Authors
Authors
Read time: 5 minutes
As the global demand for digital infrastructure accelerates, the data center construction sector is busier than ever. A key area of contention in the negotiation of the construction contracts for these projects is the extent to which the contractor will be liable for the consequences of delays, defects or non-performance.
Managing risk and resilience
This is, inevitably, a topic of concern for both developers and contractors on any construction project: The parties must always negotiate a suitable balance of risk and liability exposure that renders the contract economically viable for both. However, this can be a particularly sensitive and contentious issue for data center construction projects, given the complexities and risk inherent in such projects.
Data centers differ from conventional construction projects in a number of ways. Constructing a data center is not just about building a facility – the systems involved in these projects are technically complex and sensitive, with power infrastructure and advanced cooling systems. Developers and end users will often impose bespoke designs and high performance standards. Small mistakes can have enormous operational and financial consequences – even, for example, if the cost of fixing a defect is relatively low, the resulting costs of any downtime can be substantial, and the losses are unlikely to be covered by insurance policies. From the contractor’s perspective, then, there is significant pressure and high exposure if things go wrong.
The starting point for a developer, naturally, is that the contractor should stand behind its obligations and be liable for the losses that flow as a result – but there is an understanding, particularly in this sector, that it is in no one’s interests to expose a contractor to unrealistically high levels of liability. No contractor would sign such a contract. The liability clauses will, therefore, be fertile ground for negotiation when agreeing on the construction contract.
In data center construction contracts, it is generally understood that a contractor is unlikely to accept full exposure to certain types of losses – most notably, damages associated with loss of use, loss of profit and similar business losses, which are sometimes known as indirect and consequential losses. In the context of critical infrastructure like data centers, these losses can be significant – even crippling – and potentially disproportionate to the contractor’s error. The parties may agree that the contractor’s liability for breach of contract excludes liability for those losses, or perhaps that such liability is capped (with such cap likely to be framed as a percentage of the overall contract value).
This rule comes with some key exceptions, most notably delay liabilities. Here, the mechanism is well established: The contractor’s liability for delay in completing the works will be liquidated into a weekly sum. The more difficult negotiation may be on the weekly sum. While the data center developer will seek to cover the anticipated losses that it will incur (either its own losses, if it will operate the data center for its own use, or damages it will need to pay to end users), such sums may exceed what the contractor can commercially withstand. Typical solutions here may involve a staggered damages set-up, where the level of damages increases the longer the delay, and potentially a sub-cap on the liability for delay damages.
Contractors may also push for overall caps on their liability under the construction contract. Traditionally, contracts for the construction of buildings based on standard industry forms, such as the JCT or NEC forms, have not set overall caps on the contractor’s liability. (This contrasts with the market norms for engineering contracts, based on FIDIC or similar forms of contract, where a cap is typically set by reference to the total contract sum.) However, after a turbulent few years in the construction industry, it is becoming increasingly common for contractors to seek overall caps on liability even in conventional building projects. Contractors in the data center sector are particularly likely to raise this.
Developers may be minded to agree to overall caps in addition to (or instead of) the indirect loss exclusion. For large projects, where a cap pitched at the contract value is a big number, and liabilities are approaching that cap, the sums may be so significant that it might be unlikely that a contractor’s balance sheet could withstand that liability. From that perspective, a financial cap may not be seen as a significant concession in negotiation.
A developer may also push for carve-outs from the cap, where unlimited liabilities apply. Common carve-outs include claims relating to gross negligence and willful misconduct, proceeds from insurance, or liability for specific breaches such as intellectual property infringement. In a data center context, the developer may expand this to include data loss or cyber breaches, though contractors are likely to resist this out of concern that a single breach or data loss incident could generate liability in the tens or hundreds of millions of dollars, well beyond the project’s value or available insurance.
Ultimately, the liability profile in the construction contract will be a product of the negotiation power of the parties, and there is no standard solution. As the demand for data centers continues to surge, this will no doubt remain an area of focus for both parties as they seek to contract on grounds that are commercially acceptable to both sides of the table.
Authors