Probabilities are always a bad idea for revenue recognition.
"A man must be big enough to admit his mistakes, smart enough to profit from them, and strong enough to correct them. " John C. Maxwell
In the first week of November shares of John Wood, PLC a UK energy engineering company has seen its shares price plummet down by 60% (ca. 50 GBP from 125 GBP).
Wood, headquartered in Aberdeen, SC engages in decarbonization, digitalization, energy security and energy transition projects.
The Company had to make write-offs on contract positions related to their projects of approximately $ 1 billion (total revenue: $ 5.9 billion in 2023).
The move has resulted from a “dialogue” with their auditor (KPMG).
Let’s take a look at the issue:
Wood is active in the project construction business with typically long time span to complete and deliver the project. Based on IFRS 15 Rveneue from Contracts with Customers entities are permitted to recognize revenue (and respective costs) based on the project progress regardless of actual billings (known as percentage-of-completion method).
IFRS 15 sets out specific criteria that must be fulfilled to enable this revenue recognition pattern.
The customer simultaneously receives and consumes the benefits of the project as it is constructed, or
The control over the project goes to the customer as it is being constructed, or
The company does not create a project (an asset) with alternative use to the Company (meaning the project cannot be sold or used otherwise) and the company has a (legally) enforceable right to the payment for performance completed to date (contract).
The red flag concerning the real world case:
At Wood it was the project Aegis (and probably others) on construction of properties to house an anti-missile defense facility for the US Army. As of 2022 losses of $ 222 million were incurred and recognized in the financial statements.
The volume of at least $ 186 million of this project was legally disputed by the customer though, primarily due to:
Yet, the disputed amount was assessed by the management as “highly probable” (based on weighing of probabilities of outcomes) and retained in revenue.
Despite this optimism the management admitted: “If the actual outcome differs from these estimates and assumptions, the ultimate loss will be different and can be material.”
Some lessons from this case:
Use conservative realistic, never too optimistic assessment.
Looking at IAS 37: virtual certainty (95%+) is required for recognition of contingent assets and revenue.
Probabilities are always a bad idea for revenue recognition.
Wood employs 35,000 employees and operates in over 60 countries worldwide.