Rollforward of Regular Assets / Retirement Accounts from Year "0"

This situation applies to both subject matters, but I'll just give specific examples using "Retirement Accounts" tab in the spreadsheet output by the program.

One expects the end of year balance to equal the beginning of year balance, plus income, plus contributions, less withdrawals. This works for all years for which all the data is presented, but does not appear to work for Year 1 (beginning balances, as entered manually into the program, are not shown).

In other words, if I take the Retirement Account ("RA") balances I enter into the program, add RA income, add RA contributions, and subtract RA withdrawals, I do not get the RA Assets balance shown.

The same thing applies, as mentioned, to Regular Assets.

While the difference is small (in terms of it's potential impact on annual consumption calculations), it is highly problematic as I track progress during the year against my "goal", the "goal" being the year end balances as provided by the program. In that regard, the unexplained difference is substantial in terms of my overall activity in Year 1.

Any insight?

Comments

dan royer's picture

What you are seeing has to do with the conversion that takes place in the first year to "today's dollars." There's an inflation adjustment applied the first year to get things into 2016 dollars. The math is right on this, it's just not immediately intuitive.

We're interpreting the inputted balances for 2016 to be the end of year balances as of 31 Dec 2015 in 2015 dollars. So, the first thing we need to do is convert 2015 balances to 2016 dollars by multiplying by 1 plus your inflation rate. Then they earn the real return on investment during 2016 plus any savings and minus any withdrawals. After 2016 we're in 2016 dollars, so there are no additional base year dollar conversions.

Well, there you go. :) Thanks!

I just calculated the difference for both Retirement Accounts and Regular Accounts and each is "off" by 3%, which is the "Current Inflation Rate" on the "Inflation/Other Variables" tab of "Economic Assumptions".

If I wanted to eliminate this issue, should I be able to set the inflation rate to zero and reduce my nominal rates of return by the same amount? [note: I don't do anything particularly fancy regarding changing of future rates, both of return and inflation.]

Alternatively, I could goose up my 12/31/2015 balances by 3%.

Any downside you see to either of these practices?

dan royer's picture

If you set your nominal rate of return at 0% and inflation at 0% you will remove those "dimensions" and you'll see the math is very simple. In that case, if you had $1000 in Assets and Saving you'd see that you have no regular asset income and just the Saving added to the $1000. But you really don't want to try to work around any of this. It's reporting as it should be because you want to see your end-of-the year balances reflect the impact of inflation for the year. So you're seeing the reports as you should see, though I understand why it's confusing.

I typically try to explain this five or ten times a year (usually not very successfully) and often I call on Kotlikoff or Mike O'Connor our tax auditor to help explain--this typically involves long formulas and the user ends up convinced but never really understands it. :) I'm not trying to say it's not worth understanding, but I am saying it involves looking at the tables in terms of not only inflation but conversion to current-year dollars the first year and interest income--thus it gets pretty convoluted. I think the main thing is that in the first year we are converting the amounts to current-year dollars--we are setting up the numbers to be denominated in 2016 dollars and then that doesn't need to happen again in future years. And this is what you want so I would not try to work around that.

Appreciate the feedback. I do understand the objective. And by the way, I don't mean take the nominal rate of return to zero (necessarily), but net it down by inflation. In other words, if your nominal rate of return was 5% and inflation 3%, change inflation to 0% and nominal rate of return to 2%. So 0% only if nominal rate of return = inflation rate.

I guess my point is that by netting down the beginning balance you end up with a year 1 ending balance that doesn't 'work' from an accounting perspective; as of 'today', the 12/31/2015 balances are 'current dollars'. I guess what I'm saying is that I'd prefer to see current year dollars based on 12/31/2015 rather than 12/31/2016.

As mentioned in the OP, using the methodology as is results in a unrealistic Year 1 ending balance, which is highly problematic in terms of tracking activity during the year towards the 'goal' of achieving the end of Year 1 balance.

This may be somewhat unique to me. As a CPA/CFO part of my appreciation of the program lies in it's 'tying out' on an accounting basis. In fact, I've added to the standard spreadsheet a tab where I pull together all the information and put it into a format I'm more familiar with; if not for that exercise I likely wouldn't have noticed the issue, other than having a sense that the Year 1 ending balance didn't 'feel' right.

I have gone back and zeroed out the inflation factor and reduced my nominal rate of return by a like amount. As one may expect, the impact on the overall plan and components of the plan were minimal, but it now 'ties out' and provides a reasonable Year 1 ending balance that I can use as a goal.

Thanks again for the feedback. We may have a minor philosophical difference but my workaround satisfies me (thankfully I don't get fancy with inflation rates or nominal rates of return!).

Nick,
I don't think "the methodology as is results in a unrealistic Year 1 ending balance"
It results in the correct Year 1 ending balance!

Say your 31 Dec 2015 ending balance is $100,000 and your nominal rate of return is 5%.
Then your 2016 end of year balance is $105,000 if you have neither deposits nor withdrawals.

This is the correct balance and your "workaround" is incorrect.

Thanks,
Mike

Hi all -

I've been mulling over this the past few days as I initially had the same concern as Nick59. The asset balance for the first year seemed to be too high.

Dan, while this does arise due to inflation, I don't think it's actually for the reason you gave in your explanation. Converting 2015 dollars to 2016 dollars wouldn't be done by multiplying by (1 + inflation rate), it would be done by DIVIDING by (1 + inflation rate). 2016 dollars are worth less than 2015 dollars.

I believe that what ESPlanner is doing is giving a year-end asset balance that does NOT account for inflation. In other words, it's giving the year-end 2016 balance in 2016 dollars. This is actually awesome, and I think is what Nick59 was looking for too: a number in "real" dollars to check balances against at year-end.

You can confirm that this is what's happening by running a simple scenario. Put $100,000 in real or retirement assets with 3% inflation and a nominal return of 7% (= 4% real return). The Total Income column always shows the real return, adjusted for inflation. In all years except the first, the regular asset balance = prior year's asset balance + Total Income - Total Spending - Taxes + Savings. However, in the first, year, even though the Total Income column shows the real return, the asset balance actually reflects the NOMINAL return.

Now that I think I understand this, it actually makes sense. If you are looking at your actual balances, you will see the effects of the nominal return, because that's the actual income/increase in valuation that you're receiving.

If ESPlanner actually showed a report like the Regular Assets report in "real" dollars (as opposed to today's dollars), every number in the Regular Assets column would be higher than it is in the report now, because it would be showing amounts in each year's dollar, not today's dollar. It appears to me that ESPlanner actually shows the first year this way, which is why the number appears to be higher than expected, at least when comparing to how the balances are calculated for the other years. I don't think any particular workaround is needed though. This is giving the correct balance in current year dollars for the first year.

As mentioned, now that I understand it, this is exactly what I'd want the program to do, because I now have a real number to check balances against on Dec 31. After rolling forward to the next year, I'll update balances, all "memory" of 2016 inflation will be gone from the program and the whole cycle will start again.

Let me know if you think I've gone astray somewhere :-)

dan royer's picture

I think you are better at explaining it than I am. It was always explained to me that the UI of the program asks for end-of-previous-year dollars and that those need to be translated or denominated in current year dollars, which causes the first-year unexpected gain. It seems clear to me if I look at the regular assets report and place in the previous year amount in the row above the current year.

I like your explanation.