Issue Brief - Employee Benefit Research Institute

5 downloads 94132 Views 1MB Size Report
downloaded for free at www.adobe.com/products/acrobat/readstep2.html ..... for purposes of illustration, this Issue Brief assumes just two asset classes: (1).
Issue Brief No. 297

September 2006

Measuring Retirement Income Adequacy: Calculating Realistic Income Replacement Rates By Jack VanDerhei, Temple University and EBRI Fellow •

• •









The limitations of traditional “replacement rate” calculations: For decades, “replacement rates” have been the primary “rule-of-thumb” measure used in the retirement planning process. However, replacement rate calculations are overly simplistic and potentially inaccurate because they often are based on methodologies limited to replacement of preretirement cash flow after adjustment for taxes, savings, and age and/or work-related expenses. Importance of investment, longevity and health risks: One of the biggest weaknesses of replacement rate models is that one or more of the most important retirement risks is ignored: investment risk, longevity risk, and risk of potentially catastrophic health care costs. A new “building block” approach: This Issue Brief illustrates the problematic nature of using conventional replacement rates for retirement planning through a “building block” approach: Building Block 1 focuses exclusively on investment risk; Building Block 2 introduces longevity risk into the planning process, in addition to the investment risk from the previous level; Building Block 3 introduces the risk of catastrophic health care costs into the calculations, in addition to investment and longevity risk. A more realistic way to calculate replacement rates: Building Block 3 represents the approach that will be used in a new Web-based planning tool to assist preretirees in their attempt to choose a meaningful replacement rate for purposes of retirement planning. This resource will be released in 2007 as a free online tool called the Ballpark E$stimate® Monte Carlo. The importance of probabilities: Some retirement planning models that, by default, use average values for longevity, investment, and health costs implicitly are using a 50 percent probability of success. Since most preretirees will want a higher probability of success, the Ballpark E$stimate® Monte Carlo model also shows results for 75 and 90 percent probability of success. Individualized results: In reality, there is no “correct” single replacement rate. Even at a specified probability of success, an “adequate” replacement rate depends dramatically on the level of retirement expenditures, retirement age, gender, asset allocation, percentage of annuitization, and other variables detailed in this Issue Brief. Conversion of the savings needed to a multiple of final earnings that is needed in retirement savings can provide a clearer picture for some, so the Issue Brief presents that as well (see pages 11, 16, and 29). Examples: Variation in target replacement rates can be seen below (for the case in which there is no equity allocation of assets and none of the initial retirement wealth is annuitized): Target Replacement Rates for High-Income Individuals (single retirees making more than $40,450 per year; 4.6 percent) Probability of Retirement “Adequacy” 50% 75 90

Male retiring at 65 52% 78 119

Female retiring at 65 59% 98 128

Male retiring at 62 64% 97 149

Male retiring at 68 43% 66 97

Target Replacement Rates for Low-Income Individuals (single retirees making less than $15,000 per year; 70.3 percent) Probability of Retirement “Adequacy” 50% 75 90



Male retiring at 65 124% 229 394

Female retiring at 65 147% 292 453

Male retiring at 62 153% 285 476

Male retiring at 68 95% 206 332

While these replacement rates will be larger than those typically contemplated for some individuals, this Issue Brief explores how the purchase of annuities at the time of retirement may be used as an effective risk management technique in some cases to reduce these targets.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

Jack VanDerhei, Temple University, is research director of the EBRI Fellows Program. Special thanks to Craig Copeland, senior research associate at EBRI, for his assistance with this report. Several of the assumptions used in this report are based on findings published by Paul Fronstin, senior research associate and director of the health research and education program at EBRI, in “Savings Needed to Fund Health Insurance and Health Care Expenses in Retirement,” EBRI Issue Brief no. 295, July 2005. Note: The electronic version of this publication was created using version 6.0 of Adobe® Acrobat.® Those having trouble opening the pdf document will need to upgrade their computer to Adobe® Reader® 6.0, which can be downloaded for free at www.adobe.com/products/acrobat/readstep2.html

Table of Contents Introduction.......................................................................................................................................4 Recent EBRI Research......................................................................................................................5 Methodology.....................................................................................................................................6 Results...............................................................................................................................................8 Building Block 1: Investment Risk Only.................................................................................................... 8 Building Block 2: Investment and Longevity Risk..................................................................................... 9 Building Block 3: Investment, Longevity and Long-Term Care Risk ...................................................... 12

Conclusion ......................................................................................................................................18 References.......................................................................................................................................19 Endnotes .........................................................................................................................................20

Figures Figure 1, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65 in the Lowest Income Category.................... 8 Figure 2, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65, Retirement Income Category 2.................... 10 Figure 3, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65, Retirement Income Category 3.................... 10 Figure 4, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65 in the Highest Income Category................... 11 Figure 5 Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation). For: Males Retiring at Age 65........................................................................................................................................... 11 Figure 6, Replacement Ratios Required for Various Probability Levels of "Adequacy," by Income Category and Equity Allocation. For: Males Retiring at Age 65 ..................................................... 13 Figure 7, Necessary Replacement Rates for Retirement Income "Adequacy," by Retirement Income Category and Equity Allocation (Various Probabilities). For: Males Retiring at Age 65 ................ 13 Figure 8, Impact of Replacement Rates on the Probability of Retirement Income "Adequacy," by Retirement Income Category and Equity Allocation........................................................................ 14 EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

2

Figure 9, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65 in the Lowest Income Category ................... 14 Figure 10, Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation. For: Males Retiring at Age 65 in the Highest Income Category................... 16 Figure 11, Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation and No Annuitization). For: Males Retiring at Age 65 .......................................................................................................... 16 Figure 12, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization. For: Males Retiring at Age..................... 17 Figure 13, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Rate of Annuitization (Assumes Lowest Income Category and No Equity Investment).............................. 18 Figure 14, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization. For: Males Retiring at Age 65................ 22 Figure 15, Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 0% Equity Allocation). For: Males Retiring at Age 65 in the Lowest Income Category .............................................................................................................................. 23 Figure 16, Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 50% Equity Allocation). For: Males Retiring at Age 65 in the Lowest Income Category .............................................................................................................................. 23 Figure 17, Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 0% Equity Allocation). For: Males Retiring at Age 65 in the Highest Income Category .............................................................................................................................. 24 Figure 18, Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 50% Equity Allocation). For: Males Retiring at Age 65 in the Highest Income Category .............................................................................................................................. 24 Figure 19, Replacement Ratios Required for a 50% Chance of "Adequacy." For: Males Retiring at Age 65 in the Lowest Income Category ........................................................................................... 25 Figure 20, Replacement Ratios Required for a 75% Chance of "Adequacy." For: Males Retiring at Age 65 in the Lowest Income Category ........................................................................................... 25 Figure 21, Replacement Ratios Required for a 90% Chance of "Adequacy." For: Males Retiring at Age 65 in the Lowest Income Category ........................................................................................... 26 Figure 22, Replacement Ratios Required for a 50% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category .......................................................................................... 26 Figure 23, Replacement Ratios Required for a 75% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category .......................................................................................... 27 Figure 24, Replacement Ratios Required for a 90% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category .......................................................................................... 27 Figure 25, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization. For: Males Retiring at Age 65................ 28 Figure 26, Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation and No Annuitization). For: Males Retiring at Age 65 .......................................................................................................... 29 Figure 27, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization. For: Females Retiring at Age 65 ............ 30 EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

3

Figure 28, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation and Rate of Annuitization. For: Males Retiring at Age 62................. 31 Figure 29, Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization. For: Males Retiring at Age 68................ 32 Figure 30, Replacement Ratios Required for a 50% Chance of "Adequacy." For: Males Retiring at Age 65, Lowest Income Category .................................................................................................... 33 Figure 31, Replacement Ratios Required for a 75% Chance of "Adequacy." For: Males Retiring at Age 65 in the Lowest Income Category ........................................................................................... 33 Figure 32, Replacement Ratios Required for a 90% Chance of "Adequacy." For: Males Retiring at Age 65 in the Lowest Income Category ........................................................................................... 34 Figure 33, Replacement Ratios Required for a 50% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category........................................................................................... 34 Figure 34, Replacement Ratios Required for a 75% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category........................................................................................... 35 Figure 35, Replacement Ratios Required for a 90% Chance of "Adequacy." For: Males Retiring at Age 65 in the Highest Income Category........................................................................................... 35

Introduction For decades, “replacement rates” have been the primary measure used in the retirement planning process. This is defined as the annual amount of an individual’s retirement income, divided by his or her yearly earnings just prior to retiring. For instance, someone who retires from a job with a $100,000 annual salary and has $75,000 a year in retirement income has a “replacement rate” of 75 percent. A “Part 1” article by EBRI (VanDerhei, EBRI Notes, September 2004) reviewed how these rates have traditionally been used to establish minimum targets for future retirees by calculating the amount needed to provide the same amount of after-tax income in retirement as that received prior to retirement after adjusting for differences in savings, age, and work-related expenses. Results from one of the most commonly cited studies indicated that for a one wage-earner family retiring at 65 with a spouse age 62, the target replacement rates were between 75–89 percent (depending on income) in 2004 (Alford, Farnen, and Schachet, 2004). Previous research on projected replacement rates found that typical 401(k) participants at very large employers were well positioned to replace 85–95 percent of preretirement income when current Social Security and existing profit-sharing and defined benefit plans are taken into account (Steinberg and Lucas, 2004). However, the likely adequacy of these income replacement rates is a function of what type of postretirement health care coverage a worker has from a previous employer. Steinberg and Lucas subtracted retiree medical costs net of subsidies from retirement income levels to determine a “net” replacement income ratio, reflecting the percentage of preretirement income available to meet all needs other than medical. As a result, the overall average replacement ratio for their analysis drops from 95 percent under the high medical coverage assumption to 83 percent under the medium assumption and 80 percent under the low medical coverage assumption. This is true for employees retiring at a “normal” retirement age of 65, and who are relying primarily on Medicare for their health care benefits. Employees retiring at an earlier age will experience even larger financial modifications. While these techniques provide EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

4

simple rule-of-thumb measures for the retirement planning process,1 there have been recent attempts to conceptually redefine retirement income adequacy (Schieber, 1996 and 1998). Moreover, it is important to understand that the current methodologies (however modified) only solve for the amount of money needed to provide the same annual cash flow available prior to retirement (albeit adjusted for taxes, savings, age and/or work-related expenses and, in some cases, the cost of post-retirement health care), and ignore some or all of the risks of longevity, investment of retirement assets, and catastrophic health care costs (such as a prolonged stay in a nursing home without the benefit of long-term care insurance).

Recent EBRI Research EBRI has recently completed a simulation model—the EBRI/ERF Retirement Security Projection Model® (RSPM)—that incorporates a wide range of data in order to produce a far more inclusive and refined projection of likely retirement income. The model projects defined benefit accruals, defined contribution, cash balance, and individual retirement account (IRA) balances, Social Security income, and net housing equity for Americans born between 1936 and 1965, inclusive (VanDerhei and Copeland, 2003). At retirement age, the model simulates 1,000 alternative life paths for each family unit to assess whether the retirement accumulations will be sufficient to pay both basic (deterministic) and health-related (stochastic) expenditures for the simulated life-path, or whether additional outside savings would be required to prevent deficits in retirement. The purpose of this Issue Brief is to show the results obtained by utilizing the concepts already adopted by RSPM for the entire population of certain age cohorts and apply them to stylized examples. These results will provide useful information for individuals attempting to include such crucial factors as longevity, investment, and health care risk into their retirement planning process. In 2007, EBRI plans to roll out Ballpark E$timate® Monte Carlo2⎯ a companion Web site to its current Ballpark E$stimate® retirement planning worksheet⎯that will allow preretirees to determine the appropriate replacement rate target before attempting to determine their desired savings rates. After a brief review of the methodology, this Issue Brief takes the results of the new model and simulates what replacement rates are required to provide “adequate” retirement income 50 percent, 75 percent, and 90 percent of the time.3 Depending on which of the risk elements are introduced into the planning process and what statistical confidence is desired, the new replacement rate targets will be seen to be larger (in some cases considerably so) than the previous benchmarks. Moreover, the huge variation in the range of replacement rate targets—depending on the individual's income, degree of annuitization for initial retirement wealth, and the asset allocation of the post-retirement investments—call into question whether the use of a single rule-of-thumb measure is realistic to use in the retirement planning process. Given the huge variation of individual circumstances (such as age, health, and income) and the complexity of retirement risks that need to be dealt with—such as longevity (addressed through annuitization of assets), old-age infirmity (addressed through long-term care insurance), and asset preservation (addressed through investment allocation)—a simple one-size-fits-all replacement rate will not work for most Americans. The results of this model reveal, in many cases, the sobering (if not staggering) amounts of money needed to provide a reasonable high chance of being able to afford retirement. However, they also show the positive results that can be obtained by annuitizing assets in retirement to protect against the risk of longevity. In this regard, the model points not only to a more realistic size of the retirement income problem but also ways that individuals can begin to deal with it.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

5

Methodology Although the Web-based version of the Ballpark E$timate® Monte Carlo will allow individuals4 to input their specific information, for purposes of illustrating its capacities in this Issue Brief, we start with stylized situations whereby each of the following variables are specified:5 • Gender. • Retirement age. • Equity allocation after retirement. • Percentage of initial retirement wealth to be annuitized at retirement.6 • Preretirement income. • Initial Social Security benefit. • Defined benefit amount (if any).7 At this point the model user is given a choice of: 1. Specifying the amount of initial retirement wealth available8 and having the program simulate a probability of having "adequate" retirement income (defined below); or 2. Specifying the probability of having "adequate" retirement income and having the program simulate the amount of initial retirement wealth needed. In either case, the initial retirement wealth can be specified as (a) a dollar amount, (b) a multiple of final earnings, or (c) a replacement ratio. Once the initial values are chosen, the program runs a large number of simulated life paths9 that include the following calculations on an annual basis for either the specified level of initial retirement wealth in the first option above, or an entire vector of initial retirement wealth balances in the second option: 1. Simulate the annual expenditures and the annual rate of return. 2. Calculate the Social Security benefit, the annuitized portion of the initial retirement wealth (if any) and the amount paid from the defined benefit plan (if any) for the year. 3. If the sum of the simulated investment income and amounts paid from Social Security, the annuitized portion of the initial retirement wealth (if any) and the defined benefit plan (if any) exceed the simulated expenditures for the year, any excess amounts are invested. If the sum of the simulated investment income and amounts paid from Social Security, the annuitized portion of the initial retirement wealth (if any) and the defined benefit plan (if any) is less than the simulated expenditures for the year, any difference is removed from the accumulated retirement wealth. 4. If the accumulated retirement wealth is simulated to turn negative, it is assumed that loans are taken out to continue consumption at the specified levels.10 At the end of the simulated life path,11 the program determines whether there is a non-negative amount left in the retirement wealth balance. If so, there is “adequate” retirement income for this life path. After the entire range of simulated life paths have been run for each level of initial retirement wealth, a probability of “adequate” retirement income is computed. This value is defined as the percentage of simulated life paths with “adequate” retirement income. Replacement rates are computed by taking each of the initial retirement wealth levels and assuming they have been annuitized at retirement age at current annuity purchase prices. The initial value of the assumed initial Social Security annual benefit (before any COLAs) is added to this value.12 The sum of these values is divided by the assumed terminal earnings to produce the replacement rates. Linear interpolation is used to estimate the replacement ratio required to produce “adequate” retirement income for three different levels: 1. 50 percent of the time (this has about the same result as using averages for life expectancy, investment experience, and health care costs, as often done in some applications). 2. 75 percent of the time. 3. 90 percent of the time. EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

6

Although users of the Web version of the model will be allowed to input their own assumptions with respect to investment income, for purposes of illustration, this Issue Brief assumes just two asset classes: (1) a diversified equity portfolio with a stochastic return with a real mean of 6.5 percent and a 100 basis point investment expense, and (2) a non-equity option that is chosen to be deterministic with a net real rate of return equal to 2.5 percent.13 The model allows users to specify their inflation assumptions for several components, but this Issue Brief assumes that general inflation will be 2.8 percent per annum and that inflation for health care components (unless otherwise denoted) will be 7 percent. Retirement expenditure data may be input by the users but guidance will be provided based on four stylized examples with different combinations of preretirement income, Social Security benefits, and retirement expenditures. For purposes of the Issue Brief, retirement expenditures are based on total expenditures net of health care (which is calculated separately) for the following four retirement income categories: 1. 2. 3. 4.

Those making less than $15,000 a year (70.3 percent of single retirees based on the 2000 Consumer Expenditure Study14). Those making at least $15,000 but less than $30,000 (20.6 percent of single retirees). Those making at least $30,000 but less than $40,450 (4.5 percent of single retirees). Those making more than $40,450 (4.6 percent of single retirees).

The 2006 values for retirement expenditures net of health care vary from $15,969 for the lowest income category to $38,097 for the highest income category. 15 Initial Social Security benefits and terminal wages may be input by the users,16 but for purposes of the Issue Brief these amounts were calculated for the four income categories above from 2004 Current Population Survey (CPS) data and brought forth to 2006 by increasing the amounts by 6.9 percent. Early and late retirement assumes terminal wages are adjusted by 3.9 percent per year and that the appropriate actuarial modifications were made to Social Security benefits. Nursing home assumptions are coded into the model with separate simulations for the probability and severity of each event. The probability of being admitted to a nursing home is simulated each year as a function of age and previous health care needs category based on results from the 1999 National Nursing Home Survey (NNHS). The length of stay is simulated based on the duration of stays at nursing homes found in the NNHS. The monthly cost was based on a figure of $3,947 and adjusted to 2006 for inflation.17 Home health care assumptions were simulated in a similar fashion, with the monthly cost based on a figure of $1,280 and adjusted to 2006 for inflation.18 Other health care assumptions are bifurcated based on Medicare eligibility. If the individual is Medicareeligible for any portion of the simulated life path, then the other health care expenditures in that year are equal to the sum of the premiums for Medigap and Medicare Part B plus the simulated drug cost. The annual Medigap premium was assumed to be $1,755 adjusted for future inflation at 7 percent, and the Medicare part B premium was assumed to be $1,062 adjusted for future inflation at 3.9 percent (Fronstin, 2006). The simulated drug cost was based on age-specific quartiles of drug expenses adjusted for Medicare Part D premiums and benefits and adjusted for inflation at 8.6 percent per year (Fronstin, 2006). For ages in the simulated life path when the individual is not Medicare eligible, other health care expenditures are equal to an HMO (health maintenance organization) cost that is age-specific and adjusted for inflation at the general healthcare inflation rate.19 Taxes are based on federal income taxes for a single taxpayer using 2005 tax rates. The amount of Social Security benefits included in taxable income as a result of the 1983 amendments are coded into the program which will result in increased amounts of retirement income needed for future cohorts to pay for the same after tax amount of consumption in retirement. For purposes of this Issue Brief, it is assumed that all retirements take place in the year 2006.20

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

7

Results This section illustrates the problematic nature of using conventional replacement rates for retirement planning through a “building block” approach: Building Block 1 focuses exclusively on investment risk. As retirees increase their equity allocation, they could potentially benefit from a higher expected investment income; however, they will also face more volatility in annual results and a larger potential downside. The calculations presented in this approach use average longevity as well as average long-term care costs for the retirees. Building Block 2 introduces longevity risk into the planning process, in addition to the investment risk from the previous level. In addition to providing the retiree with decisions with respect to investments, this also provides the opportunity to mitigate overall risk through the purchase of immediate annuities at retirement age (as noted earlier, actual annuity purchase prices are utilized in the calculations). The calculations presented in this approach use average long-term care costs for the retirees. Building Block 3 introduces the risk of health care costs into the calculations. This provides the framework necessary to evaluate the potential benefits of long-term care insurance as a way of increasing overall probability of retirement adequacy.21

Building Block 1: Investment Risk Only Figure 1 shows the results of the simulations for a male in the lowest income category retiring at age 65 and living to exactly age 82. If this retiree were to retire with an account balance of $300,000 and chose an asset allocation that included no equities, he would have virtually no chance of having “adequate” retirement income. If the equity allocation were increased to 25 percent, the probability of success increases substantially to 43 percent. Additional increases in the equity allocation result in larger probabilities of adequate retirement income: 50 percent equities yields a 52 percent probability of success, 75 percent equities translates to a 56 percent probability of success, and an all-equity portfolio would provide “adequate” retirement income in 58 percent of the simulated runs. Figures 2 through 4 provide similar results for males retiring at age 65 in income categories 2 though 4 (lower to highest), respectively. Given that expenditures are assumed to increase with income, each successive income Figure 1 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 1 (investment income stochastic, longevity and health care expenses Probability of Adequacy

100% 90% 80%

Equity Allocation

70% 60% 50% 40% 30% 20%

0% 25% 50% 75% 100%

10%

$1 $ 00 0 $2 ,00 00 0 $3 ,000 00 $4 ,00 00 0 $5 ,000 00 $6 ,00 00 0 $7 ,000 00 $8 ,00 00 0 $9 ,000 0 $1 0,0 , 0 00 0 $1 0,0 , 1 00 0 $1 0,0 , 2 00 0 $1 0,0 , 3 00 0 $1 0,0 , 4 00 0 $1 0,0 , 5 00 0 $1 0,0 , 6 00 0 $1 0,0 , 7 00 0 $1 0,0 , 8 00 0 $1 0,0 , 9 00 0 $2 0,0 , 0 00 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Inst it ute, Ballpark E$timate® M onte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

8

category will require a larger initial retirement wealth for the same probability of success. Conversely, the probability of adequate retirement income will decrease for a given initial retirement wealth and equity allocation as the income category increases. The following table demonstrates this: Probability of Adequate Retirement Income for Building Block 1 for a Male Retiring at Age 65 With an Account Balance of $300,000 Equity Allocation Retirement Income Category 0% 25% 50% 75% 100% 1 (Lowest) 0% 43% 52% 56% 58% 2 0 29 47 51 51 3 0 18 39 48 47 4 (Highest) 0 0 4 17 23

The higher expected return but larger volatility of equity returns is demonstrated in these graphs. In each case, a 50 percent probability of adequate retirement income can be provided for a smaller initial retirement wealth when the assets are invested 100 percent in equities. However, the increased volatility of equities causes these lines to cross, and if a retiree desired a 90 percent chance of adequate retirement income, the 100 percent equity allocation would require the largest initial retirement wealth. Some financial planners will attempt to simplify the threshold needed for adequate retirement income by converting the initial retirement wealth to a multiple of final earnings. Figure 5 demonstrates the results of restating the previous figures into this metric for the highest and lowest income categories for 100 percent equity allocations. If the retiree were in the highest income group and desired a 75 percent chance of adequate retirement income, he would require an initial retirement wealth of 4.2 times final earnings if he were to invest 100 percent in equities. However, if the same equity allocation were chosen by a low-income retiree, he would require an initial retirement wealth of 12.1 times final earnings. Figure 6 demonstrates the probability of retirement adequacy expected for each replacement rate for the highest and lowest income categories for both zero and 100 percent equity allocations (Figure 7).22 If a highincome retiree were simply interested in a 50 percent probability of adequacy, he would require a 48 percent replacement rate if he invested 100 percent in equities and a 58 percent replacement if no assets were invested in equities. The same figures jump to 117 percent (all equity) and 146 percent (no equity) for the low-income retiree. If, instead, a male retiring at age 65 desired a 90 percent probability of adequate retirement income, the necessary replacement rate would be 66 percent for a high-income retiree with no investment in equities and 87 percent with 100 percent investment in equities. His low-income counterpart would require a 185 percent replacement rate with no equity investments and a 232 percent replacement rate with 100 percent investment in equities. Figure 8 illustrates the probable levels of adequacy and equity allocations with respect to the required replacement ratios.

Building Block 2: Investment and Longevity Risk This approach relaxes the assumption that the retiree will live to the average life expectancy (the point at which 50 percent of those reaching the retirement age are still alive and 50 percent have passed⎯currently 82 for men and 85 for women at age 65) and introduces longevity risk. Figure 9 provides the same analysis for a low-income retiree as was presented in Figure 1, although when longevity risk is introduced there will be opportunities for those with low initial retirement wealth to have a non-zero probability of adequacy (meaning those who are simulated to die within a few years of retirement before they have exhausted their initial retirement wealth). However, the additional longevity risk requires larger initial retirement wealth for those seeking a 90 percent probability of retirement adequacy. In Figure 1, a zero equity allocation would require approximately $500,000 in initial retirement wealth, whereas in Figure 9, the same probability would need approximately a $600,000 balance. Figure 10 shows similar results for a high-income retiree.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

9

Figure 2 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65, Retirement Income Category 2 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 100% 90% 80%

Probability of Adequacy

Equity Allocation 70%

0%

60%

25%

50%

50% 75%

40%

100%

30% 20% 10%

$0 $1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 3 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65, Retirement Income Category 3 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 100% 90% 80%

Probability of Adequacy

Equity Allocation 70%

0% 60% 50% 40%

25% 50% 75% 100%

30% 20% 10%

$0 $1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

10

Figure 4 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 100% 90% 80%

Probability of Adequacy

Equity Allocation 70%

0% 60%

25% 50%

50%

75% 40%

100%

30% 20% 10%

$0 $1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 5 Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation) For: Males Retiring at Age 65 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

Low Income 60%

High Income

50% 40% 30% 20% 10% 0% -

5

10

15

20

25

30

35

40

Final Earnings Multiple Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

11

Figure 11 provides similar analysis in terms of final earnings multiples and Figure 12 demonstrates the impact of replacement rates on the probability of adequate retirement income by income category, equity allocation and rate of annuitization. Figures 13 and 14 demonstrate the value of purchasing an immediate annuity at retirement age for those interested in a high probability of retirement income adequacy. Assuming a low-income male retiring at age 65 with no equity investments desires a 90 percent chance of retirement income adequacy, the necessary replacement rate without an annuity would be approximately 241 percent. If 25 percent of the initial retirement wealth were annuitized immediately, the replacement rate could be reduced to approximately 213 percent. Additional increases in the percentage of account balances annuitized result in further drops in the necessary replacement rate: 50 percent annuitization would require a replacement rate of 188 percent, 75 percent annuitization would require approximately 177 percent and 100 percent annuitization would require approximately 166 percent. Although Figure 13 was a convenient educational device to illustrate how annuitization may help those desiring higher probabilities of adequate retirement income to achieve it with lower replacement rates, it is important to note that in reality the optimal degree of annuitization will vary by income category as well as asset allocation.23 Figures 15 and 16 demonstrate how the optimal degree of annuitization varies by asset allocation for a low-income individual. In Figure 15, a retiree with no equity allocation would be better off annuitizing if he or she desired more than a 40 percent chance of adequate retirement income; however, in Figure 16, a retiree investing 50 percent of the initial retirement wealth in equities would need at least a 55 percent chance of adequate retirement income as a target before annuitization would be useful. Figures 17 and 18 show similar results for high-income retirees. The tradeoffs between equity allocation and degree of annuitization are shown directly for various probabilities of adequacy in Figures 19, 20, and 21 for those in the lowest income category. Similar analyses are provided for the highest income category in Figures 22, 23, and 24.

Building Block 3: Investment, Longevity, and Long-Term Care Risk For purposes of retirement planning, it is typically assumed that most, if not all, of the retiree expenses will behave in a predictable fashion. For example, in this model, it is assumed that knowing one's retirement income, age, and real estate status, one can make reasonable predictions regarding many of the expenditures in retirement. However, one major exception to this deals with long-term care costs. This section adds in the third of these “building blocks” to try to deal with a situation that could prove financially catastrophic to a retirement plan that otherwise has dealt adequately with investment and longevity risk. Although this Issue Brief does not attempt to provide private market alternatives to dealing with this risk, as it did with longevity risk in the previous section, plans are under way to include such options in the Web-based version of this model. Figure 25 demonstrates the relationship between replacement rates and probability of retirement adequacy by equity allocation and annuitization for each of the four income categories for a male retiring at age 65. Figure 26 shows a similar relationship in terms of final earnings multiples for the highest and lowest income category. The following provides a simple example of the impact of adding long-term care as a stochastic variable for a retiree in the highest income category, assuming no equity allocation and none of the initial retirement wealth is annuitized: Replacement Rate Needed Under: Probability of Adequate Retirement Income 50% 75% 90%

Building Block 2 50% 68 87

Building Block 3 52% 78 119

Difference Between the Two: 2% 10 32

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

12

Figure 6 Replacement Ratios Required for Various Probability Levels of "Adequacy," by Income Category and Equity Allocation For: Males Retiring at Age 65 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 250%

Equity Level Lowest Income, 0% Equity Lowest Income, 100% Equity

200%

Highest Income, 0% Equity

Replacement Rate

Highest Iincome, 100% Equity 150%

100%

50%

0% 50%

75%

90%

Probablity of Retirement Income "Adequacy" Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 7 Necessary Replacement Rates for Retirement Income "Adequacy," by Retirement Income Category and Equity Allocation (Various Probabilities) For: Males Retiring at Age 65 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) Equity Allocation3 Probability of Retirement 1 Income "Adequacy" 50% 75% 90% 50% 75% 90% 50% 75% 90% 50% 75% 90%

Retirement 2 Income Category 1

2

3

4

0% 146% 156 185 75 80 96 61 65 80 57 59 66

25% 131% 150 175 71 78 90 61 66 75 53 58 64

50% 124% 148 178 67 78 93 58 67 77 51 58 69

75% 120% 153 204 66 82 103 55 69 84 48 61 73

100% 117% 166 232 65 88 119 56 73 103 48 63 87

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 20, 2006 version. See text for definition of retirement income adequacy (90% is most likely to have adequate income in retirement). 2 See text for definition of retirement income category (1 is lowest, 4 is highest). 3 Percentage of retirement assets invested in stocks (equities). 1

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

13

Figure 8 Impact of Replacement Rates on the Probability of Retirement Income "Adequacy," by Retirement Income Category and Equity Allocation For: Males Retiring at Age 65 Option: Building Block 1 (investment income stochastic, longevity and health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

Lowest Income, No Equity

60%

Lowest Income, All Equity 50%

Highest Income, All Equity 40%

Highest Income, No Equitiy 30% 20% 10% 0% 0%

100%

200%

300%

400%

500%

600%

700%

Replacement Rate ®

Source: Employee Benefit Research Institute, Ballpark E$timate Monte Carlo, August 2006 version.

Figure 9 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

Equity Allocation 60% 50%

0% 25% 50%

40% 30%

75% 100%

20% 10%

$0 $1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

14

Similar analyses are presented for females retiring at age 65 in Figure 27. The following is a simple summary of the impact of the gender differentials. In each case, it is assumed there is no equity allocation and none of the initial retirement wealth is annuitized:

Lowest-Income Category Replacement Rate Needed for: Probability of Adequate Retirement Income Male Female 50% 124% 147% 75% 229 292 90% 394 453 Highest-Income Category Replacement Rate Needed for: Probability of Adequate Retirement Income Male Female 50% 52% 59% 75% 78 98 90% 119 128

The impact of early retirement on replacement rates is explored in Figure 28. In this case, a male is assumed to retire at age 62, take a permanently subsidized early retirement benefit from Social Security, and pay the cost of health coverage himself until he is eligible for Medicare at age 65. Figure 29 shows the impact of late retirement by analyzing the relationship between replacement rates and probability of adequate retirement income for a male retiring at age 68. In this case, we assume that initial receipt of Social Security benefits is delayed until retirement age. The following is a simple summary of the impact of various retirement ages for males. In each case we assume there is no equity allocation and none of the initial retirement wealth is annuitized:

Highest-Income Category Replacement Rate Needed Under: Probability of Adequate Retirement Income 50% 75% 90%

Male, 65 52% 78 119

Male, 62 64% 97 149

Male, 68 43% 66 97

Lowest-Income Category Replacement Rate Needed Under: Probability of Adequate Retirement Income 50% 75% 90%

Male, 65 124% 229 394

Male, 62 153% 285 476

Male, 68 95% 206 332

The tradeoffs between equity allocation and degree of annuitization are shown directly for various probabilities of adequacy in Figures 30, 31, and 32 for those in the lowest income category. Similar analyses are provided for the highest income category in Figures 33, 34, and 35.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

15

Figure 10 Impact of Initial Retirement Wealth on the Probability of Retirement Income "Adequacy," by Equity Allocation For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70% 60%

Equity Allocation 50%

0% 25%

40%

50% 30%

75% 100%

20% 10%

$$1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth ®

Source: Employee Benefit Research Institute, Ballpark E$timate Monte Carlo, August 2006 version.

Figure 11 Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation and No Annuitization) For: Males Retiring at Age 65 Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

Low Income 60%

High Income 50% 40% 30% 20% 10% 0% 0

10

20

30

40

50

60

70

80

Final Earnings Multiple ®

Source: Employee Benefit Research Institute, Ballpark E$timate Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

16

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

2

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100 65% 62 59 55 55 90 81 78 80 78 120 108 105 105 117

0% 122% 111 105 101 99 178 157 151 146 142 241 213 209 219 237 65% 61 59 57 56 83 79 75 74 75 104 101 98 98 105

25% 119% 113 105 101 100 158 151 143 143 144 213 194 182 195 213 63% 61 59 59 57 81 78 74 74 73 98 93 94 89 91

63% 61 61 59 58 77 74 74 71 72 92 88 86 85 88

Annuitization Level 50% 75% 114% 113% 112 112 105 108 106 109 103 106 151 144 146 141 140 134 139 136 141 134 188 177 180 172 178 155 179 160 179 167 62% 61 61 61 60 74 72 72 71 70 87 85 82 82 80

100% 112% 111 110 110 109 137 137 130 132 133 166 160 153 156 153 4

Income Category 3

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100 50% 46 42 41 40 68 59 58 55 57 87 76 76 78 86

0% 56% 52 49 47 47 75 69 65 64 66 102 91 87 91 105 48% 45 44 42 42 60 57 55 54 56 74 72 67 70 75

25% 55% 52 50 49 48 71 67 63 63 65 91 83 80 83 88

47% 46 43 43 42 57 55 53 53 53 67 66 63 63 66

Annuitization Level 50% 54% 52 51 50 49 67 63 61 62 62 79 76 75 75 76

Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic)

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

For: Males Retiring at Age 65

47% 45 45 44 44 55 53 51 53 51 63 61 59 59 59

75% 53% 52 51 51 50 63 63 62 60 60 75 74 72 71 71

46% 46 46 45 45 52 52 51 51 51 58 58 58 58 58

100% 52% 52 52 52 51 62 61 60 60 59 72 71 69 69 67

Figure 12 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization

Income Category 1

17

Figure 13 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Rate of Annuitization (Assumes Lowest Income Category and No Equity Investment) Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

0% Annuitization

60%

25% Annuitization

50%

50% Annuitization

40%

75% Annuitization

30%

100% Annuitization

20% 10% 0% 0%

100%

200%

300%

400%

500%

600%

700%

Replacement Rate ®

Source: Employee Benefit Research Institute, Ballpark E$timate Monte Carlo, August 2006 version.

Conclusion Most Americans have always been on their own when it comes to savings and the ability to retire well. Since 1937, Social Security has provided a base level of income for those workers who reached retirement age, became disabled, or their survivors. With an average income replacement rate around 40 percent, Social Security was never designed to be the sole source of income. The Internet age for the first time makes savings planning tools readily available to all individuals. All workers can open an individual retirement account, and savings tools can help them determine how much they need to save. The same is true for workers who have a 401(k) or similar savings plan at work. It has always been important for most employees or their advisors to be able to construct a savings plan for retirement that will allow them to determine, inter alia, the appropriate savings rates at an early age. There are various financial programs and calculators available to assist employees with this process, but many of them require the user to input a desired replacement rate or its equivalent. Although there have been many studies that provide readily available rules of thumb, often these are based on methodologies limited to replacement of preretirement cash flow after adjustment for taxes, savings and age and/or work-related expenses. One of the most problematic aspects of using the results of these models is that one or more of the most important retirement risks is ignored: investment risk, longevity risk, and risk of potentially catastrophic health care costs. The Ballpark E$timate® Monte Carlo Web site that EBRI and its American Savings Education Council and Choose to Save® programs will make available to the public in coming months, at www.choosetosave.org, specifically addresses each of these risks and allows the user to determine what replacement rate (or initial retirement wealth expressed in either dollar values or a multiple of earnings approach) will provide a 50-, 75-, or 90 percent chance of successfully providing a specified amount of nonhealth retirement expenditures as well as simulated health care expenses. Users will be able to construct individualized "what-if" scenarios that will provide instant feedback on the changes in replacement rates, dollar values, or multiples of earnings required as a result of changing retirement age, asset allocation, and/or percentage of initial retirement wealth annuitized.24 EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

18

How much of a difference will this process make in the initial determination of a target replacement rate? If it is assumed that there is no equity allocation of assets nor is any of the initial retirement wealth annuitized, the stylized high-income category male modeled in this Issue Brief would need a 52 percent replacement rate in order to have a 50 percent chance of covering his retirement expenses if he retires at age 65.25 If he is not comfortable with a 50–50 prospect of "running out" of retirement income, he could increase his chances of success to 75 percent by raising his income replacement rate to 78 percent; if he wants a 90 percent chance of success, he would have to raise his replacement rate to 119 percent. What would it mean if this high-income male took early retirement as soon as he is eligible for Social Security (age 62)? The replacement rates jump considerably: For a 50–50 chance of success, his income replacement rate would be 64 percent; for a 75 percent chance of success, his replacement rate is 97 percent; and for a 90 percent chance of success, his replacement rate is 149 percent. If, on the other hand, he decides to delay retirement until age 68, he can decrease the figures to 43, 66, and 97 percent, respectively. The same simulations for the stylized low-income category male would require significantly larger replacement rates, since non-health care retirement expenditures are not reduced proportionally with retirement income nor is the health care expense typically a function of income. In this case, a low-income male retiring at 65 would have to replace 124 percent of his annual income just for a 50 percent chance of success, while a 75 percent chance requires a replacement rate of 229 percent; for a 90 percent chance of success, he would require a huge increase in the replacement rate to 394 percent! Similar modifications for early and late retirement would take place at this income level. Whether a significant percentage of workers will be able to accumulate sufficient retirement wealth to achieve these replacement rate targets is a question that requires separate modeling. EBRI is currently in the process of updating the EBRI/ERF Retirement Security Projection Model® to account for the changes recently enacted in the Pension Protection Act of 2006 and will use the projected retirement accumulations from that model to determine the impact on various demographic groups under a variety of assumptions with respect to plan sponsor and plan participant reactions to the new provisions.26 As longevity continues to increase, as health expenses continue to climb, and as Social Security replacement rates slowly decline, the need for individuals to save for retirement has never been greater. The need has always been great for detail beyond a “rule-of-thumb” replacement target of “75 to 85 percent,” but only now are free tools being made broadly available via the Internet so that all individuals have the opportunity to do the needed planning without the necessity of seeking professional advice.

References Alford, Susan, D. Bryan Farnen, and Mike Schachet. “Light At The End Of The Tunnel: Getting On Track for Affordable Retirement.” Benefits Quarterly (4th Quarter, 2004). Ameriks, John, Robert Veres, and Mark J. Warshawsky. “Making Retirement Income Last a Lifetime.” Journal of Financial Planning (December 2001). Fronstin, Paul. “Savings Needed to Fund Health Insurance and Health Care Expenses in Retirement.” EBRI Issue Brief no. 295 (Employee Benefit Research Institute, July 2005). Holden, Sarah, and Jack VanDerhei. “The Influence of Automatic Enrollment, Catch-Up and IRA Contributions on 401(k) Accumulations in Retirement.” EBRI Issue Brief no. 283 (Employee Benefit Research Institute, July 2005). Hurd, Michael D., and Susann Rohwedder. “Alternative Measures of Replacement Rates.” Prepared for the 8th Annual Joint Conference for the Retirement Research Consortium, “Pathways to a Secure Retirement.” August 10–11, 2006. Schieber, Sylvester J. “Conceptual and Measurement Problems in Contemporary Measures of Income Needs in Retirement.” Benefits Quarterly. Vol. 12, no. 2 (Second Quarter 1996): 56–68. EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

19

____. “Deriving Preretirement Income Replacement Rate Targets and the Savings Rates Needed to Meet Them.” Benefits Quarterly. Vol. 14, no. 2 (Second Quarter 1998): 53–69. Steinberg, Allen and Lori Lucas. “Shifting Responsibility: The Future of Retirement Adequacy In America.” Benefits Quarterly (4th Quarter, 2004). VanDerhei, Jack, and Craig Copeland. “Can America Afford Tomorrow’s Retirees: Results From the EBRIERF Retirement Security Projection Model.” EBRI Issue Brief no. 263 (Employee Benefit Research Institute, November 2003). VanDerhei, Jack. “Measuring Retirement Income Adequacy, Traditional Replacement Ratios, and Results for Workers at Large Companies.” EBRI Notes, no. 9 (Employee Benefit Research Institute, September 2004).

Endnotes 1

Indeed, the choice of a replacement rate is often the first step in determining the appropriate savings rate during the accumulation process. See the interactive Ballpark E$stimate at http://choosetosave.org/ballpark/ for an example.

2

A method for computer modeling based on chance; a technique for producing distributions of outcomes of stochastic (variable) processes by running many iterations of the model process. Distinct from “deterministic” projections, which are based on set assumptions that exclude the possibility of variation. 3

The terms adequate or adequacy when used to describe the results in this Issue Brief refer to whether there was sufficient retirement income under a simulated set of circumstances to allow the individual to maintain a pre-specified consumption pattern in retirement as well as the ability to afford simulated health care expenditures. The consumption patterns used in the four hypothetical examples are averages from government surveys and in no way represent a value judgment on whether these amounts are sufficient in a social welfare context. Nor do they necessarily reflect an individual's perception of adequacy. However, the Web version of the model will allow an individual to choose what he or she believes to be an adequate level of expenses.

4

Currently, the model is designed only to simulate results for individuals, but a future version of the model will allow families to use the model in an integrated manner. 5

The model also differentiates the calculations based on whether the retiree rents or owns a house and, if the latter, whether there is still a mortgage on the house at the time of retirement

6

Currently, the only option in the model is to purchase an immediate annuity at retirement. Further enhancements are currently in the planning stages, including the use of longevity insurance. 7

For purposes of simplicity, the analysis in this Issue Brief assumes the retiree does not have an accrued benefit from a defined benefit plan. However, the lump-sum equivalent of the benefit could be computed to use with this analysis. 8

In most cases, it is assumed that the vast majority of the initial retirement wealth is composed of 401(k) and IRA balances that are not Roth variants of these programs. Although the program allows input of both taxable and nontaxable amounts, for purposes of the Issue Brief it is assumed that all amounts other than certain specified amounts of Social Security benefits will be subject to federal income tax. 9

The numbers illustrated in this Issue Brief are based on 1,000 simulated life paths.

10

In reality, retirees may be forced to decrease their consumption instead.

11

This is defined as life expectancy for Building Block 1 approach (below) and a stochastic age for Building Blocks 2 and 3.

12

It is important to realize that this methodology is implicitly adding a nominal annuity to a real annuity; however, the purchase of the latter appears to be de minimis. The Web version of the program will allow real annuities to be selected for the initial retirement wealth if the user prefers.

13

This option was chosen to reflect a guaranteed investment contract (GIC) or stable value option, where the rate of return realized is net of expenses.

14

The Consumer Expenditure Survey (CES) is conducted by the Bureau of Labor Statistics of the U.S. Department of Labor. The survey targets the total noninstitutionalized population (urban and rural) of the United States and is the basic source of data for revising the items and weights in the market basket of consumer purchases to be priced for the Consumer Price Index. The CES data allow for the total expenditures that these elderly individuals make annually by

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

20

gender and home ownership. The expenditures used in this study are the total expenditures net of health care costs for just the single individuals of the respective genders. 15

Hurd and Rohwedder (2006) report 2001 CES expenditure averages of $31,700 for those aged 65−74 and $22,800 for those age 75 or over. These numbers include health care expenditures and include expenditures from families. They also compare these numbers with those derived from the Consumption and Activities Mail Survey (CAMS) and find larger expenditure averages: 12 percent larger for the younger group and 30 percent larger for the older group. The authors speculate the differences may be due to the problem of the reference person and the difference in numbers reflects a difference in the populations represented. We are unable to discern which set of expenditure numbers are likely to be more representative; however, it is clear that the replacement rates derived in this study for any specific set of circumstances would increase if the CAMS expenditure numbers were used. 16

A link to the benefits estimators on the SSA.gov Web site will be provided in the Web site interface.

17

The estimates were a national average of monthly nursing home costs found from the NNHS. The NNHS is a nationwide sample survey of nursing homes, their current residents and discharges that was conducted by the National Center for Health Statistics from July through December 1999. The home health care use, length of stay (use), and monthly expenses were determined from the 2000 National Home and Hospice Care Survey (NHHCS). The NHHCS is a nationwide sample survey of home health and hospice care agencies, their current and discharged patients that was conducted by the National Center for Health Statistics from August through December 2000. 18

19

We utilized the CareFirst BlueChoice HMO HIPAA plan quotes for the Washington, DC area from their Web site in 2006.

20

No state or local income tax is assumed in this Issue Brief, but Web users will be able to input values to override the default values.

21

This evaluation feature will be added to the model once we determine the best way to introduce actual or generic long-term care insurance policy parameters and pricing information into the model.

22

Figure 7 provides detailed information for all four retirement income categories.

23

Ameriks, Veres, and Warshawsky (2001) found similar interactions.

24

Future versions of the model will also allow for the purchase of either generic or specific long-term care insurance policies.

25

The requisite replacement rates for females would be slightly larger in each instance due to their longer life expectancies.

26

One potentially positive note for future retirement income adequacy deals with the automatic enrollment provisions in the pension reform law. Holden and VanDerhei (2005) used a similar simulation technique to predict the impact of various combinations of default contributions and default asset allocations on future 401(k) account balances. The one that most closely approximates the provisions in the legislation provided increases of 126 percent in the average account balances for those in the lowest income quartile of all 401(k) participants.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

21

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

22

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

Probability of % of Equity Adequacy Investment 50% 0% 50 25 50 50 50 75 50 100 75% 0% 75 25 75 50 75 75 75 100 90% 0% 90 25 90 50 90 75 90 100 65% 62 59 55 55 90 81 78 80 78 120 108 105 105 117

0% 122% 111 105 101 99 178 157 151 146 142 241 213 209 219 237 65% 61 59 57 56 83 79 75 74 75 104 101 98 98 105

25% 119% 113 105 101 100 158 151 143 143 144 213 194 182 195 213 63% 61 59 59 57 81 78 74 74 73 98 93 94 89 91

Annuitization 50% 114% 112 105 106 103 151 146 140 139 141 188 180 178 179 179 63% 61 61 59 58 77 74 74 71 72 92 88 86 85 88

75% 113% 112 108 109 106 144 141 134 136 134 177 172 155 160 167 62% 61 61 61 60 74 72 72 71 70 87 85 82 82 80

100% 112% 111 110 110 109 137 137 130 132 133 166 160 153 156 153 4

Income Category 3

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

Probability of % of Equity Adequacy Investment 50% 0% 50 25 50 50 50 75 50 100 75% 0% 75 25 75 50 75 75 75 100 90% 0% 90 25 90 50 90 75 90 100 50% 46 42 41 40 68 59 58 55 57 87 76 76 78 86

0% 56% 52 49 47 47 75 69 65 64 66 102 91 87 91 105 48% 45 44 42 42 60 57 55 54 56 74 72 67 70 75

25% 55% 52 50 49 48 71 67 63 63 65 91 83 80 83 88

47% 46 43 43 42 57 55 53 53 53 67 66 63 63 66

Annuitization 50% 54% 52 51 50 49 67 63 61 62 62 79 76 75 75 76

47% 45 45 44 44 55 53 51 53 51 63 61 59 59 59

75% 53% 52 51 51 50 63 63 62 60 60 75 74 72 71 71

Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic)

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

2

Income Category 1

For: Males Retiring at Age 65

46% 46 46 45 45 52 52 51 51 51 58 58 58 58 58

100% 52% 52 52 52 51 62 61 60 60 59 72 71 69 69 67

Figure 14 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization

Figure 15 Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 0% Equity Allocation) For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% Annuitization Percentage

70% 60%

0%

50%

25%

40%

50% 75% 100%

30% 20% 10%

$$1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

Figure 16 Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 50% Equity Allocation) For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80% 70%

Annuitization Percentage

60%

0% 25% 50% 75% 100%

50% 40% 30% 20% 10%

$1

$0 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

23

Figure 17 Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 0% Equity Allocation) For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probability of Adequacy

80%

Annuitization Percentage

70% 60%

0%

50%

25% 50%

40%

75% 100%

30% 20% 10%

0, 00 $1 0 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0, 00 0

$9 0

0, 00 0

$8 0

0, 00 0

$7 0

0, 00 0

$6 0

0, 00 0

$5 0

0, 00 0

$4 0

0, 00 0

$3 0

$2 0

$-

$1 0

0, 00 0

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 18 Impact of Defined Contribution Balance on Probability of "Adequate" Retirement Income, by Annuitization (Assumes 50% Equity Allocation) For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90%

Probabilityd of Adequacy

80%

Annuitization Percentage

70%

0%

60%

25%

50%

50% 40%

75%

30%

100%

20% 10%

$0 $1 00 ,0 00 $2 00 ,0 00 $3 00 ,0 00 $4 00 ,0 00 $5 00 ,0 00 $6 00 ,0 00 $7 00 ,0 00 $8 00 ,0 00 $9 00 ,0 00 $1 ,0 00 ,0 00 $1 ,1 00 ,0 00 $1 ,2 00 ,0 00 $1 ,3 00 ,0 00 $1 ,4 00 ,0 00 $1 ,5 00 ,0 00 $1 ,6 00 ,0 00 $1 ,7 00 ,0 00 $1 ,8 00 ,0 00 $1 ,9 00 ,0 00 $2 ,0 00 ,0 00

0%

Initial Retirement Wealth Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

24

Figure 19 Replacement Ratios Required for a 50% Chance of "Adequacy" For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 140%

120%

Replacement Rate

100%

Equity Level 80%

0% 25%

60%

50% 75%

40%

100% 20%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 20 Replacement Ratios Required for a 75% Chance of "Adequacy" For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 200% 180% 160%

Replacement Rate

140% 120%

Equity Allocation 100%

0% 25%

80%

50% 60%

75% 100%

40% 20% 0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

25

Figure 21 Replacement Ratios Required for a 90% Chance of "Adequacy" For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 300%

250%

Replacement Rate

200%

150%

Equity Level 0%

100%

25% 50% 75%

50%

100%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 22 Replacement Ratios Required for a 50% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 60%

50%

Replacement Rate

40%

Equity Level 30%

0% 25% 20%

50% 75% 100%

10%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

26

Figure 23 Replacement Ratios Required for a 75% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 80%

70%

Replacement rate

60%

50%

Equity Level 40%

0% 25%

30%

50% 20%

75% 100%

10%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 24 Replacement Ratios Required for a 90% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 2 (investment income and longevity stochastic, health care expenses deterministic) 100% 90% 80%

Replacement Rate

70% 60%

Equity Level 50%

0% 40%

25% 50%

30%

75% 20%

100%

10% 0%

0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

27

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

28

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

62% 61 62 58 58 114 104 99 93 92 142 135 133 130 138

64% 61 59 59 57 110 103 95 89 85 160 156 143 143 149

65% 63 60 59 58 120 103 91 93 92 180 160 148 155 168

63% 62 58 59 59 117 104 100 89 88 155 141 132 124 141

Annuitization Level 25% 50% 75% 124% 111% 111% 114 112 111 110 105 105 102 109 107 102 106 105 229 228 224 215 205 208 184 186 205 176 177 187 176 175 177 394 301 291 296 293 280 290 283 269 289 268 281 301 293 286

0% 124% 115 109 104 105 229 210 195 175 179 394 335 302 289 353

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

2

50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Income Probability % of Equity Category of Adequacy Investment 50% 0% 1

For: Males Retiring at Age 65

61% 60 59 59 58 113 108 100 95 93 145 134 141 131 142

100% 111% 102 103 101 103 227 212 197 191 178 302 292 290 284 292 4

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

50 50 50 50 75% 75 75 75 75 90% 90 90 90 90 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

Income Probability % of Equity Category of Adequacy Investment 3 50% 0%

52% 48 43 43 43 78 70 66 63 67 119 107 96 98 111

0% 56% 53 51 50 50 97 89 80 74 79 152 125 125 121 132 51% 46 45 43 43 74 66 63 62 62 106 96 90 92 97

47% 47 45 45 43 72 69 62 59 63 94 89 87 87 84

47% 45 45 44 44 72 69 65 62 58 89 82 84 80 83

Annuitization Level 25% 50% 75% 56% 53% 52% 52 52 51 51 50 51 50 51 50 50 51 50 98 93 94 85 84 86 78 81 82 76 76 76 76 72 73 135 121 118 116 115 111 112 110 109 108 111 109 119 111 106

Option: Building Block 3 (investment income, longevity, and health care expenses stochastic)

45% 45 44 45 45 71 69 67 63 61 87 88 85 87 86

100% 51% 51 50 51 50 90 86 82 80 74 122 113 109 112 112

Figure 25 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization

Figure 26 Impact of Final Earnings Multiple on the Probability of Retirement Income "Adequacy," by Retirement Income Category (Assumes 100% Equity Allocation and No Annuitization) For: Males Retiring at Age 65 Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 100% 90%

Probability of Adequacy

80% 70% Low Income 60% 50%

HIgh Income

40% 30% 20% 10% 0% 0

10

20

30

40

50

60

70

80

Final Earnings Multiple Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

29

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

30

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

2

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100 79% 70 70 66 65 147 124 112 107 108 204 187 172 162 199

0% 147% 135 131 122 114 292 238 214 206 216 453 373 348 364 374 76% 74 67 66 64 135 118 108 106 103 192 167 162 158 162

25% 141% 142 124 119 116 276 237 212 200 212 411 358 320 318 368 74% 73 67 70 66 130 119 110 101 105 170 157 144 149 146

70% 70 69 66 64 126 115 110 104 101 158 152 139 147 140

Annuitization Level 50% 75% 143% 135% 129 122 133 128 118 118 121 118 258 256 231 234 215 221 205 211 202 199 382 337 313 316 306 299 312 281 313 307

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Income Category 1

For: Females Retiring at Age 65

71% 69 66 63 63 121 117 110 108 112 159 169 153 148 152

100% 138% 123 118 118 116 247 228 223 218 198 338 316 311 287 309 4

Income Category 3

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

Probability % of Equity of Adequacy Investment 50% 0% 50 25 50 50 50 75 50 100 75% 0% 75 25 75 50 75 75 75 100 90% 0% 90 25 90 50 90 75 90 100 59% 51 49 47 46 98 79 73 69 69 128 113 103 116 125

0% 71% 61 58 56 53 118 100 95 88 89 171 148 139 138 157 55% 52 51 48 47 88 77 71 67 69 117 107 100 101 108

25% 65% 60 60 58 55 113 98 89 86 85 155 134 128 134 136

53% 53 51 48 48 85 76 71 67 67 106 99 93 94 97

Annuitization Level 50% 63% 59 58 57 56 104 97 89 85 82 138 129 118 118 122

50% 53 50 48 48 80 75 70 69 69 99 90 90 90 93

75% 60% 60 64 59 55 103 96 93 86 87 132 124 112 122 117

Option: Building Block 3 (investment income, longevity, and health care expenses stochastic)

51% 50 50 49 49 77 74 71 70 69 102 96 92 93 93

100% 61% 58 58 57 56 99 94 92 89 90 129 123 124 121 121

Figure 27 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

31

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

2

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100 85% 83 75 72 70 139 128 121 110 109 234 203 186 188 198

0% 153% 139 135 126 124 285 259 225 222 231 476 398 391 393 419 82% 80 75 72 71 139 124 112 114 109 208 192 164 165 171

25% 150% 142 124 133 126 290 232 227 210 205 421 372 342 343 354 79% 77 75 74 73 136 127 113 112 109 179 165 162 157 167

79% 75 74 75 74 136 128 118 108 104 175 160 158 158 159

Annuitization Level 50% 75% 143% 136% 139 134 129 134 128 129 131 126 249 268 240 248 220 233 209 207 215 210 373 339 351 313 319 310 329 313 330 329 75% 75 73 74 74 130 125 120 120 116 182 163 158 162 167

100% 131% 126 127 126 127 265 244 228 224 218 358 322 321 323 335 4

Income Category 3

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100 64% 61 56 52 50 97 89 81 84 81 149 125 122 124 127

0% 75% 68 65 62 63 118 109 98 96 99 196 163 155 153 176 62% 58 57 53 55 90 83 76 78 78 130 114 108 110 116

25% 70% 68 65 63 64 110 106 95 93 91 164 145 142 141 143

60% 59 57 55 54 91 81 76 76 71 121 107 103 107 104

Annuitization Level 50% 69% 66 64 65 62 114 103 96 94 87 151 139 136 127 128

Option: Building Block 3 (investment income, longevity, and health care expenses stochastic)

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Income Category 1

For: Males Retiring at Age 62

59% 57 56 56 55 90 82 78 75 74 111 98 98 96 105

75% 68% 64 65 64 63 113 104 100 92 91 114 132 126 128 129

56% 56 55 55 56 87 81 78 77 75 111 108 102 102 102

100% 64% 64 63 64 63 107 105 99 96 91 141 137 122 138 136

Figure 28 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation and Rate of Annuitization

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

32

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

2

0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

93% 89 91 87 90 199 192 175 152 150 292 271 250 256 249 51% 51 51 50 50 94 91 81 83 79 141 129 113 122 129

53% 52 50 50 51 110 96 83 80 83 156 141 127 139 121

25%

95% 95 87 91 91 206 184 167 153 157 332 290 268 280 297

0%

52% 50 51 49 51 99 95 97 82 76 125 121 115 115 126

92% 89 90 90 87 203 185 165 165 158 266 242 236 225 235 53% 52 51 50 51 98 92 88 86 80 123 116 115 111 122

93% 90 91 90 90 200 191 175 160 155 261 242 244 251 237

Annuitization Level 50% 75%

51% 52 50 52 51 100 92 89 84 83 121 125 120 118 118

93% 92 90 91 90 196 190 176 168 154 258 256 246 238 240

100%

4

3

Income Category

50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

% of Equity Investment 0% 25 50 75 100 0% 25 50 75 100 0% 25 50 75 100

43% 40 39 38 37 66 60 55 55 56 97 86 85 85 96

46% 46 44 43 44 77 77 66 63 64 123 111 100 104 111

0%

41% 39 39 38 37 60 60 55 53 52 89 79 83 79 75

45% 43 43 43 43 76 76 67 66 63 112 101 95 94 104

25%

39% 39 38 38 38 62 60 57 52 52 81 77 73 73 76

45% 43 43 43 43 82 76 71 68 67 104 97 95 91 98

39% 38 38 38 39 63 62 59 57 52 76 73 73 73 74

44% 43 43 43 43 81 76 75 70 64 101 96 96 100 93

Annuitization Level 50% 75%

Option: Building Block 3 (investment income, longevity, and health care expenses stochastic)

Source: Employee Benefit Research Institute, Ballpark E$timate ® Monte Carlo, August 2006 version.

1

Probability of Adequacy 50% 50 50 50 50 75% 75 75 75 75 90% 90 90 90 90

Income Category

For: Males Retiring at Age 68

38% 38 38 38 38 62 61 58 56 54 74 73 73 73 73

43% 44 43 43 43 79 77 71 70 67 96 100 95 95 100

100%

Figure 29 Impact of Replacement Rates on Probability of "Adequate" Retirement Income, by Income Category, Equity Allocation, and Rate of Annuitization

Figure 30 Replacement Ratios Required for a 50% Chance of "Adequacy" For: Males Retiring at Age 65, Lowest Income Category Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 140%

120%

Replacement Rate

100%

80%

Equity Allocation 60%

0% 25% 40%

50% 75% 100%

20%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 31 Replacement Ratios Required for a 75% Chance of "Adequacy" For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 250%

Replacement Rate

200%

150%

Equity Allocation 0%

100%

25% 50% 75%

50%

100%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

33

Figure 32 Replacement Ratios Required for a 90% Chance of "Adequacy" For: Males Retiring at Age 65 in the Lowest Income Category Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 450%

400%

350%

Replacement Rate

300%

250%

200%

Equity Allocation 0%

150%

25% 50%

100%

75% 100%

50%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 33 Replacement Ratios Required for a 50% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 60%

50%

Replacement Rate

40%

30%

Equity Level No Equity 25%

20%

50% 75% All Equity

10%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

34

Figure 34 Replacement Ratios Required for a 75% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) 90%

80%

70%

Replacement Rate

60%

50%

Equity Level 40%

0% 25%

30%

50% 75%

20%

100% 10%

0% 0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

Figure 35 Replacement Ratios Required for a 90% Chance of "Adequacy" For: Males Retiring at Age 65 in the Highest Income Category Option: Building Block 3 (investment income, longevity, and health care stochastic) 140%

120%

Replacement Rate

100%

80%

60%

Equity Allocation 0%

40%

25% 50% 75%

20%

100% 0%

0%

25%

50%

75%

100%

Degree of Annuitization Source: Employee Benefit Research Institute, Ballpark E$timate® Monte Carlo, August 2006 version.

EBRI Issue Brief No. 297 • September 2006 • www.ebri.org

35

EBRI Issue Brief EBRI Employee Benefit Research Institute Issue Brief (ISSN 0887−137X) is published monthly by the Employee Benefit Research Institute, 2121 K Street, NW, Suite 600, Washington, DC 20037-1896, at $300 per year or is included as part of a membership subscription. Periodicals postage rate paid in Washington, DC, and additional mailing offices. POSTMASTER: Send address changes to: EBRI Issue Brief, 2121 K Street, NW, Suite 600, Washington, DC 20037-1896. Copyright 2006 by Employee Benefit Research Institute. All rights reserved, No. 297.

Who we are What we do Our publications

Orders/ subscriptions

The Employee Benefit Research Institute (EBRI) was founded in 1978. Its mission is to contribute to, to encourage, and to enhance the development of sound employee benefit programs and sound public policy through objective research and education. EBRI is the only private, nonprofit, nonpartisan, Washington, DC-based organization committed exclusively to public policy research and education on economic security and employee benefit issues. EBRI’s membership includes a cross-section of pension funds; businesses; trade associations; labor unions; health care providers and insurers; government organizations; and service firms. EBRI’s work advances knowledge and understanding of employee benefits and their importance to the nation’s economy among policymakers, the news media, and the public. It does this by conducting and publishing policy research, analysis, and special reports on employee benefits issues; holding educational briefings for EBRI members, congressional and federal agency staff, and the news media; and sponsoring public opinion surveys on employee benefit issues. EBRI’s Education and Research Fund (EBRI-ERF) performs the charitable, educational, and scientific functions of the Institute. EBRI-ERF is a tax-exempt organization supported by contributions and grants. EBRI Issue Brief is a periodical providing expert evaluations of employee benefit issues and trends, as well as critical analyses of employee benefit policies and proposals. EBRI Notes is a periodical providing current information on a variety of employee benefit topics. EBRI’s Pension Investment Report provides detailed financial information on the universe of defined benefit, defined contribution, and 401(k) plans. EBRI Fundamentals of Employee Benefit Programs offers a straightforward, basic explanation of employee benefit programs in the private and public sectors. EBRI Databook on Employee Benefits is a statistical reference volume on employee benefit programs and work force related issues. Contact EBRI Publications, (202) 659-0670; fax publication orders to (202) 775-6312. Subscriptions to EBRI Issue Briefs are included as part of EBRI membership, or as part of a $199 annual subscription to EBRI Notes and EBRI Issue Briefs. Individual copies are available with prepayment for $25 each (for printed copies) or for $7.50 (as an e-mailed electronic file) by calling EBRI or from www.ebri.org. Change of Address: EBRI, 2121 K Street, NW, Suite 600, Washington, DC 20037, (202) 659-0670; fax number, (202) 775-6312; e-mail: Publications [email protected]. Membership Information: Inquiries regarding EBRI membership and/or contributions to EBRI-ERF should be directed to EBRI President/ASEC Chairman Dallas Salisbury at the above address, (202) 659-0670; e-mail: [email protected]

Editorial Board: Dallas L. Salisbury, publisher; Steve Blakely, editor. Any views expressed in this publication and those of the authors should not be ascribed to the officers, trustees, members, or other sponsors of the Employee Benefit Research Institute, the EBRI Education and Research Fund, or their staffs. Nothing herein is to be construed as an attempt to aid or hinder the adoption of any pending legislation, regulation, or interpretative rule, or as legal, accounting, actuarial, or other such professional advice. EBRI Issue Brief is registered in the U.S. Patent and Trademark Office. ISSN: 0887−137X/90 0887−137X/90 $ .50+.50

Did you read this as a pass-along? Stay ahead of employee benefit issues with your own subscription to EBRI Issue Brief for only $89/year electronically e-mailed to you or $199/year printed and mailed. For more information about subscriptions, visit our Web site at www.ebri.org or complete the form below and return it to EBRI. Name Organization Address City/State/ZIP Mail to: EBRI, 2121 K Street, NW, Suite 600, Washington, DC 20037 or Fax to: (202) 775-6312 © 2006, Employee Benefit Research Institute−Education and Research Fund. All rights reserved.