Roth conversions are nothing new; the ability to convert to a Roth has existed since 1998, the year that Roth IRAs came into being. The conversion rules were liberalized somewhat in 2005; however, the opportunity to initiate a Roth conversion has proved elusive to most clients. That's because current rules prohibit households with more than $100,000 in modified adjusted gross income (MAGI) from converting a traditional IRA to a Roth IRA. Come January 1, 2010, however, the rules will change. The income cap will be lifted. As a result, an estimated $3 trillion sitting in traditional IRA accounts will become eligible for conversion.

Clearly, the lifting of the income cap will create an opportunity for many clients and prospects. It will also create an outstanding business opportunity for financial advisors. That's because decisions such as whether to convert, how much to convert, when to convert and the timing of tax payments are more complex than they first appear. Furthermore, the popular press is already alerting the public to the Roth conversion opportunity, so the demand for advice will be strong. Financial firms both large and small will fuel demand even more with their marketing.

As is often the case with complex financial decisions, there are some general rules of thumb that will be put out as guidance for those thinking about conversion, but in reality, each client's situation is unique. The only way to arrive at a satisfactory decision is to run the numbers. Unfortunately, this is not going to be as easy as it sounds. Many do-it-yourself investors will get incomplete answers because they turn to free online calculators. Some advisors may fall into a similar trap if they fail to perform their due diligence before selecting a software application to aid them in their Roth conversion analysis.

Before we discuss any specific firms or products, it may be useful to discuss a framework that readers can apply when choosing their conversion software. The discussion must begin with three key factors (just as it did with Monte Carlo software last month): the quality of the model, the quality of the inputs and the quality of the explanation. Let's look at each in turn.

An evaluation of the quality of the model involves a great many things. Clearly, a program should get the rules right. It should also let you opt to pay the tax in 2010, or you should be allowed to pay the tax over two years-half in 2011 and half in 2012.

Any decent tool should also give advisors the option of paying the taxes for the conversion out of the traditional IRA or from another source. Ideally, you want the option to model the growth of the funds used to pay for the conversion; that is, you present what would happen if the conversion did not occur and you instead took the funds earmarked to pay the income tax for the conversion and put them in a taxable investment account. Most comprehensive financial planning software will do this, but many stand-alone programs will not.

Inputs are also important. Good default options can speed data entry, lowering your costs of producing a report. If you do not like the program's defaults, you should be allowed to alter them, but some programs require you to override the defaults each time you run a plan, if they allow it at all. We indicated earlier that you have the option in 2010 of delaying payment of taxes till 2011 and 2012. If you want to model this scenario, look at how your software handles this task. Can you simply check a box to choose this method, or are additional steps required?

If you choose to postpone the taxes until 2011 and 2012, it is possible that you will pay higher taxes on the conversion, since many provisions of the Bush tax cuts are set to expire. The software should illustrate this; it should also allow you to take into account what would happen if the sunset is postponed for some, or all, tax brackets.

If you want to illustrate a scenario where you keep the money in a traditional IRA and invest the funds that would have been used to pay the conversion taxes, the assumptions you make regarding those investments will impact your findings. For example, if you assume that the money will be invested in an asset that is taxed at the client's marginal rate, and if you assume the money will be spent beginning at age 65 or 70, the results will be different than if you assume the funds are invested in a broad market ETF or Berkshire Hathaway stock that is highly unlikely to generate any tax until sold. Furthermore, if the intent is to pass the asset to the next generation, the (hopefully) appreciated asset will get a step-up in basis, reducing or eliminating the income tax impact on the heirs. Some programs allow you to easily alter scenarios for these funds; others do not.

No matter how good the software is, the insights provided by the advisor are crucial. For example, if your client is in a top marginal tax bracket, and if you believe that marginal rates will be higher in the future, you should certainly illustrate what a Roth conversion would look like under that scenario, but it is imperative to inform the client that other outcomes are possible.