[Retire Early]
ESPlanner and Consumption Smoothing.

ESPlanner and Consumption Smoothing.


Return to Table of Contents - Home Page
.............

This article was first posted July 1, 2006.


ESPlanner is a financial planning program developed by economists Laurence J. Kotlikoff, (Chairman of the Dept. of Economics at Boston University) and Dr. Jagadeesh Gokhale (a Senior Fellow at the Cato Institution.) It uses the life-cycle model of saving and consumption for which Franco Modigliani won the Nobel Prize in Economics in 1985. Consumption smoothing is the observation that individuals seek to avoid abrupt changes in their standard of living. ESPlanner uses this concept of consumption smoothing to make recommendations on the amount of retirement savings and life insurance needed to maintain a given standard of living throughout an individual's lifetime.

The ESPlanner software for individuals comes in two versions. The standard ESPlanner for $149 and an ESPlanner Plus version that includes a Monte Carlo simulator for $199. The original purchase price includes one-year of software updates. After the first year, it costs $50/year for updates and support.

The program has a detailed 83 page User's Manual that you can review before purchase. There's quite a bit of financial information that the user must input (see page 12 of the User's Manual for a listing), but it's needed for the extensive calculations the program runs through.

ESPlanner provides the user with a great deal of flexibility to model significant financial events such as starting a family, paying for college, or retiring early. The Monte Carlo simulator also allows you to change your asset allocation each year if you want to reflect a less aggressive asset allocation as you get older.

To get some simple, baseline results for the program, I input the case of a single individual with a $1 million retirement nest egg split evenly between an IRA and a taxable account. To further simplify things, I assumed that the person received no Social Security or pension benefits and planned to leave no estate

Static Analysis

The standard version of ESPlanner doesn't take asset allocation into account. It bases the consumption smoothing calculation on the investment return and inflation assumptions input by the user. (The default values are a 6% investment return and 3% inflation.) The table below shows the results for a variety of investment return and inflation assumptions.



Annual Retirement Withdrawal.
50-year-old with 50-year life expectancy to age 100.
$1 million portfolio, no Social Security or pension. Estate = $0
.----- Annual Investment Return -----
. 4% 5% 6% 7% 8%
2% inflation $29,906 $36,172 $42,680 $46,962 $49,685
3% $24,254 $29,758 $35,945 $42,397 $46,749
4% $19,352 $24,106 $29,578 $35,656 $42,035
5% $15,211 $19,233 $23,945 $29,376 $35,292
Annual Retirement Withdrawal.
60-year-old with 40-year life expectancy to age 100.
$1 million portfolio, no Social Security or pension. Estate = $0
.----- Annual Investment Return -----
. 4% 5% 6% 7% 8%
2% inflation $33,852 $39,420 $45,147 $51,187 $57,606
3% $28,506 $33,613 $38,979 $44,582 $50,504
4% $23,753 $28,316 $33,290 $38,494 $43,961
5% $19,598 $23,598 $28,103 $32,907 $37,958
ESPlanner
Annual Retirement Withdrawal.

70-year-old with 30-year life expectancy to age 100.
$1 million portfolio, no Social Security or pension. Estate = $0
.----- Annual Investment Return -----
. 4% 5% 6% 7% 8%
2% inflation $40,837 $45,781 $50,840 $56,094 $61,574
3% $35,799 $40,479 $45,251 $50,187 $55,329
4% $31,147 $35,555 $40,052 $44.686 $49,503
5% $26,961 $30,947 $35,224 $39,583 $44,091

Monte Carlo Simulation

The "Plus" version of the program includes a Monte Carlo simulator with historical data on 12 different asset classes.

ESPlannerPlus Monte Carlo Simulator
available asset classes
Fixed Income Equity
Long-term corporate bonds Large cap stocks
Long-term gov't bonds Small cap stocks
Long-term TIPS Non-US Equity
intermediate-term gov't bonds Pacific basin equity
Short-term gov't bonds European equity
- Emerging mkts equity
- REITs

To test the Monte Carlo simulator, three cases were run for our retiree with a $1 million nest egg evenly split between an IRA and a taxable account.The results are summarized in the table below.

The "Recommended Trajectory" for the retiree's Living Standard of 5% to 7% (e.g. $51,226/yr to $71,580/yr from a $1 million portfolio in Case #1 and #2) is a good bit higher than the "safe withdrawal rate" in the 4% to 5% range supported by other calculators. However, the Monte Carlo analysis does warn of a significant risk that a standard of living that high won't survive to age 100. Reducing the Recommended Trajectory for Case #1 (60-year-old) by 25% from $51,226/yr. to $38,420/yr. reduces the risk of failure at age 100 from 38% to 15%.

A 100% Fixed Income (FI) portfolio was also included for comparison. As expected, it supported a lower withdrawal rate than the portfolios with large equity allocations.

ESPlannerPlus Monte Carlo Simulator
Recommended Trajectory/Prob. of Failure at age 100

Age at death = 100, $1 million portfolio,
no Social Security or pension. Estate = $0
Retirement
Age
Case #1
75% S&P 500
Case #2
75% MPT
Case #3
100% FI
50 $47,773 / 41% failed $64,080 / 16% failed $39,228 / 39% failed
60 $51,226 / 38% failed $66,786 / 19% failed $40,701 / 41% failed
70 $57,338 / 41% failed $71,580 / 28% failed $46,580 / 41% failed
Notes:

Case #1 -- 75% S&P500 Portfolio
Both IRA & Taxable account
5% cash, 25% mid-term gov't bonds, 75% S&P500 (Large cap stocks)

Case #2 -- 75% MPT Portfolio
IRA = 5% cash 20% TIPS, 20% LC, 20% SC, 10% Pacific, 10% Europe, 10% REITs, 5% Emerging Mkts
Taxable = 5% cash 20% mid-term gov't bonds, 20% LC, 20% SC, 10% Pacific, 10% Europe, 10% REITs, 5% Emerging Mkts

Cash #3 -- 100% Fixed Income (FI) Portfolio
IRA = 5% cash 95% TIPS
Taxable = 10% cash, 45% mid-term govt, 45% short-term gov't

Probability of Failure at age 100 = percentage of the 500 Monte Carlo trials that failed to sustain the Recommended Trajectory at age 100

For users accustomed to personal financial planning programs like Intuit's Quicken or Microsoft Money, the ESPlanner user interface isn't as polished and user friendly. The program does what it says it does, but it takes a while to work through the very detailed and somewhat cumbersome data entry process. Overall, the program's ability to model changing financial circumstances over a lifetime along with detailed Social Security benefit and income tax analysis probably outweighs the data entry burden and relatively high purchase price.



filename = esplanner.html
Copyright 2006 John P. Greaney, All rights reserved.


[Retire Early]