Infratil: How Lloyd Morrison’s Listed Fund Became a Bet on Data Centres

Infratil is the rare listed company whose product is capital allocation itself: it buys infrastructure early, funds its growth, and sells when others will pay more than it can earn by holding. This article explains where the company came from, how the external-management model works and what it costs, why CDC Data Centres now dominates the portfolio, how One NZ, Longroad Energy and Wellington Airport fit, and what shareholders should watch as capital spending approaches NZ$4bn a year. It is part of the New Zealand Company Stories hub.

How did Infratil begin?

Infratil was founded in Wellington in 1994 by Lloyd Morrison, a former investment banker, as one of the world’s first listed infrastructure funds. It raised money from New Zealand retail investors to buy stakes in electricity, airport and port companies being corporatised or privatised.

The idea was new at the time. Infrastructure assets were then owned almost entirely by governments or utilities; pension funds had not yet created the asset class. Morrison saw that the reforms of the 1980s and 1990s were turning New Zealand’s power boards, airports and ports into companies whose shares could be bought, and that long-lived monopoly assets suited patient capital. Early holdings included Trustpower, a stake in Port of Tauranga and other regional utilities, and in 1998 a 66% stake in Wellington Airport, which Infratil still holds alongside Wellington City Council.

Morrison also set a tone. He was public-spirited, argumentative and a nationalist about capital: he campaigned for local ownership of infrastructure and for a higher savings rate. He died of leukaemia in February 2012. The management firm he founded, H.R.L. Morrison & Co, now simply Morrison, has run Infratil ever since.

How does Infratil make money?

Infratil makes money in two ways: from the growing cash earnings of the businesses it part-owns, and from selling those businesses at a profit. It pays a modest dividend and reinvests most of its capital, so shareholders’ returns come mainly from rising asset values.

The portfolio is deliberately tilted to what the company calls “ideas that matter”: digital infrastructure, renewable energy, healthcare and airports. Infratil prefers platforms that can absorb large amounts of new capital at attractive returns, as opposed to mature assets that pay out steady yields. A data-centre developer with land, power and customers, or a renewables developer with a pipeline of projects, fits the description. A fully built toll road does not.

The second leg is recycling. Infratil has a long record of selling well. It bought Shell’s New Zealand fuel business with the NZ Super Fund in 2010, renamed it Z Energy, floated it in 2013 and was out by 2015. It sold Tilt Renewables in 2021 at a price that reflected the scarcity of wind developers. In July 2025 its majority-owned Manawa Energy was acquired by Contact Energy, a deal described in the Contact Energy story; Infratil took Contact shares as part payment and sold a 5% block in May 2026 for about NZ$495m, keeping roughly 9%.

Not every exit flatters. RetireAustralia, held jointly with the NZ Super Fund since 2014, was sold to Invesco Real Estate for A$845m in a deal agreed in August 2025. Infratil said its return over about 11 years was close to nil. The company’s record depends on a few very large wins outweighing a longer list of average outcomes.

Who owns and manages Infratil?

Infratil is a widely held company listed on the NZX and ASX, with no controlling shareholder. It has no executives of its own in the usual sense: Morrison, a privately owned infrastructure manager, provides the chief executive and investment team under a long-term management agreement.

The arrangement is unusual for a company of this size. Morrison receives a base fee calculated on the value of the company and incentive fees tied to the performance of its investments, particularly the international ones. When assets are revalued upwards, as CDC has been, incentive fees can be very large; they ran to hundreds of millions of dollars in the years of CDC’s sharpest revaluations. The mechanism works in both directions: the weak outcome at RetireAustralia produced a negative incentive fee of about NZ$21m in FY26.

Critics argue that external management creates a conflict, since fees rise with asset values and with the size of the fund. Defenders point to the record: a total shareholder return of 18.9% a year since 1994, by the company’s reckoning. The board, chaired by Alison Gerry, is independent of Morrison and has periodically renegotiated terms. Jason Boyes, a Morrison partner, has been chief executive since 2021, when he succeeded Marko Bogoievski.

The model was tested in December 2020, when AustralianSuper made an unsolicited offer valuing Infratil at about NZ$5.4bn. The board rejected it as too low. With portfolio assets now valued at NZ$20.6bn, that decision looks sound, and it is a case study for the argument, made in the NZX story, that takeovers have drained the local market of its best companies.

Why is CDC Data Centres so important to Infratil?

CDC is Infratil’s largest asset by a wide margin, valued at about NZ$8.9bn for its 49.7% stake at March 2026. It builds and operates high-security data centres in Australia and New Zealand, and its contracted capacity has passed one gigawatt as cloud and AI customers sign long leases.

Infratil bought into Canberra Data Centres in 2016, when it was a specialist landlord for Australian government agencies. The original attraction was security accreditation: federal departments needed sovereign facilities, and CDC had them in Canberra. The customer base then widened to the large cloud providers that serve government, and from 2023 to those building capacity for artificial intelligence. CDC expanded to Sydney, Melbourne and Auckland. Infratil raised NZ$1.15bn of equity in June 2024 largely to fund its share of the build, and lifted its stake slightly in 2025; Australia’s Future Fund and the Commonwealth Superannuation Corporation are the other main owners.

The numbers show a business in a steep build phase. CDC’s EBITDAF was A$393m in FY26. Guidance for FY27 is A$680-720m, and the company forecasts more than A$1bn by FY28 as contracted halls are delivered. Operating capacity stood at 671MW with a further 572MW under construction. Data centres of this kind are leased on long contracts, typically with the customer paying for power, which makes the revenue closer to property rent than to technology sales.

Infratil’s second digital asset, Kao Data in Britain, is much smaller. In New Zealand its data-centre interests sit alongside rivals described in the Spark story, where a telco is selling down the same sort of assets that Infratil is building.

Where does One NZ fit in the portfolio?

One NZ, the former Vodafone New Zealand, is Infratil’s main source of cash. It earned EBITDAF of NZ$609m in FY26, has nearly two million mobile connections, and doubled its distributions to Infratil to NZ$180m. Infratil owns 99.8% of it.

Infratil and Brookfield bought Vodafone’s New Zealand business in 2019 for NZ$3.4bn. In 2023 Infratil bought Brookfield’s half for about NZ$1.8bn and the company was renamed One NZ. Along the way the mobile towers were sold into a separate vehicle, Fortysouth, in which Infratil kept a minority interest that it sold in April 2026 for NZ$217m. The full account of the deal and of the three-player mobile market is in the One NZ and 2degrees article.

The role One NZ plays inside Infratil differs from CDC’s. It is a mature, cash-generating business in a market with little growth: guidance for FY27 is NZ$600-640m, essentially flat. Its value to the group lies in funding. Dividends from One NZ and Wellington Airport help pay Infratil’s own interest and dividend while CDC and Longroad consume capital. A telco is also a plausible candidate for sale or listing if a buyer offers a full price.

What are Longroad Energy and the other holdings?

Longroad Energy is a United States renewables developer in which Infratil owns 42%. It builds and owns solar, wind and battery projects, and its EBITDAF rose 170% to US$121m in FY26. The rest of the portfolio comprises renewables platforms in Asia and Europe, an airport and medical imaging.

Longroad was founded in 2016 by the former management of First Wind, with Infratil and the NZ Super Fund as backers. It had about 3.5GW operating and 2GW under construction at the FY26 result, and aims for run-rate EBITDAF of US$1bn around the end of the decade. Infratil approved a further US$300m commitment in 2026. The risk is political: American tax credits for wind and solar were curtailed in 2025, and Infratil itself has cited regulatory uncertainty there.

The smaller holdings are options of varying promise. Gurin Energy (95%) develops renewables in Asia; Galileo (38%) does the same in Europe. Wellington Airport lifted earnings 2% in FY26 despite airline capacity constraints of the kind described in the Air New Zealand story. RHCNZ (56.8%) and Qscan (59.5%) are radiology groups in New Zealand and Australia, where public funding decisions weigh on returns.

What do the latest numbers show?

For the year to 31 March 2026 Infratil reported proportionate operational EBITDAF of NZ$989m, up 11%, and a net parent surplus of NZ$550m after a loss the year before. Portfolio assets were valued at NZ$20.6bn, up 13%, and the total dividend was 20.9 cents a share.

The company invested about NZ$1.3bn during the year and divested more than NZ$600m, mainly RetireAustralia, Fortysouth and surplus property, with a further NZ$1bn or more of disposals flagged. Net debt at the corporate level was about NZ$3.2bn, with gearing of roughly 24% on the company’s measure. In December 2025 S&P Global assigned Infratil an inaugural BBB+ credit rating, which widens its access to debt markets beyond the retail bonds it has long issued in New Zealand.

The forward numbers are the more striking. Guidance for FY27 proportionate operational EBITDAF is NZ$1.3-1.4bn, with most of the increase coming from CDC. Capital expenditure across the portfolio is guided to NZ$3.8-4.4bn on a proportionate basis. Infratil is, in effect, committing several years of earnings in advance to data centres and American solar farms.

Who competes with Infratil?

Infratil competes for assets with global infrastructure funds such as Brookfield, Macquarie, Stonepeak and KKR, and with pension and sovereign funds investing directly. For investors’ money on the NZX it has no close peer; in data centres its investee competes with NEXTDC, AirTrunk and Equinix.

The competitive position has changed since 1994. Infrastructure is now a crowded asset class, and mature assets are priced to yield little. Infratil’s response has been to take development risk that large funds often avoid: backing management teams to build new capacity, as it did with CDC and Longroad. Its listed structure gives it permanent capital, with no fund life forcing a sale, and New Zealand retail investors and KiwiSaver funds have been steady supporters of its equity raisings.

In Australian data centres the benchmark was set in 2024, when Blackstone agreed to buy AirTrunk at an enterprise value of more than A$24bn. That price supported CDC’s valuation, and also showed how much capital rivals can deploy. Local investors comparing Infratil with utilities should note the difference: a lines company such as Vector earns a regulated return, whereas Infratil’s returns depend on negotiating commercial contracts and on asset prices.

What are the main risks for Infratil shareholders?

The principal risks are concentration in data centres, the scale of funding required, dependence on independent valuations that have not been tested by a sale, the cost and conflicts of external management, and political exposure in American renewables.

CDC’s customers are few and powerful. A handful of cloud companies and government agencies account for most demand, and if the AI investment cycle slows, new halls could be leased more slowly or on weaker terms. Supply of power, equipment and skilled labour constrains delivery, and construction costs have risen. None of this implies that existing contracts are unsafe, but the valuation assumes that growth continues for years.

Funding is the second question. With proportionate capital spending near NZ$4bn a year against earnings of about NZ$1.3bn, Infratil relies on asset-level debt, asset sales and, when needed, new equity. It has been explicit that divestments are part of the plan. That is a reasonable strategy when markets are open and less comfortable when they are not.

What can founders and CFOs learn from Infratil?

The main lesson is that returns come from being early in a theme and then having the capital and patience to keep funding it. Infratil bought CDC when it was a niche Canberra landlord, and its largest gains came from reinvesting for a decade.

1. Back platforms, then feed them. The first cheque matters less than the ability to write the next five. Infratil chooses businesses that can absorb capital at high returns, and structures its balance sheet around that.
2. Sell when the buyer is more optimistic than you are. Z Energy, Tilt and Manawa were sold at or near full prices. Holding every asset for ever is a preference, and good allocators treat it as one.
3. Report the failures. Infratil told shareholders that RetireAustralia returned close to nil. Candour about weak investments earns credibility for the valuations of strong ones.
4. Align the manager, then check the alignment. Incentive fees paid on unrealised valuations need independent directors, published methods and clawback when values fall.
5. Match funding to asset life. Long-dated bonds, an investment-grade rating and asset-level debt reduce the chance that a good asset must be sold at a bad time.

Founders seeking infrastructure-style capital should note what such investors look for: contracted revenue, a clear use for more money, and management that owns equity. The Founders Hub covers capital-raising more broadly.

What happens next for Infratil?

The next phase is about delivery and funding. CDC has to build the capacity it has contracted, Longroad has to navigate American policy, and Infratil has to complete more than NZ$1bn of planned divestments to help pay for both.

Three developments are worth watching. The first is CDC’s ownership: a business of its size is a candidate for a partial sale or listing, either of which would test the valuation in public. The second is what Infratil sells. The remaining Contact shares, the imaging businesses and perhaps part of One NZ are all plausible sources of cash. The third is the management agreement, which attracts more scrutiny each time incentive fees are paid.

New Zealand’s own infrastructure needs, set out in the infrastructure deficit article, offer Infratil fewer opportunities than might be expected: public-private partnerships for roads and prisons earn lower returns than the company targets. Its capital is likely to keep flowing offshore, to data halls in Sydney and solar farms in Arizona. Lloyd Morrison’s fund remains a Wellington company, but most of what it owns is elsewhere.

Frequently Asked Questions