Skip to Content
US Videos

Three Main Variables in the Retirement Cost Equation

New research from Morningstar Investment Management examines the key factors that drive income replacement rates, trends in retirement spending, and the duration of retirement.

Christine Benz: Hi, I'm Christine Benz for

One of the many complicated aspects of retirement planning is figuring out how much of your preretirement income you'll need to replace during retirement.

Joining me to discuss some research on this topic is David Blanchett. He's head of retirement research for Morningstar Investment Management.

David, thank you so much for being here.

David Blanchett: Thanks for having me.

Benz: David, this is really a complicated issue. A lot of people use what's called the 80% rule as a starting point for figuring out their income replacement rate, but you really took a comprehensive look at this issue of income replacement.

Let's talk about the variables that you think are crucial to get your arms around when you're attempting to do some planning.

Blanchett: The moral of the story is, one size does not fit all. This is a very complex process to figure out how much someone needs, but there are three main variables to think about when estimating the cost of the retirement.

The first is, as you mentioned, the replacement rate: How much do you need to replace of your preretirement pay when you retire?

The second is, how does that need change over time? The most common assumption is that individuals have the initial need increased by inflation. So if you spend $40,000 in the first year, and inflation is 3%, you spend $41,200 in the second year, and so on.

The third assumption is, how long retirement lasts. People commonly use 30 or 35 years as the retirement period, but that actually varies, as each assumption does, by each individual retiree.

Benz: You found in terms of income replacement rate, there's a huge variation there. 80% does not fit all [people]. You've found that for some people it could be as low as 54%.

Blanchett: Or lower.


Benz: Or lower. And at the high-end close to 90%. What are the key factors that drive that big variation?

Blanchett: Lots of things change when you retire, obviously. You're no longer paying Social Security tax, Medicare tax. You're no longer saving for retirement.

If you're saving 15% of your pay before you retire, you're not used to consuming that money. When you retire, you don't have to replace that. Each person has a different perspective on what expenses go away at retirement. There are also tax benefits; there's a higher standard deduction. Things that make retirement cheaper--things like Social Security. Social Security isn't taxed the same way as income is.

Taken together, each person's replacement rate is going to vary. One trend that you do see, though, is that those individuals who tend to have lower income--so if you make $20,000 a year--have a higher replacement rate than those who make, say, $100,000 a year.

Benz: Let's discuss why that is.

Blanchett: It has to do with taxes and the implications of Social Security. Social Security is the biggest component for most lower-income households when they retire, and it's not taxed. But they also weren't paying tax beforehand on their earnings based upon their overall tax code. So, much doesn't change for them at all.

Someone who is making $200,000 a year has to be saving for retirement. They've got to be saving probably 10% to 15% per year, and that expense goes away when they retire. So all of a sudden that's gone, and taxes change as well. Their replacement rate is going to be lower than someone who is making $20,000 a year.

Benz: Let's discuss trends in retirement spending. There's that old adage that you have your go-go, slow-go, and no-go [years]. Did you factor in trends during retirement [spending] into this research?

Blanchett: We did, and I think that's a great way to look at how people's expenses change over time. There are the go-go years when you first retire; you can travel and do stuff. Then as you age, you slow down and your body says, hey, I'm tired, I'm not as healthy.

What I looked at was the actual changes in spending for actual retiree households in the U.S. This is actual U.S. data. What you see is that as people progress through retirement, they actually spend less and less and less. While the common assumption is that your spending increases with inflation every year, it's actually not the case. It actually goes down in real terms. So even if inflation is, say, 3%, your spending only rises by, say, 1%. It tends to tick back up at age 80-85 for health-care costs, but for the most part, you don't see this relationship where people are spending more and more each year based upon inflation.

Benz: I guess a crucial question, though, is, do people spend less as retirement goes on because they want to, or because they need to?

Blanchett: It's hard to isolate that exact variable. I actually created four different groups. What I found was that, people who have matched spending to wealth--so those that have the same overall levels of wealth to spending--tend to follow the same path over time. Those individuals, though, who have either over-saved or under-saved tend to adjust over time.

Individuals who have saved lots of money and realize, I have all this money, and I'm not spending a lot--they actually tend to spend more over time. Individuals who have under-saved spend less, but the overall theme, regardless of the group, is that, as you move through retirement, your increases go down. So you're spending less and less in real terms or inflation dollars over time.

Benz: A related question is, if someone goes through this exercise and thinks about their specific variables and comes up with something like, I'll spend 78% of what I was spending while I was working. All of that could go out the window if there are unanticipated costs. So how do those figure into all of this?

Blanchett: It's a fun exercise to make projections for 30 years, but it's kind of impossible. So, I think the best thing you can do is have a good starting point. What this research is doing is, helping someone figure out how much they have to save for retirement. I think the best thing you can do is ask yourself realistically, what do I need to live off of when I retire? That number is going to be different for every single person. Then based upon that, make some reasonable projections, and figure out what you can live off of every year, because one thing you don't want to do is over-save or under-save. You want to make sure that every year you're consuming the right amount to maximize your happiness during retirement.

Benz: Can you share some best practices for people attempting to run through this exercise, attempting to understand how much they'll need during retirement? What are some of the key steps they should take?

Blanchett: I think the first step is to figure out that replacement rate. That’s asking yourself, what do I spend today, and what's going to change when I retire? If you're saving to a pre-tax 401(k), well, that's 10% you're [hopefully not going] to need to consume when you retire. There are things like taxes. What do you spend on work clothes? What do you spend on getting to work? Ask yourself, what's going to change when I first retire?

Benz: Downsizing for a lot of people as well.

Blanchett: Exactly. Are you going to move? What happens with your overall lifestyle?

Then, how do I see my expenses changing over time? Obviously, a big concern is health care, and health-care costs are predicted to go up much faster than inflation. So, of what I'm going to spend, how will that increase over time. That's the middle part.

Then the last part is, how long is this going to last? How long do I need to save for? Is it 25 years, 30 years, 35, 40 years? Again, that varies by individual and by couple, but that assumption is really important, because obviously saving for a 20-year period is a lot different than saving for a 40-year period.

Benz: How would you suggest that investors begin to estimate that?

Blanchett: A really good approximation and what I like to use is, 30 years. It's a great starting point. But the key is to go in every year and re-address the situation. There's one way we've talked about before, the RMD method. You look at how much longer do I think I'm going to live. Let's say I have 20 more years to live. Well, then a good ballpark to take out for that year is 1-over-20 or 5%. That works for individuals and couples. The key is being conservative, but not too conservative, and going back each year and figuring out, does this still makes sense given my overall needs and lifestyle?

Benz: Update that as you go along.

Blanchett: I think that's the most important thing, because as you noted earlier, it's hard to make projections for 30 years. You've got to make sure that what you're doing makes sense for the long haul, and going in every year to revisit that ensures the safety of the overall strategy.

Benz: David, some really important research here. Thank you so much for sharing it with us.

Blanchett: Thanks for having me.

Benz: Thanks for watching. I'm Christine Benz for