“National health care is terrain so craggy and forbidding that only the foolhardy venture to cross it.”  So writes Irwin Unger in his recent book, The Best of Intentions: The Triumphs and Failures of the Great Society.  He analyzes the effects of the Medicare and Medicaid programs, initiated in 1965, and concludes that President Lyndon Johnson and colleagues created, at least in part, “a health Dracula that is now draining the lifeblood of the nation.”[1]

All the same, and despite the monsters lurking about, I enter this morning that craggy terrain leading down toward National Health Care Gulch.  The facts of national health care are disturbing.  In its first full year of operation, 1967, Medicare—then embracing Americans ages 65 and over who received Social Security—covered 19 million persons and cost slightly over $3 billion.  Expenses quickly began to soar, rising an average of 17 percent per year, nearly three times the general increase in the consumer price index.  By 1984, Medicare covered 30 million persons at a cost of $63 billion, a twenty-fold increase in just 17 years.  In 2003, Medicare covered 41 million persons at a cost of $283 billion.  In that same year, means-tested Medicaid spending totaled another $267 billion: well over half a trillion dollars when added together. 

More broadly, overall health spending in the U.S. continues to rise at two to three times the rate of general inflation.  In 2003, the increase was 7.7 percent, to $1.7 trillion.  As a proportion of the nation’s total output, expenditures for health reached 15.3 percent.  This figure is, by far, the highest in the developed world; the next closest numbers come from Germany and Switzerland, at slightly above 11 percent.  Some modern nations with universal state health insurance coverage, such as the United Kingdom, Austria, and Finland, report figures under 8 percent: about half of the American number.

Health spending in the U.S. for 2003 reached an average of $5,670 per person.  For a family of five, expenses climbed $1,765 in just one year.  According to analyst Paul Ginsburg of the Center for Studying Health System Change:

Health insurance premiums are growing faster than what people earn.  Government health care spending is [also] growing faster than federal revenues, crowding out other priorities.[2]

Moreover, there is the politically-charged issue of Americans not covered by any form of health insurance, private or public: in 2002, 43.6 million persons and growing, according to the U.S. government.

These problems are real and vast, and much debated.  Blame for soaring costs gets cast about widely.  Some indict greedy HMO’s.  Others point to overpaid doctors.  Still others cite excessive liability insurance costs for physicians, driven by trial lawyers.  Others blame new and expensive medical technologies. 

And yet, commonly forgotten in this debate are the relationships of physical health, health care, and health insurance to healthy marriages and healthy families.  Indeed, these are—I will argue today—the missing pieces in the health debate puzzle, and the touchstone for real solutions.

Moreover, the ongoing debate on health care and health insurance coincides with special attention this year to the future of another kind of insurance, old age insurance under Social Security (which is my subject for next month).  The historian in me whispers that it is time to step back and ask two basic questions: What has been the historic relationship of insurance, generally, to family life?  And, more specifically, how has government-provided, or social insurance, related to the family? 

INSURANCE: RISK-SHARING, REDISTRIBUTION, AND THE MORAL HAZARD

Insurance, broadly considered, is about the sharing of risk.  And it is important to remember that for most of human history, the extended family was each individual’s primary insurer.  When tragedy hit, when death came, when sickness struck, one’s relatives or one’s clan were there to help.  Those without families were in a precarious position, indeed.  Even in America today, some groups still organize their risk-sharing around the extended family or small community: notably, the Old Order Amish, who avoid most forms of insurance (they are, for example, legally exempted from Medicare and Social Security).  They do this, we should note, because they believe insurance of the modern sort, public and private, tends to compromise independence and to weaken family and community bonds.

Still, long ago, the rise of human trade, or commerce, began to generate demand for forms of insurance.  Traces of insurance contracts for sea-going vessels can be found in parts of the ancient world, among the Babylonians, the Phoenicians, and the Greeks.  It appears that the ancient Romans practiced some form of life insurance.  Yet the modern insurance idea developed in the Medieval era, as ties of clan and family in Europe gave way to ties of mutual interest, such as craft guilds.  Another strand of the insurance industry developed among religious bodies or ethnic groups, organized as “friendly societies”: that is, societies of friends who knew each other, who shared common beliefs and values, and who agreed to share certain risks.[3]  This approach took strong root among ethnic and religious groups in the young United States.  Both of my grandfathers, for example, were members of the Skandia Society, which provided insurance and credit for immigrant Swedes in my hometown of Des Moines, Iowa.

Even in these early forms of organization, though, the insurance process had three relevant qualities:

1. All insurance represents a displacement of “family centered” solutions to the problem of risk.

2. All insurance is redistributive.  Money is collected from the participants; some funds are held and some invested; and persons facing certain specified events will receive payments in return.  In true insurance, though, a relatively large number of the insured will receive less than they paid in, and some may receive no money back at all (as example, people with health insurance who never get sick).  In order to cover large losses, this is the only way the system can work. 

3. All insurance runs the risk of “the moral hazard.”  In the narrowest sense, the moral hazard of insurance involves a deliberate act of deception, as in the arsonist who burns down his newly insured building.  The old rule among insurance underwriters, as phrased by the eminent actuary Albert Mowbray, is that “moral hazard…is uninsurable on any terms.”[4]  Or as another insurance theorist explains, the vigilant underwriter will “refuse to accept any application which he feels is contaminated in any way by the moral hazard.”[5] 

The related, “more dangerous” problem is what insurance analysts sometimes call “the morale hazard.”  This form of the moral hazard results from the indifference created by insurance, as in persons who say: “I have insurance, so why should I worry?”  Here, the comfort of insurance alters behavior.  In some cases, people become more careless, sloppier.  In other cases, they make lifestyle decisions that increase their risks.  They may take up skydiving as a hobby, because they have life insurance.  Or they may decide to indulge in risky sexual behaviors, because they have health insurance.  Notably, eminent insurance theorists agree that “the moral hazard is perhaps more serious in health insurance than in any other branch of insurance.”[6]  As we shall see, this is true in more ways than one.

SOCIAL INSURANCE AND THE FAMILY

Social insurance can be succinctly defined as “the attempt of government to apply the principle of insurance to the prevention or alleviation of poverty.”[7]  A broader definition is provided by the actuary Ralph Blanchard:

Social insurance is any form of insurance in which the government goes beyond the regulation of [insurance] practices and the dissemination of information.  It may do so by compelling insurance, by shifting its cost, by subsidy, or by becoming itself an insurer.[8]

In modern societies, the contingencies commonly covered include sickness, maternity, accident, unemployment, invalidity, age, and premature death.  Most historians argue that compulsory insurance of this sort was a logical and inevitable result of industrial development (although the Amish, who today compete quite well in industries such as furniture making, stand as living denials of the term “inevitable”).  Importantly, “sickness insurance” was the first form of social insurance to emerge in the modern Western world.  In 1883, the Reichstag of the old German Empire, at the prompting of Chancellor Otto von Bismarck, approved a compulsory health insurance plan for German workers.  By the 1930’s, most European nations had followed suit.

A common aspect of all of these early social insurance projects was their assumption of a distinctive family system.  Except in the Communist Soviet Union, European nations built their programs on the model of a breadwinning husband and father, a homemaking wife and mother, and a home focused on rearing children.  Working women were rarely eligible for coverage through their outside work; rather, women received their benefits mainly through their roles as wives and mothers.  If a husband died, a woman and her children normally received continued health care, a death benefit, survivor’s insurance, and/or a mother’s pension.  So it might be said that these early systems were pro-family, albeit—by contemporary standards—in a fairly rigid way.

The story in America differed somewhat from Europe.  Prior to the Great Depression of the 1930’s, the Federal government had little experience in social insurance.  The one exception was The Sheppard-Towner Act of 1921.  Passed over the fierce opposition of the American Medical Association (AMA), Sheppard-Towner provided funds to states for pre-natal, maternal, and infant health education and care.  It was the first federal entitlement; there was no means test.  Visiting nurses for pregnant women were a common Sheppard-Towner project, and there is some evidence that the program did directly reduce maternal and infant death rates.[9]  Still, despite this success—or some would say because of it—the AMA succeeded in repealing Sheppard-Towner in 1929.

Among the states, there were tentative experiments in social insurance programs.  Mother’s pensions for widows were the most popular.  Several states adopted small workmen’s compensation programs.  Yet the American commitment to individual and family responsibility stymied most other initiatives.  For example, during the early 1920’s, several states studied sickness insurance.  But the lack of any sense of crisis and powerful opposition by the AMA to “socialized medicine” cut off these initiatives.

Then came the economic collapse of the early 1930’s, the election of Franklin D. Roosevelt as President in 1932, and the initiative known as the New Deal.  In June, 1934, he created The Committee on Economic Security, to study social insurance questions.  The Committee’s report led directly to passage of the Social Security Act of 1935.  This measure resurrected the Sheppard-Towner program of pre-natal and infant-care education and federalized mother’s pensions as the Aid to Dependent Children program.  The measure also created old age insurance funded by “contributions” into individual accounts, together with death and unemployment benefits for covered workers.

As in Europe, the new American Social Security system assumed a breadwinner/homemaker/child-centered home as its model.  As one of the architects of Social Security, Abraham Epstein, explained in his 1933 book, Insecurity: A Challenge for America:

It must be remembered that the American standard assumes a normal family of man, wife, and two or three children, with the father fully able to provide for them out of his own income.  This standard presupposes no supplementary earnings from either the wife or young children.[10]

The Social Security Amendments of 1939 strengthened this orientation by adding survivors insurance for male workers and a homemaker’s pension for their wives.

Unlike in Europe, though, neither the 1935 or 1939 Acts included sickness insurance.  As the Committee on Economic Security had ruefully noted, any such system of national health care would need to rest on “sound relations between the insured population and the professional practitioners”; that is, on the happy participation of the doctors.  And the AMA remained firmly opposed.  Still, the Report laid out principles for a future national system.  It should:  provide “adequate health and medical services to the insured population and their families;” “exclude commercial…agents between the insured population and the professional agencies which serve them;” and provide “cash payments in partial replacement of wage-loss due to sickness and for maternity cases and…health and medical services,” benefits that could be provided by additional payroll taxes of about 6 percent.[11]

The idea of compulsory national health insurance covering all workers and their dependents was reborn during the closing years of World War II.  In November, 1945, Harry Truman submitted to Congress a Presidential message on health care which called for such a plan.  Opposition by the AMA prevailed again.  Instead, a modest wartime expedient began to shape an alternate system.  The Internal Revenue Service had ruled in 1942 that employer-provided health insurance would be exempt from taxation.  This stimulated growth of the now dominant American alternative, where health coverage offered by private insurers is closely tied to employment.

Meanwhile, advocates for a unitary national system turned to incremental steps.  In 1956, Congress added disability coverage to Social Security.  And in 1960, the Kerr-Mills bill became law, providing medical aid to the aged poor on a means tested basis.

The big step came in 1965.  Lyndon Johnson’s overwhelming victory over Barry Goldwater the prior year, and huge Democratic majorities in both the House and Senate, set the stage for creation of the Medicare and Medicaid programs.  Since the late 1930’s, some advocates of social insurance had suggested turning next toward universal health coverage for the aged.  The political advantages of giving aid and support to this huge voting block were obvious.  In retrospect, however, it is clear that arguments used to sell these measures to the public were, at best, flawed; at worst, deliberate deceptions:

1. To begin with, advocates portrayed elderly Americans as an unusually impoverished group, a sympathetic body desperately needing tax-supported medical insurance.  Income statistics provided by the U.S. Department of Health, Education & Welfare showed the elderly having, on average, lower incomes than other adult age categories.  Yet these statistics totally ignored asset ownership, including private annuities, insurance contracts, property, private pensions, and savings, where the elderly enjoyed an overwhelming advantage.  They also failed to note that most of the elderly were, in fact, retired, meaning that their “incomes” would be relatively low.

2. Advocates for Medicare and Medicaid also spread misinformation about the coverage to be provided.  The image held up was that of the responsible elderly being financially ruined by a catastrophic illness.  In fact, the Medicare plan would only provide 60 days of hospital care; Medicaid would kick in only after the patient was financially ruined.  In short, neither measure provided catastrophic coverage.  As one honest Democrat of the time, Russell Long of Louisiana, asked during a Senate hearing:

3. Well, in arguing for your plan you say lets not strip poor old grandma of the last dress she has and of her home and what little resources she has and [then] you bring us a plan that does exactly that unless she gets well in 60 days.[12]

4. Moreover, advocates provided wildly misleading future cost estimates.  In 1965, the Johnson Administration—even using a generous inflation estimate—calculated that Medicare would cost only $12 billion in 1990.  The actual cost turned out to be $110 billion, nearly ten times the estimate.  The price tag for taxpayers was also carefully camouflaged in the bill, with payroll tax increases preplanned through 1987.[13]

5. In addition, the program was sold as a measure of social and economic justice.  In practice, though, the new Medicare tax was regressive, falling most heavily on the working poor and the lower middle class.  Indeed, the program actually saw the working poor subsidizing the medical care of the retired rich.  California Republican James Utt fumed over the false assumption “that everyone over 65 is a pauper and everyone under 65 is rolling in wealth,” but to no avail.[14]

6. Finally, advocates advanced the measure as pro-family.  It would relieve adult children of the responsibility and burden of caring for their elderly parents, they said, so allowing these younger adults to invest more resources in their own children.  This surely did not happen.  In fact, 1965 actually marked the end of the Baby Boom and the beginning of a massive retreat from children.  In practice, the advent of Medicare and Medicaid merely represented another step in socializing direct inter-generational duties and bonds.  Where families once took care of their own, that task would now be further transferred to the state.  [As an aside, I note that a much stronger “pro-family”—or at least “pro-child”—case  for incremental social insurance could have been made if proponents had turned to universal coverage for prenatal services, maternity, and infant care.  Society holds a clear interest in healthy mothers and healthy babies.  And unlike elder care, maternal and infant medical services are truly “preventative,” where you can show a clear return on timely intervention.  However, the elderly proved to be a much more attractive political constituency than were babies.]

THE MORAL HAZARD

Today, the American health care system is a sprawling, joint public-private venture.  Projections for 2005 show total health expenditures of $1,907,000,000,000, an astonishing number.  Of this, $858 billion will be covered by public insurance, $702 billion will be covered by private insurance, $262 billion out-of-pocket, and $98 billion by other private sources.  This system combines the provision of excellent care to many alongside remarkable inefficiencies and massive gaps in coverage, all fueled by the infusion of an ever higher percentage of our national wealth.

It is impossible for me—perhaps for anyone—to outline new policy measures that could set all things right.  The Gulch is too deep.  Relative to the family, though, I want to highlight two examples of “the moral hazard” that have grown within this system, and offer responses.

The first expression of “the moral hazard” here is the life-style subsidy which the existing American system now gives to non-family living.

One of the oldest and strongest findings of social science and public health is the relationship between marriage and raising children and good health.  As early as the 1860’s, William Farr, the Superintendent of England’s Statistical Department, had identified the marriage-health connection.  Examining data from France, he found the death rate for unmarried men, ages 20-30, to be 75 percent higher than that for married men of the same age; this also remained true for men in their forties.  Among women, childbearing in the era before antibiotics remained somewhat risky, and raised the mortality rate for the young and married.  However, after age 40, unmarried women also exhibited death rates 50 percent above those of the married.  As Farr concluded:

This is the general result: -- Marriage is a healthy estate.  The single individual is more likely to be wrecked on his voyage than the lives joined together in matrimony.[15]

More recent inquiries have confirmed the health-giving effects of family living.  Looking specifically at the United States of the 1980’s, Debra Umberson reported in the Journal of Health and Social Behavior that:

[T]he married have the lowest rates of negative health behaviors, and the divorced, the highest rates.  Marital status is somewhat more important to health behaviors than is parenting status.  Parenting status, however, also has a deterrent effect on health compromising behaviors.  This deterrent value depends more on the presence of children in the home than simply having had children.[16]

Some have suggested that the health giving effect of marriage primarily benefits men.  As the feminist author, Jessie Bernard, once put it: “men need marriage more than women do.”[17]  However, when sociologists Catherine Riessman and Naomi Gerstel tested this thesis in 1985, they found surprising results:

[O]n four of the five [health] indicators, separation has a more negative effect on women than it does on men.  More specifically, separated women are more likely than separated men, relative to their respective cohort, to restrict activities due to illness, to limit activity due to chronic disease, to report acute conditions and to visit physicians.  Moreover, divorce also has a more negative effect on women than men on most of these same health indicators.[18]

Bridget Maher’s book for the Family Research Council, A Family Portrait, reports more such evidence on the health-giving effects of traditional family living.  Allow me, though, to cite some very recent results.

From The Journal of Family Issues:

The consistency of the health benefit of marriage, across all domains of health is remarkable….Divorces appear to have the worst overall health profile.[19]

From Social Science and Medicine:

Having children in the home is definitely associated with more favorable outcomes in terms of perceived health, physically disabling conditions and health-related behavior for women with a husband/co-habitant [that is, a husband in the home].[20]

From The Journal of Health and Social Behavior:

Women’s risk of dying decreased as spousal income increased but the opposite was true for men (both [relationships] are statistically significant)….[I]ncreasing spousal income is an asset for women but a liability for men.[21]

From the Archives of Pediatric and Adolescent Medicine:

Exclusive breast feeding for 4 or more months appears to diminish the risk of respiratory hospitalization in infancy to one third or less the risk observed for formula-fed infants.[22]

From The British Medical Journal:

Children attending day care centers have a 1.5-3.0 times higher risk of gastrointestinal and respiratory tract infections than children cared for at home or in small family care groups.[23]

And from the journal Public Health:

This decrease in [hospital] bed usage among the married occurred despite their continuing, albeit declining, majority status within the general population…. The positive relationship between marriage and health has increased steadily since the 1970’s onwards, despite the challenges to marriage in modern society.[24]

This all sounds like good news.  But, as the last citation hints, there is a flipside to the story: the proportion of persons living in traditional married couple homes with children is declining.  Indeed, consider these recent changes in American family life.

As marriage, long term breast feeding, and maternal child care at home all are associated with good health, these turns toward non-marriage, divorce, cohabitation, childlessness, working mothers, and small children in daycare all have the opposite effect: poorer health and higher medical costs.  Indeed, these negative turns are surely caused, at least in part, by the moral hazard of insurance coverage.  Precise calculations here are not yet available, but I would reasonably estimate that about 25 percent of today’s high health costs are the direct result of these changes in household structure and behavior.

In practice, this also means that remaining intact natural families are providing a lifestyle subsidy for persons who now live outside traditional family bonds.  This lifestyle subsidy is paid directly through social insurance measures such as Medicare and Medicaid.  It is also paid through private insurance, particularly where health insurance providers are not allowed to discriminate on the basis of marital status, or where they are forced by law to cover unmarried cohabitors and homosexual partners.[25]

RISE OF THE ELDER-CARE STATE

The second form of “the moral hazard” spawned by our health care system is part of a larger pattern: the change in normative assumptions guiding social insurance.  As noted before, the welfare states that developed throughout the Western world between 1885 and 1965 openly favored a distinct family model: a breadwinning father; a homemaking mother; and children receiving full-time maternal care.  Income transfer policies, from Sweden to the United States to Australia, uniformly taxed income and wealth from those over age 40 and gave most of it to young married adults with children.  In the United States, this was achieved—sometimes indirectly—through large tax exemptions for children and the practice of income splitting in the tax code; through generous housing subsidies through VA and FHA programs; and through a payroll tax that remained small.

This all changed, though, starting—again—in 1965.  Not only were Medicare and Medicaid initiated that year; the same bill also provided for an across-the-board 7 percent increase in Social Security retirement benefits and a big jump in the payroll tax.  Other increases in pension benefits followed, including a generous indexing formula.  As a result, by the 1970’s the Young-Family welfare state was rapidly giving way to the Elder Care state.  Meanwhile, the Aid to Families with Dependent Children program and related measures grew rapidly, funneling billions of dollars more to unmarried women with children.  Where the redistributive effects of the American social insurance system of, say, 1960, had favored the male breadwinner, female homemaker, child rich family, the American system of income redistribution in, say, 1985 strongly favored the elderly and single mothers.  Married couple families with children were now the clear losers. 

Can we gauge the size of this change?  Looking at the same phenomenon in New Zealand, sociologist David Thomson calculated that persons born in 1930 will have contributed about 17 years of average pay in taxes over their lifetime and will have received back 35 years worth in benefits.  However, those born in 1955 will pay 27.6 years of average pay in taxes, while receiving at best 27.4 years in benefits back.  No net gain; perhaps a small loss.  Meanwhile, those born after 1960 will post a net loss of ten years pay.  When compared to those born in 1930, those born after 1960 will have to work up to 28 more years than the earlier couple just to enjoy the same relative standard of living.  Thomson even calls that earlier cohort “the selfish generation.”[26] 

Australian analyst Alan Tapper suggests that in light of this shift of the welfare state, certain parallel trends in family life should cause no surprise.  He points to:

…delayed marriage, less marriage, reduced fertility, an unprecedented divorce rate, and the trends towards the two-earner family, all tendencies which came into prominence in the period after 1970, when the financial forces [of the welfare state] were arrayed against the young.[27]

While calculations of this complexity have yet to be done for America, some preliminary work suggests the same story.  Laurence Kotlikoff and Jagadeesh Gokhale, writing in 1994 for The Public Interest, report that “today’s children may have to hand upwards of 40 percent of their lifetime income over to the government while their grandparents will end up paying just over a quarter of their lifetime income.”[28]  For most Americans in their 20’s and 30’s, moreover, the heaviest tax burden that they now face is the payroll tax, taking 15.3 percent of their income from the first dollar earned: part for old age, survivors, and disability insurance; part for Medicare.  Total federal dollars flowing directly to the elderly totalled about $900 billion in 2001; direct federal benefits specially flowing to young, intact families with children were almost non-existent.

RESPONSES

In short, the moral hazard distorting our health insurance system is now evidenced in both the subsidy and encouragement provided to unhealthy, non-family lifestyles and choices and in the mounting tax burden imposed on young adults, the very ones who must form families if our society is to have a future.  How might we respond?

Relative to private health insurance, the need—in Bryce Christensen’s words—is “to let actuaries be actuaries.”  Measures such as the Americans with Disabilities Act, the Health Insurance Portability and Accountability Act, and so called “community rating” schemes are—in Richard Epstein’s words—“making information a near-forbidden commodity” for health insurance companies.  With “elements relevant to the accurate pricing of risk…excluded from view,” including age, sex, marital status, and sexual preference, everyone becomes “more likely to engage in riskier activities.”[29]

In a properly ordered society, those men and women who marry, bear children, and rear them at home should reap the benefit of making such naturally healthy choices.  By committing to marriage and home, and on the basis of sound business principles, they deserve preferential treatment in health insurance, meaning significantly lower insurance rates.  Those who remain single, divorce, cohabit, or engage in non-marital sex should pay higher rates.  To cries of “Discrimination!,” the proper answer is “No lifestyle subsidy!”[30]

Turning to social insurance, perhaps American families need to acknowledge a hard lesson.  Back in the 1930’s, they surrendered both substantial autonomy to the state and their natural advantage in maintaining good health—in exchange for a Social Security system that did show “pro family” elements and that did, for one generation, deliver substantial benefits their way.  Yet, as outlined above, the scheme turned sour during the late 1960’s.  In retrospect, it appears to have been a fool’s bargain.  By the 1980’s, young families had become the net losers in this state income redistribution scheme: not only in America, but in all Western countries.  Not much has changed since, nor is it likely to, given current political realities (that is, the elderly vote; infants and children do not).  The hard lesson here is that “communitarian” policies such as social insurance, even when crafted with the best of intentions, can only work in small communities, where people can know and watch over their neighbors and so prevent “the moral hazard” from emerging.  Real family autonomy and community strength normally exist in inverse relation to government size and controls.

Regarding health care, the corollary is to begin taking prudent steps to dismantle the Medicare-Medicaid regime, so that the real benefits and advantages of life within the natural family can come back into play.  For example, this family perspective strengthens the case for the favorable tax treatment and expansion of private medical accounts.  Relative to Medicare, the goal should be to reorient the program over time away from universal coverage toward higher co-pays, means-testing, and an emphasis on catastrophic care.  Incentives such as credits against payroll taxes should be provided to families that care for ailing elderly relatives in their homes.  Medicaid should refocus on pre-natal, maternal, and infant care, the one clear place where “preventive medicine” produces powerful results and where the moral hazard seems weakest.

There are dozens of other incremental steps that might be taken.  Rather than a tedious review, let me recommend instead one simple, general standard:

With a possible exception for prenatal, maternal, and infant health measures, support all initiatives that deconstruct social insurance health coverage and that encourage market forces based on sound and thorough actuarial principles, including the consideration of household and family structures.

Under this test, I believe, the natural family can only gain.

Endnotes: