Social Security benefits are of grave importance to the millions of people seeking retirement each year. I came across a great article recently, and it got me thinking about some of the projections that would be useful for those retirees. There were a few points that were not addressed in the article that I think are critical for retirees to understand.
To make an informed decision about when to take social security, an aspiring retiree needs to answer a few questions:
- How safe do you feel social security is relative to investable assets?
- How long do you expect to live?
- How important is it to leave some wealth behind?
- How much risk do you feel comfortable accepting?
I won't try to make an empirical argument about the dangers of social security. I can tell every reader that in my preparations, I am very honestly calculating my own benefits of social security at $0 per year. I have quite a long ways to go until retirement, and while I think the program will still exist, I believe it will be thoroughly gutted before my generation is eligible to draw on the benefits for which they have paid. Therefore, on a qualitative level, I am heavily in favor of drawing out all funds as quickly as possible.
Regarding the other three questions:
I can't tell any reader exactly how long they should expect to live, but I can lay out the calculations in a way that allows them to pick the age and see the results. I can't tell a retiree how much they should seek to leave to their heirs. It wouldn't be my place, even if I knew their scenario. I can't decide for them how much risk to accept, but I can lay out the options.
I strive to answer all 3 of those questions using the charts below. It should be noted that the original content, represented in the black grid lines, was obtained from the article I referenced in the opening paragraph. It was my goal to expand on that information. My calculations will make the following assumptions:
- No person that is about to turn 62 is planning to live to an age greater than 120.
- The growth rate in social security will be roughly equal to 2.47%, the arithmetic mean of the increases from 2000 through 2014. If you're interested, the geometric mean was 2.46%. It does not make a significant impact on our calculations.
- No tax implications are considered. Any investor making this decision should seek qualified assistance from a tax planning specialist.
I include the chart here in multiple images, and then I'll explain the new columns.
The first new column is "Additional Income". This represents the difference between what you would collect in any given year as a result of choosing to start social security at the age of 62 rather than at the age 70.
The second column, IRR, represents the required internal rate of return that would need to be achieved throughout retirement for the two options to provide the same rate of income if the retiree died at the end of the year. Yes, it's a little morbid that way, but it gives readers access to the data behind the calculations.
I'm not comfortable with strictly using dividend yield stocks as a way to establish yield. I find the one dimensional calculations can ignore the quality behind the dividend. I subscribe to the idea that investors are capable of liquidating part of their holdings to create their own income. Because some investors prefer using dividend yield, I have provided the IRR for them.
Using the #% columns, an investor can find exactly how much more cumulative wealth they would have at any given age. I didn't present any numbers below 4%, because I feel returns below 4% are unrealistic long term projections. My evidence for that claim is looking at the yields on McDonald's corporate bonds.
The long term debt for McDonald has a YTM around 4%. I believe that a diversified portfolio of long term corporate bonds from high quality companies has less risk than a long term reliance on social security. Therefore, this is the lowest rate that I believe an investor can reasonably use for compounding. While it is possible that the interest rate environment may change and bonds may decrease in value, if they are being held to maturity the interest and yield have been locked in on the date of purchase. It is my assumption that the investor choosing to purchase a 30 year bond will not trade it after one year. That is the behavior of a trader, not a retiring investor.
Using the column for 4%, we can see the investor that dies prior to his 91st year of life will have been better served by taking social security at 62 and buying into a diversified portfolio of bonds.
However, if an investor is willing to take on slightly more risk to enhance their returns, the decision to take social security at 62 becomes better. By the time we get to a 7% yield, an investor could be living to 120 and still be better off having taken the payout at 62.
So, what does it take to get a 7% yield?
Using the CAPM model, we know that we can combine the risk free asset with the optimal risky portfolio to establish the lowest variance for any given level of returns. However, this is not an article about CAPM, so I'm going to cut to the chase. I'm going to construct a hypothetical portfolio through a simple process. Rather than use treasury bills as the risk free rate, I'll be using 4%, the rate on MCD's corporate debt. As mentioned earlier, I believe a diversified portfolio of high grade corporate debt is less risky than social security, so using a risk free asset as the base would create a biased comparison that understated the risk of social security.
I would suggest the following portfolio as a starting point:
This hypothetical portfolio, with each of above indexes/mutual funds weighted at 10% and rebalanced annually, from January 1st, 2000 up until April of this year has produced a geometric mean annual return of 9.208%. The hypothetical portfolio is, in my opinion, fairly diversified. I'm not going so far as to say it is the "optimal risky portfolio", but I will say that it is capable of dramatically outperforming the portfolio of the average investor on a risk adjusted basis when the risk is calculated as the standard deviation of annual returns. For the portfolio suggested above, the annual standard deviation (historically, from 2000) is 15%. The worst year returns would have been a loss of 28% in 2008. The second worst year the loss is 0.2% in 2002. Every other year the returns were positive. The standard deviation is also increased by two years, 2003 and 2009, in which the gains were greater than 30% in each year. You may also notice that in both cases the best years are immediately following the worst years. These funds were not selected carefully on the basis of great returns. They were selected for exposure to different factors. I built the hypothetical portfolio to ensure it had exposure to precious minerals, to large cap stocks, to small cap stocks, to bonds, to real estate, and to foreign investments.
Because there is no great investing skill required to buy those funds and set annual rebalancing, I feel that this hypothetical portfolio is a reasonable one for the analysis. The values being used throughout the model are historical returns, but that is part of why I have selected mutual funds or index funds for the analysis. It reduces the bias of including a single very hot company.
If the investor seeks 5% annual returns, they would need 19.2% in the Portfolio, with the rest in debts yielding 4%.
If the investor seeks 6% annual returns, they would need 38.4% in the portfolio.
If the investor seeks 7% annual returns, they would need 57.6% in the portfolio.
If the investor seeks 8% annual returns, they would need 76.8% in the portfolio.
Remember, at 7% the investor is financially ahead through their 121st birthday.
Waiting until 70 to take social security, assuming no very clear and specific tax advantages, is poor financial planning. If the retiree has the confidence to invest their money in the manner suggested, and to leave it there with annual rebalancing, I believe they will be far better off starting the payouts at 62. If the 6% target seems reasonable, and I believe that it is a very conservative allocation, then the retiree that dies before 103 has beat the system. If something terrible happens to him or her, that portfolio will be a great gift to the next generation.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: All information contained in this article represents the opinion of the author. All investors should do their own due diligence prior to investing. Nothing included in this article should be considered a recommendation to you to buy, sell, or otherwise take any action in regard to any investment opportunity. I do personally hold 4 of the mentioned funds in my retirement accounts. Seeking Alpha's system does not recognize those funds as valid symbols. I am holding MRESX, BUFBX, USAGX, and USSPX.