Estate Model Limitations

Return to Goldplan

Here are some of the pitfalls that should be considered when using GOLDPLAN® , or any computer model, for that matter.  If you encounter or can think of others please let me know.

  • RATES OF RETURN - computer models and calculators usually assume some rate of return over many years.  But you should be aware that 6% every year is not the same as 20% in one year, -14% another, 5% another, etc., even if the average rate of return turns out to be 6%. This is particularly true if a portfolio is an owner's source of current income.

    Fortunately, GOLDPLAN    ® gives you the option of illustrating a portfolio's performance using various historical indexes as well as a fixed return.  

    Testing the performance of investment portfolios using an index is a good idea, given that it is not always intuitively obvious how variable returns of a specific portfolio might affect the overall estate.

 

  • INTEREST RATE FLUCTUATIONS - the market value of assets in an estate continually fluctuate with market rates of interest.  Bonds, for example can decline substantially in a year when interest rates rise.  The GOLDPLAN® model does not attempt to make any assumptions about how market rates of interest might change from year to year or how the valuation of assets might be affected by such changes.

 

  • TAX RATES - every year IRS income tax tables change.  The model essentially assumes that the current tax structure is the same for all years. We do, however, update the tables each year and if the IRS publishes expected future tables we use those as well for the years that apply.

 

  • ASSET TURNOVER - we found it inappropriate to try to accurately account for the effects associated with the rate of turnover in investment portfolios. But keep in mind that a portfolio with high turnover may produce substantially lower real returns due to the associated taxes, fees and expenses that turnover can generate.

 

  • INCOMPLETE TAX ANALYSIS - the model cannot make every possible critical calculation  For example, you can compare how much your estate might be worth years later if you invest your retirement dollars in a traditional IRA vs.a Roth IRA.  But remember that in the intervening years, how you earn, spend, and invest your money will affect the final result.  And the model will allow you to compare "apples and apples" by illustrating how the overall estate might look after assuming the liquidation of all retirement accounts at some point in time.  But the model can not illustrate all the possible options available for taking retirement distributions, nor can it calculate how much those assets will be worth in the hands of beneficiaries after beneficiaries pay income tax.  Furthermore, the model does not make complex generation skipping tax calculations.

 

  • IRD (Income in Respect to a Decedent) - the model does not automatically attempt to calculate IRD, however, a worksheet is included that will allow you to calculate IRD and its impact on beneficiaries with respect to IRA or 401(k) plans for a single individual, e.g., a surviving spouse.

 

  • STATE TAXATION - GOLDPLAN® does not calculate state death taxes for those states that charge more than the federal credit for state death taxes. However, if such taxes are calculated outside the model, they can be inserted manually.

 

  • THE CHARACTER OF EXPENSES - the model illustrates that if your expenses today are too high, it can mean that future goals will be sacrificed. But the model cannot predict what the benefit of certain expenses might be.  Some expenses might be better characterized as investments which could actually improve the prospect for future economic success.  Examples might be spending money on educational experiences, buying a new wardrobe to enhance one's self esteem, taking a scenic cruise to Aruba and maybe meeting a wealthy widow or widower on board.

 

  • AN UNCERTAIN FUTURE - every assumption the model makes becomes more uncertain the farther out into the future you look.

These limitations do not invalidate the model, but only illustrate that its purpose is to augment your own cognitive thinking and to help point out the implications of possible unconsidered economic consequences or relationships. Not even a fancy model can predict the future. But they're useful anyway.

Your questions and comments are welcome.
Feedback



Copyright © 2001-2011 Mark A. Goldman

.


No matter how carefully a computer model is constructed, it's the knowledge and  integrity of the person using it that will determine whether or not it's truly useful.