I am VERY new to working with Pralana (3 evenings). I am having difficulty with how to include various kinds of living expenses, most of which will be essential expenses. My impression is that these should be created as "phased expenses", and without an end date if expected to be for all of remaining life. It took some time to understand this; so please correct me if I have not understood Pralana correctly.
My problem is the some of my living expense amounts can be treated as just affected by the overall inflation rate. But I want to include the potential that some of these expenses will by affected by an inflation rate that is a little larger than general inflation rate. How would I do this? It appears that everything in "phased expenses' are automatically affected by the general inflation rate, and cannot be changed. Miscellaneous expenses allows for including the general inflation rate or no inflation it appears, but not a plan/scenario selected rate different from the general rate.
Advice on how to manage this element of planning will be appreciated.
Assuming you're using the online program, take a peak under Review > Expenses > Expense statement. (Check the box for "show hidden columns".) You'll see toward the left all the internal constituents of what Pralana considers essential expenses. These essential expenses are entered under various sections of "build", and you should utilize those as much as possible: 1) healthcare, 2) children (e.g. college), 3) Long-term care, etc., 4) housing costs, 5) miscellaneous and phased expenses that you check as "essential" on each of their input screens, 6) Taxes are also essential, since you have to pay them, and are calculated automatically.
Some of these allow separate inflation rates, but not all. (Look under: Build > Get Started > Scenario Assumptions > Inflation to set separate inflation for: health care, long term care, medicare premiums, irmaa limits.) If you had a phased expense which you expected to rise greater than the general inflation, you'd need to manually "hack" it in by entering different amounts under different phased dates (e.g. manually raise the real phases amount 2% above inflation every five years).
We all can help you with any of the above areas if they are not clear from the program itself.
I see; perhaps I was expecting too much flexibility. Most of my specific expense categories are entered into the 'phased expenses' section, where only the general inflation rate can be applied. (I am unclear what you meant by '5)variable and phased expenses' as if there is a separate page labeled 'variable'? Or just the page labeled 'miscellaneous' ? It seems only healthcare and long term care expenses are designed to have a different inflation rate. ( I have no college expenses).
It does seem a little kludgy to create multiple phases just to accommodate this. I would be surprised if I am the only person who has different expenses that are subject to different inflation rates. Rather than introducing sizable step changes in expenses I know how to add in a 'supplemental' expense item for the same category that can be allowed to rise with inflation using a small fraction of the base amount. Won't be exact, but close enough that it would require creating a new phase only for every 8 or 10 years.
Nonetheless, thanks for confirming that the reason I cannot figure out how to do this in pralana is because it doesn't exist, rather than I just could not see it.
(I am unclear what you meant by '5)variable and phased expenses' as if there is a separate page labeled 'variable'? Or just the page labeled 'miscellaneous'
Yes I had a typo and meant "miscellaneous and phased". I corrected my reply.
I see the ability to set separate inflation rates also for: health care, long term care, college, medicare premiums, irmaa limits.It seems only healthcare and long term care expenses are designed to have a different inflation rate. ( I have no college expenses).
I would be surprised if I am the only person who has different expenses that are subject to different inflation rates. Rather than introducing sizable step changes in expenses I know how to add in a 'supplemental' expense item for the same category that can be allowed to rise with inflation using a small fraction of the base amount. Won't be exact, but close enough that it would require creating a new phase only for every 8 or 10 years.
Out of curiosity, what are those expenses that you're putting under phased? And can you say more what you mean by "add a supplemental expense for the same category that can be allowed to rise with inflation"? Wouldn't that require the same "kludgy" phased entries every 8-10 years as simply adjusting the phased amount itself?
It's too bad the "college" category is age-dependent: otherwise you could "hack" it for your phased expenses. 🙂
I have broken my 'day to day' living expenses out into 14 different categories (grocery, major home repairs, small home repair/additions; electronic services, insurance (non home, non medical), local travel, entertainment, dining out, car maintenance, etc) because I expect I will want to play with starting with more designated to an item as a scenario plan. These are included as phased expenses, and right now I have just the single phase of 'the rest of our lives'. I would like to consider if some of these items increase faster than general inflation; groceries may increase faster than the inflation rate for other items; or slowly over time we go to restaurants a little more frequently or to progressively higher priced restaurants, or do progressively more subscriptions to theatre, concerts, or other things. Modeling that slow progressive liberalization of choices would be most similar to an inflation factor. But there is no clean way to do that in Pralana.
Two work-arounds come to mind. Your suggestion of just every few years declare a new phase of life, and bump up the expense amount for whatever category I want to examine effects would insert a 1-year crossover of a stepped increase in expenses, which could be quite discernable in the yearly total expenses unless the phases were relatively short; but that might use up all the phases Pralana currently permits in a scenario, leaving nothing for going into long term care or other events. In contrast still kludgy approach is to modestly bump up the category expense at start of the phase so that each year it is increasing more than the general inflation amount on what would be the actual intended year-1 expense amount. There would still need to be more than 1 phase created to reset the adjustment amount after some years. But that would be a smaller amount and would allow for creating fewer phases of life where that is needed. In contrast, it means the early years of a phase have somewhat over-estimated the categories expense, but the under-estimation at the end of the phase is much smaller than for an equally long phase the other way, and the intended slow increase in the expense is what Pralana incorporates. Initially I had thought of doing it as the intended initial expense size shown explicitly and then an addition expense in the list labeled as the excess inflation adjustment proxy, but it probably is just simpler to put it all together in a single expense category item.
Kludgy both ways, trading number of short phases needed to keep the under-estimation error within a specific size vs fewer, somewhat longer phases where the phase starts with a degree of over-estimation and ends with a smaller size under-estimation. .
Another option is to utilize a personal rate of inflation (rather than CPI) as Pralana's base entry. If you are going to go into that level of detail, you would be keeping your own spreadsheet of expenses with all these categories, right? And so you'd be able to produce a weighted average inflation of all of these? With that baseline figure you would need far fewer phases, since the weighted average personal inflation would only change when something significant changed in your "mix". You'd want to use Pralana's health care estimator, though, and you'd need to carefully set Pralana's health and housing inflation settings, since they would be in relation to your personal rate of inflation rather than the cpi-u. You'd also probably want to use "real" dollars rather than "future" dollars to do your planning.
You might recommend the programmers add customized inflation rates for phased entries using the built-in Feedback option. They will be rolling out a feature voting system soon, so you'd find out how many others shared your desires.
Thanks for the suggestion. Doing it that way would seem to make it exact for every year and would not require any additional phases for incorporating that. And since in the various tables of yearly expenses, cash flow, etc, Pralana does not break down the expenses by my categories, just the aggregate essential and aggregate non-essential totals, I would actually loose nothing at all in what information the program provides me as outputs. A disadvantage would be that whenever I make a change in how much I want to input as the baseline amount for a category of expense, the fraction of the overall expenses that is the one or two with odd-ball inflation will change, so I would need to recalculate what the aggregate inflation value would need to be. To me that feels counterproductive to Pralana's ability to enable me to have all the different categories explicit and independent of each other.
In the end, perhaps, restoring a little perspective is useful. This is all just estimations and nothing can be guaranteed even semi-precise once a few years out in the future. I don't think anything I'm thinking about in this matter will have much difference than just not worrying about expense-category customized inflation.
In the end, perhaps, restoring a little perspective is useful. This is all just estimations and nothing can be guaranteed even semi-precise once a few years out in the future. I don't think anything I'm thinking about in this matter will have much difference than just not worrying about expense-category customized inflation.
Modeling is as much an art as science. I've learned over the years that simplicity is better than complexity. You want to make sure you capture the big stuff. So stick to as few asset categories as possible (stocks-bonds-bills, not worrying too much about exact types of stocks and exact types of bonds), and "big expenses" (e.g. health care), LTC, and so on. One reason I'm attracted to variable withdrawal methods (which are not really modeled much in Pralana) is that they constantly adjust, and annuitization methods in particular are self-correcting.
you wrote: One reason I'm attracted to variable withdrawal methods (which are not really modeled much in Pralana) is that they constantly adjust and are self-correcting.
I think that variable methods are what I will be very interested in. It seemed like Pralana has a variety of variable spending strategies built-in to examine. How does a variable withdrawal method differ from what Pralana has? Or if just terminology, what is insufficient in what Pralana does for preference?
@prekob Yes, Pralana has a bunch of variable methods to explore, like guyton-klilnger and fixed % with floor and ceiling. My own preference, which fits my own situation, is to use an actuarial method. This is perhaps better known as the annuitization method, since it treats withdrawals from your portfolio like a home-made annuity, doling out income evenly across time based on expected returns (using the PMT equation). Pralana contains one actuarial option which is close to the boglehead VPW method. (Unlike VPW, for better and worse, Pralana's version sets a floor for variable spending at 50% and a ceiling at 200% of the previous year's variable portion.) Another deceptively simple example of an actuarial method is RMD withdrawal, based on using the IRS RMD tables. My own preference is to use the TPAW method, which looks at the npv of your expense stream and income streams till death, covers remaining essential expenses with safe bonds, and annuitizes the leftover amount across time. This is also sometimes called "lifecycle model", but can get a bit esoteric compared to the other methods.
I agree with @jkandell. Keep it simple. Don't let perfect be the enemy of good. This is a long-term planning tool that's driving based on a whole lot of assumptions, inflation being just one of them. It will never be perfect. You should be phasing expenses anyway, as they'll certainly change as you age. Build your anticipated cost increases into those phases if you want, but be very careful trying to predict the future. Especially with historical simulations, you can backtest your plan against times of high inflation.
