Supreme Court of Canada

McClelland & Stewart Ltd. v. Mutual Life, [1981] 2 S.C.R. 6

Date: 1981-06-22

McClelland and Stewart Limited (Plaintiff-Respondent) Appellant;

and

The Mutual Life Assurance Company of Canada (Defendant-Appellant) Respondent.

1981: February 26; 1981: June 22.

Present: Martland, Dickson, Estey, Chouinard and Lamer JJ.

ON APPEAL FROM THE COURT OF APPEAL FOR ONTARIO.

Contracts—Insurance—Effective date of policy—Declaration found in application stating policy “shall become effective” when delivered, i.e. February 28—Policy stating date, consequent upon special instructions, to be January 23—Limitation clause effective if suicide within two years of effective date of policy—Interpretation of policy—The Insurance Act, R.S.O. 1970, c. 224, s. 148(2).

Appellant was the irrevocable beneficiary of a life insurance policy contracted by an author commissioned by them. The application was written on January 30, 1968 but, on the advice of the agent for reasons related to the premium payable, was backdated to January 23, 1968. The policy was delivered to the author on February 28, 1968. Following the author’s suicide on January 31, 1970, appellant claimed the policy proceeds. The assurance company refused to pay, however, relying on a self-destruction clause that limited the company’s liability to the amount of premiums paid if the insured died by his own hand within two years of the effective date of the policy. The words “effective date” were not defined in the policy, and the determination of that date formed the issue on appeal. The insurers maintained that the effective date was February 28, 1968, relying on a statement in a Declaration in the application that a policy “would become effective when delivered to and accepted” by the insured. Appellant argued that January 23, 1968 was the effective date.

Held (Martland and Estey JJ. dissenting): The appeal should be allowed.

Per Dickson, Chouinard and Lamer JJ.: It was incongruous that the policy would be partly in effect on one date and partly in effect on another date. Notwithstanding the wording of the Declaration, the language of the policy did not lead inexorably to the construction advanced by the assurance company. The policy was

[Page 7]

entirely structured around January 23. Upon the true construction of the entire contract, for the purpose of calculating any of the time periods contemplated, including the application of the self-destruction clause, the effective date of the policy was January 23, 1968.

Per Martland and Estey JJ., dissenting: There was no distinction between “become effective” used in the Declaration and “effective date” used in the self-destruction clause. Both denote the time at which the coverage of the life of the deceased by the insurer begins. The Declaration, the self-destruction clause and the incontestability clause all referred to the same point of commencement for the computation of specific times in each clause in question. The clash was between February 28 as the effective date for these two clauses and January 23 for the “policy years” computation and the consequential computation of the term of the contract itself. The two periods of time and the separate provision for their measurement could both exist in harmony within the one contract of insurance. The anomaly, if any, was for the demonstrable benefit of the insured—a lower premium. In the result, the contract read as a whole required the measurement of the two-year period in the self-destruction clause from February 28, 1968, however the premiums may have been calculated.

The doctrine of contra proferentem did not come into play as there was no ambiguity in the language of the contract. The clause in question followed almost precisely the wording of the insured’s Declaration in the application for insurance. Reading the two provisions together, the meaning was clear and unequivocal.

APPEAL from a decision of the Court of Appeal for Ontario reversing a judgment of Labrosse J. Appeal allowed, Martland and Estey JJ. dissenting.

David I. Bristow, Q.C., and R.J. Otter, for the appellant.

Colin Campbell, Q.C, and Claude Gingras, for the respondent.

The reasons of Martland and Estey JJ. were delivered by

ESTEY J. (dissenting)—The Court of Appeal of Ontario, reversing the trial judge, found that the self-destruction clause of a life insurance policy measured the time period therein set forth from the date of delivery and acceptance of the policy of insurance and not from the date from which the

[Page 8]

insurance premiums were calculated. The facts are simple. The deceased, Terence Robertson, applied to the respondent through an agent for life insurance, on January 30, 1968. After medical examinations and the completion of forms in connection with aviation coverage, a policy was issued on February 27, 1968 and delivered on or about February 28, 1968. The appellant was named as beneficiary of the policy. The trial judge found, and the parties do not dispute the finding, that the insured took his own life on January 31, 1970.

The directly relevant portion of the “general provisions” included in the life insurance policy issued by the respondent provides as follows:

SELF-DESTRUCTION—If the life insured shall, whether sane or insane, die by his own hand or act

(a) within 2 years of the effective date of this policy, or any reinstatement thereof, the liability of the Company shall be limited to an amount equal to the premiums paid;

(b) within 2 years of the effective date of any insurance subsequently added to the policy, the liability of the Company in respect of such insurance shall be limited to an amount equal to the premiums paid for such insurance.

It will thus be seen that the date of death falls within the two-year period stipulated by the self‑destruction clause if the period is measured from the date of delivery of the policy, on February 28, 1968, but is beyond the two-year period if the latter is measured from January 23, 1968 which is said on the back of the policy to be the base date for the computation of premiums. There is no dispute that, on making the application for insurance, the deceased agreed to a proposal made for his benefit by the agent to calculate the premiums from January 23, 1968 because to do so would qualify the deceased as an insured aged 46. The practice in the respondent company, if not the industry, at that time was to classify an applicant for insurance according to the actuarial tables using as his age the nearest birth date, forwards or backwards. Since the deceased was born on July 25, 1921, the adoption of January 23, 1968 as the base date for premium computation allowed the respondent to classify the insured as 46 years of age for the purposes of the insurance policy, and

[Page 9]

this resulted in a lower premium to be paid by the insured throughout the life of the contract. In fact, the application for insurance signed by the deceased includes the following, probably written by the agent:

SPECIAL INSTRUCTIONS

“Please date insurance age 46”

In the “Head Office” box adjoining those instructions there has been inserted in handwriting, probably at Head Office:

“Dated 23-1-68 for age 46”

This practice had, of course, the necessarily disadvantageous effect of causing the first premium to be paid for a period in which there was no outstanding coverage; namely, for the period from January 23, 1968 to the date of the commencement of the contract in law, February 28, 1968. Since the deceased could not have died in the intervening period (because a death in such period would preclude a contract arising), there was no coverage.

The only statutory provision directly referable to the issue now before the Court is s. 154 of The Insurance Act, R.S.O. 1970, c. 224:

154.—(1) Subject to any provision to the contrary in the application or the policy, a contract does not take effect unless,

(a) the policy is delivered to an insured, his assign or agent, or to a beneficiary;

(b) payment of the first premium is made to the insurer or its authorized agent; and

(c) no change has taken place in the insurability of the life to be insured between the time the application was completed and the time the policy was delivered.

(2) Where a policy is issued on the terms applied for and is delivered to an agent of the insurer for unconditional delivery to a person referred to in clause a of subsection 1, it shall be deemed, but not to the prejudice of the insured, to have been delivered to the insured.

Since virtually every item on the front page of the policy must be considered, it is convenient to set out the document here:

[Page 10]

The Mutual Life Assurance Company of Canada

HEAD OFFICE: WATERLOO, ONTARIO

INSURES THE LIFE OF

—TERENCE ROBERTSON—

(herein called the assured)

as follows:

PRINCIPAL SUM ASSURED: $100,000. payable on the death of the assured if the assured dies prior to January 23, 1988.

BENEFICIARY: McClelland and Stewart Limited

,if living, otherwise the C3tatc of the assured.

RENEWAL PRIVILEGE: This policy is written on the 5 Year Renewable Term plan. This plan provides that the policy may be renewed automatically without evidence of insurability by the payment of the premiums specified below for successive periods of 5 years each up to attained age 65 of the assured.

THIS POLICY MAY BE CONVERTED not later than the end of the 15th policy year as provided in the Conversion Privilege.

FIRST PREMIUM: $91.42 payable on or before the delivery of this policy.

SUBSEQUENT PREMIUMS payable during the lifetime of the assured on the 23rd day of February and monthly thereafter in every year until the premiums for 20 policy years have been paid, as follows:

Subsequent Premium Commencing A.C.P.

$ 91.42 February 23, 1968 $ 88.83

$ 134.71 January 23, 1973 $ 130.90

$ 212.88 January 23, 1978 $ 206.86

$ 339.65 January 23, 1983 $ 330.03

AGE OF ASSURED: 46 Age Admitted

POLICY YEARS date from January 23, 1968 and anniversaries thereof.

THIS POLICY PARTICIPATES in the surplus distribution of the Company.

THE PROVISIONS set forth on the following pages form a part of this policy.

SIGNED AND SEALED at Waterloo, Ontario, February 27, 1968.

“W.H.[?]” “K.R. MacGregor”

Vice-President and President Secretary

“A.G.S.”

[Page 11]

Of particular importance to the issue here is the declaration signed by the deceased in the application for insurance:

DECLARATION: I apply for the insurance as described and declare that the above and all written statements and answers given in connection with this application are full, complete and true. I agree (1) that subject to the attached “Receipt for Cash Payment,” any policy issued on this application shall become effective when delivered to and accepted by me, if the initial premium has first been paid and no change has taken place in the insurability of my life since the completion of any evidence of insurability in connection with this application; and (2) that my acceptance of the policy shall be a ratification by me of any amendments made to this application by the Company in the space headed “Amendments.”

By s. 148(2) of The Insurance Act, supra, the application for insurance and the policy are both comprised in the insurance contract on its issuance.

The learned trial judge found in effect that there were two operative dates in the policy on which, for different purposes, the policy “became effective”. In the result the trial court found that:

…there is a difference between the date on which the policy became effective for purposes of coverage as opposed to the date on which it became effective for the purpose of determining the application of the exclusionary clause.

The trial court thereupon found that the effective date of the policy for the purpose of the application of the self-destruction clause was January 23, 1968. In reaching this conclusion, the court applied the contra proferentem rule. The Court of Appeal, relying upon the Declaration included in the application for policy, stated:

A fair reading of the policy which includes the declaration of the Applicant is the policy only became effective upon delivery and acceptance. This would be February 28th, 1968. As there can be only one effective date that date should be used in the construction of the self-destruction clause.

and thereupon reversed the result.

[Page 12]

Even the most cursory examination of the policy of insurance reveals that January 23, 1968 was the date upon which at least four time elements of the contract were based:

(a) premium calculations;

(b) the duration of coverage;

(c) the conversion privilege;

(d) the “policy years” comprised in the term of the contract of insurance.

In the result, the only periods of time to be examined in the contract in which January 23 is not clearly used as the time base are the two-year periods in the incontestability clause and the self-destruction clause itself.

The policy provides that the “policy years” of the insurance contract “date from January 23, 1968”. If the expression “two years” in the self-destruction clause means “two policy years”, the appellant succeeds. If the expression refers to two years from any of the other possibly applicable dates such as February 28, the appellant fails. Section 154 of The Insurance Act, supra, is not of as much assistance as might at first glance appear. It is prefaced with the expression “Subject to any provision to the contrary in the application or the policy...”. This puts the question back to the interpretation of these two documents, the application and the policy.

Since the term “effective date” is not defined in either the contract or the statute, one must look to other terms of the contract for possible assistance in discerning the meaning of the expression. In the “incontestability” clause, it is provided that the statement made in the application for insurance “shall be incontestable after the insurance…has been in force for two years during the lifetime of the insured”. This clause requires that the period of two years be completed before the death of the insured and after the insurance has been “in force”. It is clear, for example, that if the deceased had died before February 28, 1968, there would be no liability in the respondent. The risk had not been accepted, the contract of insurance had not been executed and delivered by the respondent nor had it been accepted by the deceased insured. There is no “provision to the contrary” in the

[Page 13]

application or in the policy and therefore coverage commenced only on February 28, 1968. Therefore, any action by the insured under the incontestability clause would have had to have been brought before February 28, 1970 as by that date the contract would have been “in force” for two years. In the circumstances of this appeal, the insurance coverage clearly had not been “in force” for two years.

The wording in issue, however, is the expression “effective date” in the self-destruction clause. It is significant that the expression “in force” is not therein employed. Instead we find the words “within two years after the effective date”. This wording does not expressly make the existence of “insurance” during the two-year period a condition or requirement of this clause. The parties by a contractual term may revise the application of s. 154 as to when “a contract does not take effect”. The contract clearly cannot come into effect in the sense of creating coverage for the insured if the insured is not living at the time of the alleged commencement of the agreement. The statute, however, does not prevent, and indeed contemplates, that the parties may, by a term in the contract, agree that some or all of the terms of the contract shall “take effect” or run from an earlier date than the date on which the contract is delivered to the insured or his agent. Thus a contract may “take effect” before it is actually “in force” if it so provides. Here the parties have expressly agreed that in four instances, the contract term, the policy years, the conversion right and the premium computation, time shall run from January 23. Only with respect to the clause in question here and the incontestability clause is it argued by the respondent that another commencement or measurement date shall apply.

In the result, on either argument, the deceased paid a premium calculated with reference to a period when the contract was not “in force”; that is, if death had occurred during such a period, no claim could arise. Such an arrangement was clearly established at the behest of the deceased, or at

[Page 14]

least with his approval, for it provided a lower premium rate for the nineteen year eleven month life of the contract if the contract had continued in effect. This premium obligation required monthly payments on the 23rd day of each month commencing February 23, 1968 (before the date of delivery and hence before the effective date urged by the respondent company) and continuing during “20 policy years” which ended on January 23, 1988.

Against all the above considerations militating in favour of the appellant’s position, the respondent advances the Declaration of the deceased in the application form (and which as stated forms part of the contract of insurance) set out above in full. The operative words for our purposes are “any policy…shall become effective when delivered to and accepted by me”. Clearly the respondent is under no risk until these events transpire. No suggestion has been made in argument that this declaration is in any way qualified by s. 154. The neat and narrow question therefore is, what is the difference, if any, between “shall become effective when” in the Declaration, and “within 2 years of the effective date” in the self-destruction clause?

I am unable to make out any distinction between the expressions “become effective” and “effective date”. Both denote the time at which the coverage of the life of the deceased by the insurer begins. The term “shall become effective” in the declaration is co-extensive in meaning with “has been in force” in the incontestability clause if the effective date at the beginning of the coverage is taken to be the effective date for the commencement of the period when the contract is in force in the latter clause. Hence all three provisions, that is the declaration, the self-destruction clause, and the incontestability clause, all refer to the same point of commencement for the computation of specific times in each clause in question. The clash is between February 28 as the effective date for these two clauses and January 23 for the “policy years” computation and the consequential computation of the term of the contract itself. These two periods

[Page 15]

of time and the separate provisions for their measurement can, in my view, both exist in harmony within the one contract of insurance. The anomaly, if there is one, is created here for the demonstrable benefit of the insured. The application form reveals the step-by-step process by which the lower premium was achieved by the adoption of a computation base different by a few weeks from what it would have been were those computations based on a date on or about the commencement of coverage. In the result the contract when read as a whole requires the measurement of the two-year period in the self-destruction clause from February 28, 1968, however the premiums may have been computed.

The appellant relies (as did the trial judge) on the doctrine of contra proferentem. That principle of interpretation applies to contracts and other documents on the simple theory that any ambiguity in a term of a contract must be resolved against the author if the choice is between him and the other party to the contract who did not participate in its drafting. The rule or principle, however, only comes into play if there is an ambiguity in the language of the contract. For those parts of the contract, as in the case of life insurance, which are mandated by legislation, the rule does not operate. Vide: Consolidated- Bathurst Export Limited v. Mutual Boiler and Machinery Insurance Company[^1]. Here, assuming the statute does not limit the latitude of the author of the contract in the drafting of the self-destruction clause, the rule would come into play providing the wording of the clause leads to more than one reasonable interpretation. Schroeder J.A., speaking for the Court of Appeal of Ontario in Losier v. St. Paul Mercury Indemnity Company[^2], after enunciating the contra proferentem rule at p. 98, went on to say:

Since exceptions are inserted in the policy mainly for the purpose of exempting the insurers from liability for a loss which, but for the exception, would be covered by the policy, they are construed against the insurer with the utmost strictness; and it is the duty of the insurers to

[Page 16]

except their liability in clear and unambiguous terms.

I respectfully agree. However, the principle has no application here. The clause in question follows almost precisely the wording of the insured’s declaration in the application for the insurance. Reading the two provisions together, the meaning is clear and unequivocal.

The parties both refer to the judgment of the Court in New York Life Insurance Company v. Dubuc[^3]. In that case, the policy was by its terms expressed “to take effect as of date policy is written”. Elsewhere in the application form it was provided that “the insurance…shall not take effect unless and until the policy is delivered…’’. These two conditions in the contract were reconciled by Newcombe J. at p. 277:

Although, by the application, the policy was not to take effect unless and until delivered, there is nothing to indicate an intention that, when delivered, it should not operate according to its terms, and therefore as of 26th June.

The case, based as it is, on the precise wording of the contract then before the Court, is of no assistance here.

I would therefore dismiss the appeal with costs.

The judgment of Dickson, Chouinard and Lamer JJ. was delivered by.

DICKSON J.—Terence Robertson committed suicide in New York City on January 31, 1970. He was an author who had been commissioned by the publishers McClelland and Stewart Limited to prepare and write a biography. One of the terms of the agreement between Mr. Robertson and the publishers provided that Mr. Robertson would obtain a policy on his life in the amount of $100,000. Pursuant to his undertaking, Mr. Robertson applied to Mutual Life Assurance Company of Canada for such a policy and a five-year renewable term policy was issued in the principal sum of $100,000 payable on the death of Mr. Robertson if

[Page 17]

he died prior to January 23, 1988. McClelland and Stewart Limited was named as irrevocable beneficiary. The policy was signed and sealed at the Head Office of the assurance company on February 27, 1968 and mailed to Mr. Robertson by the insurance agent on February 28, 1968.

Following Mr. Robertson’s death McClelland and Stewart Limited claimed the policy proceeds. The assurance company refused to pay, relying upon the following provision in the policy:

SELF-DESTRUCTION—If the life insured shall, whether sane or insane, die by his own hand or act

(a) within 2 years of the effective date of this policy, or any reinstatement thereof, the liability of the Company shall be limited to an amount equal to the premiums paid;

The right of the appellant to recover depends upon whether Mr. Robertson died within two years of the “effective date” of the policy. The words “effective date” are not defined in the policy. The assurance company relies, however, on the Declaration found at the end of Part 1 of the application for the policy. The Declaration reads:

I apply for the insurance as described, and declare that the above and all written statements and answers given in connection with this application are full complete and true. I agree (1) that subject to the attached “Receipt for Cash Payment”, any policy issued on this application shall become effective when delivered to and accepted by me, if the initial premium has first been paid and no change has taken place in the insurability of my life since the completion of any evidence of insurability in connection with this application; and (2) that my acceptance of the policy shall be a ratification by me of any amendments made to this application by the Company in the space headed “Amendments.” [Emphasis added]

The assurance company says simply that according to the Declaration (which forms part of the contract of insurance, The Insurance Act, R.S.O. 1970, c. 224, s. 148(2)) the policy “shall become effective when delivered” and as this date was not earlier than February 28, 1968, the death of Mr. Robertson on January 31, 1970 occurred

[Page 18]

within “2 years of the effective date of this policy” and the self-destruction clause applies. How can the date when the policy “becomes effective” be any different from the “effective date” of the policy? The two dates must be one and the same. The parallelism of language undoubtedly lends great force to this argument.

The matter is not that simple. At the time the policy was issued it was the practice of insurance companies to determine the age of an assured to the closest six months. Mr. Robertson’s birthday was July 25. The application for insurance was written on January 30 on which date he was judged for insurance purposes to be age 47. The agent explained to him that by backdating the application to approximately January 23, he would be insurance age 46 rather than 47 and this would result in some saving in premium payments. Mr. Robertson agreed. Accordingly, in a space headed “Special Instructions”, just above the Declaration, the agent wrote “Please date insurance age 46”. In an adjoining space headed “Amendments (for Head Office Use only)” the words and figures, “Dated 23-1-68 for age 46” appear. On the face of the policy the following is written:

PRINCIPAL SUM ASSURED: $100,000, payable on the death of the assured if the assured dies prior to January 23, 1988.

POLICY YEARS date from January 23, 1968 and anniversaries thereof.

The question which must now be decided is whether the “effective date of this policy” is February 28, 1968 (the date of delivery), as the Declaration would suggest, or January 23, 1968 by reason of the wording of the policy when issued, consequent upon the “Special Instructions—Please date insurance age 46” and “Amendments (for Head Office Use Only)—Dated 23-1-68 for age”. If February 28, 1968 is the effective date of the policy the appellant McClelland and Stewart Lim-

[Page 19]

ited fails; if the date is January 23, 1968, the appeal succeeds.

Taken alone and read without consideration of the scheme of the policy the kindred language of the self-destruction clause and the Declaration undoubtedly create a formidable argument in support of the case of the assurance company. It is plain however these cannot be read in an isolated and disjunctive way. The question before us is not to be determined on a mechanical reading of two phrases set apart, but rather on a reading of the policy and the Declaration in entirety.

Let us examine each. The Declaration is clearly concerned with the date upon which the policy comes into force, in the sense of when the assurance company is on the risk. That date is the date of delivery of the policy, the initial premium having been paid, and no change having taken place in the insurability of the life of the applicant. At that time the policy comes into force and the company is at risk. When we turn then to the self-destruction clause, we are concerned with a period of time during which the liability of the company is limited in the event of suicide. That period starts to run on the effective date of the policy. The period would normally commence at the delivery of the policy but in this case the company and the assured agreed that the policy would be antedated, that it would have retroactive effect, to the date upon which the assured could be said to be insurance age 46, i.e. January 23. That was made the effective date of the policy for all purposes except, arguably, the self-destruction clause. It seems to me incongruous, to say the least, that the policy would be partly in effect on one date and partly in effect on another date, that for some purposes the effective date of the policy would be one date and for other purposes another. Notwithstanding the wording of the Declaration I do not think that the language of the policy leads inexorably to the construction for which the assurance company contends.

It is apparent upon reading the policy that it is entirely structured around the date January 23. The principal sum is payable on the death of the assured prior to January 23, 1988. The duration of

[Page 20]

coverage is to a January 23 date. Policy years date from January 23, 1968. Premium calculations date from January 23. The conversion privilege is related to January 23. I have difficulty concluding it was the intention of the parties that the effective date of the policy for the purposes of the self-destruction clause would be February 28 but for all other purposes the effective date of the policy was January 23.

I am of the opinion that, upon the true construction of the entire contract, for the purpose of calculating any of the time periods contemplated, including the application of the self‑destruction clause, the effective date of the policy was January 23, 1968.

I would allow the appeal, set aside the judgment of the Court of Appeal for Ontario and restore the judgment at trial, with costs to the appellant throughout.

Appeal allowed with costs.

Solicitors for the appellant: Hume, Martin & Timmins, Toronto.

Solicitors for the respondent: McCarthy & McCarthy, Toronto.

[^1]: 1979 10 (SCC), [1980] 1 S.C.R. 888.

[^2]: 1957 396 (ON CA), [1957] O.W.N. 97.

[^3]: 1926 30 (SCC), [1926] S.C.R. 272.