B Dr Samer Skaik
There is a moment in project meetings that anyone who works on public infrastructure will recognise.
A question is put to the engineer. It is a fair question, within the engineer’s remit, and the answer is not especially difficult. And before answering, the engineer glances at the employer’s representative sitting across the table.
It lasts less than a second and everybody in the room sees it. The contractor sees it and files it away. The engineer’s own team sees it. And what it communicates is that the person the contract designates as the impartial administrator of this contract is, in practice, checking.
I do not raise this to criticise the engineers involved, most of whom are experienced professionals in an impossible position. I raise it because it is a structural problem specific to public employers, the contract does not solve it, and it is the origin of a startling number of the disputes I see.
What the contract asks for
The standard forms ask the engineer to do two things that sit awkwardly together.
The engineer is the employer’s agent for most purposes. The employer appoints and pays the engineer, and the engineer administers the contract on the employer’s behalf. But when it comes to determining a claim or a disputed matter, the engineer is required to act neutrally: to consult with both parties genuinely, to try to bring them to agreement, and failing that, to make a fair determination in accordance with the contract, regardless of who is paying the invoice.
The 2017 editions tightened this considerably. The determination process is now structured, with defined periods for reaching agreement and for issuing a determination. If the engineer does not determine in time, the contract treats the silence as a rejection, which the dissatisfied party can carry forward. I touched on this in an earlier post about the change of form; here I want to look at what happens when the engineer’s neutrality meets a ministry.
Three ways it fails
In my experience it breaks down in three recognisable patterns.
The first is the engineer who will not decide. Determinations go unissued, or are issued so late that the contractual period has long expired, because the engineer is waiting for a position from the employer that never comes. The contractor, entirely properly, treats the silence as a rejection and escalates. The employer is then defending a claim it never actually considered on its merits.
The second is the engineer who determines and is then disowned. A determination is issued in favour of the contractor. The employer refuses to give effect to it, does not pay, and behaves as though the engineer has made an error. This is the worst of the three, because the employer has now put itself in breach of its own contract while simultaneously destroying the engineer’s standing with the contractor for the remainder of the project. Every subsequent determination is read as provisional.
The third is the engineer who is instructed. Here the employer tells the engineer what to determine, and the engineer complies. This one is the most dangerous and the least visible at the time. It usually leaves traces — in emails, in draft determinations circulated for comment, in meeting minutes — and those traces are exactly what a contractor’s legal team will look for years later. An arbitral tribunal that concludes the engineer was not acting independently will give the determinations little weight, and the employer will have lost the procedural advantage the contract gave it.
Why it happens, honestly
The usual explanation offered is bureaucracy, and that is true but not sufficient. The deeper reason is one that public officials will recognise immediately and that consultants often miss entirely.
Approving a contractor’s claim exposes an individual public servant to personal risk.
Think about it from inside the system. If you reject a claim and you are wrong, the consequence is that a tribunal decides against your organisation in four years, by which time you have moved on. If you approve a claim and someone later questions it, you are the official who authorised a payment to a contractor. Depending on the jurisdiction, that can mean an audit finding, an investigation, a referral, or worse. The asymmetry is enormous, and it is entirely rational to respond to it by never approving anything.
That is the machine the engineer is operating inside. The engineer is asking for a decision from a person for whom the safest available decision is no decision at all. Once you see it in those terms, the engineer’s glance across the table stops looking like weakness and starts looking like an accurate reading of the room.
It also explains something that puzzles people new to this sector: why so many public employers seem content to let matters proceed to arbitration over sums they could have settled. From inside, arbitration is not a failure. It is a mechanism that transfers the decision to an external body, so that no individual official has to own it. That is an expensive way to run a project, but it is not irrational behaviour by the people doing it.
The problem of who was here last year
One further factor deserves mention because it compounds all three failure modes and is almost never designed for.
Public sector project teams turn over. On a five-year contract it is common to see three project directors, two or three contract officers, and at least one change of minister. The engineer’s team turns over as well, though usually less. The contractor’s commercial staff, by contrast, are frequently the most stable people on the project, and they are the ones holding the institutional memory of what was agreed in year one.
The practical effect is that an incoming official inherits a file rather than an understanding. Faced with a claim that turns on a conversation held two years ago by someone who has left, the safest response is to defer, which loops straight back into the first failure mode. As a sole ADAAB member, I have sat in quarterly meetings where nobody on the employer’s side of the table had been present for the events under discussion, and everybody on the contractor’s side had.
There is no clever solution to this, but there is a partial one: insist that every determination, every agreed position, and every significant instruction is recorded with its reasons at the time. A file of reasoned documents survives turnover. A file of outcomes does not, because the next person cannot tell whether the outcome was principled or expedient, and will treat it as neither binding nor defensible.
What actually improves it
You cannot fix this with a clause, and consultants who try tend to produce particular conditions that make things worse. What helps is administrative, and most of it has to be done at the start.
Write down what the engineer can determine without a referral. A delegation matrix with monetary thresholds and subject-matter categories, agreed by the employer, issued to the engineer, and shared with the contractor. It costs an afternoon. It removes the single largest cause of stalled determinations, which is genuine uncertainty about whether the engineer is allowed to act.
Create a body that can say yes. Where a matter must be referred upward, it should go to a standing claims review group with a fixed meeting cycle and a defined membership — typically the project director, a finance representative and a legal one. A collective decision distributes the personal exposure that paralyses an individual, and a fixed cycle means the engineer knows when an answer will arrive.
Document reasons, not just outcomes. The protection an official actually needs is a written record showing that a claim was assessed against the contract on stated grounds. An approval with reasons is defensible in an audit. An approval without them is not, which is why officials avoid giving them. Insisting on reasoned recommendations from the engineer is therefore not bureaucratic overhead; it is what makes approval possible.
Use disagreement as designed. An employer that disputes a determination has a contractual route: give notice of dissatisfaction and refer the matter onward. What it should not do is give effect to nothing and hope the issue subsides. The contract accommodates an employer that disagrees with its engineer. It does not accommodate an employer that ignores its engineer.
Let the board carry what the engineer cannot. Where a standing dispute board is in place, the informal assistance route is available for exactly these situations — a matter both parties would like resolved but which the employer’s internal machinery cannot bless. An independent view sought jointly gives an official something to put in the file. It is not a substitute for the engineer’s function, but it takes weight off it.
A word to the engineers
Since this post has been written mostly from the employer’s side, one observation from the other.
The engineer who misses determination periods to avoid displeasing the employer is not, in the end, serving the employer. Deemed rejections and unissued determinations produce escalation, and escalation produces cost. The engineer who issues a reasoned determination within time — even one the employer dislikes — has given the employer a defensible position and a clear route to challenge it. That is what the appointment is for, and it is worth saying so to the client early, at the point where the relationship is being set rather than after the first uncomfortable determination.
The underlying point
Everything in this post comes back to a single mismatch. The contract assumes a decision-maker with authority acting to a timetable. Public administration produces distributed authority acting to a different one. Neither is going to change, so the work is in building the bridge between them, and the time to build it is in the first month of the project when nobody is under pressure.
That is not a coincidence. It is the same month in which most of the other things worth doing on these projects also need to happen, which is a subject I have written about at some length.
This post is the sixth in a series for public sector employers delivering infrastructure under internationally financed contracts. Related reading: The Engineer’s Role in Dispute Prevention, When the Banks Changed the Rulebook, and The First Four Weeks.
Similar Posts:
- Engineers should be given deadlines to assess claims
- The disappearance of the impartial engineer
- Can disputes be settled like gentlemen?
- Empowering the Engineer: Significant Advances in FIDIC 2017 Red Book Over 1999
- Understanding Contractor Remedies When the Employer Occupies and Uses Part of the Works Under FIDIC 1999
Discover more from CMGuide
Subscribe to get the latest posts sent to your email.