Anxiety Check

How are your retirement anxiety levels at the moment? I must admit my levels were high earlier this week. As described in my previous post, I recently realized that I’m (probably) not going to make it to 21-years of military service. Thus, I’m (probably) going to retire after my pension vests at 20-years. This means that I won’t secure the transferability of my GI Bill to my kids prior to retiring. As a result, I’m leaving a lot of potential money on the table.

I spiked my anxiety levels even further this week by breaking the above news to Mrs. Grumpus. She doesn’t routinely read my blog. She’s too busy with the kids and running the household to find the time to read my 3500+ word tomes (i.e. articles). As a result, if I come to some major insight about my life while writing, I can’t rely on her to read about it. This means I actually have to talk to her. Unfortunately, I’m a much better writer than a talker, so I usually make a mess of the conversation.

Conversation Results

This time was no different. In true Grumpus Maximus fashion, I dropped the retirement/GI Bill information bomb on her at the most inconvenient time possible (over breakfast with the kids on a school day). The immediate results were exactly what I’ve come to expect when I engage with such impeccable timing. I didn’t even have the numbers to back up my big idea. I was shooting straight from the hip based on everything I wrote about in my last post.

Apparently, despite all my self-reflection since starting this blog, and the specific realizations from a previous post on money and communication issues; I haven’t actually learned all that much. That said, I’ve now walked her back from the edge of the cliff, or more likely, a crime of passion. I’ve also managed to quell my retirement anxiety in the process. How I’d do it? Stay tuned and I’ll tell you.

Anxiety Fueled Vapor Lock

Before I walk you through the steps of my tried and true method for quelling retirement anxiety, let me just paint a picture for you. Based on the realization in my last post about retiring at 20 vice 21 years, my anxiety recently roared up like a monster. Part of that was probably linked to my ongoing efforts to ramp down my anti-anxiety/anti-depression medication. However, it’s safe to say that another big cause of my anxiety was the doubt my new retirement intentions created about the viability of my previous retirement plan.

What’s a person like me to do in a situation like this? Hyperventilate? Curl up in a ball in the corner? Run around with my hair-on-fire? Hide underneath my covers? Drink heavily? Sit in a catatonic state crushed by the overwhelming enormity of the situation? In the past two weeks, I feel like I’ve battled back just about all of those internal reactions in some form or another. None of them are useful, but all are normal, as anyone with anxiety can attest.

The Problem

Fortunately, I’m a prototypical male (from Mars). If there’s one thing I love to do, it’s solve a problem. As Mrs. Grumpus (who’s from Venus) can attest, it doesn’t matter if a solution is required; I’ll engineer one anyway. When done, I’ll clap my dusty hands together in slow motion, pat myself on the back, and declare it “Miller-time” like the beer commercials of my youth. The illusion of control that my planning provides is usually enough to put the anxiety geenie back in the bottle, at least momentarily. That’s where I found myself this week, ready to solve my (admittedly) self-created retirement problem; in hopes that it would help tame the anxiety monster.

Before I ran off and tried to solve the impossible though, I had to define my problem. So … what was my problem? My problem was:

I didn’t know what a 20-year pension, without a transferable GI Bill, and one less year of saving for a forever house, would do to my current retirement budget and retirement plan.

As I theorized in my previous article, a 20-year pension coupled with one less year of saving for a house might force my family and me to retire somewhere cheaper to live than California. This, in turn, might force us to pay for college when California would’ve otherwise picked up the tab under their disabled veteran program.

However, that was all supposition. Logical supposition, but supposition nonetheless. I didn’t know any of the impacts for certain, because I hadn’t actually tested my proposed plan. I hadn’t even researched what my new pension number would be. Nor had I updated my retirement budget with my latest 2018 spending data. Which meant I had a lot of work cut out for myself.

Steps to Solve the Problem

Fortunately, I’m fairly well organized in the retirement planning department, so it was easy to devise a three-step process to tackle my current dilemma. I decided to:

1. Review my written retirement plan to see what had changed
2. Update my proposed retirement budget in accordance with the changes to the plan
3. Test the new budget by running a number of scenarios with a high powered retirement calculator in order to determine my possibilities.

These steps should sound familiar. I’ve written about them numerous times before in my previous Planning articles.

Step 1: Cut a Hole in the Box?

As much as I wish my life was as exciting as Saturday Night Live music video featuring Justin Timberlake, it’s not. Thus, I won’t be dressing up in my best 1990s suit with a fake beard to review my current retirement plan. I can just look it up online because my retirement plan is a historical record on the interwebs. That’s right folks, everyone can participate in my planning review because I used my retirement plan as the basis for two of the first articles I ever published. The first article was about the need to plan, and the second was an example of my planning method. In the second article, I specifically used my plan as the reference.

For those who’ve never read my plan, or don’t remember my planning method, it’s important to note that I used the Goal; Reality; Obstacles; Opportunities or Options; and Way Ahead (GRO2W) method to build my plan. As a result, when I look at the Goal for my plan, it states:

“Retire from the military in San Diego at no more than XX years of active service in order to spend more time with family, live a fuller life, and focus on experiences. I intend to do this by providing my family of four the ability to live on my military retirement pay, the other benefits associated with military retirement, the proceeds from any other wealth I am able to accumulate while on active duty, and never return to what I consider ‘full-time work’.”

Reality Bites?

It’s also worth noting that the Reality statement for my plan also said this:

“Retirement at the XX year mark means retiring on approximately USD XXXXX gross (2016 rates) per year. Nothing to scoff at, but not an amount that would allow for carefree living when many of life’s major expenses (like college for our two children) remain ahead of us. This will be difficult in San Diego but potentially easier in other top military retiree locations like Colorado Springs or Tucson.

Adaptive and continuous planning will be required while still on active duty. Flexible execution will be required once retired. For instance, I may need to go back to full-time work for short periods if the economy slows down and investment values plummet. Alternatively, I may want to consider a lifestyle business which generates enough money to alleviate the need for large withdrawals from investments. Continuous tracking of money flows both in and out of accounts will be required decades into retirement with spending tripwires set at low thresholds.”

Translation

You may be wondering what the XXs were in both my Goal and Reality paragraphs. At the time, I hadn’t completely determined what was (mathematically) feasible in terms of years served, pension amount, and home prices for San Diego. I had two different terms of service floating around in my head: 21-years was the minimum that I needed to secure the GI Bill transfer, and 23-years was the maximum I was willing to serve if the numbers dictated it. In a later article on the use of a high powered retirement calculator to test my plan, I determined that a 21-year pension (with roughly $500K cash saved for a house) was mathematically feasible with a high probability of success. Thus, it became my primary option. I retested that plan in a March 2018 article, just to make sure.

Obviously, all that work was done prior to the 20-year retirement realization I made in my last article. Which means up until this week I’d never tested my retirement plan using a pension value based on less than 21-years of service. Nor had I modeled and tested it with less than $500K in cash saved for a house. Thus, I didn’t know what my new retirement date,  a smaller pension, and an inability to save $500K in cash for a house would do to my retirement budget. I didn’t know if it would survive my preferred retirement calculator’s simulations or show a miserably high chance of failure.

Simulation Determinations and an Updated Budget

Thus, I determined I needed to model two different scenarios.

1. A scenario where I take out a small mortgage to cover the difference between the cash we have saved and the cost of a house (which would increase our yearly retirement budget).
2. A scenario where I dip into our investments to make up the difference for the cash payment on a house (which would decrease our investment totals at the beginning of retirement).

Prior to any testing though, I needed to update both my projected income and my family’s spending numbers from the last 12 months in my retirement budget spreadsheet. None of that was particularly hard, just tedious. I started with the income side of the house and researched the value of a 20-year pension at my paygrade starting in late 2019 (when I’d retire). I did that using Department of Defense’s pension calculator. In doing so I came up with a workable total of $53,835. Not too shabby.

After that though, I subtracted out my projected yearly payment for the Survivors Pension Benefit (SPB), which is the Veteran Administration’s version of pension insurance for spouses and kids. You can read more about those types of insurance plans in the article I wrote for the Pension Series about Survivorship. Otherwise, simply note that my SPB payments come out pre-tax. This reduces the value of the pension prior to Federal and State taxes. Thus, upon running the calculations I determined the value for my pension after SPB would be $50,333.

About Social Security

Once done, I moved onto Social Security, which was a bit more complicated. First off, I realized my previous retirement budget egregiously miscalculated (too high) my projected Social Security income. I had based my projected Social Security payments on my annual Social Security statement, which provides an estimated monthly value for Social Security payments at your retirement age (67 for me). However, if you read the annual statement carefully you’ll notice this verbiage:

You have earned enough credits to qualify for benefits. At your current earnings rate, if you continue working until your full retirement age (67 years), your payment would be about …

I italicized the “At your current earnings rate” part for emphasis. That’s the kicker. The annual statement assumes you’ll continue to work and earn until your stated retirement age (67). Obviously, that’s not my plan. At some level, I’ve known for a while that there would be an impact from retiring early on my Social Security payments. However, I only realized how much when I found this calculator on the Social Security Administration’s (SSA) website and calculated it.

How it Works

The bottom line is that the SSA averages your 35-highest grossing work years into their formula to determine your monthly payment. If you don’t have 35 work years, they simply plug a zero in for each year you’re short. As any high school math student will tell you, zero as a value in the mathematical formula used to average numbers significantly reduces the final value. In my case, I’m 8 years short of 35 years of taxable Social Security income. That’s 8 zeros, which significantly drags down my final average. Bummer, but at least I have a much more accurate value prior to making any huge decision.

At the end of the day, I took my new monthly value and calculated it out for a year to get $14,580. I then halved that total to determine the value for my wife’s claim on my Social Security ($7,290). After that I took my full value, my wife’s half, and totaled it up and got $21,870. I added that figure to my retirement budget spreadsheet and moved on to expenses.

Quicken Makes Compiling Spending Categories Easy

“Track your money” is one of my personal finance mantras. I’ve written several articles (check my Resources Page) about the various software evolutions I’ve gone through. Currently, I’m back to using Quicken. After my revelations about Mint last year, and the trouble I’ve had with Personal Capital, Quicken (believe it or not!) turned out to be the easiest platform to use. By that, I mean Quicken is the easiest program to use that spits out the information I need in the format I need. For this exercise, that format is Excel.

Thus, it only took a few hours of work before I had an updated list of all my major spending categories that will carry over into retirement. Not only that, but I had Quicken spit the values from those spending categories for the last 12 months into an Excel spreadsheet. I then totaled those categories to get the total discretionary spending I plan to use for this updated retirement budget ($69,345). Easy peasy lemon squeezy.

Next, I calculated the Cost of Living difference between Honolulu and San Diego. That’s not hard if you use the Numbeo website. When I did, I found out that San Diego is 25% cheaper to live in than Honolulu … if rent isn’t taken into account. That’s a big difference, and since I’m not worried about rent when we move back to San Diego, I reduced the total of my discretionary spending by 25%. In doing so I got a reduced discretionary spending amount of $52,009 for my retirement budget. On to the next category!

Non-Discretionary Spending: Taxes

Not all of my proposed retirement spending budget is subject to a 25% cost of living differential. In fact, some of my spending is wholly dependent on the amount of money I pull in through my pension and investments — that includes taxes. Other categories, like health insurance expenses, are linked to the benefits I get from a military retirement. Thus, my next step was to determine what those categories were, and what they would cost me.

For taxes, I needed an accurate estimation of what my pension and investments would cost me at the California State and the Federal level. For the Federal taxes, I used H&R Block’s tax calculator — which incorporates the new Trump Era tax cuts. I’m sure H&R Block data mined the shit out of my information, but who cares? What’s important is that the calculations showed that I won’t owe any Federal taxes from my pension and investments. In fact, I may get paid by the government. I double checked it on another site and got the same result. I believe the reason has something to do with my kids and tax credits. The bigger picture is that Federal taxes aren’t going to be a major impact on my retirement budget, even after the kids leave the house.

It’s pretty much the same story at the State level for California, even though it taxes pensions. I ran my situation through several State tax estimators for California and got a sub $500 tax bill. That story will change over time, as tax laws change, and the kids leave the house. However, it’s good enough for now, and I’ll compensate for the unknown when it comes to modeling taxes in my retirement calculator.

More Taxes

Next, I needed to estimate property taxes. Fortunately, due to California’s unique property tax laws, that’s easy. Essentially property taxes are never more than roughly 1.00% of the assessed value of your property. Thus, if I buy a $500K house, I pay roughly $5,000 a year in property tax. There are some variations on this at the county and city level, but not enough to make a difference for the purposes of this exercise. Certainly, based on our ownership of a condo in California since 2004, my property tax bill has reflected this reality. Thus, for my calculations, I used $5K for a $500K house as my baseline. In the scenarios below in which I tested a more expensive house, I increased my property tax accordingly.

The final tax determination I had to make was for my Social Security income. Based on my research, and due to my pension, 85% of my Social Security income will be taxed once I claim it. Of course, that’s so far into the future, it all might change. For purposes of this exercise, however, I planned to tax my Social Security payments at the normal tax rate.

Non-Discretionary Spending: Healthcare

I know this won’t be the case for everyone, but for U.S. military retirees, estimating dental and healthcare premiums is easy. Access to the subsidized military healthcare system is probably more valuable than the pension itself. To estimate my healthcare costs all I did was go to two websites and compare plans. Here are the links to the Tricare Dental and Medical websites for retirees. When I ran the numbers, I got $1721 for dental and $594 for medical annually.

Flexible Retirement Planner (FRP) to the Rescue

After all that research, I totaled my estimated retirement expenses. Doing so allowed me to input that value into my retirement calculator of choice and run my simulations. The total I arrived at (with reduced COL for San Diego) was $62,307. To be clear that includes all my estimated discretionary spending plus my fixed spending categories like taxes and health care.

As I’ve written about before, FRP is my high-powered retirement calculator of choice. As its name indicates, it’s flexible. Super flexible in fact. This means I can program all the spending and saving scenarios I outlined above, and then run tests on them to provide probabilities of success or failure.

The Drawbacks

As much as I like FRP though, it is but one calculator. There are other, even better, calculators out there. However, they all suffer the same fatal flaw: they can’t predict the future. All outcomes are probabilities at best. Those probabilities depend on the accuracy of the numbers input into them. As you’ve seen from my process above, there’s already some fudge in my numbers. Inevitably then, there will be fudge in the probabilistic outcomes that FRP shows me. Keep that in mind when you run your own retirement budget tests.

I’ve developed my own unique way of hedging against this fudge. I tend to use worst-case scenario numbers. Thus, in my retirement budget plan my family’s proposed spending is higher than I’m sure it will be; investment returns are lower than both the market and my personal historical averages; Social Security and Pension payments are taxed heavier than I just outlined; inflation is higher than currently predicted; so on and so forth.

The Scenarios and Results

With that in mind, once my research was complete, I went into FRP and entered all the totals I just outlined above. I’ve screenshot the two main pages in the program where all this data is input so you can see my entries (below).

Doing so allowed me to run different house purchase scenarios to see what each input did to my overall probability of success. My scenarios included the outright purchase of a house at $500K, $600K, and $700K by dipping into our investments. They also included buying $500K thru $700K houses using either a $100K or a $200K 30-year fixed loan at a 5.5% interest rate. I listed the results with their probabilities of success below. As a point of reference, I consider anything above 95% chance of success as a solid option:

Scenario Name	Probability Of Success
50.3K Pension w/o mort 500K House	99.3
50.3K Pension 500K House w/ 100K Mort 5.5 Int	98.1
50.3K Pension 500K House w/ 200K Mort 5.5 Int	96.6
50.3K Pension w/o mort 600K House (includes +1K prop tax)	95.8
50.3K Pen 600K House w/ 100K Mort 5.5 Int (includes +1K prop tax)	93.2
50.3K Pension 600K House w/ 200K Mort 5.5 Int (includes +1K prop tax)	91.1
50.3K Pension 700K House w/ 100K Mort 5.5 Int (includes +2K prop tax)	86.5
50.3K Pension w/o mort 700K House (includes +2K prop tax)	83
50.3K Pen 700K House w/ 200K Mort 5.5 Int (includes +2K prop tax)	77.8

The good news is that there are several scenarios in which my reduced pension scenario works with both an outright house purchase and a home loan. It should also come as no surprise that the most important factor in the probability for success appears to be the home price. The cheaper the home, the higher my probability of success. The next most important factor appears to be the size of the loan and its impact on the annual retirement budget. The more money I borrow, the larger my annual budget becomes. Thus, the less likely it becomes that my pension and investments alone could support us.

What Are My Takeaways?

Thanks for asking. I learned several important things when I saw the above results. The first was that both Mrs. Grumpus and I could calm down. We now know there’s a high probability that the sky won’t fall if I choose to retire earlier than 21-years and don’t secure my GI Bill’s transferability. We can still make a life for our family in San Diego, without the need to ever work another job again, if we so choose. The likelihood is small that neither of us will ever work again, but the above outcomes will not force us into jobs we don’t want. Also, by being able to afford a California retirement, I’ll still be able to tap into the one-of-a-kind California tuition assistance program that the state runs for the children of veterans. Happy Veteran’s Day indeed!

My second takeaway was that the retirement plan is still viable. Surprisingly so since I didn’t even factor in the effects of VA Disability. Maybe I made better estimations of cost and income this time around? Or maybe the stock market’s continued run-up in the last few years makes for a much stronger net worth and provides more options than when I last ran the scenarios. In any case, the results gave me the assurance I needed to stuff the anxiety geenie back in the bottle.

My final takeaway is that I can start the retirement process now. It usually takes about a year to execute military retirement effectively, and I’m well within that window. That said, the above results are probabilities of success based on an unknowable future. I’ll need to keep an eye on everything while going through the retirement steps. If the stock market crashes in the next 12-months, or tax laws change significantly; I’ll need to re-test the numbers and re-evaluate the options.

In the meantime, I can start the retirement process and not worry about leaving the GI bill on the table. Doing so won’t subject my family to live on food stamps, or living in a van down by the river. Thus, I’ve slain my anxiety monster … for now.