Infratil: Bigger bets, cleaner cuts

I'm Brooke Roberts, one of the co-founders and co-seos of Shareses.

Today, we are joined by Jason Boyes, the CEO of Vroomvitol.

We're going to be talking about data centers with their investment in CDC, Renewable Energy,

their $1 billion divestment plan, InfraTools portfolio and a whole lot more.

But before we get started, here's some important information.

Investing involves risk. You might lose the money you start with.

We recommend talking to a licensed financial advisor.

We also recommend reading product disclosure documents before deciding to invest.

Everything you're about to see and hear is current at the time of recording.

Well, Kia ora, welcome. Thanks for coming on Sheed Lunch and chatting to us all things.

InfraTools has got such a wide-ranging portfolio.

So we'll dive deeply into them, starting with CDC, the data center side of things.

But before we do, really intrigued on what is InfraTools view in terms of what's happening in the world at the moment

and what do you think invasive should be really thinking about?

From an InfraTools perspective, we don't really use a lot of liquid fuel in the individual businesses.

So we're reasonably insulated there.

The airport obviously has an exposure through New Zealand, which is doing it really tough for us all,

like critical infrastructure, right? So we will see a bit of an impact there.

When you look out at the world there, the thing that's moving a lot that really,

we think about a lot as long interest rates.

So 10-year bond rates in Australia, the US, and New Zealand have reacted really strongly.

You know, you sort of mid to high falls in a lot of those cases, which is extremely impactful.

Over a six-month period, you're talking about 50 basis points, you know, half a percent.

On a long-dated set of cash flows, which is what infrastructure is,

that can move your valuation by like a billion dollars.

It's just a lot, right? It's a dollar or a share.

And I feel like if you've been watching our share price over the last three months,

the market really anticipated that quite early on, actually, in January,

we all came back quite bad inflation prints out of Australia and New Zealand,

and then obviously the war is kind of compounded it.

And in some ways, the market was a little bit right, I think, in terms of,

I think it was about a billion that came out as people anticipated those long rates.

Have they gone too far?

I think there's a strong case that they are anticipating quite a bad outcome, a stack-flation outcome,

where you've got no growth to offset higher interest rates and higher inflation.

It's not a zero risk, though, so you, you know, they might be right, they might be wrong.

But then on the other side, we are seeing, and you might see this as well in your flows,

as a real interest in defensive stocks, places to hide.

And I think we've definitely seen that in our share price, which has been quite resilient

since that kind of interest rate anticipation and good growth over the last week.

So, you know, places where you've got steady cash flows that aren't super exposed

to kind of growth up or down.

In the past, those have been a little bit boring for people, right?

But they've definitely seen a real comeback in the market on those sorts of things.

And then, I guess, lastly, actually, when we created this portfolio, 10 years ago now,

we were anticipating a reasonably low growth world,

and we were really trying to find places to hide, which had really resilient growth and strong downside protection.

So, again, with that kind of defensive lens in mind,

I feel like digital infrastructure has some really strong tailwinds, renewable energies

getting a kind of another shot in the arm through this.

Those elements of the business feel like they will do well, and a kind of messy world.

Yes, so a lot on your mind.

How do you manage it all, you know, and I'm just even thinking about your treasury.

Yeah, yeah.

Yeah, there's a lot of people in Morris, and there's 200 people.

There's a treasury team who are really busy.

Yeah.

And a strategy team, and then there are people who are teams that are across their individual assets.

And a lot of the job is receiving, coaxing out of them,

all their kind of current views, and then trying to merge it into twists or total changes in strategy.

So, if we dive into one part of the strategy, which is data centers,

you know, your own infertile, own CDC, about $8 billion, I believe.

When you last talked to us, owned around a 40% stake now, around a 50% stake.

Why do you have that strategy?

So, we own just under 50%, we have more than that, a bit more than 50% in voting terms.

We bought a little bit at the start of last year, as you say.

But really kind of more interesting is it is one of our biggest weights in our portfolio.

We have 40% or so off the portfolio invested in CDC data centers.

And it's a massive investment theme for us.

And we didn't do that overnight.

This is something that's emerged over the 10 years that we've held in that business.

And our approach really is to stay really close to the assets, particularly the big one.

So, we've been sitting on the board, we've been working with the management team,

we've seen how they developed over time, as well as reading externally all the stuff everyone reads about the growth in data centers.

And we're still incredibly high conviction, obviously with this way, in the portfolio, on the business.

It's kind of gone through two or three waves in our ownership time.

It was actually originally designed to take advantage of outsourcing of data centers from government.

That was good, but it got a massive shot in the end when cloud came along.

And you know, that was probably tailing off a little bit two years ago.

And then AI and AI Cloud is now really driving that whole business

on still fundamentally incredibly attractive terms.

And in a business that we believe has some real structural advantages to capture more than its fair share of that growth.

And what are some of those structural advantages?

Yeah, we talked about a bit of that in Sydney last week. Greg, I think outlined really well,

where he sees the advantage of the business today in a kind of AI-focused market for data centers.

The two that really stand out to me are one, it's capital structure.

So because of its mix of businesses, very high credit quality oriented, long-term contracted.

He often talks about he's never lost a customer.

That creates a kind of credit profile that enables them to invest earlier in land and power

in the right locations to them win contracts later on.

And that is proving to be incredibly important in this phase where the builds have got bigger and bigger.

And the contracts have got bigger and bigger.

So to buy a decent plot of land in and around Sydney, you could be talking for $500 million Aussie.

And then to develop the power infrastructure, that's $100, $200 million more as well,

before you can then go out and really win a contract in the market.

And that's capital structure is proving, I think, to be a very strong advantage.

And the other one which stands out to me anyway is what he calls this closed-loop liquid cooling.

And so data centers in general, certainly talking to my 10-ages kids, AI in general,

I think it's a bad rap and rightly so in some ways for using water.

And that's data centers that evaporate water to more efficiently call the computing infrastructure inside.

CDC does not use that technology. It uses a different technology where you fill up the data center once

and the same water just gets reticulated around in a loop.

And it's chilled using chillers on the roof and a whole heap of other smarts.

Before it goes down into the data center, so it's not actually using water.

And for a bunch of kind of, I think, happenstance reasons.

That is what CDC's always done.

I think that's going to prove to be an incredibly big advantage in terms of getting development approvals,

community acceptance and all those things, which are really important because these structures are really big.

So that's a real advantage for them as well.

Yeah, I mean, making that bet 10 years ago, that would have seemed quite futuristic really.

And now, you know, obviously that's really, you can see the trajectory of AI.

How do you assess the future demand of AI products?

I wake up in the morning thinking about CDC and AI and I go to sleep and I think about it.

It's like one of the biggest cause will probably make in our collective investment careers.

And we listen to what Greg tells us the customers are taking.

As I say, Morrison has 200 people all around the world investing in infrastructure

and a lot of them are spending time thinking about data centers, thinking about fiber,

thinking about the implications of AI.

And so you've got a lot of, I guess, inputs coming through on which to form a kind of investment view.

I would say it's been a little bit easier actually over this last period

when I think the market in general, capital markets in general have been quite skeptical

and I think in some ways undervaluing the potential benefits of AI.

Even when a genetic really came along and we could see the implications of it for one NZ, for example,

which is another great way we get on the ground views of exactly what AI can do for existing businesses.

We could see how much that changed the whole equation, not just humans calling AI,

but agents calling the APIs.

The amount of extra flow you are going to get through these data centers from that

which is obviously massively more than people were able to actually comprehend.

And so through this period, I feel like it's been a little bit easier just to take that call.

Maybe it's not been that popular, but then we can monitor through one NZ.

Does it actually turn into money?

You know, this whole ROI thing and they're like, yeah, really does.

We're getting three X return on these agent investments if we do it right.

It's hard actually to get that to the bottom line because as you probably know in your business right,

if you get 20% more efficient, you can't just change a person or move them onto something else.

You've kind of got to get to a whole times more efficient to actually see that,

but the returns are really there.

I think it could get hard again if the market runs again and the way ahead of where we think logically you can get to,

then I think that's almost a harder call to makers actually should be harvesting some from this

because actually people have got to get it themselves.

A lot of the major providers of AI technology are running at a loss

and not really clear signed profitability.

How does that plan how you think about those customers in the demand going forward?

Think about it a lot.

So you're talking about the labs, right, the Anthropics and OpenAI.

Again, when you're seeing what the technology can do, say at one,

or even talking, we were in Asia two weeks ago talking to fund managers up there

who are implementing this in their own businesses, astonished by how much value they're getting out of them.

I think what people find it hard to imagine is, you know,

what is a Bloomberg terminal cost like $10,000 a year, maybe 20?

What if your Claude subscription costs that much?

Because that actually provides the same amount of value.

Then you can suddenly see how, you know, that's 20X,

what you pay for Claude today, OpenAI today, right, professional.

Then you can see, I think, how the economic could actually work.

When you're actually getting this massive shift of value to the model providers,

now some of that will need to go to infrastructure, et cetera, et cetera.

But when you can see those, the potential for those subscription numbers to change wildly,

then I think you can understand actually these people can make money out of it,

actually quite a lot of money out of it.

It comes from other people probably.

But I think the more and more we see that, the more and more we get convinced

that people will really pay for the value that these models or tokens

and the Nvidia guys talk about provide to them, that's how you get confidence, I think.

Closer to home, you know, in Australia and New Zealand,

we're seeing rising fuel prices, hearing a lot more rhetoric about moving to electronic vehicles.

You know, you've obviously got big exposure to renewables.

And last time we talked, you spoke a lot about renewables and demand for solar,

in particular in the US, too, I believe.

How do you see the conflict in the Middle East?

How are you thinking about that when it comes to your renewable exposure?

Yeah, it's hard.

These things are quite complex. It's really hard to predict.

So maybe, and I think it's different in different regions.

So in the US, it's fundamentally very strong,

even though the rhetoric from some in central government is quite anti-renewables in some ways,

actually in fact, they preserve the tax credits, the subsidies that they have there until 2030.

So, and now that they're the most subsidies anywhere,

that anyone offers anywhere in the world for renewable energy development.

So there's this real period now, the industry just talks about filling the box,

building as much as you can in this period,

and while the power's never been cheaper,

it's come at a great time because at the same time,

demand for new energy in the US is really strong.

So we see that playing out pretty much as we were to talk about last time in quite a strong way,

which is good.

But then elsewhere in the world, we're seeing returns from kind of middle of the road

or kind of standard renewable energy projects is actually getting quite compressed.

So to give you an example in Europe,

you would have thought sort of out of your crane war,

which is obviously still ongoing,

and the energy security issues that created for Europe

that there would be more stimulus there for renewable energy.

In fact, you know, the same public dollar can only go so far,

and it's being stretched to building gas infrastructure of a backup,

and it's being spent on, you know, defense, which is usually strategic as well.

And so, renewables, while it is still a priority in Europe,

you don't get the same red or red at all.

The amount of public money that's available to go to there is less effectively

because you've got these other things that are really important as well.

And it's hard to know when that will change,

and then in Asia,

where we have a small business,

they're definitely feeling the impacts more directly and just about anywhere in the world

with this current war.

So I feel like it will get a good stimulus there,

but I think given what we've seen in Europe,

we're probably going to take a bit more of a weight and CNB very selective,

I think, on what we expect to get underway.

There was a regional energy security forum,

I think two weeks ago,

between Southeast Asian countries, New Zealand was part of that,

where there was a lot of talk about,

for example, Indonesia being able to export solar power into Singapore,

which is a project we have on the go there,

which is good to see that starting to move again,

but also Indonesia itself saying,

well, actually maybe we should have some of this power for ourselves,

because we're as exposed as anyone on this.

So I think it's very complex to figure out exactly where the ball lands

on any given year on these things.

I guess the general trend is fine,

but the short-term effects can be real,

whereas at the moment,

it's just a big green light on,

I think, you know, the US for these sorts of investments.

Talking about the US Senior Exposure to Long Road,

they're in the data center gold rush.

Can you tell us about the 1,000-mile solar project

in the relationship,

how long it has with Meta?

Yeah, so that's a reasonably big,

I think it's about 400 megawatt solar project,

and the kind of capital cost to get these things into construction

is reasonably significant,

so we bought it with a PPA,

so power purchase agreement,

with Meta who was buying basically all the power out of that business,

for a very long period of time,

we've well over 20 years,

in a market that they're using to either power or offset the power

they're taking from the grid for data centers,

and we would have multiple examples of that across the portfolio,

whether it's either Meta or Microsoft or one of these people,

taking it for their own needs,

or the utility in that region

who's buying it,

then basically on selling it.

And so that's a big driver of demand in the market,

but it's not the only driver of demand in the US,

through all this,

and if you watch Nvidia's presentations a couple of weeks ago,

they're all talking about national security,

needing to reassure these kind of critical industries

and not be as exposed as they have been in the past,

and that is definitely driving power demand as well.

Like a robotic factory,

I saw some statistics were up in the years last week,

a robotic factory can use 30 to 40 megawatts of power,

that's what a big data center used to be, right?

And that's just one of these automated battery assembly lines,

I think was the example there,

a massive amount of power, right?

We had the big, beautiful bull,

which was looking actually kind of bad news for Long Road.

It seems to be a bit more optimistic now.

What is your latest outlook on the US policy in its impact?

These were the reforms that were brought through middle of last year,

I think middle to late last year,

that were affecting those tax subsidies

they talked about for US renewable development.

In the end, kind of nobody really lost an eye

because as I said, the subsidy's got extended to 2030.

The solar actually out to 2037,

I think is the date for battery,

which is half your Capix cost now.

So it's created an environment

where everybody kind of needs to rush to build by 2030,

but it's not actually necessarily a bad thing.

And we're already seeing actually signs

with customers of the price they'll be willing to pay

when the tax credits are gone

through Long Road.

So essentially the power price has to go up to keep returns

as they were.

And we're seeing signs that the customers will do that

through the negotiations we're having with them already.

It sort of goes,

if we can't make 2030,

then we're going to need a different power price

and people agreeing to that.

So we're good to 2030

and we're starting to see that the market post 2030

is becoming a bit clearer as well.

Last time we talked about Wellington Airport,

where there's major upgrades underway,

obviously there's more to come there.

How has the revamp turn been all been received?

I think really well.

I just went through it for the first time

with the new relay open the other day.

It looks amazing, right?

It'll be nice when that kind of jet star end,

I think is the next part that is going to get a revamp

because that's pretty tight in there.

It's actually a mixture of like four buildings

so it hasn't been that easy,

but they've got a really great plan there as well.

The tricky thing for airports,

all these infrastructure things is

how can you time your Capix

when it's going to be needed?

And so through this period,

I think we'll get some good signals from airline partners

about exactly what they need when,

if we can shift around Capix to help them.

The other thing we've finished is,

I don't know if people have seen,

at the end of the runways,

there's kind of different coloured parts of the runways

at the end,

and that's called E-Mass.

It's a different type of concrete

that is designed to help decelerate planes

if they're by accident.

They shouldn't.

They shouldn't over-shoot the runway,

which I think is great to know

as a consistent user of that airport.

So that's just been completed,

and that creates,

along with some airplane improvements,

almost the same runway extension,

a usable runway,

that the full runway extension

we were talking about two or three years ago

would have created a real fraction

of the cost just through using this new technology.

So that's been completed,

and we hope to be the beneficiary of that

in Wellington at some time,

with some longer haul aircraft,

hopefully being able to be coming in and out of here again.

Yeah.

That's a fantastic innovation,

and I'm not having to do all of that build,

and you know,

the surface we're getting a bit up.

Yeah, totally right.

I think the brakes,

I don't have to do that anymore,

and you mentioned,

you know,

might be able to do more international routes,

is there any discussions about,

you know, different routes coming?

Constantly.

I mean, always the objective has been,

could we get Singapore coming into North America?

Yeah.

And certainly through this,

the crisis I guess it is,

Singapore is getting a massive uptick and traffic

as the way to get North,

if you don't want to go through North America.

So, yeah,

if we could land that,

anyone out there listening,

it's got a plane.

That would be amazing.

Yeah.

I think for the airport.

So on to infrastructure,

and, you know,

a bit more about New Zealand politics,

the Infrastructure Commission released New Zealand's

first ever national infrastructure plan,

calling for decisive action,

including accelerated electricity investment,

and asset recycling.

Yeah.

Does this plan actually change anything

in terms of your pipeline and investable opportunities,

and do you expect it to result in real change?

So, good question.

I think those elements of the plan made a lot of sense,

and things we would have been advocating for

in a long period of time,

people often ask,

if you had to choose what would you invest in in New Zealand,

I think energy is definitely one,

where any country that is long energy

just has more than enough of,

particularly renewable,

zero marginal cost,

energy is going to do really well in the future,

whether it's these factories,

exporting tokens,

or your own digital infrastructure

for education and healthcare is going to be amazing.

So, I would 100% support that.

I don't think,

I actually think a lot of the ingredients for that

investment already exists,

amongst the existing renewable energy companies,

with transpired zone plans and funding plans.

There's a few tweaks you could make,

but I think the future looks bright

if we can just get on with some of that stuff.

So, that's good.

I think, you know,

recycling of capital probably does make sense.

We can't afford everything we want,

and particularly if you can,

I don't know if it's ring fence or whatever,

that capital for reinvestment and new stuff we need.

I think when you're taking the country forward,

that would be a pretty important requirement for it.

But I wouldn't advocate doing it,

because, you know,

we've got a limited window to do it,

so we should do it,

because if you look back at the history of privatizations,

once they get ahead of steam,

people just go ahead with them.

You don't necessarily get the best number.

And if you look at the way we think about it,

we're always thinking about where to shift the portfolio,

but it's a pretty high bar to just sell something

just because we have to.

Yeah.

Because we've got a limited window to do it.

So, if you can through this infrastructure plan,

create alignment around, you know,

over the next 30 years,

we're going to do this,

rather than we're going to do it next year.

The late next year could be an absolutely horrific market

to sell stuff into,

and I just wouldn't do it just to try

and take a political window to do it.

That would be my kind of little worry to watch out for.

But I think, in general, it's a sensible idea.

Yeah, in a long-term view.

Yeah.

Your credit rating has recently changed to Triple B+.

Can you tell us what that means

to infertile investors?

Yeah, I'm glad you asked the...

So, in December last year,

we announced we've got a credit rating from Standard and Paws,

which is Triple B+.

As you say, we've never had a credit rating before.

In fact, Lloyd might turn in his grave.

He really hated rating agencies.

He thought they were the cause of the global financial crisis,

which I think you could make a decent case for.

But for us, really,

it is about becoming more clearly an institution, you know,

an institutional grade stock,

which goes with our size.

What it means, really, is we can access many more capital markets now

at a materially cheaper cost than we used to be able to.

And actually, why we never had one before really was,

Standard and Paws didn't have a way of rating investment hoarding companies like us.

Yeah.

But they have had one and developed one over the last number of years,

and more and more investment entities are using it.

And so it was a great opportunity, I think,

to kind of respond to people's questions of,

you know, you guys use a lot of debt.

How should I think about that?

We use a reasonable amount of debt.

That's how we get the equity returns.

A lot of that sits in the portfolio companies.

So for example, long road,

they would fund 90% more of their capex to build a solar farm

using other people's money, mainly debt.

And CDC is no different.

But that they will have their own credit ratings.

And then at our level,

we've always carried additional debt

at the holding company level,

which creates a number of benefits, I think, for shareholders.

But then people worry you've got debt on debt.

How do I think about that?

And so trying to make it super simple for people to think about that

through the credit rating.

And then you should see tangible benefits in terms of the cost of our funding.

But also, I think, resilience of funding sources,

because we'll better go to so many more markets to, you know,

refinance debt or extend debt if we need to over time.

And so lastly, I thought we could talk about the $1 billion

divestment plan.

Your halfway to the target of the $1 billion asset sales.

Retire Australia was held for 11 years.

Yeah.

And Immersal took an accounting loss of around 80 million on the sale,

with a lot of management time and capital tied up over those 11 years.

Yeah.

What does that experience tell you about what Immersal should

shouldn't shouldn't own?

And where they're warning signs early in the holding period,

the business wasn't going to scale the way that you might have hoped.

Do you start at the start when we were talking about data centers?

Actually, that was a big bet 10 or 11 years ago.

Well, actually, Retire Australia was made exactly the same year.

And so it was long road exactly the same year.

Anyway, and they were all about $200 to $400 million, I think.

So that class of 2016, we call it.

And so you never know how these things are going to go, right?

When it goes crazy and I'm going to go crazy.

And another one didn't work as well as we as well as we hoped.

I think you can't, you know, if we could all get two out of three in that way,

it would probably be happy.

But I think it's right to try and learn from what worked well, what didn't.

I think for Retire Australia, I might have talked about this in the past.

I think we underestimated how hard it would be to convert existing retirement villages

to the more New Zealand model, which they call a continuity of care model.

You have hospital grade care on site.

It actually took us quite a while.

I think they have actually got a really credible product to do that now,

which is great and part of this story that was marketed when we sold the business last year.

But probably the other learning would be, yeah, it's impossible to catch up

when you're behind a case after 11 years.

So there is a real cut your losses, I guess.

I think if you get behind your way better off to exit,

give somebody else the opportunity to buy and actually do that growth story.

But rather than try and keep pushing on and try and make it work.

So for us, we're probably just like everybody else, right?

So I can fix this, I can fix this, I can do this, I can do that.

But actually the weight of time on your return is absolutely savage.

And you are almost always better to leave earlier.

Leave that, you know, the management team stays in place in the queue with a week on.

Give them a good shareholder who can really fund that growth story

and then just move on to something else who probably are the big lesson.

Yeah, nice.

And then QScan has been put under a strategic review.

And there's more to be cut to achieve that billion dollar in divestment.

Is infertile exiting healthcare together?

Or is it purely a matter of kind of scalability relating to the specific businesses

that are currently in your portfolio?

Yeah, well, healthcare is a theme.

I think there's definitely a scalability element here, which didn't took about last time.

So meanwhile, if, you know, for a tarot Australia, I think kind of went sideways,

maybe slightly down, as you said, there was an accounting loss in the end.

In the meantime, these other businesses sort of did 10X.

And so for them to then do 10X from there is impossible, right?

And it's almost the same for QScan, I think, when we made that investment,

it was about 10% of our market cap, we're about $3 billion then.

Now we're 11 or whatever.

And now I'm still about the same size, right?

So they haven't mounted a key part for all good reasons to do with both their physical capacity, their scale,

and also just the market in general over there.

So I think it makes sense to recycle that capital and have it available.

They're a great business.

Just maybe it doesn't fit an Apple portfolio.

And a little bit learning the lesson from RA, right?

It's like, you can obviously you're welcome in some ways.

And it's like, well, let's move on, give them a good home, and use that capital ourselves.

The other kind of thing that was going on in the background when we set that target was

a lot of questions about how we fund CDC's growth in particular,

and long road a little bit as well, without having to come back to market to raise equity,

which we have had to do in the past as well.

And the market is very intolerant, I think, of stories where you are constantly having to come back to the market.

To raise capital, and you can see a few examples of that Australia and New Zealand at the moment.

So again, in terms of paying for the buck for shareholder,

getting that money back and even just holding it to show you can fund those things

is probably going to make shareholders more money than us just hanging around in there

and, you know, hoping it grows.

So when you look to the year ahead, you know, what excites you?

Pretty excited about the AI side, as you can probably tell.

I think the question mark a little bit out there,

although I think the industry addressed it a bit last year,

is can Australia grow as fast as Asia?

Is it going to get serious interest from US for kind of overflow of AI needs,

which would see the industry industry in New Zealand growing much faster than our natural growth rate?

So we think there's a really great case for that, certainly from the conversations we're having with customers

and being able to reveal that this year would be amazing.

I think everything else around AI is super interesting from an investment perspective as well.

And that's what one NZer doing in terms of changing culture and mindset

and training their staff to use this and then seeing the outputs that is going to be super exciting

and then finding other ways to invest in it around the theme,

I think are a sensible thing to think about given our good position in data centers

and a great position in a business that's transforming itself using this technology is really super exciting.

So yeah, I couldn't say enough about that.

So I've got quick five questions for you.

Do it.

All right. If you had to sell every asset tomorrow, except one, what are your key things?

CDC.

Yeah. Does conflict in the Middle East change your renewable equation for you?

No, slight up wait.

What's the one individual asset you think people are sleeping on?

I think still, I think long road to be honest.

Yeah.

People don't understand that business is very misunderstood, especially amongst institutions outside New Zealand.

They don't really get it.

So it's a big focus for us to explain that.

That's cool.

Hey, thank you so much.

Thank you.

Thanks for having us.

And thank you so much for joining us today.

Thanks, heaps for listening.

You, again, can listen to shared lunch on YouTube or wherever you get your favorite podcasts.

Leave us a rating and tell us what you'd like to hear about next.

Until then, Marcewa.