True estate planning is beyond the scope of this program, but it is noteworthy that Silver includes an "estate" input page. This page serves primarily as a checklist to document whether the client has wills, trusts, powers of attorney, etc. It also documents whether the client lives in a community property state. Finally, it captures some other information the program requires to perform calculations such as intended charitable bequests, credit shelter exclusions that have already been used, etc. Life and LTC information is captured next.
As you would expect, all income, pensions and lump-sum distributions are captured. When dealing with Social Security, the application can automatically calculate a benefit, or you can enter your own estimates. On the expense side of the ledger, you can either just enter annual estimates, or you can drill down using the worksheet provided. For uneven inflows and outflows, such as payments expected in the future, or intermittent expenses (like a new car every four years), there are separate input screens.
Once the income and expenses are in place, the advisor enters assets, asset allocations, and rate assumptions. The program allows the user to specify one rate assumption for each type of asset before and after retirement. For example, you can assume that the client will earn an average rate of 8% on taxable assets before retirement and 7% after. You can also enter average tax rates for pre- and post-retirement, as well as a single assumption about the cost basis for taxable assets and another single assumption for annuity assets. Experienced advisors may want to control the standard deviation assumptions used in the program's Monte Carlo simulation, but most users will allow the application to supply that number.
In my tests, navigation and data entry were simple. A simple case might take as little as 15 minutes to enter, while a more detailed one could take up to 45 minutes. Once all of the data is in place, analysis can begin.
"Silver's 'what-if' screens are fantastic!" according to Levin. "I can create many scenarios quickly." At the top of the screen is a simple graphic analysis determining whether clients will run out of money before death given their projected life expectancies. The graph assumes there will be constant average rates for returns, taxes, inflation, etc. Below the graph is a Monte Carlo probability result-the odds of success if returns are randomized.
In the first scenario I ran for an imaginary couple, both the straight-line calculation and the Monte Carlo simulation indicated that they had little chance of meeting their goal of retiring at ages 68 and 67 with $84,000 in income and $54,000 left to the survivor, given their life expectancies of ages 88 and 92. On the lower half of the page, however, it is easy to change the assumptions and generate almost instantaneous results. When an alternative scenario delayed retirement to ages 70 and 69 and delayed Social Security payments until the same ages, the straight-line calculation indicated success, but the Monte Carlo simulation indicated a success rate of only 46%. I then lowered the income need to $80,000, which boosted the success rate to 60%. Finally, we pushed back retirement one more year, and the Monte Carlo success rate went up to 75%. All of this was accomplished in a couple of minutes.
The latest version of Silver builds on the success of its "what-if" scenario generator with something new. The folks at MoneyTree call it "dynamic behavioral analysis." In a sense, this is an extension of the thought process behind Monte Carlo simulations. As most readers know, Monte Carlo gained in popularity because market returns vary over time, and the order of returns can have a huge impact on retirement income projections. Monte Carlo simulations illustrate this concept to clients.
By the same token, behavioral factors can also play a role in a retirement income distribution plan. While most retirees have fixed expenses, they also have discretionary ones. After retirement, if a retiree experiences suboptimal returns on their portfolios, it is reasonable to assume that some discretionary spending will be scaled back. Behavioral dynamic analysis can take this behavior into account by creating rules that say if a distribution exceeds a predetermined percentage of the total portfolio, that year's distribution will be scaled back by a percentage not to exceed, for example, 10%. Obviously, by scaling back spending in lean years, and incorporating this behavior into the analysis, the life of a portfolio can be extended.
Furthermore, many clients have discretion over when they will retire. Rather than run a Monte Carlo simulation where the client supplies a retirement age (say 65) and then solve for the portfolio termination date, doesn't it make more sense to set a termination date, a sustainable rate of return and the asset base necessary to fund such a plan, and then solve for the projected retirement date using Monte Carlo? So, for example, you could say that under a given set of circumstances, if you retire at age 65 your odds of success are 40%, but if you wait until age 68 they improve to 60%. Silver allows you to do this.
Overall, there is a lot to like about Silver. It is fast, intuitive and convenient. The output, while it makes some compromises for the sake of expediency, is meaningful. According to the feedback I've received from a number of MoneyTree clients, both customer service and technical support are well above average.
While it is difficult to judge exactly what the impact of dynamic behavioral analysis will be at this point, it appears that by adding additional variables to the accepted models and by offering different ways of examining the data, MoneyTree Silver adds new perspective and increased flexibility to the income distribution process. That is certainly a step in the right direction.