Plaintiff, as holder of one of defendant’s insurance • policies containing both life and disability insurance, sues under section 195 of the Civil Practice Act on behalf of herself and all other holders of similar policies issued by defendant between 1917 and 1934. The question presented is whether, during the years 1931 to 1935, the defendant failed to apportion equitably its divisible surplus among its policies or made an unlawful discrimination when it paid smaller dividends on its policies providing life and disability insurance than on policies, otherwise similar, providing only life insurance.
While the amount in dispute between the parties is trivial, the vast majority of New York Life policies contain disability benefits, approximately 1,600,000, as against 1,000,000 providing life insurance only; hence, if a legal wrong was done to plaintiff a similar wrong was done to the 1,600,000 holders of similar policies, and this alleged wrongful discrimination against her and all other holders of similar disability policies involves, for the years in question alone, approximately $15,000,000 in dividends.
The agreed statement of facts contains, in two volumes, a vast mass of detailed information regarding the types of policies involved; the basis, nature and principles of mutual fife insurance; the history and experience of the company; the basis, elements and factors in the computation of premiums, the creation and ascertainment of divisible surplus and the apportionment of dividends, with specific illustrations of the application of such factors over a wide range of experience; the company’s detailed policy records, long tables of figures, statistics and actuarial computations, etc. Such broad and complicated factual basis cannot, within reasonable compass, be discusséd _m any detail in a judicial opinion. Accordingly, the court will give only such summary of the salient facts as is necessary intelligently to present and discuss the issues, before pronouncing the conclusion of the court thereon.
Plaintiff originally had a $2,000 life plus disability policy issued by defendant on June 13, 1927. Subsequently, she surrendered this policy for two policies of $1,000 each, on one of which premiums are payable annually, and on the other semi-annually. For con*122venience in making comparisons and contrasts with other policies,' plaintiff’s $1,000 twenty-payment policy providing both life insurance and disability insurance, for a total annual premium of $30.30, will be called herein the “ disability policy; ” and a policy issued at the same time and the same age and under generally the same conditions, but providing only life insurance and without provisions for disability, will be called the “ non-disability policy.”
Plaintiff’s disability policy contains: (1) Life insurance, i. e., the company’s promise to pay $1,000 on the death of the insured (and the terms and conditions relating to such life insurance are identical with those contained in all other similar life insurance agreements made at the same time by defendant, whether in policies providing only life insurance or in policies providing both life and disability insurance); and, in addition, (2) disability benefits under the terms of which, on receipt of proof that the insured is totally and permanently disabled before sixty, the company agrees to pay the insured ten dollars per month each month (i. e., one per cent of the face of the policy each month) and also agrees to waive payment of premiums falling due during the period of disability. The annual premium is stated to be $30.30. This total premium includes the fife insurance and the disability benefits. The policy expressly provides:
“ The total premium stated on the first page hereof includes an annual premium of $2.96 for Disability Benefits.
“ Any premium due on or after the anniversary of the policy on which the age of the Insured at nearest birthday is sixty, will be reduced by the amount of premium charged for Disability Benefits. Upon written request signed by the Insured and upon return of this policy for proper indorsement, the Company will terminate this provision and thereafter the premium shall be reduced by the amount charged for Disability Benefits.”
Under these terms, on termination of the disability benefits, plaintiff’s premium would accordingly be reduced to $27.34, the exact amount of the premium on a similar non-disability policy. The policy’s provision as to dividends reads as follows:
“ Participation in Surplus — Dividends
“ The proportion of divisible surplus accruing upon this Policy shall be ascertained annually. Beginning at the end of the second insurance year, and on each anniversary thereafter, such surplus as shall have been apportioned by the Company to this Policy shall at the option of the Insured be either ” paid in cash; applied toward payment of premiums; applied to purchase of participating paid-up addition to the sum insured; or left to accumulate as a dividend deposit.
*123No default has occurred in plaintiff’s performance of the terms and conditions of her policies, and such policies are in full force and effect.
The defendant New York Life Insurance Company (hereinafter referred to as “ the Company ”) is and has been since its inception in 1845 a mutual life insurance company with no capital stock, engaged in the insurance business in what is called the co-operative or mutual plan. Under such mutual plan, all the members pay regular fixed sums or premiums into the company fund, the sums being based on age and character of insurance desired; the company’s officers manage the money, investing and reinvesting, paying out death and disability claims, matured endowments, surrender values, loans, taxes, expenses, etc., and set aside as a reserve fund the amount required by law, calculated mathematically, to be held for future protection of the members, and after setting aside funds sufficient to cover other liabilities (such as unpaid death and disability claims and other amounts embraced in a contingency reserve), return what is left over to the members annually, as their equitable share of divisible surplus. The share so distributed is called a dividend.
Since 1845 the company has issued policies covering over 6,000,000 policyholders whose lives have been insured in sums aggregating over $18,000,000,000, and it has paid to the policyholders nearly $3,000,000,000 in death and disability claims, matured endowments and surrender values, and over $1,000,000,000 in cash dividends. During the past eight years (1927 to 1934, inclusive), the company paid to its policyholders (1) over $1,100,000,000 in death, disability and other policy claims, and (2) about $480,000,000 in dividends; and it apportioned to pay to its policyholders in 1935 about $46,000,000 in dividends. In 1935 the company had over 2,000,000 policyholders, insured to the extent of about $6,661,000,000, who receive dividends annually.
Under the mutual plan, in order to provide for unforeseen contingencies, the premium to be paid by a member is fixed by the company at an amount somewhat in excess of that which the company anticipates will be necessary to cover the cost of the insurance. The member pays that amount in advance, but later receives such excess payment, if any, as a dividend, and thus gets the insurance at cost.
The premium to be paid for each age and type of policy is first determined by a purely mathematical calculation based strictly on specified mortality or disability tables and on a specific assumed rate of interest (e. g., three per cent) expected to be earned on the company’s funds. As the purely mathematical calculation of the *124net premium does not cover expenses, taxes, losses in investments, univested funds, and possible excessive mortality or disability, the company adds to the net premium an amount called the “ Loading ” (i. e., an additional amount calculated to be sufficient to cover such expenses, taxes, losses, idle money, etc.), which “ Loading,” plus the mathematical net premium, constitute the actual premium paid by the policyholder.
To offset the unavoidable increase in death and disability rates which arise from the yearly advance in age of members insured, the company must accumulate out of premiums a fund required by law to be held by every life insurance company, and known as “ Reserve.” The amount of reserve on any policy for any given age, plan, and time in force is a matter of mathematical calculation and is exactly the same when calculated by any actuary, depending on the particular mortality table, disability table and interest rate assumed.
As a practical matter, no company can tell in advance exactly what its interest earnings, death and disability claims or expenses will be. Yet if its interest earnings unexpectedly decrease (as has been the case to an enormous extent in every company since 1931), or if substantial financial losses befall (as has also been the case in the years 1931-1934), or if epidemics sweep the country (as influenza did in 1918) causing abnormal death losses, or if disability claims greatly increase (as has been notably the case since the depression), the company cannot increase the annual premiums to meet such losses. Therefore, in order to attain the highest degree of security, so that, in the distant years to come, members or their beneficiaries will surely receive the insurance at maturity, death, or disability, the company, in fixing the premium, has assumed four things:
(1) That the death rate will be precisely that of the specified mortality table (whereas, by careful medical selection, the death rate is usually less than that given in the table);
(2) That, in the case of disability policies, the payment of disability benefits will be precisely in accordance with the specified disability table (whereas, it was believed that the actual payments for disability benefits would probably be less than that indicated by such table);
(3) That the company will earn the specified rate of interest on its funds (whereas, in fact it will probably earn more than such assumed rate); and
(4) That the company’s expenses, taxes and extraordinary losses will be the exact amount of the loading (whereas, in fact the expenses, etc., are ordinarily less than the loading).
*125Consequently, there are four, and only four sources of surplus or profit to a mutual life insurance company:
1. Mortality savings. If the mortality is deferred beyond the period calculated on the basis of the mortality table used, instead of the company paying death losses at the time it provided for doing so, it keeps the money at interest longer and receives more premiums than expected. The resulting profit from such favorable mortality experience is called mortality savings.
2. Disability benefits. By careful selection of the risks insured, fewer persons may receive disability benefits, and for a shorter time, than was assumed in the disability table.
3. Interest. The company generally derives from interest a higher net rate than it assumed in mathematically calculating the premium; and to the extent that it earns more than the rate assumed, such excess interest is a profit.
4. Expenses. By economical management, the company may keep its aggregate expenses within the aggregate of all its loading. If so, then to that extent, the difference between the aggregate expenses and the aggregate loading is a profit.
There is a subsidiary element which was formerly a source of profit, to wit, lapsed and surrendered policies; but with annual dividend policies the element of lapses and surrenders is too small a factor to be calculated each year, and is calculated periodically.
Accordingly, from these four sources, mortality savings, disability benefits, interest, and expenses, the company accumulates funds in excess of legal reserves; and the excess of the funds on hand over such reserves (and other liabilities) constitutes the company’s surplus.
By the application of these so-called dividend factors, the company’s various, annual dividend rates are ascertained.
An “ economic adjustment ” factor has been also added as a fifth dividend factor in connection with recent numerous and unanticipated losses arising from defaults in interest on bond and mortgage investments, depreciation of assets, default in interest payments, and reduced interest rates.
As the policyholders in a mutual life insurance company have paid in more than was necessary, they are entitled to a return of such overpayments. The dividends of a mutual life insurance company are, accordingly, strictly speaking, not profits as in the case of an ordinary corporation, but really constitute a return to the policyholder of the amount he has been overcharged for his insurance. In fife insurance companies operating on the mutual plan the whole of the divisible surplus is distributed among the members annually as equally as may be in the proportions in which they have contributed to it.
*126Life insurance is based upon the principle of averages, and hence no company can consider an individual policy by itself and determine whether there has been an overcharge in its premiums, but every company must first determine the aggregate amount of all overcharges in premiums before it can decide how to arrive at what each individual policy shall receive back as dividend. The aggregate of all overcharges constitutes what is called the divisible surplus.
When the divisible surplus for any year has been ascertained, the company must determine how it shall be apportioned among all the policyholders, and for this purpose every mutual life insurance company divides its policies into a large number of homogeneous classes. In a “ class ” consisting of similar policies issued at the same time, under the same conditions, at the same age, with the same dividend distribution period, upon the same plan of insurance, and calling for the same annual premium per $1,000 face amount of the policy, there is not, even in the largest companies, a sufficient number of persons to give a true average rate of mortality. Accordingly, every life insurance company doing business upon the mutual plan applies to such group or “ class,” not the individual experience of such “ class ” by itself, but some ratio representing the average result of the company’s experience (1) as a whole; or (2) in some portion of its business sufficiently large to eliminate the effect of any accidental variations. The company thereby obtains for each of such “ classes ” substantially the same result that would have been obtained had the “ class ” contained as many persons in it as the company contains as a whole.
By a series of actuarial mathematical calculations covering its ninety years’ experience with over six million policyholders, the company determines from time to time its average experience with respect to mortality, disability, interest, expenses, lapses and surrenders, and this average experience is then used to determine the annual dividend rate, calculated in accordance with what is called the “ contribution ” method of ascertaining dividends; i. e., the distribution of divisible surplus to the different sources from which it is derived, so that as near as may be it is returned to the policyholders in the proportion in which they have contributed thereto. The “ contribution ” method was published more than seventy years ago by Sheppard Homans, actuary of the Mutual Life Insurance Company (The Assurance Magazine and Journal of the Institute of Actuaries, October, 1863, Vol. 11, p. 121 et seq.), and has since been followed with its principles and basic characteristics unchanged. Homans stated the fundamental principle as follows: “ Each participant should be benefited in proportion to the excess *127of his payments over and above the actual cost of insurance. * * * The first point * * * will be to determine what the actual cost of insurance, and consequently the overpayment, has been in any and every case.” He analyzed the mutual life insurance business; gave the intricate mathematical formulae of his proposed method for determining the contribution or overpayment of any policy, and then summed up the results as follows: “ The difference between the sum of his credits and the sum of his debits determines the overpayment or contribution from the policy proper.”
Daniel H. Wells, a distinguished American actuary, stated, in 1892, the basis of the contribution method as follows: “ It is of the very essence of the contribution method that no member or class of members shall be made to pay for the insurance furnished to any other member or class of members; that the cost of insurance shall not be increased to any individual or class because of the insurance of any other individual -or class.” (Papers and Transactions Actuarial Society of America, Vol. II, p. 361.)
The contribution method has been used since by practically every American and Canadian company, and is the recognized standard for dividend distribution throughout the world.
The company’s application of the contribution method is based upon keeping a debit and credit account with each “ class ” of homogeneous policies as an independent unit. It gives to the “ class ”— not the actual experience of the class itself — but an average experience which would have obtained if the class had been sufficiently numerous to produce average results.
In the creation of any mutual life insurance company’s surplus the factors employed by every mutual life insurance company in the ascertainment and apportionment of its divisible surplus are “ positive ” or “ negative ” or “ zero,” as follows: If the mortality is less than the mortality table employed, or the interest is greater than the rate assumed, or the expenses are less than assumed in the calculation of the premium, or the disability claims are less than in the disability table employed, then the respective elements of mortality interest, expense or disability are positive, otherwise, such respective elements are negative. In any case where the mortality, interest, expense or disability experience equals the tabular rate the element is zero, mdicating by such descriptive mathematical term, no contribution, either positive or negative, to the divisible surplus from the particular source to which the zero factor refers, that is, no divisible profit or loss is indicated from such element in the determination of the dividend.
*128As the controversy between plaintiff and defendant arises from defendant’s use of the disability factor in calculating plaintiff’s dividend, the use of this factor must be referred to in some detail. In 1910 the company began to issue, for the first time, life insurance policies with both death and disability benefits. From time to time, between 1910 and the present time, the company issued fife insurance policies (1) with death benefits only, and (2) with death benefits and varying forms of disability benefits, providing in the earlier years of issue only for a waiver of the premium during the continuance of disability; and in later years also providing for payment to the insured of a certain annual or monthly income during disability.
From 1912 to 1917, inclusive, in the calculation of dividends, the disability factor was positive and the company paid larger dividends on its disability policies than it paid on its non-disability-policies; in 1918 and 1919 the disability factor was (1) positive for certain forms of disability policies and (2) zero for other forms, depending on the year of issue and the type of benefit conferred; from 1920 to 1930, inclusive, the disability factor was zero and the company paid the same dividends on both lcinds of policies; from 1931 to 1934, inclusive, the disability factor was (1) zero for policies with certain forms of disability benefits, and (2) negative for other forms, depending upon the year of issue and the type of benefit conferred.
In 1912-1917 — during which period there were apparently gains to the company’s surplus from the disability policies — the company applied positive disability factors to all its various homogeneous classes of disability policies. Other policies taken out in the same years but without disability benefit provisions, had contributed nothing to surplus from disability provisions, and were not given any participation in the surplus created from such disability provisions.
During the years 1920-1930, when the company used a zero disability factor with respect to all its homogeneous “ classes ” of disability policies, the fluctuations were such that the company did not definitely settle whether certain adverse experiences were due to accidental variations in disability claims, or whether they represented a permanent trend of a greater number of persons being disabled. During the latter portion of such 1920-1930 period, not only had there ceased to be an apparent contribution to surplus from disability provisions, but the officers and directors of the company gradually realized that there was a serious possibility of a very large and utterly unanticipated drain upon the company’s contingency reserve, as a result of such disability provisions in its life insurance policies.
*129Beginning in 1931, the directors and officers of the company became convinced that, although the premiums received by the company over a series of years for the disability benefits in its life insurance policies exceeded the actual cash disbursements on account of disabled policyholders, such provisions were not contributing to the surplus; and in the exercise of their reasonable and best judgment they decided that, while the experience of the prior eleven years (1920-1930) was not conclusive, they should no longer continue to use a zero factor; and, therefore, they decided that the company should use, and thereafter it did use, a negative disability factor in the ascertainment and apportionment of divisible surplus among those disability policies. No notice, however, of the change made in 1931 in the company’s method of distributing divisible surplus was given to the policyholders.
In 1932 the company ceased to issue any policies with the former one per cent monthly income disability benefits, and greatly reduced the rate of income payments. Subsequently, it ceased to issue, and does not now issue, any policies with disability benefits providing for the payment of any kind of income.
The Insurance Department of the State of New York has approved the company’s practices in the distribution of dividends to its policyholders as equitable and as in conformity with the laws of the State of New York and the rulings of such department. In 1930, before the company adopted a negative disability factor in the apportionment of its divisible surplus to disability policies, it took the matter up with the Insurance Department of the State of New York and that department, after thorough consideration, approved the application of a proposed negative disability factor and the rate thereof by which the disability policies would have applied to them such negative factor of disability experience, while similar policies without disability benefits would not have any disability factor applied to them.
If, from January 1, 1931, to January 1, 1936, the company’s surplus had been apportioned so that non-disability policies received the same dividends per $1,000 face amount of the policy as similar policies containing disability benefits, the disability policies would have received in dividends about $15,000,000 more and the non-disability policies would have received about $15,000,000 less than they actually received.
On these and other facts detailed and amplified and expressly stipulated to by the parties in the agreed statement of facts, the plaintiff claims (1) that the disability policies are divided into two separate and distinct contracts — (a) one for death benefits, and *130(b) the other for total and permanent disability benefits — for which separate and independent premiums were paid; and (2) that the company has discriminated in favor of non-disability policies and against the disability policies, by paying on the former a larger dividend (per $1,000 face amount of insurance) than it paid on the latter, although the policies were exactly similar, except for the disability benefits, and that such difference was solely on account of the losses which the company sustained because of such disability benefits.
On the same facts the company claims (1) that all its disability policies are life insurance policies, i. e., not two contracts but one, with both (a) death benefits, and (b) total and permanent disability benefits; that such disability policies belong, for the ascertainment and apportionment of dividends, to a “ class ” different from the “ class ” to which non-disability policies belong; and (2) that the company, in the light of its prior experience, had the right, since 1931, to pay on non-disability policies larger dividends than it paid on otherwise similar policies containing disability benefits.
The question for the court to decide is whether the company had such right, or whether it violated any contractual or statutory duty in so doing and in applying a negative disability factor in connection with the apportionment of dividends to its disability policies, and whether such negative factor can be so employed in any proper application of the contribution method.
The statutes relied on by the respective parties are the following sections of the Insurance Law:
“ § 83. Distribution of surplus to policyholders. * * * every domestic life insurance corporation * * * shall provide in every policy * * * that the proportion of the surplus accruing upon said policy shall be ascertained and distributed annually and not otherwise. Upon the thirty-first day of December of each year * * * every such corporation shall well and truly ascertain the surplus earned by such corporation during said year. After setting aside from such surplus * * * a contingency reserve not in excess of the amount prescribed in this article, every such corporation shall apportion the remaining surplus equitably to all other policies entitled to share therein.”
“ § 89. Discriminations prohibited. No life insurance corporation * . * * shall make or permit any discrimination between individuals of the same' class or of equal expectation of life, in the amount of payment or return of premiums or rates charged for policies of insurance * * * or in the dividends or other benefits payable thereon, or in any of the terms and conditions of the policy.”
*131“ § 108. Discriminations under accident or health policies prohibited. No insurance corporation * * * shall make or permit any discrimination between individuals of the same class in the amount of premiums, policy fees, or rates charged for any policy of accident or health insurance, or in the benefits payable thereunder, or in any of the terms or conditions of such insurance contract, or in any other manner whatsoever.”
Counsel have been unable to find any decision in this State passing upon the precise issues submitted, or to cite any decision of a court of record whose opinions are published in any other State passing upon such precise issues. Certain rules, however, concerning policyholders’ rights with regard to divisible surplus and the apportionment of dividends have been judicially determined.
In Uhlman v. New York Life Ins. Co. (109 N. Y. 421 [1888]) it was held that the relation between a policyholder and a mutual insurance company is not that of trustee and cestui que trust, but is the relationship of debtor and creditor; that in the absence of wrongdoing or mistake the amount of divisible surplus to be apportioned as determined by the company is final and conclusive on all policyholders; that the apportionment of surplus earnings of an insurance company must be equitably made; that, prima facie, the apportionment made by the company should be regarded as equitable, since under the terms of the policy the duty of making it is cast upon the company, and it ought to be presumed that the company has performed its duty instead of presuming that it has failed to do so; but that on proper allegations of fact showing the apportionment made by the company is not equitable or has been based upon erroneous principles, the policyholder and all others similarly situated have a right to make proof of such allegations and, if proved, the court will declare a proper principle upon which the apportionment is to be made so as to become an equitable apportionment.
In Greeff v. Equitable Life Assurance Society (160 N. Y. 19 [1899]) it was held that a policyholder has a right to share in an equitable distribution of the company’s accumulated surplus, but, until a distribution is made by the officers or managers of the company, a policyholder has no such title to any part of the surplus as would enable him to maintain an action at law for its recovery. The court said: “ We think the principles which control the disposition of the surplus earnings of a stock corporation are applicable here. In those cases it has often been held that until dividends have been declared a stockholder has no right of action at law to recover any part of the fund applicable to that purpose, and that when directors have exercised their discretion in regard thereto *132the courts will not interfere unless there is bad faith, or wilful neglect, or. abuse of such discretion.” (Citing authorities.)
Equitable Life Assurance Society v. Brown (213 U. S. 25 [1909]) reiterated the rulings of the Uhlman and Greeff cases that the insurance company is not a trustee of its policyholders, and that discretion rests with the officers of the company as to what amount of surplus shall be retained and distributed and when the distribution shall be made. The court said: “ The complainant’s contention, as above stated, that there is such a trust in the fund mentioned has never been regarded as the law in the State of New York [Cohen v. New York Mutual Life Insurance Company, 50 N. Y. 610; People v. Security Life, etc., Co., 78 N. Y. 114; Bewley v. Equitable Life, etc., 61 How. Pr. 344; Uhlman v. New York Life Ins. Co., 109 N. Y. 421; and, to the same effect, Greeff v. Equitable Life Assurance Society, 160 N. Y., 19], nor anywhere else so far as any case has been cited on the subject.” And in overruling another contention of the complainant in the case the United States Supreme Court summarized the Uhlman and Greeff cases as follows: “ We think that neither the Uhlman nor the Greeff case decides any such principle as is asserted by the complainant. After holding, as already stated, that there was no trust existing between a policyholder and even a purely mutual company, reference was made in the former case to the contention of the defendant that the apportionment made by it, or under its direction, was absolutely and, at all events, conclusive upon the policyholders, it was said in the opinion that that was not an accurate statement, and that the plaintiff and others similarly situated had the right, upon proper allegations, of showing that the apportionment made by the defendant was not equitable, or had been based upon erroneous principles, and he had the right to a trial and to make proof of such allegations, and if true the court could declare the proper principles upon which the apportionment was to be made so as to become an equitable apportionment. The Greeff case simply adopted that statement in the course of the opinion,' which is chiefly devoted to the discussion of other matters.
“ There is nothing in either case to show that any other wrongdoing or fraud was in contemplation of the court than that above mentioned, viz., that the proposed or actual distribution of the money as between the policyholders themselves was not equitable, or was based on erroneous principles.”
In Miller v. New York Life Ins. Co. (179 Ky. 246; 200 S. W. 482 [1918]) the plaintiff contended that because the defendant, a mutual life insurance company, had not classified its policies with others issued in the same year, at the same age, and upon the same plan, *133and had not treated such “ classes ” or groups as independent units, the defendant had made no classification whatever. The court held that an equitable and fair classification includes in one class all policies issued upon the same plan so that participation in profits may be uniform to all the members of that single branch of the mutual enterprise, and that a further classification among such holders of uniform policies depending only upon the accidents of age and date of issue ought not to be made and must not be made when such subdivision is into such small units as will necessarily result in inequitable inequalities among members holding the same kind of policies.
On the facts stipulated and on the statutory and contractual obligations before us and under the rules of law applicable, we now determine the issues presented for adjudication.
Running through the whole of plaintiff's argument and as the basis on which, that argument ultimately reposes is the contention that the policy in question is a severable, divisible or separate contract; that there are in fact in such policy two complete agreements of insurance, one a complete agreement of life insurance for a specified premium ($27.34), and the other a complete agreement of disability insurance for another specified premium ($2.96); that plaintiff and all other holders of similar policies have the same rights under each of the several “ agreements ” as they would have if each such “ agreement " was contained in a separate policy; that a reduction in dividends paid on the “ agreement" of life insurance, merely because the policy also contains an “ agreement " of disability insurance, is to the extent of such reduction an illegal exaction of an additional premium for the disability insurance; and that the portion of divisible surplus earned on “ agreements " of life insurance must be divided as dividends under such “ agreements," while the portion of divisible surplus earned on “ agreements " of disability insurance must be divided as dividends under such “ agreements." Plaintiff compares her policy with a similar non-disability policy, and maintains that the life insurance agreement is identical in her policy with similar fife insurance agreements issued at the same time in policies without agreements of disability insurance, and that the annual premium in both instances is $27.34 for “ the agreement " of life insurance.
After consideration of the record and briefs, we conclude that this basic contention of the plaintiff, that the policy is severable and divisible in the sense that it is in legal form two separate agreements, cannot be sustained. We reach the conclusion that the policy is one agreement, a single policy, with both death and disability benefits so interwoven as to constitute a single integral *134insurance contract. While the amount charged for disability is specifically set forth (and must be so specified pursuant to the regulations of the New York State Insurance Department), the policy itself provides for one premium, $30.30, and the statement on page 2 that the “ total premium stated on the first page hereof includes an annual premium of $2.96 for disability benefits,” is to meet the provision that plaintiff, at any time while the policy is in effect, may drop the disability benefits, in which event the premium is automatically reduced to $27.34. In that sense the disability feature of the policy is, indeed severable, but it is sever-able for the purpose of being terminated, and until the privilege of dropping the disability benefit is exercised (or has been rescinded for any proper reason), the policy with the various promises therein contained is a single policy and may for dividend purposes be dealt with as a whole.
In its physical aspect, the policy is one instrument. A consideration of the provisions for disability benefits makes it clear that if all of these provisions were taken together apart from the rest of the policy, it would be impossible to spell out a separate contract for disability insurance; there would be no identifiable company, no identifiable sum to be paid (for the disability benefits cannot be obtained without the death benefit), and no identifiable person to receive the payment. The provisions for disability benefits are so intimately dependent upon and unintelligible without the death benefit provisions as to be by themselves impossible of constituting a separate contract. There was one application for the policy, not two, and provision is made for one premium. Throughout the policy the single number is invariably used, never the plural; it is referred to as “ this policy,” “ the policy,” “ this contract,” “the contract,” “the entire contract; ” the policy thus describes itself about seventy-four times in the singular. The following provision is also significant: “ The Contract.—■ The policy and the application therefor, copy of which is attached hereto, constitute the entire contract. * * * ISTo agent is authorized to make or modify this contract.”
A single set of promises arising in connection with life insurance and with disability insurance are closely interwoven and dependent one upon the other. The disability benefits are issued only when the terms thereof are printed in a life insurance policy, they cannot be obtained separately, and cannot be continued in force independently of the death benefit in the policy. When the annual premium became due, if plaintiff tendered the $2.96 specified as the amount for the disability benefits, she could not compel *135the defendant to continue the disability benefits unless the balance of the premium was paid.
From all the above it follows that the policy is a single contract containing various provisions and various promises, severable and divisible in one sense as a remedial measure for enforcement, but the policy cannot be taken, as plaintiff contends, as the equivalent of two separate policies, one the exact equivalent of a policy of life insurance with merely a death benefit, and the second an independent contract or policy covering disability benefits.
Williston points out that the words “ divisible ” and “ severable,” as applied to contracts, are often misleading: “ Some confusion has arisen from the inexact use of the terms ‘ divisible contract ’ and ‘ severable contract ’ on the" one hand, and ‘ entire contract' on the other. * * * Sometimes the words entire or indivisible
are used as meaning that there is one contract as distinguished from several contracts, and at other times the words are used as meaning more than this, namely, that there is a contract which is not divisible. A divisible contract, using that term properly, is always one contract and not several contracts. It differs in one respect only from other contracts — namely, that on performance of one side of each of its successive divisions, the other party becomes indebted for the agreed price of the division.” (2 Williston Cont. [1920 ed.] § 861, pp. 1646,1648, 1649.) He adds (§ 863, pp. 1652, 1653): “ The essential test to determine whether a number of promises constitute one contract or more than one is simple. It can be nothing else than the answer to an inquiry whether the parties assented to all the promises as a single whole, so that there would have been no bargain whatever, if any promise or set of promises were struck out. * * * The question essentially is one of fact: Did the parties give a single assent to the whole transaction or did they assent separately to several things? ”
Applying that test to the policy under review, were the life insurance provision stricken out at the outset there would have been no contract whatever, as disability benefits are issued only when the terms thereof are printed in a life insurance policy; they cannot be separately obtained.
Hence we conclude that it cannot be said that the directors of the defendant company violated either a contractual or statutory duty when, for dividend purposes, they treated the policy as a whole and considered the contribution to surplus of the whole policy, not the contribution of some portion of the policy, as the basis for determining how much of the divisible surplus should be returned in the form of a dividend upon the policy.
*136The statutory test is whether the apportionment is equitable or whether it unfairly discriminates between individuals of the same class. (Insurance Law, §§ 83, 89, 108, supra.) No fraud, bad faith, willful neglect or mistake of fact is here claimed. The alleged wrong is an abuse of discretion or the application of an erroneous principle, i. e., the use since 1931 of the negative disability factor. As indicated by the authorities cited above, the apportionment, as made by the company, must be regarded prima facie as an equitable apportionment, and the plaintiff must allege and prove facts showing that the apportionment is not equitable or has been based upon erroneous principles, and in the absence of any allegation of wrongdoing or mistake, the directors’ determination of the question must be treated as proper. (Uhlman v. New York Life Ins. Co., Greeff-v. Equitable Life Assurance Society, Equitable Life Assurance Society v. Brown, supra.)
To succeed, plaintiff must show that the principle on which the apportionment is based is so clearly erroneous as to be beyond the exercise of any reasonable discretion on the part of the company’s directors. The author of the contribution method has stated that the difference between the sum of the policyholder’s “ credits ” and the sum of his “ debits ” determine the overpayment or contribution from the policy proper. (Homans, supra, p. 124.) The negative factor was, of course, not used in connection with the policy as a whole, for otherwise the policy would not be entitled to any distribution of surplus. Its contested use is only in connection with the disability element. The “ debits ” arising from that element are set off against the “ credits ” arising from other elements before determining the net overpayment or contribution from the policy as a whole. This is the company’s procedure with regard to the homogeneous “ classes ” in question, namely, the classes of disability policies. A class of policies-, viz., a disability class, that entails such debits on its disability benefits cannot be said to contribute as a class to surplus in the same degree as another class of policies (non-disability) that does not entail such debits.
It is true that the plaintiff is the holder of a disability policy on which no claims have been made. But the plaintiff and all persons holding the 1,600,000 disability policies issued and outstanding possess and have received disability protection. To those who have become disabled the disability benefits have accrued. To those who, like plaintiff, have fortunately never become disabled, the disability benefit is a matter of protection, as is all other insurance. That being so, it cannot be said that the policyholders who possess that valuable protection as an integral part of their insurance should receive precisely the same dividend treatment *137as holders of non-disability policies who have never received any such disability protection.
The record shows that from 1910 to 1919 disability benefits were a source of surplus in the defendant’s business and received all the surplus from that source. The non-disability policies were not permitted to share in the surplus derived from that source because they had in no way contributed to its creation. From 1912 to 1919 the non-disability policies received a smaller dividend than the disability policies. From 1920 to 1930 the defendant, acting under the belief that the disability experience was approximately equal to that which had been assumed in the calculation of the premiums, and that any adverse experience was a mere accidental temporary variation due to chance, to be taken care of by the company’s contingency fund, employed what is referred to as a zero factor as to disability policies. From 1931 to 1935, as the result of its experience, the defendant, instead of using a zero factor, employed a negative factor on its disability policies, causing the dividend on the disability classification to be less than on similar policies containing no disability provision. There is thereby no discrimination, and particular classes are uniformly treated. All the policyholders of disability benefits have been treated alike in the apportionment of dividends, and all policyholders without disability benefits have been treated alike in such apportionment.
On the facts stipulated and limiting our decision strictly to these facts, including the particular experience of the company in question, we cannot say that the use of the negative disability factor since 1931 is erroneous, inequitable, beyond the exercise of the directors’ reasonable discretion, or a breach of statutory or contractual obligation.
Judgment should accordingly be entered in favor of defendant, but without costs.
Martin, P. J., Townley, Glennon and Untermyer, JJ., concur.
Judgment unanimously directed in favor of the defendant, without costs. Settle order on notice.