Wednesday, November 23, 2016

Why is the American Academy of Actuaries Painting Such a Rosy Picture of Social Security’s Long-Term Financing Problems?

Added December 8, 2016:  In response to this post, the AAA has added several caveats to its Social Security Game in an effort to avoid providing potentially misleading information about the System's long-term financial situation.  The Academy's Game with the new caveats can be found here.

From time to time, we deviate (slightly) from our primary mission to offer our thoughts on Social Security’s (the System’s) financial condition.  We do this because:
  • the System appears to have long-term financial problems, 
  • the System’s future benefit and tax provisions can be changed at any time by Congress, and 
  • changes in System benefits and taxes can affect almost everyone’s financial situation.

Thus, the financial condition of the System and the resulting uncertainty about its ability to pay scheduled benefits in the future should be an important consideration for individuals in the U.S. when developing their current spending/savings budgets.

How Big Is Social Security’s Problem?

There is considerable confusion regarding the size of the System’s long-term financing problem.  In the American Academy Actuaries’ (AAA) “Social Security Game,” the AAA claims the System’s problem can be “fixed” with approximately a 23% increase in the current System tax rate.  This is one option of many in the menu developed for the Game.

Using CBO and Trustees Assumptions to Estimate the Size of the Problem

On the other hand, as discussed in our May 17, 2016 post, Steve Goss, the Chief Actuary of Social Security has said, “Remedying OASDI’s [Social Security’s] fiscal shortfall for 2034 and beyond will require a roughly 25 percent reduction in the scheduled cost of the program, a 33 percent increase in scheduled tax revenue or a combination of these changes.”  Thus, based on the Trustees’ assumptions about the future, the Social Security’s Chief Actuary believes the problem is much larger than as indicated by the AAA.  In his recent testimony before the House Subcommittee on Social Security, Keith Hall, the Director of the Congressional Budget Office (CBO) noted that, based on CBO’s assumptions, their estimate of the System’s long-term problem was even greater than the Social Security Trustees’ estimate.

The graph below, shown in Figure 2 of Mr. Hall’s testimony, shows projected Social Security Tax Revenues and Outlays for the period 2000-2090 under both the CBO’s and Trustees’ assumptions.  This graph does a very good job of quantifying the projected shortfall between System revenues and scheduled benefits from the period 2030 to 2090 under the two sets of assumptions.  Under either set of assumptions, the shortfall is projected to be relatively constant, when measured as a percentage of the country’s projected Gross National Product for this period.   One can fairly easily see from this graph that the projected shortfall in revenues is relatively close to the 33% figure quoted by Mr. Goss under the Trustees assumptions, and something in the neighborhood of 45% under the CBO assumptions.  These figures can be confirmed by comparing projected 2090 outlays with projected 2090 tax revenues in the first section of Table 2 of Mr. Hall’s testimony.1 Under either set of assumptions, we are talking about significantly higher tax revenue shortfalls than the 23% figure claimed to “fix” the System in the AAA’s Social Security Game.  If you prefer to think in terms of necessary benefit reductions rather than required tax increases, the percentages are about 25% under the Trustees’ assumptions and about 30% under the CBO assumptions.2

AAA’s Fix 

So why has the AAA low-balled the size of the System’s long-term problem, when even the System’s Chief Actuary (using the Trustees’ assumptions) has indicated that we are looking either at much higher potential tax increases or benefit reductions?  Unfortunately, it is not clear to me why a profession that prides itself in “substituting facts for appearances” and “demonstrations for impressions” would want to provide this potentially misleading information.   In fact, Precept 8 of the profession’s own Code of Conduct expressly requires that an individual actuary “who performs Actuarial Services shall take reasonable steps to ensure that such services are not used to mislead other parties.”  It doesn’t appear to me that the AAA has taken such reasonable steps, but technically the Code of Conduct doesn’t apply to the organization representing the profession, only its individual members.

Planning Implications of Future System Reform

It is important to note that the 25% decrease in scheduled benefits (or approximately 30% under CBO assumptions) will automatically take place when the System’s Trust Fund runs out of assets if Congress fails to act prior to the Trust Fund Exhaustion date.  This is effectively an across-the-board decrease in benefits payable to beneficiaries at that time.  If Congress acts prior to the Trust Fund Exhaustion Date (by increasing tax revenue, decreasing benefits or some combination of the two), it is likely that some individuals will be less affected and some will be more affected than they would be under the default across the board benefit reduction scenario.

So, what does this all mean to retirees and pre-retirees in the U.S. who are counting on certain levels of future Social Security benefits?   That is the $64,000 question.  Will you be one of those individuals whose benefits are mostly unaffected or will you be one of those individuals whose future benefits or taxes will be significantly affected?  To paraphrase Harry Callahan in the movie “Dirty Harry” (by deleting the ending pejorative), “you’ve gotta ask yourself one question.  Do I feel lucky?  Well, do ya…?”

History has shown us that, when making changes to the System, Congress has been more inclined to reduce benefits and increase taxes mostly for those who are not close to retirement age.  Thus, it is unlikely that Congress will allow the default across the board benefit reduction scenario to take place.  On the other hand, it is also unlikely that Congress is going to place the entire burden of shoring up the System on the shoulders of our younger workers.  It appears likely that those with relatively higher incomes (young and old) will be asked to bear a significant portion of the increased cost in this next round of System reform.

The System is currently funded primarily with payroll taxes.  It is possible that the next round of System reform may involve other sources of revenue.  In any event, your current financial planning should consider the possibility that the scheduled (or actual) Social Security benefit you input in our Actuarial Budget Calculator worksheet may be reduced in some manner, your future taxes increased or some combination of the two.  Unfortunately, the changes necessary to truly “fix” the System may be larger than you thought.


1. Projected Percentage Shortfall in Revenues using Projections for 2090:
(6.34 – 4.29) / 4.29 = (20.08 – 13.59) / 13.59 = 48% using CBO assumptions
(6.14 – 4.63) / 4.63 = (17.68 – 13.33) / 13.33 = 33% using Trustees assumptions

2. Necessary Percentage Benefit Reductions using Projections for 2090:
(6.34 – 4.29) / 6.34 = (20.08 – 13.59) / 20.08 = 32% using CBO assumptions
(6.14 – 4.63) / 6.14 = (17.68 – 13.33) / 17.68 = 25% using Trustees assumptions

Sunday, November 20, 2016

Using Multiple “Data Points” to Determine How Much You Should Spend

The primary purpose of this website is to help individuals (with possible assistance from their financial advisors) determine how much they can afford to spend each year.  Our website was initially established in 2010 to help retirees with this issue, but we have recently expanded the scope of our purpose to address this issue for pre-retirees as well.  We have developed several spreadsheets that utilize basic actuarial principles to help individuals develop reasonable spending budgets.  And while we believe the development of a reasonable spending budget is an important part of an individual’s spending decision process, it is but one “data point” of several  that may be considered.  This post will discuss other possible data points that may also be useful in your spending decision process.

Applying the ABC Data Point

The Actuarial Budget Calculator (ABC) contained in this website determines a spending budget for the current year by mathematically balancing an individual’s assets (current assets and the present value of future income from other sources) with her current and future spending liabilities.  Thus, significant increases or decreases in the individual’s current assets from one year to the next can result in some volatility in the actuarially calculated spending budget from year to year.  As we have said many times in this blog, we have no problem if a retiree chooses to smooth her spending budget from year to year or to smooth her actual spending.  In fact, we have suggested that retirees consider establishing a “rainy day fund” after one or two favorable investment years to be available in subsequent unfavorable years as one approach to mitigate such fluctuations.  Thus, last year’s spending level may be another “data point” to consider.

The ABC with Recommended Assumptions Data Point

We understand that some of the users of the ABC do not use the recommended assumptions to determine their spending budgets.   These users may feel that because they invest a significant portion of their assets in risky investments, our recommended investment return/discount rate is too conservative and does not represent their best estimate.  We also understand that some retirees and/or their financial advisors may use any number of non-actuarial approaches to determine spending budgets.  And this is fine, too.  We don’t insist that the ABC using recommended assumptions is the one and only true answer.  However, we do suggest to these individuals that they also run the ABC with the recommended assumptions as another data point, for comparison purposes.  The ABC with recommended assumptions produces an actuarially calculated spending budget under the assumption that assets will be invested in relatively low-risk investments (approximately interest rates imbedded in life annuity products).  A significant positive difference between the retiree’s spending budget and this actuarially calculated budget can provide a measure of how much extra risk the retiree is “capitalizing” through his or her investment strategy.  In any event, running the ABC with recommended assumptions provides another data point that tells the retiree how far off the “actuarially balanced” track she may have strayed with her current spending strategy.

ABC Run-Out Tabs Data Point

The ABC also provides run-out tabs that show future spending and assets if all assumptions are realized in the future and spending exactly follows the budget plan.  In situations where the retiree expects to receive income from deferred sources (such as from a future sale of an asset or from deferred annuity contracts) the run-outs may show assets declining precipitously prior to receipt of the deferred income if spending continues course.  In such situations, the run-out tab information can serve as another data point in the spending decision process.

ABC 5-Year Projection Tab Data Point

In the 5-year projection tab, the ABC also provides the capability to model future investment and spending experience that differs from assumptions.  The results of this tab can be useful for developing contingency plans in the event actual experience deviates significantly from assumed experience and can also provide another data point in the spending decision process.

As indicated in our post of October 31 of this year, the run-out tabs also indicate what next year’s assets will be if all assumptions are realized during the budget year and spending exactly follows the budget, so this number is also another data point in determining this year’s spending if it looks like assets at the end of the year will be significantly different from this number.

Historical Record Data Points

We encourage retirees to maintain a record of prior years’ spending budget calculations and prior years’ expenses.  This historical information can also serve as additional data points to help with future spending decisions.  The information can also be useful in selecting assumptions about the future, particularly about future assumed increases in various types of expenses.

Your Gut Instinct

As we said in our post of October 17 of last year, “Unlike many experts who think that most retirees aren’t smart enough or motivated enough to manage their own money, I believe that most retirees possess the necessary skills to successfully manage their finances in retirement, much like they successfully managed their finances when they were employed.  Of course, for those retirees who can afford one, a financial advisor can be very helpful in this process.  However, when push comes to shove, it is you, Mr. or Ms. Retiree, who are ultimately responsible for making the investment and spending decisions that affect your financial situation during your retirement.”  This brings us to our final data point – your gut instinct.  As we have said many times in this website, it is ok to spend less than your spending budget.  It may also be ok to spend more sometimes (but, please don’t use this as your excuse for running out and buying a big boat).  Once you have gathered sufficient information, you and your significant other, if you have one, need to make the final call on your spending.

Gathering all these data points to make spending decisions may seem like a lot of extra work.  For some retirees (those with just one or two sources of retirement income, for example), it may not be necessary or worthwhile to gather this additional information.  For those with more complicated situations (including those with multiple sources of retirement income), it may.  You must find the appropriate balance between the time you spend managing your retirement and just enjoying your retirement.  We are here to help you (or your financial advisor) find the best answer for your specific situation.

Monday, November 14, 2016

Deferring Commencement of Social Security Benefits is Ok, Deferring Retirement is Better—Part II

This post is a follow-up to our post of April 28, 2014, where we looked at the effect on a hypothetical 65-year old’s annual spending budget under the following scenarios:
  1. Retiring at age 65 and commencing Social Security immediately, 
  2. Retiring at age 65 and deferring commencement of Social Security until age 70, and 
  3. Continuing to work 5 more years, retiring at age 70 and commencing Social Security at 70.
We concluded that while Scenario #2 might increase one’s spending budget by something in the neighborhood of 5%-10% over Scenario #1, Scenario #3 might increase one’s spending budget in retirement by 40% or more.

Now, in a recent study entitled, “Is Uncle Sam Inducing the Elderly to Retire”, the authors use some mysterious (to me) methodologies to conclude that the financial benefits of continuing to work an additional five years is much lower than the 40% figure we previously developed.  The authors conclude, “We find that if all elderly now working were to continue to work for five more years, they would, on average, raise their sustainable living standards (annual discretionary spending per household member with an adjustment for economies in shared living) by roughly 5 to 8 percent depending on their age and position in the resource distribution.”

As an actuary, my first reaction is to look at some numbers to see what they support.  So let’s use the Actuarial Budget Calculator (ABC) to look at a 65-year old male making $50,000 per annum gross wages.  Let’s assume he has $200,000 in accumulated savings and his home equity will cover his future expected non-recurring expenses. 

If we go to the Social Security Quick Calculator, we see that if this hypothetical individual retires and begins commencement of his Social Security benefit immediately, he would receive approximately $1,275 per month based on the assumptions made for his prior earnings history by the calculator.  The calculator also indicates that if he has no future employment income but he defers commencement of his benefit until age 70, his age 70 benefit in today’s dollars would be approximately $1,806 per month, and if he continues to work until age 70, his age 70 Social Security benefit would be $1,932 in today’s dollars.

Let’s assume that our hypothetical individual desires to have future spending budgets keep pace with inflation and uses the assumptions we recommend for the ABC.  He has no bequest motive.

For Scenario #1 (inputting an annual Social Security benefit of $15,300 – monthly benefit of $1,275 – starting immediately and $200,000 of accumulated savings), we get an annual spending budget of $24,011.

For Scenario #2 (annual Social Security of $23,928 – monthly benefit of $1,806 increased by 5 years of assumed inflation starting in 5 years), we get an annual spending budget of $25,842, an increase of 7.6% over Scenario #1.

For Scenario #3, we input an annual Social Security benefit of $25,597 (a monthly benefit of $1,932 increased by 5 years of inflation) starting in 5 years.  We then go to the new pre-retirement tab and assume that our hypothetical individual will receive annual 2% per annum pay increases, will save 10% of his pay each year and will not receive any additional pre-retirement income (such as a matching employer contribution).  Under this scenario, our hypothetical individual is expected to have a real dollar spending budget of $45,000 for 5 years and, at age 70, his real dollar spending budget is expected to decrease to $35,530 and remain at that level for the rest of his life.  Note, however, that this ultimate real spending budget is almost 48% higher than the Scenario #1 spending budget. 

Yes, he will have FICA taxes, income taxes, work-related expenses and savings that will need to be paid while he continues to work.  However, he had these expenses in prior years, so these are not new for him if he continues to work.   And, yes, he will not be receiving Social Security benefits while he works (of course he could if he wanted starting at his Social Security Normal Retirement Age of 66, but he decides to defer).

The authors are undoubtedly correct that there is some confusion in the general population regarding how the Social Security Earnings Test works.  For most readers of this blog, however, the concept is not that difficult.  Per “How Work Affects Your Benefits” prepared by the Social Security Administration, “You can get Social Security retirement or survivors benefits and work at the same time. But, if you’re younger than full retirement age, and earn more than certain amounts [generally $15,720 for 2016], your benefits will be reduced.  The amount that your benefits are reduced, however, isn’t truly lost. Your benefit will be increased at your full retirement age to account for benefits withheld due to earlier earnings.”

In their analysis, the authors assume that the Earnings Test is a “pure tax on benefits”, i.e., they ignore the increase in future benefits that results.  We respectfully disagree with the reasonableness of this assumption and, as a result, find the author’s conclusion misleading.

Bottom line:  I’m not buying the author’s argument that Uncle Sam is inducing the elderly to retire through operation of its tax and subsidy policies.  Of course, results will vary from individual to individual.   For most people, however, there is still plenty to be gained financially by continuing to work.  But, don’t just take our word for it.  Use our Actuarial Budget Calculator spreadsheet to crunch your own numbers.

Monday, November 7, 2016

Pension Actuaries Discuss Best Ways to Employ Assets to Mitigate Risks in Retirement

This post recommends two recent articles written by pension actuaries:  Mark Shemtob and Steve Vernon.

I volunteer with Mark Shemtob on the American Academy of Actuaries’ Lifetime Income Task Force.  The original mission of this task force was to “address the risks and related issues of inadequate guaranteed lifetime income among retirees.”  Mark is a consulting pension actuary like I was before I retired.  He is also a Certified Financial Planner and a Retirement Management Analyst.  His recent article, “The Retiree Nest Egg—Navigating the Risks” appears in the November/December 2016 issue of Contingencies Magazine, published by the American Academy of Actuaries.

Mark’s common sense advice to baby boomers regarding retirement planning can be summarized as follows:

  1. Continue to work (if you can) until you are satisfied you are financially ready to retire 
  2. Consider deferring commencement of your Social Security benefit until age 70 or purchasing a longevity annuity 
  3. If the sum of your Social Security and pension benefits doesn’t fully cover your fixed living expenses, consider purchasing a life annuity to cover the shortfall 
  4. Have a plan to cover future health-care costs, long-term care expenses and unexpected expenses 
  5. If your retirement spending strategy involves withdrawals from invested assets, make sure to monitor investment fee levels and selectively limit investment risk (perhaps by using a “bucketing” investment strategy that is coordinated with income to be received from other sources).
I worked with Steve Vernon for many years, and readers of this blog will recognize his name from the frequent references to his articles.  Steve was also a consulting pension actuary.  In his recent article, “6 retirement strategies from a local pro,” Steve discloses his own personal retirement strategy.  Not surprisingly, many of his 6 strategies are similar to those recommended by Mark.  Steve includes a couple of strategies that are not strictly financial.

I found the recommendations in Mark’s and Steve’s articles to be excellent and, for the most part, consistent with the opinions and recommendations we make in this website.   We may have small differences of opinion (like the best way to determine spending from investments, for example), but our thinking on retirement planning is not miles apart.   And maybe that is because we all think like pension actuaries.

Friday, November 4, 2016

You Want Software that Models the Effect on Your Retirement Spending Budget of Assuming Different Rates of Future Increases for Multiple Expense Categories? We’ve Got It!

I like to read the financial planning strategy posts from Michael Kitces.  I especially enjoy his weekly “Weekend Reading for Financial Planners.”  Michael is a good writer and is very prolific.  I don’t always agree with everything he says, but I give him big-time kudos for the expertise and energy he brings to financial planning discussions.

In his post of November 2, Michael summarizes much of the latest research on spending patterns in retirement.  I won’t summarize the research here again as you can simply read Michael’s post, and we have previously discussed much of this research in our posts of March 31 and August 20 of this year, entitled “Planning for Constant Real-Dollar Spending in Retirement – Is It Setting the Bar Too High (Parts I & II).”

At the end of his post, Michael says, “In practice, doing this kind of projected retirement spending may also be more difficult in today’s financial planning software, simply because most of the tools aren’t built to handle multiple different spending categories, each with their own inflation rates and age-banded spending cuts.”  Well, our Actuarial Budget Calculator (ABC) is not “most of the tools,” and it is built to handle 3 different spending categories:

  • essential health-related expenses 
  • essential non-health related expenses 
  • and non-essential expenses,
each with its own assumed future increase rates.  You will find this useful feature in the Budget by Expense-Type tab of the ABC.  And it wouldn’t be all that difficult to modify the results of this tab to look at more than 3 spending categories, if desired.

After you have used the Budget by Expense-Type tab to develop a current spending budget utilizing different assumptions for future increases in the 3 expense categories, you can go back to the Input tab of the ABC spreadsheet to see what single rate “desired increase in future budget amounts” produces an equivalent current spending budget (if you are curious).  For example, in the Budget by Expense-Type tab you might assume future increases equal to assumed inflation for essential non-health related expenses, inflation plus 2% for essential health-related expenses and 0% increases for non-essential expenses.  Depending on the relative mix of these expected expenses, the resulting current spending budget may be equivalent to that produced in the Input tab by assuming inflation minus 0.5% increases (or some other value) in your total recurring future spending budgets.

Happy Budgeting!