Stuff Yaron Finds Interesting

Technology, Politics, Food, Finance, etc.

ESPlanner – Figuring out life insurance, retirement and more

How much life insurance do we need? How much do we need to save for retirement? How much do we need to save for our daughter’s college education? These are basic financial questions and they are unanswerable because they require perfect (or at least reasonable) knowledge of the future and as the song goes “the future is not ours to see”. But, regardless, we still have to muddle through. So this is where the program ESPlanner can be helpful, if you understand what it’s doing and what it’s limitations are. Below I explain how our family uses ESPlanner primarily to figure out how much life insurance to get but also as a spot check for our retirement and college savings plans. [Note: Updated on 12/23/2014 to account for switching to ESPlanner Plus]

1 What it is

ESPlanner is a software program that lets you enter in (as explained in gory detail below) a ton of information about your past, present and future financial life. It also lets you enter in data about what happens if you or your significant other (if you have one) dies. The point of entering all this data is to try and figure out things like how much life insurance to get for each person, how much to save for retirement, etc. It also makes it possible to model the effects of different future financial decisions. For example, what if you want to buy a vacation home? Or retire to another state? What about taking Social Security payments early? ESPlanner has programmed into it all the federal and state tax laws (with projections of how they change over time) so it can actually capture the consequences of these decisions and let you see what happens.

2 Why we use it

I use ESPlanner primarily to help us figure out how much life insurance to get. But it’s also very useful in forcing us to think through our financial lives and understand where we think the money is going to go. To a lesser extent I use ESPlanner to spot check our retirement and college plans as well as get a general sense of how reasonable our spending and savings rates are. Outside of helping to get an actual figure for life insurance I really use ESPlanner as a sort of financial check up.

3 Limitations

ESPlanner can’t predict the future. For example, it has tons of information about a wide variety of tax laws. This is really useful in understand the tax implications of various choices (want to model choosing a Roth 401(K) versus a traditional 401(K)? ESPlanner can easily handle that) but remember tax laws can and do change. Of course, ESPlanner can let you put in some of your predictions for those changes but that isn’t useful to me since I have no idea where things are going to go. In fact, ESPlanner requires one to put in a non-trivial number of predictions about the future. Are they reasonable? Beats me.
The point is summarized by a very old software aphorism called GIGO - Garbage In, Garbage Out. If the data we put into ESPlanner is wrong then we shouldn’t be surprised when the recommendations are wrong. This isn’t ESPlanner’s fault. But still, as someone said, it’s hard to make predictions, especially about the future.
Which means one should take ESPlanner’s recommendations with more than a little bit of salt.

3.1 Inheritance Tax

ESPlanner has no logic regarding inheritance tax. For the vast majority of people that is probably o.k. because the Federal inheritance tax exemption is quite large and most states either don’t have inheritance taxes or a reasonably large exemption. But if you are lucky enough to have enough money that any of those taxes apply then using ESPlanner can get you into real trouble. If your estate is big enough for this to be a problem then anyway I suspect you have to hire an estate planning attorney to help you figure out how to structure your estate to minimize taxes. You will need to talk to that person to figure out what tax exposures you have and to enter those manually into ESPlanner.

4 User experience (UX)

There is now an on-line version of ESPlanner but I’ve never used it mostly because I have more than a little bit of experience in securing cloud services and so I try hard not to put any useful data in “the cloud”. Thankfully there is a software version of ESPlanner one can download but I find its UX to be pretty painful.
It took me a while to figure out how the data entry dialogs work, although once I got the basic idea they were easy. The painful part is anytime I want to change anything. Couldn’t they just let me enter a change right into the table dialogs? Please?
Another problem is that the implications and functionality of certain entries are extremely difficult to figure out and I pretty much never find the help dialogs an actual help. In most cases I just have to play around with different values in the entry to figure out what the heck it’s actually doing and ask a lot of questions in the ESPlanner website.
That having been said the introductory document is actually quite reasonable and the program does work so I think it’s worth the pain, but pain there is a plenty of pain to be had.

5 Cost

In a world of $5 programs for smart phones and free ad driven websites the cost of ESPlanner can seem a bit extreme. The basic version costs $149. But I now use ESPlanner Plus and it costs $200. After a year I have to spend another $50 (for either version) to renew the subscription so I can continue to get updates. However given the complexity of what ESPlanner does and given how small the audience of people who I suspect will ever use it I think the cost is reasonable. ESPlanner contains federal tax rates, state tax rates, social security rules, medicare rules, logic to predict how those rules will change thanks to things like inflation, etc. It’s a whole mess of very detailed work and the laws it is based on change every year so the code has to be updated every year. So yes, it’s expensive compared to most programs but if there’s another program in the same league that costs less I’d love to hear about it.

5.1 ESPlanner vs ESPlanner Plus

Until this year I used ESPlanner, the basic version. I have now switched to ESPlanner Plus. The main difference between the two is that ESPlanner Plus allows one to run Monte Carlo simulations of different portfolio allocation results. Since Monte Carlo simulations use a normally distributed curve to approximate stock market returns and as the stock market doesn’t follow a normal distribution the feature isn’t very useful in my mind. But it brings along a subsidiary feature that I want very much. This feature is the ability to create different portfolios of various investments and then assign them to different years. This allows me to simulate things like expected returns as I switch our portfolio from stock heavy to bond heavy as we get nearer to retirement. Unfortunately making this all work is really quite difficult as I explain in gory detail below. For a long time I didn’t care because I was sufficiently far away from retirement that the details seemed kind of spurious. But alas, that is no longer true.

6 Interpreting the results

I relate our actual settings for ESPlanner (more or less, no promises of accuracy) in section 7↓. But since those are long and tedious I first describe our how we use the results.

6.1 Life insurance

After pressing the “Create Reports” button and then ok I first check the Inputs and Assumptions section of the report. I’ve found on multiple occasions that I either entered in something wrong or had values apparently disappear. I think the disappearing value problem is caused by forgetting to press ’apply’ and not getting a warning that one has an unapplied value when changing tabs. So it’s critical to check that what you expect is actually there. I then usually walk through the Details sections to see if the numbers seem reasonable, again, another check for errors.
When looking at the report the first place I usually go to is Suggestions-Annual Suggestions to see the living standard per adult. I then go to my financial program and take our total expenses for the last 365 days, divide it by 2.7 (we estimate that our daughter costs 70% of us) and multiply that by 1.6 and divide it by 2. The 1.6 represents the assumption that 2 people live as cheaply as 1.6 and dividing by 2 normalizes the 2.7 “people” on the 2 people benchmark. So basically I take (2+1*0.7)*1.6/2=2.16 and divided that by our expenses (minus taxes) for the last year. I then compare this number to the living standard per adult number. Typically the number recommended by ESPlanner is larger than what we actually spend per person according to this calculation.
What this says is that according to ESPlanner we are over saving/under spending. It also argues that ESPlanner’s life insurance recommendation (made in the section “Suggestions - X Suggestions” where X is the current year) are too high since we don’t live at the spending level ESPlanner says we could live. But keep in mind that ESPlanner assumes that our future predictions of income are correct. The reality is that we can’t know that. With the economy in bad shape and with me working in an industry that actively discriminates against older employees I have to take into account the very real possibility that I may end up unemployed or under employed for a very long time. Sure, right now everything is roses, but who knows what the future holds?
So it’s tempting to estimate the life insurance amount down but I don’t and the reason is that if the future is better than I hope then I’ll need even more life insurance and it gets expensive to buy more later. So, since I can afford the insurance premiums now and since it’s very easy to lower the amount of life insurance/reduce our premiums without negative consequences later in case I can’t afford the premiums I tend to err on the side of accepting higher premiums.

6.2 Saving for retirement

Because we save more than ESPlanner recommends I tend to largely ignore its suggestions for saving for retirement, especially the ones that tell me to run down our savings now and save more later. The logic ESPlanner is using is called consumption smoothing and it makes perfect sense if and only if one knows the future with high certainty. Still, the fact that the numbers are lower than what we are actually saving is a good check that our savings program is reasonable. It’s also interesting to run scenarios where we retire early and compare those to our current savings.

6.3 College Savings

I also look at the College Savings amounts (available under Details - 529 Savings) and use it to check our current college savings. Once I’ve adjusted for our projected GET savings I use this to help check how much we are saving in our 529.

7 Settings for ESPlanner

In this section I walk through the settings I try to use in ESPlanner in gory detail. The sections below, except for Family Information which is filled out when creating a planner, match the tabs on the left side of the planner UX in version 2.28.0.

7.1 Family Information

Children in household I accepted the default that our daughter will “leave the household” at 19. Keep in mind that my estimates of college costs include room and board so I account for her living expenses there even after she has “left” the house.

7.2 Planning Method

There are three methods that ESPlanner supports.
Economics-Based Planning This uses consumption smoothing, the idea that one should try to spend the same amount of money every year one is alive rather than say living badly when one is young in order to live better when one is older. When in this mode the program will try to calculate how much one should save/spend each year in order to achieve smooth consumption. I don’t really use this mode anymore.
Economics-Based Planning with Upside Investing The goal with this model is similar to the previous except that in this model the assumption is that one will want to start a process where by one moves all of ones “risky” (read: stock) assets to “safe” (read: cash, treasuries, TIPS, CDs, anything federally insured) assets over some period of time. The model lets you specify how much money to put into stocks each year until the start of the transition to safe assets. Once the transition starts no more additional money will be put into stocks and the stock portfolio will be sold in equal pieces each year until the target date when all stocks are supposed to be sold. The program will then calculate what would happen if all of ones stock money went away and one only had the ability to survive on the safe assets. It will then use (bogus, see previous comments on the Monte Carlo method) numbers to show probabilities that the stock portfolio will survive and provide extra income. This model doesn’t actually represent how I intend to save for retirement. My goal is to have a stock/bond ratio where the bond part getting heavier and heavier each year. But I still intend to rebalance to that goal amount and put new money in as it’s earned since we will make the transition before we retire. None of which upside investing handles so it’s pretty useless to me.
Economics-Based Planning with Monte Carlo Simulations This is a more sophisticated beast. It’s reason for existing is that it allows one to describe ones portfolio and then run a Monte Carlo simulation to see how that portfolio might do. The Monte Carlo part unfortunately is just plain silly. It’s been known at least since the late 1960s that stocks follow a “fat tail” distribution that is not normally distributed. Or in English, good and bad things happen way more often than anything related to a Gaussian distribution (log normal or otherwise) would predict. It’s also unknown if the correlations between assets is stable, in fact, there is a lot of evidence that shows that when things go seriously south correlations tend to increase. But what the Monte Carlo model does let me do is create a bunch of portfolios representing my target stock/bond distribution over different periods of time and then set up those portfolios to be used by the program.

7.3 Planning Method - Monte Carlo

7.3.1 Build Portfolios

The program supports up to 10 portfolios, Default and Portfolio 2 - 10 (or whatever custom names I assign). Each portfolio consists of a set of assets and their mean and correlation. The goal then is to create portfolios that represent our portfolio over time. To make things more interesting portfolios can be taxable or tax exempt.
What I want to model is how my asset allocation will change over time as I move from 70/30 stock/bonds to 0/100 stock/bonds. There are two challenges. First, how do I map the transition into the portfolios I have available? The second is, what assets do I put into the portfolios?
For how to map the transition I just use algebra. In an ideal happy world I would have enough bonds when we retire that we can meet our minimum needs indefinitely. Then if by some miracle there is excess money we can put that in stocks and goose our living standard at best or just live our minimum at worst.
Currently I simplify things by using a portfolio for each decade. So my current portfolio is the 70/30 portfolio. I then calculate how many decades until I want to hit 0/100 and then divide 70 by the number of decades and subtract that over and over again. Again, just simple algebra. So let’s say that I’m in my 40s and want to be 100% bonds by my 80s. So that would add 50s, 60s, 70s and 80s or 4 decades. 70/4 = 17.5. So portfolio 0 (for my 40s) would be 70/30, 50s = 52.5/47.5, 60s = 35/65, 70s = 17.5/82.5 and 80s = 0/100. This gives me 5 portfolios. For extra accuracy I could split each decade into a taxable and tax exempt portfolio since I tend to have different assets in my taxable than my tax exempt accounts. But at some point it’s just too much spurious detail so I’m not going to worry about it. Instead I use the same portfolio for both taxable and tax exempt accounts. Note that I could easily come to regret this as our taxable and tax exempt portfolios are not the same. We tend to go bond heavy in our tax exempt money and stock heavy with our taxable. But honestly I’m reaching the end of my patience with this game so I have decided to simplify before I give up. I must admit that the endless issues I’ve had with E$Planner have worn me down.
This then brings up the problem of what to put in the portfolios. I tried to create my own assets but that turned into a complete disaster. There is something seriously screwy with E$Planner that makes it lose its mind when users enter their own assets. So I had to give up on that strategy and instead just try to use the assets that are already there.
In theory I just have two types of assets, safe and risky. Risky has a hoped for return of about 4% in real terms. Safe would have a real return somewhere between 0% (what we are currently seeing) and 2% (what we have historically seen).
For “Safe Zero”, that is, a safe asset returning 0% average real return we have to use a tiny bit of algebra. The pre-programmed cash asset that comes with the program returns an average of -3.04% and the Short Term Government Bonds return 0.6%. So we can create a synthetic 0% mean return asset as follows:
(( − 0.03X + 0.006*(1 − X)))/(2)  =  0  − 0.036X + 0.006  =  0 X  =  0.1667
So 16.67% of safe zero money should go to cash and (1-0.72)=83.33% should go to Short Term Government Bonds.
For “Safe Two”, that is, a safe asset returning 2% average real return I need to use another asset because the math requires the second asset have a return that has a larger return than twice my desired return or the math won’t work out. In other words, I need an asset with a return higher than 4%. So I picked Dimensional Large Cap International with a return of 7.12%. Since I’ll only use the median return I can largely ignore the variance and beta.
(( − 0.03X + 0.0712*(1 − X)))/(2)  =  0.02  − 0.1012X + 0.0712  =  0.04 X  =  0.3083
So 30.83% should go to cash and 69.17% should go to Dimensional Large Cap International to get the desired 2% return.
Finally for the risky 4% return honestly the return should probably be a bit lower so I’m o.k. using the Inflation Indexed Government Bonds Fund with a 3.72% return.

7.3.2 Implement Portfolios

So, putting this all together for safe zero plus risky, I would get an allocation like:
Stock Portion 70% 52.5% 35% 17.5% 0%
Inflation Indexed Government Bonds Fund 70% 52.5% 35% 17.5% 0%
Cash 5% 7.92% 10.84% 13.75% 16.67%
Short Term Government Bonds 25% 39.58% 54.16% 68.75% 83.33%
Each column is one of the 5 portfolios covering each decade.
So now I have to go to the Implement Portfolios tabs and calculate the years for each decade and enter them all in, three times, once for taxable, once for tax exempt for me and once for tax exempt for my wife. Make sure that on the last decade you set the drop down for last year to the last year the dial supports. This is the year we have set as our death year. I try to eyeball the Portfolio Choice table to make sure I set everything the same. And yes, this interface really sucks and makes experimenting with different portfolios and allocations incredibly painful.

7.3.3 Spending Behavior

This lets me spend based on different assumptions regarding the actual returns one will get on ones portfolios. This setting is not that useful to me since I’ve already put my hard guesses about portfolio returns in the “Build Portfolios” section. So I set this to Spend aggressively.

7.3.4 Precision

I just take the recommended setting.

7.4 Earnings

Set Retirement Date I picked 65 for myself based on nothing useful. After all, I believe the social security retirement age for my age cohort is 67 and I wouldn’t mind being in a position to retire by 55. I’ll probably play with those numbers but for life insurance purposes I am using 65.
Employee Wages I took my current compensation package (including stock grants and bonuses) and my expected maximum wages (based on a complete blind guestimate, but I did pick a figure reasonably close to what I make now since at my age I’m likely getting close to my maximum earnings) and calculated the geometric yearly average increase. The formula is CS(1 + X)Y = MS where CS is current salary, X is the growth rate I’m trying to solve for, Y is the number of years until I think I will achieve maximum salary and MS is maximum salary. A little arithmetic gives X = (MS)/(CS)(1)/(Y) − 1 . I assume once I hit maximum my salary will only go up at the rate of inflation.

7.5 Assets & Saving

7.5.1 Regular Assets/Regular Assets

Just a data entry exercise (although I do enter my ETFs as mutual funds not individual stocks). The only big gotcha was remember to not include any cash we are using for our reserve fund. That money is entered in the reserve fund section later.

7.5.2 Regular Assets/Upside Investing

I ignore this section because Upside Investing hasn’t proven very useful for me. It doesn’t let me change my stock/bond percentages over time and so doesn’t really represent how I invest.

7.5.3 Current Saving

Because I use upside investing this dialog is not active. Even when I don’t use upside investing I usually leave this blank to see what ESPlanner recommends.

7.6 Retirement Accounts

7.6.1 Assets

The only interesting note is that I put our Roth 401(K) as a Roth IRA because there was no option for Roth 401(K).

7.6.2 Key Ages

I left the values as given. Which means we make our last contribution at 64, start to withdraw at 65 and continue withdrawing until 100.

7.6.3 Contributions

I use a Roth 401(K) so I put my contribution into the Roth IRA column. I do get an employer match and I need to see what I can do there since I suspect ESPlanner treats the Employer match as going into a 401(K) and not a Roth 401(K). Note that I specify a 0% growth rate because I already contribute the maximum and expect the maximum contribution to only go up at the rate of inflation. If I’m reading the report correctly I believe that ESPlanner treats both the employer contribution and the Roth IRA contribution as increasing at the rate of inflation so everything appears o.k.

7.6.4 Special Withdrawals

Not planning any

7.6.5 Smooth Withdrawals

I used the defaults. This means we aren’t planning on annuitizing any of our assets. As we get closer to retirement I may change my mind. But given that apparently there is no level of financial malfeasance on the part of the financial sector that will be punished by our government the idea of handing over a bunch of money with a vague promise of a guaranteed return for life is just too absurd to accept. Which is sad because there are actually very good reasons why an annuity is a great idea both in theory and in practice. But, alas, as I get older the risk reduces so perhaps at some point I will use annuities. But for now I’m just not going to worry about it.

7.6.6 Choice of Annuity

See previous section for why this section doesn’t apply.

7.6.7 Upside Investing

As mentioned above, I don’t use this.

7.7 Pensions & Annuities

My wife has a tiny pension from her job but it’s so small it’s not worth tracking, other than that, we have no pensions nor annuities so neither parts of this section apply.

7.8 Social Security

7.8.1 Benefit Receipt

Because I expect to get a higher Social Security payment and thus will benefit more from suspending I specify myself for file and suspend. This is something I’ll play around with more as I get closer to retirement. The rest of the values are set automatically and I leave them alone.

7.8.2 Future Covered Earnings

Since all of our earnings are covered by social security we just used copy to move the figures over here.

7.8.3 Past Covered Earnings

Once upon a time the IRS used to mail out statements (they are supposed to start again soon) with a list of past earnings. We had kept our old ones so I entered the data from there and used our tax records to get the missing years. It’s theoretically possible to get the earnings statement via ssa.gov but I’ve never gotten that to work for myself. I don’t remember ever blocking access on-line but it certainly sounds like the kind of thing I’d do.

7.8.4 Medicare

Took the defaults which have us enrolling in Medicare part B at age 65.

7.9 Estate

7.9.1 Estate Planning

Desired Percentage Change in survivor’s living standard 0% for both. Based on everything I’ve read spending in retirement doesn’t usually go down unless one has absolutely no choice. One, after all, wants to enjoy retirement. Not just sit in a room waiting to die.
Special bequest No plans for one.
Funeral expenses in today’s dollars It seems like funerals cost around $10k for all the bells and whistles.

7.9.2 Life Insurance

In general I want E$Planner to make a life insurance estimate for me. That’s one of the main benefits of the program. But there is an interesting complication. My employer has a very nice policy where they let all unvested stock grants vest instantly if the employee dies. So I treat this as effectively an insurance policy set to the amount that my grants generally hang around. I use my employer’s long term stock price (not it’s current price) as the estimate for the face value. Since this is effectively term insurance it has no cash value.
I also have a mandatory life insurance policy through work so I throw that in too, again, this is effectively term insurance so it’s face value, not cash value.
Note however that since it’s possible I might inconveniently die while unemployed when I actually purchase insurance I tend to ignore these numbers. I include them here though so I can understand the cost of paranoia.

7.10 Primary Home

7.10.1 Current Home

We rent so I put in our monthly rent. In “other monthly rental expense” I put in our utilities and rental insurance costs.

7.10.2 First Change of Home/Second Change of Home

One of our big financial goals is to have a home owned free and clear by the time we retire. The idea being that this will stabilize our cash flows during retirement since we won’t have to come up with the rent every month. It will also provide some hedge against inflation. Our thinking is to try and save enough in cash to buy a residence rather than taking on a mortgage. I randomly picked 10 years from now for when we will buy the home. I set the purchase price based on housing in our area. Property taxes I got by looking at local houses on Redfin. For homeowner’s insurance I used the estimate from the New York Times rent vs buy calculator. I use 1% for maintenance costs. And I used the New York times estimate for closing costs.

7.11 Vacation Home

Neither have one nor plan on getting one.

7.12 Real Estate

We do not own any nor plan on investing in actual real estate assets directly. If we were going to invest in real estate directly we would probably use REITs. But honestly we have enough exposure to housing through our stock index funds (all the public home builders) that I don’t stress this much.

7.13 Special

7.13.1 Special Expenditures

It’s tempting to put college expenses here but that isn’t right in our case, instead college expenses are covered in the 529 section.
We buy new cars every 10 years. So I take the current cost of new cars (based on our recent purchases) and remembered to add the 9% sales tax and then set that up as an expense that recurs every 10 years until we die. Yes, I realize we probably won’t be driving cars at 90+ but that just means there will be some other expense to take its place. I do wonder if car2go or zipcar might eventually change this logic, but I’m not wiling to bet on that quite yet.
The other cost I put in here is Private School. In addition to the base cost we also have to include the optional mandatory donations. Based on looking around Seattle we can expect 1st – 5th grade to cost around $17,000 and 6-8th to cost around $17,500. It looks like high school costs somewhere in the range of $29,000 for 9-12th (no, I’m not kidding, the 1-8th grade costs are taken from Villa and high school is taken from University Prep in Seattle). We are actually planning on sending our daughter to public school but I want to make sure that we are budgeted for her to go to private school if it turns out public school isn’t working.

7.13.2 Special Receipts

I am not counting on any inheritances for modeling purposes. I wish my parents a long happy life and hope they spend every penny they have saved in a wonderful retirement.

7.13.3 Reserve Fund

The emergency fund is intended to handle unexpected expenses and/or unemployment. Once we retire we are completely dependent on our retirement savings and so we don’t need the reserve fund anymore. Put another way, it is a requirement that our retirement assets are structured so that a reserve fund isn’t necessary. So I set up the reserve fund to have a fixed amount of money from now until we retire. I set the growth rate to 1% nominal (meaning a negative return, which is mostly what I’m seeing these days).
I put the nominal rate of return at 3% (equaling the inflation rate for a real rate of return of 0%). The size of the fund is tuned to how long we want cash to keep us going in case I lose my job. We use that value for step 2. We skip 3 and we don’t worry about 4 since the goal isn’t really to increase it and we would more or less like to stabilize our spending.
Note: The reserve fund money is treated separately from the money entered in “Assets & Savings”. This confused me because I just put all of our cash in “Assets & Savings” and was surprised when our net worth was higher than it should have been. So I have to make sure to subtract the current value of the reserve fund from our Assets & Savings.

7.14 529 Plans

I have an article I wrote explaining my projections for college costs. But I have to make an adjustment to those numbers to account for the fact that we have a Washington GET plan. The way I do this is as follows: I take the UCLA out of state tuition and fees (available here) which acts as my “reference school” for college costs and I subtract from it the current UW in-state tuition and fees (available here) plus an extra 25% and I use that as the base cost. The extra 25% accounts for the fact that we are saving for 5 years of costs in the GET program but spending it across 4 years (which is explicitly allowed in the GET plan).
So the base number is $13,194 + $22,878 - ($12,394 * 1.25) = $20,579.50. Then I add UCLA Room and Board for off campus apartments as well as books and supplies to get a final yearly number of $20,579.50 + $1,599 + $14,571 + $585 + $1,638 = $38,972.50. That covers tuition, fees, room and board and books. This does not include health insurance because I am assuming we will be able to offer our daughter health care coverage under our policy. I then multiply this by my estimated rate (see article) of 2.85% real increase in yearly college costs. So for my daughter’s first year of college in 2024 this would be: 38972.5(1 + 0.0285)10 = 51618 in 2024 in 2014 dollars. Note that the 2.87% probably underestimates the benefit of the GET program since it rises at the rate of tuition/fees which typically go up faster than room and board, but oh well.
So I enter $51,618 as “tuition” (I know, it covers multiple things) for year 2024 in today’s dollars as a qualified expense to be funded by 2024. Then I multiple by 1.0285 to get $53,089 for 2025, $54,602 for 2026 and $56,158 for 2027.
Finally I put the nominal rate of return of the 529 plan as being 3%. Which means, since I’m modeling inflation as 3%, I expect zero real growth in our 529 plan’s value. Given the nightmarishly bad 529 plan options available this is actually better than I expect to get due to fees. The only reason I even use the 529 plan is that the effective return after accounting for the tax benefit is high enough to make it worth my while.

7.15 Assumptions

7.15.1 Nominal Rates of Return

Since I am using Monte Carlo simulations this entire tab is no longer available.

7.15.2 Standard of Living

Index Set at exactly 100

7.15.3 Cost of Children

Took the default of 70%.

7.15.4 Taxes

Federal Taxes/FICA/State Income Tax Left blank (you think I know? Although this is a great screen to torture ourselves with in order to play ’what if’?).
Municipal Bonds I just calculated what percentage of our taxable assets were in municipal bonds. Note that I have no target amount. I try to place bonds as much as possible into tax exempt accounts but sometimes I need to put some in taxable for a variety of reasons.
Dividends and Capital Gains 100%. All “income” received from these assets is either capital gains or dividends.
Unrealized long-term capital gains/losses on regular financial assets I just calculated the difference between our taxable asset’s current value and their original cost.

7.15.5 Benefits

Given America’s snarling hatred for its own people has risen to such an extent that it literally allows its citizens to go without water I can only assume that the funding issues with Social Security won’t be addressed. According to the Social Security trustee's report this means that in 2033 Social Security will only be able to pay 3/4s of it’s scheduled benefits. So I set a change to Social Security to 2033 and reduce benefits by 33% so I entered -33.
The same trustee report also covers Medicare. Unfortunately it doesn’t really (nor should it) try to explain how the deficits will affect premiums. Those deficits could lead to premium rises, payroll tax rises or more likely, some combination of both. I have more faith in Medicare being taken care of than Social Security for the simple reason that a whole industry of parasites has arisen around Medicare ranging from drug companies to doctors to hospitals who try to squeeze every penny they can out of the government and in the process make Americans pay literally twice as much per capita for Medical care than anywhere else in the world. The parasites will want to keep that food train going so they’ll figure out something to keep Medicare funded. This doesn’t mean premiums won’t go up, I just don’t know by how much. So my current guess will be at least by the rate of inflation (thought that has not been the case so far) so I’ll leave the real rate at 0.

7.15.6 Demographics

I left both at 100.

7.15.7 Inflation/Other Variables

Inflation is set at 3%. This has been the long term average of inflation in the U.S. although with our government printing money with abandon one is forced to wonder what the future holds. The “other” settings are at defaults.

7.16 Contingent Plans

First, enable contingent planning.
Second, check All and Copy.
Third, double click on Contingent Plans entry so you can see the plans underneath it.

7.16.1 Earnings

No change since I want to see how much it would cost for each of us to continue our life style if one of us should die.

7.16.2 Special -Special Expenditures

I removed the dead spouse’s car and added $30k a year worth of extra expenses until our daughter is 18 to deal with being a single parent household. I picked that number out of the air, btw. I also marked it as not tax related.

7.16.3 Special - Special Receipts

None

7.16.4 Social Security / Retirement Accounts / Primary Home / Vacation Home

No Change

6 Responses to ESPlanner – Figuring out life insurance, retirement and more

  1. Greta says:

    Excellent analysis! I’ve used ESPlanner since 2005, and have yet to feel complete confidence in it (even aside from the problems of predicting the future). This is because it’s so difficult to reconcile its “Discretionary Spending” and “Net Worth” amounts to my actual spending and net worth, although I will try the reconciliation you outline in paragraph 6.1.

    Coming from an accounting background, I want to reconcile our net worth to that shown in the Quicken software I use to track all our finances — not possible. Mr. Kotlikoff explained to me in a Feb. 2012 forum post that ESP treats current debt as “future debt,” and “it’s properly accounted for.”

    I will say that the user interface has improved since 2005, though not nearly enough. It is indeed “painful” not to be able to edit one’s entries. And I agree completely that “the implications and functionality of certain entries are extremely difficult to figure out,” as you say. I keep vast notes (much like your “gory details”) on how I have “played around with different values in the entry to figure out what the heck it’s actually doing.”

    Thanks for this thorough review. Nice to know I’m not alone in this approach!

    Greta Perleberg
    Wichita, KS

  2. Greta says:

    PS In case you’re wondering who on earth I am, commenting on your blog, I found this from the ESP forum post you made. ;-)
    Greta

  3. Administrator says:

    Sorry for the delay in posting your comments but I was on vacation. I have not had too much trouble getting net worth to match up (once I figured out how to handle the reserve fund and I ignore the current value of my daughter’s prepaid tuition plan since I account for it as described in section 7.13. Discretionary spending is less of a concern to me. I make sure to enter fixed key expenses like housing or college and let the program make recommendations. I always save more than the program recommends but that is mostly because of my lack of confidence in my own future projections.

  4. Richard says:

    Thanks for this careful write-up. I’ve used ESPlanner since 2006. I’m an early retiree because of this software–otherwise I’d be too worried and uncertain to be able to stop working. (Given sub-par investment returns, have to admit I’d be glad to land some part-time work but it’s not happening) At any rate, I think my situation is fairly “predictable” income-wise and expense-wise so the consumption smoothing recommendations are pretty much gospel for me and my wife.

    We track all spending on a separate Excel sheet and monitor whether we’re on pace to stay below ESPlanner’s recommended spending level.
    We also have a $200K reserve fund built-in. That means ESPlanner disregards that much of our wealth when recommending annual consumption. I had the same problem you mention with a bug in ESPlanner overstating net worth after creating reserve fund, so be careful.

    In general, the ESPlanner net worth number is an important to check on your data input and program performance. If it’s wrong, then the software is cranking out recommendations based on a mis-stated bundle of wealth at the start.

    The program is a financial education as well as a planning tool. When coupled with detailed expense monitoring, it helps you quantify how much wealth you must accumulate to meet your basic needs, including travel and restaurants and things many of us in the industrial world feel deprived not to have.

    And it allows you to game out different levels of risk to your living standard, from investment returns, inflation, SocSec changes, Medicare changes, longevity, and tax changes.

    As you say, a crucial number is the user’s selection of real, after-inflation returns on savings and investments. US Treasury TIPS used to yield a safe 2% above inflation. With that option gone for the foreseeable future, it’stempting to take greater investment risks. But I haven’t done that because I’ve got no money I can really afford to lose–without being forced back to work. So I too have pared back my projected real returns, to a measly 1 percent in the latest run.

    Fortunately I found that cutting from 2% real to 1% real actually didn’t make that much difference to spending recommendations–and it eases my mind to feel a little more bulletproof by being able to stay with safer low-yielding investments. The relatively small drop in spending recommendation due to going from 2% real to 1% real is probably a function of being at low absolute values for real return. If I were projecting 6% real and cut THAT in half, I think it would have a large impact.

    The program is kind of clunky but since my situation is fairly simple I’ve learned the functions that work for me. Every three or four months I find myself re-checking everything including my wealth inputs. If I screw this up, I’m cooked. Have also used a telephone consultant with ESPlanner ($100/hour) to re-confirm that I’m using it properly.

    It’s great to have this tool. Don’t know of anything else like it.

    Thanks again for writing!

    I’ve also relied on the life ins. function since I’m expecting (and have programmed in) an inheritance that my wife won’t get if I die first.

    We’re both projecting life span till age 95–unlikely, but a risk-buffer. And we’ve got a declining standard of living kicking in at age 75, lower still at 85.

  5. Pingback: Derived keys and per user encryption in the cloud | Stuff Yaron Finds Interesting

  6. There’s certainly a lot to know about this issue. I love all the points you have made.

Leave a Reply to Administrator Cancel reply

Your email address will not be published. Required fields are marked *