Financial innovation and standards for the capital of life insurance companies
Richard W. KopckeSince their inception, insurance companies, banks, and other financial institutions have played prominent roles in our capital markets. These intermediaries have fostered saving and investment by issuing liabilities that appeal to savers in order to purchase the obligations of investors on attractive terms. Among financial intermediaries, life insurance companies traditionally have distinguished themselves by attracting long-term savings and by providing long-term financing for investment in real estate and durable equipment by businesses.
Because financial intermediaries must bridge the often disparate interests of savers and investors, the evolution of these institutions and their products depends on the opportunities created by the needs of their customers. To a degree, financial intermediaries have been coping throughout this century with the consequences of their own success. By supplying much of the financing needed to build our modern industrial corporations and by giving capital markets a good foundation, financial intermediaries have helped create enterprises that no longer depend so greatly on intermediaries for funds. Furthermore, as the nation has become wealthier and credit markets have become deeper, savers have become more willing to assume more risk in pursuit of a greater share of the returns from their investments. These demands, coupled first with rising interest rates between the late 1960s and the early 1980s, then with the capital gains that subsequently accompanied falling rates, precipitated many of the financial innovations of the last three decades. As rising interest rates depressed their profits, the capital per dollar of assets of many intermediaries fell for a time as they attempted to offer competitive terms for savings. Some, seeking to earn higher yields or to maintain their share of savings, also made riskier investments or sold riskier liabilities. In time, many adopted a "mutual fund" approach to their business as they unbundled their services.
After profits and capital ratios subsided for financial intermediaries during the 1970s, those who supervise and regulate these enterprises adopted new methods of measuring and controlling the risks arising from financial intermediation. Regulators increasingly favored enforcing capital requirements that rise with an intermediary's holdings of certain risky assets, appraising risky assets according to their market values, and imposing prompt remedies when capital ratios become too low. These steps, of course, reinforce intermediaries' interest in redesigning their liabilities to resemble mutual funds, wherein savers implicitly provide the capital to support their investments. The returns on investments in many popular life insurance and annuity contracts, for example, depend on the performance of funds offered by life insurance companies to their policyholders.
More stringent standards for capital may reduce the risk of insolvency, but they also can impose greater costs on financial intermediaries. For intermediaries that hold assets not traded consistently in public markets, the strategy of tying capital to holdings of certain risky assets, marking these assets to market, and requiring the sale of these assets when capital appears to be deficient can entail costs that exceed the benefits. The success of this strategy for managing risk depends greatly on the nature of the risks inherent in those assets deemed risky. The conflation of risk-based capital requirements and of marking risky assets to market is a conservative policy when the values of these assets tend to follow random walks. If, on the other hand, these values tend to revert to trends over time, this policy can increase rather than diminish the risks inherent in financial intermediation. In either case, requiring intermediaries to sell risky assets into illiquid markets tends to dissipate rather than preserve their capital. Finally, the linking of capital requirements to investments in specific assets and the marking of these assets according to their disposal values take a very narrow view of the risks inherent in financial intermediaries' balance sheets, a view that can either exaggerate or diminish the magnitude of these risks. Because the risks borne by an intermediary depend on the mix of assets in which it invests and the liabilities it issues to finance these assets, prudent standards for capital should weigh the characteristics of an intermediary's entire portfolio of assets and liabilities.
This article opens by briefly discussing the role of financial intermediaries in capital markets. The next three sections describe in more detail the distinctive features of life insurance companies. This discussion first examines how insurers have reshaped their liabilities to cope with the consequences of rising interest rates and increasing competition for savings during the past three decades. It also examines how insurers have restructured their assets. This discussion then analyzes the consequences of these financial innovations for the capital of the industry as well as the distribution of capital among life companies.(1) The following section examines the issues relevant for measuring and controlling the capital of life companies, describing when some of the more common approaches are likely to work best and when their costs are likely to exceed their benefits. The final section offers the conclusions.
I. Capital Markets and the Role of Financial Intermediaries
Economic development depends on the efficient transfer of resources from savers to investors. Although savers may finance investments by making loans directly to investors or by purchasing investors' offerings of stocks and bonds,(2) much of this financing passes through financial intermediaries. Insurance companies, banks, thrift institutions, pension funds, finance companies, and other intermediaries issue their own liabilities to savers, using the proceeds to acquire the debt and equity issued by investors.
Financial intermediaries create derivative securities, essentially transforming the obligations of investors into financial assets that appeal to savers. In doing so, intermediaries encourage capital formation by fostering saving on terms that entail a lower cost of capital for investors. Without intermediation, each financial contract between savers and investors must [TABULAR DATA FOR TABLE 1 OMITTED] reconcile their frequently disparate motives. Savers, who seek more than an attractive yield from their financial assets, also value attributes such as liquidity, assurance of their family's well-being in the event of sickness or death, or an annuity for the remainder of their lives after they retire. Investors, on the other hand, ordinarily seek obligations that conform more closely to the life spans of their assets or the patterns of their earnings and cash flows. While households have had limited interest in accumulating savings in 30-year debentures, steel manufacturers have had less interest in financing furnaces by issuing demand debt or options tied to the lender's life span. Intermediaries also serve savers and investors by evaluating investors' prospects, monitoring their performance, and providing both savers and investors a dependable access to funds on terms commensurate with their risks and returns.
Mismatched Books and the Role of Capital for Financial Intermediaries
The nation's tangible assets generally have a long life span (Table 1). During the past four decades, real estate - residential structures, nonresidential structures, and land - has represented about three-fifths of tangible assets. Business equipment and consumer durable goods represent about one-fifth of this wealth. Except for inventories, which now represent about one-twentieth of total assets, the stock of tangible assets is inherently illiquid. The nation as a whole cannot sell a substantial share of these assets very quickly, except perhaps at greatly depressed prices.
Although the nation essentially is committing itself to its fixed investments in real estate and durable goods, individual investors nonetheless may "liquidate" their investments by selling them to others at "fair" prices when a market exists and when buyers and sellers are equally well informed. These transfers are most facile, and investments appear to be most liquid, when incentives for installing new assets are most inviting. This apparent liquidity often diminishes greatly for many assets, especially those lacking dependable public markets, when business activity slumps.
The nation's tangible assets are essentially financed by households' savings. Some of this financing is direct, such as the ownership of residences or [TABULAR DATA FOR TABLE 2 OMITTED] durable goods (Table 2). Most is indirect, taking the form of financial assets, which constitute about two-thirds of households' savings. In turn, about one-half of these financial assets are the liabilities of financial intermediaries, most of which are managed by depository institutions, life insurance companies, and pension plan advisors that frequently offer savers and investors contracts with guarantees of cash values or returns.
By design, many financial intermediaries bear risks in running mismatched books, risks arising from their writing liabilities with specific commitments that, in turn, are backed by their investments (directly or indirectly) in durable tangible assets. In order to fulfill their obligations to their customers, the return to and, therefore, the value of the investments behind intermediaries' assets must generally fulfill investors' expectations. In addition, savers may not attempt to withdraw a significant amount of their savings from these intermediaries when the earnings on their assets may be depressed too greatly or when new opportunities offer savers greater yields.
Financial intermediaries customarily diversify or hedge some of their risks. For example, life insurance companies purchase a variety of assets to achieve a stream of income that more closely matches the outlays required to meet their obligations. These companies also generally issue different types of insurance or investment contracts as well as maintaining other lines of business in order to manage better the volatility of their cash flows. Indeed, much of the financial innovation of the 1970s and 1980s reflected intermediaries' efforts to diversify their businesses and hedge their risks better through broader portfolios of assets and liabilities.
Other things equal, an intermediary's capacity for bearing risk diminishes as its capital diminishes or the difference between the yields on its assets and liabilities diminishes. Capital is the difference between the value of an intermediary's assets and that of its liabilities. The less capital per dollar of liabilities and hence per dollar of assets, the greater is the chance that relatively low returns on investments will prevent the intermediary from meeting its contractual obligations. Intermediaries also expect to earn a sufficiently great margin between the yields they earn on their assets and the yields they offer savers - a portion of which may be regarded as an insurance premium - to accommodate the risks they bear. Mutual fund advisors, for example, invest negligible capital in their funds, earn a comparatively small margin, and retain relatively few risks.(3) Except for fees and commissions to defray the costs of selling their shares and managing their assets, mutual funds essentially pass the returns on their assets and the risks inherent in those assets to their shareholders. Insurance companies, on the other hand, maintain considerably more capital, anticipate earning substantial margins, and retain more risk by offering savers and investors contracts with specific guarantees and options. The more savers value competitive yields and the more investors can avail themselves of competitive terms for funds in public capital markets, the more intermediaries' potential profit and capacity for bearing risk shrink.
In the past two decades, their mismatched books have taken a toll on many financial intermediaries. Rising interest rates depressed their margins as the returns offered on new investments exceeded those on many of their existing investments. Much of the financial innovation of the 1970s and 1980s reflected the strategies taken by intermediaries in order to offer savers competitive returns. As intermediaries sought assets with the greatest promise, their fortunes frequently flowed and ebbed with the value of investments in less developed countries, oil and gas enterprises, farm land, and commercial real estate. Assets increased more rapidly than capital for many intermediaries as they tried to maintain their share of the nation's savings despite the comparatively small margin between the yields on their assets and liabilities. During the 1970s and 1980s, as analysts became more concerned that the capacity for bearing risk was dwindling for many financial intermediaries, those who supervised these intermediaries adopted more formal capital requirements or increased existing requirements.
II. Liabilities of Life Insurance Companies
Since their inception, life insurance companies have managed substantial portfolios of long-term assets on behalf of those holding their life insurance, health insurance, annuity, and pension contracts. Currently, life insurers hold approximately $1.5 trillion in reserves to back the $11 trillion of life insurance in force in the United States and their other obligations (American Council of Life Insurance).
The life insurance industry comprises both stock and mutual companies. Mutual companies tend to be older and larger than stock companies, and they manage about two-fifths of the industry's assets. The financial capital of these companies, their surplus, comprises contributions by policyholders and retained earnings. Consequently, policyholders of mutual insurers generally receive shares of the earnings or surplus that these companies may approve for distribution. Stock companies tend to be younger than mutuals, and the assets of stock companies have grown more rapidly than those of mutuals in recent decades. The financial capital of stock companies comprises the proceeds from sales of equity and retained earnings. Accordingly, those who hold policies with stock companies are not buying an equity stake in the company, and the management of stock companies divides the net investment income from their assets between the interests of policyholders and stockholders.
Insurance Contracts and Savings
Since their inception, life insurers have cultivated their capacity for managing assets by offering savers a range of products, some of which verge on pure investment contracts. Although basic term or health insurance policies entail relatively little saving, policyholders may accumulate substantial savings in whole life policies, annuities, and investment contracts. The appeal of life insurers' contracts often depends on prospective yields on the investments backing them. Not only do the revenues and profits of insurers ordinarily increase with the yields on their assets, but policyholders also tend either to earn greater dividends or to pay lower net premiums as these yields rise.
That insurers' products are regarded, often greatly so, as investment contracts is both inevitable and proper. Insurers' assets both account for a substantial share of personal wealth and finance much of businesses' investment in plant and equipment. When yields are comparatively great, insurers are more willing to promote and customers are more willing to buy those contracts that accumulate assets most rapidly. Accordingly, life insurance companies, acting as financial intermediaries, help regulate the flow of financing for tangible investments depending on the yields businesses are willing to pay and the returns that savers require. By paying sufficient attention to the yields on their assets, insurers and their customers may allocate funds most efficiently to the most promising enterprises.
Term Life and Health Insurance
Even term life and health insurance policies, which verge on pure options contracts, reflect the returns insurers earn on the funds they collect when they sell these contracts. Because of the lag between the collecting of premiums and the payment of claims, insurance companies hold these premiums on behalf of their customers until claims are paid. The premiums of policy owners, therefore, produce a portfolio of assets representing an accumulation of savings to defray the cost of future contingent liabilities. With popular group health plans, for example, if sponsors and insurers are reasonably certain of the magnitude and timing of claims, premiums are essentially the discounted values of the costs of paying these future claims and administering the plan. The rate of discount generally reflects the insurers' return on assets.
The premiums on term insurance contracts mostly depend on the probability of the insured's dying and the expenses for servicing the contract. For example, a company selling $100,000 annual term contracts to 35-year old males may expect 2 of every 1,000 clients to die during the year, an expected obligation of $200 for each such contract. To cover this obligation, expenses, commissions, and profit margin, the premium for these policies might be $250 (see, for example, Gregg and Lucas 1973). If the company collects the full premium at the inception of the contract, the reserve for this policy initially is $200. As the company pays its death benefits during the course of the year, this reserve falls to zero.
A company issuing many such term insurance contracts during the course of the year will tend to hold a fairly stable quantity of reserves, averaging $100 for each contract throughout the year, which engenders a stable portfolio of assets for the company to invest. Insurers ordinarily return a share of the yield on these assets to their policyholders by charging lower premiums than described in the foregoing example or by paying rebates and dividends on their policies.
Permanent, Universal and Variable Life Insurance
The yield on life insurers' assets is of greater importance for permanent insurance contracts than it is for term policies, because these contracts combine life insurance with saving program that allows policyholders to accumulate wealth. Permanent life insurance contracts maintain a constant death benefit over many years without charging premiums that rise as the probability of death increases by setting a premium that initially is greater than that required by term insurance policies. During the early years of a permanent insurance contract, the policy accumulates cash value mainly because the premium exceeds the cost of insurance. In later years, the accrued earnings on previous years' cash values contribute to the growth of future cash values. As the cash value rises, the amount of pure insurance the company must provide diminishes. A $100,000 policy's premium is intended to achieve a cash value of $100,000 at the time the insured is 100, thereby eliminating the element of insurance. The greater the return on assets that insurers expect they will earn, the lower they need set their premiums to fund their permanent insurance contracts.
The premium on a permanent insurance policy entails a guarantee: policyholders' savings in the contract will accrue income at a rate no less than the yield assumed by the insurer to set the premium. To limit the risk created by this guarantee, insurers ordinarily set their premiums by assuming they will earn returns nearer the lower range of their forecasts of prospective yields on their assets. Insurers also typically credit their policy owners with yields in excess of these guaranteed rates of return, so permanent insurance policies usually attain full funding well before the insured reaches 100 years of age. If the yields on the cash values of a company's permanent insurance policies were not competitive, new customers would favor other contracts, and existing policyholders would consider withdrawing their cash values. The more successful are the company's investments, the greater are its dividends or rebates, and the more attractive are its policies.
Basic whole life is only one form of insurance that accumulates cash values in addition to paying for pure insurance coverage. Life paid-up at 65, policies paid-up in 20 or 10 years, and single-premium policies charge greater premiums than whole life to build cash values more quickly, thereby increasing policyhold-ers' commitment to an investment program. Conversely, modified life policies charge comparatively low premiums during their first few years, compensating for the loss of cash value by charging greater premiums in subsequent years.
Although life insurance contracts that accumulate cash values are nominally long-term agreements, they commonly allow policyholders to withdraw their cash value, either by taking policy loans or by surrendering their policies, which can be a valuable option. Older policies, in retrospect, often priced this option too cheaply by fixing the rate of interest on loans, often at 5 or 6 percent, and by not adjusting a policy's dividends to reflect the income on outstanding loans. Newer life insurance policies frequently charge prevailing rates of interest on such loans and adjust the return on a policy's cash value to reflect the rate of interest on any of its loans. Furthermore, those policies that quickly accumulate substantial cash values (such as single-premium contracts) also may impose redemption fees on early withdrawals of funds.
In addition to these traditional forms of permanent insurance, universal, variable, and universal-variable life policies have become more common since the 1970s. These newer contracts allow policyholders either a role in mariaging their cash values or freedom to alter their policy's terms, essentially by separating to a greater degree the insurance features and the investment features of the contract. With variable life, a portion of the cash value is invested in separate accounts - including equity, bond, or money market funds - rather than placed in the company's general investment portfolio; the death benefit varies with the value of these investments, but does not fall below a minimum guaranteed by the policy. Universal life allows policyholders to alter their death benefits comparatively easily and to vary the amount or timing of their payment of premiums; the rate of return accruing to cash values depends on prevailing rates of interest. Universal-variable life, combining features of universal and variable contracts, allows policyholders to choose, within limits, both the timing of payments and the funds in which the assets are invested.
Annuities and GICs
Life insurance companies also offer other financial services to their customers, including guaranteed investment contracts and annuities. Guaranteed investment contracts (GICs) are similar to medium-term notes. A saver might be offered a return of 9 percent for three years. These contracts most frequently are sold to pension and retirement plans sponsored by employers. The appeal of GICs rests on their high yields combined with their guarantee of principal, and their maturities force purchasers to review these investments comparatively frequently.
Annuities promise to pay their beneficiaries an income for a specific interval of time, often from retirement until death. Annuities may be funded by making installment payments over several years or by making a single payment (often by transferring funds from other investments, such as balances in permanent insurance policies or employers' thrift and pension plans). The value of annuities depends very much on the return that an insurer earns on its assets. The greater the yield, the smaller are the installments required to fund a given annuity, and the greater are the income payments resulting from any given investment in an annuity. Annuities often allow policyholders the option to withdraw all or a portion of their cash values before the beneficiary begins receiving income payments from the contract. Though not so common in the past, annuities today generally impose redemption fees for early withdrawals, and the assets backing the annuities may be marked to current prices to value customers' withdrawals.
III. The Composition of Insurance Companies' Liabilities
During the first half of this century, the main business of life companies was the selling of life insurance, and the whole life policy was their most popular contract. By the mid-1950s, reserves against life insurance contracts accounted for almost three-quarters of the total reserves of life companies, and term insurance represented one-sixth of ordinary life insurance in force. The commitment to whole life insurance was sufficiently great that life insurance companies managed about one-fifth of all financial assets held by financial intermediaries, a share second only to that of commercial banks. Mutual life insurance companies, in turn, managed most of the industry's assets, having written nearly two-thirds of the life insurance in force.
Between the 1950s and the 1980s, the competition that accompanied rising interest rates reshaped the life insurance industry. Today, as measured by reserves, the selling of annuities has supplanted the selling of permanent insurance policies. Reserves backing life insurance account for only about three-tenths of the total reserves of life companies, and term insurance accounts for just under one-half of ordinary life insurance in force. Although the assets backing annuities and other products have grown much more rapidly than life insurance reserves, life companies now manage only about one-eighth of all financial assets held by financial intermediaries, a share exceeded by those of commercial banks and pension fund advisors. Furthermore, stock companies, which currently manage about three-fifths of the life insurance industry's assets, have overtaken the mutual companies.
Life Insurance
Before the 1960s, interest rates generally varied between 3 and 6 percent in the United States [ILLUSTRATION FOR FIGURE 1 OMITTED]. Insurers generally priced their permanent life contracts by assuming the return on their assets would average 3 or 4 percent over the life of the contract. This return was comparatively attractive for policyholders, partly because their returns were not taxed as income as they accrued, and partly because other, more lucrative means of accumulating wealth were lacking. The mutual fund industry was negligible, people's investing directly in stocks and corporate bonds had been limited,(4) the certificate of deposit had not yet been introduced, and savings accounts at banks and thrift institutions yielded little more if any than insurance contracts.
As interest rates began ascending in the 1960s, the admonition to "buy term and invest the rest" became more compelling. The cash values of a new policy would be invested in the company's general account, and the return to these cash values would be dictated by the yields of the assets in this account. Accumulating savings in permanent life policies lost its appeal because the yield on the seasoned bonds in life companies' general accounts could not match the returns available from other intermediaries [ILLUSTRATION FOR FIGURE 2 OMITTED]. Between the mid-1960s and the early 1980s, the difference between the yield on insurance companies' investments and prevailing rates of interest on bonds tended to increase, culminating at almost 7 percentage points in 1981. Surplus and capital grew more slowly than assets for many established life insurance companies as interest rates rose. Nevertheless, newly formed and rapidly growing stock companies, which were unencumbered with sizable investments in low-yielding bonds, gained the advantage of promoting contracts that offered high yields coupled with the traditional benefit of deferred income taxes.
If their policyowners were to "invest the rest," then insurers would compete for these savings. Life companies promoted new policies in the 1980s that offered more competitive yields. During the early 1960s, insurers were allowed to establish separate accounts distinct from their general account. These accounts resemble mutual funds inasmuch as they ordinarily do not require the company to contribute capital, and the net income accruing to the accounts belongs to those who hold the contracts that are backed by the assets in these separate accounts. With the sales of universal and variable insurance policies linked to separate accounts, insurers allowed customers to invest their premiums at prevailing rates of return and to decide the allocation of their investments among accounts invested in money market securities, bonds, or equities. In 1982, universal and variable life policies represented a negligible proportion of the ordinary life insurance in force in the United States; by the early 1990s, their share had risen to more than three-tenths. In addition, many insurance agents became licensed to sell directly to their clients mutual funds and other securities, including those underwritten by subsidiaries of life companies.
Not only did rising interest rates deter sales of new permanent life contracts, they also encouraged some policyholders either to surrender or to withdraw funds from existing contracts. During the 1960s, policyholders generally could borrow against the cash values in their life insurance at fixed rates, less than 6 percent in most policies, while the returns on their accumulated cash values would continue to accrue at a rate dictated by the overall return on insurers' general accounts. In other words, those who wished to withdraw funds from their policies possessed an inexpensive option: they could borrow cheaply in order to acquire other assets without reducing the yield on their cash values noticeably.(5) Similarly, with rising interest rates, policies could be surrendered on comparatively favorable terms.
Triggered by high interest rates, policy loans and surrenders have reduced significantly life companies' cash flows. Policy loans rose from almost 4 percent of assets in 1955 to almost 8 percent of assets by 1970. During the 1970s, life companies increasingly tied their rates for policy loans to prevailing yields on bonds in order to deter the demand for loans against new policies. Nevertheless, in the 1970s the flow of policy loans and surrenders helped depress the growth of other investments by insurance companies, During the early 1980s, policy loans increased after interest rates rose sharply [ILLUSTRATION FOR FIGURE 3 and 2 OMITTED]; consequently, outstanding loans rose above 9 percent of insurers' assets. At nearly the same time, surrenders rose considerably, peaking in 1985. This greater volume of loans and surrenders diminished the growth of insurers' other assets between 1978 and 1986.
Although rising interest rates stanched life insurers' investing, falling rates have not yet fully restored this flow of funds. The volume of loans and surrenders has remained relatively high since the early 1980s, suppressing the flow of funds available for other investments. Falling interest rates have produced substantial capital gains in the stock and bond markets over much of the last 10 years. The prospect of these gains, made more accessible by flourishing mutual funds, attracted savings from life insurance companies as well as other intermediaries. Furthermore, some mutual funds, banks, and brokerage houses now offer savers variable life insurance and annuity contracts with relatively low fees; the fund, bank, or broker manages the assets, while purchasing only the necessary coverage and administrative services from insurers.
Annuities and GICs
As rising interest rates depressed the demand for permanent insurance policies, the need for the investment program inherent in these policies also became less compelling with the growth of individual annuities and pension plans sponsored by employers. After World War II, businesses increasingly established defined-benefit pension plans for their employees. These plans may be managed by their sponsors, banks and trust companies, life insurance companies, or other investment advisors. Sponsors of these plans typically review their contracts with their managers frequently, and disappointing performances eventually entail a change of managers.
Life insurers initially attracted a comparatively small share of the funds flowing into pension plans, partly because regulations limited insurance companies' ability to invest in certain assets, including common stock. After legislation in most states allowed life insurance companies more latitude for the investments in their separate accounts for pension plans, and after the Employee Retirement Income Security Act (ERISA) of 1974 required most employers to increase the funding of their defined-benefit pension plans, the growth of reserves for group annuities with life insurers increased significantly [ILLUSTRATION FOR FIGURE 4 OMITTED]. During the late 1970s and the early 1980s, the GICs issued by insurers helped attract a substantial flow of funds into defined-contribution plans sponsored by employers.
ERISA also allowed people not already covered by a pension plan to establish their own individual retirement account (IRA). In 1981, federal legislation both increased the maximum contributions permitted for IRAs and allowed people covered by an employer's pension plan to establish IRAs. During the early 1980s, insurers' sales of individual annuity contracts, especially single-payment deferred annuities, also increased rapidly, as insurers promoted the advantages of tax-deferred investments that earned the relatively high returns prevailing at that time. Consequently, the growth of reserves for IRAs and individual annuities at life insurance companies surged in the mid-1970s and again in the early 1980s.
Since the mid-1980s, sales of annuities have subsided. The growth of reserves for individual annuities has fallen much less than that of group annuities, partly because individual contracts are sold much like tax-sheltered mutual funds to people whose group coverage may no longer look so generous or so secure. But life insurers are not the sole vendors of these contracts. Mutual fund advisors, banks, and brokerage houses also are promoting their IRAs and annuities.
Increasing interest rates, greater competition, the termination of some group defined-benefit plans, and tax reforms have taken a toll on sales of annuities, especially group contracts. As interest rates rose in the late 1970s and early 1980s, employers increasingly turned from insured pension plans to trusteed plans as banks and other investment advisors offered more appealing returns at competitive fees. By the mid-1980s, insurers had successfully promoted GICs offering attractive yields, but by 1991 the sales of GICs had stalled after sponsors of pension plans became concerned about the safety of these investments.(6) The growth of reserves backing group annuities also fell as the number of people covered under group annuity contracts first grew more slowly during the late 1980s, then began to fall by the 1990s as a result of layoffs and the termination of some defined-benefit plans.
IV. Assets and Capital of Life Insurance Companies
Since their inception, life insurance companies have invested most of their assets in mortgages and bonds, which have long been regarded as suitable investments for the reserves and cash values backing permanent life insurance policies. By matching the predictable cash flows of long-term securities with the flows promised in long-term insurance and annuity contracts, insurers could confidently anticipate meeting their obligations as they came due. Furthermore, inasmuch as yields on longer-term securities generally exceed those of shorter-term securities, insurers could price their contracts most attractively by investing in bonds and mortgages, provided yields remained relatively stable. Investing predominantly in shorter-term securities would allow the company to profit should interest rates rise, but should interest rates decline, the return on assets would not be sufficient to provide adequate profits and to service the obligations to policyholders. Insurers investing predominantly in equities or in real estate would bear similar risks; a substantial slump in the value of these assets would diminish insurers' surplus and, possibly, their ability to fulfill their obligations to their policy-holders.
Before 1900, the mortgage obligations of farmers and homeowners along with the bonds issued by railroads, canals, and governments accounted for most of the assets held in the general accounts of life insurance companies. During the last three decades, bonds gradually surpassed mortgages, while commercial mortgages displaced farm and residential mortgages, as employment shifted, in stages, from agriculture to manufacturing and then to service industries (Table 3). Also, as life companies promoted their variable insurance and annuity products, assets held in separate accounts have increased significantly since the late 1970s.
Bonds and Mortgages
Before the 1980s, mortgages represented just over one-third of insurers' assets, but, by the early 1990s, they had fallen to nearly one-sixth of assets. At the same time, insurers' investments in bonds rose from just over two-fifths to more than one-half of their assets. These figures overstate the shift from mortgages to bonds, however. Since the mid-1970s, insurers have increased greatly their investment in the mortgage-backed securities issued by agencies sponsored by the federal government, which are classified as bonds on insurers' balance sheets.(7) Life companies also have purchased mortgage-backed securities from other underwriters. Accordingly, almost three-tenths of insurers' assets, directly or indirectly, were invested in mortgage loans in the early 1990s, while bonds not backed by residential mortgages represented just over four-tenths of assets.
This commitment to bonds and mortgages also is evident in credit markets. Life insurance companies hold about one-third of the corporate bonds issued in the United States (Table 4, upper panel), and in the past 15 years, they have supplied about three-tenths of the net new funds raised through bonds issued by corporations (lower panel). Since the 1960s, the share of outstanding residential mortgages held by life companies has fallen significantly, as they have essentially ceased acquiring these securities directly. Instead, as insurers turned to mortgage-backed securities, their holdings of the outstanding issues of federally sponsored credit agencies have increased substantially since the 1970s. Insurers recently have supplied about one-fifth of the net new funds for these securities.
While the importance of residential mortgages waned during the 1960s and early 1970s, life insurance companies increased their investments in commercial mortgages (Table 3). Since the late 1970s, commercial liens have represented about one-sixth of their assets. Despite this relatively constant share during the past 15 years, insurers' holdings of commercial mortgages essentially doubled between 1984 and 1990 as assets also doubled. The growth of the volume of life companies' investments in commercial mortgages closely matched that of the market for these loans over the past three decades, because insurers have held about three-tenths of outstanding commercial mortgages since the early 1960s (Table 4, upper panel).
[TABULAR DATA FOR TABLE 3 OMITTED]
Capital and Surplus
The assets of life insurance companies are investments made on behalf of their policyholders and owners. The owners' stake is the capital and surplus, or simply capital, which essentially is the difference between the value of the companies' assets and their liabilities.(8) This capital also is a financial insurance reserve that protects the interests of policyholders. With more capital per dollar of assets in their general fund, life companies more likely can meet their obligations to their policyholders despite possible losses on their investments. Although the average capital-asset ratio for the industry has remained fairly constant, near 9 percent, during the past two decades, companies with capital ratios no greater than 6 percent held approximately one-half of the industry'sassets from 1984 to 1990 (Table 5). By 1993, this median capital-asset ratio had risen to 7.7 percent.
For most of this century, the principal source of new capital for established life insurance companies and the life insurance industry has been retained earnings. During much of the past two decades, however, the growth of insurers' assets exceeded the capacity of their retained earnings to accumulate capital commensurately. For example, since 1984, insurers' assets have grown about 11 percent annually. Yet, the contributions of retained earnings would have allowed capital to grow less than half this rate [ILLUSTRATION FOR FIGURE 5 OMITTED]. Consequently, the capital ratios for [TABULAR DATA FOR TABLE 4 OMITTED] some established insurers subsided by the mid-1980s, and the issues of new equity by stock companies became an important source of capital for the industry. During the 1980s, insurers also increasingly sold variable contracts, which accumulate assets in separate accounts rather than general accounts, thereby lessening their need to raise capital. Recently, some mutual companies also have either converted to stock companies to raise new capital or explored the merits of doing so.
Table 5 Capital Asset Ratio Far Life Insurance Companies
By 1993, after the median capital ratio for the industry had risen nearly 2 percentage points, the distribution of capital among insurers became less skewed. Although insurers were able to retain more earnings and issue more equity as interest rates fell in the late 1980s and early 1990s [ILLUSTRATION FOR FIGURE 5 OMITTED], capital ratios rose mostly because the rate of growth of their reserves and assets subsided after the mid-1980s [ILLUSTRATION FOR FIGURE 4 OMITTED]. Capital per dollar of assets for many companies, especially those with the lowest ratios, had risen significantly and their blends of assets and liabilities appeared to entail less risk. Most companies, especially those with relatively low capital, reduced their investments in mortgages, real estate, and low-quality bonds. By 1993, companies for which risky assets exceeded five times capital and for which capital-asset ratios were less than 7 percent held less than one-thirteenth of the industry's assets [ILLUSTRATION FOR FIGURE 6 OMITTED]. Moreover, the outstanding GICs of most companies were no longer so great compared to their capital [ILLUSTRATION FOR FIGURE 7 OMITTED]. Only about one-fifteenth of the industry's assets were held by companies for which GICs exceeded three times capital and for which risky assets exceeded five times capital.
V. Standards for the Adequacy of Capital
Probability theory has long observed that the risks inherent in forecasting mortality and morbidity threaten the solvency of any insurer whose access to capital is limited (Peirce 1878). During the past century, both theory and practice have shown that these failures can be infrequent, isolated events when insurers maintain adequate capital and underwrite their policies prudently. As financial intermediaries, insurers also bear risks arising from their issuing guaranteed liabilities backed by assets whose returns are not guaranteed. With adequate capital and a prudent matching of their assets to their liabilities, these risks, too, may seldom threaten the solvency of insurers.
To measure and control properly the risks arising from financial intermediation, regulators and risk managers increasingly favor risk-based capital requirements, appraising assets according to their market values and taking prompt remedies when capital ratios become too low. If, for example, an intermediary holds only bonds with a negligible risk of default and with cash flows that closely match those entailed by its obligations, then its capital per dollar of assets need only satisfy some minimum capital requirement. If, instead, the intermediary invests in designated risky assets, then it would hold additional capital in proportion to its investment in these assets. Risk-based capital requirements essentially increase the price of holding those assets designated as risky. This price is greatest when these requirements force intermediaries to raise more capital. It is least when intermediaries easily can accommodate these requirements within their customary capital positions.
The details of statutory capital requirements for life insurance companies often vary from state to state. Nonetheless, the asset valuation reserves as defined in the Annual Statement adopted by the National Association of Insurance Commissioners generally apply uniform standards for risk-based capital to life insurance companies. These reserves equal a prescribed proportion of a company's investment in various assets, a proportion which varies by type of asset and by the rating of the asset. In 1988, companies maintained mandatory valuation reserves against equities and bonds. By 1993, valuation reserves were extended to mortgages, real estate, and other invested assets. Reserve ratios varied from 1 percent on highest quality bonds to 20 percent on the lowest quality; from 3 percent on the highest quality preferred stock to 22 percent for the lowest quality; from 15 percent to 30 percent on common stock, depending on the nature of the investment; and from 1.75 percent to 10.5 percent on mortgages, depending on the life company's past experience. Reserves were generally 7.5 percent against real estate and 20 percent against certain other assets, such as venture capital funds and limited partnerships. For life insurance companies, the additional burden of posting these reserves is relatively modest. In 1993, for example, the asset valuation reserve accounted for approximately one-sixth of the capital of the life insurance industry.
The success of any rule defining capital requirements depends greatly on the characteristics of the risks inherent in so-called risky assets. Greater capital requirements ultimately impose greater costs on insurers' policyholders and shareholders, costs that can exceed the benefits of holding more capital once requirements become sufficiently great compared to the risks embedded in insurers' balance sheets. The case for substantial capital requirements is strongest when insurers hold unhedged risky assets for which: (a) returns are driven by random walks, (b) expected returns are low relative to the volatility of returns, and (c) dependable markets or appropriate hedges are lacking, trading is at best sporadic, and prices either are difficult to obtain or are potentially very volatile. Should the values of risky assets tend to revert to trends, however, then the case for maintaining substantial capital requirements and for marking these assets to market becomes less compelling.
When the Values of Risky Assets Are Governed by Random Walks
If the values of risky assets are driven by random walks, then the consequence of "news" is no more likely to increase the value of these assets more than previously expected than it is likely to decrease their value more than expected.(10) Accordingly, after a sequence of unexpected losses, these assets are no more due for redeeming gains than they are due for continuing losses (Cootner 1964, Merton 1990).(11)
A very conservative policy for managing the risk inherent in these assets might require that capital equal 100 percent of the value of risky assets (Friedman 1959, Tobin 1985, and Litan 1987). Only shareholders' funds would be invested in risky assets. Whenever a company's investment in risky assets exceeds its capital, the company has placed some of the funds backing its obligations to its policyholders in these assets. In this case, should the value of the risky assets fall sufficiently, the company's obligations would exceed its resources. If none of a company's obligations to its policyholders is invested in assets that follow random walks, the company's assets almost surely will exceed the value of its obligations over time. Therefore, a 100 percent capital requirement essentially would eliminate the risk of insolvency with minimal monitoring and intervention.
The need for 100 percent capital requirements against the values of assets governed by random walks diminishes as risk managers monitor the performance of these assets more frequently and, when necessary, promptly require the company to raise new capital, to hedge its risks, or to sell some of these risky assets. Should substantial losses on these assets greatly reduce a company's capital-asset ratio, the chance that especially high returns on these assets subsequently would raise this ratio might be little greater than the chance that especially low returns would lower this ratio further.(12) Accordingly, as the value of a company's assets approaches that of its obligations, the odds of imminent insolvency increase, and conservative risk managers justifiably might require that risky assets be replaced by safer assets. Although shareholders would sacrifice relatively little by retaining the risky assets and taking the chance that their value will appreciate, this reduction of risk is prudent from the viewpoint of policyholders, because there is little to gain and much to lose by retaining these assets.(13)
With this strategy of prompt intervention, the magnitude of capital requirements depends on the frequency with which risky assets are reassessed according to their market prices or their disposal values. The more frequently these assets are appraised and the less volatile are the prices of these assets, the more capital requirements might be reduced. For example, if the value of a portfolio of assets seldom changes more than 10 percent between reviews and its assets always may be sold promptly at their prevailing prices, then a 20 percent capital requirement might provide adequate security.
Capital requirements also may be reduced to the degree the expected returns on risky assets are great compared to the volatility of their returns. (See Box.) The random walks of assets typically comprise two elements: a volatility describing the magnitude of random gains and losses, and a drift describing the expected rate of appreciation. As the drift increases compared to the volatility, the odds diminish for accumulating a given net loss during a specific interval of time. Consequently, an insurer requires less capital per dollar of assets, other things equal, to maintain a given degree of security as the expected return on its portfolio increases relative to the volatility of its returns. Finance theory predicts that assets with greater expected returns ordinarily have greater undiversifiable risks; otherwise, investors would bid up the prices, thereby bidding down the returns of assets offering unusually high returns. But, financial intermediaries presumably exist partly to acquire assets and to issue liabilities that are not priced efficiently in capital markets, perhaps for want of adequate public information, or perhaps because these assets are less suitable for other portfolios.(14)
A comprehensive measure of the capital of financial intermediaries comprises both the funds contributed by shareholders (the capital reported on their books) and the value of any additional margin they may earn from buying assets and issuing liabilities that are not priced efficiently in public markets. The lower is this margin, other things equal, the greater book capital must be in order to maintain the same degree of financial security for those holding the liabilities issued by financial intermediaries. A principal concern of regulators and risk managers, supporting their interest in greater capital requirements, is that the competition accompanying recent financial innovations has diminished both book capital and margins for most intermediaries, including many life insurance companies.
When the Values of Risky Assets Revert to Trends
Although the return on assets whose values revert to trends may be very volatile in the short run, over longer periods their returns may be predicted with much greater certitude than the returns on assets driven by random walks. Unlike the case for random walks, once the value of a trend-reverting asset deviates from its trend, the odds of its returning increase with time, while the odds of its straying further diminish. Other things equal, capital requirements for insurers financing these assets with suitable liabilities need not be as great as those warranted for assets governed by random walks. Furthermore, should capital fall, the need to mark assets to market and to enforce remedies to rebuild capital promptly becomes much less compelling when the values of risky assets do not stray off course indefinitely.
The value of a diversified portfolio of real estate investments, for example, might revert to trend more than it might follow a random walk.(15) Because developers and investors tend to err when forecasting rents and the demand for space many years in advance, the supply of space at times exceeds demand at prevailing rents; at other times, it may fall short of demand. Following a surfeit of building, rents tend to fall, and the value of existing real estate can fall well below the construction costs for new real estate for several years. Nevertheless, after a hiatus in new development, the existing space is absorbed, and rents eventually recover. As long as new construction is expected to resume in the future to satisfy new demand, the value of existing real estate is more likely to rise to match its replacement cost than it is to fall a commensurate amount.
In the long run, the rate of return to patient investors reflects the rate of change of replacement costs and the cost of capital more than it reflects the vagaries of the real estate cycle. Investors who acquire real estate when values exceed trend risk earning below-average returns. Those who invest when values are comparatively low may earn above-average returns. But, the rates of return to all investors converge to the trend rate of return as their holding periods increase.(16)
The distinction between assets whose values are governed by random walks and assets whose values revert to trends is not sharp. The values of some assets may exhibit characteristics of both processes, and the enforcement of capital requirements can determine the degree to which one or the other dominates. For example, financial intermediaries often hold assets that lack dependable public markets, assets whose values also might be driven by random walks. If, in the normal course of events, these values should happen to fall enough to depress the capital of intermediaries, thereby triggering substantial sales, this divestiture could temporarily elicit "fire sale" prices in illiquid markets. Once the liquidation by intermediaries is complete and transactions recover, the values of these assets would revert to "normal."
Other things equal, capital requirements need not be as great for trend-reverting assets as for assets governed by random walks in order to achieve the same amount of protection for policyholders. (See Box.) The longer the interval of time, the closer the correspondence between actual and expected returns on these assets tends to become. Accordingly, when insurers match trend-reverting assets to policies with sufficiently long maturities (including the terms of any embedded put options extended to policyholders), they bear comparatively little risk provided their policies' guaranteed yields are fixed according to the longer-run yields they expect to earn on their assets.(17)
The need to mark assets to market and to enforce the prompt sale of assets when capital falls also is less urgent for managing the risks inherent in trend-reverting assets. Should the value of these assets fall substantially, thereby reducing an insurer's capital-asset ratio, the chance that especially high returns eventually would raise this ratio increase with time, while the chance that especially low returns would reduce the ratio further diminishes with time. Therefore, when the value of a company's assets approaches that of its obligations and its liabilities are of sufficiently long duration, the expected losses due to insolvency may be low compared to the expected gains from retaining these assets. Whereas measuring the value of a company's capital by marking risky assets to market might be a fair appraisal when assets follow random walks, this strategy can substantially understate the value of the company when assets tend to return to trends. Indeed, if the conflation of capital requirements and market valuation force insurers to sell or hedge risky assets when their values are lower than average, then these techniques for managing risk, which are conservative for contending with random walks, can reduce the expected value of capital and policyholders' investments when the values of risky assets tend to be governed by trends (DeAngelo, DeAngelo, and Gilson 1994). If the requisite sales are too great or the markets too illiquid, then these techniques may increase rather than diminish the risks inherent in financial intermediation. These techniques also tend to increase the volatility of returns on risky assets, undermining the contribution of intermediaries to financial markets.
Risky Assets and the Value of Capital
In principle, no single asset is inherently risky or safe. An apparently risky asset, when held in properly diversified portfolios, can increase investors' expected rates of return while not increasing the volatility of their returns. Moreover, an apparently risky portfolio of assets, when matched with suitable liabilities, can promise attractive returns on capital at relatively low risk.
Therefore, insurers' capital requirements should take into account the characteristics of a company's entire portfolio of assets and liabilities. These requirements ought not depend only on the expected returns, variances, and covariances for the various assets held by an insurer; they also should depend on the characteristics of the liabilities used to finance or hedge these assets. The capital required for the assets and liabilities allocated to separate accounts or mutual funds - wherein policyholders are granted minimal, if any, guarantees of the value of their investments - can be negligible even though these assets might be invested in equities. Similarly, the capital required for the general account's portfolio, other things equal, should diminish as the characteristics of the account's liabilities correspond better with those of its assets. For example, a portfolio of bonds, which can be risky for a company issuing short-term investment contracts, is not so risky for an insurer issuing longer-term fixed-rate contracts with adequate protection from policy-holders' exercising their put options (including policy loans).
To assess the capital of insurers, their entire portfolio should be marked to market. The marking of risky assets to market values while reporting the values of other assets and liabilities according to other rules misstates the capital of insurance companies. If insurers held only high-grade bonds, and if these investments produced cash flows that matched closely the cash flows entailed by their obligations, then this misstatement might not be important. In these circumstances, if the prices of high-grade bonds fell with rising interest rates, the present value of an insurer's liabilities would fall by a matching amount, and the lower prices of bonds would not reduce the capital of insurers.(18) Nevertheless, not marking all assets and liabilities to market can bias the measurement of capital. If, for instance, the values of risky assets fall when interest rates rise, then ignoring the decline in the value of liabilities while marking down the value of risky assets understates capital. Also, even if insurers hold only high-grade bonds, to the degree the cash flows of these bonds do not correspond closely with the cash flows of insurers' obligations (taking into account any policyholders' put options), the value of capital changes with interest rates.
The Cost of Capital
From the viewpoint of policyholders, capital is a kind of financial insurance that protects insurance companies' guarantees to their customers. Shareholders require a greater return on their investment in the company than do policy owners, because this insurance shifts much of the risk of insurers' investments from policy owners to shareholders. This difference in returns essentially reflects the premium that policy-holders pay to shareholders for this financial insurance. Proper standards for capital ought to reconcile the demand for financial insurance by policyholders with shareholders' willingness to supply this insurance. Although a company with less capital bears a greater risk of insolvency, a company with too much capital may founder by not offering its customers sufficiently attractive returns.
If the assets and liabilities of life insurance companies were traded in perfect markets among investors who were equally well informed and whose assessments of potential returns on those assets and liabilities were similar, then the overall cost of funds for insurance companies would vary negligibly with capital requirements.(19) Greater capital requirements, other things equal, might force a company to depend more on equity financing and less on funds raised from those holding life, health, and annuity contracts. Although shareholders, who bear greater risks, generally require a greater return on their investment than do policyholders, the weighted average cost of funds need not increase for the company in these circumstances. As long as the company's blends of assets and of contractual commitments remained the same, the potential volatility of returns to shareholders and policyholders would diminish with more capital. Consequently, the rates of return required by shareholders and policyholders would fall enough to offset the consequences of an increasing dependence on equity financing.
Without perfect markets and agreement among investors, however, greater capital requirements can raise the cost of capital for insurance companies. Investors not privy to the information available to insurers assess the expected returns or the volatilities of the companies' portfolios differently than the companies' managers. Often the assessments of "outsiders" are too pessimistic, and among outsiders, shareholders often are better informed than policyholders. For this reason, among others, life insurance companies and other financial intermediaries traditionally have profited by transforming the obligations of investors into financial assets that appeal to savers: customers generally value the guarantees and options embedded in intermediaries' contracts more than management believes these features cost. If outsiders are inclined to bid too little for an insurer's assets, the cost of equity financing - the terms at which shareholders are willing to make a leveraged investment in the company - frequently appears too expensive to the company's managers.
Equity financing usually becomes especially expensive at those times when insurers most need new capital, after the returns on insurers' assets fail to achieve expectations or insurers' margins and earnings subside. Therefore, when insurers raise new capital in times of duress, times when outsiders are most inclined to question the past decisions of management, policyholders may pay too great a premium for their financial insurance. The consequences in this case are similar to those that occur with the forced sale of risky assets in illiquid markets.
Greater capital requirements can lower policy-holders' expected returns not only by increasing insurers' cost of capital but also by diminishing shareholders' expected rate of return from operations. Life companies derive a share of their profits from the business of underwriting life, health, and annuity contracts. More capital per dollar of assets entails more capital per dollar of liabilities, thereby diminishing the rate of return on capital from their operations, other things equal. Greater capital requirements, consequently, encourage insurers to increase premiums or to write contracts that accumulate fewer assets in their general accounts. Currently, policies that accumulate assets in insurers' separate accounts appeal most to those customers who believe the net premium on variable contracts is comparatively low as a result of the unusually large returns on stocks and bonds that generally have prevailed since the 1970s. These customers, for the time being, essentially are willing to provide capital on comparatively cheap terms by bearing much of the risk from investments in separate accounts. Also, some insurers have reduced their need to raise capital by writing the basic insurance coverage for mutual funds, banks, and others that promote their own variable insurance and annuity contracts.
VI. Conclusion
Since their inception, financial intermediaries have been evolving from specialized businesses to enterprises offering a variety of financial services. Rising interest rates impelled this evolution during the past three decades as most financial institutions promoted new products to remain competitive. As intermediaries attracted new business, their assets grew more rapidly than their net investment income and capital. The life insurance industry was no exception. This erosion of capital per dollar of assets for life insurers, like that of many other intermediaries, concerned those who supervise these companies, because insurers representing much of the industry placed a greater share of their assets in investments commonly regarded as risky. Furthermore, many of these companies had issued liabilities with potentially short maturities to finance their assets.
Since the early 1980s, most financial intermediaries have increased their capital-asset ratios. Many financial intermediaries, including life insurance companies, also implicitly raised capital from their customers by promoting a "mutual fund" business strategy. Beginning with investment advisors and trusts, then spreading to insurance companies, pension funds, and banks, intermediaries have sold accounts wherein savers expect to earn greater returns by bearing more of the risks of the investments backing their contracts.
For the time being, the mutual fund strategy appeals both to insurance companies and savers. Life insurers have not needed to raise so much capital to support contracts whose funds were invested in separate accounts. Savers have held a greater claim on the net income on assets that were acquired mostly after interest rates peaked in the early 1980s. Although this strategy has flourished with the comparatively great returns (at seemingly little risk) on publicly traded securities during the past decade, its roots run deeper. As savers have become wealthier, investors more familiar, and capital markets broader and more resilient, the opportunities for financial intermediaries to earn substantial margins have tended to diminish. In other words, the customers of financial intermediaries are less willing to pay for the services of their capital.
To cope with a diminishing margin between their return on assets and their cost of funds, financial intermediaries also had pursued more aggressively investments that seemingly were not priced efficiently by other investors, particularly investments in real estate, highly leveraged enterprises, and new ventures. Novel investments, of course, pose new risks and new lessons for intermediaries. Even if they could comprehend and control these risks sufficiently well to earn consistently above-average returns, such investments increasingly are subject to criticism by analysts and supervisors who have become wary of financial intermediaries' capacity for bearing risk during the past two decades. During the 1950s most intermediaries resided within granite walls; today, they occupy glass towers.
To measure and control the capital of financial intermediaries, analysts, regulators, and managers increasingly have favored linking capital requirements to investments in risky assets, appraising assets according to their disposal values, and enforcing prompt remedies when capital ratios become too low. The consequences of such policies, however, depend greatly on the characteristics of the assets and liabilities that constitute
COPYRIGHT 1995 Federal Reserve Bank of Boston
COPYRIGHT 2004 Gale Group