America faces a retirement crisis that is perhaps even bigger than the "47 million uninsured" crisis we face in healthcare. Based on figures from the Center for Retirement Research at Boston College, more than 60% of working-age households are at risk of not being able to maintain their standard of living in retirement. This article will focus on financial products individuals can purchase to reduce the risk of outliving their assets, including products being offered today and products that are under development. My views are somewhat unconventional, particularly with regard to current products.

I believe that the most popular retirement products do very little to reduce the risks associated with retirement. One product that can provide valuable protection against the risk of outliving assets, the income annuity, has never become popular. Right now, products being developed in academic and research institutions have the potential to do a much more effective job than the products available today. The challenge will be getting these new products built and effectively distributed.

Retirement Planning Today

Most retirement planning being done today involves planners and clients working together to assess the adequacy of savings to meet retirement needs. Planners gather client data on assets, liabilities, future sources of income and expected future expenses. They may also have the client do a risk-tolerance assessment to provide input for asset allocation recommendations.

Using these inputs, the planner will likely use tools like financial projection software, an asset mix optimizer and Monte Carlo analysis in order to create a recommended sustainable withdrawal strategy for the client. One such strategy might be that the client withdraws 4% to 4 1/2% of assets in the first year and increases withdrawals each year, based on inflation. An alternative might be that the client takes a higher percentage initially, but then varies each year's withdrawals depending on investment performance.

These approaches can work well for a client who has built up sufficient assets (or has sufficient pensions or other income) that a withdrawal rate of 4% to 4 1/2% can provide adequate cash flow. It preserves investment flexibility and liquidity, and, if the client dies early, it maximizes bequests. However, for clients of more modest means, the 4% to 4 1/2% may not be enough to sustain their pre-retirement standard of living. Raising the withdrawal percentage increases the risk that assets will not be adequate for the full span of retirement. Among the risks are: living longer than expected, incurring significant health—or long-term-care costs and experiencing poor investment performance, particularly in the early years after retirement. Such clients may be able to increase retirement cash flow and guard against the risk of outliving assets by purchasing some kind of insurance, in the form of an annuity or other financial product. Purchasing such products gives up some liquidity and flexibility, along with spending funds that would otherwise be available for bequests. However, for many, it may be a necessary tradeoff.

Popular Products, No Protection

Target-date mutual funds and deferred annuities are two products that are popular with retirees. Despite their popularity, though, these products don't manage the risk of outliving assets. Target-date mutual funds are basically balanced funds that automatically shift away from equities and toward fixed income as the client ages. Some are set up to provide for regular withdrawals, and some even adjust the amount of withdrawals based on performance of the underlying funds. To the extent that such funds help clients keep asset allocations on track, they do provide some benefit.

Another popular product is the deferred annuity—available both in fixed and variable (equity-linked) versions. This product can be used to build tax-deferred savings, and it allows the purchaser to "annuitize"—that is, turn the savings into a lifetime stream of income. Sales of deferred annuities totaled an impressive $244 billion in 2007.

However, the best measure of risk reduction is not the level of sales, but the amount of assets annuitized. Based on LIMRA data, annuitized assets have averaged about $10 billion per year since the year 2000. If we think of a "steady-state" or "input-output" model, the level of annuitizations (output) should roughly equal the deferred annuity sales (input). Given the huge disparity between $10 billion and $244 billion, there are a lot of deferred annuities that never get annuitized. These annuities end up functioning like tax-deferred, high-expense mutual funds, and they do not help reduce the risk of outliving assets.

Living Benefits: the Right Direction?

Some changes in deferred annuity products during the past few years may point them more in the direction of protecting purchasers from outliving assets. These changes involve the introduction of living benefits. First launched in the late 1990s, these benefits have gone through some transformations-the latest is called the guaranteed lifetime withdrawal benefit, or GLWB.

This feature guarantees that the annuity purchaser can withdraw up to a set percentage of the original annuity deposit each year and continue such withdrawals for life. Percentages range from 4% to 7%, depending on the purchaser's age at the first withdrawal, with higher percentages at older ages. Annual charges run 50 to 75 basis points of account value. The product may also allow for increases in the "benefit base" above the original deposit, based on investment performance of the annuity.

The big advantage of the GLWB is that it provides an income that cannot be outlived and preserves the purchaser's control over assets. A disadvantage is that it does not benefit from the mortality pooling effect of annuitization, so the guaranteed income amounts will be less than amounts available through annuitization. (I'll provide a numeric example of the mortality pooling effect in the next section.) Another disadvantage is that although the product may be designed to increase benefits based on performance of the underlying annuity funds, there is no guarantee of automatic inflation adjustments. A 65-year-old's benefit of 5% without inflation adjustments may be equivalent to only 3.5% to 4% with inflation adjustments.

A further disadvantage is that the GLWB can be difficult to build into a financial plan because the actual payments can vary greatly depending on the underlying funds' performance. Finally, purchasing a GLWB involves purchasing a variable annuity, and these products have been subject to criticism (legitimate, in my view) because of their high expense charges and tax inefficiencies.

A lot has been written about the pluses and minuses of this particular annuity feature. In 2007, for example, Ibbotson Associates performed a rigorous examination of the impact of adding an annuity with GLWB to a mutual fund portfolio. (It should be noted that the Ibbotson study has generated considerable controversy, so it's also worth reading Bob Veres' critique as well as Ibbotson's rebuttal and the discussion postings at www.financial-planning.com.) My view is that the GLWB offers a step in the right direction, but there are existing and potential products capable of doing a better job of meeting retirement needs.

Income Annuities

The income annuity, also known as an immediate or payout annuity, is designed to turn savings into an income that cannot be outlived. It can be described in the following simple terms:

• A client gives an insurance company a significant lump sum of money.
• The insurance company pays the client an income that lasts for life.
  

The most popular version is one from which the client receives a level monthly income. Variations include annuities whose payments increase annually by a set percentage or whose annual increases are based on inflation. Another is the variable immediate annuity, whose payments adjust annually based on the performance of underlying mutual funds. Income annuities may be single life or joint life—typically issued to married couples—whose payments continue until both members of a couple pass away.

Income annuities work on the mortality pooling principle: Those who die early subsidize those who live longer than expected. Because of the risk pool, annuities can provide a higher retirement income than a retiree might achieve by following a conservative program of regular withdrawals from savings. Based on Vanguard annuity rates as of August 2008, a 65-year-old female would pay $493,315 for an income annuity paying $40,000 per year (8.1% of the purchase price) for life. If this same individual were to invest at a fixed rate of 4% and base her withdrawals on making the money last 30 years, she could only take yearly withdrawals of $27,431 (5.4% of the investment amount).

Of course, she could achieve higher withdrawals by investing some portion of her funds in equities, but that would entail taking on more investment risk than with the fixed annuity. A workable solution might be to invest a significant portion of assets in an income annuity with fixed payments and then weight the remaining non-annuitized assets heavily toward equities.

Given the advantages offered by income annuities, plus other considerations like the shift away from defined-benefit pensions and worries about the future of Social Security, we might expect income annuities to be very popular. That's not the case. In 2007, income annuity sales totaled just $7.1 billion. Some rough estimates based on demographic data indicate that the potential market is between $200 billion and $300 billion.

So owners of deferred annuities are not annuitizing, and individuals with regular savings are not buying income annuities. This naturally raises the question: Why are income annuity sales falling so far short of potential? This question is important not only with regard to income annuities, but also for potential new products. Will those new products be successful, or will they, too, fall short?

Reasons for Low Sales

How do we explain the failure of income annuities? Here are some reasons based on my own thinking, reading and experience.

Aversion to large, irreversible transactions. People strongly resist making large financial commitments. It's just a fact of life. And these commitments are big. If, for example, an individual aged 65 were considering purchasing an annuity, he or she might have to invest more than $500,000 to provide a useful level of income. That's a pretty big "bite-the-bullet" amount—like buying a house and making a lifetime commitment to live there.

Underwhelming sales compensation. Commission-oriented salespeople, who represent the vast majority of those who offer financial products, do not like income annuities—one upfront commission and that's it. This explains at least part of the disparity between the $7 billion of income annuity sales and the $244 billion for deferred annuities.

Complexity of the sale. The planner needs to be able to make a comprehensive recommendation, including the appropriate age for the client to purchase the product, the features to choose, the amount of annuity to purchase, how much to leave in regular savings and the appropriate asset allocation for the money left in savings. The planner has to factor in such considerations as the value of the client's savings and pensions, plus the client's survival outlook and desire to leave bequests. For a planner who has been dealing with retirement planning by recommending mixes of mutual funds, it's a big step to consider products like income annuities and address all the appropriate planning issues.

Clients' aversion to facing reality. For some individuals, even looking at annuity pricing can bring home the sad reality that their savings are woefully inadequate. Such individuals may be temporarily happier living with the vague hope that somehow their savings will earn enough to support their retirement.

Anti-selection pricing. Income annuity products are priced for healthy people. Insurance companies are forced to price the products this way, because buyers tend to self-select based on their subjective longevity prospects. For the client in average health, an income annuity may seem like an expensive way to protect against longevity risk.

Considering this list, I do not find the low level of sales surprising. As for possible ways to address these issues, perhaps the aversion to large, irreversible transactions could be mitigated somewhat by spreading the annuity purchase over time, either before or after retirement. The sales compensation issue might be dealt with by increasing commissions or adding trail commissions, but the danger is that higher commissions may do too much damage to product performance. The complexity issue will require the development of new planning techniques and associated software, but we're unlikely to see progress along these lines at current low levels of sales. Overall, I am not optimistic about the income annuity, at least as it is designed and sold today.

Potentially Revolutionary New Products

Perhaps we can summon up some hope by looking at potential new products. The good news is that there is a lot of work going on in research institutions and universities aimed at developing products that bear the potential to revolutionize retirement planning. The challenge will be to turn these innovative ideas into real-world products that will actually succeed in the marketplace. In this section I'll discuss research work by individuals like Moshe Milevsky, associate professor of finance at York University in Toronto, and economist Mark Warshawsky, who heads retirement research for consulting firm Watson Wyatt.

The product ideas include longevity insurance, an income annuity that begins paying after a significant deferral period (e.g., 20 or 30 years); a life-care annuity, which pays a significant pop-up benefit if the purchaser needs long-term care (based on criteria like those used in long-term-care insurance policies); and a ruin-contingent life annuity-a product, similar to longevity insurance, that pays based on two contingencies: longevity and weak investment performance. Because there are two contingencies, the cost is much less than that of an income annuity or longevity insurance. In addition, new funding alternatives include spreading purchases over time (either pre—or post-retirement) and accessing additional sources of funding such as home equity.

Longevity Insurance

Longevity insurance is a new product being offered by a few insurance companies, but it has yet to produce significant sales. Longevity insurance is basically an income annuity with a long deferral period, typically 20 to 30 years, before payments begin. For example, a 65-year-old could purchase longevity insurance that would begin paying an income at age 85. This income would last for life. The person could then focus on managing his or her other assets with a goal of making them last to age 85 and not have to worry about longevity risk beyond age 85.

A key advantage of longevity insurance over a standard income annuity is price. For example, if a 65-year-old wanted to purchase a lifetime annuity that paid $40,000 in the first year, increasing at 2.5% per year thereafter, the cost would be about $600,000. If the same individual wanted to purchase longevity insurance based on the same payment stream, with payments of $65,500 beginning at age 85 ($40,000 increased at 2.5% annually for 20 years), the cost would be only $95,000. Some vendors allow individuals to purchase the product over time, which can help to further overcome aversion to large lump-sum transactions.

Perhaps the biggest obstacle to making this product a success is that it is so different. Retirement planning, including the use of annuity products, has always focused on savings. This is the first attempt to use an insurance product to deal with longevity.

Life-Care Annuity

This enhanced version of the income annuity is not yet being offered by insurance companies, although the idea for building it has been around for more than a decade. The simplest version is a standard income annuity with a pop-up benefit that pays extra income if the purchaser's health deteriorates and he or she then satisfies standard long-term-care insurance claim criteria—for example, the loss of two or more activities of daily living. The fact that this product is not yet available reflects the unfortunate reality that the tax and regulatory environment can present significant hurdles to the introduction of innovative products. Hopefully these issues can be overcome, because there are a number of reasons why this product makes a lot of sense.

First, as mentioned earlier, income annuity pricing tends to be based on the life expectancy of healthy individuals, so adding the long-term-care benefit can make the product attractive to individuals whose health is more average and allow insurance companies to offer more attractive annuity pricing. Second, this product can reduce individuals' reluctance to tie up savings in an income annuity owing to concerns that they may need extra funds to cover long-term care expenses. Third, by combining an income annuity with a long-term-care product, insurers will have less need to underwrite long-term-care policies and reject individuals who may have some health problems. This issue becomes more important at older ages.

An initial concern I had about life-care annuities was that the cost of adding the long-term-care benefit might be so prodigious that it would seriously compromise the value of the basic annuity. When I did some product modeling, however, I was pleasantly surprised to find that adding the long-term-care benefit only modestly increased the cost.

Here's an example. For a 65-year-old female, the cost of an annuity that pays $40,000 per year, inflating at 2.5%, comes to $611,000. The cost of adding a long-term-care pop-up that pays an additional $40,000 per year (with the same 2.5% annual inflation) is only $22,000. So adding what would seem to be a very appealing additional benefit adds only about 3.5% to the cost. In the marketplace, the cost may turn out to be a bit higher because of higher loads in the long-term-care piece or building higher inflation into the long-term-care benefit. But it looks like it's certainly feasible to build an attractive combination product if the tax and regulatory issues can be overcome.

Ruin-Contingent Life Annuity (RCLA)

The RCLA is another product that has made it to the drawing board but is not yet available. The brainchild of Moshe Milevsky, the RCLA is a variant on longevity insurance—but two contingencies generate payments. The first contingency is longevity. The second contingency is weak investment performance, particularly in the critical years immediately following retirement. (Phoenix and Lockwood Capital have come out with a similar product, but it's not available as a standalone—it's tied in with Lockwood funds—and doesn't offer inflation protection.)

An RCLA is more complex than other retirement products, so I'll explain its workings with an example. Let's assume a 65-year-old female retiree has $1 million in savings. She would like to set up a withdrawal plan to spend $40,000 in the first year after retirement and increase spending each year based on inflation. She could simply invest the funds and draw money out of savings. But if she did some work with a financial planner, including financial projections and Monte Carlo analysis, she would come to realize that while 4% is, indeed, a conservative rate, she could run into trouble if she lives to age 90 or more, or if she's unlucky enough to endure a bear market in the early years of withdrawals. This is where the RCLA comes to the rescue.

The RCLA is a derivative product based on a specific investment index with an inflation-adjusted withdrawal rate built in. An index appropriate for a new retiree might be something like a 50/50 stock/bond weighting. In this example, based on a 4% withdrawal rate, the index would track a hypothetical fund that performs like a 50/50 stock/bond mix with inflation-adjusted withdrawals based on 4% of the original fund. The RCLA product would make no payments initially, but if the index were to hit zero, the product would make 4% payments, adjusted for inflation, for the remaining years of the purchaser's life.

In this example, the individual would match her own portfolio allocation to the index and purchase enough insurance to match the size of the portfolio. Indications are that the lump-sum cost for covering this $1 million portfolio might be on the order of $20,000 to $40,000, significantly less than the $95,000 needed to purchase longevity insurance that begins paying at age 85. The reason the cost is less is that there are two contingencies that need to be satisfied before payments are made. But these are exactly the contingencies the client is concerned about, so there is an excellent match of needs with benefits. Think of the possibility of providing retirement insurance at a price of less than $40,000; we've come a long way from asking the client to buy a $600,000 annuity.

It's worth noting that the RCLA is quite similar to a standalone version of the GLWB rider. Like the GLWB, the RCLA allows the purchaser to preserve control of assets and liquidity. Also, as with the GLWB, the RCLA does not generate as much of a mortality pooling effect as annuitization. A possible enhancement for the RCLA would be to add the long-term-care pop-up feature. It would not be necessary for the hypothetical fund to reach zero before paying out long-term-care benefits.

Can We Get There from Here?

Several issues must be addressed in order to make these new financial products a success. The challenge is daunting. Here's what we must overcome:

Raising awareness. Most planners and the general public are not aware of the work going on in academic and research institutions to design new products for retirement planning.

Taxes and regulation. Even good products can face hurdles posed by the tax and regulatory environment. The irony is that the most innovative products tend to face the highest hurdles.

Software and techniques. It won't be enough to get insurance firms to build and offer the types of new products described here. An even bigger challenge will be developing software and planning techniques, so that planners and financial salespeople can integrate these products into an overall system of retirement planning. Introducing these new products will lead to a much more complex planning environment than the one we work in today.

Training. Following the need for new software and techniques will be the need to train planners and financial salespeople, so they can begin to offer clients these new approaches.

Compensation. The income annuity is an example of a product failure that is a least partly related to compensation. In order to get financial salespeople to promote these new products, the challenge will be to find ways to provide compensation that makes the products attractive to them, but does not carve out so much in compensation that the products become unattractive to consumers. For fee-based planners, it will be helpful if insurance companies also build low-load and no-load versions of these products.

Potential abuses. Unfortunately the introduction of new financial products often attracts dishonest salespeople trying to make a quick buck. As new products are offered, it will be important that sales-practice management keeps pace.

Insurance company risk concerns. Insurance companies may be averse to introducing new products that involve their taking on investment risk, regardless of the potential to satisfy client needs. Insurers need to meet quarterly earnings targets, and their balance sheets are subject to close scrutiny by analysts and rating agencies. They may be reluctant to introduce products that, while profitable in the long run, may produce short-term fluctuations.

Preretirement funding. Most of these new products will be more palatable if they can be bought over a number of years preceding retirement. To overcome the aversion to large irreversible transactions, it would make sense to spread purchases over the 20-year period preceding retirement.That way, a $100,000 purchase at retirement could be transformed into slightly more than $3,000 per year. This will require a different approach than today's practice of waiting until near retirement to get into retirement planning.

Finally there's the issue of how to deliver new retirement products to the public. Should there be a government role? Could these types of products be put inside 401(k) plans as investment options? There are a host of questions. However, given the complexity of the new products, there will be a growing need for purchasers to receive top-quality planning advice. There's bound to be an important role for planners regardless of how the products are delivered.

So when we look to the future of retirement planning, we see tremendous potential. Financial planners need to move things along. They must organize a task force to address the challenges and promote the success of these new products. This task force would need to focus more on implementation than on ideas. There are plenty of ideas; it's the implementation that needs to catch up. With the right people involved, such a group could tremendously improve retirement prospects for millions of Americans.

Joe Tomlinson, FSA, CFP, is an actuary and financial planner based in Greenville, Maine, who devotes a significant amount of his time to researching and writing on financial planning topics. He was previously a vice president at John Hancock Financial Services, where his primary focus was on investment and retirement products.