Thursday, 12 November 2026

Thursday, 12 November 2026

21:09:40

21:09:40

ECONOMICS · MARKETS · Capital

Researching the Economics of Change.

Researching the Economics of Change.

Infrastructure Economics

Can Waikato Buy Time Before a NZ$47.9 Million Grid Upgrade?

Transpower’s Waikato forecast contains an unusual mismatch. In 2030, just three half-hour periods and 7 MW of firm relief could postpone a NZ$47.9 million grid connection. Rokin translates the operator’s deferral estimates into annual payment ceilings and tests how they change as the required service expands. The first year leaves room for flexibility, but longer commitments need increasingly demanding capacity, location and reliability guarantees.

In August 2030, Transpower expects the Waikato 110 kV network to exceed its N-1 interconnection limit for just three half-hour periods. The largest forecast shortfall is 7 MW. Those figures sit behind a proposal for a new connection at Hamilton East, with an estimated capital cost of NZ$47.9 million in real 2026 dollars. [1][2]

Seven megawatts sounds modest until the network constraint is considered. Waikato is supplied through two Hamilton transformers, with local hydro generation providing support. After one transformer fails, the remaining system must still supply customers. A small extra load at the wrong moment can exceed that requirement. The network is being planned around what happens during a credible outage, rather than the average demand on an ordinary day. [1][3]

Transpower prefers to establish a new 220/110 kV connection east of Hamilton. Ōtorohanga remains a shortlisted alternative, and consultation is open until 2 December 2026. The preferred option is not yet an approved investment. [4][5]

Buying one more winter

Transpower’s non-transmission assessment says that 7 MW of firm, correctly located support in 2030 could allow the connection to be delayed by a year. A supplier might provide that support through controllable industrial demand, storage or a combination of resources. The offer would have to respond when the grid needs it, rather than appear available only on paper. [2]

How much could the network pay? Transpower values a one-year delay at NZ$1.2–2.0 million in 2026 present-value terms. The range reflects how much of the project expenditure could still be postponed. Transformers and switchgear have long procurement lead times, so an agreement signed too late may leave little spending to delay. [2]

To compare those benefits with a supplier’s annual availability payment, Rokin divides the cumulative deferral value by the discounted kilowatt-years that would need to be contracted. The calculation uses Transpower’s annual capacity forecasts, a 5% real discount rate and payments at the end of each year. It is a consistent price comparison, rather than a claim about the bids that suppliers would make.



Service period


Required MW at end


Indicative ceiling (NZ$/kW-year)


2030 only


7


208–347


2030–2034


35


68–116


2030–2040


79


34–58

Table 1. Rokin calculations using Transpower’s cumulative deferral values and capacity forecasts [2]. Flat real annual payment, 2026 present values, no reserve buffer or implementation costs. The numbers are indicative ceilings, not market tariffs.


For 2030 alone, the result is roughly NZ$208–347 per kW per year. Keep the service running through 2034 and the equivalent flat rate drops to NZ$68–116. Extending it to 2040 brings it down to NZ$34–58. Each range applies to the entire stated contract period, with capacity rising along Transpower’s annual forecast.


The peaks start to add up

The first winter may be manageable for a small portfolio. By 2034, Transpower expects 35 MW of relief across 55 half-hour periods. In 2035, the model requires 45 MW for the Waikato interconnection constraint and another 7 MW associated with the Valley Spur, while exceedances rise to 190 half-hours. By 2040, the two requirements total 79 MW, with 864 forecast half-hour exceedances. [2]

More contracted capacity is only part of the problem. The heat maps suggest that constrained periods spread beyond the first winter morning peak into evening hours and other months. An industrial customer may face repeated interruptions. A battery may need to recharge between calls. A portfolio able to cover three short peaks in 2030 may struggle with the later pattern even if its nameplate capacity rises.

Location also begins to matter more. From 2035, the Valley Spur has a separate voltage-stability requirement. Adding the two MW forecasts is useful for calculating the total contracted capacity in our screen, but a megawatt offered at the wrong point on the network cannot simply be moved to the Valley Spur. Technical qualification must be done before any commercial comparison means much. [2][3]


What a five-year contract might cost

Suppose a provider asks for NZ$100 per kW-year of availability and the buyer contracts 15% more capacity than the forecast requirement. The present value of payments for 2030 is about NZ$0.66 million. Against Transpower’s one-year deferral value of NZ$1.2–2.0 million, that leaves some room for other expenses.

Extend the same arrangement through 2034 and the bill rises to NZ$9.09 million. Transpower estimates a cumulative five-year benefit of NZ$5.4–9.2 million. The contract would exceed the lower estimate by NZ$3.69 million. Even at the top of the range, only about NZ$0.11 million remains before mobilisation, metering, contract administration and the cost of assuring delivery. [2]

These are chosen prices and buffers, not evidence that such a service can be purchased. The public forecast also does not provide the complete half-hour load profile needed to test response duration and recharge, or a set of firm supplier bids. Transpower is still seeking expressions of interest, with any later procurement dependent on credible offers. [2]


A short window for flexibility

Short-lived flexibility looks plausible while the shortfall is rare. By the later 2030s, the network needs far more capacity across many more peak periods. A delayed project still needs building eventually, and some expenditure may have been committed long before a supplier is ready.

Supplier bids, contract guarantees and the detailed timing of future overloads are not public. Transpower would also have to show that the particular resources offered can maintain N-1 reliability at the right locations. Those checks belong at the beginning of a procurement decision, not after a contract is signed.

Waikato has a narrow opportunity in the early years. Flexibility is worth considering when it can keep the network secure and postpone spending for less than the delay is worth. The growing number of constrained periods makes a multi-year service progressively harder to justify against a permanent connection.

RESEARCH NOTE  /  Rokin Research converts published Transpower forecasts and cumulative deferral values into annual payment ceilings. The 2030–2040 horizons are analytical scenarios, not announced procurement contracts. The model is source-verified and arithmetically reproduced; no supplier offer or technically qualified non-transmission portfolio has been established.


Sources

[1] Transpower, Shortlist overview. pp. 4, 11-14. Official source

[2] Transpower, Attachment 5: Non-transmission solutions. pp. 5-11. Official source

[3] Transpower, Attachment 3: Investment Test. pp. 5, 10-15. Official source

[4] Transpower, WRIC project page. Updated 9 Nov 2026. Official source

[5] Transpower, WRIC shortlist consultation. Consultation opened 21 Oct 2026. Official source

Rokin Research

Key takeaways

The transition’s investable bottlenecks are increasingly physical: connection queues, grid equipment, permitting capacity, and skilled delivery.

1. Seven megawatts could buy one year. Transpower forecasts just three constrained half-hour periods in 2030. Firm flexibility at the right location could defer the Hamilton East connection. 2. The first year's delay has an estimated value of NZ$1.2–2.0 million. Rokin's equivalent availability-payment ceiling is NZ$208–347/kW-year, before delivery and assurance costs. 3. The requirement grows quickly. By 2040, Transpower forecasts 79 MW across two location-specific constraints and 864 half-hour exceedances. A longer contract would need substantially more capacity and reliable delivery.