In this post we are going to revisit our hypothetical retiree, Mike, whom we talked about in our posts of April 16, 2015 and February 25, 2015. As you may recall, Mike is a single 65 year-old male retiree who is eligible to receive a Social Security benefit of $16,800 per annum and he also has accumulated savings of $750,000 but no other sources of retirement income. As we saw in the previous posts, if Mike uses the Excluding Social Security spreadsheet from this website and the recommended assumptions (and no amount to be left to heirs), he will develop an initial spending budget of $49,427 ($16,800 from Social Security plus a $32,627 withdrawal from his accumulated savings). If all the recommended assumptions are unchanged and exactly realized each year in the future and Mike spends exactly his spending budget each year, his spending budget would be expected to remain constant in real dollar terms until he reaches about age 90. If he survives past age 90, his spending budget would be expected to decline somewhat as his age plus his life expectancy starts to exceed 95 (under the mortality assumptions in the 2012 Society of Actuaries Individual Annuitant mortality table with 1% projection).
Here is Mike’s Actuarial Balance Sheet as of his 65th birthday for this base case. Present values are determined using the recommended spreadsheet assumptions (4.5% discount rate, 2.5% inflation increases and death at age 95)
 |
| (click to enlarge) |
As indicated in the previous posts, Mike is not pleased with his spending budget and he has looked at a number of alternatives to increase early year spending. Mike knows that his life expectancy is 23 years under the Society of Actuaries mortality table. Therefore, he knows that assuming a 30-year payout period is likely to be conservative and leave money unspent upon his death. He also knows, however, that if he uses his life expectancy as the expected payout period, his spending budgets will decline in future years as life expectancy does not decrease by one year for each year that a retiree ages (see our post of December 3, 2014 for a graph of this effect). Mike is also aware of experts who say that many retirees spend less in real dollar terms as they age. So, Mike feels that there is some conservatism built into the recommended assumptions that he can exploit to increase his near-term spending budgets.
Mike’s first step is to see how much his essential spending is. He determines that his essential spending needs are about $40,000 per year. With respect to his essential spending, however, Mike feels that it is important to be conservative both with respect to the expected payment period of 30 years and with respect to the desire to maintain constant purchasing power. With a little playing around with the Excluding Social Security spreadsheet and QLAC purchase rates from Immediateannuities.com, Mike sees that if he designates $415,000 of his accumulated savings to essential spending, his entire Social Security benefit and spends $70,000 to purchase a deferred annuity starting at age 85 (with no benefit for death prior to that age), he can generate an initial essential spending budget of $40,074 ($16,800 from Social Security plus $23,274 from accumulated savings) that is expected to remain constant in real dollars over the next 30 years.
This leaves Mike with $265,000 in accumulated savings ($750,000 - $415,000 dedicated to essential spending - $70,000 for purchase of the QLAC). With respect to this $265,000 that he has decided to dedicate to non-essential spending, Mike is more willing to front-load this spending. He decides that he will target his spending over his remaining life expectancy (not 30 years) and he will not build in any increases for future inflation. Using the Excluding Social Security spreadsheet, he enters 23 for expected payout and 0% for desired increases due to inflation. This gives him an initial non-essential spending budget of $17,924 and a total initial spending budget of $57,998 ($40,074 essential plus $17,924 non-essential). This spending budget is approximately 17% higher than his base spending budget 0f $49,427.
Mike knows that his total spending budget will decline in real terms from year to year if all assumptions are realized. In fact, he estimates that his non-essential spending budget will only be about $5,100 in real dollar terms at age 89.
Here is Mike’s Revised Actuarial Balance Sheet reflecting purchase of the QLAC.
 |
| (click to enlarge) |
It is important to note that even though Mike only spent $70,000 for the QLAC, the present value of benefits expected to be received under that contract is about $120,000 as Mike is assuming that he will live until age 95 (not his life expectancy assumed by the insurance company). Thus, from a pure budget perspective (and not necessarily from an investment perspective), the purchase of the QLAC is a smart move. He is using the mortality premium from the insurance contract to more cheaply fund future essential expenses than he can with his accumulated savings.
Could Mike use a conservative approach for his essential spending and a less conservative approach for his non-essential spending and still obtain his desired increased spending budget with the 4% Rule, any safe withdrawal rate rule, or the Guyton decision rules? Not bloody likely. That is why smart retirees and their financial advisors should chooose the Actuarial Approach rather than some “simple” rule of thumb.
As I said in my previous post, any spending approach that does not periodically match a retiree’s assets with her liabilities runs a significant risk of failing to achieve the retiree’s spending objectives. What are the retiree’s assets? They include her accumulated savings and the present value of retirement income from other sources (such as annuities or pensions). What are the retiree’s liabilities? These include the present value of future annual spending budgets, the present value of the amount the retiree wishes to leave to heirs and the present value of other expenses such as long-term care.
This post will illustrate, with an example, the matching of assets and liabilities achieved by the Actuarial Approach advocated in this website.
Let’s assume that we have a hypothetical retiree named Mary who is age 65. She has $1,000,000 in accumulated savings, a fixed dollar pension of $15,000 per year, a Social Security benefit of $20,000 per year and no other sources of income. She has determined that her essential expenses in retirement will be about $50,000 per year. Since believes that these essential expenses will stay reasonably constant in real dollar terms from year to year. She would also like to leave around $500,000 (in future dollars) to her daughter at her death or have that money available for long-term care or extra medical bills if needed. She also believes that she will need something like $100,000 for unexpected expenses not included in her essential expense budget, such as purchases of new automobiles or gifts to her daughter.
Mary goes to the “Excluding Social Security” spreadsheet to see how much of her accumulated savings it will take, together with her pension and her Social Security benefit to cover her $50,000 annual real dollar essential expense budget and still leave her with $500,000 at her expected death. She enters $600,000 in accumulated savings, $15,000 in annual pension, $500,000 to be left to heirs at death and the recommended assumptions (including a desired annual increase rate applicable to future budgets attributable to savings and pension of 2.5% per annum). The resulting spending budget when Social Security is added is $51,401. Since this is close to her estimate of essential expenses she decides she will dedicate $600,000 of her accumulated savings to her “essential expenses” budget along with her pension and her Social Security benefits. She may even invest these essential expense assets differently than her other accumulated savings.
To cover unexpected expenses, Mary dedicates $100,000 of her assets to this budget item. Since she feels that the amount she desires to leave to her daughter at her death can serve several purposes, Mary does not feel it is necessary to dedicate additional assets to cover rising health costs or long-term care expenses. Finally, Mary wishes to travel early in her retirement and have an active social life. She dedicates her remaining $300,000 of accumulated savings toward non-essential spending but she wishes to front-load this budget item. Therefore, she enters $300,000 in the Excluding Social Security spreadsheet with 0% increase in the annual desired increase. The result for the first year is a non-essential spending budget of $17,624. Mary knows that this budget item will not increase in nominal terms from year to year and therefore will represent a declining real spending budget as she ages.
Mary’s first year spending budget is $69,025 (a total of $51,401 from Social Security, her pension and her essential assets) plus $17,624 from her non-essential assets. The exhibit below shows Mary’s Actuarial Balance Sheet as of her date of retirement. The present values are based on the recommended assumptions and results from the Excluding Social Security spreadsheet.
 |
| Click to enlarge |
Mary will revisit her spending budget thought process at the beginning of each new year. She will use the Excluding Social Security spreadsheet and enter new data and new assumptions. Investment experience may deviate from the assumptions she used. Her spending may deviate from her budget. Assumptions may be changed. Her objectives and liabilities may change (or she may refine what is essential and what is not essential). If she continues to use the Actuarial Approach, she has the flexibility to make informed adjustments in her budgets. Each year, she will balance her assets and her liabilities. If she uses the recommended smoothing algorithm, assets and liabilities may not be perfectly matched, but she knows that the match will be close enough.
Mary knows that she has to crunch a few more numbers under the Actuarial Approach than she would have to under the 4% withdrawal rule or some other variation of this rule, but Mary feels much more comfortable with the control she has over her spending budget using this much more sophisticated (and not that much more complicated) approach. Mary also finds comfort in the fact that the approach she is using is consistent with basic actuarial principles and is not just some simple rule of thumb approach.
Once again we read in the popular press about the 4% Rule and the tinkering that will be necessary to make this rule (or some form of this rule) possibly work in retirement. In his May 13, 2015 article, 4 Reasons Why the 4% Rule Isn’t a Hard and Fast Rule, David Ning tells us that spending needs to be adjusted in retirement. His “hard and fast” advice for doing this is that “you will be tempted to spend more in bull markets” and “you should decrease spending in bear markets.” In her May 8 article in The New York Times, New Math for Retirees and the 4% Withdrawal Rule, Tara Siegel Bernard quotes several industry experts with various opinions about the 4% rule and different adjustments that might make the rule work. The experts in this area continue their search (using their Monte Carlo modeling) for a Holy Grail spending rule to replace the now-suspect 4% Rule. So, what is a poor retiree to do now without a clear, simple spending rule of thumb?
Sorry folks, but a retiree’s budget problem is basically an actuarial problem that requires an actuarial solution. The retiree (or the retiree’s advisor) needs to periodically match the retiree’s assets with her liabilities. What are the retiree’s assets? They include her accumulated savings and the present value of retirement income from other sources (such as annuities or pensions). What are the retiree’s liabilities? These include the present value of future annual spending budgets, the present value of the amount the retiree wishes to leave to heirs and the present value of other expenses such as long-term care. Any simple spending rule of thumb that doesn’t attempt to match these assets and liabilities (and most common approaches don’t) runs a significant risk of not meeting the retiree’s spending objectives.
Ok, we’ll does this actuarial solution come in the form of a simple rule of thumb like the 4% Rule? No. It doesn’t. And while periodically matching assets and liabilities requires some number crunching, the Actuarial Approach and spreadsheets set forth in this website do most of the work for you. The process is relatively straightforward and doesn’t require you to be an actuary.
As we said in our post of August 2, 2014, Are You “Most People”?, the Actuarial Approach is not for everyone. It for someone who wants more than a questionable simple rule of thumb who is willing to do a little number crunching for the purpose of taking the guess-work out of developing a reasonable spending budget.
The June, 2014 explanation of how to use the Actuarial Approach has been revised to incorporate necessary adjustments to the general process for those retirees who want to “front-load” their spending budgets instead of developing a spending budget that is expected to remain constant in real dollar terms from year to year. Here is a link to the revised explanation.
The Transamerica Center for Retirement Study has released its annual survey of workers of all ages on the subject of retirement, entitled Retirement Throughout the Ages: Expectations and Preparations of American Workers. This survey provides lots of interesting information, and the Center uses this information to make recommendations to workers, employers and policymakers (pages 18, 19 and 20).
Consistent with results from prior years, "outliving my savings and investments" was cited by workers of all ages as their most frequently cited fear (page 31). Of course, addressing this fear by developing a reasonable spending budget in retirement is what this website is all about.
Of particular interest to me, in light of my previous post about the need to strengthen Social Security financing, were the results shown on page 36--that 47% of surveyed individuals in their 60s reported that Social Security will be their primary source of income in retirement and the results shown on page 32 that many individuals are concerned about the future of Social Security. But, despite these fairly disturbing results, the Center's Recommendations to Policymakers fail to mention any action to fix Social Security at all. As I said in my previous post, I believe the first step toward increasing workers' retirement outlook in the future should be to make sure that Social Security, the foundation of retirement security for most Americans, is solid.
In my post of March 1, 2015, I briefly discussed Social Security’s financial problem. In this post, I will once again mount my steed and tilt at the Social Security financing windmills by advocating adoption of a more actuarial approach to solving the problem. Readers who desire more background on the problem, the confusion resulting from the different approaches used to measure the size of the problem, how the problem came about and how Canada solved a similar problem can read my article in the May/June issue of Contingencies Magazine, the magazine for the actuarial profession.
Briefly, the 1983 Amendments to Social Security solved the financial problem that existed at that time, which was measured using the 75-year Actuarial Balance. This measurement is still around today and is widely quoted in the press as representing the size of the problem that needs to be solved today, but it was defective as a measure of the size of the problem in 1983, and it remains defective today. The 75-year Actuarial Balance calculation fails to reflect the future deficits expected after the end of the 75-year projection period. For this reason, the Social Security actuaries have proposed a stronger measure they refer to as “Sustainable Solvency” which would also require, at the time of a measurement, that trust fund ratios at the end of the 75 year projection period be expected to remain stable or on an upward trend. Unlike the 75-year Actuarial Balance calculation, the stronger requirement for Sustainable Solvency is not well quantified in the annual Trustee’s Report and therefore, it tends to get ignored when discussing reform options.
As noted in the article, Canada faced a similar financing problem with The Canada Pension Plan and implemented sweeping reforms in 1997. These reforms included self-sustaining provisions (automatic adjustments) to safeguard desired levels of funding, which resulted in Sustainable Solvency not only at the time of adoption of the changes, but also provided a mechanism for maintaining Sustainable Solvency in the future. I call this even stronger requirement, “Self-Sustaining Sustainable Solvency.” Actuaries, who work with the concept of automatic adjustments every day, may simply call this approach “actuarial financing.” I believe that the approach adopted in Canada provides a good blue-print for similar action in the U.S.
From time to time, we hear someone call for a national conversation on retirement in light of the retirement “crisis” in this country. Without a doubt, the first step in addressing this issue has to be making sure that Social Security, the foundation of retirement security for most Americans, is solid. I believe adoption of actuarial financing for Social Security is important for keeping that foundation strong for the future.
I want to thank Jean-Claude Menard, Chief Actuary for The Canada Pension Plan, for his thoughtful comments on an initial draft of the article and for his patience with me in explaining the process used in Canada.
Regarding the Don Quixote reference above, this is not the first time that I have advocated consideration of a more actuarial approach for Social Security financing (anticipating a tax rate expected to remain level indefinitely). Back in 1982 when the National Commission on Social Security Reform was working on what would become the 1983 Amendments, I wrote a paper that was subsequently published in the 1983 Transactions of Society of Actuaries entitled, “A Better Financial Approach for Social Security.” While this paper may be available from the SoA library, it is very difficult to find (and probably worth a lot of money). If you are interested in reading my thoughts on Social Security financing from around that time, a staff member of the Conference of Consulting Actuaries was able to provide me with a pdf version of the transcript of my paper and presentation, “Social Security--There Will Be No Long-Term Solvency With Pay-As-You-Financing” from a 1984 meeting of what was then the Conference of Actuaries in Public Practice.
