Heritage Foundation Leadership for America

The year 2008 will likely go down in public-pol­icy history as a singularly important one: the year when the first wave of baby boomers began to retire. Over the next 25 years, more than 75 million boomers will begin collecting their Social Security checks, drawing down Medicare benefits, and rely­ing on long-term care under Medicaid. No event will financially challenge these important programs over the next two decades more than this move­ment of the largest generation ever into retirement.

These programs will not only face financial tests. Certainly, these will be great over the next several decades, given that none of these "entitlement" pro­grams even now can easily meet their financial obligations. Doubling the number of people in retirement will constitute a substantial growth in the dependent population of the United States. Per­haps the most important aspect of the boomer retirement is how dramatically it reminds us of the rapid growth of dependency in the United States.

Americans have always expressed concern about becoming dependent on government, even while understanding that life's challenges make most of us, at one time or another, dependent on aid from someone else. This concern stems partly from deeply held views that life's blessings are more readily obtained by independent people and that growing dependency on government erodes the spirit of self-reliance and self-improvement. This helps to explain the broad support for welfare reform in the 1990s.

This concern is also partly explained by a fear that the very nature of American democracy will change as citizens become more reliant on govern­ment. A citizenry that reaches a certain tipping point in its dependency on government runs the risk of evolving into a society that demands an ever-expanding state that caters to group self-interests rather than pursuing the public good.

These demands will probably grow as more and more Americans pay no taxes for the government services they receive. In 2004, 39.6 million adult individuals paid no taxes, and millions more paid next to nothing. This number stood at 38.8 mil­lion in 2007, or 28 percent of all taxpayers. Some will wonder how likely it is that Congress can restrain the growth of dependency-creating pro­grams when more and more Americans pay noth­ing to receive aid.

The growing realization that the flagship entitle­ment programs and the growing number of taxpay­ers with no financial stake in the government threaten to bankrupt that government has led to an increasing interest across the political spectrum in the growth of dependency-creating initiatives. Are we closing in on a tipping point that endangers the workings of our democracy? Have we, perhaps, already passed that point? Can our republican form of government withstand the political weight of a massively growing population of Americans who see themselves entitled to government benefits and who contribute little or nothing for them?

To explore these questions, we need to measure the extent to which federal social programs have grown. How much have such programs "crowded out" what were once social obligations and services carried out by community groups, family networks, and even local governments? In other words, has the civil society yielded significant ground to the federal public sector?

The Index of Government Dependency is an attempt to measure these patterns and provide data to help us ponder the implications of these trends. Table 1 contains the Index scores for 1962–2007, with 1980 as the base year. Based on the Index, Her­itage Foundation analysts have found that depen­dency on government has grown steadily and at an alarming rate in recent decades. Specifically:

• Using a benchmark index of 100 for 1980, the Dependency Indexfor 2007 stands at 238, an 8.8 percent decrease over the 2006 score of 261. The 2006 score of 261 reflected spending for hurricane relief. When compared to the 2005 score of 237, the Index continues its seemingly relentless upward growth.
  
  
• The Index has grown by 30 percent since 2001, and has more than doubled since 1980, increas­ing by 138 percent.
  
  
• Federal spending on educational subsidies for college loans has risen by 108 percent since 2001. Since 1962, higher education aid has increased eleven-fold.
  
  
• The Index's health and welfare component has risen by 196 percent since 1980 and by 25 per­cent since fiscal year (FY) 2001.
  
  
• The Index component for federal retirement outlays has grown by 180 percent since 1980, and by 27 percent since 2001.

The Fiscal Challenges Posed by Growing Dependency

Entitlements. The issue of dependency is partic­ularly salient today because more and more Ameri­cans are about to begin their reliance on government during retirement. At age 65, retirees are eligible to collect income from Social Security and health care benefits from Medicare or Medicaid.[1]  Currently, these programs make up 53 percent of all non-inter­est federal program spending. Over the next two decades, that spending will increase to nearly 65 percent of non-interest spending as 10,000 baby boomers per day retire and begin to collect benefits. Together, these programs will enable the govern­ment dependency of 80 million baby boomers.

This is particularly troubling because most of the impending users of these programs are middle- to upper-class Americans who would otherwise not be dependent on the government. Because eligibil­ity for these programs is linked to age, not financial need, multi-millionaires and billionaires collect the same benefits as do low-income retirees, such as subsidized prescription drugs through Medicare Part D.

To pay for these middle- and upper-class entitle­ments in the coming years will require unprece­dented levels of deficit spending. According to the Government Accountability Office, the amount of debt Americans are facing to pay for these commit­ments is $53 trillion—$175,000 per American.[2]  This will be an unsustainable level of debt that is sure to slow the economy and could force high rates of taxation in the future. The high costs of these pro­grams, which will be shouldered by the children and grandchildren of baby boomers, could lead to further increases in dependency of future genera­tions who would be more likely to rely on welfare in a slow economy, for instance. This snowballing of dependency—caused by Social Security, Medicare, and Medicaid—could send the country past a tip­ping point of dependency that could endanger democracy itself.

Additionally, the cost growth illustrates the bud­getary problem of allowing dependency to grow unchecked. One reason this growth will be so sig­nificant is that these programs grow on autopilot, which, in turn, perpetuates dependence because these programs are not subject to regular debate and evaluation. Unlike nearly all other program spend­ing, Social Security, Medicare, and Medicaid are mandatory spending programs that operate outside of the annual budget process. This entitles these programs to first call on all federal revenues regard­less of other budgetary priorities. Substantive policy reform is required if this automatic dependency is to be drawn down. The solution is to turn these pro­grams into 30-year budgeted programs, subjecting the budgets to debate every five years.

Other policy reforms that emphasize indepen­dence must also be part of addressing the problems inherent in these and other programs. The concept of a safety net ought to be restored to gear Social Security, Medicare, and Medicaid toward only those who truly need them. This can be accomplished by relating benefits to retiree's income and encouraging personal savings during working years.

Growth in the Non-Taxpaying Population. The challenges that Con­gress faces in reforming these entitle­ment programs are heightened by the rapid growth of other dependency-creating programs and in the number of Americans who pay nothing for them. How likely, however, is Con­gress to reform entitlements as rap­idly as it should when so many voters pay little or nothing for the other dependency-creating programs con­tained in this Index? In other words, can Congress rein in the substantial middle-class entitlements in an envi­ronment of rapidly growing depen­dency programs?

In 1962, the first year of the Index of Government Dependency, the percent­age of all taxpayers with zero tax liabil­ity stood at 20.1 percent. This number fell to 16 percent by 1969 before begin­ning a ragged, but ultimately steady, increase. By 2004, the percentage stood at 29.98 percent.[3]  It stood at an esti­mated 28.26 percent for 2007.[4]  In short, we are rapidly approaching a point where one-third of taxpayers do not pay taxes for the fed­eral benefits that many of them may receive.

The Index's Purpose and Theoretical Motivation

The 2008 Index of Government Dependency is organized into four major sections. Section one explains the purpose and theory for the Index. Sec­tion two reviews major policy changes in the five program areas. Section three features a methodology that describes how the Index is constructed. Section four discusses the Index in terms of the number of Americans who depend on government programs.

The Index of Government Dependency is designed to measure the pace at which federal government ser­vices and programs have grown in areas in which pri­vate or community-based services and programs exist or have existed to address the same or similar needs. By compiling and condensing the data into a simple annual score (composed of the scores for the five components), the Index provides a useful tool for ana­lyzing dependency on government. Policy analysts and political scientists can also use the Index and the patterns that it reveals to develop forecasts of likely trends and consider how these trends might affect the politics of the federal budget.

The Index uses data drawn from a carefully selected set of federally funded programs. The pro­grams were chosen for their propensity to duplicate or replace support given to needy people by local organizations, neighborhoods, communities, and families such as shelter, food, monetary aid, educa­tion and health care, or employment.

In calculating the Index, the expenditures for these programs are weighted to reflect the relative importance of service (e.g., shelter, health care, and food). The intensity of someone's dependency will vary with respect to the need. For example, a home­less person's first need is generally shelter, followed by nourishment, health care, and income. We weight the program expenditures based on this hierarchy of needs, which produces a weighted index of expenditures centered on the year 1980.

Historically, individuals and local entities have typically provided more assistance to needy mem­bers of society than they do today. Particularly dur­ing the 20th century, government gradually offered more and more services that were previously pro­vided by self-help and mutual-aid organizations.[5]  Lower-cost housing is a good example. Mutual-aid, religious, and educational organizations have long aided low-income Americans with limited housing assistance, but after World War II, the federal and state governments began providing the bulk of low-cost housing. Today, the government provides nearly all housing assistance.

Health care is another example of this pattern. Before World War II, Americans of modest income typically obtained health care and health insurance through a range of community institutions, some operated by churches and social clubs. That entire health care infrastructure has since been replaced by publicly provided health care coverage, largely through Medicaid and Medicare. Whether or not the medical and financial result is better today, the relationship between the person receiving health care assistance and those paying for it has changed fundamentally. Few would dispute that this change has affected the total cost of health care and the pol­itics of the relationships among patients, doctors, hospitals, and those needing care.

Financial help for those in need has also changed profoundly. Local, community-based charitable organizations once played the major role, which resulted in a personal relationship between the indi­vidual receiving help and those in the community providing that assistance. Today, Social Security and other government programs provide much or all of the income in indigent and modest households. Unemployment insurance payments provide nearly all of the income to temporarily unemployed work­ers once provided by unions, mutual-aid societies, and local charities. Indeed, income assistance is quickly becoming a government program with little if any connection to the local civil society.

This shift from local, community-based, mutual-aid assistance to government assistance has clearly altered the relationship between the person in need and the service provider. In the past, the person in need depended on help from people and organiza­tions in his or her community. The community knew the person's needs and tailored assistance to meet those needs within the community's budgetary constraints. Today, housing and other needs are addressed by distant government employees who have no ties to the community where the needy person lives.

Both cases involve a dependent relationship. However, the dependent relationship with civil society includes expectations of the recipient per­son's future civil viability or ability to aid another person in turn. The dependent relationship with the political system has no reciprocal expectations. The former, based on mutual and reciprocal aid with future aid dependent on the recipient return­ing to civil viability, is essential to the life of civil society itself. The latter is usually based on unilat­eral aid in which the recipient's return to civil via­bility is not a factor. Indeed, "success" in such government programs is frequently measured by the program's growth rather than the outcomes it produces. While the dependent relationship with civil society leads to a balance between the interests of the person and the community, the dependent relationship with the government runs the risk of generating political pressure from interest groups—such as health care provider organiza­tions, local communities, and the aid recipients themselves—to expand federal support.

The Index of Government Dependency provides a way of assessing the magnitude and implications of the change in dependency in American society. The steps taken in preparing this year's Index are described in the methodological section, and the Index is based principally on data from the Presi­dent's annual budget proposal.[6]  The last year for the 2008 Index is FY 2007. We used a simple weighting scheme and inflation adjustment to restate these publicly available data. We encourage replication of our work and will provide the data that support this year's Index to anyone who so requests.

The Index Components

We began by reviewing the federal budget to identify federal programs and state activities sup­ported by federal appropriations that fit the defini­tion of dependency. Specifically, this standard means that a reasonable argument could be made that the program or activity provides goods or ser­vices that could crowd out or constrain private or local government alternatives. Furthermore, the immediate beneficiary must be an individual.

This standard generally excludes state programs that could foster dependency. However, federally funded programs in which the states act as interme­diaries are included.

Elementary and secondary education is the prin­cipal state-based program excluded under this stip­ulation. Post-secondary education is the only part of government-provided education included in the Index. Military and federal employees are also excluded because national defense is viewed as a primary function of the federal government and thus does not promote dependency in the sense used in this research.

We then divided the qualifying programs into five broad components:

1. Housing
   
   
2. Health care and welfare
   
   
3. Retirement
   
   
4. Higher education
   
   
5. Rural and agricultural services

The following sections discuss the pace and content of policy change in these five components. (Health care and welfare are discussed separately.)

1) Housing.[7] The Department of Housing and Urban Development (HUD) was created in 1965 by consolidating several independent federal housing agencies into a single Cabinet department. The pur­pose of the consolidation was to elevate the impor­tance of government housing assistance within the constellation of federal spending programs. At that time it was believed that the destructive urban riots that broke out in many cities in the early 1960s were a consequence of poor housing conditions and that these conditions were contributing to urban decay. To this end, the two initiatives—housing assistance and urban revitalization—were combined in a sin­gle federal department.

HUD spending still largely reflects that dual mis­sion. In any given year, about 80 percent of HUD's budget is targeted toward housing assistance, and the other 20 percent is focused on urban issues by way of the Community Development Block Grant (CDBG) program. Given the nature of these programmatic allocations, HUD budgetary and staff resources are concentrated on low-income households to an extent unmatched by any other federal department.

Within the 80 percent spent on housing assis­tance are a series of means-tested housing pro­grams, some of which date back to the Great Depression. Typically, these programs provide low-income households, including the elderly and dis­abled, with apartments at monthly rents scaled to their incomes. The lower the income, the lower the rent. Traditionally, HUD and the local housing agen­cies provide eligible low-income households with "project-based" assistance, an apartment unit that is owned and operated by the government. Public housing projects have historically been the most common form of such assistance, but they began to fall out of favor in the 1960s because of the rampant decay and deterioration that followed from concen­trating too many troubled, low-income families in a single complex or neighborhood. Periodically, a new form of project-based program is adopted as "reform," but the new program tends to fall out of favor after several years of disappointing results. HOPE VI is the most recent form of project-based assistance, but high costs relative to benefits led the Administration to terminate the program in 2006.

HUD also provides "tenant-based" housing assis­tance to low-income households in the form of rent vouchers and certificates. These certificates help low-income households rent apartments in the pri­vate sector by covering a portion of the rent. The lower the person's or family's income, the greater the share of rent covered by the voucher or certificate. Vouchers were implemented in the early 1970s as a cost-effective replacement for public housing and other forms of expensive project-based assistance, but still account for only a portion of housing assis­tance because of industry resistance to terminating the lucrative project-based programs.

Finally, HUD provides block grants to cities and communities through the CDBG program accord­ing to a needs-based formula. Grant money can be spent at a community's discretion among a series of permissible options. Among the allowable spending options is additional housing assistance, which many communities use to provide assistance to a greater number of low-income households. In 2005, President George W. Bush proposed transfer­ring CDBG from HUD to the Department of Com­merce and reducing funding for the program.

Although HUD programs are means-tested to determine eligibility, they are not entitlements. As a result, many eligible households do not receive any housing assistance because of funding limitations. In many communities, the waiting lists for housing assistance are long—up to several years—and in some cases local housing authorities are no longer adding families to the list because there is no pros­pect of their ever getting an apartment.

Recognizing that HUD housing assistance can cre­ate dependency among those who receive its bene­fits, some Members of Congress have attempted to extend the work requirements of the 1996 Personal Responsibility and Work Opportunity Reconcilia­tion Act to HUD programs. Regrettably, advocates for the poor have thwarted these efforts. To date, the most that can be required of a HUD program bene­ficiary is eight hours per month of volunteer service to the community or housing project.

The complexity of HUD's changing mix of project-based housing assistance can make measur­ing dependency difficult, especially over time. For example, trends in inflation-adjusted HUD spend­ing suggest that dependency has been rising for many years.[8]  Alternative measures, however, such as periodic tabulations of the share of renters receiv­ing some form of housing assistance, indicate no change over the same period. For example, infla­tion-adjusted HUD spending increased by 11.6 per­cent from 1993 to 1999, but the share of renters receiving some form of rent subsidy fell from 18.4 percent to 17.8 percent during that same time period, perhaps reflecting the shift to the more costly HOPE VI program. Census estimates are available for only those two years, so it is difficult to determine the extent to which these numbers char­acterize the entire period. More recently, the increase in HUD assistance—especially in the CDBG program—was caused by efforts to address the rebuilding needs along the Gulf Coast related to Hurricanes Katrina and Rita.

2a) Health Care.[9]  Public health programs, par­ticularly Medicare and Medicaid, are contributing to a growing dependency on government. These two programs were enacted in 1965 to provide coverage for the elderly, poor, and disabled. Medicare delivered benefits to 43.9 million people in 2007 and 48.1 mil­lion people were enrolled in Medicaid that same year. Combined, these programs accounted for $517.3 billion in federal spending in 2006, which translates into 19.5 percent of total federal spending.[10] 

Medicare provides health care for individuals age 65 and over and for those with certain disabilities. Medicare enrollment has increased steadily since its enactment, indicating that an increasing number of people now depend on government for their health care. In 1970, an estimated 20.4 million individuals were enrolled in Medicare. By 2007, the number of enrollees has more than doubled.[11] 

Left unchanged, dependency on Medicare will only grow. During the five-year period from 2007 to 2012, 77 million baby boomers will begin to retire in large numbers, pushing enrollment to unprece­dented levels. This flood of new enrollees will not only increase the number of individuals dependent on the program, but also the demand for new med­ical benefits. While Medicare is the primary source of health care coverage for this population, many enrollees have supplemental private sources of cov­erage, such as employer-provided retiree coverage. However, the demand for new services—such as the addition of a universal prescription-drug benefit in 2003—crowds out private coverage alternatives. Two-thirds of all Medicare enrollees had prescrip­tion-drug coverage from another source before the new drug benefit was enacted.[12]  But according to a recent analysis, the new drug benefit resulted in a crowd-out rate of 72 percent. For every seven pre­scriptions paid for by the government, five would have otherwise been privately financed, resulting in a net gain of only two new prescriptions.[13]  If trends like these continue, Medicare will become the sole financier, not just the primary source, of health ben­efits to this population.

Medicaid, the joint federal–state health care pro­gram for the poor, also faces growing dependency. In 1990, 22.9 million people were enrolled in Med­icaid, a figure which has more than doubled since. Medicaid serves a diverse population of the poor, including children, adults, the elderly, and the dis­abled. While a plurality of Medicaid enrollees were children (49 percent), a plurality of spending goes to the elderly and the disabled (41.8 percent).[14] 

The structure of the Medicaid program varies from state to state because states can determine their own eligibility and benefit levels provided they meet a minimum federal standard. Many states have used this flexibility to expand eligibil­ity further up the income scale. These incremental Medicaid expansions and enactment of the State Children's Health Insurance Program (SCHIP)[15]  have allowed eligibility of more individuals for government health programs, particularly in working families that may have access to private coverage but choose to enroll in government-run programs instead.

This growing dependency directly affects taxpay­ers. Medicare and Medicaid are the two largest enti­tlement programs, and spending for both is expected to skyrocket and become even worse. By 2018, Medicare is projected to cost $879 billion, and federal spending for Medicaid is expected to reach $445 billion. The Congressional Budget Office antic­ipates that the two programs will consume 12 per­cent of GDP by 2050.[16]  Actuaries at the Centers for Medicare and Medicaid Services at the Department of Health and Human Services predict that the gov­ernment (federal and state) will fund almost one-half of all health care spending by 2015.[17] 

While Congress did not attempt to expand Medicare in 2007, the program is still adjusting to the addition of a universal prescription-drug bene­fit enacted in 2003. This benefit makes Medicare the primary source of prescription drugs for seniors. Congress did spend a tremendous amount of time on Medicaid and the closely aligned SCHIP during the summer of 2007. For the first time since its creation, SCHIP was proposed for reauthoriza­tion. Some in Congress seized the opportunity to broaden eligibility of SCHIP and Medicaid. These efforts were ultimately thwarted by two presiden­tial vetoes. However, some states continue to use existing flexibility and authority to expand the size and scope of these programs.

In its yearly survey of sources of health insurance coverage, the U.S. Census Bureau in 2007 pub­lished figures that underscore the current trend toward government dependency.[18]  The percentage of Americans with private health insurance is on the decline, mostly as a result of the steady erosion of employer-based coverage, while the percentage of Americans on government programs is rising even faster, in large part due to Medicaid and SCHIP expansions at the state level and an aging population that is becoming increasingly depen­dent on Medicare.

Government-run health care is unsustainable. Without fundamental change, there will be far greater dependency on the government for health care, fewer workers to pay for it, and less incentive for private-sector solutions. Instead of depending on the government for health benefits and services, a better alternative would be to convert the money used to administer public health programs into a direct subsidy to help those in need purchase pri­vate health care coverage.

2b) Welfare.[19]  The 1996 Welfare Reform Act, or the Personal Responsibility and Work Opportu­nity Reconciliation Act (PRWORA), replaced the decades-long Aid to Families with Dependent Chil­dren (AFDC), through which recipients were enti­tled to unconditional benefits, with Temporary Assistance to Need Families (TANF), a block grant program. Welfare reform effectively altered the fun­damental premise of receiving public aid and ended it as an entitlement. Receiving assistance was now temporary and tied to demonstrable efforts by the recipients to find work or take part in work-related activities. Self-sufficiency of the recipients became the focus. The successes of welfare reform are unde­niable. Between August 1996 and March 2008, wel­fare caseloads declined by 63.5 percent, from 4.5 million families to 1.6 million families.[20]  The legis­lation was similarly successful in reducing child poverty. Since 1996, 1.3 million children have been lifted out of poverty.[21] 

The initial years after welfare reform witnessed dramatic progress, but by the late-1990s, most states had met the PRWORA's work goals, and the motivation to further reduce dependence and encourage work among recipients waned. The national TANF caseload flatlined between 2000 and 2005, and only one-third of TANF recipients worked. In February 2006, after four years of debate, Congress reauthorized TANF under the Deficit Reduction Act. The new legislation reiterates the need to engage recipients in acceptable work activities, moving them to self-sufficiency. Once again, states are required to increase work participa­tion and to reduce their welfare caseloads using the lower 2005 caseload levels as the new baseline, which essentially restarts the 1996 reform. As required by the Congress, the Department of Health and Human Services also issued new regulations to strengthen work participation standards.[22] 

The 2006 reauthorization also contains a notable measure that begins to rectify the inattention to the other two 1996 welfare reform goals: reducing unwed childbearing and restoring stable family for­mation.[23]  The erosion of marriage and family is a primary contributing factor to child poverty and welfare dependence, and it figures significantly in a host of social problems.[24]  Troublingly, for the last four decades, the unwed birth rate has been rising steadily, from 5.3 percent in 1960, to 38.5 percent in 2006.[25]  Today, more than one child in three is born outside of marriage. In the TANF reauthoriza­tion, Congress, for the first time, enacted a healthy-marriage initiative, allocating $100 million in TANF funds per year—less than 1 percent of total TANF expenditures in the fiscal year 2006—to local organizations that provide voluntary mar­riage-centered services and skills training to recipients. In doing so, the government is finally recognizing the critical role that a stable marital and family environment plays in reducing child poverty and welfare dependence.

Despite the 1996 Welfare Reform Act and the 2006 TANF reauthorization, comprehensive wel­fare reform is far from achieved. Today's welfare sys­tem is a convoluted machinery of 70 programs, six federal departments, and a voluminous collection of state agencies and programs. A typical welfare recip­ient family could receive assistance from six or seven programs (e.g., TANF, Medicaid, food stamps, public housing, Head Start, and the Social Service Block Grant) administered by four different depart­ments.[26]  Too many of these welfare programs oper­ate on means-tested eligibility and without any real mechanism to break dependence. Twelve years after the reform, the welfare system still rewards non-work. Further reform efforts should focus on applying TANF principles to other failing welfare programs that subsidize idleness and foster depen­dency, and remove the anti-marriage bias and eco­nomic marriage penalties inherent in other means-tested welfare programs (e.g., EITC for married couples with children).

3) Retirement.[27] Since the time of President Franklin D. Roosevelt, the American retirement system has been described as a three-legged stool consisting of Social Security, employment-based pensions, and personal savings. Yet the reality is quite different. Almost half of American workers (about 71 million) are employed by companies that do not offer any type of pension plan. This propor­tion of private pension coverage has remained roughly stable for many years, and experience has shown that few workers can save enough for retire­ment without an employer-sponsored pension plan. For workers without a pension plan, the reality of their retirement is closer to a pogo stick consisting almost entirely of Social Security.

Since 1935, Social Security has provided a signif­icant proportion of most Americans' retirement incomes. The program pays a monthly check to retired workers and benefits to surviving spouses and children under the age of 18.[28]  Monthly bene­fits are based on the indexed average of a worker's monthly income over a 35-year period, with lower-income workers receiving proportionately higher payments and higher-income workers receiving proportionately less. The lowest-income workers receive about 70 percent of their pre-retirement income, average-income workers receive 40 percent to 45 percent, and upper-income workers average about 23 percent.

However, the demographic forces that once made Social Security affordable have reversed, and the program is on an inexorable course toward fiscal cri­sis. To break even, Social Security needs at least 2.9 workers paying taxes for each retiree who receives benefits. Today, the ratio is 3.3 workers per retiree and dropping because the baby boomers produced fewer children and are now nearing retirement. The ratio will reach 2.9 per retiree in 2017 and drop to 2 workers per retiree in the 2030s.

Current retiree benefits are paid from the payroll taxes collected from today's workers. Starting in 2017, Social Security will not collect enough in taxes to pay all of the promised benefits.

Since 1983, workers have been paying more in payroll taxes than the Social Security program needed. These additional taxes were supposed to accumulate to help to finance retirement benefits for baby boomers. But these excess taxes were not saved or invested for the future. Instead, the money was used to finance government programs. In return for the diverted revenue, Social Security's trust fund received special issue U.S. Treasury bonds. In 2017, when Social Security starts redeem­ing its Treasury bonds, the federal government will be required to pay off the bonds through higher taxes or massive borrowing.

Social Security's uncertain future is a problem for all workers, but especially for the roughly half of the American workforce that has no other retirement program. Few of these Americans have any signifi­cant savings, and they will depend heavily on the government for their retirement incomes.

This dependency is largely the result of govern­ment policies. By soaking up money that could oth­erwise be invested for the future, Social Security's high tax rate makes it much harder for lower-income and moderate-income workers to accumu­late any significant savings.

Government policies also discourage the growth of occupational pensions to cover a higher proportion of the workforce. Over the last few decades, the cost of traditional pension plans has skyrocketed, and thou­sands of them have shut down. Efforts to develop innovative hybrid pension plans stalled when confus­ing laws and regulations resulted in lawsuits.

While many larger employers have substituted defined-contribution plans, such as 401(k) plans, both types of plans are subject to the Employee Retirement Income Security Act (ERISA). ERISA reg­ulations are especially onerous to smaller employers, who usually lack the resources to hire a good funds manager and the necessary knowledge of the com­plex legal requirements. As a result, small businesses hesitate to offer retirement plans to their workers for fear of accidentally violating a regulation.

A simpler, less regulated account suitable to smaller businesses would go a long way toward increasing the number of workers with retirement savings. Simplified automatic enrollment proce­dures, automatic investment choices, procedures that allow savings to follow the worker from employer to employer, and better annuity choices would also help. Regrettably, until these policies move from theory to reality, Americans face increased dependence on a government-managed Social Security system that cannot possibly meet their needs.

4) Higher Education.[29]  In 2006–2007, the fed­eral government spent more than $86 billion on student aid for postsecondary education—a real increase of 77 percent over what was spent 10 years earlier.[30]  In all, more than 6 million students received federal grants and 8.5 million took advan­tage of various education tax benefits.[31]  The federal government also provided more than 18 million loans to undergraduate and graduate students.[32]  The consistent growth in federal aid for higher edu­cation follows continuous increases in higher edu­cation costs. In 2007, the College Board reported that the annual cost of tuition, fees, room, and board increased by 5.9 percent at private four-year colleges and public four-year colleges (in state).[33]  Over the past decade, the real cost of a year at four-year private and public colleges increased by 29 percent and 41 percent, respectively.[34] 

The federal government continues to increase federal spending on higher education. In September 2007, President Bush signed the College Cost Reduction and Access Act of 2007 (H.R. 2669), which amended the Higher Education of 1965.[35]  The legislation increased funding for Pell Grants, established new targeted grant programs, and reduced the interest rates on federally subsidized student loans.[36]  It also reformed aid programs to reduce government payments to lending and guar­antee agencies and increased fees for lenders. The long-term cost of this legislation is projected to be $15 billion over the next 10 years.[37] 

Despite this considerable investment in higher education subsidies, there is growing concern about rising tuition costs at American colleges and univer­sities.[38]  According to the College Board, during the 2007–2008 school year the total cost of tuition and fees increased by 6.3 percent at four-year private colleges, and by 6.6 percent at public colleges.[39]  These increases continue the general trend in recent decades of steady annual tuition increases. Between 1982 and 2003, college tuition costs increased by 295 percent, outpacing health care (195 percent), housing (84 percent), and the Consumer Price Index's category "all items" (83 percent).[40] 

In his book Going Broke by Degree, Ohio Univer­sity economist Richard Vedder argues that increas­ing government support for higher education has contributed to rising tuition costs. "Students receiv­ing grants or subsidized loans are far less sensitive to tuition increases than they would be if they were paying their own way," Dr. Vedder argues. "Where entrepreneurs in a free, unsubsidized market seek to cut costs and lower their prices to lure new cus­tomers away from businesses that are raising theirs, there is very little of that in higher education."[41] 

Policymakers should reassess both the effective­ness and equity of continuing to increase federal subsidies for higher education. First, increased fed­eral subsidies are likely to contribute to rising col­lege tuition prices.[42]  Second, the Census Bureau reports that only 28 percent of American adults have college degrees.[43]  Since college graduates are likely to earn significantly more income over the course of a lifetime than non-college graduates, tax­payers should question the equity of providing increased subsidies to this population.[44]  To address these issues and curb growing dependency on gov­ernment, Congress should reform federal higher education policies to reduce spending and refocus grants and subsidies on students who otherwise cannot afford higher education.

5) Rural and Agricultural Services.[45] Much of the rapid increase in "rural and agricultural assis­tance" dependency is rooted in farm subsidy pro­grams. A multitude of farm subsidies (e.g., direct payments, countercyclical payments, market assis­tance loans, and non-recourse loans) generally work together to compensate farmers for low crop prices. Conservation payments pay farmers to ini­tiate conservation projects or simply to stop farming their land. Export subsidies effectively lower the price of American products so that they can under­cut international competitors.[46] 

Farm subsidy supporters often describe farmers as impoverished victims of unpredictable weather and large global economic forces. In reality, farmers are doing quite well. The average farmer has a net worth of $895,756[47]  (double the national average of household wealth), and an annual income of $77,654[48]  (62 percent above the national median[49] ) despite living in a rural area with significantly lower costs of living. The failure rate for farms is just one-sixth the rate of other businesses.

Yet farm subsidies have become America's largest corporate welfare program. Two-thirds of farm sub­sidies are distributed to just 10 percent of farms, most of which have annual household incomes over $130,000. In contrast, the bottom 80 percent of farmers receive just one-fifth of the subsidies. If farm policy were really designed to help poor farmers, Congress could guarantee every full-time farmer in America an income of at least 173 percent of the federal poverty line ($34,873 for a family of four in 2004) for just $4 billion per year.

Instead of need, farm subsidies are based on two factors: which crops are grown and how much is grown. Approximately 90 percent of all farm subsi­dies go to growers of just five crops: wheat, corn, cotton, soybeans, and rice. Growers of most other crops are ineligible for most subsidy programs, regardless of need.

Farmers who plant more crops receive larger sub­sidies. This is where the economic logic of farm sub­sidies falls apart. Subsidies are intended to compensate farmers for low prices that result from an oversupply of crops, but granting larger subsi­dies to farmers who plant the most crops only encourages them to plant yet more crops, driving prices even lower and leading to calls for larger sub­sidies. Furthermore, while paying some farmers to plant more crops, the Conservation Reserve Pro­gram pays other farmers to plant fewer crops. One analyst accurately describes U.S. farm policy as "one foot on the break, one foot on the accelerator."[50] 

Eventually, Congress acknowledged the failures of centrally planned agriculture. The 1996 Federal Agricultural Improvement and Reform Act of 1996[51]  (also known as the Freedom to Farm Act) was designed to slowly phase out farm subsidies by 2002 and allow the agricultural sector to operate as a free market. After spending just $6 billion on farm subsidies in 1996, Congress overreacted to a temporary dip in crop prices in 1998 (resulting from the Asian economic slowdown) by passing the first in a series of annual emergency bailouts for farmers. By 2000, farm subsidies hit a record $30 billion. Farmers quickly grew accustomed to mas­sive government subsidies, and competition for the farmer vote induced a bipartisan bidding war on the eve of the 2002 elections. Lawmakers gave up on reform and enacted the largest farm bill in American history, projected to cost at least $180 billion over the following decade. Despite escalat­ing costs and negative economic effects, farm socialism is now the overwhelming preference of Congress and the White House.

Farm dependency will almost certainly continue. Policymakers mistakenly see farm subsidies as the solution to (rather than a significant cause of) low crop prices. Expensive disaster payments are doled out whether the weather is bad (crops destroyed) or good (crop oversupply lowers prices). Finally, farm subsidies have created an entitlement mentality among a class of farmers who will likely punish any elected officials who pursue reform. Currently, there are no plans to move farmers toward self-sufficiency.

Indeed, the current House- and Senate-passed farm reauthorization bills would not fix this system. They would increase subsidy rates, and reject President Bush's call to subsidize only those farmers earning less than $200,000 annually. Both bills contain large gimmicks to conceal 10-year spending hikes of approximately $25 billion. Even corn farmers, already ben­efiting from soaring prices resulting from federal ethanol policies, would continue to receive billions in annual subsidies. America's trading partners would likely respond to these trade-distorting policies by imposing trade restrictions, thus harming American farmers and consumers.

How the Dependency Index Is Constructed

After identifying the government programs that contribute to depen­dency, we further examined the data to identify the components that con­tributed to variability. Relatively small programs that required little funding and short-term programs were excluded. The re­maining expenditures were summed up on an an­nual basis for each of the five major categories listed in Table 2.[52]  The program titles are those used by the Office of Management and Budget for budget function and sub-function in the budget ac­counting system.

We collected data for FY 1962 through FY 2007. Deflators centered on 2000 were employed to adjust for inflation.

Indices are intended to provide insight into phe­nomena that are either so detailed or complicated that simplification through arbitrary but reasonable rules is required for obtaining anything other than a rudimentary understanding. For example, the Consumer Price Index (CPI) of the Bureau of Labor Statistics is a series based on an arbitrarily selected "basket of goods" that the bureau surveys periodi­cally for price changes. The components of this basket are weighted to reflect their relative impor­tance to overall price change. Energy prices are weighted as more important than clothing prices. Multiplying the weight times the price produces a weighted price for each element of the CPI, and what the total of the weighted prices produces is roughly the CPI score.

The Index of Government Dependency gener­ally works the same way. The raw (or unweighted) value for each program (i.e., the yearly expendi­tures of a program) is multiplied by its weight. The Index for that year is the total of the weighted values.

The Index is calculated using the following weights:

Housing: 30 percent

Health and welfare: 25 percent

Retirement: 20 percent

Higher education: 15 percent

Rural and agricultural: 10 percent.

The weights are "centered" on the year 1980. This means that the total of the weighted values for the Index components will equal 100 for 1980, which gives the Index a reference year from which all other Index values can be evaluated.

We chose the year 1980 because of its apparent significance in American political philosophy. Many analysts view 1980 as a watershed year in U.S. history because it seems to mark the begin­ning of the decline in left-of-center public policy and the emergence of right-of-center challenges to policies based on the belief that social systems fail without the guiding hand of government.[53] 

The Index certainly reflects such a watershed. Chart 8 plots the Index from 1962 to 2007. The scores have clearly drifted upward over the entire period.

There are two plateaus in the Index—the 1980s and 1995–2001—that suggest that policy changes may significantly influence the Index growth rate. During the early 1980s, the growth of some domes­tic programs was slowed to pay for increased defense spending, and Congress enacted significant policy changes in welfare and public housing dur­ing the 1990s. Both of these reduced the Index growth rate.

Chart 9 connects the Index to major public pol­icy changes. The largest jump in the Index occurred during the Johnson Administration fol­lowing the passage of the Great Society programs. The Johnson Administration not only launched Medicare and other health programs, but also vastly expanded the federal role in providing and financing low-income housing. The Index also jumped 92 percent (from 36 to 69) under the Nixon and Ford Administrations, when Republi­cans were funding and implementing substantial portions of the Great Society programs.

The two periods of relatively more conservative public policy (the 1980s and 1995–2001) stand out clearly in Chart 9. The slowdowns in spend­ing increases during the Reagan years and after the 1994 congressional elections produced two periods of slightly negative change in the Index. These periods saw significant retreats from the Great Society goals, particularly in the nation's approach to welfare. The return of budget sur­pluses during the last years of the Clinton Admin­istration, however, led to significant spending increases for all of the components, particularly education and health care. Since then, the Index has grown at roughly the same rate as it has during the past 25 years.

Calculation of Covered Population

The Index reflects the growth of federal gov­ernment programs that arguably crowd out or substitute for similar initiatives advanced by gov­ernments below the federal level or by organiza­tions within civil society. Index values do not depend on the number of people receiving sup­port through these programs, but that number nevertheless sheds additional light on what the Index illustrates.

Data on the number of people enrolled or bene­fiting from the programs listed in Table 2 between 1962 and 2007 were drawn from a variety of public sources. A significant effort was expended to elimi­nate duplicate enrollments. For example, many people who receive food stamps also receive medi­cal services through Medicaid.

Chart 10 shows the annual number of program participants from 1962 through 2007. On the eve of the Great Society programs, some 21.7 million peo­ple (or 11.6 percent of the population) received assistance through the programs listed in Table 2 that existed at the time. Today, 58.7 million people (19.4 percent of the total U.S. population) receive some level of assistance through the programs included in the Index.

Growth in income and non-financial support among program participants has accompanied the increase of people who receive assistance. Per capita financial and non-financial support (after adjusting for inflation) stood at about $6,440 in 1966. By 2007, this support had grown to about $26,000. (See Chart 11.)

Data in the Index and complementary estimates of program populations raise concerns about the ability of local governments and civil society orga­nizations to provide aid and other assistance. They also raise traditional republican concern about the long-term viability of political institu­tions when a significant portion of the population becomes dependent on government for most or all of its income.[54] 

Nearly one-fifth of Americans (19.4 percent) on welfare may or may not be sufficiently high enough to trigger this concern. However, this percentage grows to 27.3 percent when federal and state employees are included. In 1962, the sum of these two categories (Index participants and government employees) stood at 33.6 million, or 18 percent of the total population. This total grew to 82.4 million (or 27.3 percent of the total population) by the end of 2007, an increase of 145 percent. This is 2.3 times the growth rate of the U.S. population over the same period and 30 percent faster than the growth rate of the population age 65 and above. (See Chart 12.)

The annual growth rate in federal and state gov­ernment employment has generally subsided since the 1960s and 1970s. (See Chart 13.) However, the growth rate of state government employment has been positive for all but three years out of the past 39. Federal employment grew during the military buildup of the 1980s and during the military down­sizing after the collapse of the Soviet Union, which led to negative change rates in federal employment throughout the 1990s.

Conclusion

Public policy appears to be significant in the growth of the Index of Government Dependency. The rapid increase in the 1960s and 1970s coincides with a new commitment by the federal government to solve local social and economic problems that had previously been the responsibility of local govern­ments, civil society organizations, and families. The sum of government employees and the population covered by programs contained in this Index grew dramatically, even after accounting for the military buildup for the Vietnam War during the mid-1960s.

The 1980s and 1990s generally witnessed much slower growth in the Index. Indeed, if the period 1989 through 1993 had reflected the policies of the periods 1981–1988 and 1994–2001, the Index would have decreased in value. However, rather than fall, the Index appears to have resumed the growth rates attained during the Carter and George H. W. Bush Administrations.

We should be concerned about this seemingly relentless upward march in Index scores. Depen­dency on the federal government for life's many challenges strips civil society of its historical and necessary role in providing aid and renewal through the intimate relationships of family, com­munity, and local institutions and governments. While the Index does not measure the decay of civil society, it reflects its declining role in this most important aspect of life.

Our concern over the growth of the Index should be particularly high for another reason: We find ourselves on the eve of the largest retirement of peo­ple in world history at the same time that the num­ber of "taxpayers" who pay no taxes is growing steadily. Our republican form of government, with its finely balanced mixture of civil and political institutions and charitable roles, probably could withstand further but limited increases in the dependent population as defined in this paper.

Can it stand, however, against the swelling ranks of Americans who believe themselves entitled to public-sector benefits for which they pay few or no taxes? Are we completely indifferent to history's many examples of experiments in republican gov­ernment collapsing under the weight of just such a population? Are we near a tipping point in the nature of our government and the principles that tie it to civil life?

A fair reading of these trends and the data con­tained in this Index would lead almost inescapably to the view that we have reached that point.

William W. Beach is Director of the Center for Data Analysis at The Heritage Foundation. A number of policy personnel at The Heritage Foundation con­tributed significantly to this year's Index of Government Dependency. Heritage policy experts David C. John, Christine C. Kim, Dan Lips, Jennifer A. Marshall, Nicola Moore, Nina Owcharenko, Brian M. Riedl, and Ronald D. Utt contributed commentary on the policy elements. Margaret Hamlin managed the numer­ical components of the Index. Kimberly Remington and Patrick Tyrrell coordinated the process of updating the policy sections.