Where the time actually goes in a Nepali public project

Ask anyone in Nepal’s infrastructure sector why projects run late and you will get a confident answer that depends entirely on where they sit. Contractors blame procurement. Procurement blames the design. Designers blame the clearances. The ministry desks that review the files, where I sit, get blamed by everyone, and blame the quality of what arrives.

Everyone is describing a real problem. The trouble is that all of these are answers about stages, and I have come to think the delay does not mostly live inside the stages.

The gap is not owned by anyone#

Take a project from conception to mobilisation and mark two kinds of interval. The first is time inside a stage: someone is doing the feasibility study, someone is reviewing the file, someone is evaluating bids. The second is time between stages: the study is finished and accepted, and nothing happens next.

Inside a stage, there is a person whose job it is. There is a file, an officer, and usually a deadline that someone can be asked about.

Between stages there is nobody. The feasibility study is complete, so the study team is done. The next stage has not started, so no one has been assigned. The project has no owner during exactly the period in which it is not moving. When a project sits for eleven months waiting for a budget line, there is no one who was late.

Why reform keeps missing it#

Almost every reform proposal I have read targets the first kind of interval. Reduce the review period. Cap the number of resubmissions. Introduce deemed approval. These are good ideas and I support most of them. But if the waiting between stages is the larger share, then compressing a four-month review to two months buys you two months out of a delay measured in years.

It is easy to see why the reform effort goes where it goes. Time inside a stage is legible: there is a receipt date and a decision date, and the difference is a number you can put in a performance report. Time between stages is invisible in exactly the same way: no receipt, no decision, no number.

What gets measured gets reformed, and we are measuring the smaller half.

What would change if this is right#

Three things follow, if the gaps really do dominate.

First, someone should own the project rather than the stage. Not a coordinating committee that meets quarterly, but a named officer whose file the project remains in from conception to mobilisation, including through the periods when nothing is happening. The point is not that they can force the next stage to start. The point is that the question “why has this not moved” has an addressee.

Second, the handover should be a dated event. Right now, the end of a stage and the start of the next are recorded, if at all, in different systems belonging to different units. Until the interval between them is a number in a single register, it will not be managed.

Third, budget timing deserves as much attention as approval timing. A large share of the waiting I see is a project that is ready and simply out of phase with the budget cycle. That is a sequencing problem, not an approvals problem, and no amount of faster review will touch it.

What I am not claiming#

I am building a dataset on this rather than asserting it, and I want to be honest that I do not yet have the sample to prove the claim. Completed projects are far easier to trace than stalled or abandoned ones, which biases any sample toward the projects that worked. It is entirely possible that when the numbers come in, the gaps turn out to be a third of the problem rather than most of it.

But even a third would be a large, unmanaged, entirely invisible third. And the cheapest thing we could do about it, writing down the date a stage ends and the date the next one begins, costs almost nothing.

If you have stage dates for projects you have worked on and would be willing to share them, I would like to hear from you.