Cognitive economics is not yet a widely used phrase, though Italian economist Marco Novarese and I have been using it as a name for a more microfounded version of what’s typically called behavioral economics.  So I’ll explain how I use the term.

The first difference is between “economics” and “finance.” The more fundamental distinction is between “cognitive” and “behavioral.” Cognitive is about how we think, while behavioral is about what we do.  

Economics, very broadly, is the study of how resources are allocated (by individuals, and across society). Cognitive economics, on the other hand, looks at what is actually going on within an individual’s mind when he or she makes that choice.

A cognitive economist asks, for instance: What is the internal structure of their decision-making?  What are the influences on it? How does information enter the mind and how is it processed? What form do preferences take internally, and how are all those processes expressed in our behavior?

Finance, by contrast, specifically focuses on investment and the value of financial instruments. Behavioral finance focuses on the phenomena of how people behave. For example, what will they do (on average) when faced with a given choice between two ways of paying for something?

I would also distinguish between the way that behavioral finance is practiced, and what cognitive economists do. Behavioral finance is an experiment-driven field. BF people mostly start from the framework of classical economics and do experiments to find out where real behavior differs from the classical assumptions of rationality.

BF is quite practical in one sense.  It helps us imagine the ways that people might behave when confronted with a given situation. However it does not make good predictions about how they will behave. Generally it will rely on experiments to distinguish among the different possible behavior modes.

Cognitive economists start at a lower level, from a microfounded model of how people make decisions, and work upwards theoretically, to develop a self-consistent model of large-scale economic behavior. It should ultimately be able to explain or predict behavior from a minimal set of base data.

In some ways this is the same goal as classical economics, but cognitive economics offers a richer and more accurate microfoundation leading to more powerful and better micro and macroeconomic predictions.

The difference between the two disciplines is like the difference between engineering and physics. Behavioral finance is like engineering: Engineers know some of the rules about how objects behave, and they can use those rules to design and test new implementations of existing inventions, and fix things that have already been built.

Cognitive economics is like physics: Physicists know the underlying theory of how things work, and they can use that to explain how existing inventions operate, and to work out how to create new ones.

Leigh Caldwell is a London-based mathematician who specializes in behavioral economics and the economics of information. He is chief executive of Inon. He also heads Intellectual Business, a new think tank, and writes the blog, Knowing and Making.

© 2010 RIJ Publishing LLC. All rights reserved.

After reading in Retirement Income Journal about the recent government  hearings on so-called 401(k) annuities, it amazes me how often the proverbial myths of “loss of control” and no “liquidity” continue to be spewed about the use of annuitization. The reality is that annuitization, properly positioned, gives retirees more liquidity and more control over their money.

Those who favor guaranteed lifetime withdrawal benefits (GLWBs) on deferred variable or fixed indexed annuities typically focus on the “life only” feature of income annuities. But that is rarely the most optimal or appropriate annuitization option. By using Period Certain, Life or Joint Life with Period Certain, or Installment Refund payouts, retirees can often create income solutions that require even less investment than GLWBs to produce the same level of income.

Those options are the foundation of what I call the “Annuitize Part and Invest the Difference” story.

Many of the single-premium immediate (SPIA) and deferred income (DIA) annuity contracts that my clients purchase allow “living commuted value” to the annuity owner. This means that the annuity owner has the option of withdrawing cash from the annuity if he or she decides, at some future date, that liquidity is more important than preserving the income stream the client originally purchased.

The liquidity options of these living commuted withdrawal options usually have some limitations or restrictions. But the same can be said of deferred annuities with GLWBs, which may have surrender periods or roll-ups that are cancelled by withdrawals.

Since the income annuity may cost less (to generate the same amount of income) than a deferred annuity with a GLWB, the remainder of the portfolio isn’t hostage to those restrictions. It can be used for additional liquidity or potential long-term growth without harming the future income stream. The longer the client can let the unfettered asset to grow, the more liquidity they can provide.

At the same time, thanks to the Period Certain feature, much of the future income from the contract annuity comes back either to the owner or the heirs. When the Period Certain is combined with a Life or Joint Life option, the contract owner earns “mortality credits” that bring income up to the same level as the income stream from a GLWB.

The argument for the deferred variable (or fixed indexed) annuity with a GLWB is that the entire sum is liquid. This is a Catch-22. I call this type of thinking “psychonomics,” not economics, because the liquidity comes at a severe price. The anticipated level of income shrinks if too much of the money under the income rider is withdrawn.

Under the typical GLWB, if a retiree withdraws more assets in a single year than the contract permits, it diminishes the benefits. The “Annuitize and Invest the Difference” option, on the other hand, creates liquidity that does not cannibalize the income or consume the portfolio—as long as the client’s combination of income annuities and at-risk investments are properly positioned relative to the client’s age.

© 2010 RIJ Publishing LLC. All rights reserved.

Executives from Mutual of Omaha didn’t participate in the Department of Labor hearings on “in-plan” or 401(k) annuities in Washington last week. That’s not because they weren’t interested, but because they’ve been too busy implementing the kind of plan that the hearings envisioned.     

Starting last March, Mutual of Omaha Retirement Services quietly began offering some of its 5,000 institutional clients, most of whom are sponsors of small and “micro” retirement plans, a voluntary deferred income annuity option called the Lifetime Guaranteed Income Account (LGIA).

So far, says Tim Bormann, Mutual of Omaha’s retirement plans product line director, about 80% of the new plan sponsors who’ve been offered the option have agreed to make it available to plan participants. United of Omaha Life underwrites the product.  

“We’re addressing many of the concerns voiced by plan sponsors following the recent market downturn,” Bormann said in a published statement last spring. “We’re removing the ambiguity from retirement savings and reducing risk. With the Lifetime Guaranteed Income Account, a participant’s retirement income will be known, shown and guaranteed.”

The program bears certain similarities to SponsorMatch, the program launched by MetLife in 2009 that combines a deferred income annuity, funded by the employer match, with a participant-directed investment account, funded by the employee’s contribution. And it’s not unlike The Hartford’s Personal Retirement Manager “two-cylinder” deferred variable annuity.

By encouraging the purchase of future income credits, and reporting account balances as both as a lump sum and as a rate of retirement income, programs like MetLife’s or Mutual of Omaha’s accustoms a participant to regarding his or her 401(k) as a source of retirement income rather than as an investment account.

By locking in future income, such a program also maximizes the time value of money and spreads interest rate risk over many years or even decades. From a behavioral standpoint, it will presumably help diminish a participant’s natural resistance to parting with a large lump sum all at once at retirement.

One outside observer thinks the plan reflects progress for the entire field.

“As I understand the product, since the units are purchased on a per paycheck basis, the participants are also essentially dollar-cost-averaging interest rates, and, to a lesser extent, mortality,” said Jay DeVivo, founder of String Financial LLC, a provider of 401(k) participant advice and managed accounts.

“Once this generation of products becomes established and consumers become better educated on concepts such as mortality credits, I’m hopeful that products like ALDAs (Advanced Life Deferred Annuities) that are less flexible but have greater income producing leverage can flourish.”

How the LGIA works

As Bormann explained to RIJ, the process works like this: the 50 investment options under a Mutual of Omaha institutional plan include a Guaranteed Account and the LGIA. Contributions to either account earn a fixed rate that floats from month to month, but every contribution or matching contribution that’s directed to the LGIA—by employer or employee—purchases an increment of future guaranteed income and earns 50 basis points less (as payment for the guarantee).

In 2010, the rate on the Guaranteed Account has ranged from 1.5% to 2.5%. The value of future income is calculated based on an assumed interest rate of 4% and a retirement age of 65. Regarding gender issues, a unisex mortality table is used and the annuity purchases do not trigger Qualified Joint and Survivor Annuity requirements. 

“The LGIA actually has two components. There’s the account balance that’s growing at the credited interest rate, and there’s a second guarantee that says, ‘If you put in, say, $5,000 a year, I can tell you exactly how much income you’ll get when you’re 65,’” Bormann said. “The underlying investments in the two accounts are identical. The difference is that the credited interest rate is 50 basis points lower on the guaranteed side.”

The program allows plan sponsors to offer something tantamount to a defined benefit plan. “We’ve had plan sponsors ask, ‘Why don’t I just function like a pension plan, since you’re basically providing a guaranteed amount.’ One strategy we think will be attractive is the concept of offering the LGIA for employer contributions and let them offer a quasi-defined benefit plan,” Bormann said.

In designing the program, Mutual of Omaha anticipated the usual objections to in-plan annuities. For instance, participants can change their minds and move money from the LGIA to another investment option within the plan and back again, with a minimum of 60 days between round trips. In the meantime, however, they lose the guarantee—and the time that their future income could have been growing. 

Contributions to the LGIA go into the general account at United of Omaha Life, but the insurer maintains adequate liquidity to allow transfers out of the account. (The cost of maintaining the liquidity is factored into the rate credited to the GA and LGIA accounts.)

At retirement, participants can take a lump sum or buy an income annuity, on whatever payout terms that United of Omaha offers. They can also roll their lump sum into an IRA and purchase an income annuity at The Hueler Companies’ Income Solutions platform. (Mutual of Omaha, along with about a dozen other insurers, distributes its income annuities through the Hueler’s competitive bidding platform.)

If a participant leaves the plan before retirement—as, presumably, many will do—they can preserve their income guarantees when they leave by rolling their LGIA into an IRA. If the plan sponsor decides to change providers and discontinue its relationship with Mutual of Omaha, the LGIA can stay with the plan as a “frozen option,” Bormann said, meaning that it no longer accepts contributions.

As for the typical plan sponsor’s greatest anxiety—that the in-plan annuity provider they choose will someday go out of business, fail to fulfill its obligations and trigger a raft of lawsuits and claims against the sponsor—Mutual of Omaha relies its long history, its survival of the financial crisis without significant damange, and the high level of participant control built into its program, to reassure potential sponsors.

Expressing today’s savings as tomorrow’s income

One of the public policy initiatives discussed at last week’s Department of Labor and Department of Treasury hearings on in-plan options was the introduction of 401(k) account statements that display each participant’s accumulated savings as both a lump sum and a future monthly income.

Mutual of Omaha’s program does that. On their quarterly statements, and at each plan’s website, participants can see their accumulated savings expressed both as a lump sum and as the monthly income stream that they’ve already purchased through the LGIA.

During the hearings in Washington, several plan providers expressed uncertainty about how to express future income—whether to base the figure on the amount already saved or to create projections and estimates based on assumptions about future contributions and rates of return. Some asked the Department of Labor to write guidelines for expressing future income in a fiduciary or compliant manner.

There is some concern that younger participants might be discouraged by the relatively small amounts of monthly income that their past contributions have purchased. Mutual of Omaha has chosen the more conservative route and provided only income that has been purchased, while offering participants an online calculator to test a variety of strategies and assumptions.  

“One of the failings of the [retirement] industry is that we’ve focused on the account balance, rather than on how much income the participant will get at retirement,” Bormann said. “We’re trying to get them focused away from the account balance.”  

Jay DeVivo agrees. “I think it is important to reorient participants away from a ‘magic number’ they should have accumulated by retirement and toward retirement income, particularly guaranteed retirement income,” he said.

“I think the risk of outliving one’s assets has been so broadly reported and discussed, along with the demise of the pension, that it is no longer an industry talking point, but a risk widely recognized by individual investors. Since the vast majority of retirement savings is done through an employer-sponsored plan—upwards of 80% of IRA assets originate from a DC plan as well—in-plan products and guarantees are going to become the rule rather than the exception.”

© 2010 RIJ Publishing LLC. All rights reserved.

The prominence of the witnesses testified to the significance of the DoL/Treasury Department hearings this week on so-called “401k annuities” and other tools that can help plan participants turn their savings into income–either while they save, when they reach retirement, or in retirement.

Delegates from the retirement industry’s A-list made statements, including (in addition to those mentioned in today’s cover story) JP Morgan, TIAA-CREF, Lincoln Financial Group, Great-West Life, Dow Chemical, Honeywell, Hueler Companies, Vanguard, Nataxis Global Asset Management, Financial Engines, and consultants Towers Watson, Milliman, and Hewitt.

Many of these companies have introduced in-plan income products already. TIAA-CREF pioneered the group variable annuity in the 1950s. Vanguard and Hueler announced an agreement last week to give Vanguard IRA owners access to Hueler’s online SPIA supermarket, which provides institutionally priced income annuities to participants in 1,200 plan.

Prudential Retirement and Great-West offer stand-alone guaranteed lifetime income riders to participants with target-date funds. Financial Engines expects to offer an in-plan option in late 2010 or early 2011. Asset managers Vanguard, Fidelity, Putnam and Russell all sell payout mutual funds, which deliver predictable but non-guaranteed income streams.

Trade, labor, and consumer advocacy groups also contributed their thoughts. The AFL-CIO, the Investment Company Institute, the Profit-Sharing and 401(k) Council of America, the American Council of Life Insurers, the American Society of Pension Professionals and Actuaries and the Spark Institute all gave testimony. Other groups and firms can contribute statements via e-mail over the next month.

Some witnesses praised in-plan annuities, while others, like Steve Utkus of Vanguard, tried to bury them. To be sure, there are plenty of good reasons for putting income options in plans, either as savings vehicles (deferred annuities) or exit options (immediate annuities, payout funds).

Big plan sponsors have economies of scale and bargaining power that lower the costs of products, administration, and education. Plan sponsors are also in a unique position to set up a program that applies the employer match to the purchase of future income—a path that MetLife and Mutual of Omaha are pursuing. Like contributions to a defined benefit plan, that kind of program would leverage the time value of money and mitigate the interest rate risk and timing risk associated with the lump sum purchase of an annuity at retirement.    

Counter-arguments

But there are plenty of many counter-arguments for putting annuities in DC plans. For plan sponsors, there are potentially huge expenses associated with evaluating the costs and benefits of different income product vendors and in educating employees. Above, all liability for picking the wrong provider or for giving employers bad (in retrospect) advice scares them. “Fiduciaries are paranoid and rightly so,” says Sheldon Smith, an ERISA attorney and president of ASPPA.

As for participants, most of them won’t retire from their current employer/plan sponsor. From the participant perspective, the average person spends only 4.2 years in any particular job and might participate in several 401(k) plans. There are also portability issues. Participants may want to change employers or get out of income products. Employers may want to change annuity providers.

Adapting recordkeeping systems to multiple income options, or options that might change suddenly, could also pose problems, especially for small employers. Ninety-percent of plans have fewer than 100 participants. Only the largest 10% of plans, which account for 85% of all participants, may be able to cope with the legal, educational, and recordkeeping challenges of in-plan options.   

Other problems: most 401(k) accounts, even at retirement, are too small to annuitize at all, let alone big enough to allow for the ideal solution: partial annuitization. Annuity purchases can also trigger the need for spousal approval, potentially increasing paperwork for sponsors. Gender-neutral pricing rules in 401(k) plans also mean that retail annuities, which have gender-specific pricing, can offer men much higher payout rates than in-plan annuities. 

Alternate vision

The zeal for putting guaranteed income options in DC plans is limited mainly to insurers. Asset managers like Vanguard and Fidelity, which are the custodians of millions of rollover IRAs, believe that most people will consolidate their tax-deferred savings in an IRAs and then buy an annuity—or simply take systematic withdrawals. Investment advisors think along the same lines, and millions of Americans are likely to take this path. 

Even if you build in-plan annuities, will they come? There’s a lot of disagreement over whether Americans want in-plan annuity options. Even in DB plans, 90% of retirees who have the option choose lump sum payouts over lifetime income streams. Shlomo Benartzi of UCLA, and an Allianz Life consultant, cited evidence of high annuitization rates in some companies, and MetLife said participants like a partial annuitization option. But the evidence is inconsistent. Most people don’t want to give up control over their assets or even part of their assets, especially not at time they stop working, when their retirement plans are still unsettled.

There’s also the crowding-out problem. Thanks to Social Security, most middle-class plan participants will get at least half of their retirement income coming from an annuity, and have no compelling reason to annuitize their DC savings. On the contrary, they may need their DC assets to stay liquid for emergencies, bequests, weddings or simply long-deferred pleasures.

To overcome this resistance or inertia, some witnesses said, the government might have to approve a qualified default annuity option, analogous to auto-enrollment and the qualified default investment options in 401(k) plans. One witness, Josh Shapiro of the National Coordinating Committee for Multiemployer Plans, asserted that nothing short of mandatory annuitization of DC assets at retirement will really change the game. The DoL and Treasury, as well as sponsors and plan providers, have already ruled that out. 

© 2010 RIJ Publishing LLC. All rights reserved.

Of the many gnarly reasons why so few 401(k) participants have access to ‘in-plan’ lifetime income solutions, the biggest obstacle may be the plan sponsor’s fear of lawsuits from participants if the sponsor’s chosen annuity provider ever went broke.  

“The fiduciary issue is so daunting that it ends the conversation,” said David Wray, the director of the Profit Sharing and 401(k) Council of America, a trade group for plan sponsors. “Sponsors ask, ‘We’re guaranteeing payments for 30 or 35 years?’ It stops the conversation. It’s a wall.”

It’s a wall that executives at many insurers and other financial services firms would love to see breached or demolished. That’s why so many of them told a Department of Labor panel Tuesday that the DoL should identify a process or path—a ‘safe harbor’—by which they could choose an annuity provider without liability.

“The government has to stand behind us when we stand behind a decision,” Wray said. “There has to be confidence that someone will make sure the annuity payments keep coming, and it’s not the employer.” Lest the panel miss his point, he added, “If one of these programs goes down, the political fallout will be significant.”

“Fiduciaries are paranoid and rightly so,” says Sheldon Smith, an ERISA attorney and president of ASPPA.

A safe harbor already exists, but plan sponsors and others say it leaves many questions unanswered. Is the fiduciary duty the same for all types of income products, guaranteed and non-guaranteed? How often would the company have to check the strength of the provider? Can companies change their minds and switch providers without exposure to liability?  

Insurers say that their strength ratings are available for anyone to see. But, in the wake of the financial crisis, sponsors aren’t so sure they can rely on the ratings agencies or the oversight of state insurance commissioners.

Aside from a clearer fiduciary standard, insurers, asset managers and others asked the Departments of Labor and Treasury for the following:

Provide safe harbor for advice about retirement income.  Plan sponsors want to see the rules on investment advice broadened, so that sponsors can educate participants about retirement income options without future liability for poor participant choices, and so that sponsors can use plan assets to pay for the education.

Encourage partial annuitization. One of the big obstacles, real or perceived, to the use of income annuities is that participants can’t annuitize part of their plan assets or don’t know that they can choose to annuitize only a portion of their assets. Witnesses asked the government to work to encourage partial annuitization.

Provide guidance on expressing savings as retirement income in participant statements and websites. Many witnesses argued that showing plan participants how much retirement income their plan assets will buy, instead of just showing them their current account balance, would motivate people to save more and help them appreciate that their account’s purpose is to provide retirement income, not wealth accumulation. They requested DoL guidelines for providing such illustrations in a compliant way. Since so many variables go into making projections about retirement income, that won’t be easy, however.  

© 2010 RIJ Publishing LLC. All rights reserved.

The sections of the tax code that cover tax-deferred and tax-exempt savings programs are a mess—well intentioned perhaps, but a mess. Stop somebody on the street and ask him to explain the tax treatment of Roth IRA withdrawals. You might as well ask directions to Maracaibo.

People in high places have been thinking about cleaning up this mess. A thick new whitepaper called “The Report on Tax Reform Options: Simplification, Compliance, and Corporate Taxation,” includes at least eight ways to rationalize the rules governing the hodge-podge of programs that we know as IRAs, FSAs, HSAs, 401(k)s, 403(b)s, 529s, SIMPLE-IRAs, SARSEPs, etc.

The report contains both good and bad news. The good news is that lots of interesting alternatives to the status quo exist—like making contribution limits the same for on IRAs and 401(k)s, or letting half the people skip their Required Minimum Distributions.

The bad news is that any change the tax law will inevitably gore somebody’s ox. What one person sees as the elimination of a wasteful and ineffective tax incentive will strike other people as a confiscatory tax hike. And no matter what happens, unintended consequences are likely follow. 

But something apparently needs to be done. According to the report, the government “spends” $118 billion a year on tax-deferred retirement, health care and education accounts, but the wrong people are using them. Of the $118 billion in uncollected taxes, 84% stays with people who make $100,000 or more a year. That’s not the kind of social engineering this administration is aiming for.

Here are the eight suggestions:

1. Consolidate Retirement Accounts and Harmonize Statutory Requirements. There’s currently a hodge-podge of related but distinct tax-deferred or tax-exempt savings programs in the tax code, from 401(k)s and 403(b)s to 529s, from IRAs to FSAs and HSAs, and from SIMPLE plans to SARSEPs. They have different rules regarding eligibility, contribution limits, and withdrawals.
2. Integrate IRA and 401(k)-type Contribution Limits and Disallow Nondeductible Contributions. The advisory board suggested raising the limits on deductible contributions to IRAs (currently $5,000 a year) to the same level as the limit on 401(k) plans. This would eliminate a lot of paperwork for people who currently make IRA contributions that exceed the $5,000 limit. On the other hand, it might reduce tax receipts and would disproportionately favor high-earners. 
3. Restrict use of IRA assets for non-retirement purposes. One of the redundancies of the current system is that people can use the money in their IRAs for certain medical or educational expenses instead of retirement.  The advisory board recommends imposing strict limits on the use of funds in retirement saving accounts for non-retirement purposes. Taxpayers would instead be encouraged to use 529 Plans for college savings and HSAs or FSAs for medical bills.
4. Clarify and Improve Saving Incentives. The report proposes an expansion of the existing Saver’s Credit to a match from the government as an incentive to low-income workers to save. The existing credit offers a tax deduction, which doesn’t help low-income workers. The report also proposes an “automatic IRA” rule, which would require employers in business for at least two years and with more than ten employees to default their employees into an IRA and start deducting contributions from their payroll. Employees could drop out of the program if they wished. 
5. Reduce Retirement Account Leakage. Under current rules, it’s relatively easy for people to squander their tax-deferred savings by spending them when they change jobs or by borrowing against them and not re-paying the loans before they change jobs. The report suggested restricting access to 401(k) and IRA money, but acknowledged that restrictions would create paperwork for employers and hardship for some employees.  
6. Relax non-discrimination rules for small plans. Many owners of small companies are discouraged from offering their workforce a 401(k) plan because the rules limit their ability to discriminate in favor of their most-valued employees by contributing more for them than for rank-and-file workers.  One solution would be to replace existing non-discrimination rules and apply the rules associated with SIMPLE 401(k) plans.
7. Eliminate RMDs for small accounts. Eliminating the distribution requirement for those with tax-deferred savings under $50,000 relieve would eliminate the burden of calculating RMDs for half of retirement account owners over age 70½, while exempting only 6% of retirement assets from taxation each year. Today, retirees over age 70½ must multiply their tax-deferred assets by an age-specific factor to determine the amount of savings they must remove from their tax-deferred account(s) and pay taxes on. For people with several tax-favored accounts, the calculation and the decision regarding which account to tap for the distribution can be complicated. One caveat: People with small accounts are the ones most likely to be in circumstances that force them to take distributions each year, whether required or not.    
8. End the ‘Curse of the MAGI.’ Today, retirees must fill out an 18-line worksheet to calculate the percentage of their benefits is subject to income tax. Much of the complexity comes from having to determine whether zero, 50% or 85% of benefits should be included in modified adjusted taxable income (MAGI). The panel suggested a return to pre-1993 system of taxing Social Security benefits and other income separately.

Conservatives might bridle at the suggestion that people in higher tax brackets shouldn’t reap most of the benefits of tax deferral or exemption, or that a tax break is a “cost” and that a failure to collect taxes is equivalent to an expense for the American people. As we’ve seen in the debate over extending the Bush tax cuts, they regard the closure of a tax break as a tax hike.

But there’s a Democrat in the White House, and his administration holds the liberal viewpoint that tax breaks should either pay for themselves by helping to achieve a public policy goal—in this case, by encouraging people who make under $100,000 to save more—or be eliminated. It would probably be politically impossible to eliminate popular tax breaks; a judicious pruning or tweaking might be the only option.    

© 2010 RIJ Publishing LLC. All rights reserved.