Agilent: Back To The Lab - [Business Breakdowns, EP.223]

0:03This is Business Breakdowns. Business Breakdowns is a series of conversations with investors and operators diving deep into a single business. For each business, we explore its history, its business model, its competitive advantages, and what makes it tick. We believe every business has lessons and secrets that investors and operators can learn from. and we are here to bring them to you. To find more episodes of Breakdowns, check out joincolossus .com. All opinions expressed by hosts and podcast guests are solely their own opinions. Hosts, podcast guests, their employers or affiliates may maintain positions in the securities discussed in this podcast.

0:45This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. This is Matt Russell, and today we are breaking down Agilent. If you aren't familiar with Agilent, it is a $30 billion market cap company at the time of this recording, and they focus on one of the more interesting niches, equipment and instruments sold into laboratories. So you can think of this equipment being sold to the life sciences sector, diagnostics, applied chemical markets, some very unique R &D work. My guest is Mark DeVos, fund manager at Troy Asset Management. And Mark brings us through the story of Agilent.

1:28This is something that was born within Hewlett Packard, spun off a little more than 25 years ago. And Mark helps break down what can be a very complicated concept around things like chromatography and some of the other things that Agilent's involved in. But this is a deeply ingrained industry player with a unique razor and razor blade model and a business that has evolved over time. These episodes tend to be my favorite, and this one lives up to the others. Please enjoy this breakdown of Agilent. All right, Mark, I am excited to have you here to talk Agilent. We were laughing a bit last week before the episode.

2:12One of the key things is just going to be establishing what Agilent actually is, what they do. I love these niche industries and the niche businesses that don't get as much airtime or coverage. Let's just start there. Do your best simple overview of what Agilent does as a business. Agilent is a leading provider of instruments, services, and consumables for labs. Sounds simple. You can picture one of their typical Agilent instruments to be about the size of a small fridge maybe on a lab bench, something like that. And its core use case is to tell you what molecules are in a sample that you're testing.

2:49A real world use case of this might be you've got a pharmaceutical company producing a drug. Let's say they're making aspirin and they take one tablet off the production line for every thousand tablets that they make or something like that. Those samples could be run through an Agilent piece of equipment and they would be testing to check they have the right amount of the active ingredient in them. And importantly, that there are no contaminants in them as well. Adden will sell them the instrument that can analyze that sample and also provide the ongoing service supports and consumers needed to run the instrument.

3:18If I'm thinking about this right, it's mostly biomedical type work that they're operating around and specifically in the labs of those businesses. Is that a fair way to categorize their end market? Yes. The business itself is actually pretty broad today. they're in 110 countries today 285 ,000 labs so it's pretty broad split between you know america's europe and apac in terms of the actual markets the kind of end customers pharmaceutical customers are the largest end market for them so that's about a third of their sales or so academia and government you can imagine universities use the stuff that's another 10 or so about 20 of the business is chemicals or advanced materials which is interesting so you could think of battery testing for evs or semiconductor production and another 20 is actually food and environmental testing.

4:03So this would be for contaminants, perhaps in your water supply, something like that. Is there any way to measure how big of a market this actually is? I'm always curious with these niche industries that revolve around very big end markets as it relates to drugs or anything along those lines. How do you categorize just overall market size for an Agilent? Agilent do publish a sort of total addressable market stat, and it's $160 billion, which is unsurprisingly a really big number when you think about all the companies and businesses in the world that need to test samples. Adjunct's revenue today is about $6 .5 billion, much smaller than that TAM number.

4:39In reality, actually, when you dig into those niche and markets, they generally have a number one or two position in most of their key markets. It's generally a pretty stable competitor set. If a company publishes a TAM number, it's probably going to be a big one. That's a fair way. But interesting to get that context of six and a half versus 150, just for perspective. Let's go back to the origin of the business. Where did Agilent come from? What is the origin story? How long has it been around? Anything that's key to its history that you would include? I think Agilent has a really interesting history.

5:13It actually starts way back in 1938, Hewlett -Packard or HP, started by two electrical engineers, so Bill Hewlett and Dave Packard working in a garage, a pretty well -known story. And the first product they developed, of course, was actually a testing instrument. It was used to test audio equipment used by Disney in the production of Fantasia, the movie, way back then. Over the years, HP built out their testing and measurement business, both electrical testing, but also over time for molecular analysis. And this included the acquisition of a small business called F &M Scientific in 1965. That business had quickly become the leader in what's called gas chromatography instruments.

5:47Their instruments were used for the first full -scale testing for steroids in the 1972 Munich Summer Olympics. The business continued to grow over the years that followed. In 1999, HP focused more on its computer business, and they spun out several businesses, including testing and measurement, into a new business called Agilent. HP kept their computing and imaging business with everything else going into Agilent. It was a bit of a mixture of assets that they inherited, but there was also some real gems in there. There's obviously years or decades of R &D, particularly in chromatography. In the years that followed, the Agilent management team refocused on its molecular testing business.

6:25So they sold their medical business to Philips in 2001. They actually had a pretty successful semiconductor component business that they sold in 2005 to KKR and Silver Lake. This would go on to become a Vargo, which then became Broadcom after they bought it. So it's been tremendously successful on its own. In 2014, that's the most recent significant change. So Agilent spun out their electronic measurement business into a separately listed company called Keysight Technologies. It's still listed today. It's doing well. That's really important. At the time, it was about 40 % of sales they were spinning out.

6:55And Keysight, whilst it was a great business, was selling equipment predominantly used for testing electronic devices during manufacture. So this was a really cyclical business and didn't really closely align with the rest of what Agilent was doing. The business that remained was entirely focused on analytical equipment for the life sciences, diagnostics, and chemicals market, so much more focused. What the management team was trying to do was have a much more clean, predictable, less volatile business, but still have those long -term growth drivers. It's very interesting when you find these family trees, standard oil being one that broke up, but all essentially in the oil business in various forms.

7:36HP being a new one that prior to this episode, I didn't appreciate just how much had been born out of that family. Quite some impressive market caps on those businesses that you referenced. You mentioned the electrical testing versus molecular testing feels very different to me in terms of, yes, they're both testing something at the surface level, but molecular versus electrical feels like it's a very wide gap. Was there a certain cultural DNA or recruiting mechanism that they got people attracted to the overall HP brand that allowed them to do that? This is a bit of a segue, but I am curious how they were able to foster innovation in two relatively different categories, at least to my brain.

8:18I think they inherited a lot of people. So Agilent is still based in Santa Clara. I think they're about 20 minute drive their current HQ from the original HP garage where it all started. There's obviously this sort of HP way culture as well, which we obviously own the shares. We've been impressed with the way that they run the business for the long term, the way they look after their people. You kind of get a sense for it, but we experienced that in COVID where right off the bat, they were first out the door saying no one in this company is losing their jobs or anything like that. I think that builds up loyalty and it's not unusual when we meet managers.

8:48They're generally agilent lifers that started as engineers. years. It's interesting where you might have a separation, but there still exists some type of relation to the original family tree and some of the talent that comes along with it. You mentioned when we were talking before a bit of a razor, razor blade approach with their business. How would you categorize what they're doing beyond selling just equipment outright? In terms of how I think about the business, about a third of it is selling instruments, the largest category of which is chromatography instruments. These are large sort of initial sales.

9:23And then two thirds of the business is selling consumables, servicing or software, which are mostly related to the instruments they're selling. So it's that roughly one third, two third split. I've mentioned chromatography a few times. What this actually is, is a machine that forces a sample, it could be a liquid or a gas at pressure through something called a column. You can think of a column as a metal tube that's stuffed with material that slows down the molecules in that sample at different rates. This means that the liquid or gas separates into different molecule groups as it passes through that column.

9:55And then on the output of the machine, you've got a sensor that detects the molecules now that they're separated. And depending on the type of sensor, it could also quantify how much of those different molecules are coming out. So it's a powerful tool. Typically, when they're selling one of those instruments, a chromatography instrument, it will cost around $100 ,000. Obviously, there's a huge range, but let's go with that for now. and it's got a useful life of about six to 10 years. Over that useful life, the customer will typically spend another $100 ,000 on ongoing consumable and servicing spend.

10:25That spend is split roughly equally between physical consumables and service costs. So just talking about the consumables for a second, the bulk of that expense is those columns that we just described. They've got a limited period of use, perhaps maybe 2 ,000 samples, something like that. And you need to replace them as the columns responsible for separating out that material. and it's degraded by high temperature or extreme pH levels, things like that. And that impacts the resolution of the instrument. These are used for pretty critical applications. You basically have to replace it often. It wouldn't be unusual to hear a customer replace it every month or so.

10:58Servicing is a bit different. It's more focused on repairs, recalibration, cleaning, or a service called qualification where they will be validating the machine is operating correctly. So we're saying Agilent is one of only two companies that has the ability to service other companies' equipment in the same lab as well. So it's a pretty big benefit if you're dealing with a large customer. And increasingly, Agilent services are sold as part of a broader service contract that's agreed when they sell the new instruments. One of the ways to judge it would be they publish a number that's the percentage of instruments out there that have active service contracts.

11:29It's called the attach rate. And it's been steadily increasing over time. So it's gone over the last five years from a sort of high 20s number to a low 30s percent. And we expect it to continue to grow about a percentage point a year. This is a great business. It sounds boring, but the bulk of agents customers are operating in regulated industries where they need to spend on consumables and servicing if they're using the instrument. The example I gave earlier, quite often the quality control process will be written into the FDA approval for a drug, and they might even mention the exact instruments and consumables being used.

11:59Once that happens and it's approved, it's expensive for the customer to move to a different instrument, and also there's some risks attached to that. So it gives a really nice recurring revenue on that consumable service spend. Is it right to think about the Agilent equipment being used and the assembly line essentially once a drug has already passed the approval rather than in the original testing or drug development stage? They straddle both. I think probably you're right, though, in that the sales split would skew towards production. They definitely make those cutting edge instruments which you use for your research and development.

12:33There's just a lot more production out there. When you look at the innovations, there's definitely innovations around making the machines higher resolution, more accurate. But actually, a lot of the innovations are focused on throughput, where this is a lab that's testing 10 ,000 sums a day. And if you can do it with one less instrument, then that's a big saving. My next natural question was going to be, what governs the decision for a lab to buy another piece of equipment, another Agilent system? Do you see labs with multiple on their assembly line? Or is there typically just one associated with each drug?

13:07I'm trying to understand whether it's capacity constraints or what drives that? Typically in a production center, you'd have a lab attached to it and they would have several instruments as part of that. And generally they will choose a single primary supplier. They may have a sort of secondary supplier there as well. But just for simplicity in terms of having spare parts on site, ongoing service and support, it's generally just easier to have one instrument. When they're considering going from that fourth to that fifth machine, they're considering the service they've received. I think most customers would probably highlight service as the primary driver of their decision.

13:40It's worth saying the machines are complex. There is a technical capability, so it's not like there's 500 options out there. It's a reasonably small group that you've got to choose from. But Agile is this one -stop shop where you can get your instrument, you can get your consumables, you can get your servicing all in one package. Talk a little bit about the competition. You mentioned there's one other competitor that can actually service third -party machines as well. What does the competition look like? Maybe we can start just in the equipment segment and branch out from there. Generally, they're competing, I'd say, against two main businesses.

14:12There are others, but two main ones, one being a company called Waters and another called Thermo Fisher Scientific. Waters is a strong competitor in liquid chromatography, but it's more narrowly focused on the pharmaceutical markets. So around 60 % of their sales are from pharmaceutical customers. They're also a smaller business. So revenue for water is about half that of adjuncts, around $3 billion. Adjuncts just above $6. Thermo Fisher is almost the opposite. It's a huge business, $43 billion of revenue built largely through acquisitions. And if you were to open a Thermo Fisher scientific catalog, you would see there's a much broader array of products.

14:47So they sell everything, lab fridges, test tubes, hot plates, centrifuges, lab shakers, water baths, etc., as well as the chromatography instruments that actually compete with Agilent. So maybe a bit less of a specialist in that sense. What drew us to Agilent was its diversified exposure by end market, but also its focus on niches where it has high market share. Can you talk a little bit about those niches? If you take a Waters who's entirely focused on pharma, what would be the other categories or niches within those categories that would be highlights for an Agilent? Gas chromatography is the golden business within Agilent.

15:27That's obviously where it came from. So F &M Scientific originally, they were a gas chromatography specialist. Agilent is still by far the market leader in gas chromatography. They're probably two -thirds of the global market for those instruments. They really dominate there. The difference between gas and liquid is predominantly driven by the sample. So gas, unsurprising, you're testing a gas. The way you get it to a gas is you heat it or you burn it. if you're a pharmaceutical company and you're testing a complex protein or something, you can't heat it up because you'll destroy it in the process.

15:57So therefore, you have to test it as a liquid. So generally, gas chromatography skews what we'll call applied markets. So you can think of that as battery testing, environmental testing for contaminants and pollutants and things like that. They're really uniquely strong, I would say, in applied markets for gas chromatography. These examples, they're reminding me of refineries from oil and gas, just in the way that they can handle this input and turn it into many bright products or outputs. There's a lot of differences, but it's interesting how much similarity there is. One of their customers is testing the outputs of refineries, basically.

16:31There are other areas as well where they're pretty strong. So liquid chromatography is the other one. They're probably about a third of the market. Waters probably has about a third of that market as well. And then the balance is smaller players and thermos showing the risk. They're also pretty strong in spectroscopy, so that's 40 -50 % of the market. I'd say that gas chromatography and applied is a post -child market. I would assume that a lot of the consumables and maintenance has somewhat of a reoccurring or straightforward measurement from year to year. The new equipment sales, you have this six to eight -year useful life.

17:05Are there major cycles, boom -bust oriented, that don't just revolve around customer losses? From an industry perspective, have there been tailwinds or headwinds that have resulted in shocks over time? In short, yes. Agilent's obviously pretty diversified by in -market, which has meant when there's been cycles in some of their smaller markets, they've kind of been able to weather it. One end market's a bit weaker, another market's doing a bit better. Generally, they struggle when their largest pharmaceutical companies, when that's doing badly, which we've just hopefully seeing the other side of now, they're just coming out of a sort of downturn there.

17:39Effectively, what happened was you obviously had the COVID -19 pandemic. There was a lot of investment from pharmaceutical companies into building out lab capacity and spending on R &D, and they bought well above the normal level of instruments. So you saw this boom in instrument sales growth. And then, of course, the CFO turns around and goes, we've got lots of instruments. We don't need to buy any more for a couple of years. You see the other side of that. It looks like we're just returning back towards normal purchase levels now. So hopefully that's behind them. The other notable area is China.

18:05So China is at 18 % of group sales. It's half of that Asia -Pacific region. Obviously, there's been challenges in China. It's a market that's led quite heavily by government grants as well to local companies that then go on and buy Agilent pieces of equipment, basically. We've been through a sort of down cycle there. Again, it looks like that's improving at the moment. For something like the COVID shock, where you had this major investment, when you think about the useful life of that equipment, and at some point towards the end of the decade, you'll come up on that six to eight years, is the expectation that there would be renewals?

18:39Do pharmaceuticals tend to take capacity offline? if it's related to something which has a shorter life cycle. How does that work just in terms of the visibility that you might have on the six to eight years and the replacement cycles coming through and showing up in revenue? I think just talking about pharmaceuticals, there should be growth really in pharmaceutical R &D spend over the long term. We expect to see growth there driven by increasing demand for healthcare, whether that's aging populations. I know the World Health Organization has population over 60, nearly doubling to 2050. Things like that are just going to drive healthcare spend, in our opinion, in one direction.

19:16And that's a tailwind for Agilent. The reality in the short term is that if an instrument's six years old, they could, of course, run it until it's eight years. There's nothing. It doesn't stop working. The parts are still available. They definitely could. What the drives are for that upgrade tend to be these instruments are operating at high temperature with sometimes nasty chemicals inside them. All components in them will eventually wear out. You've got the consumables, which are designed to be replaced. But you can think of all the valves, the heaters, everything else will just start to break more after a long period of time.

19:46And lab downtime is problematic for these pharmaceutical companies. They'll want to be getting rid of these instruments before they start to see a fair amount of downtime, which will naturally happen as they become older instruments. They'll just be looking to upgrade. The other thing is there are typically reasonably significant upgrades to the platform every three years or so. So after six years, you're probably two years, maybe eight years, three upgrade cycles in. And there should be a meaningful upgrade in terms of either resolution or throughput, which should help justify the upgrade spend effectively.

20:20With the competition, you mentioned that service is a major driver of the decision making there. When it comes to innovation, has there been a historical trend line for were Agilent or maybe a competitor where innovation has really driven market share gains or someone moving up the chain in terms of what they've captured of the market? I don't think so. We talk to the companies a fair bit and they talk about reasonably stable market shares overall. I think if you do a really good product launch after a couple of years, you might gain a percentage point of market share. These are very entrenched, stable market shares.

20:57Waters actually had to chart in one of their investor days, which showed sales of a specific type of column over time, we're talking over decades, and it was basically a flat line. And this was because of that point around it being tied into drug certifications. And once it's tied in, people just use the instruments again and again and again. If it works, and they're getting good service, they generally don't change it. So that's why that service component is so key. You mentioned there's stable market shares. Do those tend to be stable with the same existing customer base? Is there a lack of switching that tends to happen?

21:29It's pretty rare that a customer will build out a whole new lab. They'll generally be upgrading one section of a lab. There's that incumbency bias. We have seen companies talk about some opportunities where you're getting pharmaceutical companies thinking about maybe reshoring labs. That's one of the few examples where you're getting whole new labs being set up. And that's an opportunity where you can kit out a lab. Increasingly, the service part is being folded into the discussion on instruments as well. And this is benefiting those large incumbents. Agilent has 4 ,000 service technicians that they employ directly.

22:02Pretty hard to compete with that. 75 % of their technicians hold a degree in chemistry or biochemistry. 25 % have a master's or PhD. So these are really skilled, bright people that are out doing 2 ,500 on -site service calls a day. They're trying to fold that into the discussion around instrument sales more, I'd say. The point on the risk associated with switching providers might be the most meaningful thing. I'm curious, do you have any data points or measurement for what the cost of this equipment represents to the end customer just in terms of, is it a large percentage of the overall cost of production or anything else that represents a large or relatively small piece of that?

22:43We think about a production facility that maybe has five or so of these sorts of instruments, maybe slightly different types. Over those five instruments, let's say over an eight -year period, they're spending a million dollars on the instruments and servicing and consumables. It's sort of a rounding error in the scale of a pharmaceutical production facility over eight years, both from a revenue -generated, profit -generated, or the other costs in that business. This is one where obviously the cost of a product failure, so if you're failing to spot a contaminant or a problem or something like that, that cost is almost unquantifiable.