Full Text: Treasury Report On Fraternal Benefit Societies.

The full text is available of the Treasury Department's "Report to The Congress on Fraternal Benefit Societies," dated January 15. The report examines the economic rationale for tax exemption and assesses whether, under that rationale, an exemption for the activities of fraternal benefit societies is justified.

====== FULL TEXT ======

DEPARTMENT OF THE TREASURY

January 15, 1993

The Honorable Dan Rostenkowski

Dear Mr. Chairman:

Section 1012(c)(2) of the Public Law 99-514 the Tax Reform Act of 1986 provides that the Treasury Department shall conduct a study of the operation of section 501(c)(8) of the Internal Revenue Code of 1986 organizations which received gross annual insurance premiums in excess of $25 million for the taxable years that ended during 1984. Pursuant to that directive, I hereby submit the "Report to Congress on Fraternal Benefit Societies."

I am sending a similar letter to Representative Bill Archer.

Sincerely,

Alan J. Wilensky

Enclosure

* * *

January 15, 1993

The Honorable Daniel Patrick Moynihan

Dear Mr. Chairman:

Section 1012(c)(2) of the Public Law 99-514 the Tax Reform Act of 1986 provides that the Treasury Department shall conduct a study of the operation of section 501(c)(8) of the Internal Revenue Code of 1986 organizations which received gross annual insurance premiums in excess of $25 million for the taxable years that ended during 1984. Pursuant to that directive, I hereby submit the "Report to Congress on Fraternal Benefit Societies."

I am sending a similar letter to Senator Bob Packwood.

Sincerely,

Alan J. Wilensky

Enclosure

TABLE OF CONTENTS

Chapter 1 Introduction and Summary

Background

Chapter 2 Description of Special Exemption of Fraternal Benefit

Tax Treatment of Fraternal Benefit Societies

Chapter 3 Rationale for Tax Exemption

Economic Rationale

Chapter 4 Overview of Fraternal Benefit Societies' Financial Data

Data

Chapter 5 Insurance Activities of Fraternal Benefit Societies

Introduction

Chapter 6 Fraternal and Charitable Activities of Fraternal Benefit

Overview of Fraternal and Charitable Activities

Chapter 7 Summary and Policy Options

Summary

Appendix 1 Congressional Mandate: Tax Reform Act of 1986

Appendix 2 1988 Treasury Department Survey and Survey Supplement of

Appendix 3 List of Large Mutual Life Insurers Used in Analyses

Appendix 4 Description of Payment and Surrender Cost Indices

Appendix 5 Statistical Information on Life Insurance Cost, Efficiency

LIST OF TABLES

1 Total Receipts and Expenses of the Seven Largest Fraternal

2 Average Receipts and Expenses of the Seven Largest Fraternal

3 Average Number of Paid Employees of the Seven Largest Fraternal

4 Insurance in Force of the Seven Largest Fraternal Benefit

5 First Year Premiums of the Seven Largest Fraternal Benefit

6 Average Ordinary Life Insurance Policy Size of Selected Fraternal

7 Means and Standard Deviations of Projected Surrender Cost Index

8 Means and Standard Deviations of Level Premiums, Payment Index,

9 Means and Standard Deviations of Measures of Efficient Operations

10 Means and Standard Deviations of Surplus Measures for Large

11 Number of Lodges and Number of Members of Fraternal Benefit

12 Measures of Local Fraternal Service for All Fraternal Benefit

13 Fraternal Expenditures by All Fraternal Benefit Societies

14 Test Statistics for Means and Standard Deviations of Projected

15 Test Statistics for Means and Standard Deviations of Level

16 Test of Significance of Type of Organization, Fraternal Benefit

17 Regression Parameter Estimates for Lapse Rate Equation of Large

18 Test of Significance of Type of Organization, Fraternal Benefit

19 Test of Significance of Type of Organization, Fraternal Benefit

20 Regression Parameter Estimates for Investment Expense Ratio

21 Test of Significance for Surplus Measures of Large Mutual

CHAPTER 1

INTRODUCTION AND SUMMARY

BACKGROUND

A Treasury Department Report to the President, Tax Reform for Fairness, Simplicity and Economic Growth (Treasury I), included a proposal to repeal the tax exemption for certain insurance companies and impose tax on the insurance income of fraternal benefit societies. /1/ This proposal was not included in the Tax Reform Act of 1986 (TRA) (Public law 99-514). However, section 1012(c) of TRA /2/ required the Treasury to study certain fraternal benefit societies that benefit from tax exemption under section 501(c)(8) of the Internal Revenue Code of 1986. /3/

The purpose of this study is to examine the operations and to assess the taxation of the insurance income of these fraternal benefit societies. Fraternal benefit societies provide insurance and other benefits for members, charitable goods and services for the community, and fraternal or club services for members. Similar to other tax-exempt organizations, fraternal benefit societies are taxed on income from business that is unrelated to the organization's exempt purpose (unrelated business income). However, insurance income is not treated as unrelated business income. This treatment of insurance income differs from the treatment of insurance income earned by other tax-exempt organizations and commercial insurers, both of which are taxed on their life, and property and casualty insurance business income. /4/

As discussed in the General Explanation of the Tax Reform Act, /5/ Congress was interested in the determination of whether fraternal benefit societies were engaged in large-scale insurance activities, the nature and scope of which were inherently commercial rather than charitable. Under such circumstances, tax exemption of their insurance income may be inappropriate. Therefore, Congress requested Treasury to obtain information regarding use of untaxed revenues from insurance activities of these organizations.

As a result of Congress' request, Treasury conducted a study which examines the economic rationale for tax exemption and assesses whether it justifies tax exemption for the activities of fraternal benefit societies. The study also assesses four potential uses of funds from the tax exemption on insurance income: (1) whether the tax exemption for insurance income is financing charitable and fraternal activities; (2) whether the fraternal benefit societies are rebating the tax exemption to their policyholders in the form of lower prices for life insurance; (3) whether the exemption is in essence subsidizing less efficient insurance operations than those of comparable commercial insurers; and (4) whether the tax exemption has been used by fraternal benefit societies to accumulate surplus in excess of that accumulated by comparable commercial insurers.

To analyze these issues, Treasury gathered data on seven fraternal benefit societies which met Congress' stated study parameters. /6/ A special survey was conducted to collect information on receipts, expenses, and operations for each of these fraternal benefit societies. This information was supplemented with publicly available data.

PRINCIPAL FINDINGS

o The insurance activities of fraternal benefit societies are

o The benefits to society from charitable services, the

o A major economic argument for exempting an organization from

o Analysis of the cost of comparable insurance policies

o Analysis of certain measures of operating efficiency indicate

o Comparison of the rate of surplus accumulation and level of

o Fraternal benefit societies provide many charitable services;

POLICY OPTIONS

The Congress may wish to consider the following options relating to the tax treatment of fraternal benefit societies:

NO CHANGE IN CURRENT TAX TREATMENT

Fraternal benefit societies perform valuable social, commercial, and charitable functions. The charitable services provided by fraternal benefit societies benefit society as a whole. Fewer of these charitable goods and services are likely to be provided unless current tax treatment continues. The economic distortions caused by the special treatment of fraternal benefit societies are relatively minor in comparison to other policy priorities. Thus, Congress could decide not to change the fraternal benefit societies' current tax treatment.

MODIFY TAX TREATMENT OF FRATERNAL BENEFIT SOCIETIES

If Congress decides to modify the tax treatment of the insurance activities of fraternal benefit societies, it may be appropriate for a fraternal benefit society that continues to have insurance activities which are a substantial part of its business, to be taxed as a mutual life insurance company. If a fraternal benefit society has only minor insurance activities, then the fraternal benefit society could be allowed a choice of being subject to section 501(m) and paying UBIT on insurance income or converting to a section 501(c)(10) domestic fraternal society and being prohibited from selling insurance. Because of the administrative burdens of being subject to taxation, modification of tax treatment could be limited to only large fraternal benefit societies that do not subsidize insurance for low-income members. In recognition that the fraternal benefit societies incur large charitable expenses, consideration could be given to increasing (or raising) the limitation on deductible charitable contributions for these entities, which is presently 10 percent of taxable income, or permitting a deduction for a portion of their combined charitable and fraternal expenses. Permitting the latter allows the fraternal benefit society to provide such services while limiting their compliance costs and the Internal Revenue Service's administrative costs.

ORGANIZATION OF THE REPORT

The study describes the special exemption for fraternal benefit societies and the rationale for such treatment. This perspective is followed by an overview of the current operations of the surveyed fraternal benefit societies. The study then presents data and analysis of the insurance activities of fraternal benefit societies and their commercial counterparts and a description and analysis of the funding of fraternal and charitable activities. Finally, the study presents a summary and policy options.

FOOTNOTES TO CHAPTER 1

/1/ U.S. Department of the Treasury, Tax Reform for Fairness Simplicity and Economic Growth, Report to the President [Treasury I]U.s. Govt. Print. off. (November 1984), at pp. 286-287.

/2/ The due date for this study was extended from January 1, 1988, to July 1, 1992 by section 11831 of the Omnibus Budget Reconciliation Act of 1990 (Public Law 101-508). See Appendix 1 for a copy of the Congressional mandate for this report and the extension of the due date.

/3/ Unless otherwise indicated, all "section" references refer to the Internal Revenue Code of 1986.

/4/ In 1985, the commercial life insurance industry paid $2.9 billion of income tax after credits on $7.4 billion of taxable income while mutual life insurance companies paid $1.3 billion of tax on $3.4 billion of income according to the U.S. Department of the Treasury, Interim Report to Confess on Life Insurance Company Taxation, (June 1988), p. 14.

/5/ Staff of the Joint Committee on Taxation, General Explanation of the Tax Reform Act of 1986, JCS-10-87, U.S. Govt. Print. off. (May 4, 1987), at pp. 584-586.

/6/ Fraternal benefit societies that received gross annual insurance premiums in excess of $25 mIllion in 1984 were to be studied.

END OF FOOTNOTES

CHAPTER 2

DESCRIPTION OF SPECIAL EXEMPTION OF

TAX TREATMENT OF FRATERNAL BENEFIT SOCIETIES

The Revenue Act of 1913 provided tax exemption for nonprofit organizations and included in that definition fraternal benefit societies that operate under a lodge system. /1/ The current tax treatment of fraternal benefit societies, as described in section 501(c)(8), has remained virtually unchanged since 1913 with only minor statutory modifications.

Tax-exempt organizations are exempt from corporate income taxation on income earned in related business activities. Income from related activities for organizations which are described in section 501(c)(8) includes income from the provision of insurance and other benefits as well as income from their fraternal and charitable activities. Similar to other section 501 organizations, /2/ fraternal benefit societies are subject to the unrelated business income tax (UBIT) on income that is unrelated to their exempt purpose. In general, income derived from non-members, for example from the rental of a lodge to a non-member, is taxable as unrelated business income. However, fraternal benefit societies are unique in that their insurance income is exempt from tax and can be used to finance their charitable and fraternal services. Income producing activities that are unrelated to the exempt purpose of a tax-exempt organization are generally subject to UBIT. Thus, without this specific exemption in the tax law, insurance operations of fraternal benefit societies would be taxable. Section 502 also operates to deny tax exemption for feeder organizations which perform commercial activities even if all commercial profits are spent on charitable activities.

In addition to exemption from the federal corporate income tax, fraternal benefit societies are exempt from certain federal excise and employment taxes. Based on the federal tax exemption, many states and local governments also exempt fraternal benefit societies from income, sales, use, and property taxes. /3/ However, payments to fraternal benefit societies, such as premiums and fees paid by members, are not deductible as charitable contributions by the payors.

DEFINITION OF A FRATERNAL BENEFIT SOCIETY

A fraternal benefit society must meet three requirements in order to qualify as a section 501(c)(8) organization. First, the society must have a fraternal and beneficial character which distinguishes it from mutual insurance companies. Section 501(c)(8) describes the other requirements that an organization must meet in order to be exempt from corporate taxation. Specifically, fraternal benefit societies, orders, or associations must:

(A) operat[e] under the lodge system or for the exclusive

(B) provid[e] for the payment of life, sick[ness], accident, or

This last characteristic, providing insurance and other benefits, distinguishes section 501(c)(8) organizations from domestic fraternal societies which are tax-exempt under section 501(c)(10). Section 50l(c)(8) organizations must provide insurance and other benefits to members and their dependents, as compared to domestic fraternal societies which are prohibited from providing insurance. Domestic fraternal societies must devote net earnings ". . . exclusively to religious, charitable, scientific, literary, educational, and fraternal purposes." /4/ Fraternal benefit societies may also devote net earnings to these activities, but they are not required to do so to maintain their tax exemption. The fraternal and beneficial character, as well as a lodge structure, however, appear to be common to both types of organizations.

While there is little legislative history defining a fraternal benefit society, the courts have provided their own definition, stating that the term is used

* * * to designate an association or society that is engaged in

Usually the fraternal bond is based on a combination of shared religious, ethnic, occupational, or moral characteristics. This required fraternal characteristic allows the organization to prescribe conditions for membership and to establish rules of conduct. Persons who fail to meet membership requirements or to follow the rules can be excluded. This fraternal bond is reinforced by members doing charitable, educational, patriotic, and sometimes religious services for others, including non-members.

The second aspect of the fraternal characteristic is the lodge structure. Organizations that do not have a lodge structure have been denied exemption under section 501(c)(8) even though the groups may have fulfilled the fraternal and insurance requirements. /6/ Treasury regulations /7/ describe operating a lodge system as:

* * * carrying on its activities under a form of organization

The lodge structure is generally a central governing body with local branches called lodges. The lodges operate under charters from the central organization and follow prescribed rules as to ritual, qualifications of members, and use of funds. Societies without central organizations are not considered to have lodge structures. /8/ The lodges are largely self-governing and elect or appoint representatives to the central organization. The lodge must hold regular meetings at least once a year for its members. /9/ The central and local organizations must be active and not just governing structures. /10/ An organization that does not operate under the lodge system but operates exclusively for the benefit of members of an association that has a lodge system may still be eligible for section 501(c)(8) treatment. /11/

The third requirement for section 501(c)(8) status is that a fraternal benefit society must provide insurance or other benefits to its members and their dependents. The insurance benefits include life, sickness, and accident insurance. Other benefits have been broadly interpreted to include those services that help spread risks among members. For example, annuities that protect against the loss of earning power are an example of "other benefits". While not every member of the society must be covered by the insurance benefits, all must have the option of coverage and a substantial number must be covered. /12/ While life insurance is within the definition of benefits that can be offered by fraternal benefit societies, it is unclear whether property and casualty insurance is within the definition. /13/

SUMMARY

To be eligible for tax exemption as a fraternal benefit society, an organization must have a fraternal bond, operate within a lodge structure, and provide insurance or other benefits to its members. Fraternal benefit societies are similar to domestic fraternal societies in their lodge structure and fraternal nature. Fraternal benefit societies, however, unlike domestic fraternal societies, are not required to devote net earnings to charitable or fraternal purposes and are required to provide insurance and other benefits to members. Domestic fraternal societies are prohibited from providing insurance.

FOOTNOTES TO CHAPTER 2

/1/ Bruce R. Hopkins, The Law of Tax-Exempt Organizations. Fifth Edition, New York: John Wiley and Sons (1987), p. 4.

/2/ Exempt organizations are described in section 501.

/3/ Hopkins, p. 32.

/4/ I.R.C. section 501(c)(10).

/5/ National Union v. Marlow, 74 F. 775, 778 (8th Cir. 1896) as cited in Hopkins, p. 371.

/6/ Revenue Ruling 55-495; 1955-2 C.B. 259; Revenue Ruling 63- 190, 1963-2 C.B. 212.

/7/ Treasury Regulation section 1.501(c)(8)-1.

/8/ Revenue Ruling 55-495, 1955-2 C.B. 259.

/9/ GCM 34607, 1971.

/10/ I.T. 1516, 1-2 C.B. 180, 1922.

/11/ Rev. Rul. 73-192, 1973-1, C.B. 224.

/12/ Rev. Rul. 64-194, 1964-2 C.B. 149, and Polish Army Veterans Post 147 v. Commissioner, 24 T.C. 891 (1955), aff'd 236 F.2d 509 (3d Cir. 1956) cited in Hopkins, p. 372.

/13/ The issue is whether "accident or other benefits" applies to accidents to persons and property or is limited to persons.

END OF FOOTNOTES

CHAPTER 3

RATIONALE FOR TAX EXEMPTION

ECONOMIC RATIONALE

Tax-exempt organizations are exempt from the corporate income tax on related business activities. /1/ Related activities are generally those activities performed by the exempt organization which are necessary to produce the goods and services that are related to their exempt status. For example, an exempt school may earn net income through the provision of educational services and that activity would not be subject to tax. Net income earned from the production of goods and services which are unrelated to an organization's exempt status is essentially subject to the corporate income tax, since unrelated net income is subject to the unrelated business income tax (UBIT). The following discussion focuses on the rationale for exempting income from the corporate income tax.

In general, organizations are exempt from income tax on certain activities in the United States either due to the goods and services they provide or their form of organization. There are the traditional tax-exempt section 501(c)(3) organizations that operate for "religious, charitable, scientific, testing for public safety, literary, or educational purposes." /2/ The exemption from tax is based on the type of goods or services that these organizations produce. There are also club-type or mutual organizations which operate more as conduits, combining members' resources to purchase or produce goods and services. In general, if revenues are received from members and returned to members, then these organizations are not taxed on their retained earnings.

A major economic rationale for exempting an organization from tax on the income from goods or services that the organization produces, is that without this exemption, the market system may experience a deficiency in the amount or quality of such goods and services. This result is referred to as a "market failure". A market failure occurs if the market price does not reflect the benefits conferred on the community as a whole. /3/ For example, scientific research often benefits an entire community. Too little research may be provided if the private compensation received for producing research does not reflect the social benefits produced. Subsidies can increase the private financial returns to research so that more research would be produced. While subsidies can be provided for each unit produced, income tax exemption of the producing organization is an alternative way of providing a subsidy.

Under this same rationale, tax exemption may be justified because the organization subsidizes goods or services for low income individuals or households. /4/ The external benefit is generated by the character of the benefitted group, and not necessarily by the character of the good and service. Donors receive an external benefit because they are concerned with the welfare of lower-income individuals. A subsidy can be implemented by the tax-exempt organization that directly provides goods and services to lower income people at less than cost. For example, a soup kitchen provides food to low income people. Alternatively, the organization may subsidize purchases of goods and services by lower income consumers with income from sales at higher prices for the same goods and services. For example, a day care center may charge lower income clients less for care of a child than higher income clients.

Alternatively, because of the complex nature of the goods or services provided, consumers may not be fully able to evaluate whether they are receiving the appropriate quantity or quality. /5/ Thus, the consumer may rely on the identity of the producer to ensure that the consumer's interests are appropriately considered. /6/ For example, hospitals that maximize profits to shareholders may have an incentive to provide lower quality care to patients who are unable to discern whether they have received the best or even appropriate treatment. A tax-exempt nonprofit hospital arguably may not have as strong an incentive to lower the quality of care and so may be more likely to provide the level of care which may be needed. Some consumers may believe tax-exempt nonprofit organizations are more likely to act in their interest than a comparable taxable organization.

The above reasons for providing tax exemption are often used to determine whether government should fund or produce certain goods and services. One economic benefit of tax-exempt organizations is that they may offer more diverse choices than could be provided by government alone. This variety of choices benefits consumers who vary in the quantity and quality of goods and services they demand. /7/ Often production by the tax-exempt sector supplements government provision. /8/

If the subsidy is provided as an income tax exemption, then the rate of subsidy varies not by the good or service produced, but by the income taxes foregone. As a result, the subsidy varies by organization, while in theory the subsidy rate should vary in proportion to the external benefit generated by each unit of good or service provided by the organization. It may, however, be less costly to administer an organization exemption than to subsidize particular goods and services.

Tax exemption may also be allowed for organizations that act as conduits and pool members' resources. The rationale for providing tax exemption in these situations is that imposition of tax is not appropriate when the organization acts only as an intermediary in organizing members' activities, which could have been accomplished by the members directly.

APPLICATION OF ECONOMIC RATIONALE TO FRATERNAL BENEFIT SOCIETIES

Fraternal benefit societies provide "insurance and other benefits" for members, charitable goods and services for the community, and fraternal or club services for members. For most fraternal benefit societies, life insurance forms the vast majority of insurance and other benefits that are provided to members. As discussed in Chapter 5, insurance products offered by fraternal benefit societies are essentially the same as those provided by commercial insurers. The study examines the external benefits that these services provide, and whether these services are of such a complex nature that adequate consumer information may be lacking. Finally, the study examines the conduit nature of fraternal benefit societies.

In the economics literature, the provision of life insurance and other benefits is generally not considered a good or service with significant external benefits to society as a whole. However, the life insurance industry has argued that subsidies for life insurance encourage purchases of such insurance for protection of dependents after the death of a wage earner. In turn, these purchases result in less demand for government assistance by families. /9/ This is also the rationale offered for subsidizing the savings element in cash value life insurance. /10/ The subsidy, granted to the savings element of cash value life insurance, is currently available to all policyholders, not just members of fraternal benefit societies. Thus, life insurance currently receives preferential tax treatment.

Insurance is not a type of product for which consumers may lack access to information on the appropriate quantity or quality that they need. There is no evidence that commercial insurance is inferior in quantity and quality to insurance sold by fraternal benefit societies. Both commercial and fraternal benefit societies sell similar products and are required by state regulation to provide consumers with similar information on insurance products. /11/ Therefore, the rationale for tax exemption based on consumers being unable to make informed choices on products may not be applicable in the provision of life insurance.

One argument for the tax exemption of insurance activities of fraternal benefit societies would be to permit or increase cross- subsidization by charging higher income policyholders more for the product than lower income policyholders. Similarly, some fraternal activities redistribute income among members. For example, a fraternal benefit society may provide scholarships to dependents of members or subsidize a retirement home for members. Information presented later in this study indicates that there is relatively little cross-subsidization of insurance or redistribution of fraternal services. Consequently, the rationale for tax exemption based on cross-subsidization does not generally apply to the fraternal benefit societies.

In general, charitable activities of fraternal benefit societies provide significant external benefits or are redistributive. Therefore, tax exemption for these goods and services may be justified in order to encourage the provision of these activities by fraternal benefit societies. The distinction between charitable and fraternal activities is that charitable activities generally benefit both members and non-members, while fraternal activities benefit only members.

Some conduit or club-type organizations are exempt on earnings from members. The net income of conduit organizations is essentially zero because any excess earnings are returned to members. With respect to fraternal benefit societies, examples of such activities would be socials or club picnics where the fraternal benefit society acts as a conduit. The conduit function may provide a rationale for tax exemption of certain fraternal services.

SUMMARY

Fraternal benefit societies provide three types of services: insurance, charitable, and fraternal. The insurance activities of the fraternal benefit societies are similar to activities of commercial insurers and, therefore, do not appear to have significant benefits that have been used in other contexts to justify tax exemption. The significant external benefits from charitable services, the redistributive nature of some fraternal services, and the use of the conduit organization form for providing fraternal services may, however, justify continuation of tax exemption for these activities of fraternal benefit societies.

FOOTNOTES TO CHAPTER 3

/1/ This discussion excludes an analysis of the effect of the deductibility of contributions to certain types of tax-exempt organizations.

/2/ I.R.C. section 501(c)(3). When discussing charitable activities, the study includes religious, scientific, testing for public safety, literary, and educational activities.

/3/ In the economic literature, such benefits are labeled "external benefits" or "externalities." Public goods are a specific case of a good with external benefits. In the contrary case, such as pollution, the market price may not reflect the harm conferred on the community as a whole as a result of an activity. These effects are called "negative externalities."

/4/ A market failure also occurs when the market does not properly distribute the benefits.

/5/ In the economics literature, this circumstance is referred to as a case of "asymmetric information."

/6/ In the economics literature, this situation is referred to as the principal-agent relationship, where the agent, for example a doctor, acts in the interest of the principal, the patient.

/7/ In the economics literature, this is described as consumers having heterogeneous demand.

/8/ Burton A. Weisbrod, The Nonprofit Economy. Cambridge: Harvard University Press (1988), p. 26.

/9/ While other reasons for the tax preferred treatment of life insurance have been offered by the commercial life insurance industry, this is the primary argument. See U.S. Department of the Treasury, Report to the Congress on the Taxation of Life Insurance Company Products (March 1990) for a more complete discussion.

/10/ Currently, cash value life insurance is subsidized to the extent the savings component, or "inside build-up," is not taxed as it accrues.

/11/ As discussed in more detail later, fraternal benefit societies tend to sell smaller insurance policies than commercial insurers. However, this may reflect differences in the income of policyholders.

END OF FOOTNOTES

CHAPTER 4

OVERVIEW OF FRATERNAL BENEFIT SOCIETIES' FINANCIAL DATA

DATA

Congress required Treasury to study those fraternal benefit societies, as described in section 501(c)(8), that received gross annual insurance premiums in excess of $25 million for taxable years ending in 1984. Seven fraternal benefit societies were identified as meeting this requirement: Aid Association for Lutherans, Independent Order of Foresters, Knights of Columbus, Lutheran Brotherhood, Mennonite Mutual Aid Association, Modern Woodmen of America, and Woodmen of the World. The study did not examine the approximately 200 other fraternal benefit societies which, according to the National Fraternal Congress of America, were in existence in 1984.

Information about insurance as well as fraternal and benevolent activities was obtained by surveying the seven societies. /1/ Both revenue and expenditure data were collected as well as information on these organizations' general and insurance operations. Materials were also provided by the seven fraternal benefit societies on their organizational structure, the history of their tax exemption, and the types of fraternal and benevolent services they provide. Publicly available data on these seven societies, as well as other fraternal benefit societies, were collected. More detailed data on the life insurance products and operations of the seven fraternal benefit societies and comparable commercial insurers were used for analyzing life insurance costs and operating efficiency, as well as surplus accumulation.

OVERVIEW OF FRATERNAL BENEFIT SOCIETIES' RECEIPTS AND EXPENSES

As shown in Table 1, the seven organizations studied had total receipts (including investment income) of approximately $3.4 billion in 1985. Gross receipts have approximately tripled every ten years since 1955, with an average annual growth rate of 10 percent. /2/ Prior to 1955 the annual growth rate was a relatively modest 2 percent, beginning in 1930 when receipts were $71.3 million.

As a percentage of total receipts (including premium and investment income), insurance related receipts increased slightly from 95 percent in 1930 to 97 percent in 1985. While premium income still accounts for the majority of insurance receipts, net investment income has become a more important income source over the 1930 to 1985 period. /3/ In 1930, net investment income was $10.2 million, or about 15 percent of insurance related receipts. By 1985, net investment income had climbed to $1.2 billion, or 35 percent of insurance receipts. For the 1975 to 1985 period when comparable figures are available, investment income as a percentage of total income rose from 21 to 29 percent for commercial life insurance companies. /4/

As shown in Table 2, the average receipts for fraternal and charitable activities for the seven fraternal benefit societies have also grown, from $0.7 million in 1930 to $2.7 million in 1985. Thus, receipts for fraternal and charitable activities grew at a slower rate than total receipts. Fraternal and charitable receipts are presented as averages for the seven fraternal benefit societies because some surveys presented totals without breakdowns between fraternal and charitable. Membership fees account for over 90 percent of fraternal receipts although they are 1 percent or less of total receipts. The growth rates of the fraternal benefit societies' receipts vary over the 1930 to 1985 period and may reflect uneven data quality for 1955 and 1965. Gifts received by the fraternal benefit societies from members to be used for charitable purposes climbed to almost one-third of fraternal and charitable receipts in 1985. These contributions are primarily the funds collected by the fraternal benefit societies that are passed on to section 501(c)(3) charitable organizations.

As shown in Table 1, total expenses have grown in a manner similar to total receipts, though total expenses have remained below total receipts. Insurance related expenses were $61.8 million and accounted for roughly 90 percent of the total expenses in 1930. While insurance related expenses had climbed to $3.1 billion by 1985, those expenses represented 95 percent of total expenses.

As shown in Table 1, insurance and annuity benefits represented approximately 55 percent of the total expenses related to insurance for the seven fraternal benefit societies between 1930 and 1985. Policyholder dividends (refunds) have increased significantly as a portion of the fraternal benefit societies' insurance operations. In 1930, refunds were $0.6 million and accounted for about 1 percent of insurance payments to owners and beneficiaries. In 1985, refunds were $428.5 million, or about 20 percent of benefit or annuities and policyholder dividend expenses. This proportion mirrors that of commercial insurers who paid 22 percent of their benefit payments as policyholder dividends. /5/

For the seven surveyed fraternal benefit societies, average fraternal expenses were $15.4 million in 1985 as illustrated in Table 2. Fraternal expenses were about 3 percent of total expenses between 1930 and 1985. These expenses are divided into two categories, non- contract benefits and other fraternal expenses. Non-contract benefits include insurance type benefits, such as payment of adoption or burial expenses. Non-contract benefits are approximately 1 percent of total expenses and have increased from an average of $0.2 million in 1930 to $5.0 million in 1985. Other fraternal expenses include the lodge administration costs as well as fraternal, social, recreational, benevolent, educational, religious, and charitable activities. These expenses have grown at approximately the same annual rate as total expenses, from an average of $0.2 million in 1930 to $9.4 million in 1985.

Charitable expenditures for the 1930 to 1985 period were negligible in comparison to total expenditures, even though they rose to an average of $1.0 million in 1985. However, some of the surveyed fraternal benefit societies' charitable expenses could not be separated from fraternal expenses and do not appear in the charitable total. This suggests that the charitable contribution amounts in Table 2 may understate the amount actually contributed to charities by fraternal benefit societies.

As shown on Table 1, net receipts of the seven fraternal benefit societies have grown from $4.1 million in 1930 to $194.5 million in 1985. For 1975 and 1985, when the most complete data are available, net receipts have been positive for insurance activities and negative for fraternal and charitable activities. This suggests that insurance income subsidizes fraternal and charitable expenditures.

In 1985, the seven fraternal benefit societies had the equivalent of approximately 13,000 full-time paid employees. /6/ As shown in Table 3, since 1965, average employment per fraternal benefit society has increased about 2 percent per year. /7/ While insurance expenses account for approximately 95 percent of total expenses, only 71 to 75 percent of equivalent full-time paid employment went for insurance activities. Employment on fraternal activities accounted for 19 percent of equivalent full-time paid employment.

The activities and operations of fraternal benefit societies vary and, thus, the categories in which the employees were counted in Table 3 for purposes of the study may vary. /8/ For example, a fraternal benefit society may not count a commissioned field representative as an employee because the representative acts as an independent contractor. Another fraternal benefit society may treat a similar representative as an employee because the representative may sell only insurance offered by the society. Similarly, local employees may have a combination of roles such that dividing their time by activity is difficult.

SIZE OF INSURANCE ACTIVITIES

The seven largest fraternal benefit societies account for roughly one to two percent of the life insurance activity in the United States, as measured by insurance in force, insurance purchased, premium income, and assets. /9/ In a few states, however, such as Wisconsin and North Dakota, over 15 percent of ordinary life insurance policies were provided by the seven fraternal benefit societies in 1985. /10/ While not a major provider of insurance in the country, the seven fraternal benefit societies have an important role in regional markets.

Fraternal benefit societies offer primarily life insurance, but also provide annuities, and accident and health insurance. Of the insurance coverage sold in 1985, approximately 95 percent of the policies and premium income of the seven fraternal benefit societies was from life insurance and the remainder from accident and health insurance. /11/ In contrast, commercial life insurers received approximately 60 percent of their premium income from life insurance. /12/ Thus, life insurance makes up a larger portion of the business of fraternal benefit societies than their commercial counterparts.

Smaller fraternal benefit societies generally do not offer annuities, except for settlement purposes. Similarly, the smaller fraternal benefit societies may not offer accident and health insurance. In 1985 the seven fraternal benefit societies studied accounted for almost 90 percent of the life insurance in force, 78 percent of the assets, 82 percent of the written life insurance, 55 percent of the accident and health insurance, and 81 percent of benefits and refunds of all fraternal benefit societies. /13/ The seven fraternal benefit societies studied represent the majority of fraternal benefit society insurance activity in the country.

FOOTNOTES TO CHAPTER 4

/1/ Copies of the original survey questionnaire and a supplement are in Appendix 2.

/2/ The Mennonite Mutual Aid Association (MMAA) was formed in 1966. Excluding the MMAA from the analysis does not materially affect the trends described.

/3/ Net investment income includes income from investment of insurance reserves and accumulated surplus.

/4/ American Council of Life Insurance, Life Insurance Fact Book Update, Washington: American Council of Life Insurance (1987), p. 27.

/5/ American Council of Life Insurance (1987), p. 17.

/6/ The Mennonite Mutual Aid Association (MMAA) has no employees but acquires management and staffing services from Mennonite Mutual Aid, Inc. Those employees are treated as employees of the MMAA for statistical purposes.

/7/ While data were collected on paid employment by the seven fraternal benefit societies in 1930, 1955, and 1975, the variation in the number of fraternal benefit societies reporting amounts and the type of employees included in the count made the data suspect. As a result, these data were not reported in Table 3. Data for 1965 and 1985 were more complete. Employment in 1964 was reported in 1965 for one of the fraternal benefit societies.

/8/ Independent agents are not included.

/9/ The National Fraternal Congress of America, 1985 Statistics of Fraternal Benefit Societies, 1986 Edition, Naperville: The National Fraternal Congress of America (1986), pp. 88, 125, 128, 131, 133, 134, and 186, and American Council of Life Insurance (1987), p. 4.

/10/ All fraternal benefit societies accounted for over 25 percent of ordinary life insurance policies in Wisconsin and North Dakota in 1985. The National Fraternal Congress of America (1986), pp. 42-43, and 56-57 and American Council of Life Insurance, Life Insurance Fact Book, Washington: American Council of Life Insurance (1986), p. 18.

/11/ The National Fraternal Congress of America (1986), pp. 88, 125, 128, 131, 133, 134, and 186.

/12/ American Council of Life Insurance (1987), p. 4.

/13/ The National Fraternal Congress of America (1986), pp. 88, 125, 128, 131, 133, 134, and 186.

END OF FOOTNOTES

CHAPTER 5

INSURANCE ACTIVITY OF FRATERNAL BENEFIT SOCIETIES

INTRODUCTION

In this chapter, the study explores in greater detail the insurance activities of the seven fraternal benefit societies. Specifically, the study analyzes the information to determine: (1) whether the insurance sold by the seven fraternal benefit societies is significantly different from the insurance sold by commercial insurers; (2) whether the seven fraternal benefit societies use their tax exemption to offer members insurance coverage at prices lower than the prices offered by comparable commercial insurers; (3) whether the tax exemption is used to offset inefficient insurance operations; and (4) whether the tax exemption has been used to accumulate surplus in excess of that accumulated by comparable commercial insurers. The results of these examinations are used in assessing the insurance activities of the fraternal benefit societies being studied.

DIFFERENCES BETWEEN FRATERNAL BENEFIT SOCIETY AND COMMERCIAL

All of the surveyed fraternal benefit societies indicated that the insurance products they provide cannot be purchased from commercial insurers, /1/ as there are four distinct characteristics of the insurance they offer. First, unlike commercial insurers who sell to the public, fraternal benefit societies generally have member agents selling insurance to society members. Second, the fraternal benefit societies offer open contract insurance such that the policyholder may be assessed additional premium payments or have benefits reduced to prevent insolvency of the insurer. Third, the societies provide membership benefits that are not part of the insurance contract and for which a premium payment is not charged. These types of benefits include adoption grants, fetal death payments, and payments if a member is diagnosed with a type of cancer. Finally, fraternal benefit societies sell more juvenile insurance than commercial insurers.

MEMBERS ARE GENERALLY THE AGENT

According to the survey, most of the fraternal benefit societies sell insurance only to members. However, because fraternal benefit societies must comply with state insurance regulations, the fraternal benefit societies continue to provide insurance to a policyholder who is no longer a member as long as he or she continues to pay the premiums.

The majority of the seven fraternal benefit societies have insurance agents who are members and who do not sell other commercial insurance. Of those fraternal benefit societies that have agents who also offer other commercial insurance, the fraternal benefit society typically does not offer that type of commercial insurance. For example, the agent may offer commercial accident and health insurance to complement the life insurance offered by the fraternal benefit society. Approximately 3 percent of all the insurance products sold by agents of the seven fraternal benefit societies were insurance products not offered by the fraternal benefit societies.

CONTRACTS CONTAIN AN ASSESSMENT PROVISION

Second, unlike commercial insurers, fraternal benefit societies use an "open contract" with an "assessment provision." The open contract means that the insurance contract references the society's constitution and bylaws, such that any change in either affects the contract. For example, if the fraternal benefit society becomes insolvent, the policyholders may be assessed additional payments to make up the deficiency, or may have their benefits reduced. The assessment provision enables fraternal benefit societies to raise premiums or lower benefits if there is a financial need. As a result of this provision, fraternal benefit societies are exempt from contributing to state guaranty funds designed to protect policyholders. Thus, the fraternal benefit societies self-insure against insolvency.

In contrast, commercial insurers have closed contracts that contain the entire agreement between the company and policyholder. Deficiencies as the result of a commercial insurer becoming insolvent are generally protected by state guaranty funds, however, the court or state insurance commissioners may permit liens, reduce benefits, or place moratoriums on withdrawals from policies. Policyholders of the insolvent company are protected from total loss, because solvent insurers doing business in the state are assessed for the shortfall. /2/ However, these assessments may generally be used by the solvent insurers to offset future state taxes and to provide a deduction for Federal taxes. /3/

OTHER BENEFITS PROVIDED

Fraternal benefit society members also receive benefits that have a risk-pooling or charitable characteristic. These include non- contract benefits, such as fetal death benefits and insurance for children whose health would make them "uninsurable." (See Chapter 6 for further details.)

JUVENILE INSURANCE

Fraternal benefit societies provide juvenile life insurance for children from age 1 day to 14 or 15 years. The price of adult individual policies is generally not affected by the sale of juvenile insurance. Juvenile insurance is often used as a way to introduce future members to the fraternal benefit society operations.

With juvenile insurance from fraternal benefit societies, the child owns the insurance certificate while an adult exercises control on behalf of the child. The fraternal benefit society retains the right to replace the controlling adult if the society deems it is in the best interest of the child. When the child reaches age 21, the child obtains full ownership rights. /4/ While commercial insurers also offer life insurance for juveniles, the insurance is generally owned by the adult purchasing the certificate and not the child. The company cannot replace the controlling adult as in the case of fraternal benefit society insurance. /5/

CONCLUSION

These four distinctions in insurance programs between fraternal benefit societies and commercial insurers do not appear to result in separate markets for life insurance from fraternal benefit societies versus commercial insurers. First, while agents and most policyholders must be members of the fraternal benefit society, there is no requirement that they participate in the society's fraternal and charitable activities. Participation may be encouraged, but it is not required.

Second, the open versus closed contract distinction does not appear to be a significant factor in a policyholder's decision on purchasing life insurance. Open contracts provide members a larger incentive to monitor the financial stability of the fraternal benefit society. Both fraternal benefit societies and commercial insurers, however, must meet similar state insurance regulations to prevent insolvency. In addition, the power to change premiums or benefits has rarely been invoked, as weaker fraternal benefit societies usually seek a merger with stronger ones. Similar1y, courts or state insurance commissioners may permit liens, reduce benefits or place moratoriums on withdrawals on policies offered by insolvent commercial insurers, such that the effect is similar to the assessment provision of an open contract.

Third, while the non-contract benefits associated with being a member of a fraternal benefit society are not offered by commercial insurers, these benefits are not contractually guaranteed, so their payment is similar to other charitable and fraternal expenditures. Finally, the ability of the fraternal benefit society to change the controlling adult for a juvenile policy is unlikely to be exercised, and thus does not provide a meaningful distinction from commercial insurance.

MENNONITE MUTUAL AID ASSOCIATION

The seven fraternal benefit societies surveyed sell life insurance, accident and health insurance and annuities. Life insurance is the primary product offered by the surveyed societies. However, one society, the Mennonite Mutual Aid Association (MMAA), offers primarily health insurance.

The sales material indicates that the health insurance sold by MMAA may be more expensive on average than that offered by commercial insurers. This higher average price allows the MMAA to sell insurance coverage at a lower price to members who may not be able to afford commercial insurance either because of health problems or lower incomes. Thus, insurance for these members is subsidized by healthier or wealthier members. This type of cross-subsidization was not generally cited as an activity of the other surveyed fraternal benefit societies.

The insurance activities of the MMAA appear to be significantly different from those provided by the other six large fraternal benefit societies. The MMAA provides primarily health, as opposed to life insurance, and cross-subsidizes health insurance costs. The other fraternal benefit societies offer small amounts of accident and health insurance as compared to life insurance. Thus, information on accident and health premiums and on costs was not collected from the other surveyed fraternal benefit societies. As a result, the following discussion of the insurance activities of fraternal benefit societies focuses on the six large fraternal benefit societies that sell primarily life insurance.

OVERVIEW OF LIFE INSURANCE OFFERED BY SURVEYED FRATERNAL BENEFIT

The conclusion that fraternal benefit societies and commercial insurers operate in the same life insurance market is supported by the finding that the types of life insurance provided by fraternal benefit societies are similar to those provided by commercial insurers.

Life insurance is generally sold in two basic forms, either as term insurance or as cash value insurance. /6/ Term insurance is characterized as having benefits payable to a beneficiary only when an insured dies within a specified period. The contract is without a significant investment element. Premiums generally pay for the pure insurance, or mortality charge, and an administration, or loading charge. The mortality charge is based on the estimated probability of death during the term of the policy, while the loading charge covers operating expenses and anticipated profit.

Cash value insurance (typically whole life) generally has benefits payable upon the death of the insured or upon surrender of the contract. Premiums have an additional investment, or savings, component. The investment component arises because during one or more of the early years of the policy, the policyholder pays a higher premium than is necessary to cover that year's mortality and loading charges. The "excess" premium accumulates in a fund held by the company for the benefit of the policyholder. This accumulated investment fund is generally referred to as the policy's "cash value" or "cash surrender value."

Table 4 shows the value of insurance in force /7/ offered by the surveyed fraternal benefit societies and commercial insurers. In 1980, fraternal benefit societies had $48.0 billion of insurance in force, consisting of $44.6 billion of ordinary life insurance /8/ and $3.4 billion of other insurance. By 1985, total insurance in force had more than doubled to $93.4 billion with $88.2 billion being ordinary life insurance and $5.2 billion being other insurance. Commercial life insurers had $4,063.6 billion of insurance in force in 1981, /9/ of which only $1,978.1 billion was ordinary life insurance. The other insurance category had $2,085.5 billion of insurance in force in 1981. By 1985, the total insurance in force had grown to $6,053.1 billion, of which more than half, or $3,247.3 billion, was made up of ordinary life insurance, with other insurance accounting for $2,805.8 billion insurance in force.

The "other" category includes group, credit, and industrial life insurance /10/ which are not offered by fraternal benefit societies, as well as accident and health insurance. Thus, the study only examines the ordinary life insurance categories.

For the fraternal benefit societies, term insurance in force was $12.1 billion and cash value life insurance $26.6 billion in 1980. By 1985, the term insurance grew slightly to $12.4 billion while cash value life insurance in force more than doubled to $55.2 billion. A major component of cash value life insurance is the universal and variable insurance /11/ that accounted for $37.6 billion of insurance in force in 1985. For commercial insurers, term insurance in force grew from $760.1 billion in 1981 to $1,201.9 billion in 1985. Cash value life insurance increased more over the period from $1,218.0 to $2,045.4 billion, including an increase in universal and variable insurance from $8.6 to $598.7 billion between 1981 and 1985.

Cash value life insurance accounts for approximately two-thirds of the ordinary life insurance in force of both fraternal benefit societies and commercial insurers. Most cash value life insurance requires a series of level, or equal, premium payments, typically over the life of the contract or a stated number of years. However, universal or variable life insurance in force /12/ represented over half of the cash value life insurance in force of fraternal benefit societies and about one-third by commercial insurers in 1985. Term insurance makes up the remainder of the ordinary life insurance in force of both fraternal benefit societies and commercial insurers.

In the 1980's, fraternal benefit societies shifted, as did commercial insurers, from providing term and level premium cash value life insurance to newer products, such as universal and variable life insurance. Table 5 shows the percentage of first-year premiums /13/ by type of ordinary life insurance for the surveyed fraternal benefit societies and commercial life insurers. First-year premiums for term insurance dropped from 26 to 2 percent of the total for fraternal benefit societies between 1980 and 1985. Over the same period, first- year premiums for term insurance declined from 18 to 11 percent for commercial life insurers. By 1985, universal and variable life insurance premiums accounted for 97 percent of the total first-year premiums for fraternal benefit societies and 42 percent for commercial insurers. The figures on universal and variable life insurance premiums for fraternal benefit societies may, however, overstate the extent of the shift to these products because the premiums are not annualized. For example, the total premium payment of a single premium policy would be included in the first year figures for the fraternal benefit societies while only 10 percent of the same premium would be included for commercial insurers. /14/

Table 6 presents the average ordinary life insurance policy size issued in a given year as well as the average insurance in force per year. /15/ The size of policies issued by the fraternal benefit societies tends to be smaller than the weighted average for all commercial insurers, though the Aid Association for Lutherans, Independent Order of Foresters, and the Lutheran Brotherhood have a few years where the average is larger. Only on four occasions does the average for any of the fraternal benefit societies exceed the average for all mutual insurers.

Including both old and new contracts, the average policy in force of the fraternal benefit societies tends to be smaller than the weighted average for the industry and for the mutual life insurers. Only the Aid Association for Lutherans has a number of years where the average policy size exceeds the average for the industry and for mutual insurers. In general, the fraternal benefit societies issued and have in force smaller average policies than comparable commercial insurers. This could result in fraternal benefit societies having higher administrative costs than commercial insurers with similar amounts of total insurance.

COMPARISON OF THE PRICE OF LIFE INSURANCE OFFERED BY FRATERNAL

As the insurance products sold by fraternal benefit societies are similar to those sold by commercial insurers, the study compares the prices charged by fraternal benefit societies to those charged by commercial insurers. Because the amount of policyholders dividends may vary from year-to-year, the study examines the prices of whole life policies prospectively and retrospectively to assess whether fraternal benefit societies charge a significantly lower price compared to commercial insurers. /16/ For the former, the study examines cost data for whole life contracts sold in 1988 projecting the 1988 dividend into the future. With the latter, the study analyzes cost data for whole life contracts sold 10 or 20 years ago. Both of these analyses rely on dividing the price of a policy into its various components.

The study selected one sample of commercial companies, large mutual life insurance companies, for comparison purposes. As discussed in more detail later, A.M. Best's comparisons of retrospective policy prices suggests that pricing of life insurance policies by larger companies /17/ is significantly different from that done by smaller companies. Since the six fraternal benefit societies that are most active in the life insurance business have larger operations based on premiums, the comparison group selected was larger life insurance companies with the same organizational capital structure. The study selected mutual life insurance companies for comparisons because, similar to fraternal benefit societies, mutual insurance companies raise capital from policyholders and cannot raise capital by issuing stock. Appendix 3 lists the large mutual life insurance companies which were used in the analyses. /18/

The total amount paid for life insurance varies, depending upon the premiums paid, mortality charges, dividends paid, policy fees, and the time value of money. /19/ In general, the higher the premiums, mortality charges, and policy fees, the more costly the policy. Two approaches used to assess the price of insurance are the interest-adjusted payment index and the interest-adjusted surrender cost index. /20/ These approaches require that similar policies be compared in order to avoid adjusting the face amount of insurance. In both situations net premiums (or gross premiums less dividends paid to the policyholder) are increased by an interest rate while total net costs are discounted at the end of the period under examination. The cost is then calculated to be the average annual amount needed to cover death benefits, expenses, and profits by the insurer. /21/ The primary difference between these indices is that the surrender cost index measures the annual average out-of-pocket cost if the policy is terminated, while the payment index measures the annual average cost at a point in time without terminating the policy. As a result, the former includes a terminal dividend and cash value payment which can be quite large. For both indices, the lower the value, the lower the cost of insurance.

For purposes of this analysis, the study assumed the market for life insurance policies to be the United States, such that companies charge one price to all customers for their policy. It is possible that market boundaries may in fact be smaller geographic regions and policy prices vary according1y by region (which may be served by subsidiaries). A fraternal benefit society lacking a subsidiary structure can charge only one price, but is likely to suit the price it charges to its major market. Data, such as cross-price elasticities, would be needed to assess market boundaries, but are not available. Accordingly, the analysis assumes that insurers charge one price for a given policy.

The Treasury survey of fraternal benefit societies asked for the surrender cost index for whole life policies issued in 1988 for smokers and non-smokers, women and men, at age 25 and 55 years old, for $25,000 and $100,000 policies. The cost surrender index assumes the dividend rate for 1988 will continue for the ten- or twenty-year period being analyzed. For similar policies, the study compared the distribution of the surrender cost index for fraternal benefit societies to that of large mutual life insurers.

Table 7 shows the results for the surveyed fraternal benefit societies and large mutual life insurers where data from at least ten mutual insurers are available. For the fraternal benefit societies, the analyzed cost per $1,000 of insurance would be $2.18 for a 25 year old male who does not smoke and bought a whole life, level premium policy for $25,000 and surrendered the policy after twenty years. The sample standard deviation indicates how much variation there is among the surveyed fraternal benefit societies in their life insurance costs. The fraternal benefit societies that offer the $100,000 policy have a lower cost per $1,000 of insurance for the larger than for the smaller policy, however, there is wide variation as to how much less. Some fraternal benefit societies have a relatively small differential of 2 percent, while other fraternal benefit societies switch from having a positive cost for the smaller policy to a negative cost for the larger one. A negative cost indicates that the premiums and investment income more than offset the cost of the insurance policy. Similarly, all policies have lower costs for the twenty-year versus the ten-year cost horizon, which range from a 5 percent difference to costs going from positive to negative. The twenty-year cost indices for the $100,000 policies are more likely to have investment income that offsets the costs of the policy. This is particularly true for policies insuring a 25 year old person, as payments to beneficiaries are not as likely.

As shown on Table 7 for the policies sampled, the mean cost for similar life insurance policies is less for the life insurance offered by the large mutual life insurers than the surveyed fraternal benefit societies. However, the surrender cost indices of fraternal benefit societies are not significantly different /22/ from those of large mutual life insurers.

As an alternative to examining projected policy costs or prices, the study examined the 1988-through-1991 cost indices using data on dividends actually paid over the previous ten- or twenty-year period. One of the advantages of this approach is that actual dividends, as opposed to projections of the 1988 dividend rate into the future, are used in assessing value in the payment and surrender cost indices. /23/ Again, the major difference between these indices is that the surrender cost index includes the terminal dividend and cash value while the payments index does not. For both indices, the lower the dollar amount per $1,000 of insurance, the lower the cost.

Best's Life/Health Review's annual dividend comparison was used as the basic data source for determining whole life policy costs of large mutual insurers. /24/ These data have only been available since 1988 with respect to policies that are ten or twenty years old. The annual Best's studies are of policies that were close to the industry average for the year they were issued. All the policies are participating whole life issued to 35 year old males. /25/ Any policy fees are included in the premium price. Similar data for policies offered by fraternal benefit societies were obtained from Best's Flitcraft Compend for 1988 through 1991. While these whole life policies are representative of the type of policy sold at the time, variable and universal life insurance policies became more popular in the 1980's. As a result, these historical cost indices do not assess the cost of variable and universal life insurance policies.

Table 8 shows the mean and sample standard deviations of the premium, payment index, and surrender cost index for selected policies and years of the sample of larger mutual life insurers and most of the fraternal benefit societies. In general, average premiums were lower for fraternal benefit societies than for large mutual insurers, but the differences are not statistically significant. The lower premiums charged for twenty-year policies by fraternal benefit societies are significant only for 1989. However, premiums tend to overstate the cost of insurance in comparison to the cost indices. The payment index for a twenty-year policy for $10,000 is significantly lower for fraternal benefit societies than for large mutual life insurers only for 1989. In general, the distributions of both the payment and surrender cost indices are not significantly different for fraternal benefit societies and large commercial insurers. /26/ This suggests that similar policies offered by fraternal benefit societies and comparable large commercial insurance companies are priced the same.

Fraternal benefit societies appear to sell life insurance policies that are similar to those sold by comparable mutual insurers. Fraternal benefit societies increased sales in the 1980's of new insurance products, such as universal and variable life insurance, just as commercial insurers did. Finally, analysis of insurance cost indices suggests that fraternal benefit societies' life insurance is similar in price to that offered by comparable mutual insurers.

COMPARISON OF THE EFFICIENCY OF FRATERNAL BENEFIT SOCIETIES AND LARGE

Analysis in the previous section suggests that fraternal benefit societies charge approximately the same price for comparable insurance policies as is charged by commercial insurers, therefore, the tax exemption provided fraternal benefit societies appears not to be passed on by means of a lower price to policyholders for insurance coverage. Another potential use of the tax exemption would be financing inefficient operations. For example, some fraternal benefit societies noted on their surveys that they have offices operating in regions with few members. This increases their operating expenses and possibly would not be done by a taxable competitor. This chapter attempts to examine whether the surveyed fraternal benefit societies are as efficient as their comparable taxable competitors, large mutual life insurers.

According to the 1988 Treasury survey, the seven fraternal benefit societies underwrite essentially all of the insurance that they issue with only about 1 percent of all the insurance in force being reinsured by another company. Fraternal benefit societies are not agents selling other organizations' insurance, but operate in the same manner as commercial insurers.

An efficient insurance operation is generally characterized as being financially strong and having low operating, primarily underwriting and investment, expenses. The different types of expenses are often related to each other. For example, underwriting expenses, such as mortality experience, are influenced by the type of policies, or lines of insurance, that the company sells. To assess whether fraternal benefit society expenses are low in comparison to comparable mutual life insurance companies, the study gathered measures related to operations from the Best's Insurance Reports: Life/Health (Best's). Specifically, the study collected Best's overall rating of company performance, whether required interest obligations could be met, mortality experience, renewal expenses, lapse rates, investment expenses, and net yield ratios for six of the societies. In addition, similar data for 1980 through 1990 were collected for the sample of approximately twenty-six large mutual life insurance companies.

For the life insurance industry, Best's collects data on the financial condition and operating results of many of the commercial insurers and fraternal benefit societies operating in the United States. Based on the data collected, Best's provides an opinion of the organization's relative financial strength and ability to meet contractual obligations. Quantitative analysis examines the organization's profitability, leverage, and liquidity. In addition, Best's qualitatively evaluates the company's spread of risk, reinsurance activity, quality and diversification of investment, adequacy and valuation of reserves, and management. Based on this evaluation, Best's assigns a rating ranging from A+ (Superior) to C- (Fair) to approximately 60 percent of the organizations examined. The other 40 percent are not assigned a rating due to lack of data, etc. /27/

As shown on Table 9, the fraternal benefit societies included in the analysis received Best's ratings of Superior (coded as 1) for all the years from 1980 through 1990. While most large mutual insurance companies also received the Superior rating, there is some variation in 1990. However, the difference in ratings is not statistically significant in 1990. /28/ Fraternal benefit societies were rated as favorably as large mutual insurance companies for 1980 through 1990.

In addition to the overall Best's ratings, various aspects of the operations were also examined. Specifically, Best's assesses whether the company has adequate total net investment income to meet issued policy commitments. Their assessment of the required interest varies from "more than ample," coded as I, to "less than required," coded as 3. Most insurers were determined to have "more than ample" investment income with a few having "sufficient" amounts.

Based on the year by year analyses of Table 9, fraternal benefit societies are more likely to have "more than ample" required interest than comparable large mutual insurance companies. /29/ large mutual insurance companies are closer to having "ample" required interest income. This suggests that fraternal benefit societies are at least as efficient as their commercial counterparts with respect to meeting policy commitments out of investment income.

To assess the underwriting experience of the companies, Best's assessment of the mortality experience was examined since Best's adjusts the mortality ratio (expected to actual) for the age of business and the types of business. For example, mortality rates are likely to be lower for new business and insurance sold to a group of younger people than for continuing insurance sold to older people. Best's rates mortality experiences from "most favorable," coded as 1, to "unfavorable," coded as 4.

The "most favorable" rating is relatively rare so that fraternal benefit societies and large mutual insurance companies received on average a rating closer to 2 or "very favorable." Table 9 shows the annual mean ratings for both groups. Only the mean values for 1980 through 1983 are significantly different with fraternal benefit societies having less favorable ratings for 1980 through 1982. Annual regression analysis suggests that only for 1980 and 1981 are the mortality experiences of the selected fraternal benefit societies significantly less favorable than those of large mutual life insurers. /30/

Best's also computes an average renewal expense ratio as part of its assessment of insurance expenses. These ratios are on a per $1,000 basis and are adjusted for the higher cost of first-year business and variations in premiums. For example, the cost of issuing new life insurance is higher than that of maintaining old business. Similarly, premiums vary depending on whether the company is a low net cost or high net cost operation. In general, the lower the renewal expense ratio, the less costly the insurance offered by the company.

As reported in Table 9, the renewal expense ratio is around $4 per $1,000 of insurance for fraternal benefit societies and large mutual life insurers. The ratio is not significantly higher, or less favorable, for fraternal benefit societies despite the fact that large mutual insurers generally have a smaller variation in their expense ratios. /31/ With respect to renewal expenses, fraternal benefit societies appear to be at least as efficient as large mutual insurers.

Because most expenses are associated with selling a policy, the longer the policy is in force, the lower the underwriting expenses per year. The lapse rate is the amount of ordinary life insurance terminated in a year, other than by death or maturity, divided by the amount of insurance outstanding (the ordinary life insurance in force at the beginning of the year plus the prior year's new business issued). The rate is only calculated for policies he1d for more than one year because the first year lapse rates tend to be twice those of later years.

For the 1980's period the lapse rate is 8.3 percent for fraternal benefit societies and a less favorable 10. 1 percent for large mutual life insurance companies. As shown on Table 9, the lapse rate is lower for fraternal benefit societies than for large mutual life insurance companies for every year though the difference is not statistically significant for 1980, 1981, and 1984 through 1986. Annual regression analysis suggests that the difference in lapse rates is also not statistically significant for 1987 and 1990. /32/ These analyses indicate that fraternal benefit societies have comparable lapse rates to those of large mutual life insurers.

The study examined whether the investment expense ratio was significantly more for fraternal benefit societies than for large mutual insurance companies. This ratio is the total investment expenses divided by the total gross investment income. The lower the investment expense ratio, the lower the operating costs, and hence, the more favorable for the operation.

On average for the 1980's period, the investment expense ratio is 4.6 percent for fraternal benefit societies and 8.8 percent for large mutual insurance companies. As shown on Table 9, for every year presented, the ratio is lower for fraternal benefit societies. In no year is the investment expense ratio significantly less favorable or higher for fraternal benefit societies. Annual regression analysis suggests that the investment expense ratios of fraternal benefit societies and large mutual insurance companies are not significantly different for 1984 through 1990 when differences in the types of investments are controlled for statistically. /33/

Finally, the study examined whether the net yield ratio was significantly less for fraternal benefit societies than large mutual life insurers. The net yield ratio is net investment income to net invested assets that are adjusted for accrued investment income. If the ratio was significantly less, it would suggest that investment management is less efficient in fraternal benefit societies than comparable commercial insurers.

For the 1980 to 1990 period, the net yield ratios are very similar for fraternal benefit societies at 9.4 percent and 8.9 percent for large mutual insurers. Moreover, for 1980 through 1985, the yield rates are significantly greater for fraternal benefit societies than for large mutual insurers. /34/

Together, these measures of financial and operating health suggest that the six largest fraternal benefit societies were at least as efficient as large mutual insurers during the 1980's. While the comparisons suggest that the fraternal benefit societies are more efficient in some respects, this may reflect the samples used in the analyses. Moreover, only the six largest fraternal benefit societies were compared to a larger set of twenty-six large mutual insurance companies.

COMPARISON OF SURPLUS ACCUMULATION BY FRATERNAL BENEFIT SOCIETIES AND

In Chapter 4 the increasing importance over time of net investment income is demonstrated. This investment income is generated by earnings on policy reserves and surplus. The study examines whether the tax exemption has been used by fraternal benefit societies to accumulate surplus in excess of that accumulated by comparable large mutual life insurers. /35/ Accumulated surplus is particularly important for mutual life insurance companies and fraternal benefit societies, as outside capital cannot be raised by issuing stock. In addition, fraternal benefit societies do not participate in the state guaranty fund designed to protect policyholders. Instead, the fraternal benefit societies self-insure against insolvency.

Life insurance is generally priced using conservative estimates of investment earnings rates and higher mortality and expense rates. The use of conservative assumptions in pricing a policy ensure the viability of the policy and company. Life insurance is generally a long term contract, thus, the security of the policy is dependent upon the long term financial security of the company selling the contract. Amounts held as life insurance reserves that are accumulated from premiums and investment income are the primary guarantee that the company can meet policy obligations. State insurance regulation ensures that reserves meet minimum standards.

The assumptions used in pricing policies also result in profits for the insurer. Some of these profits may be returned to policyholders as policyholder dividends, or distributed to stockholders in the case of a stock insurance company. However, profits are also retained by the company as accumulated surplus, or the difference between assets and liabilities. This surplus can operate as a contingency fund, such that the larger the surplus for a given block of outstanding policies, the more secure the company and the long term viability of the policies. However, gross accumulated surplus values do not reflect the quality of the investments, mortality, or expenses, all of which can affect the financial security of the company.

Accumulated surplus is also used by companies to finance newer policies. In general, policy expenses are greatest for the initial years that a policy is in force. As a result, the premium and investment income for a block of policies sold at the same time may not be adequate to meet the financial obligations of the policy. The company essentially borrows funds from accumulated surplus to meet this financial need. In later years expenses become smaller such that premiums and investment income for the block of policies are adequate for funding the necessary reserves and generating accumulated surplus. As a result, the larger the accumulated surplus, the more policies a company can write.

The study analyzes the financial statements prepared in accordance with statutory accounting practices required by regulators as they appear in Best's Insurance Reports: Life/Health. /36/ Statutory surplus is the sum of special surplus funds and unassigned surplus. The former is accumulated surplus that is earmarked for specific contingencies. Despite the uniform definition, there can be variation in what is included in accumulated surplus. For example, companies may assign funds to a special surplus fund or to a special reserve in order to hold funds for a contingency. In addition, accumulated surplus may vary from year-to-year reflecting changes in reserve assumptions and income from operations. For this reason, the accumulated surplus for 1965, 1970, 1975, 1980, 1985, and 1990 is examined.

Because of the conservative nature of statutory accounting, which is likely to understate the value of assets and overstate the value of liabilities, accumulated surplus is likely to be understated. For example, some assets, such as furniture and equipment, may be excluded from those counted in statutory assets, while the major liability of policy reserves is valued using conservative estimates of future mortality, interest, premiums, and dividends.

The annual growth rates of accumulated surplus for the 1965 to 1990 period were examined for fraternal benefit societies and large mutual life insurers in order to assess whether fraternal benefit societies added to accumulated surplus at a greater rate than their commercial counterparts. /37/ Table 10 shows that while the rates are similar for some periods, the growth rates are significantly greater for fraternal benefit societies for the 1965 to 1970 period (6.8 versus 3. 1 percent) and 1980 to 1985 period (12.2 versus 6.2 percent). /38/ This 1980 to 1985 period is also when net yield ratios were significantly greater for fraternal benefit societies than large mutual insurance companies.

To determine whether the amount of accumulated surplus is reasonable, the ratio of surplus relative to risk is important. Two measures of risk are used in the study: total liabilities (of which the major component is reserves) and insurance in force. The problem with using surplus to total liabilities is that some types of insurance, such as group life insurance and accident and health insurance, may require smaller reserves relative to the amount of risk. On the other hand, surplus to insurance in force ratios give too great a weight to these types of insurance.

As appears on Table 10, both surplus to risk measures are higher for the large fraternal benefit societies than for large mutual life insurance companies. In all cases the relative amount of surplus is significantly greater for fraternal benefit societies. /39/ These results suggest that fraternal benefit societies have added to accumulated surplus at a greater rate and have greater accumulated surplus levels than comparable mutual insurers. Thus, it appears that the tax exemption may, in part, finance additions to and be retained in surplus.

SUMMARY

In this chapter, the study examined whether the insurance offered by fraternal benefit societies is significantly different from that offered by commercial insurers. While there are some distinctions, the study concluded that the insurance policies of fraternal benefit societies appear to serve the same markets as those served by commercial insurers. In examining the trends in the types of life insurance issued by fraternal benefit societies and comparable commercial insurers, the study found that both groups began to offer variable and universal life insurance contracts during the 1980's. This suggests that fraternal benefit societies responded to the same market forces as commercial insurers. Specifically, as commercial insurers altered the mix of products offered, so did fraternal benefit societies.

The study assessed whether the price of insurance (cost to the policyholder) charged by fraternal benefit societies was significantly less than that charged by commercial insurers. It examined a number of specific comparable policies both prospectively and retrospectively, and generally found that fraternal benefit societies charged prices that were not significantly less than those charged by comparable large mutual life insurers. This suggests that the corporate income tax exemption was not being rebated to the policyholders.

The study examined whether the fraternal benefit societies operated less efficiently than comparable large mutual life insurers. Using various measures of operating expenses and financial strength, the study concluded that fraternal benefit societies appear to be as efficient as their taxable counterparts.

Finally, the study analyzed whether fraternal benefit societies have added to accumulated surplus at a greater rate and have significantly greater accumulated surplus positions than comparable large mutual insurance companies. The study found that fraternal benefit societies added to accumulated surplus at higher rates during some periods and have significantly higher relative levels of surplus than large mutual life insurance companies. This suggests that the corporate income tax exemption is in part financing accumulation of surplus by fraternal benefit societies.

FOOTNOTES TO CHAPTER 5

/1/ Economists would describe the issue as whether insurance from fraternal benefit societies is in a market separate from that of insurance from commercial insurers.

/2/ Kenneth Black Jr. and Harold Skipper, Jr., Life Insurance. Eleventh Edition, Englewood Cliffs: Prentice Hall (1987, pp. 188-190 and 581-582.

/3/ James Barrese and Jack M. Nelson, "Distributing the Cost of Protecting Life-Health Insurance Consumers," Statement presented before the Senate Committee on Judiciary Subcommittee on Antitrust, Monopolies, and Business Rights (April 28, 1992), estimate that in 1990, 14 percent of the assessments by guaranty funds willbe allocated by insurers to policyholders, equity holders, and employees.

/4/ 1988 Treasury survey of Fraternal Benefit Societies.

/5/ Black and Skipper, p. 111. Both fraternal benefit societies and commercial insurers offer primarily term insurance to juveniles.

/6/ More details on the different types of life insurance products can be found in U.S. Department of the Treasury (March 1990).

/7/ Insurance in force is the sum of the face amount, or the amount that would be paid in the case of death or at the maturity of the policy, and dividend additions of policies outstanding at a given time. American Council of Life Insurance (1986), p. 126.

/8/ Ordinary life insurance is life insurance usually issued in amounts of $1,000 or more with premiums payable on an annual, semi- annual, quarterly, or monthly basis. See American Council of Life Insurance (1986), p. 126, for more details.

/9/ Comparable data for 1980 were not available, so data from 1981 were used.

/10/ Group life insurance is typically issued for a group of people under a master policy. As part of a loan transaction, a debtor may purchase a credit life policy, a term insurance policy for the amount of the outstanding loan. Industrial life insurance is issued in small amounts, usually less than $1,000 of insurance, with premiums payable on a weekly or monthly basis. See American Council of Life Insurance (1986), pp. 123-125, for more details.

/11/ In a variable contract, the cash value is invested in a portfolio of assets segregated from the general investment accounts of the company. The policyholder can exercise a measure of control over the investment of his or her funds. A universal life policy allows the policyholder to vary the schedule of premium payments and the level of death benefits.

/12/ Universal and variable insurance policies permit policyholders greater flexibility and potentially higher returns from investing in cash value life insurance.

/13/ First-year premiums from policies that are exchanged for another policy are included.

/14/ Data on the U.S. is based on a LIMRA survey cited in Table 4.5 of U.S. Department of the Treasury (March 1990).

/15/ These data are from Best's Life/Health Review (November 1991), pp. 62-64 and (December 1988), pp. 77-80. Averages for all commercial insurers and mutual life insurers are from A.M. Best Company, Best's Aggregates and Averages: Life/Health, Oldwick: A.M. Best Company, Inc. (1991 and 1989).

/16/ This analysis does not account for other characteristics that may influence the price of the policy. For example, an insurer with a more extensive sales agent network may have more expensive insurance because the policyholder is paying for additional service. To the extent that both commercial insurers and fraternal benefit societies vary in a similar manner in the types of characteristics they offer, however, the average price should be the same. Evidence presented later suggests that this is the case.

/17/ Larger companies issued at least $100 million of ordinary life net premiums written in 1988, $7 million of direct ordinary life 4premium income, $200 million of ordinary life face amount issued, $500 million of ordinary life face amount in force, and were licensed in at least twenty states.

/18/ Our working hypothesis is that large fraternal benefit societies and large mutual insurance companies make up a population of companies that are similar in their operations. Our two samples from this population differ only in whether they are fraternal benefit societies or mutual life insurance companies.

/19/ This analysis does not account for the other benefits that fraternal benefit societies provide.

/20/ See Appendix 4 for more details about these measures.

/21/ Black and Skipper, pp. 161-162.

/22/ This result is not surprising given the small number of policies in the comparison. In general, the smaller the sample size, the greater the difference must be in order for the difference to be considered significant. Appendix 5 contains the test statistics for Table 7 and the other tables in this section. A 5 percent level of significance was used in a one-tail test. The other statistical tests in the report also use a 5 percent level of significance in one-tail tests.

/23/ See Appendix 4 for more delails on how the cost indices are computed.

/24/ Best's Review. Life/Health Insurance Edition, "10-Year Dividend Comparisons," (November 1988) pp. 107-114, "20-Year Dividend Comparisons," (November 1988), pp. 97-106, "10-Year Dividend Comparisons: $50,000 Policy," (August 1989), pp. 68-82, "10-Year Dividend Comparisons: $25,000 Policy," (August 1989), pp. 84-94, "20- Year Dividend Comparisons: $10,000 Policy," (July 1989), pp. 86-96, "20-Year Dividend Comparisons: $25,000 Policy," (July 1989), pp. 97- 104, "10-Year Dividend Comparisons," (August 1990), pp. 86-93, "20- Year Dividend Comparisons," (July 1988), pp. 64-72, "10-Year Dividend Comparisons," (August 1991), pp. 46-58, and "20-Year Dividend Comparisons," (July 1991), pp. 89-102.

/25/ with a participating policy the company agrees to distribute to policyholders part of company surpluses or profits. All life insurance policies written by mutual life insurance companies and fraternal benefit societies are participating policies. The policy would be paid up no earlier than age 85 but no later than age 100. Premiums were either level or modified for only the first five years of the policy. Policies chosen were as similar as possible to increase the validity of the comparison.

/26/ While large mutual insurers may have lower cost indices than fraternal benefit societies, we are only examining whether fraternal societies charge a lower price for a given insurance policy.

/27/ See "Preface" to A.M. Best Company, Best's Insurance Reports: Life/Health, Oldwick: A.M. Best Company, Inc. (various years), for a more detailed description of the Best's ratings.

/28/ Statistical tests cannot be done in other years because there is not enough variation. However, the study did test whether fraternal benefit societies have a more favorable, or lower numeric rating, in 1990 using a one-tailed t-test at a 5 percent significance level. The t of 0.660 with 29 degrees of freedom is not statistically significant.

/29/ The statistical tests indicate that fraternal benefit societies have significantly more interest than required in comparison to mutual insurance companies. The study is interested, however, in whether fraternal benefit societies have at least as strong required interest positions. In addition to performing t-tests on the annual means for fraternal benefit societies and large mutual insurers, the analysis of variance and the regression techniques were used to assess whether there is a significant difference. The results are the same even correcting for heteroskedasticity of the errors. The chi-squared test statistics are reported in Appendix 5.

/30/ When all years of the sample are pooled, the Durbin-Watson statistic was 0.642 for 352 observations suggesting the presence of serial correlation. Therefore, the analysis was done using annual regressions. Furthermore, since the errors from the annual regressions are heteroskedastic, the tests were made using White's asymptotically consistent variance-covariance matrix. The tests are whether fraternal benefit societies had significantly less favorable or higher numbered ratings than large mutual insurance companies. Appendix 5 reports the results of these tests.

/31/ Results from the annual regression analysis are reported in Appendix 5.

/32/ When all years of the sample are pooled, the Durbin-Watson statistic was 1.462 for 352 observations, suggesting the presence of serial correlation. Therefore, the analysis was done using annual regressions. Furthermore, since the errors from the annual regressions are heteroskedastic, the tests were made using White's asymptotically consistent variance-covariance matrix. The tests are whether fraternal societies had significantly less favorable or higher lapse rate than large mutual insurance companies. Because the lapse rates are likely to vary with the type of insurance, the study controlled for average policy sizes, the amount of whole, term, and group insurance in force, as well as the accident and health net premiums written. Appendix 5 reports the results.

/33/ When all years of the sample are pooled, the Durbin-Watson statistic was 1.462 for 352 observations suggesting the presence of serial correlation. Therefore, the analysis was done using annual regressions. Furthermore, since the errors from the annual regressions are heteroskedastic, the tests were made using White's asymptotically consistent variance-covariance matrix. The tests are whether fraternal societies had significantly less favorable or higher investment expense ratios than large mutual insurance companies. Because the expense ratios are likely to vary with the type of investments in the organization's portfolio, the study controlled for the percentage of the portfolio invested in bonds, stocks, mortgages, and real estate as well as the net yield rate. Appendix 5 reports the results.

/34/ When all years of the sample are pooled, the Durbin-Watson statistic was 0.285 for 352 observations, suggesting the presence of serial correlation. Therefore, the analysis was done using annual regressions. Furthermore, since the errors from the annual regressions are heteroskedastic, the tests were made using White's asymptotically consistent variance-covariance matrix. The tests are whether fraternal benefit societies had significantly higher net yield ratios than large mutual life insurance companies. Appendix 5 reports the results.

/35/ The following discussion is based on Black and Skipper, pp. 194-196, 380-382, and 560.

/36/ There are no generally accepted accounting principles (GAAP) for mutual life insurance companies.

/37/ Annual growth rates for a five year period are calculated.

/38/ See Appendix 5 for the related test statistics. A one- tailed t-test using a 5 percent significance level was used to test whether fraternal benefit societies have significantly higher annual growth rates of accumulated surplus.

/39/ See Appendix 5 for the related test statistics. A one- tailed t-test using a 5 percent significance level was used to test whether fraternal benefit societies have significantly higher surplus. In 1965, 1970, and 1975, however, the variation as measured by the standard deviation for surplus to liabilities for fraternal benefit societies is significantly greater than for large mutual life insurance insurers. A similar problem occurs with respect to surplus to total insurance in force in 1990.

END OF FOOTNOTES

CHAPTER 6

FRATERNAL AND CHARITABLE ACTIVITIES OF

OVERVIEW OF FRATERNAL AND CHARITABLE ACTIVITIES

In previous chapters, the study presented the broad receipt and expense data on fraternal and charitable activities. In this chapter, more detail is provided on the types of fraternal and charitable activities supported by fraternal benefit societies. Where possible, charitable activities are separated from fraternal ones.

As discussed in Chapter 4, net receipts for fraternal and charitable activities in the seven largest societies were -- $14 million in 1975 and -- $96 million in 1985. While difficult to separate, roughly 75 percent of receipts for fraternal and charitable items were from fraternal activities while over 90 percent of fraternal and charitable expenses were for fraternal expenses. With respect to expenses, non-contract benefits accounted for approximately 33 percent of fraternal expenses for the seven fraternal benefit societies. These figures include both national, or home office, revenues and expenses as well as amounts from the local level.

A major fraternal expense that occurs primarily at the national level is for non-contract benefits which are insurance type benefits available to members or their dependents. For example, the Mennonite Mutual Aid Association pays $1,500 or 80 percent of adoption expenses paid by members and burial expenses up to $2,000 for members facing financial hardships. The Aid Association for Lutherans, the Independent Order of Foresters, and the Lutheran Brotherhood have orphan benefits for dependents of deceased members that vary from $66 per month to $210 per month depending on the age of the orphan. In addition, special post-secondary education benefits are available to orphans. Finally, some of the societies provide benefits or insurance for newborns and children. For example, the Aid Association for Lutherans provides $750 for a still birth to a member and $1,500 for the death of a child of a member within 60 days of birth, and $2,500 of life insurance coverage for children of members. These membership benefits account for 20 to 30 percent of all fraternal expenses for the seven fraternal benefit societies and have grown in importance over the last 20 years.

Most of the fraternal receipts at the national level are from membership fees. While expenses related to membership are not separately shown, they are likely a function of the number of members as well as the lodge structure. Table 11 presents the number of lodges and members for 1981, 1984, and 1985 and the number of lodges for 1986 through 1989 for the seven societies. These seven fraternal benefit societies account for approximately half of all lodges in fraternal benefit societies and 60 percent of all fraternal benefit society members in the United States. The number of lodges declined between 1981 and 1984, which was primarily due to a decrease in the number of Lutheran Brotherhood lodges. However, the number of lodges has increased between 1984 and 1989. The number of members has declined through this period, as is the case for all fraternal benefit societies in the United States. Because the number of members per lodge has decreased over the 1980's, the overhead cost per member of running the lodge structure may have increased over this period.

MEMBERSHIP IN FRATERNAL BENEFIT SOCIETIES

When discussing the rationale for tax exemption, we noted that there are two broad types of organizations, the traditional tax- exempt section 501(c)(3) organization and the more club-type organization. Fraternal benefit societies fit the general club category in that they have a lodge structure and a membership united by a fraternal bond. Since fraternal benefit societies are governed by their members, determination of whether an individual is entitled to admission rests with the members.

There are four major types of fraternal bonds. First, members may share the same faith or be related to someone who does. These societies typically provide financial support for the clergy, religious orders, seminaries, individual congregations, or schools associated with the faith. Examples of this type of fraternal bond are the Aid Association for Lutherans and the Lutheran Brotherhood which require members be Lutherans, the Knights of Columbus whose members must be Catholic men, and the Mennonite Mutual Aid Association whose members are Mennonites or Anabaptists. Often these fraternal benefit societies use the congregation as the local lodge or branch within the fraternal organization.

Second, society members may share a common ethnic background. These societies often act to preserve traditional culture and provide stability and a sense of identity to members. While not included in the group of fraternal benefit societies being studied, some examples include the Croatian Fraternal Union of America, the Danish Brotherhood in America, and the Ukrainian National Association.

Third, the fraternal bond may be that members have related occupations. Often these fraternal benefit societies are concerned with occupational safety conditions and preventing work-related injuries. Though not included in the sample of seven, examples of this type of society include the American Postal Workers Accident Benefit Association, the Police and Firemen's Insurance Association, and the United Transportation Union Insurance Association.

Finally, there is the broad category that requires members to be committed to high moral standards, ethical conduct, patriotism, good citizenship, and traditional family life. Fraternal benefit societies with this type of bond vary considerably as to membership requirements. Often these societies are based on ancient Greek societies, medieval guilds, and English friendly societies. Of the fraternal benefit societies studied, the Independent Order of Foresters, Modern Woodmen of America, and Woodmen of the World have this more general fraternal bond.

About 3 percent of applicants for membership in the seven surveyed fraternal benefit societies are rejected. This low rejection rate partially reflects the pre-screening that is done prior to membership application. None of the societies waive membership requirements for applicants. New members are usually recruited by current members or at church related activities. Membership dues averaged $15 per year in 1985 with some variation. Some societies assess dues based on the member's level in the lodge hierarchy. Members of fraternal benefit societies are expected to purchase insurance from the society.

In addition, many of the fraternals have juvenile memberships and programs for young persons. Generally, full membership privileges are available for those over the ages of 16, or in some cases, 18 years old. Juvenile members tend to be children of adult members. Finally, some fraternal benefit societies have social members.

TYPES OF ACTIVITIES OF FRATERNAL BENEFIT SOCIETIES

Fraternal benefit societies believe that the fraternal bond is strengthened by fraternal and benevolent work by members. The types of activities include mutual aid or self help for members as well as charitable activities helping members and non-members. Although fraternal benefit societies collect some information on the amount of money spent, the number of hours volunteered, and the number of events, this information is difficult to break down between mutual aid and charitable activities. While some non-members join in the charitable activities, the vast majority of charitable work is performed by members.

TYPES OF ACTIVITIES

Initiation of specific activities can be either at the national or local levels. A national organization, or home office, will often provide financial, personnel, and program support to the local lodges to run a specific program. In addition, the national organization may match the funds raised by the local lodge for a project. For example, the national Lutheran Brotherhood matches local funds raised to aid Lutheran churches. However, lodges can also initiate fraternal activities for their communities and members.

Consistent with the historic functions of the fraternal benefit societies, many fraternal benefit societies provide disaster assistance relief. Major disasters, such as floods or tornados, are generally dealt with on a community scale, so both members and non- members benefit. For example, the Knights of Columbus spent $0.6 million in 1987 and $1.1 million in 1990 on disaster aid, while the Independent Order of Foresters and Woodmen of the World had various clean-up projects after Hurricane Hugo. Similarly, the Lutheran Brotherhood has a matching grant program for local lodges to aid those suffering from natural or medical disasters and the Modern Woodmen of America has a Fraternal Aid Fund to assist members who need financial assistance after floods, earthquakes, tornados, or other hardships or disabilities.

In addition to these community grants, the fraternal benefit societies generally provide funds for individual members suffering health or financial problems. For example, the Mennonite Mutual Aid Association has catastrophic aid for members who have financial hardships because of medical expenses.

Table 12 reports local statistics on various measures of fraternal activities for all fraternal benefit societies in the United States. These amounts make no distinction between fraternal and charitable activities. Except for the decline in the year 1986, which may be due to underreporting, there has been an increase in the number of events, acts of fraternal service, and hours of fraternal service between 1985 and 1989. An example of a fraternal event would be a picnic for members. A business meeting of officers or educational lecture also fits in this category. A fraternal service includes activities such as renovating a public baseball diamond or providing a wheel chair ramp for a member. The hours volunteered in the fraternal service efforts would be included in the number of hours of fraternal service. In 1985, all members averaged about 4 hours per year of volunteer work. In 1989, 46.7 million hours of volunteer work were provided by all the fraternal benefit societies in the United States. Members of the seven societies devoted 11.2 million hours in 1990.

While there is variation as to the types of activities carried out by the local fraternal benefit societies, most of the societies studied have education, scholarship, youth, disaster relief, and community aid programs. These activities are often organized and executed by committee and are in addition to the work done supporting the lodge structure.

The general community aid projects tend to provide both financial and volunteer assistance. These can be at an international, national, or local scale. For example, the Mennonite Central Committee provides funds and volunteers to assist the disadvantaged throughout the world. The Knights of Columbus are a principal sponsor of the International Special Olympics Summer Games and other national and local programs to help those who are mentally retarded. Many of the local charitable community projects include maintenance of parks, providing food and clothing to the needy, or repairing the homes of the elderly.

TYPES OF EXPENDITURES

Table 13 shows the types of expenditures at the local level made by all United States fraternal benefit societies in 1985, 1987, and 1989. These expenditures include both cash grants by the local fraternal benefit society as well as expenses in crying out an activity. Fraternal benefit societies that finance more of their expenditures nationally would have fewer expenditures appearing on Table 13, therefore, Table 13 may provide a distorted picture of society activity for certain categories of expenditures. In 1990, the total fraternal and charitable expenditures by the seven largest societies was $124.6 million, or less than half the amount spent by all fraternal benefit societies in 1989.

Approximately 50 percent of fraternal expenditures are for local activity and benevolent expenses. Religious, educational, and charitable contributions activities, such as are traditionally associated with section 501(c)(3) organizations, make up approximately 15 percent of fraternal expenditures. While specific expenditures may be classified in more than one category by different local lodges, a typical expenditure pattern is evident.

Local activity expenses account for roughly one-third of fraternal and charitable expenses at the local level. These activities tend to be the more traditional club-type activities, such as meetings, socials, or member picnics. These activities generally benefit members and may include activities of juvenile members who participate in youth groups. All of the fraternal benefit societies studied had youth programs. For example, the Knights of Columbus spent about $3.4 million in 1985 on programs for about 18,702 juvenile members (Squires) and some 22,000 scouts. /1/ By 1990 youth activities accounted for $8.2 million of expenses for the Knights of Columbus.

The next largest category is local benevolent expenses which constitute approximately 18 percent of total fraternal and charitable local expenses. These encompass expenditures that may benefit more disadvantaged members or non-members. For example, expenses in constructing a wheel chair ramp for a member or for a local nonprofit organization may be included in this category. Similarly, local expenditures by the Independent Order of Foresters for the Big Brother and Big Sister programs appear in this category.

Membership expenses can include the more traditional costs of sending out a membership newsletter, as well as providing a telephone counseling and referral service for members. For example, the Modern Woodmen of America has a family counseling program for members. These expenses account for roughly 15 percent of local fraternal and charitable expenditures.

Because the fraternal benefit societies may have a religious bond, approximately 7 percent of expenditures are for religious activities. Typically these expenditures take the form of grants to local churches for specific projects and would be treated as charitable contributions. For example, the Aid Association of Lutherans, the Knights of Columbus, the Mennonite Mutual Aid Association, and the Lutheran Brotherhood all have grant or loan programs to help fund building and renovation of churches and support religious education.

Educational expenses primarily encompass loans and scholarships for students and grants to schools. These expenses account for approximately 4 percent of fraternal and charitable local expenditures. While the post-secondary and graduate educational loan programs are usually limited to members and their dependents, the scholarship programs may vary as to whom is eligible. All seven of the fraternal benefit societies provided educational grants or loans to members or dependents. For example, the Modern Woodmen of America provided educational grants to members worth $1.3 million in 1985, and the Aid Association of Lutherans awarded loans and grants to 13,250 non-members in 1985. Fraternal benefit societies may also support educational institutions. For example, the Aid Association of Lutherans gives grants to Lutheran high schools, colleges, seminaries, camps, and churches, while the Knights of Columbus support correspondence schools and evening veterans' education services. Compared to the other five fraternal benefit societies, the Mennonite Mutual Aid Association and Woodmen of the World have less extensive educational aid programs.

Educational activities include working with members as well as the community. For example, the Independent Order of Foresters has provided educational materials, speakers, and matching funds for local groups working to prevent child abuse. Woodmen of the World has offered several programs to promote patriotism, varying from presenting over 2,600 awards for proficiency in history to providing 48,000 flags and patriotic literature to community organizations in 1990.

Charitable contributions from the local organizations also account for 4 percent of fraternal and charitable local expenditures. These expenses tend to be cash contributions to either a local or national charity. In some cases there may be overlap with institutional expenses. These contributions tend to go to charitable institutions, such as St. Jude's Hospital, or one of the national fraternal causes, such as Habitat for Humanity. These expenses include fund-raising activity expenditures but generally benefit non- members.

Approximately 3 percent of fraternal and charitable expenses at the local level are for recreational and health activities. These expenses include rental of facilities for members as well as provision of education courses. For example, the Aid Association of Lutherans supports health fairs for members and their community, and provides materials for evaluation of personal health habits.

SUMMARY

Fraternal benefit societies provide many charitable services; however, much of the combined fraternal and charitable activity appears to be more fraternal in nature, largely benefiting members. A major proportion of the combined expenses is for non-contract benefits to members, as well as support of social activities. Mutual aid is the next major category of expenditure and includes many of the benevolent activities of the societies. Charitable expenditures benefiting non-members (traditional tax-exempt organization activity) appear to be less prevalent than expenditures for the fraternal activities.

FOOTNOTES

/1/ Knights of Columbus Supreme Council Proceedings. 103 Annual Meeting, Washington, DC, August 6-8, 1985, New Haven: Knights of Columbus (1985), p. 28.

END OF FOOTNOTE

CHAPTER 7

SUMMARY AND POLICY OPTIONS

SUMMARY

The purpose of this Congressionally-mandated study is to examine the operations of large, tax-exempt, fraternal benefit societies. This study has also been conducted to assess the tax treatment of fraternal benefit societies.

Fraternal benefit societies receive more generous treatment compared to other tax-exempt organizations since other tax-exempt organizations pay unrelated business income tax (UBIT) on insurance income. Fraternal benefit societies also receive more generous treatment compared to commercial businesses, since section 502 prohibits profits from a commercial enterprise from being exempt from tax even if all of its profits are payable to charitable organizations.

The insurance products sold by fraternal benefit societies are similar to the products so1d by commercial insurers. Specifically, examination of the pricing of policies, the costs of operations, and the offering of new products suggest that the large fraternal benefit societies conduct their insurance operations in a manner similar to commercial measurers. As life insurance products have changed in the commercial markets in the 1980's, so too have the insurance products provided by large fraternal benefit societies. Analysis of the cost of life insurance suggests that the large fraternal benefit societies charge prices that are not significantly less than those charged by comparable large mutual life insurers. In addition, insurance operations by fraternal benefit societies appear to be at least as efficient as those of large mutual insurers. This suggests that fraternal benefit societies do not use their tax exemption to compete unfairly with commercial insurers in terms of price or to operate inefficiently.

The study determines that, except for the Mennonite Mutual Aid Association, the large fraternal benefit societies do not use their tax exemption for cross-subsidization of insurance. As a general rule, insurance policy costs are not lower for those with poor health.

The tax exemption of their insurance operations do appear to allow the fraternal benefit societies to finance fraternal and charitable services and to accumulate surplus, or related profits. The funds for fraternal and charitable services benefit charitable organizations and support members' charitable activities, as well as provide members with non-contract benefits and membership services. On occasion, in comparison to large mutual life insurers, the large fraternal benefit societies have had significantly higher accumulated surplus growth rates, leading to higher relative levels of surplus. These additional assets accumulated as surplus benefit policyholders of fraternal benefit societies by reducing the risk of insolvency.

POLICY OPTIONS FOR TAXATION OF FRATERNAL BENEFIT SOCIETIES

Fraternal benefit societies are exempt from the corporate income tax under section 501(c)(8) of the Internal Revenue Code. Below are options for the income tax treatment of fraternal benefit societies.

NO CHANGE IN CURRENT TAX TREATMENT

A major justification for retaining the present tax treatment of fraternal benefit societies is that the fraternal benefit societies provide charitable and fraternal services which benefit society. Without tax-exempt insurance income, fewer charitable goods and services may be provided. Although fraternal benefit societies are accumulating excess surplus and are using their preferential treatment of insurance income to pay for their fraternal and charitable activities, the economic distortions resulting from this special treatment are relatively minor when compared to other distortions in the economy. Combined with the increase in administrative and compliance costs, the net gains to society from taxing the insurance income of these organizations may be small.

MODIFY TAX TREATMENT OF FRATERNAL BENEFIT SOCIETIES

During the past 30 years, the tax exemption for insurance income of fraternal benefit societies appears to have financed accumulation of surplus and funded fraternal as well as charitable activities. Unlike other tax-exempt organizations, fraternal benefit societies can use untaxed insurance income to finance charitable and fraternal services. Similarly, compared to commercial insurers, fraternal benefit societies receive a higher subsidy for charitable activities. Therefore, Congress may consider alternatives for taxation of insurance income of fraternal benefit societies.

Fraternal benefit societies that continue substantial participation in the insurance business could be taxed as insurance companies. If Congress decided that taxation was appropriate, it could limit this modification to apply only to large fraternal benefit societies that do not cross-subsidize the insurance they sell. Fraternal benefit societies without substantial participation in the insurance business could be subject to section 501(m) and pay UBIT on insurance income or they may choose to convert to section 501(c)(10) organizations, as these organizations also have a lodge structure, perform fraternal and charitable activities, but are prohibited from selling insurance. Fraternal benefit societies that converted to section 501(c)(10) organizations would no longer receive preferential treatment with respect to charitable and fraternal expenses as compared to other tax-exempt organizations, since they would no longer be allowed to use untaxed insurance income to finance fraternal and charitable activities.

Congress could give special recognition to the social benefits resulting from the charitable activities of the fraternal benefit societies, and provide for these entities' special treatment for charitable expenses, such as raising the deduction limitation (currently 10 percent of taxable income under section 170). Alternatively, fraternal benefit societies that are unable to separate fraternal and charitable expenses could be permitted to deduct a percentage of the combined charitable and fraternal expenses to which the higher limitation would apply. Although a special charitable deduction limit would require special tax provisions applicable to a small group of taxpayers and does not create a level playing field relative to taxable insurance companies, the deduction recognizes the historical importance of charitable activities by these organizations as compared to other companies, and would target the tax benefit to goods and services that provide benefits to society. Moreover, a safe harbor provision that does not require these entities to separate their fraternal and charitable activities would lower administrative costs.

[Omitted tables 1-13 are not suitable for online presentation and are

[Omitted appendixes 1-5 are not suitable for online presentation and