How to model long term care insurance

I wish to purchase and enter long term care expenses. Is the "special expenditures" the proper method for entering or should these numbers be budgeted in "consumption"?


You got it right the first time. Special expenditures is the right way to do it since it's an "off the top" expense as far as you're concerned.

Note that due to the way we treat special expenses internally (we don't know that you won't be making those special expenses if you're dead) you may see life insurance expenses late in life. We're thinking about a solution to that problem but in the mean time, just enter non-taxable special receipts to offset the insurance costs if they turn up.


Dick Munroe

I'm just a user, but here are some thoughts. You don't have to do all of these, but they may be of interest:

1. Genworth has some good LTC market data on costs and inflation for many regions/cities around the US which can help with projecting future expenses. They also have data on median and maximum costs along with different types of care (e.g. nursing home vs. assisted living facility or other).
2. The inflation rate for LTC is different from traditional CPI (or at least has been) so you may want to factor that in.
3. Many LTC expenditures can be classified as "medical" from a tax perspective in ESPlanner Special Expenditures.
4. If you have a LTC policy, you may want to model policy payments (by you) and benefits received (to you).
5. Some policy benefits are considered "not taxable" as income per IRS and can be modeled this way in ESPlanner Special Receipts.
6. Each policy is different so you may be in a situation where the policy covers up to X years and you want to model Y years in case of an extended situation. Also, some policies assume you cover the first several months (or longer) LTC expenses before benefits begin.
7. For Contingency planning, this can be somewhat tricky to model until you get comfortable with it. For example, if married, the oldest person should only have contingency special expenditures or receipts until their last year of life. The longest living spouse can have special expenditures/receipts until the last year of their life.
8. Also, policy benefits and your expenditures may be different with one person on LTC vs. both if they overlap and one may be in an assisted living facility while the other is in a nursing home (or other possibilities) each with different expenses.
9. You may want to consider selling your remaining housing equity near end of life if LTC has impacted both of you. You can model this under "primary home", perhaps with a second change of home at this point. Unfortunately, it's not as easy to model in Contingency planning.


Brian, yours is an old response to an even older question, but would you be so kind as to describe a realistic scenario with hard numbers? Maybe attach a database example to illustrate?

I'd probably be most comfortable using the model where insurance premiums are recorded as a special expense, but I don't know how to account for typical LTC benefits, since I don't know where they'd occur in life. (As I'm managing my Mom's finances right now, and she's in an ALF, I get the part about tax-free income offsetting medical expenses.)

Also, how to model a policy where the surviving partner gets the remainder of the benefit? I guess you'd have to make an arbitrary decision as to the extent of LTC expenses attributed to the decedent, and use the remainder in the contingent life of the survivor?

Hi Chris,

This is a complex topic so I'm not sure how much justice I can do to it beyond my earlier comments, but will add some points here. FYI - there are many studies out there on different LTC topics. For example, how long men/women typically stay in various types of LTC and at what ages.

First, I use Genworth's annual "cost of care survey" for a lot of my raw data. It comes out each year in late April. There are also individual state reports with local city/county data in a number of areas such as homemakers services, home health aide services, adult day health care, assisted living facility, and nursing home care. You can see the daily rate range (min to max along with median), median annual rate and trailing 5 year annual growth rate. When I first started, I wasn't sure what to do with the info. This lead to a couple thousand cells of Excel side calcs of testing until I thought I had a handle on what I wanted to do and had it modeled to my satisfaction.

There are a ton of potential assumptions that you can make. Here are a few:

- Rate of LTC inflation (which has been different from CPI and differs between different LTC sub-categories). Converting back and forth from multiple inflation rates along with nominal/real has been a pain specifically. Some LTC policies (and ESPlanner models) are highly sensitive to inflation changes. What happens to you if inflation is low, medium, high?
- When the LTC occurs? I've assumed near EOL. How long for each person? Larry has written about using 5 years although you may want to refer to studies on this issue or your own best guess. There may be a progression from modest LTC to ALF to nursing home over time. Should you aim for median rates or higher/lower? Since there is a general aging of the population, there may be more demand (and higher prices) in the future, but that's just a guess. If you may change states/cities to be closer to family members, you may want to take that into account. Depending on the age differences of each spouse, you may have gaps between being on LTC or they could overlap. Etc.

Regarding ESPlanner inputs, you may want to include LTC policy premiums (as a special expenditure although this could just be part of consumption to simplify), paid out benefits per your specific policy (as special receipts, probably non-tax related or at least partially) up to the level of your annual (inflation adjusted) LTC expenses, and expenses for both spouses (assuming the person is married and the policy covers both spouses) as special expenditures (probably with "medical" tax treatment). Since each policy is different, you'll have to read through very carefully to capture the relevant points which may not be 100% obvious at first.

Near EOL, at the point where both spouses are in LTC or no one is left at home, consider selling all remaining home equity. This is straight forward in normal planning, but (unfortunately) not available in contingency planning if you have a previous change of home. For me, this has led me to temporarily (I hope) stop using contingency planning because it causes a big difference in overall results. I'm hoping this eventually gets addressed in the software.

Also, you may find the following article from Scott Burns interesting.

I know this doesn't give you a specific scenario with hard numbers, but the reality is that we're facing probabilities instead until they become hard numbers in the future. Once you get your core LTC model set up, try stress testing it with some "what if" analysis which should give you a better feel overall.


Chris...I'll put some #'s out there to try and make it clearer, but beware that LTC modeling is highly sensitive to your specific situation, policy, assumptions, etc. and yours may be very different from these.

Example #'s ONLY:
- Spouse #1 EOL 2040, enters LTC (ALF) in 2036, moves to nursing home in 2038
- Spouse #2 EOL 2044, enters LTC (ALF) in 2040, moves to nursing home in 2042

In this example, there are no overlapping periods of LTC between spouses, but some level is needed for 8 straight years.

- ALF (median) nominal annual expense $85,000 in 2036 and increasing by 5% / year
- NH (median) nominal annual expense (semi-private room) $175,000 in 2036 and increasing by 5% / year

Convert all of these to real $ (using your assumption for inflation in "assumptions" tab), then input into special expenditures (per above) as appropriate for each year. You can change the labels in special expenditures or receipts (e.g. "LTC ALF spouse #1") or something else which may make it easier to make sense of when reviewing later on. Note: If you use contingency planning, only input special expenditures up to EOL for the spouse expected to die first. Input the remaining special expenditures into contingency planning. Note #2: Contingency planning special expenditures may be different for each spouse so think though/plan for this carefully. It does make sense, but you are expected to put these in correctly. Note #3: If you plan to sell all remaining home equity (to pay for LTC or other) if both spouses are no longer living at home, you can do this easiest through normal (not contingency) planning.

Re: policy benefits...assume you purchased $100 / day in coverage that will grow by X% annually. This is highly sensitive to CPI. If CPI is > "X", then your coverage will probably be smaller than you hoped for. Assume this covers both spouses. Convert to real $ (per year) for relevant years (e.g. 2036 to 2044) and compute annual maximum benefits. For first year of LTC for each spouse, assume first 90 days of expenses are out of your pocket (appears to be typical) so not covered by your benefits. Enter benefits per year, up to level of expenses and as covered by your policy into special receipts. In your example, the surviving partner gets the remaining benefits, so you'll need to add that in as appropriate. This could be in contingency planning or regular planning under special receipts.

I haven't included the exact inputs as these vary wildly depending on your specifics which would be different from this example.

One final point, it is okay to cut through all the complexity and just ballpark estimates for (high) LTC expenditures and (moderate) LTC benefits that are hopefully in the right neighborhood. Clearly that's not the route I've described, but could be "good enough". Either way, there is a ton of uncertainty and garbage in/garbage out applies. That's another reason why I prefer a "safety factor" approach as reality may not match our profiles.

Hope this is helpful.


Thanks, Brian. Your examples are helpful.

We use cookies to deliver the best user experience and improve our site.