If there is one thing that most knowledgeable financial advisors agree on it is that building sustainable retirement income portfolios is more challenging than building accumulation portfolios. With accumulation portfolios, time is on your side, and you can compensate for a few years of subpar returns in the future. Once clients retire, however, a few years of bad returns can seriously hurt their ability to withdraw from their diversified portfolios.

Ever since Bill Bengen's seminal article, "Determining Withdrawal Rates Using Historical Data," appeared in the October 1994 issue of the Journal of Financial Planning, advisors have been struggling to come up with retirement income plans that allow clients to live comfortably without the risk of outliving their nest egg. Bengen's initial work focused on determining the dollar amount that clients could safely withdraw annually without outliving their portfolio, as well as the asset allocation of the portfolio. This research was the basis of the widely quoted "4% rule." This "rule" is really just a guideline for determining sustainable withdrawal rates for a given client. Bengen and others have since that time built a lot on the initial research.

To this day, many advisors and their clients approach retirement income planning in one of two ways: They either begin the process with a rule-of-thumb "safe withdrawal rate," still using 4% as the standard. Or they start out with a predetermined "need," such as $80,000 per year adjusted for inflation annually, and then select a portfolio with the expected return sufficient to provide the necessary income (taking into account Social Security and other income sources) at the appropriate level of risk. This risk is commonly defined as the standard deviation of the portfolio.

Manish Malhotra and his fledgling firm Fiducioso Advisors are trying to introduce, through his Income Discovery software, now in beta, a retirement income planning framework that is superior to the methods that have come before. The software itself is relatively easy to master, and the output provides information that clients should readily understand.

Although the Income Discovery venture is new, Malhotra is no novice when it comes to wealth management or financial software. He has spent the last 12 years working for financial firms in various roles ranging from corporate banking to IT architecture. Before creating Income Discovery, he was a senior vice president at Citigroup Global Wealth Management (Smith Barney and Private Bank) leading the IT strategy and architecture function. His team was responsible for building the IT architecture of an integrated desktop for financial advisors and private bankers, a multiyear initiative. He led technology evaluation of major financial planning software while at Citigroup. Before that, he led the IT architecture for enterprise risk systems at Bank of America. He started his career as corporate banker with a leading private financial services firm in India before moving into software at Dresdner Kleinwort's subsidiary in India. He is an engineer-MBA from top schools in India (the Indian Institute of Technology and XLRI).

Income Discovery's calculation engine is powerful, but at its core, what it does is illustrate the interplay between four factors: the level of sustainable income, plan failure rates, the potential life of a portfolio in a "bad case" scenario and the average terminal value of the portfolio. Some comprehensive financial programs can do some, if not all, of what Income Discovery does, but most can't do it as effortlessly, and most can't illustrate on one screen the interplay of the four factors as well as their effect on the income portfolio.

The other interesting aspect of the program is its ability to model the trade-offs inherent in selecting one of three retirement income strategies: a systematic withdrawal plan from a traditional diversified portfolio of stocks, bonds and cash; a joint and survivor annuity (the default choice is one that pays 100% of the benefit when both spouses are alive and then pays 75% of the benefit after the death of the first spouse); and a maturity matched portfolio (MMP).

The concept of a maturity-matched portfolio is similar to that of a bond ladder; however, in a bond ladder, you roll over the principal. With an MMP, you use the interest (if any) and the principal to supply a required cash flow for a stated period of time. For example, if you wanted to provide income for the next five years annually, you could buy zero-coupon bonds with maturities of one, two, three, four and five years. Of course, you could build this MMP in other ways, but this is a simple example. The point is that at the end of the five years, all principal and interest from the portfolio would be depleted.

Matched-maturity portfolios are particularly helpful in mitigating the risks associated with the sequence of returns during retirement. The impact of the sequence of returns can be critical in retirement because if the worst returns come during the first two years of retirement, the damage to an income plan can be severe. In a typical portfolio of stocks, bonds and cash, assuming spending is constant, the client may be forced to liquidate some assets at low prices in order to fund the cash flow needs, resulting in accelerated portfolio depletion.

MMPs help ensure against the risk of a bad sequence of returns by ensuring the required cash flow for the desired period of time. Admittedly, MMPs are not the only method of providing the needed cash flow. You could, for example, hold the money required to fund the cash flow in cash; however, with very short-term rates near zero, MMPs may be preferable for periods of over a year or two.

Whether you ultimately decide to use MMPs or not, the application's ability to illustrate a portfolio composed partially of MMPs is appealing. Even more appealing is the ability to illustrate the "cost," in annual cash-flow terms, of extending the MMP period on the fly. An alternative scenario that extends the MMP period from five to seven years, for example, can be run with a mouse click. Almost instantaneously, the client will see the impact on the asset mix and the cash flow.

Although clients may not be aware of the risks of a bad sequence of returns unless an advisor brings it to their attention, most are acutely aware of the risk at the other end of the spectrum: the risk that the portfolio will expire before the client does. Income Discovery allows the advisor to easily add longevity insurance to the mix in the form of a longevity annuity, that is, an annuity that does not begin paying benefits until some specified time in the future. If you leave the software unconstrained, it will solve for what it calculates to be the optimal percentage of assets, if any, that should be devoted to a longevity annuity. Of course, many clients will resist putting too much into such a product, so if they are not comfortable with the recommendation, you can constrain the permissible amount devoted to this product.

Speaking of annuities, Income Discovery allows for more traditional annuities as well. Currently, the advisor has to price the products and enter them manually, but Income Discovery plans to add feeds from two annuity clearinghouses so that advisors can select individual annuities and get pricing in real time.

Now that we've discussed a few of the things that Income Discovery can do, let's take a step back and briefly summarize how it works. First, you create a household. Only minimal information is required to get started. This includes names, Social Security benefits and start dates, plus total assets and any pension benefits. Next, you go to the inputs screen (it is unclear to me why this is a separate page). Here, you choose the scenarios for the simulation (historical returns or generated returns; Income Discovery provides these numbers, as well as model asset allocations, the standard deviation of returns and the cross correlations, but you are free to change them). You then enter the desired income, the time horizon in years, a maximum amount to be annuitized, if any, and the details of up to three annuity policies to be analyzed. Or, if you prefer, you can run the software without entering the desired income amount and let the program calculate what the client can safely withdraw.

You then push the "perform analysis" button. The program generates three initial "plans." Each plan includes the annual income amount, the median terminal value of the portfolio, the failure rate, the number of years the portfolio will last in a "bad case" (defined as the worst 2nd percentile of portfolio life across all scenario paths). It then displays the amount to be invested in the systematic withdrawal portfolio, the MMP and each annuity being considered.

Let's say that all of the resulting three plans show a portfolio failure rate of at least 15%, but the client insists on a computed failure rate of no more than 10%. With the click of a mouse, the advisor can constrain the failure rate to 10% using the provided drop-down menu. The program will then recalculate based on the constraint. Either the monthly income will be less, the mix will change, or both, but the client can immediately see the trade-off between the two factors. Perhaps the recommended mix for this portfolio is 20% in the diversified portfolio and 80% in an immediate annuity linked to the CPI. The client may object to allocating such a large portion of total assets to an annuity, so you constrain the model to a maximum of 40% annuities. This diverts more money to the diversified portfolio of equities and fixed income, but it lowers the monthly payment by about 10%. This demonstrates the cost of limiting annuities to the client. Armed with this vital information, they can then decide which plan they prefer.

The inputs page includes a few additional options. One allows you to illustrate a phased retirement, so if a client wants $7,000 the first ten years of retirement followed by $6,000 the next five years and $5,900 for the remainder of the plan, you can illustrate that. It also allows you to plan by income category: basic, desired and extra. The point here is to set a baseline amount of income that is absolutely necessary and then add layers of income.

I tried out a beta version, but the production version should be available by the time you read this. Since I used a beta version, it would be premature to rate the program; features are still being added. Overall though, I liked the clean, simple interface. I liked the fact that it is built on the Google framework; this ensures that it will work with both PCs and Macs. It is also browser-neutral. I had no problem running it on Chrome, Firefox or Safari. I also tried running it on a Samsung Galaxy tablet (with the Android OS) and the iPad using their default browsers. It was usable in both cases, but since there are no native apps for the tablets, and since the site is not yet optimized for tablets, the pages were not always centered on the screen.

Since Income Discovery will often be used collaboratively with clients, the layout of the pages could be made more pleasing. The ability to output to PDF so you can save whatever you've provided to the client would be nice. Context-sensitive help and a user guide are also needed. I assume that Income Discovery is working on some of these issues.

Malhotra has indicated that the initial release is only the beginning. The firm is in negotiations with one or more firms to provide MMP execution for advisors who want it. Soon, advisors will be able to include MMP in the plan without having to worry about implementation. They can outsource it to a third party that specializes in building them.

Right now, the application does not differentiate between the clients' various tax buckets (taxable, tax-deferred and Roth), so it can't do tax-sensitive optimization or stipulate the order in which the various accounts are depleted. This capability will be added in the future.

Since Income Discovery is new and a little rough around the edges, why should you buy now? Well, right now, Income Discovery is currently being offered at no charge for an unspecified period of time. Before the free period expires, Income Discovery will offer all early adopters an opportunity to subscribe at 60% off the list price ($100 versus the list price of $250 per month) on an annual subscription basis.

Whether you ultimately decide on a paid subscription or not, Income Discovery is worth checking out. It allows you to get right to the heart of income distribution planning with minimal data input, and it clearly illustrates, in real time, the impact that changing one factor has on the other elements of a plan. You can sign up at www.incomediscovery.com