Don't Trust a Number You Can't Explain

AI can hand you a confident, polished answer in seconds. So can a stock screener. But a number without context can lead you exactly where you don't want to go. 

In Episode 63, Greg walks through what happened when we ran our McDonald's ($MCD) report through an AI model. It flagged a drop in return on invested capital as a sign of trouble. We saw a 12% return during a year the economy shut down and called it a win. Same number, opposite conclusions. The difference was knowing what drove it.  

The same trap shows up in dividend screens. General Mills ($GIS), Campbell's ($CPB), Kraft Heinz ($KHC), Smucker's ($SJM), and Conagra ($CAG) all offer attractive yields, yet most of them lost money over a decade when the S&P 500 gained more than 300%. Look underneath and the story is clear: weak returns on capital and a habit of overpaying for growth. Verizon ($VZ) and AT&T ($T) tell a similar story, while Xcel Energy ($XEL) flips it, with low returns today but a real catalyst ahead. 

We close with David Booth's book Stay Calm and its first rule: follow the science. Whether it comes from AI, a screener, or a headline, don't trust a number you can't explain. 

TOPICS COVERED

[00:41] Introduction
[03:11] AI critiques our McDonald's ($MCD) report.
[07:59] Trust but verify: the limits of AI in investing
[11:34] Listener question: Are dividend opportunities fading?
[15:12] Food stocks: big yields, poor returns ($GIS, $CPB, $KHC, $SJM, $CAG)
[18:54] Verizon ($VZ), AT&T ($T), and Xcel Energy ($XEL)
[22:40] Turnarounds, value traps, and Nvidia ($NVDA)
[26:38] Book review: Stay Calm by David Booth
[30:32] How we're staying invested without overpaying
[31:38] Final takeaway: don't let AI drive you off a cliff 

 ________ 

 📖 Free Book: Dividend Growth: The Quiet Engine of Wealth 

Dividend growth investing sounds simple, but doing it well for decades is not. That’s why we wrote Dividend Growth: The Quiet Engine of Wealth—a practical guide to building a framework you can stick with when things get uncomfortable. You can get a free copy here. 

📰 Monthly Newsletter: Observations on the Market 

📧 Questions or comments: dcm.team@growmydollar.com 

________ 

Resources:

📅 Schedule a meeting: Financial Planning & Portfolio Management
📊 Getting into the weeds: DCM Investment Reports & Models 

________

If you found this valuable, subscribing and leaving a review helps more investors find the show.

Instagram | Facebook | LinkedIn | X

________ 

Disclaimer: Past performance does not guarantee future results. Every investor should consider whether an investment strategy is right for them and all the risks involved. Stocks, including dividend stocks, are volatile and can lose money. Denewiler Capital Management may or may not have positions in the publicly traded companies mentioned herein.

Transcript

Copy Transcript Copied

[00:00:00] Greg Denewiler: This is Greg Denewiler, and you are listening to another episode of The Dividend Mailbox, a monthly podcast about dividend growth. Our goal is to stuff your mailbox full of dividend checks. When they grow over time, a funny thing happens: you create wealth.

Welcome to episode 63 of The Dividend Mailbox. Today's episode has several themes running through it, but the motivation really comes from a few things that have happened recently that, to me, made a statement: "You better understand what in the world you're looking at and why it's out there." So we're gonna cover it from different angles.

One of them is gonna be through AI and our experience there. It's almost like it takes its own personality. We'll just leave it at that. But then we will move into: is it hard to execute the dividend model right now? I will say it's not easy. There's a lot of companies out there that have been great dividend growth stories in the past, but are they now?

And then we'll end with a review of a book. It's a great example of how to deal with all the stuff that's going on out there. The piece that hopefully you take away from this episode: the distractions and all of the hot topics that are going on, it is even more and more important to understand why some of these things can be extremely misleading.

Just as a quick note before we get into this, dividend growth investing sounds simple, but doing it well over long periods of time takes discipline and patience. These episodes give you pieces of how we think about it, but if you're trying to build wealth, it helps to have a clear framework you can come back to when things get uncomfortable, which is why we wrote Dividend Growth: The Quiet Engine of Wealth.

And ultimately, the real goal of it is to help you tune out the noise and make better long-term investment decisions. So if that interests you and you'd like a free copy, you can find it in the show notes or at growmydollar.com/dividend-growth-book. With that, let's get into the episode.

I'll just start with AI's been in the news. Everybody knows about it. In the investment world, it's a big topic because there are a lot of people out there trying to come up with ways to manage a portfolio using AI, taking some of the emotion out of it. And then there's the whole research piece, and this is where this episode comes in, because we're putting together the first draft of our report for McDonald's, going through why we like it, us evaluating it as why we think it's a good dividend growth story.

We ran it through an AI model and basically asked, okay, critique this, but do it in a very rigorous way. Well, one of the things that it came back with that it did not like was our return on invested capital discussion. To quote what it said, "Every incremental return on invested capital figure spanning 2019 is garbage."

And then it goes through and says, "The ROIC declined from 21%," which was actually in 2018, "to 12% in 2020 is an accounting change you've presented as performance." Well, I wanna remind you, one of the things apparently AI is unfamiliar with is that the economy shut down for a few months in 2020.

There wasn't a whole lot of traffic. McDonald's revenue dropped by basically 10%. Well, one of the reasons why it wasn't much worse is because it's not a pure fast food restaurant play. It is more of a real estate play, and we went through that whole situation in the last call. So our view is that a return on invested capital number of 12% in 2020 was a huge win, and that is indeed performance.

But AI took it as if you have a big decline like that, that is not performance. That is, things are not going very well. So that is one thing that we're going to use as an illustration. Y-you have to know what drives these numbers. Another point that AI made was, according to its definition, negative equity has a huge impact on return on invested capital.

Our definition is a little different, and there are multiple definitions out there. They're not significantly different, but they can produce different results. The way we calculate it is operating income after tax, which means interest comes after, so it doesn't matter how the business is financed, whether it's all equity, all debt, it makes no difference, with one exception.

A higher debt expense can lower your tax rate, but we're not gonna go there 'cause I don't think that's significant. But operating income after tax, and then we divide that into total assets minus current liabilities plus short-term debt. What that does is it takes the entire asset base, and we subtract out current liabilities because most of that part of the balance sheet is vendor finance.

So that's not financing that costs the company anything. You add back short-term debt because that is capital that they raise, and it does cost them something. That's our definition. You know, if you've listened to our podcast, you know, a, uh, return on invested capital is one of the big variables that we lean on, but it's not the only one, and it's not the silver bullet.

It's just one thing that helps build a mosaic of, of what we're trying to find. And part of the benefit is none of these measures are perfect, but the return on invested capital formula is not really impacted by how a business is financed. But according to AI, their comment was, "Well, yes it does." Really, the whole point of me going through this is AI comes up with a conclusion.

It's going to frame it really eloquently, most likely, and you really need to know what's behind it. 'Cause if you're using AI to come up with answers for you because you don't even really know the question, or you don't know if your answer, how it's weighted or what it's weighted to, or whether to really challenge it and say, "Well, you know what? I don't agree with that because of this, this, and this." I mean, I actually kinda laugh at it. As we read the critique that AI did on our first draft of our report, my initial comment was, "Well, you moron, you don't even know what you're talking about." And then it's like you step back and say, "It's software. It's a machine."

So AI offers a lot of value, but some of the things that it may come up with are just not valid, either to your case or they misinterpreted what you were trying to achieve. You know, in my case, it had a different definition than I had. I mean, really, you're crazy not to use it because it can save a lot of time and make things much more efficient.

But it's the infamous trust but verify. It's kinda like driving a car with some of this new software. Personally, I'm not gonna drive a car without being ready to take the wheel at any point. By the time you figure out what's wrong, it may not be pretty. That's my only point there. There's a great interview out from the CFA, the Chartered Financial Analysts group, on: is AI potentially going to take over portfolio management?

It's probably going to be a great assistant, but let's just say you're 60 years old and y-you have $2 million. Would you put it all into an AI-managed portfolio when you have no idea when they're gonna make the next change, how they will adapt going forward, what they're gonna put more weight on, less weight on?

There's just multiple things that can create problems. Are they all gonna potentially do the same thing 'cause they're following the same thing, which creates a real herd mentality? I mean, things can go bad, and by the time you figure it out, it will be potentially too late. One of the great things about having a real person, people create relationships.

AI is never going to create a relationship. When things aren't going well, a relationship can help you get through it and to understand what the issues are, what the challenges are, and give you a better feel of why you should just stay calm. We did an episode and we kinda talked about second-level thinking.

The problem with AI is it takes everything from these databases from the past and uses that information to come up with a conclusion. Well, when you get into second-level thinking and looking at management teams that are potentially going to pivot, AI cannot pivot because it doesn't know what the pivot is.

It has no idea even that the pivot exists. So one of the things that becomes a problem now is a lot of the past data shows some really successful companies that are now starting to struggle. This leads to a question from a reader who read our book and had a great question that is very relevant in the current environment we're in.

So, you know, love the mental framework this book provides about investing. Dividends feel like they're fading over time, and screening for stocks that meet the criteria seems like it would leave one with far too few to feel diversified. Something about that bothers me. Do we reach further into smaller dividends with faster growth or just pile into more of the same?

This is where I get stuck. To answer the reader's comment about what do we do about lower dividend yields and investment opportunities as they diminish, as stock prices go up and the dividends are going up slower, so yields are declining based on current valuations. Well, I think there are still some ideas out there, hence, you know, we've been talking about McDonald's.

However, you can't really build a portfolio of 30 or 50 names of them right now, or you run the risk of paying too much. But then, this is gonna sound like a contradictory statement. We've talked in the past about dividend-biased ETFs. We do use them. You have to be careful because there's some definite style differences.

Higher yield, lower yield, higher growth. I mean, they're, they're really all over the board, just pretty much like stocks are. So it definitely takes time to dig in and figure out which one works for you, but there's nothing wrong with owning an index. Right now, that's how we get invested. Not fully, but we use them to help at least get the exposure to the client.

They're, they're much broader diversified, which is the whole point of them. Over time, our plan is always if we do start to come up with more ideas, then the money comes out of the index. It's a little bit of straddling the fence. But it still does give you the income component that you're looking for. I think that really sums up well that it is a tough environment out there because several things are happening.

Number one, the stocks that are cheap are cheap for reasons, and we're gonna go into several of these situations that are cheap that look like potential dividend plays, but are problems of one form or another. And then if you do find something that's good, the stock reflects it. And with all the money out there chasing returns, they're just flat expensive.

So trying to find something in the middle ground is, is challenging. So I am going to talk about some ideas that, well, some names that have been out there forever. A lot of them are decades, some of them are even a century old. And one of them I actually, which kind of brought this up to my attention, Campbell Soup.

You know, Campbell Soup continues to go down, and I have to say, it gets tempting. The dividend yield is attractive, but you take a look at it and it's got a very high chance of not giving you the total return even close to what you want, and it may even have a dividend cut. That led to just looking at a lot of different names.

Not that long ago, a lot of the food stocks like General Mills, Campbell Soup, Kraft Heinz, Smucker's, Conagra, these are all companies that have big brand names and for a very long time were able to cha-charge a premium, earn an excess return on those brand names. They were great dividend growth stocks, and the stocks performed well.

If you do a dividend screen, a lot of these names are gonna pop up. General Mills has a 6.8% dividend. Campbell Soup, 4.7%. Kraft Heinz, 6.5%. Smucker's, 3.7%. Conagra is 4.8% on the dividend yield. Those are all yields that are attractive, but here's the thing.

If you look at the last 10 years as the baseline, the S&P is up 318%. It's been a phenomenal decade for the market as a whole. General Mills is down 21%. Campbell Soup, down 47%. Kraft Heinz, down 55%. Smucker's, up 20%. Conagra, down 37%. These are total returns. You gotta remember, these include the dividends. These companies have been horrible total return investments. You look at their 10-year growth, they've started to run into problems there. General Mills, plus 27%. Campbell Soup, plus 11%.

Kraft Heinz, minus 33%. Smucker's, plus 49%. Conagra, minus 10%. What was maybe a hint to problems coming or, you know, why these things are struggling right now, back to our trusty return on invested capital. Our basic hurdle rate for return on invested capital is 10%, and the simple reasoning for that is that usually covers their cost of debt and also their cost of equity, which simply is the long-term return of the stock market. Well, if you look over, uh, just kind of a back-of-the-envelope average, General Mills actually has averaged about 10% return on invested capital.

So they have had a little bit better return on their assets, even though the stock is down. Campbell Soup, less than 10%. Kraft Heinz, less than 5%. Smucker's, less than 10%. Conagra is down around 5%. These companies, you know, they're not earning the premium on their brands.

And one of the reasons why some of them underperformed or the return on invested capital has declined from premiums in the past is they go out and acquire trying to buy growth, and they pay too much for it, and that just hurts their cause from an invested capital standpoint. Broadening out a little bit, you look at Verizon and, and AT&T. Again, these are big dividends.

Verizon, 5.6%. AT&T, 4.7%. Now, they've done a little bit better than the, the sector that I just mentioned. Verizon's up 65%, AT&T's up 52%. But that is way, way below a plus 318% decade. Dividend growth in Verizon's only been 22% in the last decade, and it's actually down 22% for AT&T 'cause they cut their dividend several years ago.

But both of these, an indication going forward, would I be looking at these things? Personally, I would not, because return on invested capital is less than 10%. They have a hard time creating much value for their shareholders after they finance the business, and that is a challenge. Well, on the other side of the fence, so to speak, that have actually done well, there is never one silver bullet.

Companies can have high return on invested capital, but they can't do anything with it. Or in the case of our next topic, it's gonna be a company that actually doesn't have that high a return on invested capital, but the stock has performed fairly well and there is definitely a catalyst there. So if you think this is simple and it's all about return on invested capital, well, let's look at Xcel, which is the utility that covers Colorado and a big part of the West.

It's a 3.1% dividend, kinda in the sweet spot there. 10-year dividend growth, it's up 75%. They've raised their dividend. It's not a double, but it's been a nice dividend grower. Total return is roughly 150%, not the 300% of the market, but then you look at the five-year average of the return on invested capital for the last five years, it's only 4%.

Then you ask, okay, how can the stock have done so well? You know, the outlook for utilities clearly is, for the most part, positive because they are the picks and shovels of AI. They're the source of the power that drives the whole bus. They are putting on assets, they are building out the balance sheet, setting themselves up to have higher returns down the road, and these are long lead times as they build their infrastructure.

It's easy to see a catalyst behind Xcel, and the country and the world needs more power. So that may be part of the answer. It's got the dividend yield, it's got the dividend growth, um, there's some growth in the earnings, but the return on invested capital is low, so it's, it's somewhat of an art form. It's not always black and white.

However, one thing we are not going to do is we're not gonna just go out and buy a stock because it has a good dividend yield. So the names that I mentioned earlier, we're probably not gonna own any of them. There may be a catalyst there, but in our mind, they're not dividend growth stories. They have not been for the last decade, and unless you feel like there's a great catalyst there, personally, we will stay away.

Now, I will be honest, everybody's trying to find the big turnaround. Well, maybe not everybody. A lot of people are jumping on, you know, jump on the momentum wagon. The other wagon where you can make a lot of money fast is you find a turnaround where you get a catalyst, and all of a sudden, the perception changes.

It's not dead wood drifting in the lake. All of a sudden, it's a cigarette boat racing across the skyline of Miami. You know, AI may bring something up that, hey, you know, this thing is really cheap based on all these different scenarios, past valuations, big legacy companies that are potentially value traps.

But again, we're not just looking for a quick pop or, okay, you've got something that's trading extremely cheap, and maybe it even rallies 50%. We want long-term compounding of the balance sheet and their capital. As we look at some of these big legacy companies that we've talked about, they're not getting their return on invested capital.

They're buying revenue growth. So consequently, dividend may go up, but the problem is the stocks, the total return of the stocks are just not moving. And without knowing what a catalyst is and where it's coming from, quite likely they just keep trying to buy growth. So one of the things I'm going to mention, this is a dangerous place to tread, but I believe, you know, this is kinda com- you know, this is, uh, this is, this is building on the horizon, but this is just my opinion.

With Nvidia, what they're doing is they're buying other businesses in other areas where they're trying to keep paths open for growth, but they go out and they overpay for this stuff just trying to get a pathway. Because they are so profitable, you don't really see it right now in the numbers. They have so many deals tied together where they do financing for a company, either through buying into it or loaning them money, and then that company turns around and buys their product.

It's like, okay, we'll see how that shakes out in the end. Look, there's no doubt this is an unbelievably successful company that is very expensive, but for the growth, maybe it's cheap. I don't know. The only thing I'm going to say is there are signs that I think, personally, you have some flags that say, you know what?

They also are out there trying to buy growth and trying to buy new pathways, but the difference between the legacy companies that we've talked about before, Nvidia's profit, basically their margins hide anything below the surface that may be building. It's hard to see it unless you really look, and then you're also making an assumption that the company is, at some point, gonna start to get into a growth problem or the growth that they're buying is eventually going to catch up with them.

Maybe it does, maybe it doesn't, and that is what creates a market.

So everything we've talked about so far basically are different versions of how you could potentially be misled. But the key, as is always the case, you really need to do a little bit of homework to make sure that the numbers you're looking at are in fact the real numbers, or at least the numbers that mean something to you.

That brings up a good book by David Booth, who is part of Dimensional Fund Managers. They manage over a trillion dollars, and it's a great book just on how the market creates wealth long term, how they looked at what they call the science and came up with conclusions. This is how you invest. You don't tweak it.

You don't mess with it. You just stick with the numbers and let the numbers work, which is why the book is called Stay Calm: Learn to Embrace Uncertainty in Investing and Life. The thing that's kind of fascinating with the whole Dimensional story, some of them came from the University of Chicago. There's a handful of people that were Nobel laureates.

They actually had a big influence of how the whole industry transformed going from the '60s and '70s, where it basically was just buy a stock, mutual funds were all managed, and then we went into the whole passive portfolio, and a lot of that came from the research that came out of Dimensional funds. Well, the University of Chicago's business school is called the Booth School of Business, and when you donate $300 million to a school, guess what?

Even Harvard may name part of the school after you. How did he create that wealth? Obviously, he built a business, and part of the whole book is just, look, there's no secret. You just invest, understand what you're doing, stay away from predictions, stay away from market timing, and look at the science of how markets work.

And markets work through... fortunately, this kind of feeds into us. It works because companies produce cash flow, they reinvest it, and they build it over time, and it makes the company worth more money. And lo and behold, you wake up one day and compounding changed your lifestyle. It's really built on just being down the middle of the road.

There's no dramatic outperformance. They are big believers in passive index investing, but they do believe that there are ways to generate a little extra return. They look at it more in the small-cap space where it's not quite as efficient, but they are fairly significant believers in efficient markets. One of his first rules, he basically puts together a list of how should you invest.

There are basic fundamental rules, and his number one rule is follow the science. Anchor investment decisions in rigorous data-driven research. And this sets aside speculation and hearsay in favor of testable theories and verifiable results. Really have your supporting documentation for what your strategy is doing, know why you think it will work, and that's their definition of follow the science.

He goes in, on to, to, to further explain it, but basically what that helps you do, and we've said this, it's not that dividend investing is the greatest thing ever, but dividend investing for us and where we live in that space, it gives us the confidence that it will produce returns, and over time, they will create wealth and sometimes they will even outperform.

But right now, it's getting harder to execute. You know, how do you find the balance of staying invested without paying up dramatically? Well, number one, we do use some dividend-based ETFs to help give us diversification right now so that we are invested, but we're not taking risk in like a Deere or a Caterpillar, where these are good, strong core businesses, but they're very expensive because they have done very well.

There's nothing wrong with having some money in the S&P 500. And then we're out there looking for building blocks on top of it, where we find like a McDonald's. And then right now, our model portfolio has about 15% cash in it. You know, there are times when doing nothing can be a good thing. So as we wrap up, AI can do a lot, and it's doing a lot.

It can come up with some dividend strategies and dividend ideas. However, it also may come up with some value traps. I mean, everybody potentially gets caught in a value trap. You know, who's the biggest shareholder of Kraft Heinz?

I'll let you look that one up. The key is don't let AI drive you off a cliff, or make sure if you're gonna do something that is fairly significantly tied to AI, you, you better make sure you have some idea how it's making the decisions that it is making.

You know, a lot of this episode really just kinda attacks in different ways just the different things to think about and how to deal with a world that's moving very fast. You can get distracted so easy. As the book Stay Calm says, the most important thing is to have a strategy and stick with it. So that's just a little bit of word of warning. We do have our piece on McDonald's coming out soon in the next few weeks, and it's gonna really go into a deep explanation of, uh, how we've looked at it and why we're doing what we're doing.

So make sure you, if you're not a subscriber to our newsletter, subscribe. You will get it in an email when we finish it, or you can look on our website and we will post it there. With that, we look forward to episode 64.

If you enjoyed today's podcast, please leave us a review and subscribe. If you would like more information regarding dividend growth or our investment strategy, please visit growmydollar.com. There you will find previous episodes and also our monthly newsletter. If you have any questions or anything to add to today's episode, please email Ethan@growmydollar.com.

Past performance does not guarantee future results. Every investor should consider whether an investment strategy is right for them and all the risk involved. Stocks, including dividend stocks, are volatile and can lose money. Denewiler Capital Management may or may not have positions in the publicly traded companies mentioned herein.