PRC Online Newbie here: I'm early retired with no income or spouse and only 2 accounts (IRA and Brokerage) trying to optimize when to take SS and if/when to do Roth conversions based tax bracket rules. Do I need to use one of the advanced Modes? Roth & SS seem interrelated so should I do one optimization before the other to get the "optimal" results. I think the optimization engine is looking to optimize across my entire life span (based on my input life expectancy) ??? so to see the impact on actual EOL age, do I need to vary my life expectancy in the setup section or just look at my account values at different times ?
Roth Conversions are very complicated, so it's great that you have an excellent tool to analyze them. Just know that there are some points where the math is subtle, so you have to be a bit careful. I'll try to lead you through some key things (though the post has gotten crazy long).
I would look at Roth Conversions first, not SS. If you benefit much from Conversions, then your optimum SS claim age is simple - wait until 70. That maximizes the low tax space for Roth Conversions (and ACA premium credits if you need those) while the SS benefit grows.
An excellent and free site for looking at claim age optimization without regard to Roth Conversions, ACA and other real-world tax issues is opensocialsecurity.com. For single males, it will likely recommend claiming at age 64 or so, depending on the current long TIPS rate of return (the site explains why it chooses that). You can then look at the difference in inflation adjusted value of claiming at the recommended age vs. waiting to claim at age 70 (actually, go to the graph at the bottom of the OpenSS output and click on any point it will tell you the value for that claim date) and compare that to the value you are getting from Roth Conversions. In our case, it's no contest, Roth Conversions rule, so waiting to 70 is by far the right move for us.
Opensocialsecurity.com and other retirement researchers do the SS claim date optimization differently than Pralana does. OpenSS notes that if you wait to claim SS, you are increasing your inflation adjusted guaranteed income. The investment that you can make that most closely resembles how SS acts in your portfolio is bonds, specifically long TIPS. Waiting to claim is compared to taking the benefit and buying long TIPS with that benefit as that is the most similar thing you could do. If you follow the logic of that comparison, then to keep your total portfolio risk constant while deferring SS, you should spend down your bonds while waiting to claim since the bigger eventual SS benefit replaces some of them.
Pralana's SS claim optimization module assumes you are spending down your average portfolio, not just the bonds, so it ends up overly favoring early claiming vs. what retirement researchers would recommend. The best I can in Pralana right now is make four changes in asset allocation, which is not bad.
A more crucial point about modeling Roth Conversions for many folks is to select the right modeling method. If you preferentially hold your fixed income in your tax deferred (which you should for tax efficiency), then you should use Advanced Portfolio Modeling and select Mode 2. Mode 2 will keep your portfolio asset allocation constant year to year in spite of different allocations in each account. Many folks actually hold their fixed income in tax deferred but fall into the trap of using Mode 1 or Simple Portfolio Modeling. That incorrect setup will have the program give incorrect results telling you to do huge conversions. What happens in that case is the program would sees the low returns in tax deferred (due to lots of bonds) and the high returns in taxable/Roth (due to stocks) and does giant Roth Conversions to get those higher returns - but of course all it would really be doing is switching bonds to stocks, which is a bad comparison.
Lastly, set an Effective Tax Rate and age of death that is consistent with your life situation and goals.
For instance, folks that intend to do QCDs and charitable bequests should probably not be bothering with Roth Conversions at all as the very best money to give to charity is money that was never cut down by taxes.
For folks that want to pass their estate to their kids, then using mortality tables makes sense for estimating your age of death and you should enter an Effective Tax Rate to tell the program about your heirs' expected marginal tax cost to get the money out of your tax deferred account after you're gone.
If you do not have heirs you care about, your planning horizon is completely different. Then you should enter something like age 100 for life expectancy. If you die early and you didn't reach the SS claim age breakeven or you never got to RMD age, it's not like you get a worse seat in heaven. Your financial risk is living a long time, even though such great age is unlikely. So researchers use withdrawal methods like Consumption Smoothing or ABW that are designed to let you spend as much as possible during your life, but using a very old age for when you run out of money, even though making it to that age is quite unlikely.
I love that opensocialsecurity optimizes by comparing longevity-weighted, net-present-value of the social security income streams. In other words, it takes into account not just when you get the money but also your chances of being alive to benefit. (It is the only readily available program that does that.) I agree with your conclusions that a single person should "no brainer" postpone till 70. With a couple, I've elected to follow opensecurity's recommendation's over pralana (in our case me taking at 70, wife at 62), and then adjust pralana's roth recommendations around that.
good feedback. So using Pralana, I looked at the value of my taxable investments (where my SS payments end up) over time by exporting the balance sheets and compared taking SS at 62 vs. 70. Looking at the table, found I'm better off taking SS at 62 until age 95, which is well beyond my life expectancy. Then I used the tables to compute the "expected value" of my account value using a longevity PDF derived from the SSAs Actuarial Life Table and also see the result favors 62 vs 70. The answer varies depends on the taxable account RoR% and your effective tax rate. It makes sense that compounding the earlier SS payments can at some point be worth more than the higher payouts taken later. I may be using the tool incorrectly but that was my conclusion. FYI, I find using an "expected value" metric ( a single number ) to compare scenarios is easier (than comparing graphs or big balance sheet tables) and it take into account the statistical outcome of your life expectancy.
The result of a breakeven at age 95 sounds odd. Perhaps you are using a nominal return for your investments for the comparison? SS goes up each year you defer and is also adjusted for inflation, so any growth rate you use for comparison has to be real, not nominal.
As noted above, I'm not thrilled with Pralana's approach. Retirement researchers use the Long TIPS rate as the discount rate so that you are comparing a guaranteed, inflation adjusted return on both sides of the comparison, instead of comparing SS benefits to the hoped-for-but-might-not-come-close-to-happening rate of inflation adjusted growth of your risk portfolio that you input.
Operationally, that means that you spend down bonds while you wait for SS as the larger SS payments give you similar security as the bonds you are spending. In Pralana, that means reducing your bond allocation a little every couple of years while you wait.
I would caution you not to base your SS decision on the estimated RoR in your taxable account. The correct discount rate, from an actuarial point of view, is the RoR of something with similar risk, e.g. TIPs or an Annuity. If you just go off relative RoR you'll likely think to claim at 62, since it has higher ER, similar to the way "mode 1" in Pralana will often erroneously recommend huge roth conversions into your high paying taxable accounts.
Rather, the reason that with two incomes it often makes sense for one to take at e.g. 62 and one at 70 is that it is the most efficient way to get "longevity insurance" for the younger person. Due to the peculiarities of how our tax system works, a surviving spouse gets the SS of the higher of the two. So once you figure in the NPVs and longevity probabilities, it often makes sense for the person with the lower income to claim at 62, and the person with higher SS at 70. The NPV of the extra years of lower income mixed with their inheriting the higher SS of the person waiting it out comes to more than other options. But of course it depends on the particularities of the couple.
