For decades, construction contracts and subcontracts—both in New York and elsewhere—seem to have attracted more than their share of payment disputes. Owners regularly run out of money. Contractors and subcontractors regularly screw up (and run out of money). Projects regularly go over budget—but rarely stay under budget—and regularly fail (often because they ran out of money).

In response, New York, like most other states, allows contractors and subcontractors to file mechanics’ liens against projects if not paid. Those liens can eventually be enforced just like foreclosing a mortgage. Mechanics’ lien laws give contractors and subcontractors a very powerful collection technique not available to ordinary creditors, such as unpaid real estate lawyers.

Contractors tried to reduce their exposure to some of these risks by adding “pay when paid” clauses to their subcontracts. Those clauses said that the contractor didn’t have to pay the subcontractor unless the owner decided to pay the contractor. It was a great mechanism for the contractor to shift the risk of nonpayment to its subcontractors.

The New York courts decided that the mechanism was too great to actually work. It flew in the face of the New York mechanics’ lien law, which said that any waiver of the right to file a mechanic’s lien was unenforceable. And, the courts said, a “pay when paid” clause amounted to a back-door waiver of the subcontractor’s right to file a mechanic’s lien.

In 2002, the New York Legislature complicated these issues by deciding that the relationship among owners, contractors, and subcontractors required further improvement. The legislature passed a law that set standards and procedures for how and when owners are supposed to pay contractors, and contractors are supposed to pay subcontractors. The law unambiguously required the contractor to pay its subcontractors whether or not the owner paid the contractor. The owner’s nonpayment shouldn’t be the subcontractors’ problem, according to the legislature.

The 2002 law did, however, carve out an exception: If the owner appointed an agent to act for the owner in signing contracts, then the agent wouldn’t be responsible for any payments due under those contracts. That conforms to traditional principles of the law of agency.

At least one smart contractor tried to use this exception to create protections similar to a “pay when paid” clause. That contractor inserted new language into its subcontracts, requiring each subcontractor to acknowledge that the contractor merely acted as an agent for the owner, so only the owner was responsible for payment. The owner’s payments would, of course, run through the contractor on the way to the subcontractor, but the contractor was still just an agent for the owner—a channel for payment—without liability.

It didn’t work. A court decided that the contractor couldn’t shrug its shoulders and claim to be nothing more than the owner’s agent. Instead, the rest of the subcontract made clear that it was a separate and independent contract between the contractor and the subcontractor. The subcontractor signed the contract in its own name. The subcontractor’s obligations ran to the contractor, not the owner. The contractor couldn’t escape liability by claiming it was an agent.

Although this contractor lost its litigation with the subcontractor, the “agency” theory just might work in a future subcontract. A contractor would need to play it through thoroughly. The contract would need to make clear throughout that the contractor is a mere innocent “agent” of the owner, and signs and acts only in that capacity.

If the contractor were in fact a construction manager, then such agency status makes sense. In a traditional general contract arrangement, however, where the contractor expects to make a profit after paying subcontractors, it’s not so easy for the contractor to claim to be the owner’s agent. Contractors may have to come up with some other solution to the problem.