
The world is shifting from asset light and infinitely scalable to asset heavy. Across the globe, trillions of dollars are going into assets. Capital expenditure is on the rise. What is driving this? What sectors are we seeing it in? What are the challenges and bottlenecks companies face in delivering on CapEx programs? And how do new technologies and approaches to talent overcome this? Our guest is Erikans Kok, Senior Partner at McKinsey where he leads their Capital Excellence Practice globally.
Loading summary
A
Foreign. Welcome to the HC Commodities Podcast, a podcast dedicated to the commodities sector and the people within it. I'm your host, Paul Chapman. This podcast is produced by HC Group, a global search firm dedicated to the commodities sector. The world is shifting from asset light and infinitely scalable to asset heavy. Across the globe, trillions of dollars are going into assets. Capital expenditure is on the rise. What's driving this? What sectors are we seeing it in? What are the challenges and bottlenecks companies face in delivering on capex programs? And how do new technologies and approaches to talent overcome this? Our guest is Erkanz Koch, senior partner at McKinsey where he leads their capital excellence practice globally. As always, you can really support the show by leaving us a positive review on the platform you're listening on and as always, I hope you enjoy the episode. Erica Hans, welcome to the show.
B
Thank you, Paul. It's great to be on. Thank you for having me.
A
Yeah, we're lucky to have you because we're going to get a bit of a broader vision here about a huge, a global trend. We're going to take it from the US perspective, but certainly a global trend and this is kind of a return to real world assets. A going from a world of this sort of the technology companies of infinitely scalable and asset light to a world now where every single vertical and as you're going to describe is investing in assets, big capital expenditure programs. And we're going to talk about the why of that, some of the challenges and some of the solutions. But it's a story that's playing out industry wide. It's certainly very consequential to the commodities sector. We could sort of start with kind of, I guess the general thesis here, you know, why suddenly have we transformed in this world of 10, 15 years ago of, you know, if you had big CapEx programs, you were discounted by the market to today where it seems every, every company out there, especially these tech companies, are spending every last cent of free cash flow on huge capex programs. Can you just give us the meta picture there?
B
Yeah, great. Paul. I think at the, at the macro level, as you're describing it, capital expenditures are critically important for the world. We need capex for energy needs such as oil and gas and electricity. We need manufacturing capacity to produce more medicine. That's of course needed in the US but of course also globally. We need CapEx to connect us through telecom. We need to build data centers to basically support all the analytical capabilities that we're getting more and more excited about. Of course, huge expectations in the US but also globally. But equally, and this is not a new trend per se, but you know, we need to continue to build big infrastructure, roads, bridges, airports to connect us again within the US but also again globally. And so all these end markets require massive amounts of capital. So at the macro level we're just seeing a big increase across almost any sector currently. Paul, If I compare CapEx for the next five to 10 years compared to the last five to 10 years, almost every sector is going to spend more and more capital then underpinning that, which is something that has not necessarily changed. But the micro level, if I think about at the company level, executive teams are aiming to increase share price. Besides investor sentiment, there are two ways to grow share price. It's either growth and roic. And so companies that are good in capex of course are able to grow faster and are able to improve their roic, which of course they will be rewarded for at their share price. That's also not a new thing. Um, but that's just an, an evergreen concept that I think urges companies now more than ever, could I say, to just be better in capex. This is a massive increase in capex expenditures and we'll talk about it, I'm sure, during our conversation today. Paul, but that massive increase in CAPEX is making competition for labor, for equipment, for materials ever more and that's making basically the returns difficult to achieve. So I think at the macro level, lots of activity, but then also at the micro level, at the company level of course very important for companies to be very good in CapEx.
A
Well, I think we know the AI story, right? That ultimately the cloud is just someone else's computer and a lot more computers are needed for the amount of compute the world is suddenly demanding and that's forcing money into data centers. And that's part of the very visible narrative and you know, macro conditions out there. Today we had sort of the warehouse story within Covid. I'm really interested in the, in the massive why. So if return on invested capital was always important and growth, you know, share price has always been important, presumably was 15 years ago. Are we just talking about a natural capital cycle here? Can you just sharpen the why? Why are we at this stage now? Is this normal? What's not normal about it? What's the biggest whys behind a global return to sticks in the ground?
B
Yeah, it's a great question, Paul. Let me, let me touch on a couple different sectors you mentioned data centers. Of course it's a global trend. You have the four hyperscalers that of course are putting massive amounts of capital in the ground. And you know, there's even a bit of a geopolitical element to this. You know, obviously US versus China versus potentially Middle east. And so companies and you know, politicians want to sort of like win that race. And so there's a lot of geopolitical pressure to spend the capex and to go build that infrastructure at the hyperscaler level. They of course want to compete against each other, but then there's also actually a lot of private capital. Basically people that are just seeing the trend and they're seeing the returns and they want to put capital in the ground. So here you go, all the way from land developers to technology providers. And so there's a lot of excitement in that data center space because everybody sees the potential and everybody wants to be first to market. By the way, many projects as you know, go over cost and over schedule. The schedule portion of this really should not be underestimated. So the time to market is important because the sooner we'll see, you know, revenue start to flow in, the better. And these guys are all competing against each other. So that's maybe at the data center then. Underpinning the data center is of course power, which we've talked about for a long time. For example, in the US it's really gas fired power plants that are going to provide the baseloads for a lot of these data centers, particularly particular in the near term. And so there is a lot of utilities that are now wanting to spend tens of billions of dollars to go build gas fired power plants which then of course is leading to, you know, construction bottlenecks and labor shortages and equipment issues which again we'll talk about in a second. Basically the ability to power the data centers is leading to a big increase in power demand. And Paul, as you know, power demand was sort of flattish for the last 20 years in the US and is now projecting, at least we're projecting, you know, the 3 to 4% CAGR year over year for the next 20 years. Given all the demand. You have the electrification trend, which of course is not one or two year thing, but you know, if you think about the last five to seven years, electrification has been a big topic and so utilities need to invest in the grid. The transmission and distribution lines need massive upscaling basically and upscaling something that we have not invested in a ton of in the US in particular over the last 50 to 70 years. But Grid's pretty old and so we need massive investments in, in tnd, tuition and distribution, lng. We've talked about it a lot, of course. Now what's going on in the straight, right, with gas prices, LNG market becoming more and more global, the US has advantaged gas position and so you've seen quite a few announcements. Basically companies building more trains or even just completely new assets, basically building out lng. You touched on medicine in your intro, I think. Very fascinating. I think life sciences companies have not really built new capacity in the last 10 to 20 years and there's a lot of new demand for new products in the life sciences space. So that's a sector that again has not really been building massive amounts of manufacturing capacity in the last years, but is now really starting to scale up. Right. And so we're seeing 15, 16% kangaroos over the last next couple years in terms of their capex expenditures. So there's a host of sectors, I would argue, Paul, where again, some recent trends are really pushing up the capital expenditures.
A
Is this a global story as well or is it very much skewed to certain areas, certain sectors? Can you just comment on that? And then I kind of want to come back to some of the underlying themes.
B
So data centers, clearly global infrastructure and residential builds, clearly global energy needs, clearly global. If you think about what Europe wants to do in their electrification journey. And of course the Middle east had big plans. We talked about ammonia and other green sources of electricity and power a lot over the last couple of years. There are certain regions that are still and obviously quite honestly more aggressively pursuing those than the US given current policy. But so most of these Abu Dhargi pool we really see as a global trend.
A
How much of this is a narrative of there was lots of capex going on. It's always been there. It's just been dominated by China in Asia. And actually this is sort of a recognition of that security of built at home is better in a less secure world. You know, this deglobalization that we're going through also security of information, all this piece is. I mean, how much of this is actually sort of the west in general now having to reinvest in their capital stock for all these reasons because it can't rely on the way the world was being structured over the last 40 years and it's just changing rapidly. Is that a fair. Is there something there?
B
I would not attribute the majority of the activity today to that trend. There is a bit, of course, with to your point, in terms of like, you know, de globalization, of course the current administration in the US has Been quite vocal about, you know, bringing manufacturing back to the U.S. and so yes, that's part of it. And maybe another small part is there's certain supply chains like rare earth minerals, certain refining capacity that is concentrated in certain countries where again the US would say, look, you know, we're trying to bring some more of that into the US So probably part of that is linked to that, Paul. But some of the other trends I would argue are just global economics and just companies wanting to win, companies seeing opportunities again, back to the data centers, back to energy, back to life sciences, back to domestic Taylor call, you know, with the 5G trend that we've seen play out over the last five to seven years. And so the majority from my view is still just, you know, raw fundamentals and just people basically seeing opportunities to put capital in the ground to get a return. A portion of it is indeed, as you're saying, more global geopolitics playing out.
A
Is there any sense of scale here? I mean, I'm sort of interested in kind of the capital cycle bit of this, whereas I keep saying it sort of 20 years ago, you know, services were booming. The idea of having an asset was discounted by the stock market. Whereas today, presumably investors, everyone is seeing the opportunity in and around these assets. Companies that don't have a plan to win in terms of capital expenditure, in terms of asset ownership and so forth are getting discounted. Have we seen these cycles go before? And can you just give us some sense of the scale of this as well? I mean, is this us sitting in our sort of energy and commodities bubble and Houston bubble and so forth, or is this a real meaningful global trend and we're just seeing an avalanche of capex going on these last couple of years, whatever it might be.
B
Okay, yeah, Paul, that's a great question on scale. So to maybe give you a few numbers from what we're seeing, so over the next five years or so globally we're seeing about an $86 trillion capital investment. Out of 86, 15 trillion is in the US and out of that 15, we would say about 3 trillion is data centers. Another 3 is in the energy utility space, about 2 is in the infrastructure. And then maybe on the life sciences, I think we're seeing roughly the top 20 global pharma companies spending 500 billion in the next five years or so, mostly between the US and Europe to give you a bit of a sense of scale. So these numbers obviously just, just staggering and you see individual companies spending just tens and tens of millions of dollars. Obviously you Know, led by budget, the four hyperscalers. You know we saw the Nextera Dominion announcement last week. I think that company combined is planning to spend 58 billion billion per year. Right. All in the U.S. so the numbers are just staggering and quite honestly larger than we've ever seen before. All right. If you think about it, just what companies individually are spending we used to have in oil and gas of course like these massive major projects, some of these LNG facilities in Australia, other locations in the world, which of course were tens and tens of billions of dollars as well, like per project. But now we're talking annual numbers, right? Just companies spending 60, 70, $80 billion per year, which is just staggering. So I think that that to me is the numbers just keep going up and then you could argue, look, part of that is inflation. Yes. Part of that is some supply chain bottlenecks. That's all true in any case, right. Things have just come more expensive but people are just building bigger, larger assets as well, which is requiring more capex.
A
That scale is, is sufficient to drive a pretty big commodity boom. Right, that's part of the conversation. We're going to have just a quick conversation on the financing of this and there is sort of this interesting interest rate backdrop which traditionally would have been a world of higher interest rates. Capex stops or drops, the opposite is happening and it would seem to me that part of that is because all of these companies have got, especially the hyperscalers, you know, so much cash on hand. A similar story to some extent in the oil and gas world where you've had great capital discipline over the last 10 years. So there's a certain amount of, they've got cash on the books that they can invest, they're not taking on debt to do this. I don't know what the story would be in the pharmaceuticals and so forth, but is this a free cash flow driven investment? Any sense sort of there on how this is being done and so forth?
B
If I think about the hyperscalers as well as life sciences, both sectors have just incredibly high margin, incredibly high cash flow and, and hence just have the opportunity to make massive investments based on their P and L and balance sheets. And so there is a lot of just balance sheet and from a cash flow perspective, financing of these large investments. If I think about oil and gas, I mean you mentioned the point around being capital disciplined. I think it's been a hard lesson for that sector because for many, many years, particularly in the US with the shield revolution, people were really seeking growth and the returns were less of a topic basically that is, to your point, radically shifted over the last five, six years. And we don't see that changing. At least I don't see that changing in the near future. So I think publicly listed oil and gas companies, from my perspective, will continue to be very disciplined and they actually have a lot of opportunities to invest, but they will continue to stay very, very disciplined in my view, and not overextend. And then I would argue, thirdly, there's a bunch of other sectors where again, it is maybe a bit of a combination between equity and debt financing because maybe, you know, equity markets are maybe a little bit less stringent on their capital budgets. They might not have the same sort of cash flow profile as the data center players and life sciences players. And so here it might be a bit of a mix between maybe a bit more opportunistically. There's also quite a bit of private capital actually going into some of these investments. So again, back to the data centers. I mentioned this earlier. Obviously you have the, again, you have the hyperscalers, but then there's also a bunch of private equity players that are just making big investments and not too dissimilar quite. We're seeing in some other sectors as well, you know, oil and gas and other sectors.
A
Yeah, private equity is locked and loaded with lots of, lots of cash and relatively few opportunities. Right. So they're a big part of this, as you know, private capital story as well.
B
That's right.
A
And you know, I guess we're not here to discuss the, the outcome of these investments and lots of big bets are being made. But what is fascinating is this. So we've established that, you know, this is a global trend. This is really consequential. It's across every vertical has a narrative around this. What I find kind of fascinating is suddenly this is a return of investment to labor and to real world assets and to physical commodities. And as Jeff Curry would say, we've all been penalized if you were in upstream investments the last 15 years because the world loved the infinitely scalable and asset light. And that's now changing. And in fact, the infinitely scalable and asset light companies are spending more in capex and more their free cash flow on assets than any oil and gas company would dare to do with its shareholders. And the lunch that you and I shared a couple of months back, you know, I found what was so fascinating about this story was of course, where there is finite resources, it's absolutely in terms of material, in terms of labor, in terms of expertise and Knowledge we've had over the last sort of 40 years where companies have been shedding their internal corporate development teams and asset teams that could design and build all this stuff. Right. It's all been outsour and that's creating incredible bottlenecks, which has a real world consequence on if you're second in the line to teaming your, to getting the talent in for your project, that might have a material and disproportionate outcome in terms of schedule. So can you just start off by describing kind of the, the bottlenecks that you are absolutely seeing and then let's go into kind of the different categories of those. But first of all, if you start off with just a bottleneck can be really consequential. And as you said, costs inevitably go up. But schedule is the killer in some ways.
B
An interesting initial statistic here. So when we think about capital productivity, and let's just take the last 20 years in the US and it doesn't change that much if you look at the last 40 years either, for that matter, when you think about construction productivity, it has been close to flat in terms of improvement. You know, we saw sort of like a 0.4% construction productivity improvement year over year over the last 20 to 40 years. If you contrast that to just the total economy, it's about a 2% per year increase. And then if think about manufacturing, it's about a 3%. Right. So 90% increase over the last 20 years compared to like a 10% increase over the last 20 years in construction productivity. So it's a sector that has quite honestly struggled to really increase productivity. I think that's at the outset. I think secondly, capital projects are just super difficult. Many projects go over cost over schedule. As you mentioned, these are complex projects. There's a big health and safety component to it. Obviously, most importantly, we want all those men and women to return safely to their families. And it's a dangerous work environment. And it's not trivial to put 5,000, 10,000, 15,000 people on a pretty condensed work site and have that group basically be productive. And so you mentioned scheduling. I'll get there in a second. But if you think about some of the other challenges, and so with that many people, there's many stakeholders, there is many contractors and subcontractors and sub. Subcontractors. And so the interfaces that you have to deal with is just very, very complex. And then lastly, I would say if you think about digital NAI applications and the use of that in construction, it is probably one of the sectors that ranks lowest in adoption, which again, probably also goes back to this sort of overall construction productivity point that we just have not seen massive improvements in productivity in capital projects.
A
Based on that, I'm kind of interested in how that plays out between, let's say, Europe and the U.S. where Europe, you've got, for the most part, a lot more stringent policy backdrop, you know, environmental, health and safety workforce and so forth, and the US in general. But I absolutely hear you in terms of the world of material science and productivity and, you know, deploying some of these new technologies. We talk about this episode on fertilizer coming up, and you're still using the harbor process from 40 years ago, right. It would seem to me doing some construction in my personal life at the moment, like Europe has newer technology, material science available to. I don't know, I'm sort of. Is there a stark difference between those two regions in terms of that capital productivity, or is that all a bit sort of, you know, Daily Mail and Wall Street Journal sort of battling it out?
B
I think at the high level, I don't see a ton of difference. Look, you could argue there's obviously certain regions in Europe, it's even more densely populated as well, compared to the US and so it just becomes even more challenging to do actual construction jobs. Um, there's of course, some of the labor laws that are more stringent in Europe. Permitting, you could argue, is. Is difficult in Europe. But then again, if I talk to companies here in the U.S. it is a massive issue as well in the U.S. all right. To just actually get projects done. And of course, in the US there's some state by state differences.
A
You're state, county by county, state by state, then federal law. Yeah, you've got a different suite of complexity.
B
Exactly. So at the high level, Paul, I don't see continuously massive differences. Again, on the margin. It might be slightly more difficult in Europe, but that's not necessarily what's driving the complexity, in my view. It is just complex. Yeah.
A
Yeah. Okay, let's talk talent and let's perhaps to make this a bit more real world, one of the popular things we're building here on the Gulf coast is obviously liquefaction facilities for lng. Two competing liquefaction plants are thinking about getting started. How does that labor challenge play out? And how do they compete for that labor?
B
Yeah. And again, let me maybe start at the macro. I think when we look at labor shortages in the U.S. so between 2017 and 2023, labor shortage doubled, or close to doubled. We were short about 200,000 workers in 2017. It went to 380,000 by 2023. So these are welders, electricians, pipe fitters, plumbers, all the crafts that you need to basically engineer and construct these projects.
A
Skilled labor, skilled qualified labor with the appropriate licenses and certifications across a whole host of technical skill sets.
B
Exactly, exactly. And that number is growing. Like we probably believe it's probably about half a million short that we are in the US between this year and next year. So the labor shortage is increasing. So that's against a backdrop of increased capital expenditures across many, many sectors as we talked about, and not a lot of use of AI and digital. And so that labor shortage at the moment is going up, up and up, which is a very worrying trend for many, many of my clients and just for many companies in the US. So then you go to your question, right, so that's at the macro level. And so then to your question, so how do companies then compete for labor? There's a few things that companies do. You know, fundamentally you go back to contracting strategy around how do I contract for that labor and what is the relationship that the owner has with the epc, which is typically the company providing the labor. And you know, whether they have their own labor force or they contract it out. But basically. So the relationship and the trust that has been built over years I think is super important because again, these projects at times go, don't go well, which then obviously leads to claims and lawsuits and even companies going bankrupt, which is not uncommon in the EPC sector. The trust and familiarity between executive teams and the relationship that people have and basically building these projects I think is super important. But the craft itself is a pretty mobile labor source. And so to a degree, when people can make higher wages and higher per diems in the LNG facility next door, you got to do a lot to basically keep people on board. And so there's probably something about culture. There's, there's of course something around the non monetary compensation. Right, so here you go to, you know, what are the facilities in the camps like, you know, is there a camp first of all? Or you know, do people basically bring their own mobile homes? What does the meal and food package look like? Other facilities that are just attractive for people to, to stay on a certain job versus go to another job? And on the contracting strategy side, look, it also depends a little bit like who takes the risk? Is the owner taking the risk for the labor rate and labor productivity or is it the EPC or is it a mix? Right where now maybe the owner takes responsibility for the labor rate and the EPC takes responsibility for the labor productivity so that they're sort of like both tied into a joint good outcome and also they allocate risk where it is probably allocated best. And so those are some of the underlying principles in my view, Paul, of how, you know, LNG Facility 1 would compete against LNG Facility 2.
C
The energy and resources sector is experiencing unprecedented change. To help navigate this change and capture its opportunities, HC Group launched Enco Insights, a global advisory network dedicated to the sector providing senior advisors and subject matter experts to investment and infrastructure funds, law firms and corporates. Enco Insights leverages HC groups 20 years of connections in energy and commodities to give clients the expertise they need when the stakes are high and insight matters. Learn more@encoinsights.com there's so much there.
A
So firstly, just talking about sort of the human side here. These are actually many of these roles are quite lucrative. Right. I don't know if you can share sort of any rough idea. You know, we're not talking sort of $50,000 jobs here. If you're an experienced welder in high pressure oil and gas systems. Right. And are these migratory groups, they have a home base somewhere and then they hitch up a pretty nice looking trailer and move around. Like I'm sort of interested in kind of how mobile that talent is and also is that talent transferable as well in terms of suddenly you've got a data center setting up and they need kind of roughly the same skill sets. I might be making that up. They're able to pay another 30% more. They're sort of in intra competition for talent, an inter sector competition for talent. It might be sort of strange and think about this in the European context, But certainly the U.S. context, that person might live in Colorado, but is spending four months, six months at a time at a various job sites but you know, live in relative luxury in a pretty nice looking trailer type thing.
B
So first of all, you're absolutely right. Like, so these are, these are five digit, five digit roles. Jobs indeed. Well paid jobs. There's typically a bit of a split between folks that do want to travel versus folks that, you know, prefer to stay closer to home and in the region. Right. So you know, so some folks will, you know, ideally just want to work in the US Gulf coast and then, you know, some folks, you know, might want to travel to the Midwest for a stint again obviously depending on where the jobs are. But again at the moment all regions across the US there's this massive build outs again across all these asset types. And so you probably would see a little bit less travel because you know, again there's just jobs around, so why would you travel? And then I think you made other very interesting point around, you know, are these the competition for these crafts here? It is important to think about, you know, the different crafts. And so for example process facilities like petrochemicals or lng, you know, there's a lot of pipe and so, you know, a lot of welding, a lot of pipe fitters, which is just a little bit different than a data center, but a lot of electricians. And so what we saw is that the, for example, the need for electricians has gone up substantially. And we have seen folks obviously go to work for construction projects on these data centers which then puts a lot of constraints on the more traditional process facility. Again whether it's refining or pet cam or LNG on the Gulf Coast. And so you do see quite fierce competition between sectors and companies for specific crafts. Again just depending on what crafts are needed for, with for its job.
A
Just coming to. I want to give some sense of the impact this is having on wage inflation and so forth. But just before we get there, how much of this is a. Because you've got two challenging processes going on at the moment. One of course is these skill sets weren't in demand as much 20 years ago, ET cetera, et cetera. You've kind of got this great retirement going on which is people in their 50s and 60s, retiring, 70s, retiring with, with relatively rare specialized skill sets. Those retirements are being accelerated because the last five years earnings have been excellent for these people. And why carry on with the grueling schedule and the dangerous work if you've got enough? There's other inflation going on that might counterfail that. At the same time, of course you've got a global competition for these skill sets. And certainly in the US immigration has been curtailed. More challenging in general. It's also frankly less desirable compared to other jurisdictions. Me saying it like you can you just, you know, how much are those two trends exacerbating this shortage?
B
Retirement is a real thing. And then also the labor pool shrinking is a real thing. So I think both, both trends you're mentioning poor, I think spot on. Obviously construction is not the only sector that's going to struggle with retirements. Right. There's a lot of like frontline manufacturing jobs and, and people that are going to retire. Right. Just that are just working and operating all these facilities where you have a Bit of a similar issue going on. So it's not necessarily was it just contained to construction, but it is a big issue. And so then the question for many companies becomes twofold. One, you know, how do I attract talent? But two, which I think is an interesting one as well, how do I keep some of the knowledge and just the knowledge corpus that those people have and how do I try to retain that? And I'm sure we'll talk in a little bit about digital and AI, but I think that's another interesting trend where companies are trying to figure out, look, how do I retain some of the knowledge and experience and just like the ways of working while the, you know, the physical human beings might, might be retiring, basically. So those are the two things that, that companies are pushing on.
A
And how is all this manifesting? Are we seeing, you know, 20% inflation in some of these key skill sets? Are we seeing a free labor force that's, you know, contractually can move within a week from any job? So the, the nature of contracts is changing and penalties or all this. I mean there must be a reaction here both to win talent in that competition, but also retain talent contractually in that competition because, and to sharpen the argument, you know, if you're getting a weaker team and your schedule suddenly is out as a, as a, well, you tell me, can we see it manifest in, you know, you do not get the best talent, it has an immediate impact. So two questions. One would be how are we seeing that manifest contracts and compensation? And then secondly, allied to that, is it a material impact if you do not get the, the best talent available? Can we see that play out? Are you advising clients? Hey, look, if you do not tackle this element, you can see delays in your schedule of six months to a year, which has this economic impact.
B
Yeah. So on the first question on, on sort like wages and per diems, the answer is yes, absolutely. We're seeing very clearly wages and per diem going up because you know, these people are in demand. And so going accomplishers have to pay more to attract top talent or sometimes any talent. Then as I mentioned a little bit earlier, there's then also the non monetary, the non monetary element to it, right. Which is just distance to the side, deal availability, additional facilities, you know, whether it's a gym or these types of facilities to just attract and keep talent, that's clearly happening. And then to the other question around, is there a big difference between sort of like, you know, your A players, your B players and your C players? Absolutely. And, and so we Talk a lot with our clients around. You know, what do you have to do to attract and keep A talent? Because to your point, Paul, if you don't get the A players, productivity will suffer. And that of course only, you know, leads to increased cost and expanded schedules which then again leads to, you know, so like lost revenue that you'll never get back. And so now then your NPV starts to struggle, your IRR starts to hurt. And so there's a lot of, a lot of bad implications, if you will, by not getting the A players on that part of your team.
A
And I kind of know this too a little bit from my wife's role in upstream oil and gas. Right. This sort of the, the talent team at these EPCs must be super sharp and on their game and. Yeah, I can imagine it's a challenge, right?
B
Well, exactly. And Paul, if I can add one thing on that one, like for the first time in a long time, we're actually seeing EPCs making pretty decent returns. Right. So they're actually having so like double digit returns. And if you look at their, just their P and ls, which again is another clear sign of, you know, this is a lot of demand for EPC capacity and that's getting priced in. Right. And people have to pay.
A
Yeah, yeah. Okay. Right. So we, you know, this is having, this is a story right now that is having an absolute impact and for positives. Right. It's nice I think frankly to see labor in demand and these skill sets, you know, in a world of AI actually there's some things that can't be aided away. Right. You're still going to need some of these core engineering skill sets and kids facing college thinking about what they're actually going to do at this time is, is a big challenge. But they're obviously, this is the, this is the free market. You know, the market will start solving for these challenges. And we've got lots of new, we're in a new world in terms of technology, material science, engineering and all the rest of it to address some of these challenges. And some of them are going away. Right. There is a retirement avalanche coming that people need to prepare for. And we have not been spending lots of money on technical colleges and all the rest of it like we did 100 years ago, whatever it might be. How are you helping clients on the EPC side, on the, on the owner side, think about, and I guess build some robustness against these challenges.
B
So Paul, this is a great question. So we're tracking about 3,000 companies and technologies in the Construction tech space. And so we have a pretty good feel for, you know, what's out there, what's people trying to experiment with, you know, what's working and what's, what's not working. All right, but a lot of activity in the, in the tech construction space and we probably see this grow, you know, 12% year over year for the next foreseeable future. And so there's a lot of activity arguably from a low base. As I mentioned earlier, construction industry is not known for adopting radically new ways of working. And the adoption of digital and AI in construction is, is arguably very low in the construction sector compared to other sectors. So, you know, low base, a lot of activity. And so I'm personally a big believer that given all the challenges we talked about before, digital AI has to be part of the solution. And so rethinking how work gets done, infusing digital AI capabilities into the process, to me has to be part of the solution. We just have to find smarter, more efficient, more effective ways of developing and delivering these capital projects.
A
On the design bit, it's as you say, we have not. Lots of things are done the same way they have been for quite some time is this idea. And I come back to the fourth generation nukes where actually the big change would be using salt halides that don't need a thousand atmospheres of pressure. So designing things that are less complicated, for example, need less welding. Is that kind of, that element?
B
Yes, so it's absolutely part of it. Right. And so my view is across the value chain of a capital project and so all the way from engineering, procurement, contracting, construction, commissioning, there's opportunities to again infuse different parts of that with digital and AI. And so look, you mentioned the engineering part. So we're serving a bunch of our clients on what I would call agentic value engineering, which is quite simply put, how do we reuse previous work and previous engineering work in a much more efficient, effective way as well as once we have a design, how do we quickly optimize that design? Many Companies and many EPCs have value engineering capabilities. But now reimagine that in a world of agentic, where you can basically run through various options so much quicker if you basically have the. So like all the work that you've done before stored in some central tool or central system. And so imagine a world where you can basically pull on all the previous engineering work you've done before to basically first of all get to a design much faster, but then also learn from all the things that worked and didn't work to basically optimize the design going forward. And again, this might sound like a pretty obvious thing and with today's technology it is becoming more and more obvious, but what we're seeing is that it's actually not that obvious and it's really not that trivial. Because look, to a degree every project is different, which is true. But you know, engineers like to engineer and so if there are ways to leverage what we've done before in a much smarter, effective way, that basically is yielding tremendous opportunities on the engineering side to deliver projects, engineer projects basically much faster and, and much cheaper, which to me is actually super exciting because again we go back to what we said before the next five to 10 years, we're seeing much more demand for CapEx, so we need much more engineering capacity. We're seeing people retire. And so if you're the company that figures out how to, you know, quite honestly, let's say you can engineer 50% faster and 50% better and, and with 50% less mistakes, you get so much more work done with the same labor pool. So to me it's a tremendous opportunity for companies to go after.
A
Let me see if I can do the talent one, which I would say that a lot of these trades are for better or for worse sort of guild work, apprentice based in some ways I don't like. We go massively off piste here and I'm definitely saying this, not you like, like doctors, right, which is you have a specialized set of knowledge that is constrained in terms of numbers of people that can attain it through tough certifications that are warranted all the rest of it. But it is, it's guild work. It's two years of apprenticeship to become an electrician in the U.S. i mean quite challenging to do especially when you're on very low incomes during that apprenticeship work. I can imagine a world where a. I think there's a role for government here in terms of recognizing that we should. Countries probably want more engineering and well, want more trades than they do, you know, biology majors like myself generally speaking. But also there seems to be a massive opportunity in terms of the ability to train and coach talent in these new skills. Especially in a world where you can do lots of simulations. That means that you don't have to rely on the job. Quite dangerous trading, if you get it wrong to being lab based, you know, repeat 10 times using sort of AI gloves types. I mean to me would seem a really interesting and useful way of beefing up that skill set in the US whilst maintaining its value as well, yeah, absolutely true.
B
And look, when you think about job sites, people have talked and are doing a lot of like off site modular fabrication. All right, so, and that's not a new trend that probably is going to only increase. There's more and more talk about robotics in construction. And so there's definitely, I think new technologies out there that are going to take some of like very complex, dangerous jobs away from the job site. But again to your point, there'll be plenty of real work to be done for a long, long, long time, right? Installation, fabrication, the, you know, obviously the welding, the electrical work that gets, that needs to get done and all the interfaces. And so I think companies and sectors quite honestly figuring out, look, how do I make those jobs more and more attractive? And obviously pay will help. These are well paid jobs, very respectable jobs, very honorable jobs. And so to your point, you know, sectors, government jointly working to figure out how do we attract the right talent to the schools to then get into the labor force, I think is super important. And to be honest, particularly with digital and AI, I think it's getting more and more exciting because you know, I think there's a real learning journey that companies and people can go through to make the jobs even more exciting than what they are today and just make it attractive. Also from a so like a learning and development perspective for people to just learn new techniques and learn new tools to just be more efficient in their jobs.
A
I think it comes down to the same rethinking. You've still got kind of certification structures that are 100 years old effectively, right? In a world where actually you and I, on a personal basis, you know, you've got YouTube now and you can, you can fix the dryer if you've got an afternoon available and a decent YouTube, right connection. And so taking some of that opportunity and learning and recognizing that like you say, these are very well paid jobs and actually in many ways much more secure than your average service job in a world of AI. And it's just, I think it's probably more a case of education and availability and breaking down some of the rooted structures of essentially guild type behaviors that have been, and this is me again, but I feel quite strongly on that, which doesn't necessarily denote then lapse in standards but just more accessible for people to get on these programmes, leveraging more technologies and perhaps less tightly controlling the numbers. The quantum of people that can have these types of roles which historically have been done to maintain value and all the rest of it as well as safety and so forth. The issue is so stark right now. The challenges are so stark, the next person to build the liquefaction plant stands a much better chance of economic success than the one who does in two years time. Are companies today, are they recognizing the starkness of the challenge? I know you're already seeing a radical rethinking of kind of like how we approach our internal engineers, how we're collaborating with our EPCs, how we're engaging with local universities, like how, you know, we've got to move the needle here and if someone can come up with a better design, we're all ears type stuff
B
in my own client service. And what we're seeing as a practice serving clients both on EPC and the owner side on construction projects is that there's just a massive uptick in the work that we're doing, at least in rethinking how work gets done, both developing and delivering projects. As I mentioned using digital NI capabilities, a lot of companies are starting to think through specific solutions. Let's take a solution like generative scheduling where you basically put an AI next to your scheduling department to really think through. Look, what are all the permutations to basically optimize a schedule? And you know, these projects have typically hundreds of thousands of line items in their P6 schedule. And so it's very complicated to basically plan and schedule for 5, 10, 15,000 workers and do that all. And so there's some very cool new technologies out there that we're working with that a company like Alice, for example, that just really help plan and schedule jobs much better. Because what we're seeing is people want to work, but if they're not assigned a certain job, they're not as productive as probably they could or should be. And so improving scheduling is a huge winner, if you will. A lot of companies are starting to really push the boundaries and just, you know, improving their scheduling capabilities. So that's an example of a bit of a point solution that again we're seeing a lot of traction with. And there's many other examples that I could give you on that. I think the more forward leading companies are starting to really rethink entire domains. And so we talked about this engineering capability, for example, or how do I think end to end from a contracting perspective. And so companies that are really starting to rethink end to end domain and how to infuse digital analytics across an entire domain is really where we're starting to see leading players trying to push, which I think is really exciting because that is really where we believe like the largest gains can be achieved if you really start to rethink, you know, again, an end to end domain, if you will.
A
Yeah, fascinating. And I guess as well you do have kind of a consolidation, you know, in oil and gas. Right. There's only a few EPCs out there. They're working across multiple projects. That does sort of give you the capacity to sort of optimize that schedule a little bit more, capture that knowledge. I mean, there is, there's a lot, as you say, there's a lot of opportunity out there in a world where we can see demand stretched out for years to come.
B
Exactly. Right, yeah. And so we're seeing. What's actually interesting, Paul, is that we're actually seeing the EPCs like rethinking how work should get done. But the owners equally are pushing hard on this. So there's probably certain areas where we actually see the owners pushing harder than the EPCs and really starting to push the EPCs forward because it is of course also fundamentally disrupting a little bit some of the EPC internal workflows. And so I think that interaction between, you know, the EPCs are pushing certain things, the owners are pushing certain things, and together they're pushing each other, I think is very healthy because ultimately the collaboration between owner and EPC is so critical to just deliver these projects properly and, you know, again, on schedule and on budget.
A
It's been fascinating when, and I think you sort of alluded to some of this, but basically when should people call you and your group at McKinsey?
B
Yeah, so look, our group, we work across the entire life cycle of projects, starting with helping optimize business cases, helping do value engineering, we're helping with procurement, we're doing contracting strategy, all the way into creating transparency, how projects are actually performing, the interaction between owner and epc, and all the way into construction and commissioning. As we talked about at the end here, we're doing a lot of work really infusing digital and analytical capabilities into the process. We're helping optimize cost and particular also schedule. Both, of course, are critically important to deliver these projects. And look, our group, we have about 400 people globally that all they do is capital projects. So it's a sizable group. I think we touch about 3 to 400 billion worth of capital projects every year. And so we're quite proud about our capabilities. Actually. We bought our own estimating firm and we've got a whole bunch of analytical capabilities in our team. And so I always talk to my clients around, look, all these projects go off the rails. So why don't we collaborate upfront to avoid projects go off the rails? At the same time, we do also get called in right. When projects are a little bit off the rails to just try to see what can we do to recover again, either on cost and, or schedule. And so we're very proud about the client base that we serve. I think we have massive impact with
A
our clients and I think quite proud to say you're based in my hometown, which I think says all we need to say about, about Houston and being at the center of this story. Right, Texas.
B
That's right, Paul. Exactly.
A
Well, Erica's I wish whichever team you're going to support in the World cup, all the best.
B
Oh, it'll be the Netherlands. Paul.
A
Yeah, there might be some battles in my household or personal soul things when it comes to what we're going to support here. I really appreciate you coming on. I think it's a fascinating story, obviously, just scratching the very veneer of the surface in terms of the expertise that you guys bring. But hopefully we can have you back on in a, in a year or two and, and see where this story plays out because I think all of these trends you've highlighted are going to get sharper and faster. Much more than we think.
B
Yeah. Thank you for having me on, Paul. It's been a true honor. I really enjoyed the conversation and yeah, we'll see each other soon.
A
Thank you for listening. To find out more about HC Group, our global offices and our expertise in search within the commodities sector, please visit www.hcgroup.
Host: Paul Chapman (HC Group)
Guest: Erikhans Kok (Senior Partner, McKinsey & Co., Capital Excellence Practice)
Date: June 2, 2026
This episode explores the dramatic, global surge in capital expenditures (CapEx) across various sectors—a shift from the "asset-light, infinitely scalable" digital economy to a new era of "asset-heavy" investment in real-world infrastructure. Host Paul Chapman discusses with Erikhans Kok the underlying drivers, sectoral impacts, financial dynamics, labor and productivity challenges, and how technology and new approaches to talent are reshaping the landscape. The discussion offers macro and micro perspectives, showing why CapEx is at the heart of current economic transformation and why it matters so much to the commodities sector.
“Almost every sector is going to spend more and more capital...that massive increase in CapEx is making competition for labor, for equipment, for materials ever more, and that’s making basically the returns difficult to achieve.”
— Erikhans Kok [02:10]
Global needs: Energy (e.g., gas-fired plants for data centers), electrification, telecom, life sciences, and infrastructure all require huge new investments ([04:58]).
Geopolitical competition: Investments reflect a "race" for technological and energy independence, with the US, China, and Middle Eastern countries vying for leadership.
Data Centers: Hyperscalers (big tech firms) and private capital are aggressively building data centers, driving power demand and creating knock-on effects in utilities and grid infrastructure.
"There's even a bit of a geopolitical element to this...companies and politicians want to sort of win that race."
— Erikhans Kok [04:58]
Long-overdue Upgrades: Transmission and distribution grids, outdated in places, are getting attention after decades of underinvestment.
The desire for supply security, on-shoring, and deglobalization (e.g., bringing manufacturing or rare earths production home) plays some role, but fundamentals and opportunity for returns are the prime motivators ([09:36]).
"...the majority...is still just, you know, raw fundamentals and just people basically seeing opportunities to put capital in the ground to get a return." — Erikhans Kok [09:36]
Globally, $86 trillion will be invested over the next five years ([11:36]).
The top 20 pharma companies will spend $500 billion between the US and Europe.
Individual companies are now spending $50–80 billion annually—much higher than historical precedents.
"We're talking annual numbers...companies spending $60, $70, $80 billion per year, which is just staggering."
— Erikhans Kok [11:36]
Wages and per diems for skilled labor are rising; competition isn't just on pay, but also facilities and working conditions ([32:17]).
EPC (Engineering, Procurement, Construction) firms are finally seeing strong returns due to demand ([32:29]).
"For the first time in a long time, we're actually seeing EPCs making pretty decent returns...double digit returns." — Erikhans Kok [32:29]
These are highly paid jobs, but modalities (travel, site amenities, migration of talent) are shifting as demand intensifies ([26:39]).
Adoption is late but rising. Out of 3,000 construction tech companies tracked, activity is growing 12% year-over-year, but from a low base ([33:55]).
Opportunities exist across the value chain—from design and engineering (leveraging re-usable designs and generative AI), to procurement, scheduling, and modularization ([35:30], [42:02]).
Example:
"Companies that figure out how to engineer 50% faster...get so much more work done with the same labor pool."
— Erikhans Kok [35:30]
Advances include agentic value engineering (reuse of prior work and AI-powered design optimization) and generative scheduling (AI-driven planning of complex projects with thousands of schedule steps).
Apprenticeship and guild models are bottlenecks—modernizing certification and training, possibly using AI- and simulation-based modules, could broaden access ([37:47], [40:27]).
"Countries probably want more engineering and trades than they do, you know, biology majors like myself generally speaking."
— Paul Chapman [37:47]
Sector and government collaboration is needed to make these roles enticing, accessible, and adaptable.
Both owners and EPCs are rethinking workflows and structure:
"We're seeing EPCs like rethinking how work should get done. But the owners equally are pushing hard on this...and together they're pushing each other."
— Erikhans Kok [44:27]
The companies that can adapt, innovate, and secure talent soonest will be positioned for outsized returns as the CapEx boom continues.
On the paradigm shift:
"Infinitely scalable and asset-light companies are spending more in CapEx...than any oil and gas company would dare to do with its shareholders."
— Paul Chapman [16:12]
On construction productivity:
"Construction is a sector that has quite honestly struggled to really increase productivity...Compared to, say, a 3% per year improvement in manufacturing, it's about 0.4% in construction."
— Erikhans Kok [18:00]
On the stakes of labor quality:
"If you don’t get the A players, productivity will suffer. That...leads to increased cost and expanded schedules which then again leads to...lost revenue that you'll never get back."
— Erikhans Kok [31:07]
On knowledge loss:
"Retirement is a real thing. And also the labor pool shrinking is a real thing."
— Erikhans Kok [29:08]
On the upside for talent and future workforce:
"These are well-paid jobs, very respectable jobs, very honorable jobs...with digital and AI, it’s getting more and more exciting."
— Erikhans Kok [39:03]
On the growing impact of technology:
"Digital and AI has to be part of the solution…if you can engineer 50% faster and better, you get so much more work done with the same labor pool."
— Erikhans Kok [35:30]
This episode vividly illustrates a once-in-a-generation transformation in global CapEx—with consequences for commodities, labor, technology adoption, and organizational strategy. The conversation is both a clarion call and a practical guide for stakeholders navigating the challenges and seeking the massive opportunities driven by the surge in real-world investment.
For more information on Paul Chapman, HC Commodities Podcast, or guest Erikhans Kok, visit the episode links or their respective platforms.