982; and that those losses had not been disclosed. The judge held that the
additional losses for 1981 had been material to the contract and should have
been disclosed and that, accordingly, the defendants were entitled to avoid
the 1982 contract. The Court of Appeal dismissed an appeal by the
plaintiffs.

On appeal by the plaintiffs:-

Held, dismissing the appeal, (1) (Lord Templeman and Lord Lloyd of Berwick
dissenting) that under section 18 of the Marine Insurance Act 1906, n1 which
was applicable by analogy in a non-marine case, a "material circumstance"
was one that would have an effect on the mind of the prudent insurer in
estimating the risk and it was not necessary that it should have a decisive
effect on his acceptance of the risk or on the amount of premium demanded
(post, pp. 517D-E, 530H-531A, D, 532E-F, 541D-E, 550C,551H- 552A).

(2) That before an underwriter could avoid a contract for non-disclosure of
a material circumstance he had to show that he had actually been induced by
the non-disclosure to enter into the policy on the relevant terms (post, pp.
516E, 549D, 550A-B, C-D, 552A-B, 571D-G).

Berger v. Pollock [1973] 2 Lloyd's Rep. 442 approved.

Container Transport International Inc. v. Oceanus Mutual Underwriting
Association (Bermuda) Ltd. [1984] 1 Lloyd's Rep. 476, C.A. overruled in
part.

(3) That the additional losses for 1981 had been a material circumstance
that ought to have been disclosed and the judge's conclusion that the
defendants were entitled to avoid the contract should stand (post, pp.
515H-516A, 518C, 551A-B, F-G, 552B, 573G-574B).

Decision of the Court of Appeal [1993] 1 Lloyd's Rep. 496 affirmed.

INTRODUCTION:
APPEAL from the Court of Appeal.

This was an appeal by the plaintiffs, Pan Atlantic Insurance Co. Ltd. (on
its own behalf and on behalf of all members of the Pan Atlantic Group
Reinsurance Syndicate and/or the Pan Atlantic Reinsurance Group in 1982) and
Republic Insurance Co. (body corporate) (on its own behalf and on behalf of
all members of the Pan Atlantic Group Reinsurance Syndicate and/or the Pan
Atlantic Reinsurance Group in 1982), by leave of the House of Lords from the
judgment of the Court of Appeal (Sir Donald Nicholls V.-C., Farquharson and
Steyn L.JJ.) [1993] 1 Lloyd's Rep. 496 given on 3 March 1993 refusing an
application by the plaintiffs for leave to amend their notice of appeal and
to re-re-re-re-amend their points of reply in the action and dismissing the
plaintiffs' appeal from the judgment of Waller J. [1992] 1 Lloyd's Rep. 101
given on 25 March 1991 by which he dismissed the plaintiffs' claims in their
action against the defendants, Pine Top Insurance Co. Ltd.

The Court of Appeal refused the plaintiffs' application for leave to appeal
to the House of Lords from their judgment, but on 19 July 1993 the Appeal
Committee of the House of Lords (Lord Templeman, Lord Jauncey of
Tullichettle and Lord Mustill) allowed a petition by the plaintiffs for
leave.

The facts are stated in the opinions of Lord Mustill and Lord Lloyd of
Berwick.

COUNSEL:
Michael Beloff Q.C., Steven Berry and Sarah Moore for the plaintiffs. For
the purpose of the general law of non-marine insurance, a circumstance that
is not disclosed or is misrepresented is material if it is a circumstance
that (i) would have caused a prudent and reasonable underwriter to reject
the risk or accept the risk on different terms had he known the circumstance
(the "different decision" test) and (ii) if known to the particular
underwriter would actually have induced him to make a different decision
(the "actual influence" test). The relevant impact is, therefore, that on
both the prudent underwriter and the actual underwriter, and the
circumstance must be one that would lead to a different final assessment.
The contrary decision in Container Transport International Inc. v. Oceanus
Mutual Underwriting Association (Bermuda) Ltd. [1984] 1 Lloyd's Rep. 476
("the C.T.I. case") either applies only to marine cases or is wrong. For
criticisms of the C.T.I. case, see Henry Brooke Q.C., "Materiality in
insurance contracts" [1985] L.M.C.L.Q. 437; Adrian Hamilton Q.C., "Avoidance
of liability on the grounds of misrepresentation and non-disclosure, or, the
rise and rise of avoidable reinsurance," Insurance and Reinsurance Law,
September 1985, p. 131; Anthony Diamond Q.C., "The law of marine insurance -
has it a future?" [1986] L.M.C.L.Q. 25; David St. L. Kelly, "Recent
Developments in relation to Inducement in Non-disclosure and
misrepresentation" (1988) 1 Ins.L.J. 30; Dr. Malcolm Clarke, "Failure to
Disclose and Failure to Legislate: is it Material? - II" [1988] J.B.L. 298;
Clarke, The Law of Insurance Contracts (1989), pp. 452-456; Steyn J., "The
Role of Good Faith and Fair Dealing in Contract Law: A Hairshirt
Philosophy?" [1991] Denning L.J. 131, 138-140; Howard N. Bennett, "The Duty
to Disclose in Insurance Law" (1993) 109 L.Q.R. 513; Mark Humphries, "Duties
of disclosure and the prudent underwriter," Tolley's Insurance Law &
Practice, Summer 1992, p. 48; Peter Rogan, "Uberimma Fides - Lord of
Misrule?" (1992) 1 Global Reinsurance 215 and P. J. H. Rogan, "The Prudent
Underwriter - Time for Change" (1992) 79 B.I.L.A.J. 7. The position under
the Marine Insurance Act 1906 is not different, but, if it is, it does not
affect the position in non-marine cases. The burden of proof is on the
underwriter to show materiality by satisfying the two tests. In so far as
Redgrave v. Hurd (1881) 20 Ch.D. 1 departs from this basic principle of
English law it was wrongly decided: see Spencer Bower and Turner, Actionable
Misrepresentation, 3rd ed. (1974), pp. 154-155. [Reference was made to Smith
v. Chadwick (1884) 9 App.Cas. 187, 196.] No similar onus probandi should be
imposed on a party who makes an inadvertent disclosure not calculated to
induce the other to enter into the contract. The court in assessing
materiality should not assume that the prudent underwriter had accepted the
risk on the terms accepted by the actual underwriter before going on to
consider the materiality of the undisclosed or misrepresented circumstance.
The insurer should not be allowed to rely on matters that were irrelevant to
him although they might have been relevant to a more sensible person.

It is a fundamental principle that contracts should be binding. Rights to
avoid contracts are exceptions to this principle: see Clarke, The Law of
Insurance Contracts, p. 456. The relevant exception for present purposes is
that a contract should be avoidable if the consent of one of the parties was
vitiated by imposition by the other party. This is necessarily limited by
the principle that the imposition must be such as would vitiate the consent
of the hypothetical reasonable contracting party. There must be an objective
element to the test so that contracting parties can rely on contracts that
were made without objectively apparent imposition even if the other party
was subjectively (because of some idiosyncrasy) imposed upon. This equitable
exception on the ground of imposition is the origin of the modern right to
avoid all contracts for misrepresentation: see Banque Keyser Ullmann S.A. v.
Skandia (U.K.) Insurance Co. Ltd.[1990] 1 Q.B. 665 and Merchants' and
Manufacturers' Insurance Co. Ltd. v. Hunt [1941] 1 K.B. 295. Insurance
contracts differ from other contracts in that the assured can be expected to
have all of the information in respect of the risk and the insurer can be
expected to have none. This justifies a duty of good faith, leading to a
positive duty of disclosure, in insurance contracts: see Carter v. Boehm
(1766) 3 Burr. 1905. The basis of avoidance is that the risk run is in fact
different from what the insurer understood it to be and therefore the
insurer has been prejudiced by the non-disclosure because he was not able
properly to estimate the risk. The principle remains that the contract of
insurance should be avoidable for non-disclosure only if obtained by
objective imposition.

The fears of Lloyd J. in the C.T.I. case as to the impracticality of the
"impact on decision-making" test are justified. Any summary is going to be
incomplete and, left on its own, bound to be open to criticism with the
advantage of hindsight. For the issue of avoidance to arise, there will have
been heavy losses. Hindsight will often be able to detect trends and facts
that could not have been observed or considered significant
contemporaneously. On the basis that the attitude of the actual underwriter
is irrelevant, one cannot have recourse to what he thought important. The
reinsurer merely has to find a prudent underwriter, looking at his leisure
through the mass of statistics produced on discovery, who finds some
material that is not reflected in the summary and to which he could have
wished to direct his mind in assessing the risk. The experienced reinsurer
has no need of protection. Inducement of the particular insurer was
originally required. That left the insured at the risk of an insurer who
acted unreasonably. To avoid that risk, the Court of Queen's Bench in
Ionides v. Pender (1874) L.R. 9 Q.B. 531 added an additional requirement: a
non-disclosure was only material if it would have been as relevant to a
prudent underwriter as it was to an actual underwriter. [Reference was made
to Seaman v. Fonereau (1743) 2 Str. 1183; Rickards v. Murdock(1830) 10 B. &
C. 527, 540-541; Elton v. Larkins (1832) 5 C. & P. 385, 392; Harrower v.
Hutchinson (1870) L.R. 5 Q.B. 584, 590 and Henry Brooke Q.C. "Materiality in
insurance contracts" [1985] L.M.C.L.Q. 437, 439-440.] For the distinction
between the objective test of materiality and the subjective test of
inducement, see Tate & Sons v. Hyslop (1885) 15 Q.B.D. 368, 379. [Reference
was also made to Elton v. Larkins, 5 C. & P. 385, 392; Traill v. Baring
(1864) 4 De G.J. & S. 318, 326, 330;

Commonwealth Insurance Co. of Vancouver v. Groupe Sprinks S.A. [1983] 1
Lloyd's Rep. 67, 78 and Arnould on Marine Insurance, 8th ed. (1909), vol. I,
pp. 693-696, sections 554-555; Arnould, Law of Marine Insurance and
Average,16th ed. (1981), vol. II, pp. 439, 462-465, paras. 582, 611.]

The authorities in the period from Ionides v. Pender to 1906 support rather
than detract from the "different decision" test: see Stribley v. Imperial
Marine Insurance Co. (1876) 1 Q.B.D. 507, 512, 514, 514-515. [Reference was
also made to Laing v. Union Marine Insurance Co. (1895) 1 Com.Cas. 11,
17-18.]

As to section 18(2) of the Act of 1906, "judgment" primarily, if not
exclusively, means the actual decision, as opposed to the decision-making
process, and "influence" means changing the decision from what it would
otherwise be. Of seven meanings of "judgment" given in the Oxford English
Dictionary, only one refers to the decision-making process.

As regards authorities after 1906, the weight of authority also supports the
different decision test: see London General Omnibus Co. Ltd. v. Holloway
[1912] 2 K.B. 72, 77; Cantiere Meccanico Brindisino v. Janson[1912] 3 K.B.
452, 467; Shanly v. Allied Traders' Insurance Co. Ltd.(1925) 21 Ll.L.R. 195,
197; Mutual Life Insurance Co. of New York v. Ontario Metal Products Co.
Ltd. [1925] A.C. 344, 351-352; Southern Cross Assurance Co. Ltd. v.
Australian Provincial Assurance Association Ltd. (1939) 39 S.R. (N.S.W.)
174, 187-188; Gauvremont v. Prudential Insurance Co. of America [1941]
S.C.R. 139, 157; Zurich General Accident and Liability Insurance Co. Ltd. v.
Morrison [1942] 2 K.B. 53, 58, 60 and Commonwealth Insurance Co. of
Vancouver v. Groupe Sprinks S.A.[1983] 1 Lloyd's Rep. 67. [Reference was
also made to Welford and Otter-Barry, The Law Relating to Fire Insurance,
4th ed. (1948), p. 140, note (k); Traill v. Baring (1864) 4 De G.J. & S.
318, 326, 329; 4 Giff. 485, 490; Rivaz v. Gerussi Brothers & Co. (1880) 6
Q.B.D. 222, 227, 229; Tate & Sons v. Hyslop, 15 Q.B.D. 368, 376, 378, 379;
Scottish Shire Line Ltd. v. London and Provincial Marine and General
Insurance Co. Ltd.[1912] 3 K.B. 51, 70-71; Glicksman v. Lancashire and
General Assurance Co. Ltd. [1927] A.C. 139, 143; Mayne Nickless Ltd. v.
Pegler [1974] 1 N.S.W.L.R. 228, 238; Marene Knitting Mills Pty. Ltd. v.
Greater Pacific General Insurance Ltd. [1976] 2 Lloyd's Rep. 631, 642; Sun
Mutual Insurance Co. v. Ocean Insurance Co. (1882) 107 U.S. 485, 510;
M'Cartney v. Laverty, 1968 S.C. 207, 213; Hooper v. Royal London General
Insurance Co. Ltd., 1993 S.L.T. 679, 684, 688, 689, 692; Western Australian
Insurance Co. Ltd. v. Dayton (1924) 35 C.L.R. 355; Barclay Holdings
(Australia) Pty. Ltd. v. British National Insurance Co. Ltd. (1987) 8
N.S.W.L.R. 514, 515, 517, 517-519; A. T. Scotford, "Chipping Away at C.T.I.
v. Oceanus" [1988] L.M.C.L.Q. 30; and Mutual and Federal Insurance Co. Ltd.
v. Oudtschoorn Municipality, 1985 (1) S.A. 419, 444.]

The actual influence test is consistent with the language and policy of the
Act. The Act does not compel an approach in which, in the absence of waiver,
the assured must perform a duty of disclosure judged only by reference to
the prudent insurer and irrespective of circumstances affecting the actual
insurer. Section 18(2) should not be interpreted as an exhaustive definition
of the grounds for avoidance by reason of non-disclosu see

Berger v. Pollock [1973] 2 Lloyd's Rep. 442, 463 and Visscherij Maatschappij
Nieuw Onderneming v. Scottish Metropolitan Assurance Co. Ltd. (1921) 27
Com.Cas. 198, 212, 215. The authorities directly against the actual
influence test rely only on an obiter dictum of MacKinnon L.J. in Zurich
General Accident and Liability Insurance Co. Ltd. v. Morrison[1942] 2 K.B.
53, 60. They should be overruled.

If the effect of the Act of 1906 is that, in marine cases, there is now a
right to avoid notwithstanding that the actual insurer was not influenced by
the non-disclosure or misrepresentation, that effect must be limited to
marine cases. It is apparent from section 2 that the Act was intended to be
not a codification of the common law of insurance but rather a codification
of the common law of marine insurance. Further, it was not perceived as a
codification of the common law of insurance by contemporaneous authorities:
see Joel v. Law Union and Crown Insurance Co. [1908] 2 K.B. 863, 884. Marine
insurance is in general treated differently from non-marine insurance. In
non-marine cases there was originally no right to avoid or terminate the
contract except for fraudulent misrepresentation: see Anderson v. Fitzgerald
(1853) 4 H.L.Cas. 484, 504. In marine cases, by contrast, a right to "avoid"
or terminate the contract for innocent misrepresentation arose by way of an
implied condition of the contract that no material facts had been
misrepresented: see Blackburn, Low & Co. v. Vigors(1886) 17 Q.B.D. 553,
561-562, 578, 583. The positions were assimilated after the Judicature Acts
but the different history illustrates the different considerations arising
in different types of insurance.

The authorities prior to the Act of 1906 required the insurer to have been
induced by the non-disclosure. Lambert v. Co-operative Insurance Society
Ltd. [1975] 2 Lloyd's Rep. 485, which held that the definition of
"materiality" in non-marine insurance was the same as that in marine
insurance, was wrong because, when it was decided, "materiality" had a
narrower meaning and inducement was a necessary condition of the right to
avoid: see Mutual Life Insurance Co. of New York v. Ontario Metal Products
Co. Ltd. [1925] A.C. 344. Following the C.T.I. case, the law became too
favourable to the insurer in respect of the right to avoid for material
non-disclosure or misrepresentation, and the result of defining materiality
in the same way for both marine and non-marine insurance in cases subsequent
to that decision would be to exacerbate that imbalance and injustice.
Lambert should, therefore, be overruled on this point. [Reference was made
to Chalmers and Owen, Digest of the Law of Marine Insurance, 1st ed. (1901),
pp. 22-23; 2nd ed. (1903), pp. 24-25; footnote 3 to article 20(1); Arnould
on the Law of Marine Insurance, 6th ed. (1887) vol. 1, pp. 519, 526, 548;
Halsbury's Laws of England, 1st ed., vol. 17 (1911), p. 414, para. 809; 2nd
ed., vol. 18 (1935), p. 408, para. 586 and Arnould, 8th ed. (1909), pp.
693-696, sections 554-555, pp. 729-730, section 590.]

The Court of Appeal were wrong to hold that undisclosed information may be
material even though the information actually disclosed was such that no
prudent underwriter would have accepted the risk. If the prudent underwriter
would not have accepted the risk in any event, then any information could
not have had any relevant effect on his judgment; still

less could it be said that he would have reached a different decision if the
additional information had been disclosed.

As to the "errors and omissions clause" (article XV of the treaty), there is
no reason in principle why parties should not include a clause that deprives
an insurer or reinsurer of the right to avoid for inadvertent material
non-disclosure or misrepresentation. Article XV is on its wording such a
clause.

Adrian Hamilton Q.C., Timothy Saloman Q.C. and Simon Picken for the
defendants. The bedrock principle that has informed the law of insurance and
reinsurance, whether marine or non-marine, for over 200 years is that
contracts are based on the utmost good faith: see Carter v. Boehm,3 Burr.
1905. That this principle is mutual has long been clear but was recently
confirmed in Banque Keyser Ullman S.A. v. Skandia (U.K.) Insurance Co. Ltd.
[1991] 2 A.C. 249.

The law of materiality has three main features. (i) The yardstick of the
"materiality" of a circumstance that has not been disclosed, or has been
misrepresented, prior to the conclusion of the contract of insurance is the
notional influence on the judgment of a prudent underwriter. It is not the
actual influence on the judgment of the actual underwriter. Parliament so
declared in sections 18 and 20 of the Act of 1906. (ii) In order to be
"material," the non-disclosed or misrepresented circumstance must be one
that, if disclosed or properly represented, the prudent insurer would want
to take into account when reaching his decision whether or not to accept the
risk and, if so, on what terms. (iii) The consequence of material
non-disclosure or misrepresentation prior to the conclusion of a contract of
insurance or reinsurance is that the insurer may avoid the contract. These
principles were correctly upheld in the C.T.I. case [1984] 1 Lloyd's Rep.
476 and are supported by relevant legislation and an overwhelming body of
consistent, and eminent, judicial authority. Fulfilment of the duty of
disclosure does not depend on nice judgments by the broker as to the minimum
disclosure he can get away with. Otherwise, the "utmost good faith"
principle would be eroded, and against such erosion the courts have
consistently set their face: see Bates v. Hewitt (1867) L.R. 2 Q.B. 595,
604-611; Becker v. Marshall (1922) 12 Ll.R.R. 413; Greenhill v. Federal
Insurance Co. Ltd. [1927] 1 K.B. 65, 72, 76, 85-86; the C.T.I.case, pp.
490-492, 496, 510-511, 529 and Redgrave v. Hurd, 20 Ch.D. 1, 13, 22-23. The
relevant inquiry is as to whether the relevant facts were fairly disclosed
at the time when the insurance was underwritten.

It has never been the law of insurance that the insurer's right of avoidance
for non-disclosure or misrepresentation is dependent on proof that the
insurer's mind was affected or that the insurer's final decision would have
been different. The rationale of the insurer's statutory right of avoidance
is: (a) the uberrimae fidei nature of a contract of insurance, deriving from
the speculative nature of the contract and the law's odium of unequal
information in such contracts (see Carter v. Boehm, 3 Burr. 1905, 1909-1910;
Park, A System of the Law of Marine Insurances, 7th ed. (1817) vol. 1, pp.
283, 287 and Greenhill v. Federal Insurance Co. Ltd. [1927] 1 K.B. 65, 76-77
and, perhaps, (b) an implied condition of the contract that unless the facts
substantially correspond with those represented the insurer will not be
liable on the policy (see Park,p. 291

and Phillips, A Treatise on the Law of Insurance, 5th ed. (1867), vol. I,
pp. 278-279). This rationale underlies section 20(1) of the Act of 1906.
Reason, and the practicalities of proof, are against a proposition that the
representee must show what his conduct would have been if the risk had been
presented to him differently, such conduct being essentially speculative:
see Traill v. Baring, 4 De G.J. & S. 318, 330.

Arnould on Marine Insurance, 6th ed., vol. I, p. 548 does not support the
plaintiffs: there is no suggestion that inducement is necessary. [Reference
was also made to Chalmers and Owen, Digest of the Law of Marine
Insurance,1st ed., p. 22.] Flinn v. Headlam (1829) 9 B. & C. 693 does not
assist at all. The relevant principles are too well rooted for an additional
requirement, namely that of inducement, to be put in except by legislation.
[Reference was made to Zurich General Accident and Liability Insurance Co.
Ltd. v. Morrison [1942] 2 K.B. 53; Cantiere Meccanico Brindisino v. Janson
[1912] 3 K.B. 452, 459-460, footnoted to Arnould, 16th ed., vol. 2, pp.
462-465, para. 611; Arnould, 6th ed., vol. I, pp. 513-514, 532-533; 7th ed.
(1901) vol. I, p. 694 and 8th ed., vol. I, pp. 694-696, section 555.]

Neither the law nor reason or principle supports any distinction or
difference between the test for avoidance for non-disclosure in cases of
non-marine insurance from that applicable in marine insurance cases. Where
Parliaments (English and Commonwealth) have desired to impose, in specific
contexts, a different test for avoidance, they have done so by enacting
terms different from those of section 18 of the Act of 1906: see section
10(3) and (5) of the Road Traffic Act 1934; section 156 of the Ontario
Insurance Act 1914; Mutual Life Insurance Co. of New York v. Ontario Metal
Products Co. Ltd. [1925] A.C. 344 and Marene Knitting Mills Pty. Ltd. v.
Greater Pacific General Insurance Ltd. [1976] 2 Lloyd's Rep. 631, 642.
[Reference was made to Seaman v. Fonereau, 2 Str. 1183; Lynch v.
Dunsford(1811) 14 East 494, 497-498; Rickards v. Murdock, 10 B. & C. 527;
Elton v. Larkins, 5 C. & P. 385; Bates v. Hewitt, L.R. 2 Q.B. 595, 608;
Harrower v. Hutchinson, L.R. 5 Q.B. 584; Ionides v. Pender,L.R. 9 Q.B. 531,
537-539; Parsons on Marine Insurance (1868), vol. I, pp. 494-495; Stribley
v. Imperial Marine Insurance Co., 1 Q.B.D. 507, 512, 514; Rivaz v. Gerussi
Brothers & Co., 6 Q.B.D. 222, 227, 229; Phillips, A Treatise on the Law of
Insurance, 5th ed., vol. I, p. 277; Chalmers and Owen, Digest of the Law of
Marine Insurance,1st ed., p. 22; 2nd ed., pp. 24-25; Morrison v. Universal
Marine Insurance Co. (1873) L.R. 8 Ex. 197; Laing v. Union Marine Insurance
Co.,1 Com.Cas. 11; In re an arbitration between Marshall and Scottish
Employers' Liability and General Insurance Co. Ltd. (1901) 85 L.T. 757,
758;London General Omnibus Co. Ltd. v. Holloway [1912] 2 K.B. 73, 85-86;
Scottish Shire Line Ltd. v. London and Provincial Marine and General
Insurance Co. Ltd. [1912] 3 K.B. 51, 70; Cantiere Meccanico Brindisino v.
Janson [1912] 3 K.B. 452; Associated Oil Carriers Ltd. v. Union Insurance
Society of Canton Ltd. [1917] 2 K.B. 184, 191; Hamilton & Co. v. Eagle, Star
& British Dominions Insurance Co. Ltd. (1924) 19 Ll.L.R. 242; Mutual Life
Insurance Co. of New York v. Ontario Metal Products Co. Ltd. [1925] A.C.
344, 345, 350-352; Glicksman v. Lancashire and General Assurance Co. Ltd.
[1927] A.C. 139; Dunn v. Ocean Accident & Guarantee Corporation Ltd. (1933)
45 Ll.L.R. 276; 47 Ll.L.R.

129; Merchants' and Manufacturers' Insurance Co. Ltd. v. Davies [1938] 1
K.B. 196; Godfrey v. Britannic Assurance Co. Ltd. [1963] 2 Lloyd's Rep. 515,
528-529; Babatsikos v. Car Owners' Mutual Insurance Co. Ltd.[1970] 2 Lloyd's
Rep. 314, 325; Barclay Holdings (Australia) Pty. Ltd. v. British National
Insurance Co. Ltd., 8 N.S.W.L.R. 514, 517; M'Cartney v. Laverty, 1968 S.C.
207; Berger v. Pollock [1973] 2 Lloyd's Rep. 442; Highlands Insurance Co. v.
Continental Insurance Co. (Note) [1987] 1 Lloyd's Rep. 109, 113-114, 117 and
Banque Keyser Ullmann S.A. v. Skandia (U.K.) Insurance Co. Ltd. [1990] 1
Q.B. 665, 777-778.] Wherever the test is formulated in Tate & Sons v.
Hyslop, 15 Q.B.D. 368, 371, 375-377, 379, 381, it is a "consideration" or
"take into account" test. Chalmers' Digest, sections 18(1)(2) and 20(1)(2),
reflects the authorities cited: see also the directions to the jury in
Bridges v. Hunter (1813) 1 M. & S. 15, 16, 18-19. The duty is to disclose
all facts that might influence a reasonable insurer in making his judgment.
It is a strict liability: it does not matter that it is an accidental
omission that does not affect the result. There is nothing in the
authorities to show that there is an extra requirement that the actual
underwriter must have been induced. The test was correctly stated in Mayne
Nickless Ltd. v. Pegler [1974] 1 N.S.W.L.R. 228, 239A-C and approved by Lord
Fraser of Tullybelton in Marene Knitting Mills Pty. Ltd. v. Greater Pacific
General Assurance Ltd. [1976] 2 Lloyd's Rep. 631, 642. The plaintiffs'
contention that the test is the impact on the actual underwriter and/or the
impact on both the actual underwriter and the prudent underwriter is
inconsistent with the terms of sections 18 and 20 of the Act of 1906. Either
the Act codified the law or it contained a correct summary of it. Any change
in the law is, accordingly, for Parliament alone. Yet impetus for
legislative change has focused on the different issue of the all-or-nothing
remedy of avoidance conferred by sections 18(1) and 20(1) of the Act of 1906
and not on the prudent underwriter test: see the judgment of Sir Donald
Nicholls V.-C. and Law Commission Report No. 104, "Insurance Law:
Non-Disclosure and Breach of Warranty" (1980) (Cmnd. 8064), Part IV, paras.
4.4-4.17 (the proportionality principle); paras. 4.24-4.28 (consequences of
breach), cf. paras. 4.43-4.53 (duty of disclosure). It is no more
appropriate to change or subvert the "prudent insurer" test than the
"prudent insured" test applicable to the insured's right of avoidance. The
duty of utmost good faith lying on the insurer, deriving from the obligation
uberrimae fidei that falls on both parties, extends "to disclosing all facts
known to him which are material either to the nature of the risk sought to
be covered or [to] the recoverability of a claim under the policy which a
prudent insured would take into account in deciding whether or not to place
the risk for which he seeks cover with that insurer:" see the Banque Keyser
Ullmann case [1990] 1 Q.B. 665, 769F-772G; [1991] 2 A.C. 249, 268F-H,
281G-282A. Yet changing the relevant law as expressed in the C.T.I.case
[1984] 1 Lloyd's Rep. 476 in the manner sought by the plaintiffs would
introduce a novel and irrational distinction between the test for avoidance
by the insurer and the test for avoidance by the insured, a distinction
dependent solely on who (the insured or the insurer) is seeking to avoid.

The law, as it has developed since 1766 (Carter v. Boehm), is clear to the
effect that the "materiality" of a circumstance does not require the

risk itself to be changed or affected provided that the non-disclosed, or
misrepresented, circumstance "induced a confidence, without which the
[underwriter] would not have acted:" Sibbald v. Hill (1814) 2 Dow 263, 267.
The argument that a circumstance, in order to be material, must be such as
would add to the risk was revived in Rivaz v. Gerussi Brothers & Co. on the
back of supportive statements in Duer, Law and Practice of Marine Insurance,
vol. II (1846), pp. 388 and 518, but was firmly rejected by the Court of
Appeal. It would be far-fetched and impracticable if the law were to impose
an inquiry as to what the prudent insurer's "final decision" would have been
on the basis of the totality of circumstances before the actual insurer.
Such an inquiry would be far too complex: see the C.T.I. case [1984] 1
Lloyd's Rep. 476, 511, 529. [Reference was made to Spencer Bower, The Law
relating to Actionable Non-Disclosure,1st ed. (1915), pp. 11-12, 16-17,
paras. 24, 30-32; 2nd ed. (1990), pp. 32-37, 45-48, paras. 3.07, 3.08, 3.09,
3.14.] The C.T.I. case did not represent any change in the law: see Dr.
Malcolm Clarke, "Insurance Contracts and Non-Disclosure" [1993] L.M.C.L.Q.
297.

As to whether, if the prudent underwriter would not have written the risk on
the terms on which it was written by the actual underwriter on the basis of
the facts actually disclosed, that precludes further information not
disclosed being treated as material (see per Steyn L.J. [1993] 1 Lloyd's
Rep. 496, 506), the plaintiffs' proposition whereby, if there were
"sufficient" material facts known to the actual underwriter to justify a
prudent underwriter declining the risk, then a further undisclosed fact
(which would otherwise have been material) cannot be material is illogical,
contrary to the law and (as Steyn L.J. held) contrary to good sense. As a
matter of logic, there is no pecking order in undisclosed material
information such that the law only recognises "sufficient" material
information to justify a prudent insurer's decision to decline the risk and
ignores other undisclosed circumstances, however material. Indeed, there is
no natural or logical method by which a prudent insurer (or the courts) may
extrapolate, and treat as relevant, one piece of "material" undisclosed
information while excluding others. Logic dictates that every material
circumstance be taken into account and that none be subjected to an
arbitrary process of de-selection. The law follows logic, because the law
and section 18(1) of the Act of 1906 are clear in requiring that the assured
disclose to the insurer every material circumstance known to him, failing
which the insurer may avoid the contract. Not only does the law and good
sense require the materiality of every particular undisclosed circumstance
to be considered; so too does the principle of the utmost good faith
(section 17 of the Act of 1906). The disclosure to which the plaintiffs'
proposition would lend encouragement is selective disclosure inconsistent
with the principle stated.

As to whether, if the relevant impact, or a relevant impact, is that on the
prudent underwriter, and the prudent underwriter would not on the
information that was actually disclosed have written the risk at all or on
anything like the terms offered, the court must, when assessing the
materiality of the circumstance, assume that the prudent underwriter had
indeed accepted the risk on the terms accepted by the actual underwriter,
the materiality of each particular undisclosed circumstance must be

assessed for its notional effect, if any, on the judgment of a prudent
underwriter. On the evidence, the non-disclosed additional losses in the
short record were material in themselves, independently of the materiality
of the long record. There is no evidence that this insurance would not have
been written at all: it would have been written on different terms.

Beloff Q.C. in reply. As a matter of language, if one proffers information
one discloses it. "Disclose" means to put at the underwriter's disposal: see
Bates v. Hewitt, L.R. 2 Q.B. 595, 605. The idea of disclosure does not
incorporate the idea that the profferee actually accepts, copies or reads
the information. The point of disclosure in insurance is to make the
information available so that the underwriter can make of it what he will.
It is his decision what to read. At the minimum, the long record was
proffered.

If the first test propounded by Carter v. Boehm, 3 Burr. 1905 and the Latin
summary, at p. 1910 ("is in his interest to know") is reflected in the
notion of "need to know," the question posed: why does an underwriter need
to know something that would not make him (if it were known) reach a
decision different from that which he would reach without it? It is no
answer to say: it is for the underwriter to assess the significance of the
non-disclosed information. Such answer misses the point. The different
decision test does not, in contradiction of the case law (see, e.g., Bates
v. Hewitt) make the assured the arbiter of materiality. It makes the prudent
insurer the arbiter of materiality and then asks, separately, what is
material vis-...-vis him. Nor is it conclusive that the obligation is
imposed on the assured at the time when the contract is made. Whether or not
he would assess or produce what the underwriter would consider is likely to
alter his decision to enter into the policy and on what terms; he should
only be vulnerable to avoidance if his assessment or production leads him to
omit matters that would have that consequence. As to policy, the draconian
nature of the remedy requires the most stringent test (if the language of
the Act of 1906 permits). [Reference was made to Dr. Malcolm Clarke,
"Insurance Contracts and Non-Disclosure" [1993] L.M.C.L.Q. 297, 298, 299;
Clason v. Smith (1811) 3 Wash.C.C. 156; Chalmers and Owen, Digest of the Law
of Marine Insurance, 2nd ed., pp. 24-25, s. 18(1) and Arnould on Marine
Insurance, 6th ed., vol. I, p. 548.]

As to actual inducement, the Court of Appeal in Cantiere Meccanico
Brindisino v. Janson [1912] 3 K.B. 452 did not expressly approve Scrutton
J.'s ruling and decided the appeal on the basis that there had been no
misrepresentation at all: see pp. 461-462. Principle tells in favour of the
actual inducement test. It is common ground that actual inducement is a
prerequisite for avoidance of an ordinary contract for even fraudulent
misrepresentation. It would be peculiar if such a prerequisite were not
required for such avoidance of an insurance contract for even non-fraudulent
misrepresentation, a fortiori for non-disclosure. The defendants argue that
the absence of such requirement stems from the special nature of the
contract of insurance as one of uberrima fides. But the by-product of the
uberrimae fidei nature of the insurance contract is the obligation, not
attendant on other ordinary contracts, to make disclosure of material
matters. There is no reason to ascribe to it any further effect than that.
Why even under a contract uberrimae fidei should a party be able to

avoid for a misrepresentation or non-disclosure that ex hypothesi was
immaterial to him?

Nor does history provide a justification for the defendants' argument. It
appears common ground that the concept of the reasonable underwriter was
introduced into jurisprudence in Ionides v. Pender, L.R. 9 Q.B. 531 in order
to protect the assured against the idiosyncratic underwriter. To achieve
that end substitution, as distinct from addition, is not required. Before
Ionides materiality was tested against the actual underwriter. No plausible
explanation has been given as to why introduction of the reasonable
underwriter test should necessarily, or did, extrude the actual underwriter
from the overall equation: see D. St. L. Kelly (1988) 1 Ins.L.J. 30, 31-35.

Ionides v. Pender, at pp. 537-538, suggests that the test of actual
inducement remains, but in any event its subsistence was not in issue in
that case. Works of high authority support the actual inducement test:
Arnould, A Treatise on the Law of Marine Insurance and Average, 2nd ed.
(1857), vol. I, pp. 541, 565-567, 584-585; Arthur Cohen K.C. in Halsbury's
Laws of England, 1st ed., vol. 17, p. 414, note (p), Arnould,6th ed., vol.
I, pp. 513-514, 519-520 (especially), note 1, pp. 520, 548; Arnould, 8th
ed., vol. I, p. 662, note (b); pp. 640-643; pp. 693-694, sections 554-555,
notes (x), (z); p. 729, section 590; Arnould,16th ed. (1981), p. 438, para.
580, pp. 462-465, para. 611, p. 475, para. 627; Flinn v. Headlam, 9 B. & C.
693; Duer, The Law and Practice of Marine Insurance, vol. II, pp. 389 et
seq.; Parsons, A Treatise on the Law of Marine Insurance and General Average
(1868), vol. 1, p. 495 and Clason v. Smith. Spencer Bower, Turner & Sutton,
Actionable Non-Disclosure, 2nd ed. (1990), pp. 11-12, 16-17 must be
discarded on this point. As regards the 6th edition of Arnould, the
defendants have not referred the House of Lords to the important parts on
which Chalmers relied in his Digest. Section 17, 1st ed., p. 21, note 4,
refers to pp. 513 and 548. Section 18(1), p. 22, note 3 refers to p. 548,
which deals with non-disclosure, and requires (Chalmers) influence on the
actual underwriter and requires that influence to be decisive. Section
20(1), p. 26, note 5, refers to pp. 519-520, which deal with
misrepresentation, and directly to a dictum of Willes J. in Anderson v.
Pacific Fire and Marine Insurance Co. (1872) L.R. 7 C.P. 65, 68, quoted by
Arnould, 6th ed., vol. I, p. 520, note 1. In other words, again, by
reference to the dictum of Willes J. directly and via the Arnould footnote,
Chalmers requires influence on the actual underwriter and requires the
influence to be decisive. The non-disclosure and misrepresentation tests are
the same under the Act of 1906.

Their Lordships took time for consideration.

25 July.

PANEL: Lord Templeman, Lord Goff of Chieveley,Lord Mustill, Lord Slynn of
Hadley andLord Lloyd of Berwick

JUDGMENTBY-1: LORD TEMPLEMAN

JUDGMENT-1:
LORD TEMPLEMAN: My Lords, I have read the speech to be delivered by my noble
and learned friend, Lord Lloyd of Berwick, and I agree with his reasons and
his conclusions.

Prior to the judgments of the Court of Appeal in Container Transport
International Inc. v. Oceanus Mutual Underwriting Association (Bermuda) Ltd.
[1984] 1 Lloyd's Rep. 476 ("the C.T.I. case") the duty of disclosure imposed
by the common law and statute seems to have been well defined

in principle, albeit that the application of principle to individual facts
produces familiar conflicts of expert evidence which the court must resolve.
When insurance is under negotiation, the underwriter must decide whether to
accept the proffered risk and if so on what terms. In particular, the
underwriter must decide the amount of the premium which he considers an
appropriate consideration for the risk accepted. If a material fact is
undisclosed by the insured, the insurer may avoid the insurance contract. An
undisclosed fact is material if disclosure would have affected the
acceptance of the risk or the rate of premium. Materiality must be judged by
the reactions of a prudent insurer, otherwise the actual underwriter could,
after the risk has matured, convince himself and the court that he would
have rejected the risk or increased the premium if full disclosure had been
made in the course of the negotiations. It must be assumed that, failing
disclosure, the prudent insurer would have been willing to accept the risk
at the premium actually negotiated. The question then is whether full
disclosure would have caused the prudent insurer to resile from the
negotiations or increase the premium.

The authorities cited in the speech of Lord Lloyd and the citations of other
authorities by counsel establish the common law rule which was embodied in
section 18 of the Marine Insurance Act 1906 in these terms:

"(2) Every circumstance is material which would influence the judgment of a
prudent insurer in fixing the premium, or determining whether he will take
the risk."

In my opinion "the judgment of a prudent insurer" cannot be said to be
"influenced" by a circumstance which, if disclosed, would not have affected
acceptance of the risk or the amount of the premium. On behalf of the
underwriters, Mr. Hamilton submitted that a circumstance was material if a
prudent insurer would have "wanted to know" or would have "taken into
account" that circumstance even though it would have made no difference to
his acceptance of the risk or the amount of premium. If this is the result
of the judgments of the Court of Appeal in the C.T.I. case then I must
disapprove of that case. If accepted, this submission would give carte
blanche to the avoidance of insurance contracts on vague grounds of
non-disclosure supported by vague evidence even though disclosure would not
have made any difference. If an expert says, "If I had known I would not
have accepted the risk or I would have demanded a higher premium," his
evidence can be evaluated against other insurances accepted by him and
against other insurances accepted by other insurers. But if the expert says,
"I would have wanted to know but the knowledge would not have made any
difference" then there are no objective or rational grounds upon which this
statement of belief can be tested. The law is already sufficiently tender to
insurers who seek to avoid contracts for innocent non-disclosure and it is
not unfair to require insurers to show that they have suffered as a result
of non-disclosure. Of course they suffer if the risk matures but that is the
risk accepted by every insurer.

In the present case the insured failed to disclose that the claims received
in respect of the 1981 insurance amounted to U.S.$468,168 and not
U.S.$235,768. No juggling with figures can obscure the clear indication,
found by the judge, that the true 1981 figures were more

alarming than the disclosed figures and would have caused the prudent
insurer in 1982 to reject the risk or increase the premium; the insurer is
therefore entitled to avoid the policy.

I would dismiss this appeal.

JUDGMENTBY-2: LORD GOFF OF CHIEVELEY

JUDGMENT-2:
LORD GOFF OF CHIEVELEY: My Lords, I have had the opportunity of reading in
draft the speeches prepared by my noble and learned friends, Lord Mustill
and Lord Lloyd of Berwick. Like them, I, too, would dismiss the appeal.

Underlying the appeal before your Lordships' House have been two questions
of principle, of great importance to the law of insurance. The first relates
to the test of materiality in cases of non-disclosure, which in the law of
marine insurance is to be found in section 18(2) of the Marine Insurance Act
1906. There it is provided that:

"Every circumstance is material which would influence the judgment of a
prudent insurer in fixing the premium, or determining whether he will take
the risk."

Here the question for your Lordships is whether, as the appellant plaintiffs
("Pan Atlantic") have contended, it must be shown that full and accurate
disclosure would have led the prudent insurer either to reject the risk or
at least to have accepted it on more onerous terms. This has been called the
"decisive influence test." The second question is whether, for an insurer to
be entitled to avoid a policy for misrepresentation or non-disclosure, it is
enough that the misrepresentation or non-disclosure was material, or whether
in addition it must, as Pan Atlantic have contended, have induced the making
of the policy on the relevant terms. This has been called the "actual
inducement test."

I turn first to the second of these questions. Like both of my noble and
learned friends, I have come to the conclusion that, on this question, Mr.
Beloff's submission on behalf of Pan Atlantic should be accepted; in other
words, I accept that the actual inducement test accurately represents the
law. I do so for the reasons given by my noble and learned friend, Lord
Mustill. Like him, and for the reasons he gives, I conclude that there is to
be implied in the Act of 1906 a requirement that a material
misrepresentation will only entitle the insurer to avoid the policy if it
induced the making of the contract; and that a similar conclusion must be
reached in the case of a material non-disclosure. This conclusion is, as I
understand it, consistent with the opinion expressed by my noble and learned
friend, Lord Lloyd, that Parliament, by enacting the law as it did in
section 20 of the Act of 1906, must have intended to codify the common law
on materiality without touching the common law on inducement.

I turn next to the first question, which is whether the decisive influence
test is the appropriate test for deciding whether a fact which has not been
disclosed is a material fact. Here there is a difference of opinion between
my two noble and learned friends, Lord Lloyd accepting the decisive
influence test and Lord Mustill rejecting it.

On this point, I respectfully prefer the reasoning of Lord Mustill. I do so
for the following reasons.

First it seems to me, as it does to Lord Mustill, that the words in section
18(2) "would influence the judgment of a prudent insurer in . . .

determining whether he will take the risk" denote no more than an effect on
the mind of the insurer in weighing up the risk. The subsection does not
require that the circumstance in question should have a decisive influence
on the judgment of the insurer; and I, for my part, can see no basis for
reading this requirement into the subsection.

Second, in agreement with my noble and learned friend, and with both Parker
and Stephenson L.JJ. in the C.T.I. case (Container Transport International
Inc. v. Oceanus Mutual Underwriting Association (Bermuda) Ltd. [1984] 1
Lloyd's Rep. 476, 511-512, 526-527), it seems to me that the decisive
influence test faces insuperable practical difficulties, because it ignores
the fact that it is the duty of the assured to disclose every material
circumstance which is known to him, with the result that the question of
materiality has to be considered by the assured before he enters into the
contract. At that time, it is not unreasonable to expect that an assured who
is aware of, and understands, his duty of disclosure should be able to
identify those circumstances, within his knowledge, which would have an
impact on the mind of the insurer when considering whether to accept the
risk and, if so, on what terms he should do so; but it appears to me to be
unrealistic to expect him to be able to identify a particular circumstance
which would have a decisive effect. Likewise it seems to me, in agreement
with Parker L.J., that an inquiry after the event as to whether the judgment
of a prudent insurer would have been decisively influenced by the relevant
circumstance, if disclosed, would in many cases be impracticable, because
this must in the nature of things depend upon the reactions of the
particular underwriter.

For these reasons in particular, and for the other reasons given by my noble
and learned friend, Lord Mustill, I would reject the decisive influence
test. I do not propose myself to review the authorities, or to examine the
learned writings, on this subject. This is because I find myself to be in
complete agreement with the analysis of my noble and learned friend. In
particular I am, like him, convinced that the origin of the test of
influence on the judgment is to be found in the writings of learned authors
on the subject, which show no indication of the decisive influence test for
which Pan Atlantic now contend - indeed the indications are to the contrary.
In the end, as it seems to me, your Lordships are at liberty to give to the
definition of materiality in section 18(2), and indeed to that in section
20(2), an interpretation which accords with the natural and ordinary meaning
of the words used, the underlying obligation of good faith and the
practicalities of the situation - all of which are, in my opinion,
inconsistent with the decisive influence test.

I wish to add that there is, to my mind, danger in considering the two
questions now before your Lordships in watertight compartments because, as I
see it, the conclusion that actual inducement by the actual underwriter is
necessary before he can avoid the contract for non-disclosure has an impact
upon the question of materiality. If actual inducement is not required,
materiality becomes all important, because it is the sole requirement for
entitling the insurer to avoid the contract on the ground of non-disclosure.
It was, I believe, because it was thought, in the C.T.I. case and
subsequently, that actual inducement was not required, that critics of the
decision in that case promoted the idea that the test of

materiality should be hardened into the decisive influence test by
introducing into the concept of materiality something in the nature of
inducement, though attributing it not to the actual underwriter but to the
hypothetical prudent insurer. But once it is recognised that actual
inducement of the actual underwriter is required, the pressure to take any
such step disappears, and the idea of introducing any such requirement into
the concept of materiality can be perceived to be not merely unnecessary but
inappropriate. The result is that, as I have said, the definition of
materiality in section 18(2) of the Act can be given its natural and
ordinary meaning which, for the reasons I have given, is also the sensible
meaning.

In the result, I wish to express my agreement with the conclusions expressed
in the last two paragraphs of part IV, and the two short propositions set
out at the end of part V, of my noble and learned friend's speech; but, for
the reasons given by him, I would dismiss the appeal.

JUDGMENTBY-3: LORD MUSTILL

JUDGMENT-3:
LORD MUSTILL: My Lords, the two short questions upon which this appeal
depends should, after more than 200 years of history, be capable of a ready
answer. The controversy which they have aroused shows that they are not.
Although the issues arise under a policy of non-marine insurance it is
convenient to state them by reference to the Marine Insurance Act 1906 since
it has been accepted in argument, and is indeed laid down in several
authorities, that in relevant respects the common law relating to the two
types of insurance is the same, and that the Act embodies a partial
codification of the common law.

The relevant sections of the Act read:

"17. A contract of marine insurance is a contract based upon the utmost good
faith, and, if the utmost good faith be not observed by either party, the
contract may be avoided by the other party.

"18(1) Subject to the provisions of this section, the assured must disclose
to the insurer, before the contract is concluded, every material
circumstance which is known to the assured, and the assured is deemed to
know every circumstance which, in the ordinary course of business, ought to
be known by him. If the assured fails to make such disclosure, the insurer
may avoid the contract. (2) Every circumstance is material which would
influence the judgment of a prudent insurer in fixing the premium, or
determining whether he will take the risk."

"20(1) Every material representation made by the assured or his agent to the
insurer during the negotiations for the contract, and before the contract is
concluded, must be true. If it be untrue the insurer may avoid the contract.
(2) A representation is material which would influence the judgment of a
prudent insurer in fixing the premium, or determining whether he will take
the risk."

"91(2) The rules of the common law including the law merchant, save in so
far as they are inconsistent with the express provisions of this Act, shall
continue to apply to contracts of marine insurance."

I. The history in outline

The questions arise in this way. Between 1977 and 1982 Pan Atlantic
Insurance Co. Ltd. (hereafter "Pan Atlantic") wrote a quantity of direct

American liability insurance. Much of this was "long-tail" business, in
which a long period of time may elapse before claims mature. More recent
experience has shown that such business can involve the insurer in
disastrous losses. The present case is no exception. The disputes arise
under a contract of reinsurance relative to the policy year 1982 made
between Pan Atlantic and Pine Top Insurance Co. Ltd. ("Pine Top"). In
relation to the years 1977 to 1979 Pan Atlantic's casualty account had been
reinsured with other insurers for excess of loss above a certain figure.
Pine Top became the reinsurer in respect of this cover for the 1980 policy
year and later renewed for two subsequent years. Disputes subsequently arose
in relation to all three years. Judgment was given in favour of Pan Atlantic
in an action brought under the first two treaties. An appeal by Pine Top was
dismissed and the position as regards 1980 and 1981 is no longer in suit.
The present appeal is concerned only with the claim under the treaty for the
year 1982.

The broking history of this treaty was summarised by Steyn L.J., in one of
the judgments now under appeal, as follows [1993] 1 Lloyd's Rep. 496,
499-500:

"The broker who acted on behalf of Pan Atlantic was Mr. Robinson of Butcher
Robinson and Staples Ltd. The underwriter was Mr. O'Keefe. On 22 or 23
December 1981 Mr. O'Keefe gave a quotation. On 13 January 1982 Mr. O'Keefe
signed the slip on behalf of Pine Top. Pine Top took a line of 50 per cent.
on the slip. Guildhall Insurance Co. Ltd. signed for the remaining 50 per
cent. Guildhall never sought to avoid their liability under the 1982 treaty
and therefore drops out of the picture.

"The structure of the renewal for 1982 must be briefly explained. For the
1981 year the rate of premium was 10 per cent. When it came to the renewal
for 1982 Pan Atlantic wanted the rate reduced to 71/2 per cent. That could
only be done by introducing an aggregate deductible of U.S.$225,000. That
was the way in which the slip was then written and signed.

"The main features of the slip were as follows. First, it covered the
liability of the reinsured under: 'all policies and/or contracts and/or
binders of insurance and for reinsurance . . . allocated to their so-called
casualty account . . .' Secondly, the treaty was to pay U.S.$75,000 excess
of U.S.$25,000 each and every loss on an ultimate net loss basis. Thirdly,
it provided for an aggregate deductible of U.S.$225,000. Fourthly, the rate
of premium was 71/2 per cent. The slip is, of course, in commercial terms
the primary evidence of the bargain struck between the broker and
underwriter. In the usual way treaty wording was prepared by clerical staff
long after the contract was made. It was agreed on 29 October 1992. . . .
Article XV reads as follows: 'It is hereby declared and agreed that any
inadvertent delays, omissions or errors made in connection with this
reinsurance shall not be held to relieve either of the parties hereto from
any liability which would have attached to them hereunder if such delay,
omission or error had not been made, provided rectification be made upon
discovery, and it is further agreed that in all things coming

within the scope of this reinsurance the reinsurer shall share to the extent
of their interest the fortunes of the reinsured.'

"The basis of Pine Top's purported avoidance of the reinsurance contract for
1982 was the presentation made by Mr. Robinson to Mr. O'Keefe. The
undisputed evidence was that Mr. Robinson was an experienced and respected
broker. Mr. O'Keefe was an underwriter with some four years' experience.
They knew one another. They had negotiated the 1981 contract. Mr. Robinson
and Mr. O'Keefe met on 22 or 23 December 1981 and again on 13 January 1982
when Mr. O'Keefe signed the slip. The discussions covered the reduced
premium and the introduction of the aggregate deductible. There was some
discussion of the loss record, and on 13 January 1982 Mr. Robinson said to
Mr. O'Keefe that he had done 'a quick update' and there was little movement
in the figures. Mr. Robinson had available for Mr. O'Keefe's inspection two
documents, respectively called the short record and the long record. The
short record contained only the record for the years 1980 and 1981. The long
record contained the record for the 1977 to 1979 period when Pine Top was
not on risk as well as the record for the 1980 and 1981 years when Pine Top
was reinsurer. The record for the 1977 to 1979 period was so bad that it was
eventually common ground at the trial that no prudent underwriter would have
signed the slip for 1982 on the terms which Mr. O'Keefe accepted. A major
issue at the trial was whether there was a fair presentation in respect of
the loss record for the 1977/78 and 1979 underwriting years. The judge found
that, although to Mr. O'Keefe's knowledge, both the short record and the
long record were available for inspection, Mr. Robinson presented the risk
in a way which diverted Mr. O'Keefe's attention from examining the loss
records for the underwriting years 1977/78 and 1979. The loss record for
1980 and 1981 was undoubtedly disclosed. But the disclosed record was
incomplete." - The minor inaccuracy as regards 1980 need not be explored. -
"There was, however . . . a more important inaccuracy in respect of the loss
record for 1981. As against the disclosed losses of U.S.$235,768 for the
underwriting year 1981 it is common ground that the true losses for 1981
were U.S.$468,168. It was common ground that Pan Atlantic had information
about these additional losses available before the slip was signed on 13
January 1982 and that the additional losses were not disclosed."

Within a few years it became plain that the treaty thus broked was incurring
disastrous losses. As with the two earlier years Pine Top repudiated
liability. Pan Atlantic commenced proceedings, and by its defence Pine Top
sought to avoid the treaty on the ground of non-disclosure of the 1977 to
1979 loss record and related misrepresentations. In January 1991, less than
four weeks before the start of the trial, Pine Top amended to add a defence
that the 1980 to 1981 loss record as disclosed was incomplete and that this
constituted a further material misrepresentation or non-disclosure.

At the conclusion of the trial Waller J. [1992] 1 Lloyd's Rep. 101 gave
judgment in favour of Pine Top. He rejected the ground of defence originally
relied on, holding that a prudent underwriter would have wanted to check the
record for 1977 to 1979 because with long-tail insurance, where the claims
record is slow to develop, the history for the older years is a much more
reliable guide than the results for recent years; but that the broker had a
summary of the 1977-79 years available, which itself was perfectly fair, and
which the underwriter, if he was doing his job, knowing it was the most
material information, decided not to examine. The judge did however uphold
the defence based on the understatement of the losses for the year 1981. The
Court of Appeal (Sir Donald Nicholls V.-C., Farquharson and Steyn L.JJ.)
dismissed an appeal by Pan Atlantic, expressing regret at being compelled to
take that course.

II. The questions of law

On these facts two questions of law arise for decision. 1. Where sections
18(2) and 20(2) of the Act relate the test of materiality to a circumstance
"which would influence the judgment of a prudent underwriter in fixing the
premium, or determining whether he will take the risk," must it be shown
that full and accurate disclosure would have led the prudent underwriter to
a different decision on accepting or rating the risk; or is a lesser
standard of impact on the mind of the prudent underwriter sufficient; and,
if so, what is that lesser standard? 2. Is the establishment of a material
misrepresentation or non-disclosure sufficient to enable the underwriter to
avoid the policy; or is it also necessary that the misrepresentation or
non-disclosure has induced the making of the policy, either at all or on the
terms on which it was made? If the latter, where lies the burden of proof?

A further question was raised in argument concerning the effect of article
XV of the treaty. This has no bearing on the fundamental issues just stated,
and I will leave it aside for the time being.

Before summarising the views expressed on these issues in the courts below I
must describe the previous decision of the Court of Appeal in Container
Transport International Inc. v. Oceanus Mutual Underwriting Association
(Bermuda) Ltd. [1984] 1 Lloyds Rep. 476 (hereafter "the C.T.I.case"). Unlike
the present, this was a case of marine insurance and therefore directly
governed by the Act. The facts are of no present concern; once again, the
dispute arose in relation to allegations that the previous claims record and
insurance history had not been properly disclosed during the broking of a
reinsurance treaty. At the trial the judge (Lloyd J.) [1982] 2 Lloyd's Rep.
178, 187-188 expressed his opinion on the law as follows:

"In general I would say that underwriters ought only to succeed on a defence
of non-disclosure if they can satisfy the court by evidence or otherwise
that a prudent insurer, if he had known the fact in question, would have
declined the risk altogether or charged a higher premium. It seems to me
that this should be the general rule, if only because the defence under
section 18 is capable of working such great hardship on the assured. Take a
case where the fact is known to the

assured, but not the materiality of the fact. Suppose that the prudent
insurer, if he had known the fact, would have accepted the risk, but charged
a small additional premium; suppose further that there is a substantial
claim under the policy. In other jurisdictions, the assured could enforce
the claim, by tendering the additional premium. But not so in England. The
fairness of the English rule is not at once obvious and hardly seems to
reflect the duty of utmost good faith under section 17 which, be it noted,
is owed both ways. Why, if the insurer would have accepted the risk in any
event, albeit at an increased premium, should he be able to avoid the claim
altogether? Since the English law is so favourable to the underwriter in
this respect, the least that should normally be expected of the underwriter
is to show that a prudent insurer would have charged an increased rate. But
the language of section 18 goes wider than that. It is the judgment of the
prudent insurer which has to be influenced, not the rate; and I can imagine
cases where the prudent insurer could say: 'Yes, this would have affected my
judgment; but I would nevertheless have charged the same rate because of
counterbalancing factors.' Such cases, though rare, would, I think, come
within the definition of materiality; but even in such cases the defendant
must prove that the judgment of the prudent insurer would in fact have been
affected, not that it might have been affected. It can never be enough for
the prudent insurer to say 'Yes, I would have liked to know this or that
fact, so that I could have made up my mind what to do about it.'"

In the Court of Appeal three extensive judgments were delivered. It is
impossible to compress these in a manner which does them justice and I will
therefore do no more than quote some of the most important passages. First,
Kerr L.J., at pp. 491-492:

"The judge in effect equates 'judgment' with 'final decision,' as though the
wording of these provisions had been 'would induce a prudent underwriter to
fix a different premium or to decline the risk.' Thus, he said, at p. 187:
'In general I would say that underwriters ought only to succeed on a defence
of non-disclosure if they can satisfy the court by evidence or otherwise
that a prudent insurer, if he had known the fact in question, would have
declined the risk altogether or charged a higher premium.' And at p. 188 he
said: 'some "difference of action" is required in order to establish
materiality. The mind of the reasonable insurer must have been influenced so
as to induce him to refuse the risk or alter the premium.' Finally in this
context, he said, at p. 189: 'normally, at any rate, insurers must show that
the result would have been affected. Moral hazard cases, to which I shall
have to refer later, may be an exception.'

"This interpretation differs crucially from what I have always understood to
be the law and from the interpretation which [an expert witness] adopted in
much of his evidence. [The witness] interpreted section 18(2) in the sense
that 'judgment' referred to the assessment or evaluation of the risk. Apart
from other passages in his evidence, this appears from the following extract
at the beginning. He had

made a written report stating his views on all the allegations of
non-disclosure, which was treated as part of his evidence. He was then
referred to section 18(2), and there was the following exchange between him
and counsel: '[Q] In what sense do you use the word material in your report?
[A] In that sense. In the sense that the facts must have an influence on the
underwriting judgment.' In my view this is the correct approach . . ."

A little later, Kerr L.J. said:

"The word 'influenced' means that the disclosure is one which would have had
an impact on the formalities of his opinion and on his decision making
process in relation to the matters covered by section 18(2). . . . evidence
to support the materiality of the undisclosed circumstance, from this point
of view, is therefore often given by an independent expert witness whose
evidence has to be assessed by the court long after the event. He, or the
actual insurer, or both, may then be asked: 'What would have been your
reaction if you had known of this undisclosed fact?' Both would in my view
give relevant evidence on materiality if they replied: 'I would then have
regarded the risk in a different light. I would have taken this circumstance
into account. As a first step I might have asked some questions before
making up my mind about the risk. What my final decision would have been, I
cannot now say for certain. I might have declined the risk altogether, or
increased the premium, or altered the terms in some other way.' And it would
make no difference if the witness went on: 'I might even have taken a
chance, or given credit for the frankness of the disclosure, by writing the
risk as I did. But I should obviously have been told about this fact before
being asked to make up my mind.'"

Finally, after detailed consideration of various reported cases Kerr L.J.
expressed his opinion thus, at pp. 496-497:

"It follows that when sections 17 to 20 of the Act are read together, one
way of formulating the test as to the duty of disclosure and representation
to cases such as the present, in the context of Mr. Fleetwood's presentation
first to Mr. Bragg and then to Mr. Lee, is simply to ask oneself: 'Having
regard to all the circumstances known or deemed to be known to the insured
and to his broker, and ignoring those which are expressly excepted from the
duty of disclosure, was the presentation in summary form to the underwriter
a fair and substantially accurate presentation of the risk proposed for
insurance, so that a prudent insurer could form a proper judgment - either
on the presentation alone or by asking questions if he was sufficiently put
on inquiry and wanted to know further details - whether or not to accept the
proposal, and, if so, on what terms?' This is not an onerous duty for
brokers to discharge in practice."

These passages all relate to the first of the questions posed for
consideration in the present appeal. As to the second question, in a
previous decision at first instance, Berger v. Pollock [1973] 2 Lloyd's Rep.
442, 463, Kerr J. had stated the principles in a way which suggested

that the insurer could avoid the policy only if he had in fact been
influenced by the misrepresentation or non-disclosure. In the C.T.I. case,
after more extensive citation of the authorities, particularly Zurich
General Accident and Liability Insurance Co. Ltd. v. Morrison [1942] 2 K.B.
53, Kerr L.J. concluded [1984] Lloyd's Rep. 476, 495 that he had been wrong
in Berger v. Pollockand that it was not necessary to prove that the mind of
the actual insurer had been affected.

The next judgment, that of Parker L.J., began with the second question and
concluded, at p. 510, after a review of the authorities, including in
particular Marene Knitting Mills Pty. Ltd. v. Greater Pacific General
Insurance Ltd. [1976] 2 Lloyd's Rep. 631, and an examination of the wording
of the Act, that there was no requirement that the particular insurer should
have been induced to take the risk or charge a lower premium than he would
otherwise have done. On the first question Parker L.J. said, at pp. 510-511:

"A circumstance which increases the risk would, as it seems to me, clearly
influence the judgment of a prudent insurer. It would be one amongst perhaps
many factors exerting an influence, be it great or small, upon his judgment
towards declining the risk, or charging a higher rather than a lower
premium, or inserting some protective condition in the policy. In the same
way a circumstance which diminishes the risk would be one amongst perhaps
many factors exerting an influence the other way, albeit disclosure of
circumstances which diminish the risk is excused under section 18(3)(a). The
decision whether to take the risk and if so at what premium, must in each
case be a matter of the prudent insurer exercising his judgment upon some
factors influencing that judgment one way and some factors influencing it
another. Whether the influence of a circumstance which increases the risk
would or would not be great enough to cause a prudent insurer to decline the
risk or move from one level of premium to another must inevitably vary as
between prudent insurers. Whether it would or not, if it is a circumstance
'bearing' on the risk (Parsons, A Treatise on the Law of Marine Insurance
and General Average (1868), vol. I. p. 495) or having a 'tendency' towards
declining the risk or raising the premium (Phillips, A Treatise on the Law
of Insurance, 5th ed. (1867), vol. I, sections 531, 571, 676), it appears to
me that it would influence the judgment of the prudent insurer in fixing the
premium or in determining whether to take the risk.

"Indeed no other construction appears to me to accord with either common
sense or practical reality. The test submitted on behalf of the respondents
[plaintiffs] would involve the court in the task, perhaps years after the
event, of endeavouring to ascertain what a prudent underwriter would have
done, first in the light of the circumstances actually disclosed by the
assured, and secondly, on the hypothesis that, in addition to those
circumstances, the undisclosed circumstance had been disclosed. Such a task
is on its face impractical. Five experienced and prudent underwriters might
be called. At stage 1, one might say he would not have taken the risk even
on the facts disclosed; the other four might all have taken the

risk but at different premiums. At stage 2 the four remaining might all say
'we regard the fact as significantly increasing the risk' but one might say
'not, but by the narrowest margin, sufficiently to demand a change in
premium, but it would call for a change in the policy wording,' one
'sufficiently to put up the premium,' and the last, 'sufficiently to decline
the risk.' Furthermore the one who would not have taken the risk in the
first place might say that he would, had the additional fact been disclosed,
have regarded it as an additional reason for declining the risk. In such
circumstances what is the court to do? It cannot, as it seems to me, choose
one prudent underwriter rather than another. The very choice of a prudent
underwriter as the yardstick in my view indicates that the test intended was
one which could sensibly be answered in relation to prudent underwriters in
general. It is possible to say that prudent underwriters in general would
consider a particular circumstance as bearing on the risk and exercising an
influence on their judgment towards declining the risk or loading the
premium. It is not possible to say, save in extreme cases, that prudent
underwriters in general would have acted differently, because there is no
absolute standard by which they would have acted in the first place or as to
the precise weight they would give to the undisclosed circumstance."

The third member of the court, Stephenson L.J., concluded as follows after a
comprehensive analysis of the authorities, at pp. 526-527:

"What is naturally meant by a circumstance 'which would influence the
judgment of a prudent insurer in fixing the premium, or determining whether
he will take the risk?' There is a danger in dissecting a phrase into its
component parts, taking it word by word, and asking what each word means.
Does 'would' mean 'probably would' or 'possibly might?' Does 'influence'
mean 'alter?' Does 'judgment' mean 'judging,' the formation of opinion, or
'judgment,' the final decision? The judge may have been asked to decide
between 'would' and 'might;' we are asked to decide between the decision
making process of judging and the decision itself. If I did only that, I
would feel more strongly the force of what I take to be the judge's opinion
that a circumstance taken into account by a prudent insurer in the process
of considering the risk offered and the premium and the terms required, but
disregarded in the conclusion to accept the risk at a particular premium and
on particular terms, is not a circumstance which could properly be said to
influence the judgment of an insurer, prudent or imprudent. But two
considerations prevent me from adopting his construction of the words.

"The first, stressed by my brethren, is the practical difficulty, if not
impossibility, of deciding what factors would affect the result of a
hypothetical prudent insurer's consideration of a risk, whether to accept it
and on what terms; whereas there is no great difficulty in answering the
question whether any particular factor would be one which he would want to
know and take into consideration in determining whether to accept a risk and
on what terms, without having to decide whether he would ultimately
disregard it altogether

or give it much or little weight. The second consideration is the overriding
duty of utmost good faith imposed by section 17. That duty seems to require
full disclosure and full disclosure seems to require disclosure of
everything material to the prudent underwriter's estimate of the character
and degree of the risk; and how can that be limited to what can
affirmatively be found to be a circumstance which would in fact alter a
hypothetical insurer's decision? Provided that there is some information
which a prudent insurer would obviously want to know, or which a credible
expert swears he would want to know, in considering an offer of a risk, that
is a material circumstance which the greatest good faith and the rule
against concealment require the assured or his agent to disclose, subject to
the qualifications which the knowledge and conduct of the insurer or his
agent may put upon the assured's duty."

And later, at p. 529:

"I conclude from the language of the subsections in their context and from
the authorities that everything is material to which a prudent insurer, if
he were in the proposed insurer's place, would wish to direct his mind in
the course of considering the proposed insurance with a view to deciding
whether to take it up and on what terms, including premium. His mind would,
I think, be influenced in the process of judging whether to do so, either
temporarily where he can say that he would ultimately have reached the same
decision without it, or permanently where it would have led him to reach a
different decision. The difficulties of discriminating between those
considered facts which influence and those which do not is well illustrated
JEB Fasteners Ltd. v. Marks, Bloom & Co. [1983] 1 All E.R. 583, and I do not
think that these subsections require an insured to do so. He is only to be
trusted by the proposed insurer if he discloses everything a prudent insurer
would want to consider before fixing the premium or determining to take the
risk."

Turning to the present case both the trial judge and the Court of Appeal
were bound by the C.T.I. case and therefore had no reason to explore the
principles at length. The trial judge [1992] 1 Lloyd's Rep. 101, 103
directed himself that the law laid down in the C.T.I. case was as follows:

"any circumstance is material, i.e. is one which would influence the
judgment of a prudent insurer in fixing the premium or determining whether
he will take the risk, if it is a circumstance which: 'would have had an
impact on the formation of his opinion and on his decision making process.'
That is to say 'judgment' was equal to 'formation of opinion' rather than
the 'final decision.' The case also made clear that the test in relation to
non-disclosure or misrepresentation was influence on the judgment of a
'prudent insurer' and that thus the right to avoid did not depend on whether
the particular insurer was influenced as a fact in relation to determining
the premium he charged or in his decision whether or not to take the risk."

In the Court of Appeal [1993] 1 Lloyd's Rep. 496 we find the court striving,
through the medium of the principal judgment delivered by Steyn L.J., to
find a workable understanding of the ratio of the C.T.I. case [1984] 1
Lloyd's Rep. 476 which was consistent not only with the rejection of
decisive influence as the test for materiality but also with the rejection
of any requirement of influence on the actions of the individual
underwriter. It may well be that but for this second constraint the court
might have felt more free in its ruling on materiality. At all events this
is what Steyn L.J. proposed, at pp. 505-506:

"Having rejected the 'decisive influence' construction, it seems to me that
there were at least two feasible alternative solutions to be considered in
[the C.T.I. case]. The first solution was that a fact is material if a
prudent insurer would have wished to be aware of it in reaching his
decision. The second solution involves taking account of the fact that
avoidance for non-disclosure is the remedy provided by law because the risk
presented is different from the true risk. But for the non-disclosure the
prudent underwriter would have appreciated that it was a different and
increased risk. Approaching the matter in this way it is possible to say
that the test is whether a prudent underwriter, if he had known the
undisclosed facts, would have regarded the risk as increased beyond what was
disclosed on the actual presentation. . . .

"In my view we are free to choose between the two solutions. As between the
two alternative solutions, I unhesitatingly choose the second solution. In
other words, I would rule that, as the law now stands, the question is
whether the prudent insurer would view the undisclosed material as probably
tending to increase the risk. That does not mean that it is necessary to
prove that the underwriter would have taken a different decision about the
acceptance of the risk. After all, there may be many commercial reasons for
still writing the risk on the same terms. But if the concept of 'influence'
is interpreted in accordance with the second solution, it seems to me that
it is easier to fit the C.T.I. . . . decision within the framework of our
insurance law and it results in a somewhat fairer and more balanced
principle of materiality as between insured and insurer. And the
difficulties which this decision has caused in practice will be considerably
ameliorated."

In a brief concurring judgment, with whose sentiments Steyn L.J. agreed, Sir
Donald Nicholls V.-C. expressed his unease at the consequence to which
inadvertent non-disclosure had led in this case, for the result was that the
reinsurer had avoided all liability for his own bad bargain and, moreover,
had done so even though full disclosure would have resulted, not in his
declining to take the risk, but only in an increased premium. Justice and
fairness would suggest that when the inadvertent non-disclosure came to
light what was required was an adjustment in the premium or, perhaps, in the
amount of cover. But those were not options available under English law. The
remedy was all or nothing. In the opinion of the Vice-Chancellor the present
case was an unhappy example

of a case where, in the absence of a discretion, the law did not produce a
satisfactory result. Farquharson L.J. agreed with both judgments.

III. Criticisms of the C.T.I. case

In substance this is an appeal against the decision in the C.T.I. case. In
his judgment [1973] 1 Lloyd's Rep. 496, 505, Steyn L.J. said quite bluntly
that C.T.I. had proved to be a remarkably unpopular decision not only in the
legal profession but also in the insurance markets. Whether this
generalisation about the markets is correct I cannot judge, but the books
and articles produced in argument all adopt a critical stance. Nevertheless,
although the unanimous disapprobation of the C.T.I.case is striking, equally
striking is the lack of unanimity about what exactly is wrong with it. Space
does not permit a full discussion of the diverse criticisms. The following
appear to be the principal complaints.

(1) The law is too harsh, for it deprives the assured of a recovery for a
genuine loss by perils insured against even if the misrepresentation or
non-disclosure had no bearing on the risk which brought about the loss.
There is practical force in this objection, but it is not consistent with
general principle, for the vice of misrepresentation and non-disclosure is
not that after the event the underwriter has suffered from having taken on a
parcel of risks one of which led to a loss, but that a breach of the duty of
good faith has led the underwriter to approach the proposal on a false
basis. This distinction need not be explored here, for it has long been
established that the absence of a connection between the misrepresentation
or non-disclosure and the peril which caused the loss is no ground for
depriving the underwriter of the right to avoid (for an early decision see
Seaman v. Fonereau (1743) 2 Str. 1183); and this is nowadays an inevitable
consequence of sections 18 and 20 of the Act. The Court of Appeal in the
C.T.I. case could not have ruled otherwise even if invited to do so, which
it was not.

(2) The law is too harsh, for it deprives the assured of the whole of his
recovery even if full and accurate disclosure would have done no more than
cause the actual underwriter, or the hypothetical prudent underwriter, to
insist on one rate of premium rather than another. The inflexibility of an
"all-or-nothing" rule has been present to the minds of all the courts which
have heard these two cases, as the judgments of Kerr L.J. and Sir Donald
Nicholls V.-C. clearly demonstrate. It has been fully ventilated before your
Lordships, and I acknowledge the attractions of a solution which involves an
element of "proportionality." Whether such a solution would be practicable
outside the field of consumer insurance is debatable: as witness the adverse
conclusion of the Law Commission of England and Wales in Report, Insurance
Law: Non-Disclosure and Breach of Warranty (1980) (Law Com. No. 104) (Cmnd.
8064). It need not, however, be debated here. As early as 1808 it was stated
in Marshall, A Treatise on the Law of Insurance, 2nd ed., vol. I, p. 463:
"Nor can the insured, by tendering any increase of premium, require the
insurer to confirm the contract;" and there has never subsequently been any
suggestion that an intermediate solution of this kind was the common law.
Moreover, the words of the Act are plainly inconsistent with any such rule.
It may be that the question of a statutory change is due for reconsideration
in the

light of the last 20 years' experience, but this is not an area in which the
courts have any freedom of choice.

(3) The law fails to take account of whether a reasonable person seeking
insurance would appreciate that a particular circumstance was material and
ought to be disclosed. Again, there is force in this submission, at least as
regards those consumer cases where there is an imbalance of expertise and
experience between the proposer and the insurer. The position is however
quite different in a case like the present where the considerations which
weighed heavily with Kirby P. in Barclay Holdings (Australia) Pty. Ltd. v.
British National Insurance Co. Ltd. (1987) 8 N.S.W.L.R. 514 are wholly
absent. The assured here was an insurance company acting through an
experienced broker. The performance of the latter in the episode of the long
record shows that these were no shorn lambs who needed the winds of the
common law rule to be tempered. The broker knew very well what he was doing,
and took care about how he did it. But this is beside the point. The House
has not been, and could not be, invited to introduce a wholly new doctrine,
hinging upon what was, or could have been, or should have been, in the mind
of the proposer. In the field of marine insurance this would require a
fundamental amendment of the Act, and in commercial insurance as a whole
such a wholesale change to a central and long-established first principle of
insurance law could not have been made by the Court of Appeal in the C.T.I.
case any more than it can now be made by this House.

(4) The doctrine of the C.T.I. case demands more of the assured than is
feasible in modern trading conditions. This is the kind of criticism which
it is hard for a court, and particularly for an appellate court, to assess.
I would, however, make the following brief comments upon it. First, I
believe that a substantial part of the criticisms, to the effect that the
broker in order to play safe will be forced to disclose hundreds of that
speedy placing of risks which is such a positive feature of the London
market, are based on an interpretation of Kerr L.J.'s pronouncements in the
C.T.I. case which is wider than the Lord Justice intended. Secondly,
although the physical bulk of placing material is likely in modern times to
have been swollen by photocopies, electronically transmitted documents and
computer printouts there will, I believe, be many cases where the core of
material of which good faith demands the disclosure is relatively small and
easy to identify. The present case is a good example. Finally, some of the
critics come close to saying that the central obligation of good faith and
its embodiment in the Act are out of date in modern conditions. This was not
an option open to the court in the C.T.I. case, or to any other court.
Undoubtedly, commercial law must be responsive to changes in commercial
practices if it is not to founder, and established principles must be
applied sensitively in new situations. Thus, once the court has reached a
conclusion on the true content of the obligations created by the Act, in the
light of any relevant previous decisions, it must translate them into
practice by reference to conditions prevailing, not in 1906, but at the time
when the risk was written. But it was not for the Court of Appeal,

any more than it is for this House, to alter the meaning of the statute.
Only Parliament can do that.

(5) The effect of the C.T.I. case has been to deter overseas interests from
placing risks in the London market. Again, it is not possible to judge the
factual accuracy of this complaint. The comment is however obvious that if
overseas interests take business elsewhere because English law insists that
they and their brokers make fair presentations in good faith this may be
business which the London market can well do without; and there is no need
to emphasise at the present time the dangers of judging the success of an
insurance market by volume alone. Moreover, whilst I accept that if that
good quality business is being driven away there is reason to look carefully
at whether the rules are being properly applied, if the rules established by
Act of Parliament are having a deleterious economic effect it is for
Parliament, not the courts, to change them.

Thus far, I have summarised and briefly discussed various of the criticisms
to show that, although they have not been overlooked, they do not point
towards a solution of the problems now before the House. The literature does
however also develop in considerable detail a number of other groups of
criticism which are directly in point.

(6) The Court of Appeal in the C.T.I. case set the standard of materiality
too low. The law ought to be that a circumstance is material only if its
disclosure would decisively have influenced the mind of the prudent
underwriter: if it would have made all the difference to whether he wrote
the risk, and if so at what premium. Alternatively, even if a circumstance
can be material without being decisive, the law ought to require a greater
potential effect on the mind of the hypothetical underwriter than was
acknowledged in the C.T.I. case.

(7) The decision in the C.T.I. case that a defence of misrepresentation or
non-disclosure can succeed even if the actual underwriter's mind was
unaffected is contrary to common sense and justice. Moreover, the rule is
not correct in principle, since (1) the juristic basis of the underwriter's
ability to disclaim the policy is that the misrepresentation or
non-disclosure vitiates the consent necessary for a binding contract, and
consent cannot be vitiated if the underwriter would have made the same
contract even if the circumstance in question had been properly disclosed;
and (2) to dispense with the requirement for inducement of the contract is
inconsistent with the general law on misrepresentation.

(8) If the actual underwriter would not have been influenced by the
information it cannot have been material, and hence the assured was under no
duty to disclose it.

(9) The court in the C.T.I. case failed to appreciate the importance of
Ionides v. Pender (1874) L.R. 9 Q.B. 531, and associated cases.

All of these criticisms were developed in the argument for the plaintiffs on
the present appeal, to which I now turn.

IV. Materiality

This part of the case depends on the words "which would influence the
judgment of a prudent insurer in fixing the premium, or determining whether
he will take the risk."

The main thrust of the argument for Pan Atlantic is that this expression
calls for the disclosure only of such circumstances as would, if disclosed
to the hypothetical prudent underwriter, have caused him to decline the risk
or charge an increased premium. I am unable to accept this argument.

In the first place I cannot find the suggested meaning in the words of the
Act. This is a short point of interpretation, and does not yield to long
discussion. For my part I entirely accept that part of the argument for Pan
Atlantic which fastens on the word "would" and contrasts it with words such
as "might." I agree that this word looks to a consequence which, within the
area of uncertainty created by the civil standard of proof, is definite
rather than speculative. But this is only part of the inquiry. The next step
is to decide what kind of effect the disclosure would have. This is defined
by the expression "influence the judgment of a prudent insurer." The
legislature might here have said "decisivelyinfluence," or "conclusively
influence," or "determine the decision," or all sorts of similar
expressions, in which case Pan Atlantic's argument would be right. But the
legislature has not done this, and has instead left the word "influence"
unadorned. It therefore bears its ordinary meaning, which is not, as it
seems to me, the one for which Pan Atlantic contends. "Influence the
judgment" is not the same as "change the mind." Furthermore, if the argument
is pursued via a purely verbal analysis, it should be observed that the
expression used is "influence the judgment of a prudent insurer in . . .
determining whether he will take the risk." To my mind, this expression
clearly denotes an effect on the thought processes of the insurer in
weighing up the risk, quite different from words which might have been used
but were not, such as "influencing the insurer to take the risk."

My Lords, this conclusion accords with what I regard as the practicalities.
Looking at the matter through the eyes of a court, called upon to rule after
the event on whether an undisclosed circumstance was material, the
proposition that "influence" means "decisively influence" takes as its point
of reference a hypothetical underwriter personifying the generality of those
who know their job and perform it carefully, without exceptional timidity or
boldness; it assumes that this underwriter has had before him all the
material which was before the actual underwriter; it also assumes that after
weighing up the conflicting factors which enter into a decision of this kind
the hypothetical underwriter has decided that the balance comes down in
favour of writing the risk on the terms on which it was actually written;
and then on these assumptions requires a t) have tipped the balance the
other way and caused him to reach a different decision. Even looking at the
matter after the event, this exercise presents great difficulties, for the
reasons given by Parker L.J. in the C.T.I. case [1984] 1 Lloyd's Rep. 476 in
the passage quoted above, at pp. 510-511. But the point is that it is not
the court after the event, but the prospective assured and his broker before
the event, at whom the test is aimed; it is they who have to decide, before
the underwriter has agreed to write the risk, what material they must
disclose. I am bound to say that in all but the most obvious cases the
"decisive influence" test faces them with an almost impossible task. How can
they tell whether the

proper disclosure would turn the scale? By contrast, if all that they have
to consider is whether the materials are such that a prudent underwriter
would take them into account, the test is perfectly workable.

Furthermore, the argument for Pan Atlantic demands an assumption that the
prudent underwriter would have written the risk at the premium actually
agreed on the basis of the disclosure which was actually made. Yet this
assumption is impossible if the actual underwriter, through laziness,
incompetence or a simple error of judgment, has made a bargain which no
prudent underwriter would have made, full disclosure or no full disclosure.
This absurdity does not arise if the duty of disclosure embraces all
materials which would enter into the making of the hypothetical decision,
since this does not require the bargain actually made to be taken as the
starting point.

There is also a logical difficulty, not addressed in argument. When an
underwriter seeks to avoid a policy for non-disclosure and the dispute comes
to court, the arguments will naturally focus on the particular item of
information which has been withheld; and, if the argument for Pan Atlantic
is right, on the question whether if it had been disclosed it would have
made all the difference to the hypothetical underwriter. It is natural that
in an inquiry thus conducted ex post facto little regard will be paid to the
other items of information which should have been disclosed and were
actually disclosed. As already noticed however this forensic exercise misses
the point of the duty to act in good faith, which calls for a decision in
the real commercial world about everythingwhich ought to be disclosed. If
there are (say) six items of information bearing on the risk, it will in
many cases be easy to say that all of them ought to be disclosed. Yet if the
narrower interpretation advanced by Pan Atlantic is right, it would be
necessary for the assured and the broker to decide in advance whether any of
them would in itself be enough to turn the scale, and the answer might
logically be that none of them were. This answer would be absurd.

Accordingly, treating the matter simply as one of statutory interpretation I
would feel little hesitation in rejecting the test of decisive influence.
Since however the opposing view has been persuasively advanced, the study
must be widened to see what light is shed by decisions and writings before
and after 1906.

I will begin with an argument which, if correct, leads directly to a
solution, without the need for close scrutiny of the language used in
judgments delivered before 1906. In its simplest form the argument is as
follows. During the early development of the law the courts were concerned
to assess the impact of the non-disclosure on the mind of the actual
underwriter, and to ask: If the circumstance in question had been disclosed
to the actual underwriter would his decision have been different? Later,
when in cases such as Ionides v. Pender, L.R. 9 Q.B. 531 it was recognised
that the test should relate to the reaction of the hypothetical prudent
underwriter rather than the actual underwriter, there was no reason why the
required degree of impact on the underwriter's mind should have been
changed; and it was not changed. The question is still whether, on balance
of probabilities, it is shown that a prudent underwriter in the position of
the actual underwriter would have acted differently if the circumstance had
been disclosed.

Whilst this argument has the attraction of simplicity, I believe it to be
unsound, for more than one reason. In the first place it founds upon an
equation between materiality and actual effect which was absent even from
the earlier law. This is most clearly seen in regard to misrepresentations.
I must return later to the relationship between misrepresentation in marine
insurance law and misrepresentation in the general law of contract, but the
texts and the relevant cases leave no room for doubt that whereas if the
representation inducing a contract was either fraudulent or a "warranty" of
the contract its falsehood would invariably give a right to avoid an
innocent misrepresentation inducing the contract would give the underwriter
a right to avoid only if it was material. Proof of actual effect was not
necessarily proof of materiality.

The long-standing controversy about whether the converse proposition might
also be true - whether a material misrepresentation or non-disclosure would
vitiate the policy even if it had no effect on the mind of the actual
underwriter - also speaks against the argument now being considered. Judge
Duer (Duer, The Law and Practice of Marine Insurance, vol. II (1846), pp.
680 et seq.) and Professor Parsons (Parsons, A Treatise on the Law of Marine
Insurance and General Average, vol. I, pp. 367-369) believed that it would,
whilst Joseph Arnould and Willard Phillips tended to the opposite view: see
Arnould, A Treatise on the Law of Marine Insurance and Average, 2nd ed.
(1857), vol. I, p. 541 and Phillips, A Treatise on the Law of Insurance, 5th
ed. (1867) vol. I, p. 371. I must consider this question at a later stage,
but for the present it is not the merits of the controversy which matter but
its existence; for it would have been meaningless if a decisive influence on
the actual underwriter was equivalent to materiality.

Another pointer to the distinction between materiality and effect is found
in the list of exceptions contained in section 18(3) of the Act. These
exceptions were recognised by Lord Mansfield in Carter v. Boehm (1766) 3
Burr. 1905 and were therefore well established by the time of Ionides v.
Pender. It is convenient to state them in the form in which they now appear
in the Act:

"18(3) In the absence of inquiry the following circumstances need not be
disclosed, namely:- (a) any circumstance which diminishes the risk; (b) any
circumstance which is known or presumed to be known to the insurer. The
insurer is presumed to know matters of common notoriety or knowledge, and
matters which an insurer in the ordinary course of his business, as such,
ought to know; (c) any circumstance as to which information is waived by the
insurer; (d) any circumstance which it is superfluous to disclose by reason
of any express or implied warranty."

The significance of these exceptions is that they were not written back by
Lord Mansfield into his definition of materiality, but were aimed at the
duty of disclosure and the consequences of failing to perform it. This is
what one would expect. The materiality or otherwise of a circumstance should
be a constant; and the subjective characteristics, actions and knowledge of
the individual underwriter should be relevant only to the

fairness of holding him to the bargain if something objectively material is
not disclosed.

The next objection to the argument is that it first posits a former state of
the law where the effect on the decision of the actual underwriter was the
only test, the concepts of the hypothetical underwriter and the influence
upon his judgment being unknown; and goes on to assert that the innovation
of Ionides v. Pender, L.R. 9 Q.B. 531 was to substitute these concepts as
the decisive test. The first stage of this argument is, I believe,
historically unsound, as will appear from the early works of reference
quoted hereafter. Leaving this altogether aside, I cannot find any support
for the second stage in Ionides v. Pender itself. The facts of the case were
simple. Commissions and other ancillary interests in goods were insured by
the plaintiffs under a marine voyage policy at greatly excessive values. The
vessel sank in circumstances which gave rise to suspicion. At the trial the
jury was unable to reach a conclusion on whether the ship was cast away and
whether the overvaluations were fraudulent, but did find that it was
material to the underwriter to know of the overvaluation and that the fact
of it was concealed. On these facts the sole issue was whether the
non-disclosure was material. This was the culmination of a long-standing
controversy about the nature of the "risk" to which the duty of disclosure
was related. Was the duty of disclosure confined, as Duer had argued, at p.
390, to those facts which affected the intrinsic nature of the risks - i.e.
those which affected the probability that the subject matter would be lost
or damaged by a peril insured against? Or were Phillips and Arnould right to
say that "risk" should be given a wider meaning so that the duty extended to
anything which would probably influence the insurer's ultimate decision,
including what later came to be called the "moral hazard?" The Court of
Queen's Bench decided in favour of the latter view. This was certainly very
important, but the decision was concerned with the kind of risk which was
the subject of the duty to disclose, and not with either the standard
imposed by the duty or the identity of the person by reference to whom the
extent of the duty was to be ascertained; and there is nothing in either the
texts or the decisions cited in argument which bore on these questions at
all. The nearest that can be found in the entire report is the following
passage from the judgment of the court, delivered by Blackburn J., at p.
539:

"We agree that it would be too much to put on the assured the duty of
disclosing everything which might influence the mind of an underwriter.
Business could hardly be carried on if this was required. But the rule laid
down in [ Parsons, A Treatise on the Law of Marine Insurance and General
Average,vol. I], p. 495, that all should be disclosed which would affect the
judgment of a rational underwriter governing himself by the principles and
calculations on which underwriters do in practice act, seems to us a sound
one."

As I read this passage its purpose was the following. The law postulated by
Duer was comparatively simple to apply. One simply looked at the intrinsic
nature of the perils insured against and inquired whether they were enhanced
by the non-disclosure. This was a question which was capable of reasonably
objective ascertainment. The problem

with the wider test, preferred by Parsons and in the event by the court
itself, was that if the duty of disclosure was widened to include all the
additional matters which might influence what an underwriter might think or
do the duty would be at the same time impractically wide and impossible of
ascertainment. The court therefore answered Duer's objection by emphasising
that it was matters which might influence the mind not of "an underwriter"
(Blackburn J.'s words) but of a hypothetical reasonable underwriter, whose
standards of materiality would be at once bounded and possible to fix. This
qualification was an essential part of the court's decision on the
controversy which was the only issue before it, but I cannot find anything
to suggest that the court thereby intended to set the whole law of
disclosure off in an entirely new direction. Nor indeed does it appear that
this is how the case was viewed after the event, for in the 5th edition of
Arnould (1877), published only three years later with the high authority of
Mr. David Maclachlan, the very few references to Ionides v. Pender, L.R. 9
Q.B. 531 give no hint that in the respects material to the present appeal
anything at all remarkable had been decided.

For these reasons I would reject Ionides v. Pender as the key to the
problem, and must look elsewhere to see whether there is anything to suggest
that the interpretation which I have proposed is wrong. Since the
controversy relates to the interpretation of an Act designed to reflect the
existing common law inquiry will naturally begin with the decisions of the
courts before 1906. With one crucial exception, nothing would be gained by
rehearsal of the score of authorities through which the House was taken with
great care during argument, for although it is possible to find expressions
in the various judgments which support one or another of the conflicting
interpretations, examination shows that in none of them did it make any
difference which formula was used. Either the case was not about the extent
of the duty at all, or the outcome would have been the same however the test
was described. Accordingly, without intending disrespect to the diligent
arguments or to the judgments in the C.T.I. case [1984] 1 Lloyd's Rep. 476,
where importance was attached to a number of the older authorities, I shall
not occupy space by analysing them, and will pass immediately to Carter v.
Boehm, 3 Burr. 1905, which not only contained the first and most extended
exposition of the doctrine but was also the starting-point for the opinions
of the notable scholars in England and the United States whose treatment of
the subject has had such a powerful influence on the development of the law.
Lord Mansfield said, at pp. 1909-1911:

"First. Insurance is a contract upon speculation. The special facts, upon
which the contingent chance is to be computed, lie most commonly in the
knowledge of the insured only: the underwriter trusts to his representation,
and proceeds upon confidence that he does not keep back any circumstance in
his knowledge, to mislead the underwriter into a belief that the
circumstance does not exist, and to induce him to estimate the risqu , as if
it did not exist. The keeping back such circumstance is a fraud, and
therefore the policy is void. Although the suppression should happen through
mistake, without any fraudulent intention; yet still the underwriter is
deceived, and the policy is void; because the risqu run is really different
from

the risqu understood and intended to be run, at the time of the agreement.
The policy would equally be void, against the underwriter, if he concealed;
as, if he insured a ship on her voyage, which he privately knew to be
arrived: and an action would lie to recover the premium. The governing
principle is applicable to all contracts and dealings. Good faith forbids
either party, by concealing what he privately knows, to draw the other into
a bargain, from his ignorance of that fact, and his believing the contrary.
But either party may be innocently silent, as to grounds open to both, to
exercise their judgment upon. Aliud est celare; aliud, tacere; neque enim id
est celare quicquid reticeas; sed cum quod tu scias, id ignorare emolumenti
tui causa velis eos, quorum intersit id scire. This definition of
concealment, restrained to the efficient motives and precise subject of any
contract, will generally hold to make it void, in favour of the party misled
by his ignorance of the thing concealed. . . . Men argue differently, from
natural phenomena, and political appearances: they have different
capacities, different degrees of knowledge, and different intelligence. But
the means of information and judging are open to both: each professes to act
from his own skill and sagacity; and therefore neither needs to communicate
to the other. The reason of the rule which obliges parties to disclose, is
to prevent fraud, and to encourage good faith. It is adapted to such facts
as vary the nature of the contract; which one privately knows, and the other
is ignorant of, and has no reason to suspect. The question therefore must
always be 'whether there was, under all the circumstances at the time the
policy was underwritten, a fair representation; or a concealment;
fraudulent, if designed; or, though not designed, varying materially the
object of the policy, and changing the risqu understood to be run.'"

Whilst it is true that this decision has been criticised on the facts, and
that the wide general contractual duty of good faith which Lord Mansfield
propounded has long since ceased to hold sway, the courts have never been
deflected from the high standard of duty prescribed in his judgment. The
assured is not to keep anything back which goes to the computation of the
"contingent chance," for otherwise there is no "fair representation," and
the underwriter is led to approach the "risk understood to be run" on a
false basis. Such is the principle on which insurance law has been developed
and insurance contracts have been made for more than 200 years and I would
do nothing to dilute it now. I can see no room within it for a more lenient
test expressed solely by reference to the decisive effect which the
circumstance would have on the mind of the prudent underwriter.

It may however fairly be objected that it is not sufficient to take Lord
Mansfield's statement and apply it to the facts of the individual case, for
directly in the case of marine insurance, and indirectly in the present
instance, the court must give effect to the words of the Act; and Lord
Mansfield's formulation did not use the word "influence" or refer to the
prudent underwriter. How did these features enter the law?

As regards the test of influence on the mind I feel little doubt that the
origins lay in the writers of the texts. These were a far more potent source
of general principle than the scattered decisions of the courts; Sir
Mackenzie Chalmers' Digest (Chalmers and Owen, A Digest of the Law relating
to Marine Insurance, 1st ed. (1901); 2nd ed. (1903)) shows that, as one
would expect, the draftsman of the Act had them fully in mind; and so of
course did Blackburn J. when he expressed the test in slightly different
terms (Ionides v. Pender, L.R. 9 Q.B. 531, 539). (It may be noted that the
word "influence" does not appear in Ionides v. Pender.) As early as 1808 we
find in Marshall, A Treatise on the Law of Insurance, 2nd ed., vol. I, p.
467:

"Every fact and circumstance, which can possibly influence the mind of any
prudent and intelligent insurer, in determining whether he will underwrite
the policy at all, or at what premium he will underwrite it, is material."

We see the word "influence" employed by Parsons, A Treatise on Maritime Law
(1859), vol. II, p. 155:

"the question as to the materiality of a representation, is not whether the
fact stated actually did or possibly could affect the risk, but whether it
would naturally tend to influence the insurer in his estimate of the risk."

A very similar formulation appears in Phillips, A Treatise on the Law of
Insurance, 5th ed., vol. I, p. 274: "tending to influence [the
underwriter's] estimate of the character and degree of the risk to be
insured against." The word also appears repeatedly in Arnould, A Treatise on
the Law of Marine Insurance and Average, 2nd ed., pp. 565-568. In the 6th
edition of the same work (1887), edited by Mr. Maclachlan, we find at p. 518
a reference to "facts 'tending to induce the underwriter more readily to
assume the risk by diminishing the estimate he would otherwise have formed
of it'" followed by the statement that "Facts, the statement of which may
reasonably be presumed likely to have such an influence on the judgment of
the underwriter are called 'material facts'." In Duer, The Law and Practice
of Marine Insurance, vol. II, we find at p. 382 that the materiality of
facts is expressed in terms of "their probable influence on the estimated
value of the risks."

These references to the influence of the misrepresented or undisclosed fact
on the mind of the underwriter are significant, not so much because they
demonstrate what could safely have been assumed, namely that sections 18 and
20 of the Act reflected a well-established understanding of the law, but
because the same authors who were content to speak of influence also
expressed the test in ways which bear closely upon the present issue. Thus,
as shown by the passages quoted (and there are others which I need not set
out), they employed the qualifying words "tending to," which to my mind are
quite inconsistent with a hard-edged criterion expressed in terms of a
decisive influence on the working of the underwriter's mind. Elsewhere in
these works the authors used expressions other than "influence the mind" to
explain the principle, plainly without any notion that they were saying
something essentially different, and again

the terminology cannot be reconciled with the narrower view of the duty to
disclose. According to Parsons, A Treatise on Maritime Law, the material
fact must be such that it would "naturally and reasonably enter into the
estimate of the risk, or the reasons for or against entering into the
contract of insurance;" and the whole object of the rules as to
representation, misrepresentation and concealment is "to enable the insurers
to judge accurately of the risk they undertake" (pp. 173, 174). So also
Duerspeaks, at p. 390, of facts which a prudent and experienced underwriter
would deem it proper to consider; Arnould, 6th ed., pp. 514 and 518, also
refers to the underwriter's estimate of the risk.

My Lords, these and similar passages do not conclusively establish that the
test of decisive effect is unsound, for the discussions in the books are
long, and short turns of phrase cannot safely be plucked out of context.
Nevertheless, I can see nothing in them to suggest that before 1906
materiality was understood as extending only to such circumstances as would
definitely have changed the underwriter's mind; and they furnish substantial
support for the view that the duty of disclosure extended to all matters
which would have been taken into account by the underwriter when assessing
the risk (i.e. the "speculation") which he was consenting to assume. This is
in my opinion what the Act was intending to convey, and what it actually
says.

Before coming to the cases decided after 1906 I should mention Rivaz v.
Gerussi Brothers & Co. (1880) 6 Q.B.D. 222, which was relied on to suggest
that a test expressed in terms of the underwriter's assessment of the risk
is contrary to authority. I disagree. As clearly appears from the argument
for the defendants (reported at pp. 225-226) the case was concerned with the
same controversy between the views of Duer and Parsons on whether the duty
of disclosure extended beyond matters directly affecting the probability of
a loss by perils insured against, as had previously been resolved in favour
of the wider view in Ionides v. Pender, L.R. 9 Q.B. 531. That the word
"risk" must be understood in the wider sense is now beyond dispute, but this
has no bearing on the principle which I derive from the authoritative texts,
and indeed from both the letter and the spirit of Carter v. Boehm, 3 Burr.
1905, that it is the relevance to the underwriter's intellectual process
when assessing the risk which determines the scope of disclosure.

I pause for a moment to consider the other conspicuous feature of the
earlier law, namely the presence in the equation of the hypothetical prudent
underwriter. Just when and how this feature was added cannot be deduced from
the materials now available, but it is at least as old as 1823 (see
Marshall, A Treatise on the Law of Insurance, 2nd ed., vol. I, p. 467), and
may well be much older. It is a fair assumption that at least one reason
must have been that the principles stated by Lord Mansfield required fair
dealing, and it would have been unfair to the assured to require disclosure
of matters which a reasonable underwriter would not have taken into account.
It is possible that another reason was that until the Evidence Act 1851 (14
& 15 Vict. c. 99) a party to a suit could not give evidence on his own
behalf, so that the actual underwriter could not testify on either
materiality or causation, and any decision on the latter would have to
proceed by assuming that the underwriter was reasonable

and asking what effect the information would have had on a reasonable
underwriter in his situation. It is unnecessary to pursue this aspect
further, and I mention it only to show that the argument based on Ionides v.
Pender, L.R. 9 Q.B. 531 cannot be sound, since it posits a change in the law
which (if it was a change at all) had occurred several decades before.

Returning to the standard of materiality, I pass to the decisions after
1906, of which by far the most important is Mutual Life Insurance Co. of New
York v. Ontario Metal Products Co. Ltd. [1925] A.C. 344. A life assured had
answered questions in a proposal form in a manner which was not strictly
accurate. The policy was governed by the Ontario Insurance Act 1914,
sections 156(3) and (4) of which provided:

"(3) The proposal or application of the assured shall not as against him be
deemed a part of or be considered with the contract of insurance except in
so far as the court may determine that it contains a material
misrepresentation by which the insurer was induced to enter into the
contract. (4) No contract shall be avoided by reason of the inaccuracy of
any such statement [i.e. in an application for a policy] unless it is
material to the contract . . ."

The insurers sought to avoid the policy in reliance on the misstatements in
the proposal form. The trial judge found that the misrepresentations were
not material. The Judicial Committee of the Privy Council upheld that
decision. I must quote from the opinion of the Board, at pp. 350-352, at a
little length:

"The main difference of judicial opinion centres round the question what is
the test of materiality? Mignault J. thought that the test is not what the
insurers would have done but for the misrepresentation or concealment, but
'what any reasonable man would have considered material to tell them when
the questions were put to the insured.' Their Lordships are unable to assent
to this definition. It is the insurers who propound the questions stated in
the application form, and the materiality or otherwise of a
misrepresentation or concealment must be considered in relation to their
acceptance of the risk. On the other hand, it was argued that the test of
materiality is to be determined by reference to the questions; that the
insurance company had by putting the question shown that it was important
for them to know whether the proposer had been in the hands of a medical man
within five years of his application, and, if so, to have had the
opportunity of interviewing such medical man before accepting the risk. The
question was therefore, they contended, a material one, and the failure to
answer it truthfully avoids the contract. Now if this were the true test to
be applied there would be no appreciable difference between a policy of
insurance subject to section 156 of the Ontario Insurance Act, and one in
the form hitherto usual in the United Kingdom. All of the questions may be
presumed to be of importance to the insurer who causes them to be put, and
any inaccuracy, however unimportant in the answers, would, in this view,
avoid the policy. Suppose, for example, that the insured had consulted a
doctor for a headache or a cold on a single occasion and had concealed or
forgotten the fact,

could such a concealment be regarded as material to the contract? Faced with
a difficulty of this kind, the appellants' [defendant company's] counsel
frankly conceded that materiality must always be a question of degree, and
therefore to be determined by the court, and suggested that the test was
whether, if the fact concealed had been disclosed, the insurers would have
acted differently, either by declining the risk at the proposed premium or
at least by delaying consideration of its acceptance until they had
consulted Dr. Fierheller. If the former proposition were established in the
sense that a reasonable insurer would have so acted, materiality would,
their Lordships think, be established, but not in the latter if the
difference of action would have been delay and delay alone. In their view,
it is a question of fact in each case whether, if the matters concealed or
misrepresented had been truly disclosed, they would, on a fair consideration
of the evidence, have influenced a reasonable insurer to decline the risk or
to have stipulated for a higher premium.

"Applying this test, the evidence which has impressed their Lordships most
is that of Dr. McCullough - a witness adduced by the appellants and who, as
their medical examiner in Toronto, was the person by whom they would
naturally be guided in accepting or declining the risk. Now Dr. McCullough
states that if Dr. Fierheller's name had been mentioned, he would have noted
it in the answer to question 18, but he also emphatically states that if he
had known at the time all that Dr. Fierheller deposed to in evidence, he
would still have sent up the case with a recommendation for acceptance. In
other words, having, as the result of his own examination, passed Mr. Schuch
as a healthy man, his opinion would not have been altered by his prior
medical history as now ascertained in great detail. Dealing with the
evidence as a whole the learned trial judge came to the conclusion that 'if
the facts as stated in the evidence of Dr. Fierheller with relation to the
condition of Schuch and his treatment had been known to the defendant
company, it was not at all probable that they would have refused the premium
and the issue of the policy, nor do I think they would even have required
the examination which the officials now think they would have required.' In
this finding their Lordships substantially concur, although they would have
expressed the finding somewhat differently and would have preferred to say
that had the facts concealed been disclosed, they would not have influenced
a reasonable insurer so as to induce him to refuse the risk or alter the
premium. Their Lordships, therefore, concur in the conclusion of the trial
judge that the non-disclosure or misstatement was not material to the
contract and therefore, under the law of Ontario, is not a ground for
avoiding it."

As I read it, in this passage their Lordships were stating the following
propositions. (1) Materiality was not to be judged by reference to the
opinion of a reasonable assured. This had been the common law for very many
years. (2) The fact that the insurers had inserted a question in the
proposal form was not ipso facto a demonstration of its materiality. (3) The
fact that if the information had been correctly stated it would

have caused the insurers to delay in reaching a decision was not in itself
enough to make it material. (4) The test was whether the information would
have influenced a reasonable insurer to decline the risk or to have
stipulated for a higher premium.

Now I must say at once that this decision cannot in my opinion be put aside
simply because it arose in the context of a statute which required proof of
actual inducement. The discussion in the passage quoted was concerned with
materiality, not causation. Here, I must differ from the Court of Appeal in
the C.T.I. case [1984] 1 Lloyd's Rep. 476 and to this extent from Samuels J.
in Mayne Nickless Ltd. v. Pegler [1974] 1 N.S.W.L.R. 228, although the
general tenor in the latter case is broadly in accord with my own opinion.
This being said, however, I do not consider that the Mutual Life Insurance
case establishes Pan Atlantic's argument. The dispute centred on the first
three propositions which I have summarised. Once the insurers' contentions
on these were rejected the outcome of the case was inevitable, and the Board
had no occasion, when stating the general law on materiality, to examine the
quite narrow distinction which this House is now considering, and which had
no relevance to the decision of the appeal before the Board. I would make
the same comment about other expressions of the test for materiality in
cases decided since 1906. None of the disputes from which they sprang made
it necessary for the court to distinguish between the various
interpretations of the test so minutely examined in the arguments before the
House.

In these circumstances I consider that the House is free to interpret
sections 18(2) and 20(2) in the way proposed, and to give them a reading
which reflects not only the words actually used but also the insistence on
fair dealing and equal bargaining which has imbued this branch of the law
from the beginning. In reaching this conclusion I have concentrated on the
Act, notwithstanding that it does not apply directly to the present dispute,
because the principles have been largely developed in relation to marine
insurance. It is, however, accepted that the laws applying to the two types
of policy is in the relevant respects the same, and once it is established
that sections 17 to 20 of the Act correctly embody the common law of marine
insurance, as in my opinion it is, the problems now facing your Lordships
may be addressed simply by applying the words of the Act. When at a later
stage examining the facts of the present dispute I will go straight to the
language of sections 18(1) and 20(2).

V. Inducement

I turn to the second question which concerns the need, or otherwise, for a
causal connection between the misrepresentation or non-disclosure and the
making of the contract of insurance. According to sections 17, 18(1) and
20(1) if good faith is not observed, proper disclosure is not made or
material facts are misrepresented, the other party, or in the case of
sections 18 and 20 the insurer, "may avoid the contract." There is no
mention of a connection between the wrongful dealing and the writing of the
risk. But for this feature I doubt whether it would nowadays occur to anyone
that it would be possible for the underwriter to escape liability even if
the matter complained of had no effect on his processes of thought.

Take the case of misrepresentation. In the general law it is beyond doubt
that even a fraudulent misrepresentation must be shown to have induced the
contract before the promisor has a right to avoid, although the task of
proof may be made more easy by a presumption of inducement. The case of
innocent misrepresentation should surely be a fortiori, and yet it is urged
that so long as the representation is material no inducement need be shown.
True, the inequalities of knowledge between assured and underwriter have led
to the creation of a special duty to make accurate disclosure of sufficient
facts to restore the balance and remedy the injustice of holding the
underwriter to a speculation which he had been unable fairly to assess; but
this consideration cannot in logic or justice require courts to go further
and declare the contract to be vitiated when the underwriter, having paid no
attention to the matters not properly stated and disclosed, has suffered no
injustice thereby. How then does it happen that the Act seems to contemplate
that once a material misrepresentation or non-disclosure is established the
underwriter has an invariable right to avoid?

One possibility, suggested by Mr. D. St. L. Kelly (1988) 1 Ins.L.J. 30,
33-34, developing an interesting thesis centred on Ionides v. Pender, L.R. 9
Q.B. 531, is that the omission of an express requirement of inducement was
due to an oversight by the draftsman. With respect, this is a proposition
which I could not accept unless forced to do so by want of any other
explanation. Sir Mackenzie Chalmers was a most learned and careful scholar.
Throughout the prolonged struggle to get successive Marine Insurance Bills
through Parliament the drafts had been closely scrutinised and revised by
persons of great knowledge and experience: see the foreword to Chalmers and
Owen, A Digest of the Law relating to Marine Insurance, 1st ed. It would be
most surprising if an elementary proposition of marine insurance law had
simply slipped through the net, and the fact that it did not is borne out by
the fact that the concluding sentences of sections 18(1) and 20(1) did not
appear in the 1894 version of the Bill. Attention must have been directly
focused on the connection between misrepresentation and non-disclosure on
the one hand and the right to avoid on the other and it seems impossible
that a need for inducement, if such was the common law at the time, could
simply have been forgotten.

In these circumstances there appear to be three reasons why the Act may have
taken the form which it did. First, the common law did not require
inducement and was correctly reproduced by the Act. Second, the common law
did require inducement but the promoters of the Act wished the law to be
changed and Parliament did change it. Third, the common law did require
inducement and the Act, properly understood, is to the same effect. To make
a choice we must look behind the Act to the developing history of marine
insurance law. For this, once again, particular regard must be paid to the
scholarly writings. The older cases have practically nothing useful to say
on the matter, very possibly because what Duer, The Law and Practice of
Marine Insurance, vol. II, p. 566, para. 10, called "the known prudence of
underwriters" would make it comparatively rare for an underwriter to ignore
a material consideration. I should add that until the modern change in the
climate of business it

would have been even more unusual for insurers to risk having it said in
public proceedings that they were at the same time careless enough to ignore
something which a prudent underwriter would have taken into account and
anxious enough to avoid paying a claim to rely on a breach of duty which had
done them no harm.

A fact which at once captures attention is the existence, almost from the
outset, of a controversy about the need for inducement. I have already given
references to the conflicting views of the 19th century scholars. To modern
eyes the controversy is puzzling. The doctrine that a contract of marine
insurance is uberrimae fidei had been firmly established for decades. How
could there still be any doubt as to a point which, although rarely arising
in practice, was of fundamental theoretical importance, the more so given
that it is nowadays a truism that an innocent misrepresentation will lead to
rescission?

My Lords, I believe that the key to the problem is provided by MacGillivray
and Parkington on Insurance Law, 8th ed. (1988), pp. 231-232, para. 577,
where it is pointed out that at common law an innocent misrepresentation did
not affect the validity of the contract or afford a defence to an action
upon it, unless there was a total failure of consideration. This proposition
was asserted as late as 1867 in the judgment of the Court of Queen's Bench
delivered by Blackburn J. in Kennedy v. Panama, New Zealand and Australian
Royal Mail Co. Ltd. (1867) L.R. 2 Q.B. 580. Thus, in the field of life
insurance, which was governed by the common law, the law on
misrepresentation took a shape which was quite unrecognisable from what it
is today: see, for example, the difficult case of Anderson v. Fitzgerald
(1853) 4 H.L.Cas. 484. It was not until, after the Judicature Acts, the
equitable doctrines governing rescission for misrepresentation had
infiltrated the general law of contract that a route to the protection of
the underwriter in non-marine cases became apparent.

It was against this background that the doctrine of good faith in relation
to marine insurance was enunciated and developed. Originally, Lord Mansfield
had proceeded in Carter v. Boehm, 3 Burr. 1905 on the basis of a general
doctrine of good faith applicable to all contracts, and this doctrine was
propounded by Park J. in his influential early work on insurance, A System
of the Law of Marine Insurances, 1st ed. (1787); 2nd ed. (1790). This
general principle did not prevail, but marine insurance continued to be
treated as an exceptional case in which non-disclosure and misrepresentation
would ordinarily vitiate the contract even though they would not have had
that effect at common law. What was never clearly spelt out was how this
result was achieved. Various theories were advanced: that the policy failed
for want of agreement on the subject matter; that non-disclosure was
constructive fraud; and that contracts of marine insurance were subject to
an implied condition precedent that there had been full and accurate
disclosure. (See for example Arnould, A Treatise on the Law of Marine
Insurance and Average, 2nd ed., vol. I, p. 548; Phillips, A Treatise on the
Law of Insurance, 5th ed., vol. I, pp. 278-279; Parsons, A Treatise on
Maritime Law, vol. II, p. 153; Kent, Commentaries on American Law, 1st ed.
(1828); 14th ed. (1896), vol. III,

p. 282; and Duer, The Law and Practice of Marine Insurance, vol. II, p. 380.

This uncertainty led to differences of opinion on more than one topic.
Conspicuous examples were the questions whether the contract was rendered
void ab initio or merely voidable, and the mechanism whereby the validity of
the contract was compromised. To a great extent the source of the power to
avoid has been made academic by the creation of the statutory power under
the Act, but the controversy has still not been resolved: see, for example,
the line of cases beginning with Blackburn, Low & Co. v. Vigors (1886) 17
Q.B.D. 553 and (1887) 12 App.Cas. 531, 539 and continuing through William
Pickersgill & Sons Ltd. v. London and Provincial Marine and General
Insurance Co. Ltd. [1912] 3 K.B. 615 and Merchants' and Manufacturers'
Insurance Co. Ltd. v. Hunt [1941] 1 K.B. 295, 312, 318 to Mackender v.
Feldia A.G. [1967] 2 Q.B. 590, March Cabaret Club & Casino Ltd. v. London
Assurance [1975] 1 Lloyds Rep. 169 and Banque Keyser Ullmann S.A. v. Skandia
(U.K.) Insurance Co. Ltd. [1990] 1 Q.B. 665. A valuable recent study of the
authorities is contained in Clarke, The Law of Insurance Contracts, 2nd ed.
(1994), pp. 549-551, para. 23-1A. Since the reasons why misrepresentation or
non-disclosure vitiated the contract must have had at least some bearing on
the need for proof of inducement, if the present task had been to
reconstruct the common law as it stood in 1873 just before the decision in
Ionides v. Pender, L.R. 9 Q.B. 531, some taxing problems would have required
solution. Particular scrutiny would have been required of the observations
of Turner L.J. in Traill v. Baring (1864) 4 De G.J. & S. 318, 330. But there
is no need for this, since the problem is to ascertain the law which formed
the basis of the partial codification contained in Chalmers and Owen, A
Digest of the Law relating to Marine Insurance and reproduced with some
modifications in the Act. This law had by 1906 become a fusion of the common
law and equity. What one would expect to find, as regards misrepresentation,
is that the newly available equitable remedy should be applied in the field
of insurance, in the same way as in the law of contract at large. In my
opinion this is just what one does find. It will be recalled that the
predecessors in the Digest of what were to become sections 18(1) and 20(1)
of the Act did not contain any explicit reference to the avoidance of the
policy. What the Digest did contain, however, was article 93(2), 2nd ed., p.
132, in the following terms:

"The rules of the common law, including the law merchant, save in so far as
they are inconsistent with the express provisions of this Digest, and in
particular the rules relating to the effect of fraud, illegality,
misrepresentation, and mistake, apply to contracts of marine insurance."
(Emphasis added.)

This passage is supported by a note "as to fraud and misrepresentation, see
Leake on Contracts, 3rd ed. (1892), pp. 291, 330 . . ." The passage in Leake
thus called up stipulates, as regards fraud, that the party is entitled to
avoid by showing that he was induced to agree by the fraud of the other
party. Furthermore, in his note to article 17 of the DigestChalmers wrote
that "the ordinary rules of law as to voidable contracts apply to
insurance." (Digest, 1st ed., p. 22; 2nd ed., p. 24).

Subsequently, when Parliament passed the Act, the addition of the new
sentences in paragraphs 18(1) and 20(1) was accompanied by the omission of
the words in article 91(2) which I have underlined, but the remainder of
article 91(2) was unchanged. Precisely why these alterations were made is
now a matter of speculation, but at all events I see no room to doubt that
in the minds of those who prepared the statute the ordinary rules of law
were to apply. The note to article 17 survived into Chalmers's work when it
mutated into a commentary on the new Act (Chalmers and Owen, The Marine
Insurance Act 1906, 1st ed. (1907); 2nd ed. (1913)) and there is no trace in
the commentary of any suggestion that Parliament had deliberately changed
the law.

At this stage I pause to note that the footnote 4 to article 17 of the
Digest called up pp. 513-514 of the 6th edition of Arnould, A Treatise on
the Law of Marine Insurance (1887), where it was stated that the contract is
vitiated (or "void," as Arnould put it) wherever misrepresentation or
concealment "has entered into the making of it." This is a hint, although
probably no more, that the editor was contemplating a need for actual
inducement.

Similarly, when one turns to the contemporary texts, to see whether any
change in the relevant law was perceived after the Act, a comparison may
first be made between the 7th and 8th editions of Arnould (of 1901 and 1909
respectively), where the editors deal with misrepresentations of material
facts. In the former, we find, at p. 641:

"555. Even where the representation is of material facts, yet, if it
satisfactorily appears that it did not influence the judgment of the
underwriter, its falsity will be held not to avoid the policy."

For this proposition the editors cited Flinn v. Headlam (1829) 9 B. & C. 693
and appended the following note:

"Phillips (A Treatise on the Law of Insurance, 5th ed. vol. I, s. 681, pp.
371-372) is of opinion that the assured cannot be allowed to prove that a
material misrepresentation did not influence the underwriter. The editors
submit that the rule stated by Arnould is correct, although the evidence in
Flinn v. Headlam may not have justified its application. In equity it is
clear that even a fraudulent misrepresentation gives no right to rescind a
contract, when it has not influenced the party to whom it was made."

In the next edition the same passages appeared, but with the following
addition to the note:

"The Marine Insurance Act, however, if the language used in section 20,
subsection 1, is construed literally, supports Phillips's view. Such a
construction involves an anomalous state of the law. For it is clear that,
apart from marine insurance, even a fraudulent misrepresentation gives no
right to rescind a contract, when it has not influenced the party to whom it
was made."

Turning to the title "Insurance" in Halsbury's Laws of England, 1st ed.,
vol. 17 (1911), a text carrying the very high authority of Mr. Arthur

Cohen K.C., we find that the author first states as a general proposition,
at p. 404:

"789. A contract of marine insurance, like any other contract, is voidable
on the ground of fraud, and any fraudulent misrepresentation made in order
to induce the underwriter to enter into the contract entitles him to avoid
the policy, unless it is proved, either that he knew the true state of facts
at the time of contracting, or that he did not rely on the
misrepresentation."

Later, in a footnote, at p. 414, to a passage dealing with the innocent
misrepresentation of a material fact, the author says:

"(p) It is submitted that it is rightly argued in Arnould on Marine
Insurance, 8th ed. (1909), vol. I, section 536 (pp. 674-675), that as a
fraudulent representation will not avoid the contract (see p. 404, ante) if
it did not influence the mind of the contracting party, this must a fortiori
be true of an innocent misrepresentation."

No reference is made to the point on the literal construction of section
20(1) which had concerned the editors of the 6th edition of Arnould.

Next, I turn to a series of decisions after the passing of the Act. The
first is Cantiere Meccanico Brindisino v. Janson [1912] 2 K.B. 112. This
raised questions of misdescription and non-disclosure. In his judgment on
appeal [1912] 3 K.B. 452, 460 Vaughan Williams L.J. relates that the trial
judge had pointed out that

"it is sufficient to avoid the contract if the statement was in fact made,
and it was untrue, and that it is not necessary that the underwriter should
have relied on it . . ."

Yet the report of the judgment of Scrutton J. [1912] 2 K.B. 112 omits any
discussion of representation, and as reported at (1912) 106 L.T. 678, 680
the judge said only: "The defendants have not satisfied me that the alleged
misrepresentation was made . . ." It therefore appears that, whatever
exactly may have been said at the trial, it was obiter.

The next case was Visscherij Maatschappij Nieuw Onderneming v. Scottish
Metropolitan Assurance Co. Ltd. (1921) 27 Com.Cas. 198. A vessel sank in
suspicious circumstances whilst greatly overinsured. The trial judge found
that the vessel had been cast away and also held that the overvaluation was
a material fact which the owners had wrongfully failed to disclose. The
Court of Appeal upheld his decision on the first issue, but doubted the
conclusion on non-disclosure given the sparseness of the evidence. As
Scrutton L.J. observed, as at pp. 215-216:

"The underwriters have not taken the course, which in my view should always
be followed, of coming into the box and saying what they knew, and what it
was material that they did not know, and how it influenced them."

Whilst anything said by Scrutton L.J., particularly on a commercial matter,
must always command the greatest respect, I do not find that these dicta
take the matter very much further forward. Nor am I able to derive help from
Mutual Life Insurance Co. of New York v. Ontario Metal

Products Co. Ltd. [1925] A.C. 344, since this turned on an express provision
in the Ontario Insurance Act 1914 to the effect that:

"The proposals or application of the assured shall not as against him be
deemed a part of or considered with the contract of insurance except in so
far as the court may determine that it contains a material misrepresentation
by which the insurer was induced to enter into the contract" (section
156(3)).

Next, there is Zurich General Accident and Liability Insurance Co. Ltd. v.
Morrison [1942] 2 K.B. 53. This arose under the Road Traffic Act 1934 which
enabled third parties to recover from insurers the amount of any judgment
obtained in respect of a liability covered by a policy. The scheme of the
Act was that the insurer's right to avoid was preserved as against the third
party if, in an action commenced within a given time, the insurers procured
a declaration that the policy had been "obtained" by the non-disclosure of a
material fact or by a representation of fact which was false in some
material particular. The circumstances of the case were that the insured had
obtained a motor vehicle policy with the plaintiffs after completing a
proposal form in which he stated that he had driven regularly and
continuously during the past 12 months and had driven cars for three years.
An accident caused the death of the third party's husband. The insurers
brought an action against the third party amongst others claiming to avoid
the policy, relying on an allegation that the facts represented in the
proposal were untrue and that the insured had failed to disclose that the
driver had failed a driving test. The action failed. For present purposes
three passages from the judgments of the Court of Appeal are material:

"The evidence entirely fails to convince me that the insurers, had they
known of the failure to pass the test, would have declined to issue a policy
on precisely the same terms as those on which they did issue the policy in
question:" per Lord Greene M.R., at p. 58.

Another passage conveys the same idea, at p. 59:

"Under the general law of insurance an insurer can avoid a policy if he
proves that there has been misrepresentation or concealment of a material
fact by the assured. What is material is that which would influence the mind
of a prudent insurer in deciding whether to accept the risk or fix the
premium, and if this be proved it is not necessary further to prove that the
mind of the actual insurer was so affected. In other words, the assured
could not rebut the claim to avoid the policy because of a material
representation by a plea that the particular insurer concerned was so
stupid, ignorant, or reckless, that he could not exercise the judgment of a
prudent insurer and was in fact unaffected by anything the assured had
represented or concealed. But under the provisions of the Road Traffic Act
1934, I think this general rule of insurance law is modified. The section
requires the insurer to establish that the policy was 'obtained' by
non-disclosure or misrepresentation. In such a case as this, therefore, I
think the plaintiff must establish two propositions: (1) that the matter
relied on was 'material' in the sense that the mind of a prudent insurer
would

be affected by it, and (2) that in fact their underwriter's mind was so
affected, and the policy was thereby obtained:" per MacKinnon L.J., at p.
60.

"[After holding that the fact not disclosed was immaterial.] I am convinced
.. . . that had the fact been disclosed the assured would have got exactly
the same policy that was issued in this case and, consequently, that it was
not obtained by non-disclosure of a material fact:" per Goddard L.J., at p.
65.

No authority was cited for the observations of MacKinnon L.J. on what the
position would have been in the absence of the express statutory requirement
that the policy had been obtained by the non-disclosure, and the passage is
plainly obiter.

Finally, there is Marene Knitting Mills Pty. Ltd. v. Greater Pacific General
Insurance Ltd. [1976] 2 Lloyd's Rep. 631. This was concerned solely with the
question whether the fact that the predecessor in business of the assured
had suffered a series of fires was a material consideration which should
have been disclosed. In the course of an opinion delivered by Lord Fraser of
Tullybelton upholding the decision that this fact was material the Judicial
Committee of the Privy Council took note that the trial judge had directed
himself by reference to a passage from the judgment of Samuels J. in Mayne
Nickless Ltd. v. Pegler [1974] 1 N.S.W.L.R. 228, 239:

"Accordingly, I do not think that it is generally open to examine what the
insurer would in fact have done had he had the information not disclosed.
The question is whether that information would have been relevant to the
exercise of the insurer's option to accept or reject the insurance proposed.
It seems to me that the test of materiality is this: a fact is material if
it would have reasonably affected the mind of a prudent insurer in
determining whether he will accept the insurance, and if so, at what premium
and on what conditions."

In my opinion this citation is no assistance here, for Lord Fraser
immediately went on to say that so far as materiality was concerned it was
unnecessary to pursue a possible difference from the test applied in Mutual
Life Insurance Co. of New York v. Ontario Metal Products Co. Ltd. [1925]
A.C. 344, since the earlier fires were obviously material whichever test was
applied; and as regards the full sentence of the passage quoted there was no
occasion to consider whether it was correct, since materiality and not
causation was the sole suspect of the appeal.

My Lords, in my judgment little or nothing can be gleaned from the 20th
century cases to indicate a solution to the problem of causation. Before
stating my own opinion on this problem there are two more points to be made.

First, one suggested explanation for the absence from section 20 of any
requirement that the misrepresentation shall have induced the contract is
that any such requirement had been swept away 30 years before in Ionides v.
Pender, L.R. 9 Q.B. 531. Consistently with the views already expressed I am
unable to accept this, and I should add that even if the effect of Ionides
v. Pender had been to make the influence on the hypothetical underwriter the
benchmark of materiality I am unable to see

why this should not have left behind such requirements of actual causation
as had previously formed part of the common law. However, as I have said,
although Ionides v. Pender was an important case it did not in my opinion
have the effect contended for.

Secondly, it has been suggested that the absence from the Act of any
reference to causation stems from a disciplinary element in the law of
marine insurance. The concept is that persons seeking insurance and their
brokers cannot be relied upon to perform their duties spontaneously; that
the criterion of whether or not the misrepresentation or non-disclosure
induced the contract would make it too easy for the assured to say that the
breach of duty made no difference; and that accordingly the law prescribes
voidability as an automatic consequence of a breach by way of sanction for
the enforcement of full and accurate disclosure. For my part, although I
think it possible to detect traces of this doctrine in the earlier writings
I can see nothing to support it in later sources; and I would unhesitatingly
reject any suggestion that it should now be made part of the law. The
existing rules, coupled with a presumption of inducement, are already stern
enough, and to enable an underwriter to escape liability when he has
suffered no harm would be positively unjust, and contrary to the spirit of
mutual good faith recognised by section 17, the more so since non-disclosure
will in a substantial proportion of cases be the result of an innocent
mistake.

For these reasons I conclude that there is to be implied in the Act of 1906
a qualification that a material misrepresentation will not entitle the
underwriter to avoid the policy unless the misrepresentation induced the
making of the contract, using "induced" in the sense in which it is used in
the general law of contract. This proposition is concerned only with
material misrepresentations. On the view which I have formed of the present
facts the effect of an immaterial misrepresentation does not arise and I say
nothing about it.

There remain two problems of real substance. The first is whether the
conclusion just expressed can be transferred to the case of wrongful
non-disclosure. It must be accepted at once that the route via section 91(2)
of the Act and the general common law which leads to a solution for
misrepresentation is not available here, since there was and is no general
common law of non-disclosure. Nor does the complex interaction between fraud
and materiality, which makes the old insurance law on misrepresentation so
hard to decipher, exist in respect of non-disclosure. Nevertheless if one
looks at the problem in the round, and asks whether it is a tolerable result
that the Act accommodates in section 20(1) a requirement that the
misrepresentation shall have induced the contract, and yet no such
requirement can be accommodated in section 18(1), the answer must surely be
that it is not - the more so since in practice the line between
misrepresentation and non-disclosure is often imperceptible. If the Act,
which did not set out to be a complete codification of existing law, will
yield to qualification in one case surely it must in common sense do so in
the other. If this requires the making of new law, so be it. There is no
subversion here of established precedent. It is only in recent years that
the problem has been squarely faced. Facing it now, I believe that to do
justice a need for inducement can and should be implied into the Act.

In forming this conclusion I have taken into account the contrary opinions
expressed in Spencer Bower, Turner & Sutton, The Law relating to Actionable
Non-Disclosure, 2nd ed. (1990), para 3.09, p. 36. Whilst attaching great
weight to any proposition from this source, I believe that both principle
and justice require the conclusion which I have expressed.

Accordingly, I consider that the instinct of Kerr J. in Berger v.
Pollock[1973] 2 Lloyd's Rep. 442, was right, and that the adoption of the
contrary view by the Court of Appeal in the C.T.I. case [1984] 1 Lloyd's
Rep. 476 should not now be upheld.

Finally, there is the question whether this conclusion holds good for
non-marine insurance. The problems raised by the wording of sections 18(1)
and 20(1) do not here arise. The general considerations are however the
same, and I feel no doubt that they should lead to the same conclusion.

Before embarking on this long analysis I suggested that the questions in
issue were short. I propose the following short answers. (1) A circumstance
may be material even though a full and accurate disclosure of it would not
in itself have had a decisive effect on the prudent underwriter's decision
whether to accept the risk and if so at what premium. But, (2) if the
misrepresentation or non-disclosure of a material fact did not in fact
induce the making of the contract (in the sense in which that expression is
used in the general law of misrepresentation) the underwriter is not
entitled to rely on it as a ground for avoiding the contract.

These propositions do not go as far as several critics of the C.T.I.case
would wish, but they maintain the integrity of the principle that insurance
requires the utmost good faith, whilst avoiding the consequences, to my mind
unacceptable, of upholding Pine Top's arguments in full.

VI. The facts

Although the underlying facts are of great importance to the parties they
have no general significance and have already been investigated in both the
courts below. Although they were explored in much detail, I may take them
quite shortly.

As regards the first presentation by the broker, if I correctly understand
the findings of the trial judge the broker had in his possession the long
record; he knew that it contained materials which showed a bad claims
experience; he let the underwriter know that he had it with him; he hoped
that the underwriter would not ask to see it, and conducted himself in such
a way as to reduce the likelihood that in fact the underwriter would do so;
in the event the underwriter did not ask to see it, with the result that he
accepted the risk in ignorance of facts which the broker himself, who had
taken the trouble to divert attention from them, must himself have thought
were significant. Notwithstanding that the trial judge held that there was
no material non-disclosure, and that the Court of Appeal saw no reason to
interfere, I would have wished to look very closely at this course of
events, if the appeal had turned on this question alone; for the conduct of
the broker scarcely seems redolent of the utmostgood faith required by
section 17, or of the fair representation on which Lord Mansfield in Carter
v. Boehm, 3 Burr. 1905, 1911 insisted. The point is however of no practical
importance, since quite apart from the question

of causation it seems to me that an argument of waiver would be
unanswerable. I am therefore content, in company with the Court of Appeal,
to leave the decision of the trial judge undisturbed.

As to the second presentation by the broker I have had the opportunity to
read in draft the speech to be delivered by my noble and learned friend,
Lord Lloyd of Berwick. Although the test of materiality which he applies is
not the same as the one which I myself would prefer I so entirely concur in
his opinion that the misrepresentation was material and in his reasons for
arriving at that conclusion that it would be pointless to cover the same
ground again. I will add one brief rider. Differing from the C.T.I.case and
hence from the principle which the Court of Appeal was bound to apply in the
present case I have concluded that it is an answer to a defence of
misrepresentation and non-disclosure that the act or omission complained of
had no effect on the decision of the actual underwriter. As a matter of
common sense however even where the underwriter is shown to have been
careless in other respects the assured will have an uphill task in
persuading the court that the withholding or misstatement of circumstances
satisfying the test of materiality has made no difference. There is ample
material both in the general law and in the specialist works on insurance to
suggest that there is a presumption in favour of a causative effect. It is
not necessary for present purposes to give the proposition this formal
label, or to explore it in detail. For present purposes it is sufficient to
say, in company with my noble and learned friend, that on the facts of this
particular case the position as regards causation is so clear that the
appeal can be decided in favour of the indemnitors without the need for
remittal to the trial judge.

VII. Inadvertent omissions

Pan Atlantic maintains that the non-disclosures relied upon by Pine Top were
inadvertent and therefore excused by the opening words of article XV.
Notwithstanding the energetic argument of Mr. Beloff I am quite satisfied
that article XV is inapt for this purpose. The matter is thoroughly
discussed by Steyn L.J. who has said all that is necessary on the matter.

VIII. Conclusion

For these reasons, although I differ in certain important respects from the
view of the law which the Court of Appeal was constrained to apply I would
dismiss the appeal. In conclusion I wish to acknowledge the painstaking
research which founded the arguments addressed on appeal, and in particular
the deployment of modern academic and other writings. Throughout its long
history the law of marine insurance has owed as much to commentators as to
the courts, and although the views of these writers are not fully reflected
here, I have taken them carefully into account.

JUDGMENTBY-4: LORD SLYNN OF HADLEY

JUDGMENT-4:
LORD SLYNN OF HADLEY: My Lords, I have had the advantage of reading in draft
the speech of my noble and learned friend, Lord Mustill. I agree with him
that the "decisive influence" test is to be rejected and

that a circumstance may be material for the purposes of an insurance
contract (whether marine or non-marine) even though had it been fully and
accurately disclosed it would not have had a decisive effect on the prudent
underwriter's decision whether to accept the risk and if so at what premium,
but that an underwriter who is not induced by the misrepresentation or
non-disclosure of a material fact to make the contract cannot rely on the
misrepresentation or non-disclosure to avoid the contract.

I also agree that no sufficient reason has been shown for interfering with
the trial judge's findings of fact. I, too, would dismiss the appeal.

JUDGMENTBY-5: LORD LLOYD OF BERWICK

JUDGMENT-5:
LORD LLOYD OF BERWICK: My Lords, in these proceedings Pan Atlantic Insurance
Co. Ltd. and others ("Pan Atlantic") claim payment under a contract of
reinsurance dated 29 October 1982. The reinsurers are Pine Top Insurance Co.
Ltd. now in liquidation ("Pine Top"). The defence is non-disclosure and
misrepresentation. This appeal provides your Lordships with the opportunity
to consider the correctness of the decision of the Court of Appeal in
Container Transport International Inc. v. Oceanus Mutual Underwriting
Association (Bermuda) Ltd. [1984] 1 Lloyd's Rep. 476 ("the C.T.I.case").

Pan Atlantic's business included the insurance of liability risks, mostly in
the United States. The business was classic "long-tail" business; that is to
say, claims took a long time to be advised, and still longer to be settled.
The history starts with the years 1977 to 1979, during which, as will be
seen, very serious losses were incurred by Pan Atlantic. But Pine Top were
not affected by these losses, since they were not then party to the
reinsurance. They came on risk for the first time in 1980. The underwriter
who accepted the risk on behalf of Pine Top for that year was Mr. Oliver.
For the following year, 1981, the risk was accepted by Mr. Neil O'Keefe. The
contract was renewed for 1982, with Pine Top taking 50 per cent. of the
risk. There were two meetings between Mr. O'Keefe and Mr. Robinson, of
Messrs. Butcher, Robinson & Staples Ltd., the brokers, during which terms
were negotiated. The first of these meetings was on 22 or 23 December 1981,
and the second on 13 January 1982, when Mr. O'Keefe signed the slip. I will
return to those meetings when I come to consider the facts in greater
detail.

For a considerable time after the inception of the 1982 contract, Pine Top
continued to pay reinsurance claims covering the three years for which they
were on risk, namely, 1980, 1981 and 1982. Then in 1985 they ceased paying
altogether. Pan Atlantic recovered judgment under R.S.C., Ord. 14 in respect
of sums due for 1980 and 1981. In March 1987 they commenced these
proceedings in respect of the 1982 year. The defence was served in May 1987.
The only non-disclosure or misrepresentation alleged at that stage related
to the 1977-79 years, in which, as I have said, very serious losses were
incurred. The experts were agreed that these early years gave the best
indication of how the business was likely to develop. By contrast, the later
years were described by the judge as having very little relevance.

Pine Top's case was that Mr. O'Keefe was never shown the figures for 1977 to
1979. If he had been, there would have been a copy of the record

on his file. Pan Atlantic's case was that Mr. Robinson took the figures with
him to both meetings with Mr. O'Keefe, and that they were available on the
table for his inspection. Waller J. [1992] 1 Lloyd's Rep. 101, 106 accepted
Mr. Robinson's evidence. He found that there was a "perfectly fair
presentation" of the years in question. His finding was upheld in the Court
of Appeal. I will say at once that, despite Mr. Hamilton's arguments to the
contrary, I can see no basis on which the House could possibly disturb that
finding.

A few weeks before the case came on for trial, Pine Top amended their
defence to add certain further allegations of non-disclosure and
misrepresentation in relation to 1980 and 1981. So far as 1980 is concerned,
it was common ground that Pan Atlantic had failed to disclose a claim of
U.S.$25,000. But as it was also common ground that that non-disclosure was
"trivial" in the context of disclosed claims of U.S.$406,277, your Lordships
are not directly concerned with the 1980 year. The dispute thus centres on
1981. Pine Top's case is that there were two additional claims in respect of
that year which had already been advised, but which were not disclosed to
Mr. O'Keefe. The figure disclosed was U.S.$235,768. It should have been
U.S.$468,168. Pan Atlantic concede, and have conceded ever since the service
of the amended defence, that there was inadvertent non-disclosure in respect
of these two claims. It is not suggested that the non-disclosure was
deliberate. The issue is whether the admitted non-disclosure entitles Pine
Top to avoid the contract. Waller J., applying the test laid down by the
Court of Appeal in the C.T.I. case, as he was obliged to do, held that the
non-disclosure was material. The Court of Appeal, applying a different test,
reached the same conclusion. The question which now arises for the first
time in your Lordships' House is what is the correct test to apply for
avoidance by an insurer of a contract of insurance or reinsurance on the
ground of non-disclosure.

Mr. Beloff, on behalf of Pan Atlantic, submits that the insurer must show
(1) that a prudent insurer, if he had known of the undisclosed fact, would
either have declined the risk altogether or charged an increased premium,
and (2) that the actual insurer would himself have declined the risk or
charged an increased premium. Mr. Hamilton, on behalf of Pine Top, submits
that it is sufficient that the fact is one which a prudent insurer would
have "wanted to know" or would have "taken into account," even though it
would have made no difference to him in the result. According to Mr.
Hamilton, the impact on the actual insurer is irrelevant.

Waller J. adopted Mr. Hamilton's test even though, as he himself said, he
felt some unease in doing so. Why, he asked, at p. 14, should the failure to
disclose the two additional claims in the 1981 year entitle Pine Top to
avoid all liability "when their underwriter appears simply not to have been
concerned to look at or take account of the most material part of the record
in considering this risk," that is to say, the figures for 1977 to 1979? The
Court of Appeal [1993] 1 Lloyd's Rep. 496 rejected Mr. Hamilton's test, but
did not adopt Mr. Beloff's. The leading judgment was given by Steyn L.J. The
test which he proposed was whether a prudent underwriter would regard the
undisclosed fact as

tending to increase the risk. If so the insurer might avoid, whether or not
he would have declined the risk or charged an increased premium. Farquharson
L.J. agreed. Sir Donald Nicholls V.-C. also agreed, but like the judge felt
uneasy at the outcome. In his view, justice and fairness required that in a
case of inadvertent non-disclosure, such as the present, it ought to be
possible for the court to adjust the premium, or perhaps restrict the cover.
But this, as he pointed out, is not an option available under English law.
In the absence of some express provision in the contract, English courts are
bound to adopt an all-or-nothing approach. There have been various proposals
for reform, which have so far come to nothing. In the meantime, Sir Donald
Nicholls V.-C., at p. 508, regarded the present case as "an unhappy example
of a case where . . . the law manifestly does not produce a satisfactory
result."

Pan Atlantic now appeal from the decision of the Court of Appeal. Pine Top
seek to support the decision on the alternative ground, decided against them
below, that there was non-disclosure of the 1977 to 1979 record.

I will start by considering the two halves of Mr. Beloff's argument. But
first it is convenient to mention certain preliminary matters, by way of
clearing the ground. These were largely non-contentious. The C.T.I.case
[1984] 1 Lloyd's Rep. 476 was a case of marine insurance. In the present
case your Lordships are concerned with non-marine insurance. It is not
suggested that there is any relevant difference between the two. The law of
marine insurance is, of course, codified by the Marine Insurance Act 1906.
Section 18(1) and (2) provides:

"(1) Subject to the provisions of this section, the assured must disclose to
the insurer, before the contract is concluded, every material circumstance
which is known to the assured, and the assured is deemed to know every
circumstance which, in the ordinary course of business, ought to be known by
him. If the assured fails to make such disclosure, the insurer may avoid the
contract. (2) Every circumstance is material which would influence the
judgment of a prudent insurer in fixing premium, or determining whether he
will take the risk."

The Act of 1906 did not change the law. It restated the common law as to
non-disclosure or "concealment" as it had first been laid down by Lord
Mansfield in Carter v. Boehm, 3 Burr. 1905 (itself a case of non-marine
insurance) and as it was developed in a number of marine and non-marine
cases throughout the 19th century. It would have been possible for
non-marine insurance to have diverged since the passing of the Act of 1906.
But it was not suggested that this had happened.

Secondly, the duty to disclose every material circumstance known to the
assured before a contract of insurance is concluded (section 18 of the Act
of 1906) is closely linked with the duty to ensure that every material
representation is true (section 20). Both are illustrations or consequences
of the rule, set out in section 17, that a contract of insurance is a
contract of utmost good faith. In practice, non-disclosure and
misrepresentation are often joined as defences in the same pleading. They
were joined in the C.T.I. case and so they are here. Often, as here, the
alleged

misrepresentation adds nothing. It is but the converse of the non-disclosure
("impliedly represented that [the proffered] information . . . gave a true
and fair picture"). It would be unsatisfactory to reach a different
conclusion on the same facts depending on which way the case is put.

Thirdly, whereas the right to avoid for non-disclosure is peculiar to
contracts of utmost good faith, the right to avoid for misrepresentation is
part of the general law of contract. The reason for the special rule
relating to non-disclosure in insurance contracts is that in the usual case
it is the assured alone who knows "the special facts, upon which the
contingent chance is to be computed:" see Carter v. Boehm, per Lord
Mansfield, at p. 1909. The purpose of the rule is to rectify that imbalance.
But in the case of misrepresentation there is no need to differentiate
between a contract of insurance and any other contract. There is no reason
to put the insurer in a more favourable position than other contracting
parties, and no justification for doing so. The ordinary law suffices.

Lastly, the duty of disclosure operates both ways. Although, in the usual
case, it is the assured who knows everything, and the insurer who knows
nothing, there may be special facts within the knowledge of the insurer
which it is his duty to disclose, as where (to take the example given by
Lord Mansfield in Carter v. Boehm) the insurer knows at the time of entering
into the contract that the vessel has already arrived. Thus the obligation
of utmost good faith is reciprocal: see Banque Keyser Ullmann S.A. v.
Skandia (U.K.) Insurance Co. Ltd. [1991] 2 A.C. 249, per Lord Bridge of
Harwich, at p. 268, and Lord Jauncey of Tullichettle, at p. 281. Nor is the
obligation of good faith limited to one of disclosure. As Lord Mansfield
warned in Carter v. Boehm, at p. 1918, there may be circumstances in which
an insurer, by asserting a right to avoid for non-disclosure, would himself
be guilty of want of utmost good faith.

The prudent insurer

I have already mentioned the three possible tests which were canvassed in
argument before your Lordships. Before discussing them, it is appropriate to
quote certain passages from the C.T.I. case [1984] 1 Lloyd's Rep. 476 on
which Mr. Hamilton strongly relied.

The leading judgment was given by Kerr L.J. After referring to section 18(1)
and (2), Kerr L.J. said, at p. 492:

"The word 'judgment' - to quote the Oxford English Dictionary to which we
were referred - is used in the sense of 'the formation of an opinion.' To
prove the materiality of an undisclosed circumstance, the insurer must
satisfy the court on a balance of probability - by evidence or from the
nature of the undisclosed circumstance itself - that the judgment, in this
sense, of a prudent insurer would have been influenced if the circumstance
in question had been disclosed. The word 'influenced' means that the
disclosure is one which would have had an impact on the formation of his
opinion and on his decision-making process in relation to the matters
covered by section 18(2)."

A little later he said:

"[The prudent insurer] is in a hypothetical position, and evidence to
support the materiality of the undisclosed circumstance, from this point of
view, is therefore often given by an independent expert witness whose
evidence has to be assessed by the court long after the event. He, or the
actual insurer, or both, may then be asked: 'What would have been your
reaction if you had known of this undisclosed fact?' Both would in my view
give relevant evidence on materiality if they replied: 'I would then have
regarded the risk in a different light. I would have taken this circumstance
into account. As a first step I might have asked some questions before
making up my mind about the risk. What my final decision would have been, I
cannot now say for certain. I might have declined the risk altogether, or
increased the premium, or altered the terms in some other way.' And it would
make no difference if the witness went on: 'I might even have taken a
chance, or given credit for the frankness of the disclosure, by writing the
risk as I did. But I should obviously have been told about this fact before
being asked to make up my mind.'"

Parker L.J., at p. 507, accepted the proposition advanced by Mr. Waller that
an insurer is entitled to avoid a contract under section 18(1) if there was
undisclosed before the conclusion of the contract any circumstance which a
prudent insurer would take into account when reaching his decision whether
to take the risk or what premium to charge and that the position of the
particular insurer is irrelevant.

Stephenson L.J. said, at p. 527:

"Provided that there is some information which a prudent insurer would
obviously want to know, or which a credible expert swears he would want to
know, in considering an offer of a risk, that is a material circumstance
which the greatest good faith and the rule against concealment require the
assured or his agent to disclose . . ."

Later, he said, at p. 529:

"I conclude from the language of the subsections [sections 18(2) and 20(2)
of the Act of 1906] in their context and from the authorities that
everything is material to which a prudent insurer, if he were in the
proposed insurer's place, would wish to direct his mind in the course of
considering the proposed insurance with a view to deciding whether to take
it up and on what terms, including premium. His mind would, I think, be
influenced in the process of judging whether to do so, either temporarily
where he can say that he would ultimately have reached the same decision
without it, or permanently where it would have led him to reach a different
decision."

Mr. Beloff's criticism of the C.T.I. test can be encapsulated in a series of
rhetorical questions. If the prudent insurer, knowing of the undisclosed
fact, would have accepted the risk at the same premium and on the same
terms, what right has the actual insurer to complain? What injustice has he
suffered? If the risk run is different from the risk understood or intended
to be run, then, as Lord Mansfield made plain in Carter v.

Boehm, 3 Burr. 1905, the insurer can avoid; and rightly so. But if the
prudent insurer would have accepted the risk at the same premium and on the
same terms, it must be because, so far as he is concerned, the risk is the
same risk. How, as a matter of ordinary language, can a circumstance be
described as material when it would not have mattered to the prudent insurer
whether the circumstance was disclosed or not? It is obvious that the
insurer cannot be required to disclose every circumstance, however remotely
related to the assessment of the risk. Why then should he be required to
disclose a circumstance which would not in fact have made any difference?
How in those circumstances could it be said that the actual insurer's
consent had been vitiated? And if not, on what other juristic basis could he
claim the right to avoid the contract?

Mr. Hamilton's answer to this line of criticism was that it is the insurer
who is entitled to decide whether to accept the risk or not, and if so at
what premium. So it is for the assured to disclose everythingwhich the
insurer would want to know, or would take into account, in reaching that
decision.

I do not find this answer satisfactory, not because, as is sometimes said,
it makes the insurer judge in his own cause, but rather because it blurs the
edges of the prudent insurer test. The purpose of the test, as will be seen
when I come to the authorities, and in particular Ionides v. Pender, L.R. 9
Q.B. 531, was to establish an objective test of materiality, not dependent
on the actual insurer's own subjective views. The test should therefore be
clear and simple. A test which depends on what a prudent insurer would have
done satisfies this requirement. But a test which depends, not on what a
prudent insurer would have done, but on what he would have wanted to know,
or taken into account, in deciding what to do, involves an unnecessary step.
It introduces a complication which is not only undesirable in itself but is
also, in the case of inadvertent non-disclosure, capable of producing great
injustice.

In the C.T.I. case [1984] 1 Lloyd's Rep. 476 Stephenson and Parker L.JJ.
attached great importance to the difficulty, as they saw it, of applying the
test proposed by Mr. Beloff. Indeed it was one of the two factors which
Stephenson L.J. regarded as decisive. He said, at pp. 526-527:

"two considerations prevent me from adopting [the judge's] construction of
the words. The first, stressed by my brethren, is the practical difficulty,
if not impossibility, of deciding what factors would affect the result of a
hypothetical prudent insurer's consideration of a risk, whether to accept it
and on what terms; whereas there is no great difficulty in answering the
question whether any particular factor would be one which he would want to
know and take into consideration in determining whether to accept a risk and
on what terms, without having to decide whether he would ultimately
disregard it altogether or give it much or little weight."

With great respect, I take exactly the opposite view. What the prudent
insurer would have wanted to know is as nebulous and ill-defined as the

alternative is precise and clear-cut. Parker L.J. made the same point at
greater length, at pp. 510-511:

"The test submitted on behalf of the respondents [plaintiffs] would involve
the court in the task, perhaps years after the event, of endeavouring to
ascertain what a prudent underwriter would have done, first in the light of
the circumstances actually disclosed by the assured, and secondly, on the
hypothesis that, in addition to those circumstances, the undisclosed
circumstance had been disclosed. Such a task is on its face impractical.
Five experienced and prudent underwriters might be called. At stage 1, one
might say he would not have taken the risk even on the facts disclosed; the
other four might all have taken the risk but at different premiums. At stage
2 the four remaining might all say 'we regard the fact as significantly
increasing the risk' but one might say 'not, but by the narrowest margin,
sufficiently to demand a change in premium, but it would call for a change
in the policy wording,' one 'sufficiently to put up the premium,' and the
last, 'sufficiently to decline the risk.' Furthermore, the one who would not
have taken the risk in the first place might say that he would, had the
additional fact been disclosed, have regarded it as an additional reason for
declining the risk. In such circumstances what is the court to do? It
cannot, as it seems to me, choose one prudent underwriter rather than
another. The very choice of a prudent underwriter as the yardstick in my
view indicates that the test intended was one which could sensibly be
answered in relation to prudent underwriters in general. It is possible to
say that prudent underwriters in general would consider a particular
circumstance as bearing on the risk and exercising an influence on their
judgment towards declining the risk or loading the premium. It is not
possible to say, save in extreme cases, that prudent underwriters in general
would have acted differently, because there is no absolute standard by which
they would have acted in the first place or as to the precise weight they
would give to the undisclosed circumstance."

I am unable to accept the reasoning in this paragraph. Five experienced and
prudent underwriters are just as likely - in my view more likely - to
disagree about what they would want to know as about what they would have
done. If it were always possible to say what a prudent underwriter would or
would not want to know, as Parker L.J. seems to have thought, it is
surprising that so many contested cases of non-disclosure have come before
the courts since the C.T.I. case was decided.

In truth, wherever the line is drawn, there will always be expert witnesses
prepared to give evidence on either side; it is then always for the court to
decide what evidence to accept, as in every other case of conflicting expert
testimony. I see no great practical difficulty on that score. On the
contrary, the certainty and practicality of Mr. Beloff's test are among its
strengths.

That brings me to the central question. What does section 18(2) of the Act
of 1906 mean? In particular, what is meant by the words "would influence the
judgment of a prudent insurer?" If I ask myself what the phrase as a whole
means, I would answer that it points to something more

than what the prudent insurer would want to know, or take into account. At
the very least it points to what the prudent insurer would perceive as
increasing, or tending to increase, the risk. On this aspect of the case I
agree entirely with the powerful analysis of Steyn L.J. in the Court of
Appeal [1993] 1 Lloyd's Rep. 496. He said of the C.T.I. case, at p. 505:

"Having rejected the 'decisive influence' construction, it seems to me there
were at least two feasible alternative solutions to be considered in the
C.T.I. case [1984] 1 Lloyd's Rep. 476. The first solution was that a fact is
material if a prudent insurer would have wished to be aware of it in
reaching his decision. The second solution involves taking account of the
fact that avoidance for non-disclosure is the remedy provided by law because
the risk presented is different from the true risk. But for the
non-disclosure the prudent insurer would have appreciated that it was a
different and increased risk."

Having set the scene in that way, Steyn L.J. unhesitatingly rejected the
first solution and chose the second. In other words he rejected the test
proposed by Mr. Hamilton. Whether, if he had been free to do so, he would
have chosen the test proposed by Mr. Beloff we shall never know. But it
matters not. The reasons given by Steyn L.J. for rejecting the first
solution and preferring the second are based largely on the judgment of Lord
Mansfield in Carter v. Boehm, 3 Burr. 1905. I would adopt those reasons, and
will not repeat them. I would only add that the increased risk theory of
materiality fits in neatly with the specific provision of section 18(3)(a)
that the assured need not disclose a circumstance by which the risk is
diminished.

If I analyse the phrase word by word, I reach the same conclusion, but I am
carried one stage further. The ordinary meaning of "influence" is to affect
or alter. "Judgment" is a word with a number of different meanings, so it is
not possible to identify the ordinary meaning in the abstract. In a legal or
quasi-legal context it is often used in the sense of a decision or
determination, as in "the judgment of Solomon" or "the judgment of Paris,"
or the formal judgment of a court of law. Kerr L.J. in the C.T.I. case
[1984] 1 Lloyd's Rep. 476, 492 considered that it meant not the decision
itself, but what he called the decision-making process. I accept that the
word may bear that meaning. But it is not the primary meaning given in the
Oxford English Dictionary, as Kerr L.J.'s judgment may suggest, and I see no
reason to give it that meaning in the present context.

In a commercial context "judgment" is often used in the sense of
"assessment." A market assessment means a judgment as to what the market is
going to do, not the process by which a stockbroker arrives at that
judgment. That is, in my opinion, the sense in which the word is used in
section 18(2) of the Act of 1906. Parker L.J. in the C.T.I. case, at p. 510,
attached importance to the words "in fixing" the premium and "in . . .
determining" whether to take the risk. But I do not regard these words as
pointing to a decision-making process, rather than the decision itself.

Finally, there is the word "would." Kerr L.J. in the C.T.I. case, at p. 492,
in a passage already quoted refers to things which the insurer might have
done if he had been told of the undisclosed fact. In my

judgment it is never enough to show that a prudent insurer might have
declined the risk or charged an increased premium. It is necessary to show
that he would have done.

My provisional conclusion, before coming to the authorities, is that Mr.
Beloff succeeds on the first half of his argument, and that in order to
avoid a contract for non-disclosure it must be shown that a prudent insurer,
if he had known of the undisclosed fact, would either have declined the risk
altogether, or charged an increased premium. This goes further than Steyn
L.J. in the Court of Appeal, but not by much. For in all ordinary cases
where the prudent insurer would have perceived an increase in the risk, he
would presumably charge an increased premium. There might be special
circumstances in which the actual insurer would decide, for his own reasons,
to incur an increased risk at the same premium. But this consideration
should not affect the objective application of the prudent insurer test. My
reasons for preferring Mr. Beloff's test are that it does full justice to
the language of section 18 of the Act of 1906. It is well-defined, and
easily applied. It does something to mitigate the harshness of the
all-or-nothing approach which disfigures this branch of the law, and it is
consistent with the reasons given by the Court of Appeal for rejecting the
test proposed by Mr. Hamilton.

In the course of his judgment [1993] 1 Lloyd's Rep. 496, 504, Steyn L.J.said
that the C.T.I. case had proved to be a remarkably unpopular decision not
only in the legal profession but also in the insurance markets. I have no
reason to doubt that view. It certainly seems to accord with the many
articles and publications to which we were referred, some of which are
mentioned by Steyn L.J. in his judgment, and all of which, without
exception, are critical of the C.T.I. decision. Steyn L.J.'s judgment in
this case, with its decisive rejection of the "want to know" or "take into
account" approach, has come far to meet these criticisms. Your Lordships are
free to go a little further. If, instead, we were to accept Mr. Hamilton's
argument, we would be reverting to the position as it was before the
decision in the court below. I suspect that this would cause widespread
regret.

The authorities prior to 1906

A very large number of cases were cited on both sides. Mr. Beloff submitted
that they did not help greatly one way or the other, since the precise point
seems never to have arisen and the result would in each case have been the
same whichever test had been applied. In these circumstances the language
cited by judges, however eminent, is of little assistance. I agree with this
submission. I only mention the cases at all because they were regarded as
helpful by the Court of Appeal in the C.T.I.case [1984] 1 Lloyd's Rep. 476.

The first case cited by Kerr L.J. was Carter v. Boehm, 3 Burr. 1905. In my
opinion, that case provides strong support for the increased risk test
favoured by the court below, and indirectly for the test proposed by Mr.
Beloff. It provides no support at all for the "take into account" or "want
to know" approach.

Next I mention Ionides v. Pender, L.R. 9 Q.B. 531. This case is important
because, as appears from Sir Mackenzie Chalmers's A Digest of

the Law relating to Marine Insurance, 1st ed., p. 22; 2nd ed. (1903), p. 25,
it was the foundation of section 18 of the Act of 1906.

The case concerned a policy on goods, which were grossly overvalued. The
vessel was lost in circumstances which suggested that she might have been
scuttled. To a claim on the policy, the defence was non-disclosure of the
excessive valuation. It was argued for the plaintiffs (see at p. 538) that
the excessive valuation was extraneous. It had no direct bearing on the
risks insured, such as perils of the sea. They cited Duer, The Law and
Practice of Marine Insurance, vol. II, p. 388 (see at pp. 537-538). But
Blackburn J., giving the judgment of the court, at p. 539, preferred the
rule laid down in Parsons, A Treatise on the Law of Marine Insurance, vol.
I, p. 495, that all should be disclosed which would affect the judgment of a
rational underwriter. In other words, materiality is not limited, as Duer
thought, to the risks "considered in their own nature." It covers also what
we would now call the moral hazard. The defendants called underwriters in
support of their case. They said that the overvaluation was a material fact.
One class of underwriters regarded the risk as "speculative," and would have
declined it altogether. Another class would have added 25 to 30 per cent. to
the usual premium. It will be noted that the evidence related to what the
underwriters would have done, not what they would have wanted to know, in
order to arrive at a decision. So the facts of the case reinforce Mr.
Beloff's argument. So does the judgment. Blackburn J. said, at p. 539:

"We agree that it would be too much to put on the assured the duty of
disclosing everything which might influence the mind of an underwriter.
Business could hardly be carried on if this was required."

Two years later, Blackburn J. had to consider a similar point in Stribley v.
Imperial Marine Insurance Co. (1876) 1 Q.B.D. 507. The action was on a
policy taken out on 27 February 1874. On 21 January the owner had received a
letter from the master, dated 9 January, that the vessel had been delayed by
bad weather, and that he would write again before sailing. That was the last
which the owners heard. The defence was that the letter ought to have been
disclosed. The jury found in favour of the plaintiffs, but the court ordered
a new trial. There are observations in the judgments of Lush J. and Quain J.
which are helpful to Mr. Hamilton. But Blackburn J. said, at p. 512:

"The question was, whether this letter, and the time which had elapsed since
it was received, ought to have been communicated to the underwriter. I think
the test is, whether a fair and reasonable underwriter, looking at this
letter and the circumstances under which it was received, would say, 'I
think this is a speculative risk, which I will either decline to take, or,
if I do take, it shall be at a greater premium than is usual.'"

Once again, there is no hint in Blackburn J.'s judgment that the test
depends on what the fair and reasonable underwriter would want to know. It
depends on what he would have done.

The next case was Rivaz v. Gerussi Brothers & Co. (1880) 6 Q.B.D. 222. There
is a lengthy citation from the decision in the judgment of

Parker L.J. in the C.T.I. case [1984] 1 Lloyd's Rep. 476, 507. The facts
were that the defendants, who were merchants, had systematically and
fraudulently undervalued shipments of fruit from Greece on certain open
policies, with the result that those policies were in truth all but
exhausted. It was held that the plaintiff, who underwrote two later policies
to follow the earlier policies, was entitled to avoid for non-disclosure.

The point in the case was exactly the same as the point in Ionides v.
Pender, L.R. 9 Q.B. 531. It was argued that the non-disclosure did not
affect the insured risks in the narrow sense. Once again the defendants
relied on Duer, The Law and Practice of Marine Insurance, vol. II, p. 390.
But Brett L.J. pointed out in the course of argument, at p. 226, that the
rule laid down by Duer had not been accepted by Lord Blackburn. On the
facts, the non-disclosure was obviously material, first, because the later
policies would be called on sooner than the plaintiff had been led to
expect, and, secondly, because the undervaluation of the earlier policies
was fraudulent. In the latter respect, it is another example of what we
would now call "the moral hazard."

Tate & Sons v. Hyslop (1885) 15 Q.B.D. 368 illustrates better than any other
case to which we were referred the dangers of taking general statements out
of context. The plaintiffs insured a consignment of sugar in course of
carriage to their refinery at Silvertown, including all risks of
transshipment into lighters. The sugar was damaged while lying in lighters
on the Thames. It was the practice of underwriters in such cases to exercise
their right of subrogation against the lightermen. The lightermen found that
the liabilities so imposed on them were onerous. So in April 1882 an
association of Thames lightermen published a notice that they would no
longer accept liability as common carriers, but only for the negligence or
wilful acts of their servants. The underwriters at Lloyd's responded at
once. They passed a resolution by which, if the lighterage was to be subject
to the new terms, the premium would be increased by up to 2s. 6d. per cent.
The plaintiffs had an arrangement with one of the lightermen whereby his
liability as common carrier was excluded. They failed to disclose this
arrangement when placing the insurance. The jury found that the
non-disclosure was material. The Court of Appeal refused to disturb the
verdict.

The judgment of Brett M.R. is of particular interest. He held that if the
only effect of the arrangement was to minimise the underwriters' right of
salvage, that is to say, the right of recourse against the lighterman, that
would not be a material fact. But here the non-disclosure was material
because the underwriters had made known to the market that they would charge
an increased premium if the liability of the lighterman was limited.

"The authorities show that the materiality is not as to the risk, but as to
whether it would influence the underwriters in entering upon the insurance
or the terms on which they would insure." (p. 376.)

So far from supporting the views expressed by the Court of Appeal in the
C.T.I. case, Tate & Sons v. Hyslop is, if anything, a case the other way.
The verdict of the jury was upheld because there was evidence that
underwriters would in fact have charged an increased premium if the
arrangement had been disclosed. The case is thus in direct line with

Ionides v. Pender and Rivaz v. Gerussi Brothers & Co. All three cases
establish that materiality is not limited to the insured risks, in the
narrow sense, as Duer thought; but covers everything which would influence
or affect the mind of underwriters, so as to decline the business, or
increase the premium.

Finally, the Court of Appeal in the C.T.I. case [1984] 1 Lloyd's Rep. 476
attached importance to Traill v. Baring, 4 De G.J. & S. 318. The case is a
well known authority for the proposition that a representation which is true
when made must be corrected if, before the contract is concluded, it becomes
untrue to the knowledge of the representor. So far as I know, it has never
been mentioned in connection with non-disclosure until it was relied on by
Samuels J. in Mayne Nickless Ltd. v. Pegler [1974] 1 N.S.W.L.R. 228, 239 and
cited by Mr. Hutchinson in his argument in Commonwealth Insurance Co. of
Vancouver v. Groupe Sprinks S.A. [1983] 1 Lloyd's Rep. 67 (see at p. 78). It
is true that there is a passage in Turner L.J.'s judgment which supports Mr.
Hamilton's argument. But the passage in question is at odds with the
judgment of Knight Bruce L.J. who said, at p. 326:

"It appears to me, I repeat, that the plaintiffs are entitled to assert, and
to be believed in asserting, that they would not have acted as they have
done if they had known, as they ought to have been informed by the society
represented by the defendants of, the real facts."

In these circumstances I attach little weight to the isolated observation of
Turner L.J.

Returning now to the mainstream of authority leading up to the passage of
the Act of 1906, I can find little if anything to support Mr. Hamilton's
construction of section 18(1) and (2). On the contrary, the evidence in two
of the cases, Ionides v. Pender, L.R. 9 Q.B. 531 and Tate & Sons v. Hyslop,
15 Q.B.D. 368, was plainly directed to what underwriters would have done,
rather than to what they would have wanted to know; and the emphasis in
these cases, as well as Rivaz v. Gerussi Brothers & Co., was on repudiating
the restricted view expressed by Duer, The Law and Practice of Marine
Insurance, vol. II, pp. 388-390, in preference to the wider view expressed
by Parsons, A Treatise on the Law of Marine Insurance, vol. I, p. 495 and
Phillips, A Treatise on the Law of Insurance, 5th ed., vol. I, s. 531, p.
277. It is this wider view which is reproduced in the Act of 1906. There is
nothing in Parsons or Phillips which suggests that materiality extends to
what insurers would have "taken into account" or "wanted to know," even
though it would have made no difference to the result.

The authorities since 1906

The leading authority on the meaning of "material" since the passing of the
Act of 1906 is Mutual Life Insurance Co. of New York v. Ontario Metal
Products Co. Ltd. [1925] A.C. 344. The case was one of life insurance. The
headnote reads:

"When statements made by an insured person upon his application for a policy
of life insurance are not made the basis of the contract

but are to be treated merely as representations, an inaccurate statement is
material so as to vitiate the policy if the matters concealed or
misrepresented, had they been truly disclosed, would have influenced a
reasonable insurer to decline the risk, or to have stipulated for a higher
premium; it is not sufficient that they would merely have caused delay in
issuing the policy while further inquiries were being made."

Kerr L.J. in the C.T.I. case, at p. 495, regarded the decision as being "no
authority for present purposes" because it depended on the particular
provisions of the Ontario Insurance Act 1914. Parker L.J. and Stephenson
L.J. took the same view. I am unable to agree.

Section 156 of the Act of 1914 provided:

"(3) The proposals or application of the assured shall not as against him be
deemed a part of or considered with the contract of insurance except in so
far as the court may determine that it contains a material misrepresentation
by which the insurer was induced to enter into the contract. (4) No contract
shall be avoided by reason of the inaccuracy of any such statement unless it
is material to the contract. . . . (6) The question of materiality in any
contract of insurance shall be a question of fact for the jury or for the
court if there is no jury."

Kerr L.J. seems to have thought that, because the Ontario statute required
proof of inducement, the case did not help on materiality. But the reasoning
of Lord Salvesen, tendering the advice of a strong Board, shows that this is
not so. Materiality, as a separate consideration, lies at the heart of the
case.

A policy was taken out on the life of the deceased. One of the questions in
the proposal form was whether he had seen a doctor within the last five
years. It was found that the answer to this question was inaccurate. The
deceased had been seen by a Dr. Fierheller. The question then arose whether
this non-disclosure was material. The test of materiality suggested by the
insurers, at p. 351, was

"whether, if the fact concealed had been disclosed, the insurers would have
acted differently, either by declining the risk at the proposed premium or
at least by delaying consideration of its acceptance until they had
consulted Dr. Fierheller."

The Board accepted the first half of this proposition, but rejected the
second, at pp. 351-352:

"If the former proposition were established in the sense that a reasonable
insurer would have so acted, materiality would, their Lordships think, be
established, but not in the latter if the difference of action would have
been delay and delay alone. In their view, it is a question of fact in each
case whether, if the matters concealed or misrepresented had been truly
disclosed, they would, on a fair consideration of the evidence, have
influenced a reasonable insurer to decline the risk or to have stipulated
for a higher premium."

The Board was thus deciding that something which the insurers would have
wanted to know, or taken into account, and which might have led

on to further inquiries, with consequential delay, is not by itself
material, unless it would have led to a different result.

The Board then went on to apply the test which they had just laid down, at
p. 352:

"Dealing with the evidence as a whole the learned trial judge came to the
conclusion that 'if the facts as stated in the evidence of Dr. Fierheller
with relation to the condition of [the deceased] and his treatment had been
known to the defendant company, it was not at all probable that they would
have refused the premium and the issue of the policy, nor do I think they
would even have required the examination which the officials now think they
would have required.' In this finding their Lordships substantially concur,
although they would have expressed the finding somewhat differently and
would have preferred to say that had the facts concealed being disclosed,
they would not have influenced a reasonable insurer so as to induce him to
refuse the risk or alter the premium. Their Lordships, therefore, concur in
the conclusion of the trial judge that the non-disclosure or misstatement
was not material to the contract and therefore, under the law of Ontario, is
not a ground for avoiding it."

In my opinion, the Mutual Life Insurance Co. case remains the leading
authority on the application of the prudent insurer test, and the meaning of
materiality in English law. It was so treated by Lord Greene M.R. in Zurich
General Accident and Liability Insurance Co. Ltd. v. Morrison[1942] 2 K.B.
53, 58, by Jordan C.J. in Southern Cross Assurance Co. Ltd. v. Australian
Provincial Assurance Association Ltd. (1939) 39 S.R. (N.S.W.) 174, 187-188
and in the current edition of Spencer Bower, Turner & Sutton, The Law
relating to Actionable Non-Disclosure, 2nd ed., p. 32. It cannot be
relegated into the background or treated as turning solely on the provisions
of the Ontario statute. If that be right, then it provides strong support
for Mr. Beloff's argument.

Only three other cases need be mentioned. The first is Mayne Nickless Ltd.
v. Pegler [1974] 1 N.S.W.L.R. 228. I find the case confusing, since Samuels
J. does not always distinguish clearly between the proper application of the
prudent insurer test, and the separate question whether the insurer must
prove actual inducement. The test of materiality which he proposes, at p.
239, is:

"It seems to me that the test of materiality is this: a fact is material if
it would have reasonably affected the mind of a prudent insurer in
determining whether he will accept the insurance, and if so, at what premium
and on what conditions."

To this test, there could be no possible objection, except that it is hardly
very helpful since it merely restates section 18 of the Act of 1906 in
almost identical language. But in so far as Samuels J. treated the Mutual
Life Insurance Co. case as depending on the Ontario statute, I would
respectfully disagree.

In Marene Knitting Mills Pty. Ltd. v. Greater Pacific General Insurance Ltd.
[1976] 2 Lloyd's Rep. 631, Yeldham J. applied the test stated by Samuels J.
When the case reached the Privy Council, it was argued that

the test was inconsistent with the Mutual Life Insurance Co. case. Lord
Fraser of Tullybelton did not find it necessary to deal with the point,
since the undisclosed facts were obviously material, "whatever may be the
precise words in which the test of materiality is formulated" (p. 642).
Indeed the facts of the case - a policy of fire insurance in which the
plaintiffs had failed to disclose two previous fires at different premises -
afford a good example of the need for, and application of, the prudent
insurer test. As already mentioned, Samuels J.'s formulation of the prudent
insurer test is unexceptionable. The only problem in the case is that in the
previous paragraph Samuels J. had rejected any need for the insurer to show
actual inducement. It is in that respect, and in that respect alone, that
there is any conflict with the Mutual Life Insurance case. I return to that
aspect of the case when I deal with the second half of Mr. Beloff's
argument.

Finally, I should mention Barclay Holdings (Australia) Pty. Ltd. v. British
National Insurance Co. Ltd., 8 N.S.W.L.R. 514, in which the Court of Appeal
of New South Wales declined to follow the C.T.I. case [1984] 1 Lloyd's Rep.
476. Glass J.A. described it, at p. 523, as sounding "a discordant note."
Kirby P. said, at p. 519:

"As expressed by Samuels J. in Mayne Nickless Ltd. v. Pegler the issue is
not whether the insurer would have been interested in the information or
would have liked to have had it in order to consider it. It is whether the
insurer, acting reasonably, would have been affected in deciding the
critical questions mentioned. Such a test is to be preferred to one which
affords the insurer the privilege of insisting upon the disclosure of any
material whatsoever that could have had an impact on the formation of the
insurer's opinion and on its decision making process, even though, in the
end, such information was not critical to or determinative of the
conclusions finally reached: . . ."

I have not attempted to deal with more than a few of the many cases cited. I
mention them only because the Court of Appeal in the C.T.I. case regarded
them as important in establishing what Kerr L.J., at p. 492, had "always
understood to be the law." My own view is that they do not help greatly one
way or the other; but on balance are in favour of Mr. Beloff's argument. For
reasons mentioned earlier, I regard that argument as compelling.

Since writing the above, I have had the opportunity of reading in draft the
speech of my noble and learned friend, Lord Mustill. He attaches great
weight to the view of the 19th century textbook writers, as does my noble
and learned friend, Lord Goff. I have not examined these writings myself,
since they were not mentioned in argument, save in passing, by Mr. Hamilton
or Mr. Beloff.

The actual insurer

I now turn to the second half of Mr. Beloff's argument, which I can deal
with more shortly.

In Berger v. Pollock [1973] 2 Lloyd's Rep. 442, Kerr J. considered at length
the question whether an insurer can avoid a

policy for misrepresentation or non-disclosure without his going into the
witness box to say how he would himself have been affected. Relying
especially on the judgment of Scrutton L.J. in Visscherij Maatschappij Nieuw
Onderneming v. Scottish Metropolitan Assurance Co. Ltd., 27 Com.Cas. 198, he
answered that question in the negative. It should be the practice, he said,
at p. 463, to call the underwriter concerned "in all doubtful cases even if
an independent underwriter or broker is called as well."

"Otherwise one could in theory reach the absurd position where the court
might be satisfied that the insurer in question would in fact not have been
so influenced but that other prudent insurers would have been. It would then
be a very odd result if the defendant insurer could nevertheless avoid the
policy. I do not think that this is the correct interpretation of section 18
despite the generality of the language used in subsection (2)."

Twelve years later in the C.T.I. case, Kerr L.J. was persuaded to change his
mind. He said, at p. 495, that he had been wrong in his reasoning in Berger
v. Pollock, perhaps because he had not been referred to the judgment of
MacKinnon L.J. in Zurich General Accident and Liability Insurance Co. Ltd.
v. Morrison [1942] 2 K.B. 53. Kerr L.J. held that any effect on the mind of
the actual insurer was irrelevant, and the other members of the court
agreed.

Mr. Beloff submits that Kerr L.J. was right the first time, and that it
would indeed be absurd, as Kerr L.J. then thought, to allow an insurer to
avoid for non-disclosure when the undisclosed fact would, ex hypothesi, have
made no difference to him. It was common ground that in the ordinary law of
contract a party seeking to avoid for misrepresentation must show that he
was induced to enter into the contract by the misrepresentation. Why then,
asks Mr. Beloff, should the law be more favourable to an insurer seeking to
avoid a contract of insurance? Why should utmost good faith require the
assured to disclose a fact which the actual insurer would not recognise as
material?

Mr. Hamilton conceded that until the middle of the 19th century the test,
whatever it may have been, required an impact on the mind of the actual
insurer. The prudent insurer did not emerge until Ionides v. Pender, L.R. 9
Q.B. 531. Mr. Hamilton argued that the effect of Ionides v. Pender was to
substitute the prudent insurer for the actual insurer, and that the actual
insurer thereafter disappeared from the scene. I can find no support for
this theory in Blackburn J.'s language. It is true that Blackburn J. refers,
at p. 539, to "a rational underwriter." But this was by way of addition, not
substitution. It was because the issue was whether the evidence of the
underwriters called by the defendants in that case was properly left to the
jury, that is to say, evidence that underwriters do in practice charge an
increased premium or decline altogether, if the overvaluation is so great as
to make the risk speculative. It was held that this was a rational practice
and that the question was therefore properly left to the jury.

If proof of inducement were no longer required, following the decision in
Ionides v. Pender, it would be impossible to explain Brett L.J.'s citation

six years later in Rivaz v. Gerussi Brothers & Co., 6 Q.B.D. 222, 229 of
section 531 of Phillips, A Treatise on the Law of Insurance, 5th ed., p.
277, where the author says:

"Concealment in insurance is where, in reference to a negotiation therefor,
one party suppresses, or neglects to communicate to the other, a material
fact, which, if communicated, would tend directly to prevent the other from
entering into the contract, or to induce him to demand terms more favourable
to himself; . . ."

It is interesting to note that Parker L.J. emphasised the words in italics
in the above quotation from Phillips in his judgment in the C.T.I.case, at
p. 508, but did not draw what would seem to be the inevitable conclusion.

Successive editions of Arnould, A Treatise on the Law of Marine Insurance
and Average support Mr. Beloff's argument. Thus in the second edition, dated
1857, vol. I, we find at p. 584:

"208. Concealment, in the law of insurance, is the suppression of a material
fact within the knowledge of the assured which the underwriter has not the
means of knowing, or is not presumed to know; by a material factis meant,
one which, if communicated to the underwriter, would induce him either to
refuse the insurance altogether, or not to effect it except at a higher
premium."

This was, of course, before Ionides v. Pender. But there is a similar
passage in the 6th edition, dated 1887, vol. I, p. 548:

"A material fact in this connection is one which, if communicated to the
other of the parties, would induce him either to refrain altogether from the
contract, or not to enter into it on the same terms."

The editor of the 6th edition was Mr. David Maclachlan. He has always been
regarded as a writer of great authority. Clearly it did not occur to Mr.
Maclachlan that Ionides v. Pender had made any great change in the law, with
the result that inducement was no longer necessary.

In the current edition (16th ed. (1981), vol. 2, p. 475, para. 627) we find:

"A material fact is one which is likely, if communicated to the other of the
parties, to induce him either to refrain altogether from the contract or not
to enter into it except on more favourable terms. Defined in these terms,
the principle is equally applicable to the assured and the underwriter. The
contract is one uberrimae fidei, and on the plainest principles of equity
such a contract which one party has thus been induced to enter upon from his
ignorance of the thing not disclosed may not be enforced against him by the
other who has failed to disclose it."

The editors return to the same point at p. 489, para. 641:

"Even though a circumstance may be material, in the sense that it would
influence a prudent insurer, if the underwriter concerned would not have
been influenced by that circumstance if disclosed, he cannot rely on the
non-disclosure to avoid the policy" (citing Berger v. Pollock [1973] 2
Lloyd's Rep. 442).

Turning to misrepresentation, the distinction between materiality and
inducement, and the need to prove both, is even more clearly stated. Thus in
the 8th edition, dated 1909, we find, at p. 694:

"555. Even where the representation is of material facts, yet, if it
satisfactorily appears that it did not influence the judgment of the
underwriter, its falsity will be held not to avoid the policy."

In footnote (z), at pp. 694-695, the editors draw attention to the then
recently enacted section 20 of the Act of 1906. It is pointed out that on a
literal construction of section 20 inducement is not required. But such a
construction would involve

"an anomalous state of the law. For it is clear that, apart from marine
insurance, even a fraudulent misrepresentation gives no right to rescind a
contract, when it has not influenced the party to whom it was made."

This footnote is reproduced in substantially the same terms in the current
(16th) edition of Arnould, at p. 463.

Arnould's view of the law was expressly approved and adopted by Mr. Arthur
Cohen K.C., another eminent authority in this branch of the law, in
Halsbury's Laws of England, 1st ed., vol. 17, para. 809, p. 414.
Furthermore, Sir Mackenzie Chalmers in his Digest (Chalmers and Owen, A
Digest of the Law relating to Marine Insurance, 1st ed., p. 22, s. 18(1), n.
3; 2nd ed., p. 25, s. 18(1), n. 1) refers to the passage which I have
already quoted from the 6th edition of Arnould, at p. 548, as the foundation
for clause 18(1) of the Marine Insurance Bill (H.L., 1894-99).

I turn now to the main authorities in favour of Mr. Hamilton's contention
that any impact on the mind of the actual insurer is irrelevant. The first
is Cantiere Meccanico Brindisino v. Janson [1912] 3 K.B. 452, where Vaughan
Williams L.J., at p. 460, quoted Scrutton J. at first instance [1912] 2 K.B.
112 as having said that it was sufficient to avoid a contract of insurance
that a material representation was untrue, and that it was not necessary
that the underwriter should have relied upon it. The second is Zurich
General Accident and Liability Insurance Co. Ltd. v. Morrison[1942] 2 K.B.
53 where MacKinnon L.J. said, at p. 60:

"What is material is that which would influence the mind of a prudent
insurer in deciding whether to accept the risk or fix the premium, and if
this be proved it is not necessary further to prove that the mind of the
actual insurer was so affected. In other words, the assured could not rebut
the claim to avoid the policy because of a material representation by a plea
that the particular insurer concerned was so stupid, ignorant, or reckless,
that he could not exercise the judgment of a prudent insurer and was in fact
unaffected by anything the assured had represented or concealed."

The latter observation is cited as authority for a statement to the same
effect in the current edition of Spencer Bower, Turner & Sutton, The Law
relating to Actionable Non-Disclosure, 2nd ed., p. 37. In the first edition
(1915), para. 32, p. 17 Mr. Spencer Bower cited the judgment of

Turner L.J. in Traill v. Baring, 4 De G.J. & S. 318, 330 for the proposition
that it is enough to show that disclosure would have given the insurer
pause. I have already expressed my reservations about Traill v. Baring in
this connection. But in any event, it was decided 10 years before Ionides v.
Pender, L.R. 9 Q.B. 531, so it cannot provide support for the radical change
said to have been brought about by the latter case.

In Mayne Nickless Ltd. v. Pegler [1974] 1 N.S.W.L.R. 228 Samuels J. said, at
p. 239: "Accordingly, I do not think that it is generally open to examine
what the insurer would in fact have done had he had the information not
disclosed." But it is far from clear whether Samuels J. was here referring
to the prudent insurer or the actual insurer; I suspect the former. If so, I
have already given my reasons for disagreeing. Finally, Mr. Hamilton
referred us to the passage in Sir George Jessel M.R.'s judgment in Redgrave
v. Hurd (1881) 20 Ch.D. 1, 21, where it is said that inducement can be
inferred from proven materiality, as a matter of law. Despite receiving the
"embarrassing imprimatur" of so eminent a judge - I quote from Spencer Bower
and Turner, The Law of Actionable Misrepresentation, 3rd ed. (1974), p. 154
- this heresy has long since been exploded by, among others, Lord Blackburn
in Smith v. Chadwick (1884) 9 App.Cas. 187, 196.

That brings me to the language of section 18 of the Act of 1906 itself. Mr.
Hamilton relies strongly, as was to be expected, on the last sentence of
section 18(1) which provides: "If the assured fails to make such disclosure,
the insurer may avoid the contract." He points out that there is nothing in
section 18 which requires the insurers to prove inducement. Nor, he says,
can such a requirement be implied. According to the language of section
18(1) it is sufficient for the insurer to prove the materiality of the
undisclosed fact. That was clearly the view of MacKinnon L.J. in the Zurich
General Accident case and of Scrutton J. in Cantiere Meccanico Brindisino v.
Janson.

This is a formidable argument. But there is a stronger argument the other
way. It was common ground that section 18 was based on Ionides v. Pender. If
Ionides v. Pender had brought about a substitution of the prudent insurer
for the actual insurer, and if the actual insurer had thereafter "dropped
out," as Mr. Hamilton argued, then it would be natural to read section 18 of
the Act of 1906 in the manner for which Mr. Hamilton contended. But for
reasons already mentioned Ionides v. Pender brought about no such change.
This is one of those rare cases when it is permissible to look at the
pre-existing law as an aid to the construction of a codifying statute. Of
course it would have been open to Parliament to go beyond the common law.
But this does not appear to have been the draftsman's intention, since the
sentence on which Mr. Hamilton relies was omitted altogether from clause 16
of the Marine Insurance Bill as originally introduced in the House of Lords
in 1894, a clause which in all other respects is identical to section 18.

Nor does the original Bill contain the second sentence in what was to become
section 20. In the case of a misrepresentation in the ordinary law of
contract it has always been necessary for the party seeking to avoid the
contract to show that he relied on the misrepresentation. It seems most
unlikely that Parliament, by enacting the second sentence of section 20,

intended to exclude this rule of common law for no apparent reason. It is
much more likely that the intention was to codify the common law on
materiality, without touching the common law on inducement. This is
consistent with the comment on section 17 in Chalmers and Owen, A Digest of
the Law relating to Marine Insurance, 1st ed., p. 22; 2nd ed., p. 24:

"The contract is often said to be rendered void by concealment or
misrepresentation, but it is clear that it is only voidable at the option of
the party prejudiced, and that the ordinary rules of law as to voidable
contracts apply to insurance."

There is no presumption that a codifying Act covers the whole of the
relevant common law: see Shiloh Spinners Ltd. v. Harding [1973] A.C. 691,
724-725, per Lord Wilberforce. In any event, section 91(2) of the Act
preserves the common law save in so far as it is inconsistent with the
express provisions of the Act. The short answer to Mr. Hamilton's argument
is therefore that section 18 does not exclude inducement. It is dealing with
a different subject matter altogether.

As for what MacKinnon L.J. said in the Zurich General Accident case [1942] 2
K.B. 53, 60, it is reassuring to notice that his observations were obiter,
and not reflected in the judgments of Lord Greene M.R. and Goddard L.J. It
is sometimes forgotten that Lord Goddard's knowledge of insurance law was
almost as great as that of MacKinnon L.J.

I conclude that what Kerr L.J. said in Berger v. Pollock [1973] 2 Lloyd's
Rep. 442, 463 was a correct statement of the law, and that his second
thoughts in the C.T.I. case [1984] Lloyd's Rep. 476, 495 were erroneous. It
follows that Mr. Beloff is entitled to succeed on the second half of his
argument, as well as the first.

If your Lordships accept this conclusion, the position will be as follows.
Whenever an insurer seeks to avoid a contract of insurance or reinsurance on
the ground of misrepresentation or non-disclosure, there will be two
separate but closely related questions: (1) Did the misrepresentation or
non-disclosure induce the actual insurer to enter into the contract on those
terms? (2) Would the prudent insurer have entered into the contract on the
same terms if he had known of the misrepresentation or non-disclosure
immediately before the contract was concluded? If both questions are
answered in favour of the insurer, he will be entitled to avoid the
contract, but not otherwise.

The evidence of the insurer himself will normally be required to satisfy the
court on the first question. The evidence of an independent broker or
underwriter will normally be required to satisfy the court on the second
question. This produces a uniform and workable solution, which has the
further advantage, as I see it, of according with good commercial common
sense. It follows that the C.T.I. case was wrongly decided, and should be
overruled.

Facts

I come at long last to the facts. Mr. Hamilton argued that Waller J. [1992]
1 Lloyd's Rep. 101 should have found that Mr. Robinson failed to disclose
the figures for 1977 to 1979. It was common ground that these

figures were material. Indeed, the experts were agreed that they gave a much
better indication of the business than the figures for 1980 and 1981. Mr.
Hamilton's difficulty is that the judge found as a fact, at p. 106, that the
figures were available for Mr. O'Keefe to look at, and that they were "a
perfectly fair presentation" of the years in question. It was no part of Mr.
Robinson's duty to compel Mr. O'Keefe to look at the figures if he did not
wish to do so. Mr. Hamilton relied on the comment that Mr. Robinson had
broked the risk in a way which was designed to concentrate Mr. O'Keefe's
mind on the later years, and that, in so doing, he had successfully taken
Mr. O'Keefe's "eye . . . off the ball" (per Waller J., at p. 114). But there
was never any suggestion of bad faith on the part of Mr. Robinson; and the
comments on which Mr. Hamilton relied cannot undermine the judge's finding
that the presentation was perfectly fair. This was the view of the Court of
Appeal. It is sufficient to say that I agree.

Turning to 1981, it was common ground that Pan Atlantic failed to disclose
the two additional claims which had already been advised before Mr. O'Keefe
signed the slip on 13 January 1982. The question is whether that
non-disclosure was material. Mr. Beloff mounted an elaborate attack on the
judge's finding in that respect. It occupies some 25 pages of the
appellants' printed case. The attack had three prongs: (1) Since the judge
applied the wrong test of materiality, his findings do not help one way or
the other. (2) The losses disclosed for the years 1977 to 1979 were so bad
that they eclipsed the two undisclosed 1981 claims. No prudent insurer would
have accepted the risk on any terms, so the undisclosed claims would have
made no difference. (3) The findings made by the judge, when applying the
prudent insurer test which he favoured, were in any event vitiated by two
errors of fact.

It is convenient to deal first with the factual errors. The judge assumed
that the undisclosed claims fell in the fourth quarter of 1981, and that
there had therefore been a very rapid acceleration of claims in that quarter
from U.S.$235,768 at the end of the third quarter to U.S.$468,168 at the end
of the fourth quarter. In truth, over three-quarters of the two undisclosed
claims fell, not in the fourth quarter, but in the second and third
quarters. So there was not the rapid acceleration in the fourth quarter
which the judge assumed.

Secondly, both the judge and the Court of Appeal attached importance to
tables which the judge himself prepared showing the development of loss
ratios over the years 1980-82. In particular the 1980 year showed a loss
ratio of 15 per cent. at 8 December 1980, whereas the 1981 year showed a
loss ratio of 30 per cent. at 30 September 1981. The tables thus appeared to
show a doubling of the loss ratio for the 1981 year at 12 months' maturity.

But, as Mr. Beloff pointed out, the tables are not comparing like with like.
If one takes year-end figures, the loss ratio for the 1980 year at 31
December 1980 was not 15 per cent. but 48 per cent., and the loss ratio for
the 1981 year at an equivalent maturity was 58.8 per cent. Moreover, the
judge seems to have made certain assumptions as to the premium figures for
the years in question, which may well not be accurate. If the premiums are
corrected, then the ratio for the 1980 year becomes 59.8 per

cent. and the ratio for the 1981 year drops to 45 per cent. or perhaps 54.4
per cent., in which case the 1981 year was showing an improvementon the 1980
year. None of this was explored in evidence. Mr. Beloff's criticism is that
in attempting to draw any conclusion from the tables of loss ratios the
judge was engaged on a frolic of his own.

I agree with Mr. Beloff that the tables do not, on the face of them, compare
like with like. But the judge was fully alive to this point. Indeed, it
reinforced his conclusion as to the materiality of the 1981 non-disclosure
that the "dramatic rise" in the fourth quarter occurred not in one year only
but in two years running.

As for the rise in the fourth quarter of 1981, I accept that it was not
quite so "dramatic" as the judge thought. But I do not accept that this
vitiates the judge's conclusion. The fact remains that the figure disclosed
by Mr. Robinson was U.S.$235,768, whereas it should have been U.S.$469,168.
The judge was entitled to regard the difference as material, on the test
which he applied, whether the undisclosed claims fell in the fourth quarter
as he thought, or in earlier quarters. Indeed it might be thought that the
earlier the claims were advised the more serious the outlook became. On any
view, the judge was entitled to regard a loss ratio of 58u8 per cent. at the
end of the first year of what was expected to be long-tail business as very
high indeed, and something which a prudent insurer would want to know about
or take into account. It follows that it would not be right to disturb the
judge's finding on the test which he applied, and the Court of Appeal were
right to take that view.

The problem remains that the judge did not apply the right test. What
follows? Mr. Beloff argues that it was for Pine Top to obtain the necessary
findings on the basis of whatever test should turn out to be right. Since
they did not obtain such findings, their defence must fail.

Whether or not this argument is theoretically correct, I regard its
application on the facts of the present case as artificial, impractical and
unfair. I take that view for two reasons. The first is that the Court of
Appeal, applying the increased risk test, found that the non-disclosure of
the two additional 1981 claims was material. For reasons already mentioned,
there is little or no practical distinction between the increased risk test,
advocated by the Court of Appeal, and the test which I believe to be
correct. Secondly, Pan Atlantic made a specific written request for a
finding that "an objective prudent reinsurer would not have come to any
different final decision if the [additional] losses had been disclosed and
included in the loss statistics." But the judge made no such finding,
presumably because, in his view, it was not justified on the evidence. In
those circumstances I would be unwilling to decide the case against Pine Top
on the ground that they have failed to discharge the burden of proof. Nor
would I regard it as appropriate to remit the case to the judge for a
further hearing, so long after the event.

So it comes to this: that although the judge applied the wrong test in law,
as he was obliged to do in the light of the C.T.I. case [1984] 1 Lloyd's
Rep. 476, his conclusion on the facts must be accepted.

The same applies to his finding [1992] 1 Lloyd's Rep. 101, 113 that the
non-disclosure "might well have influenced" the actual insurer, Mr. O'Keefe.
Mr. Beloff argued that this falls short of proof of

inducement. He may be right. But once again Pan Atlantic had asked for a
finding that Mr. O'Keefe would not have taken the additional claims into
account "in his decision." Once again the judge declined to make that
finding.

The second prong of Mr. Beloff's three-pronged attack, which raises a novel
and teasing philosophical question, does not call for separate
consideration.

The overall result is that the appellants win on the law but lose on the
facts. It follows that I would dismiss the appeal. I recognise that this
result may be less than satisfying to the plaintiffs. But even if the case
had gone back to the judge, and they had won on the facts, it might well
have proved a Pyrrhic victory.

DISPOSITION:
Appeal dismissed.

Plaintiffs to pay three-quarters of defendants' costs in House of Lords,
amount to be certified by Clerk of Parliament if not agreed.

SOLICITORS:
Solicitors: Ince & Co.; Alsop Wilkinson.

M. G.

But, of course, I am speculating.
I wonder how much she drank - Or more importantly how much the insurance
companies let you drink before they deny liability.

I'm curious how insurers would react if for example there was a fire at the
Munich Oktobrefest.

But, of course, I am speculating.
I wonder how much she drank - Or more importantly how much the insurance
companies let you drink before they deny liability.

I'm curious how insurers would react if for example there was a fire at the
Munich Oktobrefest.