How to Make Money Last in Retirement

We're living longer. And even if we've saved for our retirement, will we have enough? When should we file for Social Security benefits? And how much should we be taking out of our retirement portfolios each year? A recent panel at the Morningstar Individual Investment Conference moderated by Adam Zoll tackled these questions and more. The panelists included Morningstar director of personal finance Christine Benz, financial planner Mark Balasa of Balasa Dinverno Foltz, and retirement specialist Mark Miller. Here's an edited excerpt of the conversation.

Adam Zoll: Mark, when you have a client who comes to you and says, "I'm thinking about retiring; I'm not sure if I have enough," how does longevity enter into the discussion?

Mark Balasa: Let's say you're 65; your life expectancy is usually in your early 80s. Fidelity and J.P. Morgan have done some nice work showing that a married couple, if they are 65 at the moment, has a 90% chance of one of them making it to 80, a 70% chance of making it to 85, and a 50% chance of making it to 90.

So, we try to help them better understand that it may not be the case for you individually but, as a couple, someone is going to be here probably longer. We need to be really conservative when we are running projections--should it be around 85, 90, or 95? We help to encourage them to think about it in those terms.

Christine Benz: While you can use some of the tables that are out there as a starting point for estimating your life expectancy, you also certainly want to factor in your own health history, familial health history, your parents' longevity, and your grandparents' longevity. If your family has a history of longevity, you also want to think about the role of cognitive decline because we also tend to see a very strong correlation between longevity and a higher prevalence of cognitive decline. An NIH study found that roughly 25% of people in their 80s were experiencing some sort of cognitive decline or dementia, and that number jumps even higher if you are in your 90s.

The combination of that greater longevity and potentially the need for long-term care would argue for a larger pool of money that you would need to set aside for your own retirement.

• Read more: Concerned About Longevity: 4 Mistakes to Avoid

Zoll: A lot of people talk about working longer into their retirement years. Mark, you have written about employment prospects for older workers. How realistic is that expectation?

Mark Miller: It's a great aspiration, and it's a good idea to work even a few additional years. Even working an extra three to five years is very salutary for your retirement plan in terms of additional years of contributing to retirement accounts, fewer years of drawing down from that money in large sums, and then the opportunity to delay your Social Security claiming. Together, those things can have a very dramatic impact on your retirement plan.

There is also, however, an acknowledgment that there is a lot of age discrimination in the workforce. The numbers also tell us that roughly half of people retire earlier than they expected because of a health problem or because they are providing care for somebody else, job loss, or job burnout. So, I think working longer is a great strategy, and it is a great thing to try to do. It is not a plan, though.

Benz: T. Rowe Price did a really valuable study several years ago pointing to the value of working longer. For people who can do it, as Mark said, it's very impactful. But they had an idea where they found that making additional retirement-plan contributions, say, post-age sixty, you tend not to get a lot of tax benefit from doing that.

The idea was, if you can stick it out, continue working, maybe stop making those additional contributions--assuming your retirement plan is in semi-good shape--start spending some of that money, and try to enjoy maybe a little bit of what retirement has to offer while you are still earning that paycheck.

I think that there are many happy mediums that people can consider in addition to maybe sticking it out in your main career. Another idea is perhaps some sort of part-time career that you find more enjoyable.

Miller: That's a big trend. It doesn't necessarily mean sticking with the thing you've been doing for 30 years. People are transitioning into interesting new roles, new jobs, entrepreneurs, doing consulting work, mixing it up, as you were saying.

• Read more: Need to Work in Retirement? Think Small

Zoll: For many people, Social Security will be the only guaranteed lifetime income source. Can you provide some sort of basic rules of thumb or guidelines in terms of how to best treat that benefit?

Miller: At a high level, I think delayed claiming is something most people should think about--especially married couples--because, as has just been pointed out, one person in the couple is likely to well exceed the longevity numbers, and that's where Social Security provides a huge benefit. Oftentimes, you have situations where people get into their late 80s or 90s, and they've exhausted their savings, and then Social Security is still there as a bedrock. Delayed claiming is something that people ought to think about. Full retirement age now is 66. It's heading up to 67, and that's the fulcrum against which the claiming runs.

Then, you can get into some of these more complex strategies that are becoming more popular, such as "file and suspend," where for couples that reach their full retirement age, the higher earner can file but then suspend the benefit. The other lower-earning spouse can then file for a spousal benefit, while the higher earner waits to claim later, perhaps as late as 70. That points to the other broad strategy. In terms of delayed claiming, it's most important for the higher earner to delay as long as possible because, in a married couple, that's where you get the biggest bang for your buck.

Zoll: How does Social Security fit into asset-allocation decisions? Some people say that Social Security should be seen as if it were a bond that is going to continue to pay you throughout your life. How would you recommend viewing Social Security within the asset-allocation framework?

Benz: My view is that people should look at their income needs during retirement, subtract from that certain sources of income that are coming in the door--pension, Social Security payments, maybe some sort of fixed annuity payments. The amount that is left over, that's what the portfolio will need to replace. Then, I think you create a sensible asset allocation for that portfolio plan. So, I wouldn't necessarily consider Social Security to be a fixed-income substitute directly. Indirectly, yes. But I think there are some tricky behavioral issues that can crop up if someone has this all-equity portfolio because, over here, is their bond--Social Security. The person in 2008 might see their all-equity portfolio drop by 30% or more; they might not be comfortable with that level of volatility in the actual portfolio.

Balasa: Conceptually, it makes great sense, as it does with the actual pension plan, to say, "Social Security is a bond equivalent; so, because of that income stream, I have a $2 million bond portfolio, by default, relative to the rest of my assets." That sounds good in theory, but it's the behavioral aspects that are tough. If you go back to 2008-09, if you had taken that approach--including Social Security--with a typical investor, that means they would have had a lot of equities, and they would not have been able to stay the course. So, although it's nice in theory, maybe on the margin it works, but as a complete solution, no, I don't think it works for a typical client.

• Read more: 3 Ways to Maximize Social Security Benefits

Zoll: I want to move now to a broader discussion of asset allocation. If you are trying to build a portfolio that's going to last for a 25- or 30-year retirement or longer, what are the basic components that you need to focus on?

Benz: I think a balanced portfolio really is a great starting point. You absolutely need equities for the longevity risk. You need that growth portion of your portfolio. When you think about the fact that starting bond yields have historically been a pretty good predictor of what you might expect from bonds over the next decade, well, at 2% on the Barclays U.S. Aggregate Bond Index right now, that's not a return engine for many retirees. They'll be lucky to keep up with inflation at that level. So, you absolutely do need stocks in the portfolio.

In talking to retirees recently and hearing from our users, my concern is that there is some complacency about equity-market risk--that, in fact, retirees are perhaps too comfortable with their equities. They have forgotten what it felt like during the bear market. So, if anything, I think many retirees seem to be erring on the side of having too much equity risk in their portfolios. You absolutely do need stabilizers, and that's why my model portfolios include cash, they include high-quality bonds--miserly yields and all. But the idea is that those are the ballast for the return-engine piece of your portfolio.

• Read More: Christine's Model Bucket Portfolios for Retirees

Zoll: I suspect one reason that seniors are focused on equities as an alternative to bonds is that there is this interest-rate anxiety that's been with us for years now. But you still feel that it is important to have that bond exposure in your retirement portfolio?

Balasa: Yes, definitely. To Christine's point, the expectation for rates to rise obviously is here. It's been here for a long time, but eventually it's going to happen. Earlier this week, I was at a conference where Ben Bernanke spoke, and he talked about his view about what's going to happen. It's pretty much like everyone else's view--that there will be a rate increase this year. But what they also say in the same breath, though, is that if the rate increases are going to be as advertised--which is always an unknown--there could still be positive bond returns for the Aggregate Index and other shorter-term maturities. And as bond rates increase over time, that's good long-term returns for bondholders.

Again, to Christine's point, if you go back to 2008-09 and you look at the anxiety that came out of an event like that, people sometimes need to be reminded that it's classic behavior. It feels good; it feels comfortable. Expectations for bonds are pretty modest, so what's the other alternative? All equities. So, there's no question, in our opinion, that bonds still have place in the portfolio.

Zoll: Traditionally, retirees are encouraged to have some sort of inflation protection in their portfolios, whether it be real estate, commodities, or TIPS. We haven't really had to worry about inflation for a long time now. Mark, do you think that inflation protection still has a place in a retiree portfolio?

Balasa: I think there is always a place for it. I will go back to Bernanke's comments earlier this year. He said, as far as the eye can see, in his opinion, inflation is not going to be a concern; he listed some of the obvious reasons--energy prices, and so on.

With that being said, the central banks--either ours or the ECB or Bank of Japan--might make a mistake. There are always opportunities. For that reason, I would say yes [to inflation protection]. But when you put these inflation hedges in the portfolio, you still need to be aware of valuations. Commodities haven't done well lately; real estate has had quite a run; TIPS, of the three, perhaps may be the most attractive, in our opinion. My point is that you shouldn't put inflation hedges in the portfolio blindly. Pay attention to valuations.

Miller: I would just add, though, that while it's probably right that overall general inflation is probably going to stay quiet, one thing that's important for seniors is that they are more affected by health-care inflation, disproportionately. And that continues to run, unfortunately, higher than general inflation--less than it has, but there is still a debate as to whether health-care inflation could take off again in a way that could be really damaging. I don't know what the solution to that is, but I think it needs to be considered in the overall risk picture with respect to inflation that affects retirees.

Benz: I love that idea of customizing your own inflation targets--thinking about your consumption basket. Another area--and, Mark, you may know more about this--but I believe that food costs, historically, take up a larger share of retiree budgets than they do for the general population. The cereal boxes seem to be shrinking. The pasta boxes are shrinking, and food costs are on the rise in many cases. So, that's something to keep an eye on.

• Read more: Is Your Retirement Portfolio's Asset Allocation on Track?

Zoll: Let's talk about some rules of thumb for maximizing your tax-advantaged accounts in retirement. Some say that any Roth assets should be held and withdrawn last because of the tax-free compounding and the value of that. Christine, are there instances when withdrawing from Roth assets earlier would make sense?

Benz: There has been a lot of research in this area. The standard rules about sequencing withdrawals in retirement indicate that you'd go through taxable accounts first, then tax-deferred, and then leave Roth until last. But that withdrawal sequence doesn't make sense in every case. The goal should be, on a year-by-year basis, that you try to keep the person in the lowest possible tax bracket, and that can mean pulling assets from a variety of account types. That means you don't need to be dogmatic and shouldn't be dogmatic about those withdrawals. A tax advisor who is well versed in these issues can really give retirees a helping hand on this problem.

Miller: One other thing that's interesting about the tax question has to do with something we talked about earlier--people working longer. The ordinary-income picture starts to get interesting for people who are past that usual retirement age but may be still working. Then, there might be some Social Security income, income coming out of an IRA or 401(k). It gets more complicated. It also starts to trip some unwanted consequences, such as greater taxation of Social Security. Plus, you move into these surcharges on Medicare premiums that kick in around $85,000 in adjusted gross income for an individual, significantly higher Medicare premium. So, there is a variety of things that are now in the picture with the "working longer" part of the story that weren't there before.

Benz: That's true. And RMDs are in the mix, too. We've had a very strong equity market--a very strong market, overall--that has lifted all boats. So, one thing that a lot of retirees are grappling with right now is that their RMDs are higher than they had been for many years. So, that would argue for perhaps pulling from some of those accounts where you would face lighter tax consequences to do so, such as Roth or perhaps a taxable account [if you need more money in a given year].

Zoll: This discussion of RMDs is a perfect segue into our next topic, which is spending and drawdown of your assets in retirement. For many years, this 4% rule has been the rule of thumb. You withdraw 4% of your assets during your first year of retirement and make an adjustment for inflation every year thereafter. Nowadays, with bond yields as low as they are, some experts are saying maybe 3% is a more comfortable way to go about it.

Mark, how do you guide your clients in terms of giving them some input with regard to how much they can safely take out each year?

Balasa: It's one of those rules of thumb that is really tough. For us, it's rarely correct because there are so many other extenuating circumstances. There's a pension, there's Social Security, there's an inheritance, we're going to downsize our home, we're going to sell the business. All of these things come into the mix. If all you have is a single pot of money and you're going to live all by yourself, maybe the 3%-4% rule works. But for most people, you have to do some scenario-planning on your own or with an advisor because there are so many variables. That rule of thumb doesn't answer most people's specific circumstances.

Miller: There has been a lot of interesting research suggesting that 4% is a little on the high side. There has also been some great stuff written, suggesting a more flexible approach: Assess it as you go.

There has also been some terrific research indicating that spending in retirement doesn't just move in a straight march upward; it tends to taper off when people hit their 80s. They are less active; they are spending less on entertainment and travel. So, the straight-line version of this is really not going to get it done for most people.

Benz: I think common sense is really helpful, too. You can run some of the scenarios, and I'm sure that would help you fine-tune things. But certainly, pulling back on distributions in those terrible market years will go a long way toward preserving your principal and making your money last. I think it's also important to consider time horizon. Certainly, younger retirees--people who have, say, a 40-year time horizon in retirement--should not be using 4% as a starting point for distributions. They should be lower. And people who, for whatever reason, want to have a very conservative asset allocation should also be more careful; 4% is probably too high if they have a portfolio that's heavy on bonds and cash.

• Read more: Don't Be Dogmatic About Retirement-Portfolio Withdrawals

Zoll: Health care is probably one of the biggest swing factors in terms of making your money last in retirement. How should people look at the health-care question, in terms of figuring out their spending pattern in retirement?

Miller: I would start with Medicare because most of us are going to file for it at age 65. Medicare does a really good job of smoothing out the more general, routine health-care costs and some not-so-routine costs as well. If you need to go into the hospital, it covers a great deal of that.

But you need to be on top of Medicare. There are some rules of the road to follow--for example, making sure you file on time to avoid penalties for late filing. The basic decision is whether you want traditional Medicare or the managed-care version, which is called Medicare Advantage. It can save some money for some people, but I still regard traditional Medicare as the gold standard because it gives you the greatest flexibility in terms of seeing providers. It costs a bit more, and it requires more paperwork because you're putting together the different layers on your own. You sign up for Part B, which is outpatient; then you add a drug plan, which is Part D. And you probably add a Medicare supplement plan. So, there are more moving parts. But for a lot of people, it's the way to go because they don't want to worry about just seeing in-network doctors.

Zoll: What about long-term-care insurance? Some feel that it's not worth the high cost in many cases. Others feel that it provides great peace of mind, and therefore, it is worth the cost. Mark, is long-term-care insurance is something that you recommend for your clients?

Balasa: It depends on the scenario. In many cases, our typical client can self-insure. They have enough assets that they don't have to worry about getting it. That being said, some still like it because of the psychology around it. They've seen how their own families have been devastated by end-of-life costs, and so they might buy just a starter policy, even though they don't really need it. Others, of course, want it and they can't get it because they can't pass the underwriting. So, it comes back, unfortunately, to specifics. Do you have enough assets? And it's a catch-22 because if you don't, then it's hard to afford the premiums. But there is a case to be made for people whose children aren't around or if they face longevity risk, and so on. It comes back to doing the actual scenario-planning for them.

Miller: Mark, how do make a decision about whether a client can self-insure?

Balasa: The typical monthly cost of long-term care in today's dollars is somewhere between $8,000 and $10,000. We then use a conservative average stay of around four to five years. We'll put that into their plan, inflate it, and see whether the client can withstand that.

Miller: The long-term-care insurance policies, that whole market is a dysfunctional area. It's not growing; a relatively small number of people buy it. There have been massive price increases because the underwriters have had trouble figuring out how to properly price these things. Some people think we're through the worst of that, but the increases have been scary. There have been 40% increases. And for people who have seen those increases, it usually makes sense to hang on and ride through it, because you've already invested in the policy. But it's been very, very tough.

• Read more: New Strategies for Covering Long-Term Care Costs

Zoll: Mark, when I spoke with you in advance of this session, you said that one of the challenges that your clients face in retirement planning is that not only are they trying to think about their own financial well-being for the rest of their lives, but they may also be financially supporting an elderly parent, possibly an adult children. How do you help them strike this balance of meeting these competing needs? I hate to call them that, but that's really what they are.

Balasa: That's exactly what they are, and this comes back to the question about the 3% to 4% spending rule. This kind of situation blows that thing apart. For us, it comes back to the same thing many families have to face. You've got one pitcher of water and you've got seven cups; you can't fill them all up, so what are your priorities? For them, we would say, "OK, what do you need?" Because the overriding thing that we hear the most often is, "I don't want be a burden on our children, so how can I solve for that?"

Next, they might say, "I have to take care of my parents because of X and the kids may need help, too." There are all kinds of scenarios. So, we try to say, for the family, "Let's scenario-plan: If we do this for you and we take a little bit down, if we help Mom and Dad, how much is left for the kids?" We go back and forth until they come up with something that balances all of the competing needs.

Zoll: Some people are fans of annuities. They like that lifetime income stream. Others talk about the high cost and wonder if it's really worth it. There are also various types of annuities. What do you think about annuities, and are there specific kinds of annuities that you would recommend and others that you would not?

Benz: I would tend to start with a very plain-vanilla annuity type. The problem is here, we've got interest rates in the mix. Interest rates are really low. Annuity payouts from the single premium immediate annuities are arguably as low as they can go, too. So, it's a timing question. I think it might be an interesting product for someone who wants to just add that baseline of assured income, but the timing issue is certainly there.

Zoll: Mark Balasa, you have said that you don't generally recommend annuities for your clients.

Balasa: You can make a scenario for them, but in our experience, it's been less often than I think they are typically sold. A lot of times, the cost is an issue, and also the quality of the insurance company backing the annuity. And then, of course, the big concern is people losing the assets at death. There are different payout options for survivors, but when you factor in that the money doesn't pass on to the family, that's for many families a showstopper.

Miller: I think if you are interested in an annuity, the types to look at would be the single premium or the newer form, the so-called deferred-income annuity, which is an interesting way to add some insurance. It doesn't start paying income until you reach an advanced age, but cost is an issue that needs to be considered. Annuities are still a relatively small part of the overall retirement picture in terms of what the marketplace is telling us. People are not buying them in droves.

Benz: And it's still a really difficult area to do your due diligence, especially with the more complicated annuity types. You really have to know what you are doing before purchasing or even researching such products. And unfortunately, the good information sources are pretty scant.

Miller: One of the pluses that I could mention, though, for annuities is when you talk about this cognitive-decline issue. It is a way of automating that check to arrive every month no matter what's going on. It can be looked at as an automated insurance feature.