Last time we talked about using a “bond tent” strategy to reduce sequence risk in the years just before and just after retirement.
In FIRE communities, an alternate strategy that has become popular is the “yield shield.”
A yield shield strategy involves holding primarily investments that pay a high dividend yield.
The theory is: if the investment pays a dividend yield of 3-4% that might be all you need to cover your safe withdrawal rate. If you don’t ever have to touch the principal, sequence risk might disappear entirely.
But is it really that simple?
This week, in part 2 of our 3-part series on asset allocation, we continue our discussion with Karsten Jeske, CFA, about the pros and cons of a yield shield strategy. We start by wrapping up our glide path discussion from last week, then dive into a critique of the yield shield.
- How early retirees with kids should plan for key expense milestones during retirement that traditional retirees have already dealt with (college, buying a home, etc)
- Pros and cons of a yield shield strategy
- Whether investing in dividend kings or dividend aristocrats helps to address the cons
- The importance of looking at total return when analyzing the yield shield, and why a high dividend doesn’t translate into higher total return
- Reasons why the yield shield can fall short (and examples where it did)
- The right way to define success of a yield shield strategy
- Why Karsten doesn’t fundamentally believe the yield shield does better than a plain vanilla stock index
This yield shield critique has been very controversial in FIRE communities. Are you persuaded by it? Or do you believe the yield shield performs better? Why or why not? Let me know by leaving a comment.
Don’t miss an episode, hit that subscribe button…
If you liked this episode, be sure to subscribe so you don’t miss any upcoming episodes!
I need your help, please leave a listener review 🙂
If you liked this episode, would you please leave a quick review on Apple Podcasts? It’d mean the world to me and your review also helps others find my podcast, too!
Links mentioned in this episode:
- Dividend ETFs: VYM, NOBL, SDY, DVY
- ERN’s Yield Shield critique: part 1, part 2, part 3
- ERN’s critique of the dividends-only approach
- Asset allocation: How to use a bond tent to reduce sequence of returns risk (HYW068)
- The shockingly un-simple math behind retirement safe withdrawal rates (Part 1) (HYW035)
- The shockingly un-simple math behind retirement safe withdrawal rates (Part 2) (HYW036)
- Schedule a private 1:1 consultation with me
- HYW private Facebook community
Read this episode as a post:
Andrew Chen 01:22
Today’s episode is part two of a three-part series on asset allocation. Last time, we talked about glidepaths and bond tents in the years leading up to retirement and the first few years after retirement to mitigate sequence risk.
This time, in this episode, we’re going to be talking about a dividend investing strategy commonly called in the FIRE community as the “yield shield,” and how that, in theory, might help mitigate sequence risk by reducing the need to draw down your principal in retirement. But we talk about some of the tradeoffs of that and whether that’s actually a fallacy.
And then, also stay tuned for the next episode, part three of the series, where we’re going to talk about an alternate dividend strategy where you pair rental real estate with your classic portfolio to see if that produces better sequence risk outcomes. And we also talk about, next time, the crossover point after which, as a retiree, you’ve effectively neutralized your sequence of returns risk, and you’re virtually guaranteed to outlive your savings.
So, it’s an action-packed agenda today and in the next episode as well. I hope you enjoy it, and let’s get to the show!
Early retirees with kids may face substantial expenses during retirement that traditional retirees have already dealt with: things like kids’ college tuition, buying a home, whatever. How should early retirees plan for these type of expense milestones in terms of their asset allocation and safe withdrawal rate planning?
Karsten Jeske 02:54
I can tell you what we do. We have a 529 college savings plan, and we still contribute to that. We have a seven-year-old daughter, and we have been contributing into her college savings plan since she was born.
Actually, I contributed into a plan with myself as the beneficiary before she was born. And then when she was born, I changed the beneficiary from myself to my daughter.
Very easy to do. I think you just have to sign one paper and send it in by mail. Probably now you can do it online.
This is what we do. This should hedge the college education. There is a saying in German that says, “Small children, small worries,” which, by extension, means “Bigger children, bigger worries.”
The worries and the expenses, by the way, too, they only go up as the children age. That’s definitely the way we plan, so I certainly think that we may probably never completely reduce our expenses that come from our daughter.
Here’s another example. What if our daughter is done with all of her education? So, paying for a wedding, then helping with the down payment, and then she moves somewhere and it’s in a high cost of living area.
Our whole point about early retirement was we can afford early retirement because we don’t have to live in San Francisco anymore. We can move to Southwest Washington State.
It’s very affordable here. Houses are relatively inexpensive. You can get twice the size of the house for one-third the cost relative to San Francisco, so it’s perfect.
But we might have to move back into a high cost of living area again later in retirement. That high cost of living area is not just the cost of living. It’s also the taxes.
Or, obviously, we could also then just say, “Sorry, we can’t afford it, so we’ll just have to visit only a few times a year.”
But there are a lot of potential expenses through the entire life span. And then you have grandkids. That potential expense channel starts.
Andrew Chen 05:32
So I guess the way to handle that, just factoring into your calculations to these expense milestones, and then trying to amortize toward them.
Because unlike the retirement date for the early retiree, which may have some flexibility on the left or the right, the kid is going to go to college, presumably. Let’s assume that that will be the case. And that’s generally going to happen around a fixed date, around 18, and she’s going to end around 22.
So, you know those dates are, relatively speaking, less flexible. And you know it’s going to spike during those periods, as an example. Is the right way to think about that just to factor that into your budget and amortize toward that, so that it doesn’t actually hit you as a spike?
Karsten Jeske 06:17
Well, I’m not amortizing anything. What I would assume is that exactly x, x+1, x+2, and x+3 years into retirement, we have more expenses because we’re going to pay for the kids’ education.
But effectively what we’ll do is, as you said, we amortize it. We put money aside every month.
And then, of course, people are going to say, “But what if there’s a stock market crash right around the time she starts going to college?” Then I’m just going to say, “Look, kid, we’re going to pay you one-fourth of what that portfolio is worth right now for your college expenses.”
If it’s more, that’s fine. Then we don’t even need to withdraw it all, and we can save it for grad school or for grandkids’ education.
And if it’s less, then you just get a student loan. It’s not like I have to hedge this exactly down to the last penny. I don’t even know what college she will go to.
We live in Washington State, so there’s some good state schools: University of Washington in Seattle and Washington State in Pullman. These are both really good public universities. I hope she’ll pick one of them and not get blinded by the expensive private universities.
And if she goes to those places, even at seven years old, I think we already have more than four years’ worth of tuition at seven years old. And if you factor in living expenses and some other expenses, housing and stuff, maybe that adds to it. So I think we’re good with that.
I’m more worried about that long-term prospect where we say, “We feel compelled that we have to move to a higher cost of living area.” That’s probably the more unpredictable and also expensive risk on the horizon there.
If you really want to commit to exactly hedging a certain amount of money, then you get into the issues. One is the glidepath. What kind of asset allocation do you choose along the way?
Right now in our 529 plan, we are 100% equities because it’s so far in the future. Again, if the money is not quite enough, well, sorry, kiddo, you just have to chip in a little bit yourself. But I’m pretty confident that there’s actually more than enough for college plus maybe even two years of grad school by the time we get there.
Now, is it going to be enough for private college and then medical school afterwards? Probably not. But then we’ll have to check if and how we want to do that.
You don’t have to help your kids with the wedding. You don’t have to help your kids with the house down payment. I feel a little bit compelled to do so because I look at what the federal debt is doing right now.
Somebody has to pay for all of this, and it’s our kids’ generation. So I feel that I’m probably compelled to help her a little bit, get a start.
I don’t want to spoil her, obviously. You don’t want to be these helicopter or snowplow parents that take care of all potential obstacles. So you have to find a fine balance there.
Imagine our daughter finds a job in one of these high cost of living areas. She pays rent there, and the rent is so high that basically her entire salary goes to the rent. She’ll never have enough money to save for a down payment, and she will be on this financial hamster wheel forever.
I would feel guilty if I didn’t help her out a little bit. And that potentially could be way more than you would ever spend on a public college education. So, we definitely think that there might be some big expenses in the future.
Our daughter is not going to eat much every month. If you look at the food budget or clothing budget or even entertainment budget, what we spend right now, we can buy clothing for her at Walmart, or we even get some stuff secondhand. She’s not complaining about them.
Now, once she gets older, that’s going to be a little bit more expensive. So this is where that phrase comes from: “Small children, small worries.” Everything just gets more expensive and more worrisome as kids age.
I have absolutely no illusion that, for some reason, we’re going to save money once our daughter gets older. It only gets worse. But again, there are ways to budget for it.
I don’t necessarily think that our spending path will be flat throughout our retirement. It might be flat, it might be slightly increasing, until we reach some sort of a traditional retirement age.
And in the future, there could be some time when we start scaling down a little bit because we travel less. But then that could also coincide with a time when health expenditures and nursing home and long-term care and stuff like that hits.
Andrew Chen 12:10
So, I guess, as you were thinking about planning for this, did that just factor into how large your portfolio you believed had to be to even get to the retirement date?
Karsten Jeske 12:23
Andrew Chen 12:23
And maybe in the early years, children are small, then the withdrawals are under-withdrawn because they’re going to be balanced out in the latter years with more aggressive withdrawal. Are these roughly the framework in which you were thinking?
Karsten Jeske 12:39
Yeah. Again, even with completely flat withdrawals, flat means you take your current baseline budget and you simply adjust it by inflation.
In real terms, it’s flat. In nominal terms, it goes up just with inflation rate.
I was always planning for the prospect that our expenses might even go up a little bit as we age. You can factor that into safe withdrawal analysis. You can model some additional cash flows later in life to hedge for nursing homes.
It turns out that if you retire in your 40s and you want to hedge some huge expenditures for nursing home, and we’re talking about $100,000, $150,000 extra. It’s not $150,000, period, but $150,000 extra on top of your baseline expenses potentially.
Even though it sounds like a big number, it’s not even going to make that much of a difference on your safe withdrawal rate today because that is so far in the future, and nobody knows what the CAPE at that time will be.
The near-term expenses, they are the big troublemaker from the sequence of return risk point of view. But what happens 40 years down the road in retirement, it’s not really going to make that much of an impact on today’s safe and failsafe withdrawal rate.
I’ve looked at all of these different things that can go wrong with your expenses. These big one-time expenses, or even the ongoing expenses very late in retirement, you want to kick the tires and check what difference that would make. It’s probably not going to make that much of a difference.
What is much more insidious and dangerous is a slow creep of your expenses going up by, say, 1% every year.
Because everybody else around you, they are adjusting. They are spending not just in line with inflation. They are adjusting their expenses more in line with per capita GDP growth, and that’s probably somewhere in the neighborhood of 1%, maybe a little bit more than 1%.
Maybe as a traditional retiree, you can get away with “I’m not going to do the nominal GDP. I’m just going to go with the CPI.”
But if you’re a very early retiree, and for 40 years, everybody else around you is raising their standard of living by 1% every year, and you don’t, after 40 years, you’ve fallen behind by quite a bit.
So, the more insidious problem in your safe withdrawal analysis is more of this slow creep up in your expenses. And it may only be 1% or so per year, and from year to year, you hardly even notice the difference.
But if you do this every single year, say, to keep up with the Joneses, keep up with your neighbors, and to keep up with your not early retired relatives and friends, then that definitely makes a big difference in your retirement finances: the fact that you just keep raising faster than the inflation rate your withdrawal amounts.
So, that’s the more insidious problem. Expenses later in retirement, 20 years down the road, is probably not the biggest problem.
I have this Google sheet where you can model your own retirement simulations. You just model that as an external cash flow need.
Or it could also go the other way around. You could say that, at some point, you will be eligible, say, for a company pension. You can take that out as a lump sum, so you put that back into the portfolio.
It’s really important to model these additional flows, whether it’s consumption creep or the big cash flows 10 years, 15 years in retirement, as, say, college education. We account for that. That’s one of the reasons why we were very conservative with our portfolio.
I guess we could have retired already in 2016, and we waited another year in 2017 and said, “Let’s just give it another year.” And then in 2018, I finally retired.
In hindsight, you could almost argue it might have been a year or two too late, that we could have retired a little bit earlier. But some of these spending uncertainties, they definitely convinced us to wait a little bit longer.
Andrew Chen 18:05
Yeah, it does make sense. This cash flow modeling Google sheet, share it with me and I’ll link to it in the show notes. I think that would be helpful for folks.
Karsten Jeske 18:17
Andrew Chen 18:18
I want to talk a little bit about the “yield shield.” This is one popular theory I know you feel pretty passionately about, for mitigating sequence risk, is to simply hold high dividend yielding assets, like dividend-paying stocks.
To use industry jargon, if you hold dividend kings or dividend aristocrats, which are two flavors of dividend stocks that have a track record of increasing dividend payouts pretty consistently, then perhaps you reduce the risk that a dividend gets cut or suspended during down markets.
And the theory is that if you can support your retirement cash flow needs solely using dividend payments, then you don’t ever need to worry about sequence risk because you never need to sell any principal.
Now, you’ve written in a blog post that concentrating your holdings in dividend stocks or dividend yield index provides maybe a bit of a false security, and doesn’t even alleviate sequence risk necessarily either.
Can you explain the intuition behind why you think that: why holding, say, a plain vanilla S&P 500 index fund might actually be a better safeguard against sequence risk, even though it doesn’t pay a high dividend yield?
Karsten Jeske 19:25
First of all, there’s so many different facets of this issue, and I have written four blog posts on this. The sequence of return series 29, 30, and 31, they dealt specifically with the yield shield and why it hasn’t worked so well.
Then part 40 deals particularly with, just looking at the equity side of the portfolio, can we just live off of the dividends? We never touch the principal and just live off of the dividends.
So, before I even get into the details, we have to get one thing out of the way and really clear. When somebody makes a claim that “Here is a shield,” implicitly, probably even explicitly, they made the case: “This is something that’s shielding you from sequence risk.” And it’s marketed as something that helps you every time.
When I want to falsify a statement like that, all I have to do is I have to come up with one example where it didn’t work. I don’t have to prove that this fails every single time. All I have to prove that this needs to be debunked is “Here’s an example where it didn’t work,” and I’m done.
Because I got a lot of hate mail. I’ve written 200 blog posts; I’ve gotten maybe 10,000 comments on all the blog posts; I’ve gotten countless emails. They were all very friendly.
This was the only blog post where I got hate mail, very nasty comments, and then some really threatening even, emails. So I must have stepped on some people’s toes there.
But this is basically the straw man argument that people make: “So you are saying that this fails every single time. Here was an example where the yield shield actually performed better than just the plain index market portfolio.”
I’m not saying that the yield shield fails every single time. I only showed that it failed that one time.
By the way, I wrote this article in 2019. I should probably do an update just to shake up things a little bit with the comments and the emails, and show that, by the way, it also failed in 2020.
I was surprised by it because I thought, “This might be something that only very occasionally failed.” And since it failed in 2007, 2008, 2009, I almost said, “Now maybe is a good time to implement this, because what are the odds that it fails twice in a row?”
But guess what, it failed twice in a row. 2020, it also performed very poorly.
So, this is the first thing I want to say. In order to disprove the claim that this is a shield or some kind of a hedge against sequence risk, I simply have to prove, “Here’s an example where it failed pretty miserably.” This is the first thing I want to say.
The second thing I want to say is that for two out of three subcomponents of the yield shield, I would never ever claim that the failure is because of the dividend yield. I’m not saying that because there is higher dividend yield, that’s why it failed.
I just say it failed because of something else. It failed despite the higher dividend yield.
And then, by the way, when the yield shield works better than the passive index portfolio, it’s also not because of the dividend yield. It’s also because of some other factors.
So, this is true for two subcomponents of the yield shield. The first subcomponent is that, within the U.S. equity allocation, you just look for higher dividend yield stocks.
And they didn’t even go to the dividend aristocrats or dividend lions or something. There are some different, more active dividend yield portfolios.
I think there’s a Vanguard high yield dividend fund. I think it’s ticker VYM. That has slightly higher dividend yield, and then they replace part of the domestic portfolio with the VYM.
It turns out that that was actually not the worst offender during 2008 because the VYM roughly performed in line with the S&P 500. And it was actually a little bit worse than the S&P 500, but I’m not going to make a big deal out of that. But what I’m going to make a big deal out of it is that, despite the higher dividend yield, it did not help you.
And the reason is that everybody, especially in the personal finance community, and especially in the FIRE community, they all pontificate every day how bad it is to do active stock picking. And now suddenly they say, “Now we pick stocks based on the dividend yield, and they are supposed to do better than the overall index.” There’s just this logical inconsistency in here.
To me, I look at the total return. And it turns out that if a stock has a higher dividend yield, then my first inkling is they pay a higher dividend yield, but then the price return is also going to be lower.
They will have lower capital gains. All else equal, they’ll have lower capital gains to make up for the dividend yield.
You can’t just say that “We have two stocks: one has a dividend yield of 2%; the other one has a dividend yield of 4%.” And you’re telling me that the 4% dividend stock is going to outperform the 2% stock by two percentage points every single year? No.
Andrew Chen 25:56
Right. It should be arbitraged away.
Karsten Jeske 25:57
It should be arbitraged away. In fact, the same people that make all of these claims about how bad active investing and stock picking is, they always make this case: “These markets are efficient.”
Now, if markets are slightly inefficient, because it takes a little bit of time and effort to dig through hundreds of pages of balance sheets and cash flow statements. Do you have an accounting background?
Andrew Chen 26:28
I’m also a CFA.
Karsten Jeske 26:29
Yeah. See? So, you could at least make the case that the active stock pickers that do all of this really careful work for the stock picking, and they go through company statements, and they talk to the CEOs and the CFOs, and they dial into the earnings calls, and they gather information that way, at least you could argue that a lot of that information is out there, but it’s not that easy to gather it.
But the dividend yield is probably one of the most visible and obvious characteristics that your corporation has. And suddenly, the same people that are poo-pooing active stock investing and stock picking are then reverting to active stock picking.
And then the measure that they use for the active stock picking is a measure that even Grandma can look up if she goes to Yahoo Finance. So, it just doesn’t make any sense.
So, for me, personally, I’m not saying that the higher dividend stocks are going to underperform. I’m simply saying that they’re not going to consistently outperform. Or they will outperform during certain market environments, and they will underperform during other market environments.
But please don’t call it a shield. Nothing is going to shield you here.
It’s like marketing a parachute and it works half the time, and then I pointed out that this parachute, unfortunately, killed the jumper under this environment. It didn’t work. Then people say, “Well, yeah, but in this environment, it did work.”
It’s still a pretty bad parachute if it fails x% of the time. And by the way, it’s not my job to prove that the parachute fails every single time.
I pointed out that out of the last five bear markets, certainly it failed in 2007, 2008, 2009. I don’t have enough data for all the different instruments they used to go back all the way because I don’t have preferred shared data, and I don’t have all the indexes that they used going all the way back through all the recessions.
But certainly over the last five recessions and bear markets, I identified one where it doesn’t work. There’s probably another one where it would have failed again, but just to be generous, I say out of five, one failed.
And guess what, now we have the sixth bear market, and in that bear market, the thing failed again. So, we just went from a 20% failure rate to a 33% failure rate.
Andrew Chen 29:10
And just to be clear, when you say “fail,” how do you define failure?
Karsten Jeske 29:16
Imagine you had retired at that date. There are two people: one just did a plain 60/40 portfolio, 60% S&P 500/40% 10-year treasuries; and the other one did all these bells and whistles. Who came out ahead?
By the way, it would not have failed in the sense that you ran out of money. Because, as I said, the 2007, 2008, 2009 bear market, as bad as it looked in real time, from a withdrawal rate and sequence of return risk point of view, it wasn’t one of the worst bear markets. That’s just for full disclosure there.
When I say failure, it means it did worse than just the plain vanilla passive portfolio. So, the first element is you take the U.S. stock portfolio and move it into higher-yielding U.S. stocks.
Then again, I looked at some of the more active U.S. dividend portfolios. There’s the NOBL, which has a relatively short history, so that was not available in 2007.
But there was a different one. I think it’s one of the State Street ETFs. That did a little bit better, but again, it’s not like anything that will save the day.
So, my lesson is that it was a bit of a hit or miss. Sometimes it works; sometimes it doesn’t.
But these are all still U.S. corporations. They’re all still constituents of the big U.S. indexes.
It’s a bit of a hit or miss, it’s a bit of a crapshoot, but the deviation between investing in U.S. dividend yield stocks versus the broad market is not huge. We are talking about relatively low tracking error between the indexes.
The bigger tracking error, of course, is if you shift from U.S. stocks to non-U.S. stocks. So now it’s the same asset class, but it’s a different universe. It’s different countries.
You have FX risk. You have the exchange rate risk. You have the risk of a lot of non-U.S. stocks have very different sector compositions.
There’s some people who say that the fact that European stocks have been underperforming recently is simply because they don’t have enough IT exposure. If you look sector by sector, the underperformance of European stocks is actually not that significant as if you look at the overall index. So, a lot of the underperformance comes from different sector compositions.
And then, on top of that, sometimes the FX movement goes in your favor. Sometimes it goes against you.
You can point out that, for example, 2001, every country went into a bear market. The bear markets happen everywhere: in Europe, in Japan, in Australia, in New Zealand. It doesn’t matter where you are.
Everybody gets hit with the bear market. There’s actually not that much diversification during the bear market because everybody just dumps every stock worldwide. It’s not just in the U.S.
But there is some diversification for the subsequent recovery because, for example, the recovery that followed after the 2002, 2003 bear market bottom, the recovery in the mid-2000s, was actually much better in the non-U.S. stocks. The non-U.S. stocks did a lot better during that recovery, but they did a lot worse than during the recovery in the 2010s.
And again, it has nothing to do with the dividend yield. The rationale in the yield shield is that you should invest in non-U.S. stocks because they tend to have higher dividend yield.
In the times when it works, it works, but it’s just because they have different sector compositions. You might have had good FX movements during the 2000s because the euro appreciated a little bit, and it had really nothing to do with the dividend yield.
Again, I’m not saying that the yield shield does not work because of the higher dividends. It’s neither because nor despite. It just doesn’t work because of some other factors.
Andrew Chen 34:10
I was curious, if we redefine the yield shield, don’t call it a shield, but just call it a strategy where what we’re trying to simulate is the thing that we described earlier, where as long as you don’t have to draw down the portfolio, then the sequence risk is mitigated.
It doesn’t matter if the good returns or the bad returns are first or second, and that the opposite comes later. You’re just trying to avoid that double dip where you have a bad return environment and you have withdrawals on top of that. And the way you hedge against that is by taking a withdrawal from the dividend, even if that comes at the expense of some of the total returns.
Because I agree: a company that is paying out a high dividend, almost by definition, means it doesn’t have as good capital investment opportunities. So, what do I do with the cash? I’m going to pay it out in the dividend.
Karsten Jeske 34:57
Yes, but even that doesn’t work.
Andrew Chen 35:03
I guess what I’m getting at is the risk that, in this definition we’re trying to mitigate, not that you’re trying to outperform this comparison portfolio, but rather, that you’re just trying to avoid running out of money.
Because the average retiree, I think, who is not steeped in financial research, isn’t thinking so much about “I want to be the best possible performing portfolio of all of them.” They just don’t want to run out of money.
Karsten Jeske 35:33
Again, let me stop you there. But even that doesn’t work.
Andrew Chen 35:37
Okay. Tell me.
Karsten Jeske 35:40
The reason is that, obviously, dividends will be cut. Even the ones that are not cut, they may not keep up with inflation. The stocks that you will never touch, there’s no guarantee that that principal will stay the same.
It will go down. It might move sideways. It might not keep up with inflation.
One little case study I did was because this is one thing that people always bring up: Australian stocks. Australian stocks have fantastic dividend yield. They have 4% dividend yield.
And let’s just put all your money into Australian stocks. You get a 4% dividend yield and never have to worry about sequence risk again.
I did the simulation. If you had done that in 2007, first of all, your portfolio would be down, both the nominal and real terms. And then, even the dividend income is also down because Australian stocks probably still pay the same 4% yield, but they also pay the 4% off of a lower base, so your dividend income also declined.
So, it’s not that easy. Even in the U.S., you look at some of the highest dividend yielding companies.
Somebody in the comments section made the case that right at the time when I wrote this, there was one very extremely high dividend yield stock. I’m not going to name the name, but it’s in the business of providing cable internet service to both businesses and households.
And that had a 13% dividend yield. Two to three days after I published that, they announced, “We’re going to cut our dividend by half.”
So, even this whole “It’s better from a cash flow point of view,” it’s not going to work that well in the sense that you might replace the sequence of return risk with basically sequence of dividend income risk. Because it means that if the stock market does well, then your dividends are going to increase too.
Or if you have a bear market short term, dividends will be cut. There are still some companies that used to be some of these dividend aristocrats as of 2007. They cut the dividend down to, say, symbolic one cent per share.
Again, I’m not going to name any names, but some financial institutions, obviously, they did that. Part of it, they had to do it just because of the cash flow problem. Part of it is because of the whole regulation, what came out of the global financial crisis.
They’re probably not even allowed to pay out much in dividends in the years. And I guess they got used to that cash flow arrangement. “If we don’t pay out as much in dividends, we have more money for investments, and we have more retained earnings that we can use for acquisitions and capital expenses.”
I guess capital expenses is not that big a deal for banks, but they probably got accustomed that “This is really good. We don’t have to pay out as much in dividends. We just do the retained earnings thing.”
We actually prefer that over paying out a lot of dividends.
And then, of course, in hindsight, you could have always said, “Yeah, you could have just avoided that.” But who knew in 2006 and 2007 that the financial sector is going to get hit so badly?
Andrew Chen 39:31
Let me ask one more question around this, which is, if an investor focuses on the dividend kings, which are the universe of companies that have paid 50 years of increasing dividends without exception.
And if you look at the actual names, they tend to be in very defensive industries, like Procter & Gamble or utility companies, where you need to consume the thing regardless of whether the market is up or down because you need to live like a human being.
And if you focus on the dividend kings and you build a portfolio around that, the goal you’re trying to solve for is not that you’re looking for 100% success, or even performing better than an alternate 60/40 or passive index portfolio, but just that the only thing you’re trying to solve for is reducing the risk that you will run out of money.
Is it still the case that you say: that it does not achieve that goal better than some of these other strategies?
Karsten Jeske 40:46
We have the VYM, which is not the dividend king, which is just a broad index. You just allocate according to dividend yield and nothing else. There’s no additional “quality screen.”
It’s not just the dividend, but it’s also the quality and the consistency of the dividends.
There is a fund by State Street that has a ticker SDY, and that has a long enough history, and there is an additional quality screen. For full disclosure, that did a little bit better than the S&P 500. So, if you had replaced your 60% S&P 500 in your portfolio with that SDY fund, you would have done a little bit better in 2008.
Not remarkably better. It’s not that this would have saved your butt, but it would have done a little bit better during the 2008-2009 bear market. But it would have done very poorly in 2020.
And there is the other fund, but unfortunately, that doesn’t have a long enough history. That is the NOBL. I think it’s a power shares dividend aristocrat index.
That got completely hammered in 2020, so that did not as well as the S&P 500. That might actually still be down for the year. I have to look up the numbers, or you can go on Yahoo Finance and check that out.
It turns out that in 2020, what were the highest-performing stocks in the S&P 500? It’s the zero dividend payers. It’s the Amazons.
Andrew Chen 42:20
That is truly a black swan event. I guess what I’m wondering is if I’m your retail investor trying to think about how you can do early retirement, and I say, “I read the Trinity study, and it says 97% success rate.”
But I know that’s not going to be true going forward because if you actually look at an updated Trinity study that accounts for, say, the more recent 20-30 years of data, then you see that that 97% drops, in many cases, by maybe 60%. It’s not very comforting.
But if I shift my strategy to a dividend king strategy, and all I’m trying to beat is that mid-60%, I’m trying to just get that higher, does it increase the chances that that happens?
Karsten Jeske 43:11
That means you have to believe in stock picking. I don’t know if there’s a technical term for it, but I call it the “all else equal fallacy,” where I have two stocks, they have different dividends, and the dividend is something that is really tangible. You observe that.
And then the price return is something extremely uncertain. Because there is great uncertainty, the best initial assumption is actually not a very good assumption. But the best initial assumption that illiterate people would do is that “Because I have no clue about where the price return comes from, the best initial assumption is that I assume the same price return for both stocks.”
One has a higher dividend yield; the other one has a lower dividend yield. Boom, there we go.
Andrew Chen 44:04
No, but I’m saying not those people. Let’s take those people aside, and let’s say the investor accepts that the price return will be lower, because it has to be.
And you take away the stock picking risk or subjectiveness by just saying, “I’m going to build a market weighted index based on dividend kings, and I’m not going to try to be an active picker. I’m literally just going to build a passive index that follows the market weights of the dividend kings.”
Something like this, so you can make it more mechanical, and you are acknowledging upfront that I’m willing to sacrifice on price return. Because what I’m solving for isn’t the highest return. What I’m solving for is avoiding running out of money.
Karsten Jeske 44:45
Right. Again, if the total return is the same, then my simulations will give you the exact same outcome for the index return versus the dividend return.
Or you could go the other way around and say, “If I strictly withdraw only the dividends and I keep the principal in place,” then you’d probably do better than with the index portfolio where you strictly take out a certain amount of money.
But then the problem, of course, is that you also withdrew less during the crisis, because I can guarantee you that some of the dividends will eventually be cut.
And this is one of the problems. You have these dividend aristocrats, and they have a very long history of paying dividends. Of course, every stock has a very long history of paying dividends until it stops paying dividends.
Well, guess what happens to a stock that is even at the brink of cutting their dividends. They know they’ll fall out of the index, and then they have to make this conscious decision. “Look, we have this very long history of making dividends.”
They will have to pay out dividends, and they are burning cash at a time when this might be a good time to preserve some cash and do some investments because money is cheap. And we could do some acquisitions with that money and buy other assets for pennies on the dollar, but we are wasting that money on dividends.
So, this whole dividend yield thing, maybe I’m too much of an economist. There’s this saying about an economist that “walks past a $100 bill on the street.”
Well, it can’t be a real $100 bill. Somebody would have already picked it up. Maybe I’m becoming now so rational and so much in favor of market efficiency that I’m missing something.
But I think that the reason why the dividend yield indexes have not done so well, of course, you could argue that this is a sector issue. If you look at the sector composition of both the dividend yields, the VYM or the SDY or the NOBL, it’s very different sectors.
And you missed the big run-up in tech stocks during 2020. And then you got hammered, obviously, with these dividend-paying stocks because it’s energy. Energy got clobbered.
Energy got clobbered also in 2008-2009, I believe. Financial, obviously, that used to be a traditional high dividend payer. Low price-to-earnings ratio, high dividend payers in the 2000s, so financial got hammered there.
You never know what kind of shape the next crisis is going to be. I always say, “I do the index rates. I’ll be done with it.”
And if I do something else from the index weights and it works out really well, I think, “Look how smart I am.” Well, maybe you’re not smart. Maybe you’re just lucky because you inadvertently picked the right sectors.
And then I would also be afraid that some of these traditional dividend payers, they have this really bad “damned if you, damned if you don’t” issue, this catch-22 when they get into cash flow problems: “Do we keep paying the dividend? We’re just going to be depleting our cash right at the wrong time.”
“Or do we cut the dividend, and then our stock is just going to get clobbered because everybody is going to dump that stock?” And then everybody who holds that stock in the NOBL, until they announce the dividend cut, it’s already going to go down in price because everybody anticipates the potential dividend cut.
And then, once it gets dropped from the index and everybody rushes and has to sell it, then all of these passive index ETFs, they have to wait until it’s actually announced. And then there’s going to be a lot of people that can front-run these big index and ETF reweights.
By the way, I think there is something of a stock picking alpha that people can still pick up. It’s basically front-running some of these index reweights. That’s something that is traditionally done.
I’ve never done it, and I think the company where I worked, we never did that. But there are some hedge funds that do that, and they can make some money off of that by front-running ETF reweights.
Because the way that ETF is run, I think it has to be pretty explicitly stated what they do it and when they do it and what are the rules there. There are some smart people that look up these rules, and they just front-run the big trades that are coming from the big index players.
But anyways, as attractive and as reasonable and as logical as that may sound, it always circles back to: show me the total returns. If this is something that will consistently beat the overall index in total returns, I would be on board.
Andrew Chen 50:33
But the goal is not to beat the index. The goal is to give up some of the upside in return for mitigating the sequence risk.
Karsten Jeske 50:42
Yeah. But again, you don’t mitigate the sequence risk. The sequence risk is just going to hit you in a different way.
For example, I wrote a blog post. This is no longer specifically about the yield shield. This is part 40 of my safe withdrawal rate series.
It’s exactly that question. What if we just simply withdraw the dividends? And I did it with the S&P 500, but I guess you could do it with some other index.
But if the total returns are the same between the dividend portfolio and the S&P 500, it simply means that the sequence risk takes on a different shape. Instead of running out of money eventually, you run out of purchasing power in the short term. You just have to cut your expenses.
For example, you could argue that if you look at just the price return in the S&P 500, certainly over the last 100 years or so, just the price return, even adjusted for inflation, never dropped below the 100%. Over a 30-year window, just looking at the price return, you preserved your capital.
And there might have been one time where it dropped a little bit below, but it wasn’t much. So, as a working assumption, not only did you preserve your capital, but just purely through the price return, absolutely you grew your capital over 30 years.
But the problem with that is if you look at: how did the dividend income look along the way over the 30 years? First of all, sometimes you start with very low dividend yield, and the dividend yield is so low that even I, as somebody very conservative, would not pick a withdrawal rate that low.
It’s also volatile. The dividends, there’s some year over year volatility. There’s also some very long drawdowns.
If you look at the dividend income, I think it’s between the 1960s and 1980s, there was something like a 30-year window where you were underwater. So you started with your withdrawals, and over the next 30 years, you were underwater and you would withdraw less than what you would have withdrawn initially.
Again, I don’t have that same data for your NOBL dividend aristocrats or dividend kings, but I bet you some of that same pattern would have emerged there too. And it has to do with the fact that the dividend aristocrats look good. If you had known 25 years ago who are the dividend aristocrats today, and you had invested in those same stocks 25 years ago when potentially many of them were not even aristocrats yet.
Andrew Chen 53:29
That’s a good point.
Karsten Jeske 53:30
So, there’s a little bit of hindsight buyers where they can show you these are the stocks we have in there, and then people look them up and say, “Oh my god, they all did really well.” But the problem is there are a few stocks that dropped out of the index because they got clobbered in 2007-2008 or they went out of business in 2007-2008. That is one of the problems.
Again, it all comes back to total return. Either you get clobbered with the sequence risk where you run out of money after 30 years, or you might have to cut your dividend income along the way. So, that dividend shtick just doesn’t work as well as some people want to make it.
But then again, I give you this: that dividend stock style bias, I totally agree. On average, it sometimes works better than the index, and it sometimes works worse than the index. So, it’s a bit of a hit or miss, and there are some examples where it doesn’t work, so they shouldn’t call it a yield shield.
Just before we finish up this point, I talked about within the U.S., shifting into higher dividend stocks, then shifting from U.S. stocks to international stocks, that’s also hit or miss. That’s the 60% portion, the equity portion of the yield shield.
And I concede that it’s a hit or miss. It’s a crapshoot. Sometimes it works better; sometimes it works not so well.
The fixed income portion: When they start moving out of safe government bonds and into higher yielding bonds, on the bond side, certifiably, it’s going to hurt you during every single bear market. If you try to pump up the yield on the fixed income side, absolutely that will hurt you every single time during every single recession.
So, this whole stock market pumping up the yield, this is not the battle I want to fight, because there are some people that feel very passionately about the dividend side. If somebody wants to do the yield shield, and you want to do that only on the stock side, absolutely, you have my blessing.
If it doesn’t work out so well, don’t come crying to me. If it works better for you in the next recession, you can say, “Look how smart I am.”
But the one thing that will certainly hurt you during every single recession and bear market is pumping up the yield on the fixed income side. I had that little table in that blog post where I looked at the yield of different things that they want to put in the yield shield.
You start from the safest asset to the riskiest asset. The riskiest one is preferred shares. This is a hybrid between stocks and bonds.
If you go along the line, say you start with government bonds, and then you do the aggregate bond index, which includes also investment grade corporate bonds, and then you completely drop the government bonds. You go only investment grade corporate bonds.
And then you go a little bit more into the high yield sector. And then you do the preferred shares.
The yield goes up, but unfortunately also, the correlation with stocks goes up. So you gain yield, but you also lose diversification against your stock market portfolio. And that is definitely going to hurt you if you are right around the stock market.
Preferred shares got hammered really badly in 2008-2009. Some of the high yielders or corporate bonds, they got hammered.
And you get this dichotomy: the safe government bonds, they start with the lowest yield, but they also perform the best during the downturn. And then, the more you go into the high yield, the worse the whole thing is going to perform throughout the recession.
And you have the same thing. Companies fail. All the yields go down.
If you look at the dividend flows along the way, basically, preferred shares, there’s a lot more optionality. For example, if a company says it has a corporate bond and it stops paying interest payments on the corporate bonds, then it’s going in default or it’s going into bankruptcy.
But on preferred shares, you have a lot more leeway as a corporation. You could just say, “It’s a hybrid between stocks and bonds. We decide not to pay the dividend this quarter.”
That definitely happened during the 2007 and 2008 recession. These kind of assets get hammered really badly during recessions and bear markets. The same thing happened again in 2020.
So, there is definitely the one time where I could say that, certainly on the fixed income side, the yield shield, not only will it fail once, which is really all I had to show, but on the fixed income side, it will fail in every single bear market. It will hurt you in every single bear market.
So, don’t consider this a shield. Consider this just basically equity risk through the backdoor. Because the higher you go up in the yield, the more hidden backdoor equity exposure you have in your fixed income portfolio.