Levy Killed, Approval Ten Months Later: Pakistan's EDF Moves From a Treasury Account to a Bank Investment

There is a gap between two dates, and it is the hardest number in this story. Pakistan's Export Development Surcharge stopped on 1 December 2026 — notified through SRO/2335(I)/2026, on the Prime Minister's direction, which itself followed a Prime Minister's Office working-group recommendation. That surcharge was the Export Development Fund's principal income line. The second date belongs to the Commerce Ministry summary that asked for an alternative income stream for the Fund: 6 October 2026. Between the two sits roughly ten months. For ten months the Fund had lost its main revenue line by statute and had no replacement approved. What has now emerged is not a grant. It is permission to carry money out of a government account into a commercial bank and invest it in government securities. The approval is the headline; the conditions are the story.

Start with what the Fund is. The EDF is a statutory fund inside Pakistan's federal fiscal architecture, held in the Public Account. That means its balances do not lapse at the fiscal year's end, it sits outside the consolidated fund proper, and it has its own board. Its income came from a levy on export value — a flow, rising and falling with the size of exports each year. The state removed that levy to cut exporters' cost of doing business, which is the phrase the official record uses.

The legal sequence ran as follows. The EDF Act was amended, a process completed with Presidential assent on 26 January 2026. On 17 February 2026 the Federal Government notified the appointment of new Board members. The matter surfaced at the Board's 93rd and 96th meetings, and the Board asked the Finance Division to facilitate moving its own corpus out of a government account. The Finance Division backed the investment but attached a legal-form requirement: because the money would leave a government account for a commercial bank, Federal Government approval was needed under Section 23(2) of the Public Finance Management Act, 2026. The Commerce Ministry routed the case to the Economic Coordination Committee, which considered the proposal and approved it, paving the way for investment in government securities.

The least discussed and most important element in that chain is the Finance Division's posture. It did not block the proposal. It supported the investment and simultaneously demanded the statutory form. That is unusual in South Asian fiscal reporting, where the regulator either quietly steps aside or performs obstruction. Here the control was identified before the transfer and the correct approval channel was used rather than bypassed. On compliance, that deserves the credit the record gives it.

The problem is not compliance; it is the mandate. Only two destinations are named — fixed-income government securities, or an interest-bearing account at the National Bank of Pakistan. Nothing else. No duration limits, no concentration limits, no credit rules, no prohibited-instrument list, no benchmark. Granting a public fund investment authority without a published mandate is a governance gap, not a technicality. It is the most recognisable weakness in sovereign fund management: the power is issued, the limits are left unwritten, and unwritten limits mean every future decision stays formally valid and publicly unappealable.

Now the part fiscal reporting usually skips. The money leaves a Treasury account, arrives at a commercial bank, and from there buys government paper. In plain terms, the state lends its own money back to itself through an intermediary and pays that intermediary's spread. On the EDF's ring-fenced books the transaction will show a clean gain. For the consolidated public sector, the benefit after intermediation, cost and tax treatment sits well below the headline yield, and in some configurations can turn negative. A meaningful slice of the 'new income' is a reclassification of value, not a creation of it. That is the same discipline I learned auditing rights cycles: the contract figure is never the argument; the assumption underneath it is.

The second calculation concerns substitution. A levy is a flow — a share of export value, every year. Investment income is a return on principal — a single-digit percentage annually under normal conditions. These two scales are not substitutes for each other. The primary document half-admits this: it speaks of maintaining the Fund's 'sustainability', not of restoring its funding capacity. Sustainability and restoration are different claims, and a document making the smaller claim should be read as honest about its own ceiling. The reasonable prior, then, is that levy replacement will be partial at best. Note where that prior comes from — an absence, not a dataset. The report carries no corpus size, no forgone revenue, no expected yield, no tenor. From years of working with match data I have kept one habit: what is not measured is still a finding, and it tells you who decided the question was not worth asking. The missing figures here are not a press-release convenience; they are a governance choice.

The third element deserves a name: mission drift. A body constituted to develop exports is being repositioned as a holder of government paper. No clause in the record imposes an export-linked deployment condition on the income earned. The resources remain ring-fenced in the exporters' name; the mechanism has moved to the market's.

One smaller measurement that still speaks: latency. The issue surfaced at the Board's 93rd and 96th meetings. Board sittings are numbered sequentially and generally held at a regular cadence, so the plain inference is that the question sat unresolved across roughly three sittings. That is an inference, not a stated fact, and it should be carried at medium confidence.

Here is the part where I usually get argued with. The easiest reading is institutional decline. I find that cheap. First, approval is not execution: nothing in the record says funds have moved or securities been bought. What exists is permission and a path, both procedural categories. Second, three mundane explanations for the absent figures exist — the summary genre of a government release, commercial sensitivity in ECC papers, and the possibility that the numbers had simply not been fixed at briefing stage. I accept the first two, and the third as well, though it is the most expensive: if no return estimate existed at the moment of approval, on what basis was approval granted? Third, and strongest: non-lapsable money parked in the Public Account carries a real cost while it sits idle. Placing it in government paper can be rational treasury management, and sounding the mission-drift alarm against that is over-reading an absence.

So I will keep the arithmetic explicit. My assessment: roughly a 60 per cent probability that investment income partially offsets the lost levy and the Fund continues at a reduced scale; roughly 25 per cent that the offset proves visibly inadequate and the Fund returns for a different form of fiscal relief; roughly 15 per cent that this becomes a durable template, with other idle Public Account balances converted into government securities. What would prove the pessimistic read wrong should be written down in advance: a published investment mandate carrying duration and concentration limits; a first public disclosure of corpus size and expected yield; and an export-linked condition on deployment of the income. If all three arrive within two quarters, the alarm should be withdrawn. If none does, what happened was reclassification, not reform.

Looking forward, three markers are worth tracking — the published investment policy document, the first Finance Division or ECC disclosure carrying amounts, and the EDF's next annual report. If none of them is public by mid-2027, this case will not be filed in South Asian fiscal history under reform. The interesting question is not whether the Fund bought securities. It is whether a single document will exist stating where its money stops — and until that order appears, we are all keeping accounts on a silence that can be measured but has never been weighed.