Social Security

Medicare was implemented in 1965 to cover the medical costs of the oldest members in society.  In 1965,  the U.S. life expectancy was only 70 years old.  Now, however, life expectancy at birth is over 78 years.  Medicare is now not just covering the oldest of the old, it also covers the “moderately” old since we are living so much longer.

An NBER working paper by  John B. Shoven, Gopi Shah Goda examines what eligibility ages for programs such as Medicare and Social Security would be today and in 2050 if adjustments for mortality improvement were taken into account.  The authors conclude the following:

“We find that historical adjustment of eligibility ages for age inflation would have increased ages of eligibility by approximately 0.15 years annually. Failure to adjust for mortality improvement implies the percent of the population eligible to receive full Social Security benefits and Medicare will increase substantially relative to the share eligible under a policy of age adjustment.”

• John B. Shoven, Gopi Shah Goda (2008) “Adjusting Government Policies for Age Inflation” NBER WP #14231.

If you were offered an actuarially fair lump-sum payment, would you give up half of your Social Security benefits? This is the question asked by Brown, Casey and Mitchell in their 2008 NBER working paper.

Overall, about 60% of respondents from the HRS data set preferred the lump-sum payment. The authors find the following individuals are more likely to prefer a lump-sum payment over the annuity:

• those with shorter expected longevity or who are in poor health,
• those with more education,
• those with less financial sophistication conditional on education,
• those who believe Social Security benefits will be cut.

Predictably, individuals who think they live longer will choose the annuity since they will get paid over a longer period of time. We also see that those who believe that there is significant political risk (i.e., Social Security benefits will be cut) are more likely to choose the lump-sum benefits.

Men aged between 63 and 67 have only 5% of their assets in private annuities. People have claimed that this was due to one of the following reasons:

• Adverse selection leads to high load factors on annuities, making them a poor value.
• Because they are made up of a fixed payment, Annuities are subject to inflation risk.
• Social Security may be a substitute for private insurance annuities.

The study finds evidence to contradict all three of these hypotheses. Social security benefits are actuarially fair, and inflation-adjusted. If people really benefited from Social Security annuities, than they should be hesitant to give up this stream of payments. Yet three of five people still would prefer a lump sum payment over the Social Security annuity, a type of annuity that offers significant advantages over those offered in the private market.

Maybe Social Security isn’t as valuable as once thought.

• Brown JR, Casey MD, Mitchell, OS. 2008. “Who Values the Social Security Annuity? New Evidence on the Annuity Puzzle,”NBER WP #13800.

There is an interesting pair of blog posts by Ezra Klein and Andrew Sullivan.  Mr. Klein advocates a more centralized health care system while Mr. Sullivan is opposed to expanding the government’s role in health care.

Who do you agree with?

Over the past two decades, the share of total Social Security spending accounted for by Social Security Disability Insurance (DI) has increased from 10 percent in the 1970s to 17 percent today. In 2005, cash payment to DI beneficiaries topped $85 billion. Authors David Autor and Mark Duggan try to explain this phenomenon in their 2006 working paper.

The researchers main conclusion are:

• Most of the extra cost of the DI program is due to the liberalization of screening process that occurred due to a 1984 law. This law gives more weight to the applicants pain and discomfort and more credence to medical evidence from the applicant’s physician. In recent years, nearly 40 percent of total DI awards were granted during the appeals process, up from 20 percent in the late 1970s.
• The value of DI has increased. While the real dollar value of DI has not increased significantly over time, income inequality has increased. Thus, for the individuals on the bottom on the income distribution, DI now appears relatively more attractive since real wages have declined in the lowest income deciles.
• The aging U.S. population has not had a significant impact on the increase in DI rolls. The authors claim that the demographic shift towards more older individuals in society only accounts for 6% of the increase in DI.

It is always difficult to determine whether an increase in the rolls of a welfare program (such as DI) is ‘good’ or ‘bad’ for society. The population which makes up the increase is likely made up of deserving and undeserving individuals. Further, many individuals are partially disabled. What if a factory worker is injured and can only work 5 hours/week? Are they disabled? What if they can only work 10 hours? 20 hours? It is difficult to know where to draw the line.

In an earlier paper (”Rise in Disability“), the authors use adverse demand shocks between 1984 and 1998 and find that “a growing fraction of discouraged and displaced workers are seeking DI benefits,” suggesting that a higher percentage of ‘undeserving’ individuals may be going on the DI rolls.  The question society must ask itself is how much money will be ‘wasted’ on the inevitable phony DI claims in order to ensure that those truly disabled are able to received adequate DI benefits.
Two suggestions made to reform the system are to make it easier for individuals to obtain health insurance so DI is not serving as the insurer of last resort and, secondly, to introduce graduated DI payments depending on the severity of the injury.

• Much of the information used for this summary came from the NBER Fall 2006 Bulletin on Aging and Health.
• Autor and Duggan; (2006) “The Growth in the Social Security Disability Rolls: A Fiscal Crisis Unfolding” NBER Working Paper No. 12436.
• Autor and Duggan; (2003) “The Rise in Disability Rolls and the Decline in Unemployment,” Quarterly Journal of Economics, 118:1, pp. 157-206.

Recently I came across a paper by John Williamson describing a Notional Defined Contribution Model as a replacement for state pension system. Countries who have adapted the Notional Defined Contribution (NDC) model include Sweden (1994), Italy (1995), Latvia (1996), the Kyrgyz Republic (1997), Poland (1999), and Mongolia (2000). The model is in essence a pay-as-you-go framework, but benefits are closed linked to contributions. Policymakers hope this will encourage labor force participation.

Individual have an account which details how much money (in taxes) they have paid into the system. This account, however, is entirely virtual in that individuals do not have access to the fund in these accounts. The government chooses a “virtual” rate of return usually based on economic factors such as price inflation (e.g.: in Italy and Latvia) or a mix of wage and price inflation (e.g.: in Poland and Sweden).

Those in favor of this plan claim that since benefits are mostly tied to contributions, this may encourage labor force participation during a person’s working years; on the other hand the NDC model is less redistributive than a guaranteed benefit pension program. There is no financial market risk since the accounts are contemporaneously paid out from the current labor force to current retirees, but there is demographic risk. If a country has few workers and many retirees, the “virtual” rate of return must be lowered to account for this. The NDC plan has lower administrative cost than one in which individuals have access to their accounts because of economies of scale and lower transaction costs, but there is no incentive for individuals to increase their savings.

To me, this plan sounds exactly like the current U.S. Social Security system. Every worker has an account with their Social Security Contributions to date. These funds, however, are not owned by the individual but are paid contemporaneously to retirees. Upon retirement, an individual receives a Social Security pension based on the value in the account. (This is a simplification since Social Security payments are subject to a cap, and the payments also depend on the number of working years).

A more sensible plan is the Chilean system of personal accounts. According to the Cristian Science Monitor, (”In Britain and Chile“):

“The second and main pillar is the obligatory monthly payroll deduction of 12.3 percent. Ten percent goes into the worker’s own account, administered by one of six private pension funds, while 2.3 percent covers administrative fees. Unlike in the US, the payroll tax is funded entirely by the employee. At retirement - age 60 for women, 65 for men - they take out what they put in, plus accumulated gains. Currently 3.6 million Chileans, or 65 percent of the 5.5 million-person workforce, are actively contributing under this system.”

There are a few caveats. The Cato Institute (”Empowering workers“) notes, “This percentage applies only to the first $22,000 of annual income. Therefore, as wages go up with economic growth, the ‘mandatory savings’ content of the pension system goes down.” Also, if an individual works at least twenty years, they are eligible for a state-sponsored minimum pension. Thus, individuals will not be destitute in old-age, even workers with lower wages over their lifetime.

The privatization of a portion of the state pension program was less successful in Britain, however. “Pension sellers, working on commission, frequently coaxed… [workers] into unsuitable products, often overstating the potential for returns.” Still, keeping retirement savings in the hands of individuals will lead to more savings and ensure sufficient funds at retirement for all individuals; especially when paired with a minimum state-provided benefit.

The future of Social Security is in question. Even Federal Reserve chairman Ben Bernanke warns of the rapidly approaching Social Security “fiscal crisis.”. Individuals at the beginning or middle of their prime working years are unsure of how large (or small) their Social Security benefits will be when they retire.

An NBER working paper by Gomes, Kotlikoff and Viceira (WP #12859) aims to calculate the cost of this uncertainty due to government indecision regarding Social Security benefits. Below I describe the model the authors derive, briefly explain the empirical section of the paper and state their conclusions.

Model

Individuals have constant relative risk aversion (CRRA) utility functions of the form:

• U=C^1-γ(1-γ)^-1 (1)

The agents learn at time L whether the social security benefits they will receive in retirement will be large, B, or small, b. A_L are the assets accumulated at time L. T represents the last year of one’s life and R represents the year of retirement. Optimal consumption is given by:

• A_L+B(T-R)=C_B(T-L), with a large retirement benefit; and (2a)
• A_L+b(T-R)=C_b(T-L), with a small retirement benefit. (2b)
• A_L+=A₀-CL (3)

Expected Utility and first order conditions are given by:

• EU= (1-γ)^-1{ C^1-γL + (T-L)[pC_B^1-γ + (1-p)C_b^1-γ] } (4)
• FOC: C^-γ= + pC_B^-γ + (1-p)C_b^-γ (5)

One can solve equations (2a) and (2b) for C_B and C_b respectively, plug equations (2a), (2b) and (3) into (4), and take the derivative with respect to C to arrive at equation (5). The authors aim to estimate how the agents’ expected utility changes when the date (L) when the government reveals the whether the benefit level is high (B) or low (b) changes. More formally, the authors find that ¶EU/¶L < 0 when the individual is risk averse (i.e.: γ>1). In words, expected utility decreases when the date the individual is informed of the social security benefits is moved further into the future.
The authors model income as a concave, quadratic function of age which has an error term which is MA(1) with probability π and ln(0.1)»-2.3 with probability (1-π). Thus, income is hump-shaped throughout one’s lifetime with some persistence in the idiosyncratic error term. Also, there is some probability of a negative income shock, which can be thought of as the probability an agent loses their job. The individual invests their savings optimally at each age, dividing their assets between stocks and a risk-free asset.

The authors then calibrate their model using parameter estimates from other sources. The authors conclude that the excess burden of government indecision can be as high as 0.6 percent of the agent’s economic resources. Individuals who are a) more risk averse, b) have more income uncertainty, c) face a larger cut in benefits or d) have higher marginal tax rates are the subpopulations most adversely affected by the government indecision.

While I am usually skeptical of life-cycle model papers which predict the future using calibration and do not feel the author’s estimates are very precise, it is important to realize that increased uncertainty due to delays in social security reform can lead to suboptimal asset allocation and consumption decisions by individuals planning for retirement.

One Social Security reform often suggested is to change the indexation of benefits from a wage inflation measure to a price inflation measure, such as the Consumer Price Index (CPI).  Critics argue that the price of healthcare has increased more than overall CPI and since the elderly spend a higher percentage of their income on healthcare, the CPI may not accurately reflect their annual cost of living increase.  This is certainly true.  According to the Bureau of Labor Statistics website, during the last 20 years overall prices have increased 85.3%, but medical care prices have increased 176.3% and medical care service prices have increased 188.6%.

This phenomenon was the inspiration for a Berndt, Cockburn, Cocks, Epstein and Griliches (1998) study on pharmaceutical price inflation for the elderly.  The authors wanted to test whether or not price inflation for pharmaceuticals is different for the elderly as compared to the non-elderly. 

The authors constructed an elderly and non-elderly aggregate pharmaceutical price index and find that elderly drug prices increased 33.1% over 6 years (1990-1996) while non-elderly drug prices increased 32.9% in the same period, an insignificant difference.  The authors also look at 3 types of medications in more detail: antibiotics, antidepressants and calcium channel blockers.  They find that when the elderly proportionally use more generics (as in the case of anti-depressants) the elderly price inflation is lower than that of the non-elderly; when the elderly use more branded medicines (as in the case of antibiotics), then elderly price inflation is higher than that of the non-elderly.  Their paper suggests no persistently higher rate of price increase for the pharmaceutical products used by the elderly as opposed to the non-elderly; the only differences in price inflation are due to the frequency of generic drug usage.

Throughout the past week, I have spoke of the work disincentives many social security programs create.  The question is: how do we measure these disincentives.  The economics literature has given three different metrics to measure implicit social security wealth a retiree has and I will discuss each in turn.

Accrual

The accrual method measures how much the value social security benefits would increase (or decrease) if a person decides to postpone retirement by one year at age ‘t‘.

• Accrual=SSW{t+1} - SSW{t}

SSW_{t} is one’s implicit social security wealth if they retire at age ‘t’.  This is calculated by taking the nominal benefits at each future date and multiplying them by the probability of surviving to that date as well as a discount factor. 

Take the example below:

Here, a pension for someone who retires at age 64 is $1000 per year.  I assume that everyone dies at age 68.  The total value of the individuals SSW is $2654.  What if the person decided to postpone retirement to age 65?

In my example, the individual receives $2000 per year if they retire at age 65 (instead of 64).  After taking into account the probability of surviving to each age as well as the discount factor, the person’s new SSW is $3309.  Thus the accrual amount is $655.  By postponing retirement, the individual increases their Social Security Wealth.  In many systems, this amount can be a large negative amount which gives individuals an incentive to retire early.

Peak Value

The peak value calculation is similar to the accrual method. 

• Peak value=SSW{r*}-SSW{t}

This method takes the SSW at the age of retirement (’r*‘) where r* is the age of retirement which maximizes the value of social security benefits and subtracts the SSW which would result from retirement this year.  This method may be optimal to capture the fact that many people retire at a target year and don’t calculate their own incentives each year.

Option Value

The option value is the most sophisticated method.  It takes into account utility from not working, but requires the researcher to model a utility function and assume parameter values.  First, let us calculate the value of retirement (V)at date ’s’.  This is equal to the discounted expected utility of wages earned between the current date t and date ’s-1′ plus the discounted expected utility of social security benefits earned between date ’s’ and the end of life.

The option value is equal to: V(r*)-V(t).  This is the difference between the discounted expected utility from retiring at r* (the date that maximizes utility) and the discounted expected utility from retirement today (date ‘t’).  If the option value is positive, the individual will continue to work.  If the option value is negative, the individual will retire. 

A more comprehensive treatment is given in Stock and Wise (1990).

Stock and Wise (1990) “Pensions, the Option Value of Work and Retirement”, Econometrica, Vol. 58(5), pp. 1151-1180

The public pension system in the United Kingdom differs significantly from its peers in continental Europe.  First, private and occupational pensions play a much larger role in retirement than in the rest of Europe.  Secondly, UK projects that public pension savings will actually decrease as a percentage of GDP (from 4.5% in 2000 to 4.1% in 2050) despite an aging population.  This is due to a decrease in benefit generosity and a push for more individuals to self-insure for retirement.  

Pensions are financed through an 23.8% payroll tax on weekly earnings between £89 and £591 (11% paid by the employee, 12.8% paid by the employer).  In 1968, 80% of males between the ages of 60 and 64 were in the labor force, but now this has dropped by 1996 to just under 40%. 

First Tier: The Basic State Pension

This is a flat-rate contributory benefit payable to men over 65 and women over 60 (gradually increasing to 65 for women by 2020). To qualify for the £73 weekly benefit, individuals must pay into the system for 90% of their working life.  There is no mean-testing for this program, and delaying the receipt of benefits by one year will increase weekly payments by 10%. 

Second Tier: SERPS, Occupational and Personal Pensions

The State Earnings Related Pension Scheme (SERPS) was introduced in 1978 and pays a pension equal to a fraction of an individuals qualifying annual earnings each year since 1978.  Originally it aimed to replace 25% of an individual’s best twenty years’ earnings (up to a limit), but subsequent cuts have move the replacement rate to 20%.

Individuals can elect to opt out of the SERPS program and instead participate in private pensions (an individual retirement account) or an occupational pension.  Occupational pensions currently cover 45% of all workers.  Since 1988, individuals could also leave SERPS and their occupational pension in favor of a private savings scheme.  By the mid-1990s one-quarter of all employees had taken out a personal pension.  Blundell, Meghir and Smith note that “there is a serious issue over the number of older workers who were ‘mis-sold’ personal pensions by financial advisers who wrongly advised them that they would be better off leaving their occupational pension scheme.” It is not surprising that 22% private-sector occupational schemes plans are defined contribution (or a defined benefit/defined contribution hybrid) as compared to only 2% in the occupational schemes in the public sector. 

Other Programs

The state also has a Minimum Income Guarantee program.  This is a flat-rate, noncontributory, means tested program to help those with little income.  All individuals aged 60 and older who are unemployed receive the Minimum Income Guarantee.  The incapacity benefit is for disabled individuals.  One must undergo a medical examination by a general practitioner (GP) and be certified unable to do any work to receive this benefit. 

For the average Spaniard, going into retirement means collecting a public pension and nothing else.  In 1990 less than 5% of the employed population had a private pension; while this figure has risen to 30% in 1999, still less than 1% of the population derives more than 10% of their retirement consumption needs from private savings or private/employer pensions.  Public Pensions expenditures currently account for 9.5% of annual GDP; this compares to 4.5% of annual GDP spent on Social Security in the U.S. in 2002.

The pattern of Labor Force Participation in Spain is similar to that of France, Germany and Italy.  Just under half (46.5%) of the population retires by age 60, although reform efforts have attempted to slowly increase the de facto age of retirement. 

Social Security

There are a variety of Social Security schemes but the largest is the Régimen General de la Seguridad Social (RGSS) which covers about 70% of all workers.  Special programs for specific classes of workers include: Régimen Especial de Trabajadores Autónomos  (RETA) for the self employed, Régimen Especial Agrario (REA) for agricultural workers, and the Régimen Especial de Trabajadores del Mar (RETM) for sailors. 

The RGSS plan is a pay-as-you-go system and contributions are 28.3% of gross income (23.6% paid by the employer and 4.7% paid by the employee).  Entitlement to an old age pension requires 15 years of contributions and is conditional on having reached 65 years old.  Pensions are paid out as follows:

• Pension=a*BR 

‘BR‘ is the base reguladora which is calculated by summing an inflation adjusted measure of earnings in the last 96 months before retirement dividing this number by 112.  Those with more than 15 years of service receive a value of ‘a’ which increases from 0.5 as the number of years of service increases and is capped at 1 for those with 35 or more years of working.  Thus, people who worked over 35 years can expect to have about 85% of their average real earnings replaced through the public pension system–subject to a pension minimum and maximum. 

Early Retirement

Under the RGSS, early retirement is available but individuals are charge a penalty of 8% for each year retired under 65.  Some individuals in hazardous jobs (railroad workers, airline workers, and of course bullfighters) or workers laid off for industrial restructuring can retire at age 60.  These groups represents about 10% of early retirees.  For government workers, the Régimen de Clases Pasivas (RCP) allows workers to retire at age 60 if they have accumulated enough work experience.

A more common means to reach retirement is through unemployment insurance (UI).  The UI program in Spain ranges between 120 and 720 days, however, a special program exists for those aged 52 and above.  This ‘UI 52+’ program includes those who a) would otherwise be eligible for a pension, except for their age, and b) have an income below 75% of the monthly minimum wage.  The benefits are paid until the individual reaches RGSS eligibility at age 65 and are set at 75% of the monthly minimum wage.  Other individuals elect to go on early disability–which involves a complicated set of eligibility rules which will not be discussed here.

While France, Germany and Italy struggle to reform their gigantic old-age pension system, in the Netherlands the problems are not nearly as acute.  The Dutch have a Social Security system, but it is not as large as in the countries to its south.  A large portion of the a Dutch individual’s pension comes from their employer; thus the fiscal strain on the government is not as severe.  On the other hand, the Netherlands ranks only behind Italy in terms of the amount of individuals between ages 56 and 65 who are out of the labor force.  This is mostly due to the fact that workers can leave the labor force and rely on the generous Dutch safety net (Unemployment Insurance, Disability, etc.) to replace the majority of their income until they reach the age for Social Security eligibility.

Pensioen (State Old-Age Pension) 

The Dutch social insurance (sociale verzekeringen) program for its state Old-Age Pension (Pensioen) is a flat rate benefit that all residents receive after they have reached 65 years old.  The rate is generally 1/2 the after tax statutory minimum wage (minimumloon), with supplements for single individuals and for married couples with one spouse who is younger than 65.  This way, all individuals are assured at least a living standard of a minimum wage worker, which is about €1350 per month ($20,700 USD/year).  This pay-as-you-go system is financed through a 17.9% payroll tax up to a maximum and pensioen expenditures account for 5% of GDP.

Other Programs

Although the official retirement age in the Netherlands is 65, a variety of other programs allow workers to retire much earlier.  Disability insurance (DI) “guarantees employees who lost more than 80 percent of their earnings capacity a benefit equal to 70%…of their daily wage (up to a maximum).”  In the 1980s, this program was such a popular means for employers to get rid of less productive elderly workers that the Dutch government had to reduce the benefit (the benefit was 80% of wages before 1985) and increase the stringency of the eligibility requirements. 

Unemployment benefits (UI) are a less desirable means of getting to early retirement, since benefits are of limited duration.  For people aged sixty of above, however, one can expect to receive UI benefits equal to 70% of their prior earnings up to age 65. 

Private Pensions

Almost 80% of Dutch male residents receive a pension from a private employer or occupation group by the time they reach age 65.  In fact, over half of dutch male residents receive a private pension between the ages of 60 and 64.  Early retirement (ER) was common during the 1980s as a means to reduce unemployment, but as the costs of ER plans have increased and unemployment has dropped to 5.8% as of May 2006, flexible retirement plans are more common where benefits increase as the age one chooses to retire rises.

Overall, the Netherlands has a relatively minimal public pension system (compared to other OECD countries), but generous social welfare benefits and private pensions have made early retirement an attractive option for many Dutch. 

Japan is an incredible success story.  After so much suffering in World War II, this country now has the second largest economy in the world and the longest life expectancy of any country in the OECD. This success however, may become a fiscal nightmare in the future. In 1998, 16.2% of the Japanese population was 65 or older (about the OECD average), but by 2050 this number will jump to 32.3%. Spending on public pensions accounted for 9.3% of GDP in 1997 and this number is only expected to increase. The 1994 Pension reform attempted to resolve some of the fiscal problems facing the nation. Benefits were to be reduced 5%, eligibility for KNH (see below) benefits was to increase from age 60 to 65, and wage indexation of benefits was to be terminated. Instituting American-style 401(k) programs have also been proposed.  

While these fiscal issues need to be resolved, longevity should be celebrated. Further, Japan has among the highest elderly labor force participation in the OECD. For those aged 60-64, 74.8% of men are working and 40.1% of women are working. Of all the public pension programs I have examined so far, the Japanese system seems to be the most sensible.

Kiso Nenkin

This is a basic flat rate pension give to all residents age 65 and up. There are no earnings tests or work history requirements needed to receive his benefit, but the pension is not large. Often, this is the only benefit housewives receive.

Kosei Nenkin Hoken (KNH – Private Employees Pension)

This program covers 85% of all workers; the rest are covered by Kyosai Kumiai (Public Employees Pension), but since the programs are very similar I will only discuss the KNH. The pension benefit is calculated as follows:

• Pension=(infl. adj. avg. wage)*(contribution years)*(0.0075)

This means a typical individual who worked 50 years will earn 30% of their career monthly earnings. Factoring in the Kiso Nenkin benefit of a husband and his wife, a typical couple will have a pension of about 50% of their annual wages. The program is financed through a payroll tax of 17.35% split evenly between the employer and the employee.

One can receive partial pension benefits beginning at age 60 through the Zaishoku program (see below), but full pension benefits do not occur until age 65. After age 65, one can delay pension benefit receipt with some actuarial adjustment.

Zaishoku Pension

The Zaishoku Pension is an earnings tested pension for employees aged 60-64 and is part of the KNH system. If one works between 60-64, an automatic 20% reduction in benefits occurs and other pension benefits are taxed away as earnings increase. Implicit tax rates are: 50% for earnings above ¥220,00 ($1,910) per month and 100% for earnings above ¥340,000 ($2900) per month.

Unemployment/Disability

Unemployment Insurance (UI) is limited to 300 days in Japan. Some people do use UI as a form of early retirement but since benefits are limited to a year, this has a moderate impact on the number of people who retire.

Unlike most European countries, very few Japanese use the Disability program as an interim income support. The rules to qualify for this program are strict and require full disability. Also, the benefits are similar to the KNH program so only the truly disabled generally apply.

Kosei Nenkin Kiken (KNK – Employee’s Pension Fund)

The employer-provided pension program covers about 2/3 of all workers. Workers can choose a lump sum payment or an annuity. Often, the KNK benefit is paid in conjunction with a mandatory retirement at age 60. Due to the seniority system, middle-aged workers are often expensive and firms use mandatory retirement to unload these workers; those who leave their job through mandatory retirement often find other part-time work, but finding full-time work in Japan for those over 60 is often difficult.

Like the French, Germans retire early (on average at between age 59 and 60) and for Germans, this early retirement is  seen as a social achievement.  Some polls, however, show that younger Germans do not believe they will receive an adequate pension in their old age and attempts at reform are underway.

The Germans have had a public pension system for over one hundred years, and their system has often been the model which other countries emulate.  The German pay-as-you-go system in it current form began in 1957 with a pension benefit and a flexible retirement age.  In 1972, the retirement age for old age pension was lowered, and old age disability pensions were created.  The “Century Reforms” of 1992 and 2001 decreased benefits throughout the pension system in an attempt to stem the fiscal problems which occur in an aging society.  The 1992 reform included a variety of changes such as making pensions dependent on pension type (discussed later) and indexing the average pension benefit for the entire system to net wages instead of gross wages as done previously.  The 2001 reform will gradually reduce replacement rates from 72% in 1997 to 64% in 2030.  Let us first look at the private sector:

Gesetzliche Rentenversicherung (GRV)

This public retirement insurance covers 85% of the German workforce and is mandatory with exemptions for civil servants and some self-employed.  Generally this is the only money individuals receive in old age, since individual retirement accounts and occupational pensions do not play a major role in Deutschland.  The program is financed 70% through a payroll tax split evenly between employees and employers (the tax is currently 19.5% of gross wages), and 30% through earmarked indirect taxes (VAT and a gas tax).

The GRV offers flexible retirement, with full benefits occurring at age 63 for most men, but age 60 for women, the unemployed and the disabled.  GRV pays no spouse benefits, but there are survivor’s benefits which are 60% of the deceased mate’s former benefits.  As of 1998, the average replacement rate was 70.5%–meaning one could make almost three-quarters one’s working salary during retirement. 

The formula to determine one’s pension depends on the following four factors:

1. Relative earnings position - computed by averaging average pension contributions over working life
2. Years of Service
3. Pension Type - There are 5 types.  The retirement age/necessary years of service are in parentheses. 
• Normal (65/5), Long Service (63/35), Women (60/15), Older Disabled (60/35), Unemployed (60/15)
4. Average pension for the nation as a whole.

As you can see, from point (3), while retirement is officially at age 65, many people retire earlier.  Two of the most common routes to early retirement are:

• Go on the Unemployment rolls at age 57.  One will be able to collect Unemployment Insurance (UI) for three years or negotiate a severance package with one’s employer. At age 60, an individual will be able to qualify for a pension.
• Go on disability at age 60.  One must prove to be at least 50% physically disabled and pass an earnings test to qualify.    

Civil Servants

Civil Servants receive an even more generous package than those in the private sector.  Replacement rates for civil servants are 75% of gross (pre-tax) wages, compared to 70% of net (after-tax) wages for the private sector.  Benefits are a function of 1) years of service, 2) last gross wage, and 3) a new adjustment factor for early retirement. 

According to the OECD Structural Policy Indicators paper, France has one of the highest implicit tax rates on continued work during its old age pension in the entire OECD.  In 1998, the implicit tax was over 80%.  Subsequent reforms have reduced the implicit tax to about 50% in 2003; yet only Luxemburg and the Netherlands have higher implicit tax rates on working while receiving old age pension than does France. 

Thus, it is not surprising to find that the average age of retirement in France is 59 years of age, and only 53% of individuals over between 55 and 59 are labor force participants.  According to the OECD Observer, these numbers are much lower than in Denmark, where the average retirement age is 65 for men and 62 for women.  It may be sensible that 86.2% of individuals over 65 receive some public benefits, but it is surprising that 23.7% of people aged 55-59 receive public benefits. 

There are two major reasons why retirement occurs at such an early age.  First, in 1983 France lowered the age for Social Security eligibility from 65 to 60.  Secondly, generous Unemployment Insurance allows older workers to leave the labor force even earlier than age 60.  Let us now examine the programs which make up France’s old age pension plan.

Social Security

This is the largest program in France.  The pension amount is calculated according to the following formula:

• Pension=a*[max(N,T)]/T*[infl. adj. avg. wage of best max(N,T) years]

N is the number of months an individual has been employed over their career, with T being a maximum cutoff.  The ‘a‘ term is determined by 1) the age of retirement and 2) N; and ranges between 0.25 and 0.50.  A later retirement date and more years of service moves ’a’ closer to 50%.  There was a maximum monthly pension of €2353 as of 2002.  In the 1970s and early 1980s, the benefits were indexed to average wage growth, but this was subsequently replaced with a price inflation measure in the mid 1980s.

Complementary Schemes

These plans are defined contribution plans (but not fully funded) and are organized by the private sector.  Workers receive a pension equal to:

• Pension=Sum{c(t)*w(t)/PP(t)}

c(t) is the contribution rate and w(t) is the gross wage each year.  PP(t) is the purchase price of a ‘point’ (salaire de rèfèrence) of pension benefits.  The Complementary Schemes are organized by occupation and benefits are limited to a maximum amount.  Executives belong to one organization (Association Gènèrale des Institutions de Retraite des Cadres - AGIRC) and non-executive employees belong to another organization (Association des Règimes de Ratraite Complèmentaire - ARRCO). 

Civil Servants have a separate pension system in which individuals can receive a full pension at age 60 as long as they have had 15 years of service.  There is also a mandatory retirement age of 65 for these workers.

Unemployment Insurance and the FNE

Most individuals who qualify for Unemployment Insurance (UI) receive declining benefits as the duration of their unemployment spell increases.  For workers over 58, however, this is not the case.  For these workers, they receive full UI benefits until the age they are eligible for Social Security.  This makes applying for UI prior before age 60 a very attractive option. 

Further, the state has set up the Fonds National pour l’Emploi (FNE) to actually increase the rate of retirement among elderly workers.  If a firm agrees, the state will give additional benefits to elderly workers who are laid off.  While this seems nonsensical (why is the state paying firms to fire workers), the government claims that the FNE will open up positions for younger workers.  According to the 1 July 2006 Economist article (”Output Demand and jobs“), the unemployment rate in France was 9.3%, so finding positions for younger workers is a significant problem (see problems with the Contrat de Première Embauche).

In Denmark the official retirement age is 65, but on average most people retire at age 61.  Around 8%-9% of GNP is spent on government income transfers to the elderly.  Among men between the ages of 55-65, only Sweden, the U.S., and Japan have higher labor force participation (LFP) rates.  Why do we see such high LFP in Denmark (which is generally seen as having a very generous social insurance system)?

The reason for this is that pensions generally range from 40%-70% of average prior wages (adjusted for inflation).  The proportion is closer to 70% for low wage earners and closer to 40% for high wage earners.  Thus, we see the earliest retirement among people in the manufacturing and service industries who generally have lower wages. 

I will now briefly describe the retirement programs in Denmark:

Old Age Pension (OAP)

This is a pay-as-you-go system which has a residency and age criteria as its only requirements.  At first this program gave a base amount to everyone 67 years and older but “from 1993 forward, this amount has been means tested against earnings from work, but not against capital income.”  OAP is financed from general tax revenues.

Social Disability Pension (SDP)

This is a disability program in which eligibility depends on health criteria, although these eligibility rules have become more flexible over time.  Individuals enrolled in SDP automatically switch to the OAP program once they reach 67 years of age.  Like OAP, this program is financed from general tax revenue. 

Post Employment Wage program (PEW) or Efterlhn

The goal of this program was to allow elderly workers to retire early to open up jobs for younger workers.  In reality, the PEW program has become one of the major means by which Danes are able to take early retirement.  On can enter the PEW program at age 60 as long has the individual has a minimal employment attachment when they apply. 

Other Programs

The Transitional Benefits Program (TBP) gives an early retirement option for workers aged 55-59 who are members of an Unemployment Fund and who have been unemployed for 12 of the last 15 months.  In 1994, this benefit was extended to individuals 50-54 years old as well.  

Public Employees have a Public Employee Pension (Tjenestemandspension) which is a defined benefit program as well. 

The Special Pension Program was enacted in 1999 in which 1% of wages go towards a defined contribution fund.  The program currently represents a small portion of the overall retirement needs of the elderly but may grow in importance in the future. 

How to get to retirement?

• If you are under age 50, your only option is to apply for disability through the SDP program. 
• Between 50 and 59 years old, one can still apply for disability (SDP) or an individual can apply to the TBP program if they meet the requirements. 
• Between age 60 and 66, retirement is popular because 1) one can enroll in the PEW program and 2) many labor market pensions give their employees benefits beginning in this age range.  This is the route most people take; the average age of retirement in Denmark is 61.
• At age 67, nearly all individuals in Denmark are retired.  All citizens and most residents are eligible for the OAP program once they reach age 67.   

In Chapter 2 of Social Security Program and Retirement around the World, economists Michael Baker, Jonathan Gruber and Kevin Milligan look at Canada’s Income Security Programs (IS).  According to their research, total expenditures on the three largest elderly transfer programs in Canada cost $22.7 billion in 1998-1999 or 20% of the federal budget.  Also, like most OECD countries, Canada has an aging population.  In 1975, contribution per capita towards public pensions exceeded these pension benefits per capita by $200.  By 1998, benefits exceeded contributions on a per capita basis by about $200.  The authors also note a trend for workers to retire earlier despite increased life expectancy.  In 1960, 87% of men aged 55-64 were labor force participants, but by 1999 this figure had dropped to 61%. 

Below I will discuss the major Canadian programs which transfer money to the elderly:

Old Age Security (OAS) System

Begun in 1952, this is the oldest federal social security program in Canada.  The program replaced means-tested, provincially-administered social security programs which were established in 1927.  The OAS program originally transferred cash to the elderly over 70 years of age but this age limit was later reduced to 65.  A uniform grant of about $420 (in 2000) is given to all who meet certain minimal residency and work requirements.  Since 1972, the OAS benefit has ben indexed to inflation (CPI).  This income is taxable and there is a clawback of 15% for high income individuals.

Canada Pension Plan/Quebec Pension Plan (CPP/QPP)

This is the largest of Canada’s IS programs.  It is financed through a payroll tax of 3.5% on all income between $3500 and the Year’s Maximum Possible Earnings (YMPE) which was $42,100 as of July 2006.  CPP/QPP averages an individual’s earnings between ages 18 and 65 after excluding 1) time spent on disability, 2) time caring for small children, 3) the remaining 15% of eligible months with the lowest earnings.  These amounts are then adjusted for inflation by multiplying eligible earnings by [(YMPE: year earnings accrued)/(average last 5 years YMPE)].  Once an inflation adjusted monthly income is calculated, individuals receive 25% of their previous income subject to a maximum. 

After 1987 for the CPP (and 1984 for QPP) individuals could claim their benefits at age 60 with an actuarially fair reduction in monthly payments.  Fifty percent of Canada’s citizens do claim their benefit early.  Benefits are paid at an individual level, but widows and surviving children under age 18 are entitled to the benefits of the deceased family member.

Guaranteed Income Supplement (GIS) 

This program is intended for OAS recipients with low incomes.  Individuals who wish to qualify for GIS must re-apply every year in order to receive the program’s income guarantee of about $500 per month for married families and $325 per month for single households.

Spouse Allowance (SPA)

For spouses aged 60-64 who are married to OAS recipients or widows of the same age, the SPA program gives them a cash transfer.  The benefits, however, are taxed away at 75% if the individual begins to work. 

Analysis 

Baker, Gruber and Milligan use the Statistics Canada Longitudinal Worker File data set to analyze retirement patterns in Canada.  Their major findings are that the IS system as a whole increasingly provides disincentives to work after age 61.   More specifically, while the present discounted value (PDV) of IS benefits decreases if one continues to work age 61; at age 69, working an extra year would reduce the PDV of IS benefits by 43%!  Reforming the IS system would be one way to induce more elderly individuals to continue to work. 

The Gruber and Wise book evaluates how the following two reforms will affect a variety of OECD countries:

• Delay official retirement age 3 years
• “Common Reform”

The Common Reform is an option favored by many economists since it seeks to minimize the distortions of an individual’s decision of when to retire.  The four characteristics of the policy are:

1. Set the normal retirement age at 65
2. Allow early retirement between age 60 and 65
3. For those who retire early, benefits will be reduced in an actuarially fair manner
4. For those who retire late, benefits will be increased in an actuarially fair manner.

Gruber and Wise use 6% as the benefit reduction amount per year for those who retire early and also for the benefit augmentation amount per year for those who retire after 65.  This would not distort the retirement decision much if at all, since the net present value of both benefits would be the same for each individual. 

This seems like a complicated scheme, but is it worth the hassle?  Is early retirement really a problem? 

According to Gruber and Wise, in 1960 82% of Americans aged 60 to 64 were working or seeking work, but by 1995 this number dropped to 53%.  While this may seem like a steep decline, it is small compared to other countries who allow workers to collect social security at even earlier ages than the U.S.  For instance, Belgium had a labor force participation rate of over 70% for its workers aged 60 to 64 in 1960, but in 1995 this number had dropped to just below 20%.  Germany has seen a drop from above 70% to under 40% in the same time period.  Thus, the common reform, seems a sensible way to give the elderly a minimum level of subsistence, while reducing distortions in the retirement decision.

Social Security has done much to reduce poverty rates in the elderly.  The program, originally known as Old Age, Survivors and Disability Insurance (OASDI), was signed into law in the United States by FDR in 1935.  At that time, poverty among elderly in the U.S. exceeded 50%.  The Center on Budget and Policy Priorities now claims that only 8.7% of the elderly live below the poverty line in the United States.

Like every government intervention, the Social Security Program had unintended consequences.  According to economist Martin Feldstein (JPE 1974), Social Security crowds out personal saving by 30%-50%.  This means the amount of money individuals would have saved for retirement in the absence of Social Security is significantly higher than the amount they currently save.  Also, retirement programs can also induce individuals to leave the labor force (retire) before they would have otherwise. 

Regarding the second issue, Jonathan Gruber and David Wise have published a book titled “Social Security Programs and Retirement around the World.  The book examines how the structure of Social Security and retirement decisions are related.  For instance, in the United States there are two popular ages to retire: age 65 (when individuals receive full retirement benefits) and age 62 (when individuals can receive early retirement benefits).  On the other hand, in Germany citizens receive early retirement at age 63, normal retirement benefits at age 65 and we see corresponding spikes in the proportion of people leaving the labor force at those ages as well as age 60.  Age 60?  Germans who qualify for disability insurance at age 60 or about receive generous retirement-like benefits; hence the reason for the jump in the amount of people who retire at that age. 

During the rest of this week, I will be examining different countries and their Social Security plans.  Most of the material will come from the Gruber and Wise book, but I will also use other material.