Social Security is the largest program in the federal budget, accounting for 23 percent of all spending. In 2013, Social Security outlays of $810 billion will far exceed outlays for the second-largest program, national defense, at about $650 billion.1
Social Security has two components: Old-Age and Survivors Insurance (OASI) and Disability Insurance (DI). OASI will cost $668 billion in 2013 and pay benefits to 47 million retirees and survivors, while DI will cost $142 billion and pay benefits to 11 million disabled people.2 This essay focuses on OASI, while a related essay discusses DI.
Social Security faces severe financial problems. The current "pay-as-you-go" structure of the program will be unsustainable as the number of elderly people receiving benefits soars in coming decades. Unless the program is restructured, Social Security will be only able to pay a fraction of promised benefits to future retirees. Furthermore, the rising costs of Social Security will damage the economy and create pressures to impose large tax increases on workers.
This essay explores Social Security's problems and describes reforms that would both improve federal finances and create a more secure retirement for future generations of Americans. The Cato Institute has developed a detailed reform plan that would turn Social Security into a system based on personal retirement accounts. Individuals would take ownership of a portion of their payroll contributions, which would be deposited in a savings account. With a partly privatized system, young people would be able to build a secure retirement nest egg free from the political risk of future benefit cuts.
Federal policymakers have discussed overhauling Social Security for years, but they have not acted on reforms. Meanwhile, many other countries have shown that privatized retirement systems based on personal accounts can benefit workers, retirees, and the overall economy. It's time that America phase out its old-fashioned Social Security system based on taxes and government control, and move to a system where individuals take charge of their own financial futures.
History of Social Security
President Franklin Roosevelt signed Social Security into law in 1935 after years of political debate regarding a government-run retirement system. The first old-age pension bill was introduced in Congress in 1909, and by the mid-1930s about half of the states had enacted government pension programs.3 Other nations had also introduced systems, including Germany in 1889 and Britain in 1908.4 By the time that the United States enacted Social Security, more than 30 countries in Europe, Latin America, and elsewhere had established such programs.5
Another precedent for Social Security was the expansive system of disability, survivors, and old-age benefits provided to Civil War veterans and their families. The pensions began modestly in 1862, but eligibility and benefit levels were repeatedly expanded in the late 19th century.6 Republicans were the main supporters of the growth in Civil War pensions, but both parties have been responsible for expanding Social Security.
Popular movements were created to lobby in favor of a Social Security system. In 1927, economist Abraham Epstein established a group called the American Association for Old-Age Security.7 In 1933, the group changed its name to the American Association for Social Security, which was the first significant use of the phrase "Social Security."8 In 1933, prominent doctor Francis Townsend proposed a national retirement plan for Americans aged 60 and older to be funded by a national sales tax, and the plan gained widespread popularity.9
In 1934, Roosevelt said that he wanted legislation "to provide at once security against several of the great disturbing factors in life," such as disability and old age, and he created a Committee on Economic Security to pursue the issue. In his annual address to Congress in 1935, Roosevelt called for legislation to provide aid for the unemployed, disabled, and the aged.10 The Social Security Act was signed into law by Roosevelt on August 14, 1935. The Supreme Court gave its constitutional blessing to the Act in 1937, calling the program "conducive to the general welfare."11
Social Security was originally quite modest. The initial taxes and benefits were fairly low, and the program only provided retirement benefits to covered workers. Also, it only covered a portion of the U.S. workforce, excluding workers in agriculture, state and local governments, nonprofit organizations, and others.
The program began with a 2 percent tax on wages, with half paid by the employer and half by the employee, although economists say that both halves of the tax have the effect of reducing workers' wages. The amount of wages subject to tax was capped at $3,000, making the maximum tax $60. Taxes were imposed beginning in 1937, while benefits were paid beginning in 1940.
Social Security did not remain a modest program. Many activists and politicians of the era dreamed of creating a cradle-to-grave welfare state, and they viewed the 1935 Social Security Act as just the beginning. In 1939, Congress added benefits for dependents of retired workers and surviving dependents of deceased workers. In 1956, Congress added benefits for disabled workers.
Congress also continuously expanded the workforce covered by the program. The share of all U.S. civilian workers covered by Social Security grew from 55 percent in 1939, to 90 percent by 1970, and to more than 95 percent today.12
Meanwhile, politicians sought favor with elderly voters by frequently boosting benefit levels. In 1967, President Lyndon Johnson signed amendments to Social Security that increased average benefits by 13 percent.13 In 1969, President Richard Nixon signed provisions that increased benefits by 15 percent.14 In 1971, Nixon signed another 10 percent across-the-board increase. Then a month before the elections in 1972, policymakers increased benefits across-the-board another 20 percent. At the time of this increase, Social Security beneficiaries received a letter in the mail telling them that the benefit increase was "enacted by Congress and signed into law by President Richard Nixon."15
In their expert history of Social Security, Sylvester Schieber and John Shoven discuss some of the reasons for the rapid expansion of the program.16 One reason was a good sales job by Social Security advocates: "The use of terminology of insurance and contributions is intended to skirt the perception that people are being ‘taxed' to finance benefits and to mask the huge welfare payments being made to many participants in the program."17 This was evident in a glowing 1939 article in Life magazine, which called Social Security an "insurance" program that would pay retirees an "annuity."18 The article included numerous poignant photos of people being helped by the benefits, but little mention of the costs.
Another reason for Social Security's political success was the creation of an expert bureaucracy in Washington that controlled the flow of information about the program and proactively agitated for an expansion in benefits. When Republicans criticized the program leading up to the 1936 elections, for example, the new Social Security Board distributed 50 million leaflets in favor of it and even released films to movie theaters that advertised the program to the public.19
Both the Social Security Board and the occasional Social Security Advisory Councils repeatedly pushed for expansion during the program's first four decades.20 The Advisory Councils were supposed to provide independent critiques of the program, but between 1937 and 1971 none of the six boards did. Instead, "most called for major expansion," notes one historian.21
Big businesses and labor unions were generally supportive of Social Security expansion. As for politicians, Social Security expansion was an easy sell in the early decades because they could offer ever higher benefits in the short term, while pushing off the costs in the form of unfunded obligations to the future. Schieber and Shoven explain that Congress "could regularly increase benefits without having to increase the payroll tax rate to do so. If Congress could act like Santa Claus, the Social Security actuaries were the elves that supplied them with gifts to distribute regularly to the voting public."22
By the mid-1970s, it was becoming clear that Social Security couldn't finance all the benefits that had been promised. In 1975, 1976, and 1977 the program ran cash-flow deficits for the first times since 1959. Deficits were projected to continue and the program's trust fund was expected to be exhausted by the early 1980s. Congress and President Jimmy Carter tried to solve the problem with an adjustment of benefits and large tax increases. Carter said that the changes would "guarantee that from 1980 to the year 2030, the Social Security funds will be sound."23
He was wrong. By the early 1980s, projections again showed that Social Security was headed for insolvency. In May 1981, President Ronald Reagan proposed a variety of benefit cuts for future retirees, but these proposals were fought back quickly by leaders of both parties on Capitol Hill.24 Even though Reagan's proposed trims to the program were very modest and entirely defensible, his political advisers got cold feet and backed down quickly.25
Reagan changed his strategy and established the bipartisan National Commission on Social Security Reform.26 The commission was chaired by Alan Greenspan and included representatives from Congress, business, and labor. By 1983, facing the possibility that checks might actually be delayed for millions of retirees, the commission proposed a set of changes including raising the retirement age from 65 to 67 over time, advancing scheduled payroll tax increases, increasing taxes on the self-employed, and taxing Social Security benefits above an income threshold. Supported by both President Reagan and House Speaker Tip O'Neill, Congress quickly passed the changes.
Once again, supporters pronounced Social Security fiscally sound, and the program's trustees predicted solvency until 2063.27 Once again, they were wrong. Despite the tax increases in the 1983 deal, it was apparent by the 1990s that Social Security's finances were deteriorating rapidly. President Bill Clinton launched an effort to study various proposals to fix Social Security, including proposals for personal retirement accounts. Unfortunately, the Monica Lewinsky scandal soured relations between the White House and Congress, and no reforms were forthcoming.
President George W. Bush came to office determined to reform Social Security. His first effort included a bipartisan commission that produced a report in 2001 with various options for adding personal accounts. Unfortunately, that effort was derailed by the terrorist attack on 9/11. Bush launched another reform effort during his second term. He discussed Social Security personal accounts during his 2005 State of the Union address, and his administration issued a study with detailed guidelines for reform. By this time, however, the president's credibility had been diminished by the war in Iraq, the poor federal response to Hurricane Katrina, and other failures. Also, many Republicans in Congress were too timid to support Bush's efforts. In the face of opposition from elderly lobby groups and labor unions, the Bush reform effort was shelved.
Social Security Basics
Social Security is the largest program in the federal budget, accounting for 23 percent of all spending.28 Social Security spending of $810 billion in 2013 is expected to far exceed spending of about $650 billion on national defense, which is the second-largest federal program. The Congressional Budget Office projects that Social Security spending will soar to $1.42 trillion by 2023, almost double the projected defense spending that year of $731 billion.29
Social Security has two components: Old-Age and Survivors Insurance (OASI) and Disability Insurance (DI). OASI will cost $668 billion in 2013 and pay benefits to 47 million retirees and survivors, while DI will cost $142 billion and pay benefits to 11 million disabled individuals.30
OASI and DI are funded mainly by a 12.4 percent payroll tax on wages up to a dollar cap, which is $113,700 in 2013 and is adjusted annually. Additional funding is provided by income taxes on some recipients of Social Security benefits and interest payments on Social Security trust fund assets.
Old-Age and Survivors Insurance (OASI)
Social Security has become the primary source of retirement income for most Americans. About 37 percent of all income of individuals age 65 or older comes from Social Security benefits.31 Almost three quarters of elderly Americans have no private pension, and almost half have no income from assets.32 But note that this lack of private retirement savings is substantially a result of the existence of Social Security. If Americans had not grown up within a welfare state, assuming that the government would pay for their retirement, most people would have saved much more.
The Social Security system is not based on savings. Rather, it is a pay-as-you-go retirement system funded by taxes. Unlike a private pension system, it is not backed by a pool of real assets available to pay future benefits. Instead, future benefits will be paid to the extent that tax revenue is available in coming years when benefit promises become due.
Also unlike a private pension plan, Social Security benefits have only a vague relationship to the contributions that workers made. In determining a person's benefits, the Social Security Administration (SSA) calculates the Average Indexed Monthly Earnings (AIME). All earnings upon which a person has paid Social Security taxes until the age of 60 are indexed for wage growth. The indexing formula is based on the ratio of the average national wage in the year the individual turns 60 to the year to be indexed. Wages earned after age 60 are not indexed. The SSA then takes the average of the 35 years in which wages are highest to determine the career average annual earnings; these wages are divided by 12 to arrive at the AIME.33
Then a formula is used to find the person's Primary Insurance Amount (PIA). For people reaching the normal retirement age of 66 in 2013, the formula is 90 percent of the first $791 of the AIME, plus 32 percent of the next $4,768 of the AIME, plus 15 percent of the remaining AIME.34 This creates a "progressive" slant to benefits such that lower-income workers get a higher share of their previous wages replaced than do other people.
However, while "Social Security is designed to provide low-income individuals with higher returns on their payroll taxes, their shorter expected life spans almost fully offsets the benefit formula's progressivity," notes Cato Institute's Jagadeesh Gokhale.35 Similarly, Harvard University's Martin Feldstein finds that the progressivity of the benefit formula "is offset by the longer expected life of higher income individuals, their increased use of spouse benefits, and the later age at which they begin to work and pay taxes."36
The normal retirement age for Social Security is currently 66 and will be rising to 67 in coming years. However, workers can retire as early as 62 and take lower monthly benefits, or they can retire later than 66 and receive higher monthly benefits. Social Security provides benefits to spouses of workers who have not worked long enough to be eligible for their own benefits. At age 62, such spouses will receive benefits equal to 50 percent of the worker's PIA.37
Social Security also provides benefits to the spouses of deceased workers. A surviving spouse over 60 receives benefits equal to 100 percent of the worker's PIA. If the widowed spouse is under 60, she is eligible for benefits equal to 75 percent of PIA if she is caring for children under age 16. Minor children of a deceased worker are eligible for survivors benefits equal to 75 percent of PIA. However, a family's total benefits are capped according to a mildly progressive formula.
Disability Insurance (DI)
Social Security Disability Insurance (DI) provides benefits to disabled workers, who are people with physical or mental impairments that prevent them from performing "substantial" work for at least a year. Earnings of $1,040 or more per month are generally considered substantial work, and the disability must be considered permanent.
Disability benefits are generally equal to the individual's PIA, computed as though the person was age 62 at the time of disablement. If the person is receiving workers compensation or other types of government disability benefits, the Social Security disability payment may be reduced.
The cost of the DI program has soared. The estimated 2013 spending of $144 billion is double the spending in 2003 of $72 billion.38 Disability Insurance is a very troubled program. It is riddled with fraud and waste, and the program's costs are rising rapidly even though the rate of actual disability for working-age adults in America has not risen.
The Looming Financial Crisis
The large baby boomer generation is beginning to retire in droves and average life spans in the nation are continuing to rise. As a result, the number of elderly people is projected to rise rapidly in coming years, which will push up the costs of Social Security and other entitlement programs. Unless policymakers cut Social Security and other programs, the fiscal and economic outlook for the nation looks grim.
When Social Security was created in 1935, the life expectancy for males at age 65 was less than 12 years. Today it is 18 years, and by 2045 it is expected to be 20 years.39 As the number of elderly people grows, the number of workers available to support them will not keep up. One reason is that birth rates have fallen in recent decades from about three children per woman in the 1960s to about two today.40 The Social Security trustees project that by 2030 the number of people age 65 and older will increase 70 percent while the number of working-age people (ages 20 to 64) will increase by just 6 percent.41
Those changing demographics are driving Social Security's financial imbalances. The imbalances are evident in projections of the program's "income rate" and "cost rate," which are the revenues and expenses of Social Security expressed as a percentage of taxable wages. In 2012, the income rate was 12.83 percent and the cost rate was 13.79 percent, which created a 7 percent deficit. By 2035, the income rate is projected to be 13.16 percent, while the cost rate is projected to be 16.98 percent, creating a 22 percent deficit.42
However, Social Security is not technically in deficit yet because it is credited with interest earnings on its trust fund, which are not included in the above income rate. When interest earnings are included, Social Security will begin running deficits by 2021.43 After that date, the program can continue paying benefits by drawing on the trust fund until it is exhausted in 2033. At that point, by law, Social Security benefits will have to be cut by 23 percent.44
Until recently, the SSA regularly notified workers of the pending cuts in benefits when the trust fund is exhausted. During the George W. Bush administration, the SSA's annual mailing of personal benefits statements to workers contained the disclaimer:
Your estimated benefits are based on current law. Congress has made changes to the law in the past and can do so at any time. The law governing benefit amounts may change because, by 2040, the payroll taxes collected will be enough to pay only about 74 percent of scheduled benefits.45
Unfortunately, this highly visible warning to workers about the program's financial problems was discontinued when the SSA stopped mailing paper copies of the personal benefits statements in 2011.
More important, trust fund accounting does not reveal the huge unfunded obligations that Social Security has created. The idea that the program is financially sound until the trust fund is exhausted in 2033 is a myth. Despite the government's use of the phrase "trust fund," it has long been recognized that there is no real pool of assets backing up Social Security. Even the glowing article on Social Security from Life magazine in 1939 admitted:
The most criticized feature of the Act has been its scheme for financing old-age insurance, under which it was planned to pile up a fantastic 47-billion-dollar reserve by 1980. The joker in this is that the government has been spending Social Security tax money for ordinary expenses and putting its own I.O.U.'s (i.e. bonds) in the reserve fund. Thus, when the time came to pay old-age annuities partly out of the interest on the bonds, the funds could be raised only by taxing the people a second time.46
Such concerns about building a large "reserve" (which was not a real store of wealth) led to Social Security amendments in 1939 that set the program on a pure pay-as-you-go basis. The trust fund became a small contingency fund or bookkeeping entry of no economic importance. The Clinton administration's federal budget in 2000 laid out the reality of the trust fund:
These [Trust Fund] balances are available to finance future benefit payments and other Trust Fund expenditures—but only in a bookkeeping sense. …They do not consist of real economic assets that can be drawn down in the future to fund benefits. Instead, they are claims on the Treasury that, when redeemed, will have to be financed by raising taxes, borrowing from the public, or reducing benefits or other expenditures. The existence of large Trust Fund balances, therefore, does not, by itself, have any impact on the Government's ability to pay benefits.47
A better way to measure the financial trouble facing Social Security is to compare the promised total future benefits to the program's total future taxes on a present value basis. Such calculations show that Social Security faces a staggering unfunded obligation of $23 trillion.48 This enormous figure is the amount that future benefit promises exceed future revenues, which makes it clear that the program is not sustainable in its current form.
Economic Harms of Social Security
Social Security faces long-term financial imbalances, but that is not the only reason why it should be restructured. The program's $23 trillion financing hole will be handled one way or another in the future. But Social Security also causes damage to the U.S. economy right now, and this damage is imposed whether or not the program faces any long-term financing troubles.
Social Security creates labor market distortions. The 12.4 percent payroll tax puts a large wedge between what employers are willing to pay for workers and the after-tax wage received by those workers. By increasing the cost of hiring workers, the demand curve for workers in the economy will shift down, fewer workers will be hired, and economic growth will reduced. Depending on labor market conditions, workers suffer from some combination of reduced after-tax wages and fewer employment opportunities.
One way that economists measure the costs of such tax distortions is to calculate "deadweight losses." These are the overall costs to society incurred when taxes interfere with efficient decisionmaking. Deadweight losses rise rapidly as marginal tax rates rise, in particular rising with the square of the marginal tax rate. This is an important issue with respect to payroll taxes because payroll taxes are piled on top of federal and state income taxes on wages. When considering both payroll and income taxes, tens of millions of middle-income families face marginal tax rates on earnings in the 30 to 40 percent range.
In a recent study, Social Security expert Jagadeesh Gokhale calculated that the average marginal federal and state tax on wages is about 35 percent.49 With that figure, he estimated that the deadweight losses imposed by the Social Security payroll tax was $161 billion in 2012. Thus, in transferring $603 billion in payroll taxes from workers to retirees in 2012, the program had the side effect of causing $161 billion of collateral damage as a result of labor market distortions.
Harvard's Martin Feldstein has also made estimates of the damage caused by payroll taxes. He calculates that every dollar of increased revenue from raising payroll tax rates causes about 50 cents of added deadweight losses.50 So let's say that Congress was considering raising the Social Security payroll tax from 12.4 percent to 14.4 percent this year. That increase would not just hit workers with a tab of $117 billion, it would also cause $59 billion of added damage to the economy from labor market distortions.51
The deadweight losses caused by taxes are a key shortcoming of pay-as-you-go government retirement systems. By contrast, retirement systems based on private savings can reduce or eliminate this cost. With retirement accounts that individuals own, workers perceive that their contributions create a direct value to them. As a result, the contributions generally do not reduce peoples' willingness to work as payroll taxes do, and thus labor markets are not distorted.
Another distortion caused by Social Security is that it encourages workers to retire earlier than they otherwise would. Labor force participation rates for men in their early 60s have plunged in the United States and other countries in recent decades. These drops in work effort appear to be partly caused by the growth of government retirement systems, particularly in those countries that pay benefits starting at an early age.52
In an American Economic Review paper, Jonathan Gruber and David Wise summarized their cross-country examination of the effects of Social Security on retirement: "One explanation for the striking decline in labor-force participation is that social-security provisions themselves provide enormous incentive to leave the labor force early, thus by their very structure exacerbating the financial problems that they face."53 They note that "there is a strong correspondence between the age at which benefits are available and departure from the labor force."54
Martin Feldstein and Jeffrey Liebman made similar conclusions after examining cross-country studies: "[I]n country after country, relaxation of early retirement rules and expansion in benefits available at younger ages were followed quickly by trends toward early retirement."55 In the United States, this early retirement effect may be muted because Social Security benefits are adjusted based on the age of retirement.
Nonetheless, to the extent that Social Security induces early retirement, it worsens the system's demographic problems because it means that fewer workers will be producing the output and paying the taxes needed to support the growing numbers of retirees. As such, federal policymakers should be looking for ways to reform Social Security and the tax code to reduce disincentives for people to continuing working into their 60s. After all, Americans in their 60s are increasingly healthy and have a wider range of non-physical work open to them in the modern economy than in the past.
Perhaps the most important economic distortion caused by Social Security is that it substantially reduces personal savings.56 Time-series studies by Martin Feldstein have "consistently found that Social Security crowds out a significant share of overall private saving."57 The reduction in saving results in a lower capital stock in the economy, which suppresses productivity and wages. Savings are the seed corn of growth in the economy, and so to the extent that Social Security reduces it, we are all worse off.
Unfortunately, the growth of the welfare state over the last eight decades has created large disincentives for personal savings. Generations of Americans have grown up assuming that the government will take care of them when they are sick, unemployed, and retired. They have responded by consuming more, and by putting less effort into planning and saving for rainy days and their old age. The expansion of government has encouraged people to become spendthrift and dependent.
Transforming Social Security into a system of individual savings accounts would give Americans more responsibility and control over their own financial futures. It would get people into the habit of saving, while engaging them in active planning for their own retirement. Ownership encourages individual responsibility.
Many Americans believe that Social Security is an "earned right." They think that because they have paid Social Security taxes they are entitled to receive Social Security benefits. The government encourages this belief by calling payroll taxes "contributions," as in the Federal Insurance Contributions Act.
However, in the 1960 case Flemming v. Nestor, the U.S. Supreme Court ruled that workers have no contractual right to their Social Security benefits: "A person covered by the Social Security Act has not such a right in old-age benefit payments … To engraft upon the Social Security system a concept of ‘accrued property rights' would deprive it of the flexibility and boldness in adjustments to ever-changing conditions which it demands and which Congress probably had in mind when it expressly reserved the right to alter, amend or repeal any provision of the Act."58 Thus, Congress can cut benefits any time that it chooses.
The upshot is that Social Security has turned older Americans into supplicants, dependent on the political process for their retirement benefits. If they work hard and pay Social Security taxes their entire lives, they earn the privilege of going hat in hand to the government to get some of their money back in retirement. But because politicians have been irresponsible with Social Security's finances, future retirees under the current program will receive a poor return on their lifetime of tax payments.
By contrast, under a retirement system based on individual savings accounts, workers would have full property rights to their savings. They would own their accounts and the money in them the same way that they currently own their individual retirement accounts (IRAs) and 401(k) plans. With Social Security privatization, retirement benefits would not depend on the whims of politicians.
Options for Reform
When considering Social Security's unsustainable finances, there are a few basic ways of tackling the problem. As former president Bill Clinton pointed out, policymakers can: a) raise taxes, b) cut benefits, or c) get a higher rate of return through private investment.59
The first option has been used repeatedly over the last eight decades to temporarily prop up the program. The initial Social Security tax in 1937 was 2 percent on the first $3,000 of wages for a maximum tax of $60, which is equal to about $1,000 today. In 2013, the tax is 12.4 percent and the cap is $113,700, for a maximum tax of about $14,000. Thus, even after adjusting for inflation, today's maximum tax is about 14 times larger than it was originally.
If policymakers continue "fixing" Social Security in the future with tax increases, it would be very harmful to the economy. As noted, the damage (deadweight losses) caused by tax increases rises more than proportionally as tax rates rise. Since tens of millions of middle- and higher-income workers already face high marginal tax rates on their wages, future increases in payroll taxes would be even more damaging than past increases.
Projections from the Social Security trustees indicate how high the payroll tax would have to rise to match promised future benefits. Restoring the program to solvency would require raising the rate from 12.4 percent to 16.6 percent, which would be a 34 percent increase.60 If that tax increase were put in place in 2013, it would be like raising the already huge payroll tax burden of $723 billion to $969 billion.
However, the situation is worse than that because as tax rates rose, the tax base would shrink because people would reduce their productive activities and increase their tax avoidance.61 That shrinking of the tax base would mean that the government would have to raise tax rates even higher than they planned in order to fund rising Social Security benefits.
Another problem is that at the same time that policymakers would be pursuing payroll tax increases for Social Security, they may also decide to raise payroll and income taxes to fund Medicare, which also faces a huge financing gap. The combined effect of such huge tax increases would dramatically shrink wage and income tax bases, perhaps prompting policymakers to raise tax rates even higher. The U.S. economy would fall into a fiscal death spiral of rising rates and reduced economic growth.
Tax increases are not the answer to solve America's problem of overpromised elderly entitlement programs. Federal taxes are already too high. Instead, a much better solution for Social Security would be to cut the growth in benefits. There are a number of ways to do that, as the Congressional Budget Office has described:
- Change the COLA. The SSA provides a cost of living allowance (COLA) to Social Security recipients each year based on a measure of inflation. However, most economists think that the current measure of inflation is too high. Switching to a more accurate measure would reduce the annual COLA by about 0.25 percent for an average year. The CBO finds that such a reform would reduce Social Security spending by about 3 percent relative to current policies by 2050, or from about 5.9 percent of GDP to 5.7 percent.62
- Raise the Normal Retirement Age. The normal retirement age (NRA) for Social Security is currently 66, but it will be rising to 67 in coming years. In 1940, the average life expectancy at age 65 was 11.9 years for men and 13.4 years for women.63 Today men age 65 are expected to live 17.9 years and women 20.4 years. Thus, while retirees are expected to live six or seven years longer than when Social Security was established, policymakers have shifted up the NRA just two years. A CBO option to raise the NRA to 70 in coming decades would reduce Social Security spending by 12 percent relative to current policies by 2050, or from 5.9 percent of GDP to 5.2 percent.64
- Raise the Early Retirement Age. Raising the early retirement age (ERA) from 62 would not reduce Social Security spending in the long run. However, raising the ERA would have the advantage of helping to boost economic growth by encouraging skilled and experienced workers to stay in the workforce longer.
- Price Indexing of Initial Benefits. A retiree's initial level of Social Security benefits is based loosely on their earnings history, and then indexed by the SSA for the economywide growth in nominal wages over the years. An alternative calculation method would be to adjust past earnings only by inflation, which would exclude the increases stemming from real growth. The CBO finds that this reform would reduce Social Security spending by 29 percent relative to current policies by 2050, or from 5.9 percent of GDP to 4.2 percent.65 This single reform would bring Social Security into permanent long-run solvency by slowly and steadily trimming the growth in benefits over time.
- Partial Price Indexing. An alternative to full price indexing would be partial or "progressive" price indexing, which would reduce the growth in benefits for middle- and upper-income workers. The benefits for lower-income workers would not be changed, and so the benefit structure of Social Security would be flattened over time. The CBO finds that partial price indexing that left the bottom 30 percent of workers unaffected would reduce Social Security spending by 18 percent by 2050, or from 5.9 percent of GDP to 4.8 percent.
Any of these changes would move Social Security in the direction of financial solvency, and thus reduce pressures to pursue damaging tax hikes.
However, individual retirement security would still be left in hands of politicians, and people would not have the security of ownership of their own retirement assets. Simple benefit cuts would also not solve the labor market distortions created by Social Security or reduce the anti-savings effect of the program. So cutting the growth in benefits would be a good reform step, but it wouldn't be as beneficial as moving to a retirement system based on personal savings accounts.
Personal accounts have long been discussed as a superior alternative to the current pay-as-you-go Social Security system. They would give workers the chance to earn higher returns on their contributions to the system and allow them to offset the benefit reductions that will occur under current law as the system's finances deteriorate. Personal accounts would also encourage greater work effort and thus boost economic growth.
More than two dozen countries have moved to retirement systems based on personal accounts. Chile led the privatization trend with reforms in 1981 that switched from a pay-as-you-go-system based on taxes to a system based on individual accounts funded by contributions of 10 percent of worker wages up to a cap.66 The tax treatment of Chile's accounts is similar to U.S. 401(k) plans, whereby the contributions are on a pre-tax basis, earnings compound tax-free, and withdrawals are taxed at retirement. Individuals can choose between competing private firms to manage their accounts.
Upon retiring, workers in Chile may convert their accounts into an annuity, take planned withdrawals, or do a combination of these two distribution options. Excess account assets above a defined threshold may be withdrawn as a lump sum. The government provides a safety net for workers who have not accumulated sufficient assets for a defined minimum pension amount.
The Chilean system has been very successful. For one thing, the dramatic increase in private savings has had a very beneficial effect on economic growth in what has become South America's most prosperous nation.67 Chile's reforms have inspired many other nations to switch to retirement systems mainly or fully based on private accounts, including Australia, Mexico, Sweden, Poland, Latvia, Peru, and Uruguay. For further reading on Social Security privatization around the world, see these Cato essays.
The Cato Institute proposed a detailed plan for reforming and partly privatizing Social Security in 2004.68 The actuaries at the Social Security Administration examined the Cato plan and determined that it would restore the program to long-term solvency.69 As such, the proposal remains a good starting point for considering major restructuring of Social Security today. Full details of the Cato plan are available in this report.
Under the Cato plan, Social Security would not be changed for current recipients or workers age 55 and older. Workers younger than age 55 would have the option of remaining in the current system or entering a new personal account system. If they remained in the current system, they would continue paying the 12.4 percent payroll tax and receive a level of retirement benefits payable with revenues available under current law. As such, the growth in benefits would need to be reduced over time to match those revenues, which could be accomplished by switching to price indexing for benefits beginning in 2020.
The other option would be a personal account system. Workers would be allowed to divert half of their payroll taxes (6.2 percentage points) to a retirement savings account that they fully owned. In return, they would agree to forgo their future accruals of retirement benefits under traditional Social Security. The remaining 6.2 percentage points of payroll taxes would be used to finance the transition to the new system and to pay for survivors and disability benefits. Once the transition was complete, this portion of the payroll tax would be reduced to the level needed to pay just survivors and disability benefits.
Workers choosing personal accounts would receive a zero-coupon bond in recognition of their past tax payments to Social Security. The bond would be calculated to be able to pay the accrued benefits up to the date that the individual chose the personal account option. The bond proceeds would be deposited in workers' accounts until they became eligible to make withdrawals.
Contributions to individual accounts would be on a post-tax basis. Interest, dividend, and capital gain earnings on investments within individual accounts, and eligible withdrawals from the accounts, would be exempt from income taxes. Thus, the tax treatment of Social Security accounts would be similar to Roth IRAs.
One advantage of personal Social Security accounts is that they would help encourage a general savings habit among workers. Workers would have the option of depositing up to an additional 10 percent of earnings into their accounts. These voluntary contributions would have the same tax treatment as the mandatory portions of personal accounts.
Funds deposited in individual accounts would be invested in real private-sector assets under a three-tier process:
- Tier I. Collection of personal account contributions would be handled by the employer, as are payroll taxes today. Employers would send Social Security contributions to the U.S. Treasury, telling it how much is from employees who have chosen personal accounts. The Treasury would transfer that portion to a private-sector custodian bank, which would invest it in a money market fund. The following year, the contributions would be reconciled to individuals' names using W-2 forms, and the amounts with interest would be distributed to the default accounts for individual workers under Tier II.
- Tier II. Workers would initially have a choice of three diversified investment options. The default portfolio—where funds are placed if no other choice is made—would hold stocks and bonds in a 60/40 percent weighting. The other two funds would hold stocks and bonds at different weights. Younger workers could choose a fund with more stocks, while older workers could choose a fund with more bonds.
- Tier III. Once workers had accumulated a certain level of funds, they could participate in a larger range of investments, similar to choices under 401(k) plans. The financial companies managing funds under Tier III may choose to offer additional services, such as retirement planning software, to attract assets from Tier II. Workers would be free to allocate their savings among different competing companies and funds. Administration costs would be likely greater under Tier III than Tier II, but people could leave their funds in Tier II plans if they wanted.
At retirement, workers would be able to choose between an annuity, a programmed withdrawal, or a combination of the two. The programmed withdrawal would be based on twice the person's life expectancy. If a person choosing this option died before the account assets were exhausted, the remaining assets would be inheritable. Workers who choose to purchase an annuity that provided at least annual income equal to 120 percent of the poverty level could withdraw any additional funds as a lump sum.70
Finally, the federal government would provide a safety net insuring that no worker's retirement income fell below the poverty level. Workers whose private account assets were insufficient to purchase an annuity of that size would receive a subsidy to enable them to purchase it. This safety net would be funded through general federal revenues.
Moving to a partially privatized system would save money in the long run, but the transition would need to be financed. That is because when workers choose the personal account option, payroll tax revenues would be redirected from the payment of current benefits to being saved in the accounts. This near-term funding gap would be a one-time event, and financing it would have the positive effect of reducing the government's long-term obligations for the program. The transition to personal accounts would move the government's need to close the funding gap forward in time, but that would be good thing because we can either "pay a little now or pay a lot later."
In sum, it is very unfair for policymakers to continue delaying needed reforms, and to essentially impose on young people the need to "pay a lot later." We know that the current structure of Social Security is unsustainable, and we know that payroll taxes and the anti-savings effects of the program are damaging to the economy. We also know that privatization of government retirement systems works, as more than two dozen nations have demonstrated. So it is time to let young people take control of their financial futures and give them the freedom to build secure retirements with Social Security personal accounts.
1 Congressional Budget Office, "The Budget and Economic Outlook: Fiscal Years 2013 to 2023," February 2013.
2 Congressional Budget Office, "The Budget and Economic Outlook: Fiscal Years 2013 to 2023," February 2013. See the supplementary tables.
5 Peter Ferrara and Michael Tanner, A New Deal For Social Security (Washington: Cato Institute, 1998), p. 13.
6 Claudia Linares, "The Civil War Pension Law," Center for Population Economics, University of Chicago, November 2001. See also Ivan Eland, "From War to Warfare," The American Conservative, February 27, 2013.
9 Daniel Beland, History and Politics from the New Deal to the Privatization Debate (Lawrence, KS: University Press of Kansas, 2005), pp. 66–67.
10 Daniel Beland, History and Politics from the New Deal to the Privatization Debate (Lawrence, KS: University Press of Kansas, 2005), p. 72.
11 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 48.
12 House Ways and Means Committee, "2000 Green Book," October 6, 2000, p. 47.
15 Edward R. Tufte, Political Control of the Economy (Princeton: Princeton University Press, 1978), 29–33.
16 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999).
17 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 8.
18 "Social Security," Life, August 7, 1939.
19 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 54.
20 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), Chapters 8 and 9.
21 Martha Derthick, quoted in Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 140.
22 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 154.
23 Quoted in Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 182.
24 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 187.
25 David Stockman, The Great Deformation (New York: Public Affairs, 2013), pp. 99–102.
26 Daniel Beland, History and Politics from the New Deal to the Privatization Debate (Lawrence, KS: University Press of Kansas, 2005), 152.
27 Sylvester J. Schieber and John B. Shoven, The Real Deal: The History and Future of Social Security (New Haven: Yale University, 1999), p. 197.
28 Congressional Budget Office, "The Budget and Economic Outlook: Fiscal Years 2013 to 2023," February 2013.
29 Congressional Budget Office, "The Budget and Economic Outlook: Fiscal Years 2013 to 2023," February 2013.
30 Congressional Budget Office, "The Budget and Economic Outlook: Fiscal Years 2013 to 2023," February 2013. See the supplementary tables.
31 Social Security Administration, Office of Research, Evaluation, and Statistics, "Income of the Population 55 or Older, 2010," March 2012, Table 10.1.
32 Social Security Administration, Office of Research, Evaluation, and Statistics, "Income of the Population 55 or Older, 2010," March 2012, Table 8.A1.
33 Dawn Nuschler, "Social Security Primer," Congressional Research Service, R42035, April 2012.
35 Jagadeesh Gokhale, "Social Security Reform: Does Privatization Still Make Sense?" Harvard Journal on Legislation 50, no. 1 (Winter 2013): 187.
36 Martin Feldstein, "Rethinking Social Insurance," National Bureau of Economic Research Working Paper no. 11250, March 2005, p. 4.
37 Dawn Nuschler, "Social Security Primer," Congressional Research Service, R42035, April 2012, Table 3.
38Budget of the U.S. Government, Fiscal Year 2014, Historical Tables (Washington: Government Printing Office, 2013), Table 13.1.
39 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 31, 2013, p. 92.
40 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 31, 2013, p. 86.
41 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 31, 2013, p. 89.
42 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 31, 2013, p. 50.
43 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 2013, p. 44.
44 Social Security Board of Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May, 2013, p. 11.
46 "Social Security," Life, August 7, 1939.
47Budget of the U.S. Government, Fiscal Year 2000, Analytical Perspectives (Washington: Government Printing Office, 1999), p. 337.
48 Social Security Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 2013, p. 17. This is an "infinite horizon" estimate.
49 Jagadeesh Gokhale, "Social Security Reform: Does Privatization Still Make Sense?" Harvard Journal on Legislation 50, no. 1 (Winter 2013): 177–79.
50 Martin Feldstein, "Rethinking Social Insurance," National Bureau of Economic Research Working Paper no. 11250, March 2005, p. 19.
51 Old-Age and Survivors Insurance (OASI) and Disability Insurance (DI) payroll taxes in calendar 2013 are estimated to be $723 billion.
52 Jonathan Gruber and David Wise, "Social Security and Retirement: An International Comparison," American Economic Review 88, no. 2 (May 1998): 161.
53 Jonathan Gruber and David Wise, "Social Security and Retirement: An International Comparison," American Economic Review 88, no. 2 (May 1998): 158.
54 Jonathan Gruber and David Wise, "Social Security and Retirement: An International Comparison," American Economic Review 88, no. 2 (May 1998): 162.
55 Martin Feldstein and Jeffrey B. Liebman, "Social Security," National Bureau of Economic Research Working Paper no. 8451, September 2001, p. 46.
56 Martin Feldstein, "Rethinking Social Insurance," National Bureau of Economic Research Working Paper no. 11250, March 2005, p. 35.
57 Martin Feldstein and Jeffrey B. Liebman, "Social Security," National Bureau of Economic Research Working Paper no. 8451, September 2001, p. 40.
59 William Jefferson Clinton, "Remarks by the President via Satellite to the Regional Congressional Social Security Forums," July 27, 1998.
60 Social Security Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 2013, p. 64.
61 The Social Security trustees appear to take into account only a small and partial behavioral effect of rising tax rates. See Social Security Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 2013, p. 63, endnote 1.
62 Congressional Budget Office, "Reducing the Deficit: Spending and Revenue Options," March 2011, p. 58. This option would switch from CPI-W to chained CPI-U.
63 Social Security Trustees, "The 2013 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds," May 2013, p. 92.
64 Congressional Budget Office, "Reducing the Deficit: Spending and Revenue Options," March 2011, p. 63.
65 Congressional Budget Office, "Reducing the Deficit: Spending and Revenue Options," March 2011, p. 60.
66 For details on Chile's reforms, see www.josepinera.com. See also Ian Vasquez, "Testimony on Chile's Pension System to the Subcommittee on Social Security, House Committee on Ways and Means," June 16, 2005.
67 Jose Pinera, "Retiring in Chile," New York Times, December 1, 2004.
68 Details of the plan are in Michael Tanner, "The 6.2 Percent Solution: A Plan for Reforming Social Security," Cato Institute Social Security Paper no. 32, February 17, 2004.
69 See Michael Tanner, "A Better Deal at Half the Cost: SSA Scoring of the Cato Social Security Reform Plan," Cato Institute Briefing Paper no. 92, April 26, 2005.
70 And note that workers could opt out of the personal account system at any time that they had accumulated enough savings to purchase an annuity equal to 120 percent of the poverty level.
by David Seif, Harvard University
A common argument is that investment-based Social Security reform will improve economic efficiency by increasing the perceived link between retirement contributions and retirement benefits. Under this argument, individuals perceive the monies paid into the Social Security system as a pure tax. They fail to recognize that payments to Social Security will generally increase their future Social Security benefits. With personal retirement accounts, in contrast, the link between contributions and future income would be clear, and the economic distortions would be reduced.
My dissertation explores this issue by studying individuals’ labor supply response to two forms of payments made to Social Security, OASDI taxes and the earnings test. All workers in Social Security-eligible work pay OASDI taxes on income up to an annual limit, while the earnings test in its current form reduces the benefits of certain young Social Security recipients in years that their labor income exceeds a certain level.
The data for my dissertation are from the Health and Retirement Study (HRS), a longitudinal survey, much of which can be linked to Social Security earnings records. The HRS is conducted in biannual waves, with the first in 1992 and the most recent from 2004. It includes 30,207 individuals who are observed at least once, and 129,553 person × wave observations, of which nearly 70 percent are of individuals aged 55-80.
To examine the earnings test’s effect on labor supply, I exploit the 2000 elimination of the earnings test for individuals over the Full Retirement Age (FRA) with both a difference-in-difference-in-difference (DDD) regression approach and by fitting a Cox proportional hazard model. In 2000, the earnings test was eliminated for anyone who had attained the FRA in a previous calendar year, and it was vastly reduced for anyone older than the FRA who reached the FRA that same calendar year. The 2000 change provides a natural experiment with which I can compare two cohorts—one that was affected by the change in the earnings test, and another that reached the FRA too early to be affected by it.
To study the effect of OASDI taxes on labor supply, I, along with co-authors Jeffrey Liebman and Erzo F.P. Luttmer, exploit complexities in Social Security rules that lead to a large number of discontinuities in marginal Social Security benefits from additional work. These discontinuities are used to estimate the effect of tax-benefit link on both the extensive labor supply (retirement) and intensive labor supply (earnings and hours worked) margins.
Back to 2008 Dissertation Fellowship Program Recipients