Using the Actuarial Approach to Determine Your Total Spending Budget in Retirement

Despite being named "How Much Can I Afford to Spend In Retirement?", most of the posts on this website have focused on how much of your accumulated savings should be withdrawn each year.  While we believe using the Actuarial Approach described in this website should be an integral part of the process of determining your annual spending budget in retirement, it generally does not provide the final answer.  Other sources of retirement income, including Social Security, annuity income and expected income from employment must also be considered.  This article describes the recommended steps to determine your total annual spending budget.  The post as of June 21 of this year and the linked article also provide examples of the process and how to use the "Excluding Social Security V 2.0" spreadsheet in this website.

Another Example of the Actuarial Approach for Determining Your Annual Spending Budget in Retirement

When I talk to people about the Actuarial Approach recommended in this website, I usually get one of two responses:  1)  It is too complicated or 2) your spreadsheet is too simple and doesn't handle every situation.  The more conflicting feedback I get like this, the more I tend to believe that, on average, the method is just about right. 

Last week, Kathleen Pender, a business reporter for the San Francisco Chronicle noted that my approach was more complicated than the 4% Rule. 
 
Readers of this website pretty much know how I feel about the 4% Rule.  First, it is not necessarily all that simple, and second, the little extra effort you may have to take to apply the Actuarial Approach should help you sleep better at night knowing you are on track with your spending budget. 

This week I also received feedback from another actuary who said that my simple spreadsheet (Excluding Social Security v 2.0) would be better if it handled more situations.  Specifically, he suggested that users should be able to input expected income to be received in the future that may not be paid for life as well as payments generally paid from pension plans that are not fully indexed with inflation.  I responded to this actuary that, since every future year will produce deviations from assumed experience (such as different investment return, deviations from the spending budget, etc.),  it is more important that the retiree follow the process we recommend in this website than it is to have the world's most accurate spreadsheet. 

But, for the retirees out there who have sources of retirement income that don't seem to be anticipated by the simple spreadsheet, here is an example that may help you develop your spending budget.

Rosemary Retiree retired on January 1, 2013 at age 65.  She had $250,000 in accumulated savings/investments, an annual pension of $12,000 per year indexed with the CPI minus 1% per year, a series of equal periodic payments of $15,000 per year payable shortly after the beginning of the year for the next 10 years, and a Social Security benefit of $18,000 per year. 

With respect to the 10-year certain payments of $15,000, Rosemary determines the present value of such payments at 5% interest to be $121,617.  To this amount she adds her accumulated savings of $250,000 and enters $371,617 as accumulated assets in the V 2.0 spreadsheet on this website together with the recommended assumptions and an amount desired to be left at death of $10,000.  With respect to the partially indexed pension annuity, she can either choose to treat this source of retirement income as a fully indexed pension similar to Social Security or she can choose to treat it as a fixed annuity payment.  She knows that either approach is not 100% accurate, but the difference will just be one more gain or loss that will be treated the same way as all the future gains and losses.  She chooses to treat the partially indexed annuity as a fully indexed annuity and does not enter an amount in the spreadsheet. 

So, her total spending budget for 2013 is $46,046.  This is comprised of $16,046 from accumulated savings (from the spreadsheet), $18,000 from Social Security and $12,000 from the pension. 

During 2013, the CPI increased by 1.3%.  Rosemary received one of her $15,000 periodic payments, $52,000 in investment income, $12,000 in pension payments and $18,000 in Social Security payments.  Let's assume that she only spent $40,000 in 2013, so her total assets at the end of 2013 (not counting the present value of her remaining 9 structured payments) equals $307,000.  To this amount, she adds the present value at 5% interest of her remaining 9 periodic payments for total accumulated savings of $418,948.  Her Social Security benefit has been increased by 1.3% to $18,234 and her pension payment has increased by 0.3% to $12,036.

At the end of 2013, Rosemary enters her new accumulated savings of $418,948 and a payment period of 29 years.  All the other input items in the spreadsheet are the same as entered in 2013.  The resulting spendable amount is $18,559 to which she adds her 2014 Social Security benefit of $18,234 and her 2014 pension benefit of $12,036 to get a total actuarial spending value of $48,829.

Using the algorithm recommended in this website to smooth gains and losses, Rosemary then checks to see whether last year's budget amount increased with 1.3% inflation ($46,645) falls within the 10% corridor around this year's actuarial value of $48,829.  Since it does, she uses $46,645 as her total spending budget for 2014.

Rosemary knows that since the actuarial value for 2014 exceeds her budget, she has some accumulated gains (primarily from favorable investment experience and under spending her 2013 budget) that she can use in later years to offset possible losses. 

Forget the 4% Rule--Use Our Approach Instead

David Ning has followed up his April 30, 2014 post questioning the relevance of the 4% Rule with another post indicating that the 4% Rule is an incredibly powerful tool but many investors will need to customize it to make it work.  As indicated in our response to his first post, the better solution is to use the approach we recommend in our website rather than fuss around with "customizing" the 4% Rule.

Accumulating Savings for Retirement of 8x or 10x Pay is Better than 4x or 6x

Recently two long-time associates of mine at Towers Watson, Gaobo Pang and Dr. Syl Schieber released American Workers' Retirement Income Security Prospects: A Critique of Recent Assessments arguing that "some recent assessments of what workers should save and when they should save it dramatically understate the adequacy of retirement savings for many households."  In this article and a summary article published by Towers Watson, they suggest that accumulated savings of four times or six times annual pre-retirement gross earnings will be sufficient to replace pre-retirement standards of living for the average worker.  They use their "life cycle structure of consumption" approach to develop lower replacement targets than developed by others and argue, "why should...workers reduce their working lifetime consumption by a total of two or three or more years of earnings in order to achieve a higher standard of living in retirement."  They express concern that, "if workers are led to believe that they are so far off a reasonable savings path that an adequate retirement income is beyond reach, it raises a question of whether some workers will not become so discouraged that they will give up on saving for the future."
 
My friends at Towers Watson make good arguments why a one-size savings rule of thumb will not fit all workers.  For example, in addition to different replacement rates provided by Social Security benefits, some workers will have other supplemental sources of retirement income, other retirement income needs, and others will retire earlier or later than age 65.  And while I share the authors' concern that there may be many workers who are discouraged about saving for retirement by big numbers, I don't believe the answer lies in giving workers a smaller target that will likely fall short of meeting their needs, and I worry that workers may reduce their current savings efforts in light of this research. 

As I indicated in my August 11, 2013, my friend and actuary, Steve Vernon, had a great, concise two-step process for retirement planning:

Step 1:  Plan to support the life you want using your best estimate of the future regarding the economy, capital markets, your life expectancy and so on.

Step 2:  Be prepared in the event that your forecasts are wrong

In other words (and the words I use throughout this website), step 2 requires you to assess and manage your risks in retirement.  And I believe that if you have no other sources of retirement income other than Social Security and you are a medium or above earner, you will be in a better position to manage these risks if you accumulate at least 8X or 10X your annual earnings at retirement rather than 4X or 6X.

The authors developed their target savings multiples by assuming hypothetical workers born in 1949 retire this year, commence their Social Security benefit this year and somehow purchase a very favorably priced indexed joint and survivor annuity using assumptions for mortality and interest used by Social Security actuaries.  Those excited about the authors' resulting lower savings targets should note that under current law, future age 65 Social Security replacement rates will be somewhat lower than the rates used by the authors for individuals born after 1954 (reaching about 7% lower for anyone born after 1959).  In addition, because of Social Security's financial difficulties, future benefit reductions of something like 25% may be required sometime in the 2030s when the OASDI trust fund is projected to be exhausted if tax rates are not increased. 

The two tables shown below illustrate total retirement income (Social Security plus withdrawals from savings) for the medium earner and high earners described in the authors' paper using the authors' data, except instead of purchasing a hypothetical annuity at retirement, the workers are assumed to follow the withdrawal strategy and recommended assumptions described in this website.  I leave it up to the reader to determine which level of savings under these circumstances best produces the reader's desired level of retirement income.  Remember that the income shown is before taxes.  

(click to enlarge)

 

(click to enlarge)

Why is it prudent for workers to set higher target multiples of savings than 4X to 6X earnings?

Here are a few reasons:

There are multiple risks in the many years of expected retirement that can work against the retiree:  poorer than expected investment income, longer than expected longevity, higher than expected inflation.  These risks can be managed to some degree with lifetime insurance products, but these products bring their own set of risks with them.   As I have said many times in this blog, it may make sense to manage these risks by combining life insurance products with a sound withdrawal strategy. 

As noted above, Social Security benefits commencing at age 65 (a large portion of the projected total income for the hypothetical medium and high earner shown in the tables above) are scheduled to decrease in the future from current replacement levels, and may need to be reduced even further to keep the program financially solvent.  Further, many people do not wait until age 65 to commence Social Security benefits.  Some workers find that they are no longer employed at that age.  Social Security statistics show that for whatever reason about half of all Social Security retirement benefits commence at age 62.  So actual Social Security benefits may not produce as high a replacement rate as shown in these tables. 

Some individuals may want to leave amounts to heirs upon their death.

Some individuals may want to be able to do more and live better in retirement than they did prior to retirement.

Bottom line:  Targeting savings of just 4X or even 6X pay at retirement may be cutting it too thinly for some individuals.  Better that you make your own decision by crunching your own numbers based on your situation and desires rather than relying on someone else's rule of thumb estimate.    

Is the 4 Percent Rule Still Relevant for Retirees? Uh, Hells No.

Thanks to David Ning for posing this question in his recent post.
 
Readers of this website will note that over the past four years about half of the posts have been raves against the 4 Percent Rule, so you will undoubtedly not be surprised to know that I agree with Mr. Ning's conclusion on this one.  However, it is one thing to criticize the 4% Rule and another thing to offer readers a solution (other than Mr. Ning's, "perhaps the safe number is now 3.5 or 2.8 percent.")

The solution is to use a better strategic withdrawal approach.  Fortunately, the actuarial approach recommended in this website is a better approach than the 4% Rule.  If you have Excel on your computer, it is very easy to use.  See our post of December 27 of last year for an example. 

Deferring Commencement of Social Security Benefit is OK, Deferring Retirement is Better

There have been a number of articles recently touting the benefits of deferring commencement of one's Social Security benefit.  For example, in my post of December 8 of last year, I quoted the author of one such article who said, "Rather than annuitizing retirement wealth, participants can get a much better deal by spending down retirement assets and deferring Social Security."  This post will examine a hypothetical situation in an attempt to quantify how much more retirement income a person could have by deferring commencement of his Social Security benefits.

Let's look at a 65 year old male with accumulated savings of $800,000 in 2014.  His Social Security Normal Retirement Age is 66.  He is eligible for a Social Security benefit of $2,000 per month (plus an inflation adjustment) if he retires at age 65 and commences his Social Security benefit at age 66, or $1,867 per month (.9333 X $2,000) commencing immediately if he retires at age 65 and commences at age 65.  Assuming annual inflation increases of 3% per annum, retirement at age 65 and deferral of his Social Security benefit until age 70, he would receive a benefit of about $3,060 per month (1.32 X $2,000 plus 5 years of assumed 3% per annum inflation increases).

The table below shows total spendable income available to the individual from accumulated savings, Social Security and, in the third scenario from a life annuity developed by using the simple spending spreadsheets found in this website and the assumptions recommended for use with the spreadsheets.  Readers are reminded that the spreadsheets provided in this website are designed to provide level real dollar total spendable retirement income from year to year over the expected payout period.

(click to enlarge)

Under Scenario 1, the individual retires at age 65 and commences Social Security immediately.  He uses the actuarial approach in this website (Excluding Social Security 2.0 spreadsheet) to determine his annual withdrawals from accumulated savings.

Under Scenario 2, the individual retires at age 65 but defers commencement of his Social Security benefit until age 70.  He uses the Social Security Bridge spreadsheet to determine the extra withdrawals from his accumulated savings that will, when combined with deferred Social Security benefit, provide level real dollar spendable income from year to year.  In this instance, the present value at 5% interest of those extra "bridge" payments is $152,480.

Under Scenario 3, the individual retires at age 65 and commences Social Security immediately.  He takes $152,480 of his $800,000 retirement nest egg and buys a life annuity.   I used a purchase rate of $14.03 per each dollar of annual income obtained from Incomesolutions.com to determine the immediate life annuity payable at age 65.

As the table shows, total retirement income is comparable for the first three scenarios with the deferral strategy (Scenario 2) being somewhat better than the immediate commencement and no annuity purchase strategy (Scenario 1) but only slightly better than using the same amount of money to purchase a life annuity (Scenario 3) at current annuity purchase rates.

But, if this person really wants to get a significant increase in retirement income, he needs to work another five years.  Scenario 4 assumes that he continues to work until age 70 and retires at that point.  His accumulated nest egg in 2014 is assumed to earn 5% per annum and he is assumed to contribute $10,000 each year and his employer is assumed to contribute $2,000 each year.  While his Social Security benefit should increase somewhat as a result of five additional years of earnings and tax payments, we have assumed it would remain the same.  Under these assumptions, his benefit at age 70 is at least 35% higher than the benefits payable at age 70 under the other strategies.

Take Away:  Deferring receipt of Social Security may be able to get you a little more retirement income (depending on assumptions employed and actual experience), but if you really want more retirement income, you need to defer both your retirement and Social Security benefit commencement date.  This strategy works for you in two ways--it should increase your accumulated retirement nest egg and it reduces the expected payout period. 

Back to How Much Savings is Needed

The primary focus of my posts is on how much accumulated savings can be spent each year in retirement.  But since that amount is generally linked with how much accumulated savings one has, readers will forward me articles aimed at pre-retirees looking for advice on how much they need to have saved at retirement. 

I have addressed this issue somewhat in this article and most recently in my post of May 31 of last year where I concluded that if you don't have sources of retirement income other than Social Security, you probably shouldn't be too worried about over-saving until you have accumulated at least 10 times your annual salary. 

This past week, I read "Have You Saved Enough to Retire?" and "Why not even $1 million may not be enough for retirement".

The first article references Fidelity's very optimistic "8 times" rule, which claims that if you are an "average worker", retire at age 67, die at age 92 and accumulate savings of 8 times your pay at retirement, you can enjoy total retirement income (Social Security and withdrawals from accumulated savings) of approximately 85% of your gross pre-retirement pay.  Based on the recommended assumptions in this website (5% investment return, 3% inflation and payments until age 95), I come up closer to a 70% replacement rate.  Of course, Fidelity assumes that you will earn an average nominal rate of return of 7.5% per annum with 2% inflation (a real rate of return of 5.5% per annum).  If I input Fidelity's assumptions in the "Excluding Social Security 2.0" spreadsheet on this website, I do get a replacement rate of 88%, so, at a minimum, I can confirm Fidelity's calculations, if not the reasonableness of their assumptions.

The second article argues that $1 million in accumulated savings at retirement may not be enough.  Whether it is enough will depend on many things, but I can provide some spending numbers associated with having $1 million in accumulated savings and you can make up your own mind.  If Social Security income is around $30,000 per year (about the maximum in 2014) and no other retirement income sources exist, then total annual gross retirement income (Social Security and withdrawals according to the approach outlined in this website) for an individual retiring at age 67 with $1 million of accumulated savings will be about $75,000 to $80,000 depending on how much of the $1 million is used to purchase an immediate annuity (based on current annuity purchase rates) and/or whether the retiree delays receipt of Social Security benefits.  This level of retirement income will probably be enough for many individuals who retire in the near future.  Keep in mind, however, that it assumes retirement occurs at age 67 with $30,000 in annual Social Security income.   

As an actuary, I tend to be fairly conservative.  I wish I could tell pre-retirees that you don't need to accumulate all that much to afford to retire, but in all good consciousness, I just can't.  If you want, you can certainly plan your retirement by assuming that you will earn a 5.5% real rate of return on your retirement assets, there will be no reductions in Social Security benefits, or you will retire later than age 67, etc.  Or, you can follow the more conservative path and save more now for your retirement. 

Delaying Commencement of Social Security--Internal Annual Real Rate of Return Calculations

Dr. Wade Pfau has produced another fine article of interest to individuals who are in a position to delay commencement of their Social Security benefits.   As discussed in several of our previous posts (most recently December 8, 2013), there are at least two ways to delay commencement of Social Security benefits.  One can keep working, or one can retire prior to age 70 and use their accumulated savings to "bridge" the period of delay by paying themselves what they would have received had benefits commenced earlier than age 70.  We have a spreadsheet to enable individuals to see how such a bridging approach could affect their total retirement spending budget.

In Dr. Pfau's article, he has determined internal annual real rates of return assuming death occurs at various ages for individuals with a Social Security Normal Retirement Age of 66 assuming commencement at age 62 vs. delaying commencement until age 70 and subsequent death at a later age.  He also assumes no changes in current Social Security law.  So, for example, he has determined a 3.2% internal annual real rate of return for an individual who dies the day before his 85th birthday (at age 84) resulting from his decision to delay commencement until age 70.  So, if inflation is 3% per annum, this example individual who elects to delay commencement of Social Security until age 70 and use his accumulated savings to bridge the payments he would have received will be financially better off provided he does not earn more than approximately 6.2% per annum (3.2% real) on his accumulated savings and the Social Security benefits he receives.

As shown in the article, internal annual real rates of return are lower than 3.2% if the individual dies earlier than age 84 and higher for deaths occurring after 84.  So, if you definitely know that you are going to live past your mid-80s, the delay strategy appears to be a good one unless you anticipate earning relatively high real rates of returns on your investments or changes in the Social Security law. 

Several Wade Pfau Posts

A first for me this week--An individual actually asked me why I hadn't posted on this site in over two weeks.  Of course it was just Bob, a fellow Baritone in my singing group, but I will take his comment as encouragement (which is unusual for Bob as he is usually all over me for singing the wrong notes and/or the wrong dynamics).  

Readers of this blog will know that I frequently refer them to Wade Pfau's Retirement Researcher Blog for excellent discussions of retirement income issues.  This past week, Wade posted two more such discussions.

In his March 27 post,   he has listed and classified known retirement income strategies.  Readers with good eyes will see that the Actuarial Approach recommended in this website resides in the bottom box as a Variable Spending strategy residing under the general classification of Probability-Based Approaches.

In his March 24 post, Wade refers to two of his articles discussing risks in retirement and presents a Taxonomy (classification) of Retirement Risks.  I would add "not spending enough" to the lower right hand quadrant of his chart as this risk can be almost as problematic as spending too much.

As I have mentioned several times in this blog, it would not be unreasonable for a retiree to manage these risks in retirement by diversifying their sources of retirement income.  For many retirees, I believe the best approach involves combining Social Security, life insurance annuity (or long-term care) products or pensions, with a good systematic withdrawal strategy.   

Shopping Around for Withdrawal Strategies That Will Give You a Higher Withdrawal Rate?

As discussed in our post of March 1, the JP Morgan withdrawal strategy produces a higher withdrawal rate than the withdrawal strategy recommended in this site when our recommended assumptions are used.  Several readers commented that this was a big plus in the JP Morgan column, especially for individuals who want to front-load spendable income earlier in retirement or for others who have lifestyle spending goals that simply aren't achieved by using more conservative withdrawal strategies.  We caution retirees against selecting a withdrawal strategy simply because it produces a higher initial withdrawal rate.

The withdrawal approach set forth in this website produces a total retirement spending budget that is designed to remain constant in inflation-adjusted dollars if all input assumptions are correct and the spending budget is actually spent each year.  That part of our approach is just simple math.  Our approach also gives the retiree a recommended process for dealing with the inevitability that the input assumptions will not be correct or actual spending will differ from the budget amounts.  To make it a little easier for some to use our approach, we have recommended what we believe to be fairly conservative input assumptions:  currently 5% nominal investment return, 3% desired increases/inflation and an expected payment period until age 95 (or life expectancy if later).

If the withdrawal rate using our approach with recommended assumptions is lower than desired,  a retiree is free to change the recommended input assumptions to reach a higher withdrawal rate using any kind of rationale the retiree chooses (i.e., I can earn an annual investment return greater than 2% real, I won't live longer than my life expectancy, I can live with retirement income that decreases in inflation-adjusted dollars, etc.).  The retiree can also simply choose to spend more than the budget amount and live with the consequences.  Retirees should of course note that higher initial withdrawals will mean lower subsequent withdrawals, all things being equal, and there are no guarantees when you choose to fund some or all of your retirement through periodic withdrawals from accumulated savings, irrespective of the withdrawal strategy you use.