The Art & Science of Life Insurance Distribution [1 ed.] 1625422113, 9781625422118

Table of contents :
COVER
TABLE OF CONTENTS
FOREWORD
PREFACE
Chapter 1 INTRODUCTION TO DISTRIBUTION
Chapter 2 THE HISTORY OF LIFE INSURANCE DISTRIBUTION
Chapter 3 THE PRIMARY LIFE INSURANCE DISTRIBUTION CHANNELS
Chapter 4 THE FUNCTIONS OF DISTRIBUTION IN THE LIFE INSURANCE BUSINESS
Chapter 5 THE MANAGEMENT OF DISTRIBUTION
Chapter 6 DISTRIBUTION COMPENSATION
Chapter 7 THE ECONOMICS OF LIFE INSURANCE DISTRIBUTION
Chapter 8 USING A PROJECTION MODEL FOR COMPENSATION DESIGN
Chapter 9 WHICH DISTRIBUTION CHANNEL IS BEST?
Chapter 10 THE FUTURE OF LIFE INSURANCE DISTRIBUTION
Chapter 11 CONCLUSION
BIBLIOGRAPHY
INDEX
BOOK EDGE
BACKCOVER

Citation preview

Douglas J. Bennett, FSA Walter H. Zultowski, Ph.D. I I

Douglas J. Bennett, FSA Walter H. Zultowski, Ph.D. ACTEX Publications, Inc. Winsted, CT

Copyright © 2014 by ACTEX Publications, Inc. All rights reserved. No portion of this book May be reproduced in any form or by any means Without the prior written permis�ion of the Copyright owner. Requests for permission should be addressed to ACTEX Publications PO Box 974 Winsted, CT 06098 Manufactured in the United States of America 10 9 8 7 6 5 4 3 2 1 Cover design by Jeff Melaragno Library of Congress Cataloging-in-Publication Data Bennett, Douglas J. The art and science of life insurance distribution I Douglas J. Bennett, FSA, Walter H. Zultowski, PhD. pages cm Includes bibliographical references and index. ISBN 978-1-62542-211-8 (pbk. : alk. paper) 1. Life insurance-­ Marketing. 2. Selling--Life insurance. 3. Insurance companies. I. Zultowski, Walter H. II. Title. HG8876.B355 2014 368.320068'8--dc23 2014011141

ISBN: 978-62542-211-8

TABLE OF CONTENTS

Foreword .................................................................................................. iii Preface ...................................................................................................... v Chapter 1 Introduction to Distribution ................................................... 1 Chapter 2

The History of Life Insurance Distribution ........................... 9

Chapter 3 The Primary Life Insurance Distribution Channels............. 23 Chapter 4 The Functions of Distribution in the Life Insurance Business ...................................................... 37 Chapter 5 The Management of Distribution ........................................ 45 Chapter 6

Distribution Compensation ................................................. 51

Chapter 7

The Economics of Life Insurance Distribution ................... 61

Chapter 8 Using a Projection Model for Compensation Design .......... 69 Chapter 9 Which Distribution Channel is Best? .................................. 89 Chapter 10 The Future of Life Insurance Distribution .......................... 97 Chapter 11 Conclusion........................................................................... 99 Bibliography ......................................................................................... 10 I Index ..................................................................................................... 103

FOREWORD Ask what has been the greatest change to come to the life insurance business in recent decades, and the answer is arguably how companies get product to the end consumer - i.e., distribution. It wasn't long ago that the vast majority of individually-purchased life insurance was sold by agents working primarily for one company. This changed as various forms of independent agents and financial advisors began selling life insurance over the years. More recently, the evolution of on line technology has opened up new ways of selling life insurance, either directly or in combination with other distribution methods such as call centers. At the same time, the discipline of marketing - as practiced in consumer package goods - also gained greater traction in the business. The result has been a breakdown of the "one-size-fits-all" mentality when it comes to life insurance distribution. This has been replaced by the viewpoint that, to be most effective, distribution channels are best matched with various market segments and life insurance products. Today, multiple distribution approaches, formerly seen as incompatible, are being operated simultaneously in companies, albeit not without challenges. One of our objectives in writing this book was to discuss how these challenges can be successfully managed. It is out of this viewpoint that one comes to appreciate that successful distribution is both an art and a science. The Society of Actuaries (SOA) has for years been involved in helping to educate its students and members about marketing and distribution in the life insurance business. More recently, the Marketing and Distribution Section of the SOA was looking to update its educational materials, and to focus more on distribution given all the changes that have occurred in this company function in recent years. The Section supported the effort when ACTEX decided to publish a book on this topic. The authors are grateful to the leadership of the SOA's Marketing and Distribution Section for its enthusiasm for the importance and relevancy 111

of this text to the actuarial community, and for the support necessary in making it a reality. We also owe a special "thank you" to Gail Hall, FSA, MAAA, President of ACTEX, for her advice, guidance, and patience as we worked on the text. We are also appreciative of the editorial work provided by Marilyn Baleshiski, and of the cover design provided by Jeff Melaragno, both also from ACTEX. We would also like to thank our reviewers who provided invaluable support and constructive criticism at various stages of this work. They are: Jeremy J. Brown, FSA, EA, MAAA Keith Dall, FSA, MAAA Janet Deskins, FSA, MAAA R. Dale Hall, FSA, MAAA, CERA Neville S. Henderson, FSA, FCIA, MAAA Jay M. Jaffe, FSA, MAAA Anne M. Katcher, FSA, MAAA Walter S. Rugland, FSA, MAAA Marcia S, Sander, FSA, MAAA Finally, for their non-actuarial support provided throughout our lives, we dedicate this work to our respective spouses, Sue and Linda. Doug Bennett and Walt Zultowski

IV

PREFACE Life insurance is unique. People don't usually wake up one day saying; "I think I'll buy some life insurance today." Since it is a low demand product that is more often "sold" rather than "bought," it is not surprising that distribution is a major function in life insurance companies. As such, it is also a major topic of debate and discussion both within and between companies. Over the years, for example, arguments have been waged as to which distribution channel is the most effective and efficient. Even within companies, arguments have raged over the "cost" of distribution, whether distribution is more appropriately viewed as an expense to be managed or an investment to generate business, and whether operating multiple distribution channels results in unavoidable conflict. The fact that there are two clear "sides" in these debates, (i.e., home office vs. field sales representatives), often makes these deliberations even more contentious. One common misperception in the business is that distribution is an exact science that can be managed according to established formulas such as may exist in other company functions. In practice, distribution is more of an art than a science, and needs to be managed within the context of the needs of the consumer being served, and the specific product being sold to meet those needs. To have an appreciation for the distribution function in a life insurance company, it is important to understand the role of distribution in business in general, as well as how various distribution systems have evolved in the life insurance business to what they are today. Additionally, it is important to understand the functions and roles served by distribution, and how distribution compensation operates in the industry. It is only with this overall picture of life insurance distribution that one can make informed decisions regarding the selection and management of life insurance distribution channels. A guiding principle throughout this text is that there is no one best distribution channel in the life insurance business, nor is one distribution channel necessarily better than another. Rather, we would argue that certain distribution channels are better than other channels for specific situations, typically defined by market needs and products. The objective of this text is to help the reader better understand how best to align markets, products, and distribution channels. v

The good news is that the processes of distribution channel selection and management are not difficult to understand. This is the case once one understands and appreciates how distribution evolved to its current state in the industry today, the functions and roles served by distribution, how to use compensation to drive the behaviors desired in a distribution channel, and how distribution serves as one leg of the industry's proverbial three-legged stool comprising market needs, products to serve those needs, and distribution.

Vl

CHAPTER ONE INTRODUCTION TO DISTRIBUTION The term distribution has multiple meanings in the business world. In the investment world, for example, it is used to refer to the process of paying out funds, as in a "dividend distribution." In the computer business, it is used to mean the circulation of programs, as in the "distribution of software." In the marketing function, however, "distribution" specifically refers to the process or paths involved in moving a product from a manufacturer to an end consumer. It is this meaning of distribution, and more specifically, distribution in the life insurance business, that will be the focus of this text. THE BASICS OF DISTRIBUTION

There is no question that distribution plays a key role in efficient and effective marketing. Marketers have traditionally talked about the "4P's" of the marketing function as being place, product, price, and promotion. The Four P's of Marketing PLACE (i.e., Distribution) Manufacturer ___.

PRODUCT

___. End Consumer

PRICE PROMOTION These are the four key elements that comprise the marketing function or "marketing mix". "Place" in this case is distribution. Interestingly, though, distribution is probably the element of the "marketing mix" that gets the least attention, at least in comparison to product, price, and promotion. This is despite the fact that it garners a very large share of the 1

2 -{>- CHAPTER ONE

overall cost of marketing, and as we'll see, can also be the most complicated of the "4P's." Perhaps this is because the distribution function is often delegated to, or aligned with, the sales function. On the one hand, this could be viewed as recognition of the importance of distribution, such that it deserves its separate place right alongside of marketing. On the other hand, as will be discussed later, doing so often causes more problems than it solves. Distribution Terminology Systems involved in getting a product from a manufacturer to an end consumer are typically referred to as channels, while the individuals or organizations involved in distribution are typically termed distributors, sales people, or intermediaries. It is important, however, to recognize that not all distribution channels involve human interaction. We can think of numerous products and services that are marketed directly to consumers. This is also true in the life insurance business, in which direct marketing through the mail, for example, has been done for years. It is only within more recent years, with the advent of online technology, that this form of distribution has garnered more attention in life insurance. Levels of Distribution In earlier times, there was no need for distribution. Artisans practiced their crafts in the backs of their shops, and consumers purchased their products in the fronts of their shops. It is likely that the term "store fronts" was conceived in this time period. This environment still exists today in the form of factory stores, although they are most likely not a manufacturer's primary distribution method. Such systems would be termed "two-level" distribution systems, in that they involve just the manufacturer and end consumer. One can also envision, however, how the artisans of years ago needed to devise a system to bring their products to consumers who were unable to travel to their shop, and/or needed an additional person to peddle their wares while they worked to create more goods in the backs of their shops. Thus, the concept of distribution was born. Today, simple two-level systems can be replaced by multiple-level systems, as additional inter­ mediaries are added to the network of delivering a product from the manufacturer to the end consumer. Four and even five level systems are not uncommon today. Each intermediary takes the product to the next step in the chain and in return is compensated for that activity. One will often see

INTRODUCTION TO DISTRIBUTION {,- 3

simple systems referred to as short channel distribution systems, while multiple-level systems of intermediaries are referred to as long channel distribution systems. Two Level

Levels of Distribution

(Short Channel Distribution) Manufacturer Multi-Level

Manufacturer

End Consumer --- Retailer

End Consumer

(Long Channel Distribution) Manufacturer ...Wholesaler ... Retailer ... End Consumer

Retail vs. Wholesale Distribution A key concept in understanding distribution is that of retail distribution vs. wholesale distribution. Simply put, any individual or organization that has direct contact with an end consumer is considered to be involved in retail distribution, or a retailer for short. A retailer can take several forms ranging from an actual store to an individual salesperson, or in more recent times a direct mail or online system. Any individual or organization that only brings a product to another intermediary is considered to be involved in wholesale distribution, or a wholesaler for short. One will also see these two types of distribution referred to as "B to C" (i.e., business to consumer), or "B to B" (i.e., business to business). It is also important to recognize that these two forms of distribution are not mutually exclusive when it comes to their usage. Manufacturers may use retail distribution for certain products, and/or in specific markets or territories, and act as wholesalers for other products, and/or in other specific markets or territories. Others may use both systems for the same products, and/or in the same markets or territories. The challenge that this presents is one of channel conflict. The management of channel conflict is a subject in itself, especially when it comes to the life insurance industry, and will be discussed later in this text. The Functions Provided by Distribution The discussion up to this point has suggested that the primary purpose of distribution is to provide the end consumer with access to product. This

4 � CHAPTER ONE

is a large part of distribution, and with many commodity products is probably the only purpose of distribution. Depending on the nature of the product, however, distribution serves purposes that go beyond providing mere access to product. With a very technical or difficult to understand product, for example, an additional role of the intermediary might be to provide education or technical assistance in the selection and/or use of the product. This assistance might be provided to another intermediary, as in a B to B environment, or directly to the end consumer in a B to C environment. Finally, an additional function provided by distribution is that of purchase corifirmation. For products which consumers might equivocate in purchasing, an intermediary can serve to get that consumer "over the hump" in their purchase decision, or reaffirm the appropriateness of the purchase for consumers experiencing "buyer's remorse." What is the Right Distribution Channel(s) to Use? This, of course, is not an easy question to answer. The selection of one or more distribution channels is based on numerous factors such as: •

the nature of the product (e.g., is it simple or complex?),

•

the characteristics of the market (e.g., is it well educated?),

•

the simple availability of appropriate intermediaries,

•

the characteristics of the intermediaries available (e.g., the ability to understand and communicate product features),

•

resources available for distribution.

Thus the challenge when it comes to distribution management is selecting the channel that is right for the specific situation. Finding the most efficient and effective match or alignment of product, market, and distribution is a major theme of this text. LIFE INSURANCE IS DIFFERENT While the previous discussion of marketing and distribution is intentionally academic in nature, and applies more readily to the marketing and distribution of consumer packaged goods, it needs to be recognized that life insurance is different. In fact, it is a challenge to

INTRODUCTION TO DISTRIBUTION - CHAPTER Two

Assurance Society in 1859 [8, Table l ]. Once the country returned to normalcy following the Civil War, the life insurance business witnessed several significant changes. It was during this period, for example, that the general agency system emerged and was refined. This was inevitable as the growth in the mutual companies' business, and its spread west, necessitated the development of a field management organization to oversee an increasingly large and increasingly dispersed field force. With increasing competition came more aggressive marketing [7, p. 13 8], increasing sales commissions, and sales practices that heretofore were considered inappropriate for this traditionally well-mannered industry. For the first time, agents could be heard disparaging their competition, and replacement of other companies' business also began to appear. Companies also responded to the increased competition in the business with creative new products, some of which were subsequently ruled illegal. The best example of this was the "deferred dividend policy" created by The Equitable in 1867, also known as a tontine [9]. A portion of the premiums for this policy went toward the policy itself, while another portion was directed into an investment fund with a set maturity date that benefited a restricted group of policyholders. Agents aggressively marketed this new product, often with inflated estimates of future returns, and shortly after the turn of the century, it is estimated that two-thirds of the life insurance policies in force in the country were tontines [9]. This, along with other sales practices lead to the well-known Armstrong Investigation in 1905, which established regulations on sales practices and controls on sales expenses, many of which still exist over a century later. Other companies, during this time of heavy compeht10n, chose to compete by avoiding competition head-on, and instead looked to do business in under-served niche markets. It was during this time, for example, that fraternal benefit societies began to underwrite and distribute life insurance. [13, p. 93] These organizations were voluntary service and social benefit associations organized around a shared characteristic such as ethnicity, religion, or profession. While these organizations have operated throughout the history of the country, it was during this time period that they grew significantly, and expanded their member services to include life insurance, which was typically sold to working-class Americans. The life insurance operations of these societies were most often organized as mutual companies, and by the turn of the century there were over 600 [8] of these in existence. While they are no

THE HISTORY OF LIFE lNSURACE DISTRIBUTION

�

15

longer the competitors of the main line life insurance companies that they once were, many of them have survived to the current day. One of the industry's best examples of niche marketing, however, was also launched during this time period. Companies such as John Hancock (1862), Metropolitan Life (1868), and Prudential Life Insurance Company (1875) began to market industrial insurance [8, Table 1]. This concept was imported from England, and involved selling low face amount policies to lower income families. Agents collected premiums (often ranging from $.05 to $.50 [8]) on a door-to-door basis every week coinciding with the day these workers were paid. Medical examinations were often not required for these small amounts of coverage, and policies were also often written on the lives of children in these families. This life insurance business model, known as the debit or home service business, still exists today, mostly in rural and inner city markets. It is, however, a fraction of what it once was, and home collection of premiums is not commonly found. Yet it is significant to note that many of the giant companies in the business today were built on the nickels and dimes spent by lower income Americans out of love for their families.

1900-1949 By the tum of the century, many of the life insurance distribution models that characterize the industry today were firmly in place. Yet there continued to be changes in the marketplace that resulted in further evolution of the business and its distribution methods. One of the biggest changes in the country was the mass distribution of the automobile, which obviously created the need for automobile insurance. This also spawned a variant of the agency system that remains a major force in life insurance distribution today; i.e., the multi-line exclusive agent. Farmers and other rural residents, complained that they should not be charged the same rates for auto insurance as drivers in the much more congested cities. As a result, Farm Bureaus started offering general insurance, and over time added life insurance to their portfolios. The agents of these companies to this day sell a wide portfolio of products, but only for their company (i.e., they are exclusive to their companies), and have expanded way beyond their rural, Farm Bureau heritages. Another innovation came in 1911, when Equitable Life Assurance Society wrote the first group life policy covering 125 employees of the Pantasote

16 -¢> CHAPTER Two Leather Company [8]. Within eight short years, a total of 29 companies were issuing group life insurance [8]. Such policies typically provided coverage to workers without applications or medical examinations as a part of their employment contract, with premiums fully paid by employers. As this business developed, it took a separate direction than the individual life insurance business, and companies in this line of business typically developed distinct group insurance departments and specialists who marketed these products to corporations rather than individuals. In fact, in many companies, group insurance was viewed as a competitor to individually purchased life insurance. Group life should not be confused with individual life insurance marketed at the worksite, which appeared on the scene later in the century. In worksite marketing, individual employees are solicited at their place of work for individual life insurance coverage, often sold as coverage to "supplement" their group policy. Such policies are, however, owned by the individual, not the employer, with the premiums being paid by the employee. Such payments, though, are usually made through payroll deduction. Finally, during this time period, several companies found and exploited niches in the marketplace for specific product offerings and services. Northwestern Mutual, for example, is often credited with developing applications for life insurance in the business market for such purposes as insuring key employees or arranging for the orderly transfer of a business in the event of an owner's death [12]. Prior to the advent of Social Security in 1935, other companies actively marketed life insurance as a way of ensuring a comfortable retirement. Phoenix Mutual Life, for example, was known for decades during this time period as the "Retirement Income Company."

1950-1969 These two decades were a relatively quiet period for the life insurance industry. Both the business and the country were settling down after the turmoil of the Second World War, and there was little new when it came to the individual life insurance product and the way it was sold. The industry basically sold whole life, term, and endowments through career life insurance agents, albeit endowments were significantly declining in popularity by this period of time. Given the dramatic changes that were about to emerge in the decade of the seventies, it is important to take note of the interest rate environment

THE HISTORY OF LIFE INSURACE DISTRIBUTION

'¢--

17

of the decade of the forties and most of this period. Not only were short­ term interest rates historically low for most of the 1940s, but also the relative spread between long and shorter interest rates was historically high. This was ideal for selling a product such as whole life that provided consumers at least partial access to long-term interest rates. Once the interest spreads narrowed towards the end of the 1960s, the advantage of whole life's static long-term returns began to disappear.

1970 -1999 This period in the industry's history is best described as the "product revolution." Perhaps the tipping point to this revolution was in 1979 when the Federal Trade Commission (FTC) published a report that strongly criticized whole life insurance. Among other things, it made the strong suggestion that consumers would be better off if they bought term, and invested the difference. This appeared to launch greater interest and competition for term life insurance, but it was coincidently a time when other forms of permanent life insurance - universal life, variable life, and variable-universal life - also arrived on the scene. This product revolution was not independent of changes occurring in distribution, and served to fuel some of these changes. Specifically, the 1980s saw more and more companies exiting the agency building business, as many viewed it as a more expensive distribution channel. Instead of recruiting and training new agents, their strategy turned to selling through independent agents (many of whom had "outgrown" their agency management structures) or agents of other companies. In the life insurance industry, this strategy is known as using a brokerage distribution channel. At this point, it is important not to confuse the life insurance business's use of the term brokerage with how this term is used to describe investment advisors, property and casualty agents, or agents who sell employee benefits. In the context of life insurance it simply means selling products through independent agents or career agents of other companies. The company that has chosen to be a "brokerage company" has simply decided to compete as a manufacturer and wholesale distributor of product, and not as a retailer to the consumer. Not surprisingly, it is mostly these companies who argue that their customer is the producer and not the end buyer. Brokerage business was nothing new to the industry, but until now had been mainly reserved for specialty products or substandard business that

18 -¢>- CHAPTER Two

a company did not want to write. Many of the main-line mutual companies fought the trend toward cheaper term and/or the proliferation of new products sweeping the industry, allowing brokerage companies to gain market share among agents whom the career companies had spent money to recruit and train. This became a major strain on the successful financial management of the agency system. Whether the growth in the brokerage distribution channel was a result of companies exiting the agency building business because of expenses, or of career agents outgrowing the need for support provided by an exclusive agency, the product revolution created an ideal environment for such a transformation. These new products gave agents an opportunity to introduce customers, both new and current, to new and potentially valuable features and benefits, especially when compared to traditional non-par whole life or yearly renewable term. With current clients, in many situations agents could rationalize replacing in-force policies with ones better suited to the current economic environment. To the extent these clients had built up equity in their current policies, products were designed such that this equity could be rolled right into the new policy. For new customers, the transparency of the new products, especially the concept of credited interest made the products easier to sell in the eyes of many agents. It became an investment sale rather than simply death protection. For many career agents, with current clients as a primary market, this meant they had a product, not necessarily even manufactured by their current company, which was relatively easy to sell to those clients. The relative ease of the sale reduced the need for them to rely on the support and services provide by their company to effectively convert their clients to these new products. For new agents, if they could identify customers who already owned some life insurance, they could position themselves to do essentially the same thing without relying on the extensive training and support provided by career companies. At the same time, companies that did not use career agents recognized that there was a growing pool of agents that would sell their new products without the expensive financing, training and sales support. They could use the money they did not need to spend supporting a career agency force to increase agent commissions or their products' competitiveness. This left career companies with competitors that oftentimes had more attractive

THE HISTORY OF LIFE INSURACE DISTRIBUTION {>- 19

agent compensation packages or products, while they were spending money on expensive agent support systems not always valued by their agents. For many but a handful of companies using career agents, the only solution was to abandon that form of distribution. The last decade of this century also witnessed experimentation with a wide variety of new ways to distribute life insurance to the public. This experimentation was enabled during this time by the slow erosion of the 1933 Glass-Steagall Act that had erected barriers between various financial institutions, culminating in the 1999 signing of the Gramm­ Leach-Bliley Act. This Act allowed financial institutions to participate in each other's lines of business. As a result, this period saw attempts to market life insurance through kiosks in shopping malls, department stores, grocery stores, commercial banks, savings & loans, credit unions, and stockbrokers. Of these, the most successful were banks and stockbrokers, but even here these success stories were for specific products and/or in specific sales situations ( e.g., estate planning) that were mediated by a life company wholesaler. Direct response marketing of life insurance, while never achieving a large industry market share, also grew during this time period. The primary success stories here, were in very targeted marketing approaches or in affinity group marketing. Examples would include USAA, serving the insurance needs of military officers; Gerber Life, selling small juvenile policies; and Amica Life, serving the life insurance needs of its parent company's auto and homeowner clients. Since this time, each of these companies has widened its marketing efforts, and this entire business model has shifted away from purely direct mail, to direct mail combined with call centers. The direct response business continues to evolve today as companies experiment with online marketing through the internet and the use of social media. It should be noted that the 1990s witnessed maJor scandals and investigations involving inappropriate and aggressive sales practices by agents. In addition to costing many companies hundreds of millions of dollars in lawsuits and penalties, these also resulted in the emergence of large and powerful compliance departments in life companies, and increased scrutiny by company broker-dealers in agents' sales of variable products.

20

-- CHAPTER Two

2000-Today This brings us up to the present, and the distribution landscape continues to evolve. The tum of the century brought with it significant consolidation of companies through merger and acquisition activity, some of which was driven by overseas companies. (It is interesting to note, however, that as this text is being written, acquisition activity by overseas companies has slowed, and several European parent companies are actually divesting themselves of their North American subsidiary companies. This is in part driven by greater financial pressure on these parent organizations resulting from new, and more stringent, solvency reserve requirements on the European continent.) During the first decade of this century, the industry also saw a wave of demutualizations as companies sought new ways to raise capital. Today, the number of major life insurance companies operating as mutuals can be counted on two hands. In a sense, the industry has completed a 150year journey from the mid-l 800s, when the number of stock companies could be counted on two hands. This is significant for the distribution of life insurance in the fact that few stock companies are actively involved in the recruiting, hiring, and training of new agents. Stock companies that maintain retail agency systems also appear to be putting greater resources in the pure manufacturing and wholesaling sides of their business. This has caused more than one industry executive to observe that the industry has "too much manufacturing capacity chasing too little distribution." Consequently, in many companies today, "distribution is king." With fewer companies recruiting and hiring new life insurance agents, there is a lack of new blood to fuel the independent agent system and a succession challenge in the independent agency world. This is resulting in a wave of mergers, acquisitions, and restructuring of independent agencies. Many of these agencies are building their own support structures and services for their producers (e.g., marketing), as companies are trending toward pure product manufacturing only and discontinuing such services. Recent years have also seen agents banding together in various types of producer groups, often in order to pool production for the purposes of achieving higher compensation bands. Finally, although not new to this decade, recent years have also seen growth in so-called Independent Marketing Organizations (i.e., IMOs). These are typically large independent agency-like organizations

THE HISTORY OF LIFE INSURACE DISTRIBUTION � 21

that provide many of the sales and support services traditionally provided by companies. With many companies exiting the agency building business in the final decades of the century, IMOs may be viewed as filling this void.

22 {>- CHAPTER Two

CHAPTER THREE THE PRIMARY LIFE INSURANCE DISTRIBUTION CHANNELS Today there are ten primary channels utilized in the distribution of individual life insurance. We will define each according to the degree to which they involve human intervention and the complexity of the consumer need that they serve. DIRECT MAIL Some will find it hard to believe that the use of direct mail as a marketing methodology has existed in the life insurance business for over a century. Direct mail advertising was pioneered by Phoenix Mutual life in 1912. Direct mail marketing, without any human intervention in the sales process, has been consistently utilized in the industry over this time, but has not realized more than two to five percent of annualized premium in recent years. Pure direct mail marketing is not as popular as it once was due to the increasing challenge of breaking through an overly saturated method of reaching the consumer. It has also been found that direct mail is more effective when combined with some human intervention, typically licensed sales representatives in a call center. While direct mail marketing continues in the industry today, it is usually employed only for guaranteed acceptance products (i.e., small face amount policies sold without any underwriting, and in which there is only a return of premium plus some percentage if the insured dies in the first two years of the policy). These are usually sold on the lives of juveniles or seniors for final expense purposes. INTERNET DIRECT Not surprisingly, a significant amount of direct marketers' attention in recent years has been focused on trying to make use of the internet for life insurance sales. Aside from using the internet as a prospecting tool to 23

24 {>- CHAPTER THREE

direct prospective buyers to agents (which for this analysis would not be considered a fonn of direct response marketing), there have been two primary ways in which companies have attempted to harness the internet for the purposes of marketing life insurance. The first has been through the use of individual company websites or microsites established for selling to a specific target market. These marketing approaches typically are used for term life insurance and involve simplified issue (defined as the use of a small number of medical questions coupled with background checks of such things as motor vehicle records and/or prescription data bases, but without medical exams). For most companies, though, technology is utilized up to the point of policy issue, where the buyer is mailed or delivered his or her policy in the traditional fashion. A few companies have experimented with the automation of the sales process all the way to the point of policy delivery, such that the actual policy can be downloaded to the buyer's computer or printed out to a flash drive. The second, and more popular, use of the internet in the sale of life insurance, however, has been through the development of so-called aggregators. In its simplest form, these are buying sites that have been developed by non-insurance company commercial firms that allow a potential buyer to "shop" for life insurance among various companies that sell through a given aggregator. These aggregator sites deal nearly exclusively in tenn life insurance, and compete primarily on the basis of price. As with company-specific websites, they also generally involve simplified issue underwriting, but also allow for the purchase of policies in larger face amounts than is typically the case in direct mail marketing. The next generation of internet direct marketing, which is still evolving as of the writing of this text, involves the use of social media in the selling of life insurance. CALL CENTER-MEDIATED DIRECT RESPONSE Industry research continues to show that, regardless of how simple the policy, many consumers view life insurance as a complicated product and desire personal assistance during the purchase process. This, of course, is a problem for pure direct response marketing methodologies. Thus, the advent of call centers and call center technology in the general business world provided a logical addition to direct response marketing

THE PRIMARY LIFE INSURANCE DISTRIBUTION CHANNELS � 25

methods. Today it is more common to see direct solicitation of life insurance through a variety of mediums, e.g., direct mail, television, or the internet, with prospects then being directed to a toll free telephone number for completion of the sale. Such call center-mediated direct response marketing methods are often also combined with affinity group or association marketing methods, whereby solicitations are made to prospects with something in common such as a college alumni or professional association. Perhaps one of the most successful of these approaches has been New York Life's agreement with AARP to solicit its senior members for life insurance. This arrangement by itself has propelled New York Life to becoming one of the top companies in the industry in direct response marketing. HOME SERVICE Beginning with the discussion of the home service distribution, we move into channels of distribution for individual life insurance that involve personal, face-to-face interaction between a prospective client and a sales intermediary. Home service, the current terminology for what had traditionally been called industrial or debit life insurance, is one of the oldest forms of individual life insurance distribution. As discussed in Chapter 2, this appeared on the U.S. life insurance industry scene in the second half of the 1800s, and was the bedrock upon which some of the giant companies in the industry were built. Agents in this system were assigned a territory or book of business (i.e., their "debit") in which they came to know the families in their area, sold them life insurance, collected premiums on a weekly basis, and delivered death claims. Considering the fact that industrial workers in those days were paid in cash on a weekly basis and did not have checkbooks or access to automatic bank drafts or payroll deduction, the debit system was a brilliant marketing approach. Agents were in their assigned households on a regular basis, and in many cases were considered "part of the family." Not only were they on the scene when the breadwinner came home with his or her weekly salary to collect policy premiums before they could be spent on other things, but they were also eyewitnesses to family life events, many of which spelled the need for additional sales on the lives of adults or children in the household.

26 � CHAPTER THREE

Another aspect of the old home service business that was also brilliant from a marketing point of view was the way in which its agents were compensated. The system did not pay commissions on individual sales, but rather on the growth of the agent's entire block of business. This provided agents with motivation to work to retain policies that had already been written, and to replace or reinstate policies that clients may have discontinued. Elements of this approach can still be seen today in the property and casualty insurance business, and in various unsuccessful attempts to institute level commission systems. For the most part, though, this approach to compensation in the Home Service distribution channel went by the wayside, as these companies moved away from the collection of premiums at the home of the client. It is easy to understand how social and labor force changes caused this distribution system to fall out of favor over the course of the century. In the early 1980s, this system also came under criticism for the cost of its policies and the sales practices of home service agents, particularly involving the overselling of policies on the lives of juveniles in low income households. Moreover, home service companies found themselves challenged by issues such as the increasing cost of this distribution system and the safety of their agents, more than a few of whom were robbed or even murdered while collecting premiums in their territories. As a result, the majority of companies utilizing this distribution channel have elimi­ nated the collection of weekly premiums by agents, and have switched to "monthly debit ordinary" life insurance, which more resembles life insurance typically sold in the business today. Home service agents continue to sell in lower income segments of the population, albeit at a significantly reduced number than in the past. Today, this distribution channel is also primarily found operating in inner city markets and in rural areas of the South. Additionally, many of the companies using this channel have shifted to selling final expense life insurance to their lower income clients. These are smaller face amount life insurance policies meant to cover burial and other expenses left behind upon the death of the insured. WORKSITE Not to be confused with employer-sponsored group life insurance, worksite distribution involves the selling of individual life insurance at a place of employment. Also called voluntary benefits or supplementary benefits, these are typically small face amount policies for which

THE PRJMARY LIFE INSURANCE DISTRIBUTION CHANNELS � 27

employees can sign up, and are usually paid for via payroll deduction. In worksite distribution, there is an "enrollment period" during which employees are provided with written solicitation material and/or an informational presentation, and then given the opportunity to meet with an "enroller" during company time. The meeting with the enroller generally lasts for fifteen or twenty minutes at most, and does not involve hard sales pitches. Potential prospects often have already decided to purchase prior to their enrollment meeting, and the role of the enroller is to answer questions and complete the sale. The term "supplementary benefits" reflects the fact that the policies purchased by employees through this form of distribution are often products that are in addition to coverage that they have under other policies, both group and individual, and the buyers of these products tend to be middle income consumers. Industry research has shown that consumers are increasingly receptive to worksite distribution as a way of buying life insurance. This is likely due in large part both to the ease of the purchase process and the convenience of payroll deduction, but also due to the perception that their employer has vetted the quality of the life insurance company prior to opening their doors to them. There are a limited number of companies that are the leaders in this form of distribution, as it tends to be a specialty business. In addition to life insurance capabilities, a worksite distribution company must also be an expert in administrative and payroll systems. The last thing an employer needs is for a life insurance payroll program to cause problems with its regular payroll systems. INDEPENDENT MARKETING ORGANIZATIONS (IMOS) IMOs are independent agency-like organizations that recruit and train new agents, as well as provide many of the support services that agents have traditionally received from their companies or agencies (e.g., sales lead generation). In their early days, these organizations tended to specialize in a single product. More recently, these organizations have expanded their offerings, but still most often focus on a specific set of products (e.g., final expense life insurance, indexed annuities, or Medicare supplement) sold in a specific market (e.g., the senior market). Their focus is usually in the lower to middle income market. Agents in this system are provided a constant flow of leads by the IMO; sales

28 � CHAPTER THREE

presentations are highly structured, and geared to high levels of productivity. While the individual IMO may have access to products from multiple companies (often created for them by companies as proprietary products), agents only sell the products of that IMO. With many companies having exited the building of career agencies, or cutting back on support services in recent years, IMOs can be viewed as recreating many of the traditional functions of the career agent company at the local level. The big difference is that the IMO system is built for high productivity, and no business is brokered outside of the IMO. MULTIPLE-LINE EXCLUSIVE AGENT {MLEA) The use of the term multiple-line in this distribution system often causes this channel to be confused with that of the independent property and casualty agent channel, or the so called Big I agent distribution system. Yes, both sell property and casualty insurance, but that is where the similarity ends. Big I agents are truly independent and sell all types of property and casualty coverage for multiple companies. Many life insurance companies over the years have attempted to sell life insurance through Big I agents with little to no success. The challenge has been that many independent property and casualty agents are very successful handling these lines of business, and have no motivation to sell and service life insurance. It should also be recognized that the property and casualty sales process is very different from that of the life insurance sales process. For one thing, ownership of many property and casualty products is mandated. The property and casualty sale is basically a "one and done" process, whereas the life insurance sales process often requires the producer to shepherd larger cases through underwriting in the home office. Thus, in addition to not having the motivation to sell life insurance, many pure property and casualty agents do not have the mindset for selling life insurance. Multiple-line exclusive agents, as the name suggests, sell only the products of their company or those of companies with whom their primary company has a selling agreement, and have been trained to be conversant in a wide range of both life and property and casualty products. They typically focus on the middle income market, although many of these agents have wealthy clients as they handle the insurance for their large homes and multiple cars. Often, however, they are uncomfortable with more sophisticated

THE PRIMARY LIFE INSURANCE DISTRIBUTION CHANNELS � 29

applications of life insurance such as estate planning or charitable giving. The oft-seen result is that they have wealthy individuals as clients, but only for their property and casualty coverage. In many MLEA companies, the agency force reports through the property and casualty company. The life company subsidiaries do not have a distribution system, but rather serve as "manufacturers" of life and annuity products that are sold through the parent property and casualty company's agents. One unique characteristic of the multiple-line exclusive agent system is the extremely high retention rates of its agents. While many career agency companies struggle to achieve four-year agent retention rates of fifteen or even twenty percent, MLEA companies often have four-year agent retention rates in excess of eighty percent. This is due to the fact that multiple-line agents also receive compensation from the property and casualty products they sell, which are easier to sell than are life products. The flip side of this, however, is that MLEA companies often have difficulty getting their tenured agents to increase their life insurance production, as they are comfortable living off the renewal commissions from their large property and casualty books of business. CAREER AGENT When one historically heard the term "life insurance agent," this referred to an agent who was hired and trained by a specific life insurance company. In the decades of the fifties, sixties, and seventies, before the growth of independent agents, the number of career agents reached an all-time high. This occurred as some of the large home service companies exited this end of the business and moved upscale. The ranks of career agents also swelled as other companies began serving the more advanced life insurance needs of the upscale market for things such as estate planning and business insurance. In more recent years, the number of career agents has declined dramatically, as many companies formerly involved in recruiting and hiring career agents exited retail distribution to concentrate on manufacturing product and selling through independent agents. This decision was most often driven by the financials of the career agent system, which can be a very expensive distribution channel if not managed properly. Industry estimates place the cost of recruiting, hiring, and managing a single career

30 - CHAPTER FOUR

The level of sophistication found in these needs analysis approaches will vary according to both the socio-economic status of the client, as well as the distribution system in question. A needs analysis seeking to understand the financial needs of an older, high-net-worth couple will obviously be more involved than that for a recently married young couple. When it comes to differences by distribution channel, it is generally true that needs analysis approaches employed by the direct response marketers are less complex than those typically utilized by agents or financial planners working in a face-to-face sales situation. The notable exception to this is in the independent marketing channel (IMO). Despite being a face-to-face distribution channel, IMO agents are often selling one or two specific products aimed at serving specific consumer needs. Thus, there is little role for comprehensive needs analysis, except for what might relate to the need for the specific product(s) they are selling. PRESENTING AND CLOSING

Based on the findings in the needs analysis, the next step for distribution is to present a solution(s) that responds to the client's needs, and to close the sale. With today's technology, this presentation usually include one or more illustrations. Such illustrations usually show the policy's basic values, along with projections as to how the policy is likely to perform over time. At this stage of the sales process, the agent's presentation is likely to include several product illustrations (e.g., 10 versus 20 year term) as well as products from multiple companies. Given the complexity of many modem policies, illustrations have become a necessary part of the sales process. Because of the potential for abuse, illustrations are regulated by state insurance departments. These regulations define the language that may be included in any illustration used in the sales process, requirements with respect to elements illustrated, and the requirement that the company appoint an illustration actuary who must attest to the company's compliance with life insurance illustration regulations. Once the client has decided upon the type of policy that he or she wants to purchase, the agent will initiate the application and underwriting process to determine the client's eligibility for coverage and the pricing of the policy based on his or her medical condition. In the traditional life insurance application process, this will involve the asking of a fairly long list of medical and lifestyle questions, and the arrangement for the client

THE FUNCTIONS OF DISTRIBUTION IN THE LIFE INSURANCE BUSINESS -¢- 41

to meet with a paramedical examiner who will take body measurements and collect blood and urine samples. In more recent years, with pressure to decrease the length of time needed to issue a policy, companies have instituted various forms of simplified issue underwriting. This typically involves a shorter number of questions, and may utilize things such as prescription database checks and motor vehicle database checks in place of paramedical examinations. In recent years, companies have also instituted various forms of tele-underwriting in which the underwriting information is taken over the phone by an underwriter, often while the agent and client are together. These new approaches to underwriting began with smaller size policies, but their usage has expanded to larger cases over the years. Beyond simplified issue underwriting, many companies also offer guar­ anteed issue policies. These tend to be small face amount policies that are provided with neither a medical exam nor any underwriting questions being asked. They are marketed as "you can't be turned down." With these policies, however, no benefits are paid if the insured dies within two years of being issued this coverage. In such situations, only the premiums paid into the policy, plus some interest percentage, are returned to the beneficiary. Of interest is the fact that involvement in underwriting used to be considered a major part of the agent's job, as the agent was face-to-face with the client and could help the company assess the client's insurability. In fact, it is out of this traditional role that agents were known as field underwriters. Over time, however, this function of the agent diminished significantly or even disappeared, as agents often sided more with clients in coaching them through the underwriting process, as opposed to being there to provide protective value for the company. Today the term "field underwriter" is rarely used, and while agents are involved in discussing clients with home office underwriters, especially in cases involving large amounts of life insurance, it is frequently more often as an advocate for getting the best rates for the client. In the past, involvement in the underwriting process (as a partner of the home office) would have been considered a significant function of distribution. In today's world of life insurance, it is a neutral, application-taking function at best, and an adversarial function to home office underwriting at worst.

42 -- CHAPTER SIX

salesperson has produced $500,000 of GDC, the payout could be, say, 70% of GDC. This table of percentages is typically called the GDC grid. As some insurance companies turned to brokerage firms for sales (either independent or their own), they needed to convert commissions based on a percent of premiums to GDC to be consistent with brokerage firms' compensation practices. But since life insurance products do not normally have a disclosed expense load, most of these companies had to first define a GDC for their products. This is further complicated by the fact that most investment products have sales compensation in only the first year, while many life insurance companies view at least part of renewal commissions as a form of deferred sales compensation. In the investment products world, there may be customer-account-based trailers paid as compensation for asset retention, but these are not really a form of sales compensation. A common approach is to define the GDC for a life insurance product as equal to the agent's commission plus any other field management overrides. One might alternatively use the distribution expense load or margin priced into the product. Then the salesperson's percentage is set to essentially pay out the same amount of dollars as the traditional commission approach would pay. The grid is used to mimic any production bonus in the commission approach. To replace renewal commission, a renewal GDC is typically defined and the grid is extended to include renewal years. One of the advantages of the GDC approach is its transparency - everyone knows what everyone is being paid. This is not always true for life insurance products, since an expense load or some other proxy for the GDC is not always disclosed. The potential lack of transparency is further complicated by the fact, as discussed elsewhere, that life insurance companies often pay agency managers or GAs a myriad of bonuses, expense allowances, and fees that were never intended to be disclosed to the salesperson. How this additional revenue is accounted for in a GDC approach is not consistent across companies. Unless this is somehow disclosed, the salesperson is left not knowing how others are benefiting from their efforts. In recent years, a number of companies using career agency disctribution channels have converted their compensation programs to be GDC based. Using a GDC based approach does not change the underlying economics of a particular distribution channel. It might make it easier for companies

DISTRIBUTION COMPENSATION � 59

to motivate desired behaviors on the part of their salespeople, or give them access to distribution channels that are not accustomed to compen­ sation based on a percent of premium. NEW YORK STATE REGULATION Through its Regulation 4228, New York is the only state that puts limits on how much a company can spend selling life insurance. New York laws define these as outside limits, which control how much a company can pay producers in commissions, both first year and renewal. These limits grew out of the Armstrong Commission analysis of insurance company malfeasance at the end of the 19th and beginning of the 201h centuries. One of the practices that the Armstrong Investigation cracked down upon was that of selling tontines. As noted in Chapter 2, these were insurance arrangements whereby a group of people pooled money, and the last person left standing collected the money. Instead of putting up reserves to pay for deaths as they occurred, company executives used these funds to spend lavishly on themselves and their producers. New York State stepped in and put an end to such practices by putting a limit on selling expenses not only for sales made in New York, but also in other states if the company is licensed to conduct business in New York. This is called the extraterritorial nature of Regulation 4228, and is the reason that today one will often see companies setting up New York subsidiaries, so that the Regulation does not apply to their non-New York business. Decades ago, the topic of "New York" and "non-New York" companies seemed to be a bigger issue than it is today. At that time it seemed that companies who didn't employ New York subsidiaries to conduct business in the state were somehow seen as higher quality companies. The fact that no non-New York company has ever consistently captured a huge market share over a long period of time suggests that the New York limits are likely close to what is economically feasible in the marketplace. While not as big a factor today, these limits still exist and need to be considered when working in New York State or for a company that is domiciled in the state of New York. The only other example of sales expense limitations in the industry is in the area of variable products, where the Securities and Exchange

60 � CHAPTER SIX

Commission (SEC) establishes such limits. Limits were originally put in place for investment products, and were subsequently carried over to variable life and variable annuity products. In closing this discussion of compensation, it should be clear that agent and agency manager compensation contracts can be very complicated. So when introducing a new contract or significant contract changes, companies will take the time to educate their agents and managers on how to maximize their compensation under the new contract. While this might seem counterintuitive, the logic is that, if the contract is designed correctly, the sooner agents and managers can figure out how to maximize their income, the better it will also be for the company. Elements of compensation contracts can also have consequences that were not envisioned when the contract was designed. In the early days of Universal Life, for example, one company decided to pay commissions based on the policy's face amount instead of the premium. This did not have much impact at the point of initial sale. Agents figured out, however, that they could get another first-year commission by increasing the face amount on in force policies, and paying for this increased face amount with accumulated values in the policy. The unintended and undesired consequence of this practice was to increase the number of underfunded policies in the company's Universal Life block of business. Similarly, when producer groups began courting large producing agents of career agent companies, several career companies attempted to combat this by developing special contracts that allowed groups of agents to combine their production. While this might have served to keep more production with the agent's primary company, it also resulted in the stacking of production simply to qualify for higher production bonuses. Without managing this, companies quickly realized that they were paying more for what they might have gotten even without these special contracts.

CHAPTER SEVEN THE ECONOMICS OF LIFE INSURANCE DISTRIBUTION While compensation plays a significant role in the topic of distribution, it is only one part of a larger discussion of the economics of any specific life insurance distribution system. There are two sides to the consideration of the economics of distribution. First, and almost taken for granted, is the revenue (however it is defined) that will be generated by the distribution channel. Then there is the cost, which, as we have described, can take on many different structures depending on the specific channel being considered. In theory there should be something similar to the classical demand curve in general economics that relates revenue to how much a company spends on its distribution channel. Unfortunately the complexity of the cost structures of the various distribution channels, as well as products being sold, make it difficult if not impossible to divine such a relationship for insurance distribution. In addition, the demand curve can be shifted by the structure of compen­ sation. Agents, for example, will at times value first year commissions over renewal commissions even when adjusted for the time value of money. This is likely unique to a specific company and its distribution channel, but identifying where these shifts might be possible are important when trying to maximize the revenue to cost relationship. Because of the difficulty in identifying a generic revenue/cost curve, the rest of this discussion of the economics of life insurance distribution will focus on estimating the cost of distribution to ensure charges necessary to cover costs are included in the price of the products sold. Distribution management must be relied upon to provide current and future revenue estimates when considering variations in distribution cost structures. Distribution expenses, including marketing costs, are like other operating expenses that must be covered by revenue from customers through 61

62 � CHAPTER SEVEN

premiums or fees and charges in the insurance contact. The fact that distribution costs are generally a company's largest expense makes it critical that provisions are made in setting the premiums and charges that allow the company to ultimately recover the cost of distribution. Besides being larger than most other operating expenses, the unique nature of distribution costs complicates the process for determining the charges. Distribution costs, like new business issue costs, are often front­ loaded. That is, they are incurred before the matching revenue is received. In the independent distribution channel described previously, for example, a company may pay a distributor 120% of the first year premium as a commission. Yet, as a result of market competition, the company is usually unable to include that level of charge in the first year premium. Similarly, most of the distribution channels discussed in this text include some type of significant investment that does not immediately result in sales. This might include agent recruiting and financing costs for a career agency distribution channel, fixed payments to independent agencies to gain access to their producers, or the cost of technology or infrastructure for a direct channel such as a call center. Not only can the recovery for this type of cost not be charged to the customer in the year that it is incurred, it typically is so large that it cannot be charged to a single year's cohort of customers. While complex, these are issues an actuary is generally equipped to deal with using readily available pricing tools. Very complex compensation schemes may require software modification, but as long as the appropriate contingences are modeled, estimated distribution costs can be determined and appropriate charges included in the policy premiums. A detailed discussion of the mechanics of life insurance pricing is beyond the scope of this text. Moreover, it will likely be different for each distribution channel. Rather, this discussion focuses on the issues to be considered when modeling distribution costs and determining appropriate charges. An approach will also be described that, depending on a company's distribution management philosophy, might reduce some of the complexity in setting assumptions with respect to distribution­ related contingences. It makes the actuary an invaluable resource as distribution management develops its ongoing strategy. The first step in estimating the cost of distribution is getting agreement among the various stakeholders as to what payments will be included.

THE ECONOMICS OF LIFE INSURANCE DISTRIBUTION

�

63

This may seem like an obvious step, but in a typical actuarial pricing model where multiple expense types must be combined into one or two inputs, it is easy to miss or double-count an expense. There is also the question of what is to be included as overhead and what is included directly. Marketing costs are an example. Some companies treat them as a corporate overhead expense, others have separate marketing and distribution units within a line of business, and others do not make a distinction between marketing and distribution. Getting agreement is critical. Common payments that might be included in an estimate of distribution expenses include: •

Commissions - This could include first year and renewal commissions (the latter are considered deferred sales compensa­ tion), but not service fees. The same is true for asset trailer commissions. These should be included if they are a sales commission, but not if they are considered compensation for retaining assets. Note should be made of any commission variation due to agents' length of service or volume of sales either within a particular product line or across all lines. Commission chargebacks for lapses or deaths should also be noted as well as any vesting.

•

Overrides - The same first year/renewal split is important, though oftentimes there is either no renewal override, or it is for a much shorter period of time. As with commissions, overrides might also vary depending on the length of service of the writing agent and/or aggregate production of the agency.

•

Incentive Compensation - To the extent that there are home office employees involved in distribution that receive at least a portion of their compensation in the form of sales incentives, these costs need to be included.

•

Bonuses - Bonuses may be used instead of varying commissions for production levels. Bonus will sometimes be paid for superior levels of product persistency or an agent's ranking with respect to sales. Agency managers may be paid bonuses for attaining specific recruiting goals or growing the agencies' aggregate production.

64 -- CHAPTER SEVEN

•

Expense Allowances - Agents will sometimes be paid an allowance to reimburse them for out-of-pocket expenses. This is less prevalent today as many people consider it to be just another commission. General Agents may receive allowances for specific expenses or participate in cost sharing programs.

•

Overhead - This refers to any home office costs allocated to the distribution function. If home office personnel are involved in sales and paid a salary, that expense should be included. It would be useful, however, to identify this separately since, rather than pure overhead, it indirectly ties to sales. Overhead should also include any agency expenses paid directly by the company, including agency management salaries.

•

Recognition - This includes awards and agent conventions.

•

New Agent Financing-This should be limited to the net cost of financing. Some of the cost of financing is recovered from the commissions on the new agent's sales and some general agents are required to share in the cost of financing.

•

Benefits - This includes the cost of providing insurance and/or retirement benefits not included in any of the overhead expenses. Sales representatives must typically attain a certain level of sales to receive subsidized benefits.

Whether or not a company incurs many of these distribution costs is contingent on more than just whether a policy is sold and a premium is paid. The agent's entire first year commission may only be paid if the policy stays in-force for 13 months. Production bonuses or enhanced commission rates may also be contingent on an agent's accumulated yearly sales. Such modeling is no different than the modeling of contingent events done when pricing a product. The major difference is the creditability of the data on which the actuary makes probability assumptions. The contingences included in any distribution cost model are highly dependent on the cost structure, in particular any compensation contracts within the specific distribution channel. The following list provides high­ level examples of what should be considered when modeling distribution costs.

THE ECONOMICS OF LIFE INSURANCE DISTRIBUTION � 65

•

Not Contingent - Some expenses are a given, at least in the short-term. The company is going to incur them regardless. Examples of these are agency or call center rent and certain overhead items.

•

Sales - Many distribution costs are primarily dependent on sales. This includes first year commissions and overrides (which could also be contingent on agents' total sales), recognition and benefits costs (here the contingency may be related to prior year sales).

•

Persistency - This includes both policy and premium persistency. Renewal commissions and overrides are only paid if the policy stays in-force. First year commissions might be reduced as the result of chargebacks if a policy lapses prior to the 13 1h month.

•

Selling Rep's Longevity - Hiring, training, and financing costs are dependent on the amount of agent hiring and how long these recruits are retained. Depending on whether or not renewal commissions are vested, the cost of renewal commissions may be contingent on agent retention. Field manager compensation can be contingent on the length of service of the selling agent.

•

Other - Production bonuses can be contingent on the selling agent's overall production. Compensation packages have recently been offered that allow agents to "bundle" their sales with other agents to qualify for larger bonuses, so the contingency in this case is whether or not the agent is part of one of these groups.

This list is not meant to be exhaustive. Rather, it is meant as a guide to the types of contingencies that may be included in distribution cost projection modeling in order to capture the nuances of the cost structure. With care, an actuarial product pricing model can be used to project distribution costs. This may be done in one of two ways: 1.

With a modification of the basic pricing model to include these additional contingencies, or

2.

By modeling these costs separately and then combining them into a single input to the pricing model.

66 {> CHAPTER SEVEN

It is beyond the scope of this discussion to demonstrate this in detail. While the first approach is certainly possible, the added contingencies can quickly make the model very complex. To the extent that expense factors must be combined to fit the model's inputs using the second approach, validation of results becomes more difficult. One approach that has been taken is to price products with a distribution loading. This could be a first year only loading, or first year and renewal loading. This loading is intended to be the revenue for a distribution business unit, either real or notional. This business unit is expected to manage distribution costs in line with the expected revenue. With this approach, distribution costs are modeled separate from pricing using a stand-alone model of the distribution channel. An approach to defining the distribution margin (revenue) is to treat it as the balancing item in the pricing exercise. This assumes that the competitive market sets the premium or charges, and mortality and persistency is also a function of the market. Interest rates are determined by the investment environment, and the company's risk tolerance and operating expenses are capped at what they would cost if outsourced. The distribution margin is set to a level that results in the desired profit. Under this approach, the distribution business unit must manage its expenses to the revenue it generates (distribution margin) from sales. Alternatively, the business unit first estimates distribution costs for a specific level of production. This becomes an input to the pricing model and other elements in the product design are adjusted to accommodate the desired level of distribution expenses. Observations •

Any split of the distribution margin between first year and renewal will depend on the extent to which the distribution business unit is expected to share in the quality (persistency and possibly mortality) of the business risk.

•

Contingencies specific to distribution expenses are not included in the pricing model if the distribution loading approach is used.

•

How margin is spent is not critical to the pricing exercise though sales forecasts, to the extent they might be dependent on the level of sales compensation, could be important inputs to the pricing model.

THE ECONOMICS OF LIFE INSURANCE DISTRIBUTION � 67

•

It is critical to clearly define which expenses are considered distribution expenses and are included in this margin, and which are covered by another margin built into the product pricing.

The distribution business unit is expected to manage the unit's total expenses to stay within the distribution margin generated by current year sales and, to the extent that deferred margin is priced into the product, appropriate prior year sales. Business unit management must decide how best to spend its revenue. This includes sales compensation, investments in the channel itself (e.g., agent hiring/training, agency acquisitions, call center infrastructure) and overhead, both the business unit's and that of corporate allocations. Investments such as agent hiring and training or call center infrastructure are unlikely to produce significant current year sales, and thus not much revenue. The business unit needs to model its expected ongoing expenses and one time investments, so as to match expenses to revenues over an agreed-upon period of time. An example of such a model is described in the next chapter. In practice, many of the distribution specific contingences described in this chapter will need to be considered to properly manage expenses. The granularity of the model will also vary depending on the structure of the distribution channel. A direct channel, for example, could likely be modeled in aggregate, whereas a career agency channel would best to be modeled at the agency or even agent level. To the extent that renewal compensation is considered sales compensation and the agent is still active or is vested, renewal expenses must also be considered in the model. The strength in treating the distribution margin as revenue and then modeling the expected distribution expenses is a significant improvement in available management information and the ability to clearly measure the economic value of distribution. The distribution business unit is left to allocate that revenue as it sees fit. Absent a viable revenue/cost curve, distribution management can, with this model, quantify the expected changes in revenue to justify proposed changes in costs.

68 � CHAPTER SEVEN

CHAPTER EIGHT USING A PROJECTION MODEL FOR COMPENSATION DESIGN PURPOSE AND LIMITATIONS To better illustrate many of the concepts discussed in the prior chapter, the authors have developed an EXCEL spreadsheet model that is intended to demonstrate many of the compensation design concepts discussed in the prior chapters. A copy of the EXCEL file is available in a copy-protected file at ActexMadRiver.com. As part of the output of the model, an assumed distribution allowable is compared to projected expenses over a 10-year timeframe. Through a wide range of inputs the model is designed to be generic enough to project results for most of the distribution channels that we have described. Three examples are presented in this chapter in order to demonstrate how the model might be used for specific distribution channels. The results are presented in Tables 8.1 through 8.6. In real life a far more sophisticated model would be used, one that likely aggregates the results of the producing units represented by this model into an agency or even a distribution business unit model. The assumptions, design elements and factors presented are purely illustrative. This model was developed for educational use only, and should not be used as a substitute for a detailed actuarial analysis of a proposed compensation desi gn. These examples are not intended to compare one distribution channel versus another. As we will discuss in the next chapter, more goes into the determination of which distribution channel is best than just expected expenses. ASSUMPTIONS When designing compensation, it is common practice to define certain elements of the design based on production credits (PCs) or sometimes first year commissions. PCs are a way for companies to standardize the 69

70� CHAPTER EIGHT

value of a sale across different product lines with different profit profiles. Companies often set a PC equal to the first year commission, but not always. For this model the PC equals 55% of universal life premium, 40% of term premium, 4% of annuity deposit and 100% of GDC. To make the model as generic as possible, assumptions are made for what is called a producing unit. This can be an agent, a call center sales rep or even a group of agents in the case of a PPGA or IMO distribution channel. Assumptions are made concerning the producing units' sales growth and mix of business over their first 10 years of service. For simplicity, the product in the first two years is fixed. Production in the 10 th year is entered and the model grades production from the second to 10 th service year uniformly. The overall cost of hiring or adding one more producing unit can be tested by making assumptions about the survival rate of the producing unit. The annual survival rate is an input item to the model for each of the first four years, as well as a single rate for the fifth year and later. The inputs are the percentage of units surviving to the end of the particular service year. As a simplification, terminations are modeled to occur at the end of the year. To use the model to assess the difference in compensation by service year, simply use 100% as the survival rate each year. The distribution allowable is defined as a percent of total PC for the producing unit's sales for the year. The allowable can be varied by level of production. The rational for this model variation is that agents, for example, during their early service may require additional support not required by an experienced agent, and this might be funded with allowable that might otherwise be available for distribution. Alternatively, there may be fixed management overhead expenses that as a percent of PC are smaller for high producing experienced agents, resulting in an increase for distribu­ tion allowable at higher levels of production. ADDITIONAL MODEL INPUTS AGENT

Sales Growth Track The agent growth track is defined by the PCs the agent is expected to sell in the l0 1h year of service. In the model example, an agent sells 25 ,000 PCs in his or her first service year and 45 ,000 in the second. The sales results grade uniformly to the 10 1h year sales, another input item.

USING A PROJECTION MODEL FOR COMPENSATION DESIGN-¢-- 71

Business Mix This is the assumed distribution of the agent's sales each year by product type. POLICY PERSISTENCY & SIZE

Life Lapse The user inputs the assumed lapse rate for both first year and renewal years for all life products. Size This is the average size policy measured in terms of premiums. ANNUITY FUND GROWTH & TERMINATION

Account Value Growth The user inputs the assumed growth in the annuity value due to interest or fund appreciation. Termination The user inputs a termination rate due to partial withdrawals, full surrenders, and annuitization. VALUATION

Distribution Allowable The allowable is defined as a percent of the PC for a particular product. This is the amount that is priced into the product to cover first year agent compensation expenses. In this model, any deferred sales compensation paid as a renewal commission is covered by a renewal year allowable. The percentage can vary by production level to reflect any economies of scale savings by agent production level that is available to cover compensation expenses. The user enters the level at which there are no further increases. The steps are calculated to be uniform between O and the maximum. Discount Rate This is the discount rate to be used to calculate the present value of the difference in projected distribution allowable (revenue) and total compensation for each producing unit.

72