Documentation a Government Body Will Accept

A Public Body Asks Whether the File Is in Order Before It Asks Whether the Project Is Good

A private counterparty who receives a strong proposal in an imperfect format reads it anyway. A public authority frequently cannot. Submissions to ministries, state industrial bodies, licensing authorities and subsidy administrators pass through a compliance check before they reach anyone whose job is to form a view about the proposal, and a submission that fails that check is returned without its merits having been considered at all. The applicant experiences this as a rejection of the project. It was not. It was a determination that the file was not yet in a state in which the project could be assessed.

The reason is not bureaucratic temperament. A public official acts under a delegated authority, spends public money, and can be asked afterwards — by an auditor, a legislature, a court, a successor in the same chair — to justify what they did and on what basis. That accountability produces a documentary culture with a specific property: the file must be able to speak for itself, years later, to a reader who was not present and who cannot ask anyone a question. Every requirement that looks like formality serves that property. A prescribed format ensures the file contains what a later reader will need. An attestation records who is answerable for a figure. A completeness checklist protects the official from having proceeded on an incomplete record.

This article sets out what that means for the document: why format compliance operates as a gate before merit is considered, how a submission is condensed into a file note by someone junior and what survives that condensation, why a figure without a traceable source is worse than no figure at all in a public file, what annexure discipline actually requires, what each desk in a multi-department circulation is asking as the file passes it, and why a sanction is the beginning of the conditions rather than the end of them. Where our transaction advisory practice prepares documentation for a public-authority audience, it is built to this reading — and none of what follows is confidential or peculiar to any particular body.

One caution belongs at the front. Public administration differs materially between jurisdictions, between departments in the same jurisdiction, and between schemes administered by the same department. Nothing in this article states the procedure of any named authority, and nothing here substitutes for the published rules of the specific scheme being applied to. What is offered is the general shape of the reading — which is remarkably consistent — and the discipline it implies for anyone preparing the file.

Format Compliance Is a Gate, and Most Files Fail at It Rather Than on Merit

The first person to handle a submission is generally not assessing it. They are checking it against a list: are the prescribed forms present and in the prescribed version, are the annexures numbered and cross-referenced as required, are the attestations signed by persons holding the offices the scheme specifies, are the financial statements of the years the scheme requires and in the form it requires, is the applicant eligible on the face of the documents. That check has a binary outcome, and it happens before anyone forms any view whatever about the project.

Applicants systematically underestimate this stage because it has no analogue in commercial dealing. A bank that likes a proposal will ask for the missing document and keep reading. A scrutiny clerk working to a checklist frequently cannot: the file is incomplete, and an incomplete file returned is the correct disposal under the procedure they are working to. The judgement they are exercising is about the checklist, not about the applicant, and appealing to the strength of the project has no purchase at that desk because the strength of the project is not what is being determined.

The most expensive version of this is version drift. Schemes publish forms and revise them; a form downloaded some months earlier, or reused from a previous application, may have been superseded. The content may be identical in substance and the file is still returned, because the officer cannot certify that the applicant has furnished the prescribed form when the prescribed form is a different document. Confirming the current version of every prescribed document on the day of filing is a five-minute task that removes an entire class of failure, and where we act for an applicant it is done as a matter of course rather than as diligence.

The corollary is worth stating plainly, because it changes how preparation time should be allocated. Effort spent perfecting the analytical case has no value if the file does not clear the gate, and effort spent clearing the gate has no value if the case then fails on merit. Both are necessary and they are not substitutes. In the mandates our advisers run, the compliance requirements are established from the published scheme documents before drafting begins, so that the substantive work is done inside a structure already known to be acceptable rather than fitted to one afterwards.

Your Submission Will Be Summarised by Someone Junior, and the Summary Is What Travels

A file in a public office accumulates notings. A dealing official reads the submission and records, in a few paragraphs, what has been received, what the scheme requires, whether the requirements are met, and what they recommend. The next officer reads that noting, may read parts of the submission, and adds their own. By the time the file reaches the person with authority to decide, it carries a chain of notings — and the reading, in practice, runs down that chain rather than through the original document.

This has a consequence that is easy to state and hard to internalise: anything in the submission that does not make it into the first noting has, for most practical purposes, not been read by the decision-maker. A critical technical certification, a completed statutory clearance, a signed offtake arrangement, an independent study supporting the demand assumption — each of these is decisive if it appears in the noting and close to invisible if it does not. And whether it appears depends not on how important it is but on how easy it was to extract while reading under time pressure.

The remedy is structural rather than rhetorical. Each substantive point should be stated once, plainly, in a form that can be lifted into a noting without rewriting: a short declarative sentence, the figure, and the source. Long paragraphs in which the key fact is embedded in the fourth clause do not survive; the official summarising is not hostile, they are simply extracting, and what is hard to extract is left. A summary table at the front of a submission — project, cost, means of finance, employment, timeline, eligibility basis, each with the annexure reference beside it — is frequently the most valuable page in the file, because it is the page the noting is written from.

The register matters here as much as in a credit paper, and for the same underlying reason. A noting is written in flat administrative prose because it must be defensible, and promotional language does not survive translation into it. Where a submission describes a project as transformative, the noting records what the project is. Where a submission states that the plant will produce a specified volume, sourced to a specified technical report, the noting can record that. The second survives the condensation with its meaning intact because there is nothing in it to strip out. Our advisers draft to that register deliberately: the objective is not to impress the reader but to be adoptable by them.

One further point about the chain. Because each noting is signed and dated, and because the file is a record, an official who adopts a figure has associated their name with it. That is a real professional exposure, and it explains behaviour that applicants sometimes misread as excessive caution. An official who cannot see where a number came from will not adopt it; they will instead record that the applicant has stated it, which is a materially weaker position in the file, or they will raise a query. Making adoption easy is therefore the single most useful thing a submission can do for its own progress.

In a Public File, a Figure Without a Source Is Worse Than No Figure at All

In a commercial negotiation an unsupported figure is a weak position that can be argued, conceded or traded. In a public file it is something different in kind: it is a number an official cannot adopt. They cannot record it as established, because nothing establishes it. They cannot rely on it in a recommendation, because a later reader will ask what it rested on and the file will not answer. Their available options are to attribute it explicitly to the applicant — which weakens it and signals that it is unverified — or to raise a query. Either way the applicant has spent a cycle and lost credibility on the surrounding material.

This is why an unsourced figure is worse than no figure. A submission that does not quantify something the scheme does not require quantified has simply not addressed it. A submission that asserts a specific number with nothing behind it has introduced into a permanent record a claim that cannot be substantiated, and has invited a scrutiny of every other number in the same document. The cost is not confined to the line that failed. Once a reader has found one figure they cannot trace, they read the rest differently, and the queries returned will reflect that.

What counts as a source is narrower than applicants generally assume, and it is worth being precise about the categories. A published statistic from an identified official or industry publication, with the publication and period stated, is a source. A technical report by a named professional carrying their qualification and date is a source. A signed contract or a firm quotation, referenced to the annexure that reproduces it, is a source. An operating history drawn from audited accounts, with the years identified, is a source. A figure derived by calculation from any of these is a source if the calculation is shown. What is not a source: a market size described as widely accepted, a growth rate attributed to industry expectations, a comparison to unnamed similar projects, or a number that appears in the model with nothing behind it.

Derived figures deserve particular attention because they are where well-prepared submissions most often fail. A revenue projection is not a source even when every input to it is sourced; it is a calculation, and what must be traceable is the calculation. Showing the volume assumption, its source, the price assumption, its source, and the arithmetic between them converts an assertion into something an official can adopt on their own authority. Where our financial due diligence team prepares a submission, every quantitative claim is traced to a citable source before filing, and the trace itself is documented — because the question is not whether the figure is right but whether the file demonstrates that it is.

There is a version of this that catches even careful applicants: the figure that was sourced when it was written and is no longer current when it is read. A tariff that has since been revised, a statistic superseded by a later release, a quotation past its validity period. In a permanent record, a stale citation is a defect rather than an inaccuracy, and it is generally noticed. Dating every source, and confirming currency immediately before filing, closes it.

The Annexures Are Not an Appendix; They Are Where the File Is Actually Proved

In commercial documents an appendix is where supporting material is put so it does not interrupt the argument, and readers rarely go there. In a public submission the relationship inverts. The main document makes assertions; the annexures are the evidence those assertions rest on, and the scrutiny works from the checklist to the annexure and back. An annexure that cannot be located against its checklist item is treated as not filed, however clearly the main document refers to it, because the officer's task is to record that the requirement was met and they cannot record what they cannot find.

Discipline here is mechanical and unglamorous, and it repays effort out of all proportion to the effort involved. Every annexure numbered, in the order the checklist lists them. Every reference in the main text naming the annexure number and, where the document runs to length, the page within it. Every annexure carrying a cover sheet stating what it is, so the officer does not have to infer from the first page. Documents in the language the scheme requires, with certified translations where they are not. Copies attested where attestation is required, by whoever the scheme specifies. Where a valuation forms part of the file, our valuation team prepares it to the format the scheme prescribes rather than to a general-purpose structure, because a valuation in the wrong form is returned on the same basis as a missing annexure. Pagination continuous across the whole file, so that a query can be answered by a page reference rather than by a description.

Two failures recur often enough to name. The first is the annexure supplied but not indexed — present in the bundle, absent from the list, and therefore invisible to a scrutiny working down the list. The second is the composite annexure, in which several distinct requirements are satisfied by one bundle of documents; the officer must then tick several checklist items against a single entry, and where the correspondence is not explicit they will frequently decline to. Both are cured by making the index match the checklist item for item, which takes an hour and removes a category of return.

There is also a question of what should not be in the annexures. Volume is not persuasive in this context and is faintly counterproductive: a file padded with material the scheme did not ask for makes the required material harder to locate, and an officer hunting through a hundred unnecessary pages for the one required certificate is an officer forming an impression. The discipline is to file exactly what is required, indexed exactly as required, plus the specific evidence that supports the specific quantitative claims made — and nothing else. Where our advisers assemble a submission, the index is built from the scheme checklist first and the documents fitted to it, rather than the reverse.

The Same File Is Read by Several Departments, Each Asking a Different Question

A submission of any significance does not sit with one officer. It circulates — to a technical desk, a finance desk, a legal desk, sometimes an environmental or land or labour desk, and to whichever authority holds the delegation to approve at the amount involved. Each desk reads the same file and each is asking something different, on its own criteria, and each can stop the file independently. The applicant frequently sees only the aggregate outcome and cannot tell which desk generated which query.

The practical consequence is that a submission must satisfy several distinct readings simultaneously, and a document optimised for one of them tends to underserve the others. A file written to satisfy the finance desk — cost, means of finance, viability, cover — can be thin on the technical basis of the capacity figure that the finance analysis depends on. A file written by engineers can be immaculate on process and silent on how the working capital cycle was derived. Neither is complete, and a file that is incomplete for one desk stops for everyone, because the desks are sequential more often than parallel.

Sequencing itself is worth understanding, because it explains most of what applicants experience as unexplained delay. A file moves from desk to desk and each movement has a queue in front of it. A query raised at the fourth desk frequently returns the file to the applicant, and the answer re-enters the circulation — not always at the point it left, and sometimes at the beginning. A single avoidable query at a late desk can therefore cost more elapsed time than several early ones. This is the strongest argument for preparing the whole file to the standard of the most demanding desk at the outset: the marginal cost of doing so is small, and the cost of not doing so is compounding.

It also explains something that puzzles applicants: a query about a matter that was, they believe, already answered. Each desk reads independently and against its own criteria, and an answer that satisfied the technical desk may not address the question the legal desk is asking about the same paragraph. The productive response is to answer the question actually asked, in the terms the asking desk uses, rather than to point out that the ground has been covered — which reads, in a file that is a permanent record, as an applicant declining to answer. In the mandates our advisers run, query responses are drafted to be filed as they stand: self-contained, in the vocabulary of the desk that raised the point, and complete without reference to earlier correspondence. Working through a network of 80+ specialist consultants across 22 countries, our advisers see the same query patterns recur across very different jurisdictions, which is what makes them worth describing at all.

A Sanction Is a Conditional Instrument, and the Conditions Are the Real Document

Applicants treat the sanction letter as the end of the process. It is more accurately the point at which the obligations become specific. A public sanction is conditional almost without exception: conditions precedent to be satisfied before any disbursement, undertakings to be maintained through the life of the benefit, reporting at stated intervals, and a right of recovery if the conditions fail. Reading the sanction as an approval and the conditions as formalities is the most expensive misreading available at this stage, because the conditions are enforceable and the recovery provisions are real.

Conditions precedent are the immediate issue and they are frequently underestimated on timing. A requirement to furnish evidence of the promoter's contribution having been brought in, of the balance funding having been tied up, of a clearance obtained, of a specified insurance in force — each takes elapsed time, and several must generally be satisfied together before the first disbursement. Where the sanction carries a validity period, as it commonly does, the conditions and the validity interact: an applicant who begins work on the conditions after receiving the sanction may find the window closing while a third party takes its own time. Reading the likely conditions from the scheme documents before sanction, and starting on the slowest of them early, converts weeks into days.

The continuing undertakings are where the longer exposure sits, because they persist after everyone has moved on. Commitments as to employment maintained, capacity operated, location retained, activity not changed, assets not disposed of — each is a condition of the benefit continuing rather than a description of intent, and each is capable of being breached by an ordinary commercial decision taken years later by someone who never read the sanction. A recovery demand arising from a restructuring nobody connected to the original file is a foreseeable event, and it is foreseeable precisely because the conditions outlive the institutional memory of them. Extracting the continuing obligations into a short schedule held by whoever will actually be operating the business is a modest step that prevents it.

Reporting obligations deserve their own mention because failure here is so often inadvertent. Periodic returns, utilisation certificates, progress reports — each with a form, a period and a due date, and each a condition of continuing entitlement rather than an administrative courtesy. A missed return is a breach on the record even where everything substantive is in order, and it is the kind of breach that surfaces at the least convenient moment, typically when a subsequent application or a later inspection brings the file back into view. Where our transaction advisory practice supports a sanctioned project, the conditions and reporting obligations are extracted into a dated schedule at the point of sanction, because the sanction letter itself is filed and rarely read again.

Write for the Record, Not for the Reader in Front of You

The discipline this article describes reduces to a single reorientation, and it is the mirror of the one a credit proposal requires. A commercial document is written for a reader who will decide. A public submission is written for a record that will be read by people who were not present, at times nobody can predict, for purposes nobody has yet formed. The official in front of you is one reader among several, and their most pressing constraint is that whatever they write down will be read by someone else later. A submission that makes that easy — complete, indexed, sourced, phrased so it can be adopted rather than merely believed — progresses. One that does not is returned, and the return will be about form, because form was as far as anyone got.

None of this favours a well-resourced applicant over a modest one, which is worth saying plainly. The requirements are published, the checklist is available before drafting begins, and the discipline is mostly clerical: confirm the current version of every form, index the annexures against the checklist, cite every figure to something a stranger could locate, state each material fact in a sentence that can be lifted, and read the conditions on the day the sanction arrives rather than the day something goes wrong. A small applicant who does those things is filing a better file than a large one that has not, and the scrutiny desk cannot tell which is which and is not trying to.

Where we act for an applicant, this is the whole of the preparation method, and there is nothing in it that a competent internal team cannot do. It is offered as an observation about how public files are actually read — drawn from work delivered by a network of 80+ specialist consultants across 22 countries — rather than as technique, because the reading is consistent enough to be described and the discipline it implies is available to anyone willing to apply it. The published rules of the specific scheme always govern, and they should be read first. What this article adds is why those rules take the shape they do, which is generally the thing that makes them possible to follow well rather than merely to follow.