Welcome to 25 for 25, the Credit Sites 25th anniversary edition of our No More Risk Better podcast. Five episodes, 25 questions, insights from our expert analysts and industry veterans. We revisit the origins of credit sites, share hard-won lessons from across cycles, and unpack the art and science of credit analysis. We'll highlight how we collaborate and explore the future of credit research, what's changing, and what still matters.
If you want to know more so you can risk better, you're in the right place. Let's dive in, celebrating 25 years of Credit Sites. Hello everyone and welcome back to the Credit Sites podcast, No More Risk Better. Today we have another one of our special episodes, our 25 for 25 series celebrating the 25th anniversary of credit sites.
The theme for today's episode is the art and science of credit analysis. We all know that data has been quite to the buzzword lately, and we want to talk to some of our experts on how we transform data into decisions. On that, we have some of our best analysts joining us today, Andy DeVries, our head of U.S. Investment Grade, and also the head of the energy and power team, an expert on all things utilities, data centers, and demand for electricity.
Andy, thank you for joining. Great to be here. We also have Eric Axon. He is a co-head of U.S.
High Yield and our head of healthcare, very focused on all things pharma, hospitals, what's happening in the political realm. Eric, thanks for joining. Yeah, thanks for having me. And then we have Mary Pollack, who is in our London office.
She is our head of Euro Real Estate, and always a delight talking to Mary as she gets a lot of good intel from our clients across the pond and is an American in the London office. So always interesting to hear from her how the U.S. and Europe-based investors think about things differently. So Mary, thanks for joining us.
Yeah, I'm excited to be here. Thanks for having me. All right. So let's just jump right into the credit call and analysis side of this conversation, because as research analysts, we have this kind of tricky dynamic where people are always looking for summaries and synthesis and feedback on what companies are doing and saying, but I think where our expertise is really valuable is making a call, having a distinct opinion on which way things are going to go.
Andy, this is something that you are truly passionate about. You always have a pitch. You always have a credit call. Can you walk us through one that you've maybe done recently that is memorable for you and how you think about the analysis that goes into these credit calls?
Sure. So we use a lot of data with our analysis of NRG and Vistra. These are two power producers unaffiliated with utilities. They sell their power into the open market and to one-off contracts.
NRG is short power via its retail book. I won't go into the details there. And Vistra is long power. So in Texas, solar went from around 2 gigawatts five years ago to 20 gigawatts last summer.
Right now, you're on 32 gigawatts, and this is an 87 gigawatt peak market. So obviously, all that solar has a really negative impact on power prices. So last fall, we upgraded NRG from underperform to market perform. Then a few weeks later, NRG raised EBIT guidance on lower supply costs, which means weaker power prices.
So we used some data there, but really crunching the numbers heading into this summer. Again, solar is now 32 gigawatts, but more importantly, batteries in Texas are up to 13 gigawatts. So we did a lot of work on battery storage because until three, four years ago, battery storage essentially didn't exist anywhere. So we used California as the blueprint because they had the most amount of batteries on their grid going to very strong incentives there.
So we crunched a lot of data, and we found that once you went over 10% of peak market demand, the amount of hourly price spikes significantly, significantly declines. So we used $75, $100 megawatt hours of thresholds, and we found that once you went over 10% in California, that's when the price spikes just fell off a cliff. So applying that same 10% to Texas, we saw that this summer, you were over that 10%. You're on 12 gigawatts now, so well over 10%.
So again, we turned positive on NRG versus Vistra. It's all relative value these days. And sure enough, as the summer unfolded, you just saw no price spikes at all, and NRG stocks up 85% versus Vistra up 55%. Both very, very strong.
Not as big a difference in the fixed income market, but you still haven't seen NRG outperforming Vistra. And a quick note on the data is, going to ChatGPT and other AI tools, anyone can crunch the numbers. It's finding the actual numbers to crunch. So in this case, we went to the Cal ISO website and got hourly data there, and then in Texas, we used Department of Energy EIA data to get that data.
So actually finding the data is harder than crunching the numbers. I would also just say NRG has since made a large acquisition to close their short position, and so they're no longer short power. So now you're in this game of short-term pain from solar versus long-term gain from data centers, and you can figure out when that flips, you can make a lot of money. Yeah, could you just let me know when that's going to flip so I can make a lot of money?
That would be super helpful. We're crunching the numbers as we speak. Yeah, of course. Andy, I like your point about you know, you can use tools now to crunch numbers so easily, right?
Like, now we're all quants. But I think that a big part of the analysis step is knowing what numbers to crunch and how to find those numbers. It doesn't seem like ChatGPT and other elements have quite gotten to a point where they can just kind of run this type of analysis from the hoop to nuts. So I'd be curious to hear from really any of you, like, how do we think about even the first step in knowing what the appropriate analysis is?
Luckily for us, the federal government's tracking this, and it's all free. So the EIA, Energy Information Administration, is a statistical arm of the DOE, Department of Energy. So you can poke around their website. They will probably be the last ones to launch a GPT, but others will eventually figure out how to scrape that data.
A lot of other data, as you can imagine, costs a lot of money. Yes, that is for sure. That is very expensive. Eric, I'd be curious to hear, like, how are you thinking about kind of balancing this quantitative analysis, the actual crunching of the health numbers, and the qualitative, right?
We all know within credit analysis, there is some degree of kind of subjectivity or this instinctual feel, like, what do we think of management, those types of things? Yeah, I mean, it's a good question. It certainly deserves a good balance. My view is that the quantitative analysis sort of serves as the bedrock on which you can overlay them, the qualitative.
You know, both extremely important, but I think the quantitative, at least in our process, sort of always comes first. Whether it's a news break, whether it's an earnings release, first step is always getting model updated, inputting the known financials and then moving from there. And of course, you can further update your models as you incorporate things like comments from management or your own assumptions. But again, that first building block for us is generally the quantitative side of things.
I think an easy example that I was taking through ahead of this podcast is like a large acquisition. On the credit side, it's really the first question you want to answer is what's the impact on leverage. So there you have to make sure your balance sheets are in order for the target and the acquirer. You have to estimate the EBITDA being acquired.
You have to input assumptions around deal financing, kind of those first building blocks to getting that initial sense of what's the leverage impact. And then from there, you can start overlaying that qualitative. So, you know, what's management talking about in terms of priorities for deleveraging versus capital allocation? You know, what's the trajectory of EBITDA and free cash flow and kind of folding in some of those harder to predict sort of elements.
But then that would give me a sense of, you know, what's the deleveraging potential? What's the impact on potential ratings? But all that kind of starts from that. What's the financial impact on day one?
I think on the qualitative side, too, where it really becomes powerful, I think, is in covering sectors over time and management teams over time. So I cover the pharma sector. It can be highly acquisitive. So we're sort of always trying to read tea leaves on management's acquisition appetite.
You know, what are the M&A needs and wants? Where are holes in the portfolio? But then how aggressive is management speaking about M&A versus the past activity? And that's a highly qualitative assessment to kind of boiling down, like, what's the M&A-related event risk for this particular company or for the sector as a whole?
So heavily qualitative there. And then I think when that marriage between the quantitative and the qualitative happens and when it works is where you can really come up with strong views and good recommendations. Yeah, absolutely. I mean, I find in strategy, we have all these frameworks and scenario analysis, but having the perspective of historic cycles, what kind of drove those pinpointing inflection points and trying to identify those going forward is a big part of making strategy work versus just being kind of like, well, lots of things can happen.
Sometimes that's what I feel like my job is. And, you know, thinking about from a junior perspective, we have a lot of people joining the world of credit, joining the world of finance in a time of massive technological innovation, so much data to digest. You know, I think about our colleagues on the buy side who are just inundated with reports and research and data and information. Andy, you have a great junior team and you always take a really proactive interest in juniors, including those on the strategy team, which I appreciate.
How do you help junior analysts who are newer to the world of credit develop some of those more instinctual things? You know, we can have juniors crunch data all day long, but how do we coach them on, you know, what the appropriate assumptions in a model are going to be? I think it's pretty simple. If you're a junior, you just have to read all the docs, like read every single document, the proxy statements, the loan agreements, obviously the 10Ks and Qs, you know, take a few pounds of salt and even the sell side equity research.
You just got to see what's out there. I found the proxies are great. Like who's getting paid in stock? Who's getting paid based on EBS growth?
Who's getting paid in cash? Might not care about the stock price. We had a management change at Explorer Infrastructure, formerly NextEra Energy Partners, and they didn't do conference calls. They're former bankers.
They stopped posting conference calls and they're saying, okay, we're selling this asset to D-Lever. We're walking away from these assets to our lenders, handing the keys over, and yet our EBITDA is going to be the exact same because we're building all these things over here. So it's like, wait, you're walking two assets and you're spending this. It's going to be an EBITDA wash?
Like, that seems a little too coincidental. So you pull up the proxy statement. They're getting 30% of their pay in stock. Like, okay, maybe that actually does give it a little more clarity.
So I just think reading all the docs is absolutely key. Yeah, definitely reading those docs. Can you just dump them into ChatGPT for summaries? I think these teams for the proxies and management have to get paid.
I think they have consultants to make it as obscure as possible. I feel like it should be like a binary, like drop-down menu that people could then feed into a ChatGPT. Right now it's not there. Maybe that's something we should actually do at credit sites is score the management compensations and how it gets paid.
All right, I'm going off the deep end here. No, I mean, I think that that is such an important part of credit work. When I was on the research team at Wells and we would have a new deal, the first question from sales was always, you know, who is management? Do we know them?
How are they compensated? Like, let's think about that as we're trying to sell this line deal. It's a pretty important facet. Eric or Mary, I'd be curious about kind of your views on developing juniors and their credit instincts.
Have you seen anything that's really worked in terms of helping juniors get to that point where it becomes a bit more instinctual or maybe something that hasn't worked? Yeah, I'm happy to jump in here. You know, I think about instincts versus process, particularly for junior folks. You know, instincts are great, but I think they can absolutely lead you to the wrong conclusions at times.
And so, you know, I prefer to lean on process, at least for, you know, earlier stage career folks. And, you know, going through the proper steps to arrive at an investment recommendation, particularly for complex situations. You know, we've all been in that position where you just feel overwhelmed at times. And I like to lean on saying, how do you eat an elephant one bite at a time?
You know, and so I find that, like, as I start to work through my process step by step, that investment thesis starts to take shape. And so I don't know if it's as much instinct as it is kind of staying disciplined to process on my side. Yeah, that's such a good point in having that framework for, like, what is step one? What is step two?
I think that in so many of the things that we are doing, there is so much information and having some discipline around how you approach these problems can be really beneficial. I guess on that, Mary, I'd be curious, as you are thinking about the experience that you've had, you have transitioned from one sector to another, you should cover telecom, you should cover real estate. How has your experience shaped how you approach your work? And think about that process in analysis over time.
That's a really interesting question, because I think when I look back at my career, I think this is true for many things in life. I mean, the first six months, two years, a couple years, the learning curve is so steep. And after three years, you're like, I know so much, I know so much. And I think after five years, you're like, you know, I've got this, like, credit research.
And then another five years go by, and you look back at what you wrote, when you're about five years in, you're like, I think it's about 15, 90% of that. But it's difficult to sometimes identify when 15 years have gone by, like, what, after that initial, just, like, learning the ins and outs, nuts and bolts process, like Eric was talking about, and doing the, like, detailed work that Andy was talking about, okay, what else has shaped my experience from there? And I definitely think, for me at least, I mean, credit, a lot of what we do is about pricing to risk. And there's something about living through instances where tell risk plays out.
One example that everyone in the market will relate to is 2020, like, seeing the COVID pandemic happen. Yes, maybe beforehand, you can see how one could have imagined it happening, but then seeing how it impacted markets, impacted industries, like, we all learned a lot in that period of time about how quickly the world can change. And that's also true for specific companies and specific industries. And I actually identified two earnings calls in my career that I remember feeling like the world changed.
And one was KPM, which is the Dutch Telco. And this was probably 2011-ish. It was the CEO talking about watching his kids use WhatsApp to send text messages. Because at that time, Telco pricing in Europe was done by text by minute.
So he was saying in real time, we have to change our entire pricing model for mobile, because if not, this is obsolete in a world of data. And the other one would be Disney's earnings call, sort of early in the transition to, or not early, now it feels early, when people are talking about this shit away from linear media, when they said ESPN is not immune. Until then, it was like, sports, sports are different. And they were like, sports aren't different enough.
And those days, like, watching the share price react, watching credit spreads react, it's like, sure, we look back and the writing's on the wall. But the truth is, like, you have to see it. And you have to see the way the market reacts sometimes before you can figure out how to incorporate that going forward. Yeah, it's so true.
Like, having those lived experiences is so valuable. And I think that's one of the reasons that the Credit Six team has fared so well during some of these things, where the experience has really informed the analysis. Like, I think back to the regional bank meltdown in 2023. And our bank's team had a very high conviction view, like, this is not systemic.
This is not 2007-2008. Like, over again, this is, you know, a very kind of concentrated area of risk. And they have such a great call. And, like, had they not lived that experience of 2007-2008, perhaps it would have been a different outcome or a different analysis overall.
All right. So let's do a little bit of a speed round. I think that we all know there's a lot of information out there. Even when I go on the Credit Six website, you know, I get all the LFI reports.
I get all the Covenant Review reports. I get all the Credit Six reports. Just on that, our website alone of really great information, sometimes it's really hard to parse through all the noise, all the headlines, and really identify something that is actionable, something that clients can take away and say, okay, I have a decision that I can make and transact in right now. So given your depth of experience, how do we sift through the noise and find those actionable insights?
Miriam, I'm going to put you in that seat and start with you. Sure. As an analyst that loves data and detail, this is not always easy for me either. But I would say it's sort of a clue to getting to, you know, a high value-add takeaway or piece of analysis.
I think sometimes it's like identifying the question you can't answer. For example, like, if management's provided guidance, a revenue target for 2026, and you cannot figure out through the KPIs that you have confidence in how they get there, if you have confidence in your analysis, and you can say, like, there's a gap here. Like, management is something that's not making sense. Like, I don't know how they're getting there.
Like, you can make the case for your numbers. Like, that's really valuable. Or, I mean, in European real estate, governance is a huge issue. And we do a lot of talking about what management teams do, but often it's actually not a management team.
It'll be a founder, largest shareholder manager, who is saying one thing on the conference call, one thing over the table, and then they've done everything else historically. And so just being able to say, like, you don't have really good insight. into what's going to happen next, like identifying what's going to be really difficult for investors to answer, I think can help draw your attention to, okay, what do we need to do next when we're looking at this situation? Yeah, that's a great one.
Eric, I'll go to you next. How do you sit through the noise and find those actual insights? Yeah, I think from my perspective, unfortunately, there's not really an easy answer there. I think you're certainly staying highly informed with what's happening with your sector, what's happening with broader markets, certainly within the companies that you cover.
That's sort of a never-ending process, but I do think that when you have these moments of heightened news flow, it does help you sit through what's a key data point versus what's a throwaway. I think having sources that you really trust and can rely upon and that have kind of delivered for you over time is super helpful in that identification process as well. And then, again, I go back to the process and I'm kind of blocking out external opinions and noise as you kind of work through that process and then not taking it well as you try to build up to your own unique thesis on a situation. So no easy one answer, but certainly I think knowing your sector and knowing the companies they're in and what's really important and critical for them is, I think, a good first step in being able to sit through noise versus something that's important.
Yeah, so a lot of prioritization there. All right, Andy, let's go to your next. Noise and actionable insights. What do you do?
Thanks, Wendy. I think the absolute key is finding out this is going to happen. What do the market think is going to happen? So, again, going back to my example of NRG and Vistra, Vistra's been saying they're going to get to investment grade at the senior and secure level for eight years now.
Absolutely no one thinks it's going to happen. But tiny little tweaks. Instead of saying we're going to get to investment grade metrics, they just said we're going to get to investment grade ratings. And then also they're fiercely competitive with NRG.
So no chat GPT is ever going to tell you that, but they're fiercely competitive with NRG. NRG's levering up to do a deal and can't get to investment grade. So maybe there's a chance to say, hey, we're actually pull ahead. And then later on top of that, Constellation has investment grade ratings.
They're the same business. They're four to five EVD that are multiple points higher. Their stock rate is higher than NRG and Vistra. Some of that's based on expectations for new nuclear data center deals.
But some of it's because they buy G. You say, you know what? Maybe actually now is the chance that they're going to do it because they're trying to get their own data center deals. And big tech's probably going to want to sign that deal if you don't have investment grade ratings.
So a couple of different things, three different things all changed. And we went out and said, we think there's a 50% chance this happens. And unanimously, every client wrote and said, you're crazy. You're drinking the Kool-Aid.
They've been saying this for years. And I highlight those three differences. And they say, oh, yeah, that is a change here. So just seeing what consensus is.
And that's when you're going to figure out how to make a trade and make some money. Yeah. I mean, Andy, sometimes it's really hard to figure out what consensus is, though. Not if you talk to people.
You just got to talk to people all day long. That's true. Maybe I'm a little bit suspicious in my role as strategist because I'll be like, you know, what do you think the pain trade is? Is it spread going tighter or longer?
And I feel like sometimes people talk to me about what they wish their portfolio was rather than what it actually was. I mean, that's probably how I think about my own personal portfolio as well. Data Into Decisions, that balance of the art and science of credit analysis. We hope you enjoyed this episode.
If you have follow-up questions for me, Mary, Eric, or Andy, you can always find us on the CreditSites.com website or reach out to your sales representative. Please like, share, subscribe. We really enjoy doing these podcasts. And now you can see our smiling faces on YouTube.
Much to the chagrin of Andy. But we'll get him. We'll get him on board. All right.
Thank you, everyone. Thanks, man. We'll see you next time.