Last December, I argued in this column that Bangladesh should graduate from least developed country status on schedule in 2026. I argued against deferral for three reasons.
First, it sends a wrong signal to the investors, that even though we passed all the markers for graduation we are not ready to accept it, which means our fundamentals remained weak and the economy was weaker than we had been projecting.
Second, the delay would fix nothing. We would be buying time but time does not fix our woes.
Third, I asked that the debate shifts from when or whether to graduate to preparing for graduation so that we benefit from it.
- Our export cannot remain reliant on GSP and other facilities forever.
- We cannot be a cheap labour driven economy forever.
- We cannot be an economy that relies on one sector forever.
For me, LDC graduation is a stress test. But the deferral has happened and we now have three years. Does it make things better or worse?
The terms of the deferral
The resolution grants the extension on an exceptional basis, and the graduation framework does not treat a domestic political transition as grounds for one. This was a concession and it is not a precedent.
Attached to it is an obligation to implement a national smooth transition strategy and report progress to the Committee for Development Policy every year. The Committee has also revised its supplementary indicators for the 2027 review to a set of 44, calling the change a shift from once-off measurement to continuous assessment.
We will be judged every year, in public, on a trajectory rather than a snapshot, and no stated ground is left on which to ask for more time.
What is the case for the deferral?
- LDC graduation would hurt our export earnings
Merchandise exports were $48 billion in FY2025-26, of which readymade garments accounted for $38.70bn, about four-fifths of our export.
The European Union took $19.06bn of those garment exports. Nearly three-quarters of everything we sell abroad moves under duty-free terms we hold only because we are a least developed country, and that share has to find new terms by 2029.
Two things change when that access ends; the tariff is the smaller of them. The first is the rule of origin.
Everything But Arms lets imported fabric made up into garments here qualify as ours, while standard preferences require both the fabric and the garment to be made in Bangladesh.
We import about 60% of our woven fabric, so a large part of the woven trade would fail that test and bear the full tariff whichever scheme we sit under.
The second change is the duty itself, as European Union tariffs on apparel run at about 11.8% at the most-favoured-nation rate and 9.6% under standard preferences, against zero today.
The published estimates of what this costs differ by a factor of three. UNCTAD's Trade Preferences Outlook 2025 puts the annual loss above $17.5bn, a fall of 32%, with 77% of it in the European Union alone.
SANEM puts it near 11% of exports and the UN readiness assessment at between 5.5% and 15%, where UNCTAD's own earlier modelling also sat.
The spread is not a dispute about arithmetic. It turns on whether our garments pay the full tariff or fall back on standard preferences.
Against last year's $48bn, the bill is somewhere between $2.6bn and more than $17bn a year, and we are preparing without knowing which.
- This loss then reaches the wider economy
Our external position is already resting on remittances rather than on exports. The trade deficit reached $27.28bn in FY2025-26 and the current account closed about $1.6bn in deficit. Record remittances of $35.6bn held reserves at $36.21bn in mid-October. An export loss at the modest end of those estimates lands on that arithmetic, as no second earner can absorb it.
The timing is worse than the amount. External debt stood at $78.22bn in March and 62% of it is concessional. Debt service rises from $4.11bn in FY25 to a projected $7.6bn by FY29, the year the preferences end, and graduation closes the windows that made the debt cheap.
The debt is already about 140% of annual exports, and that ratio worsens from both directions if exports fall while servicing rises at market rates.
- The macroeconomic outlook has worsened since last December
Last December, the IMF put our growth near 3.5% with inflation easing. In October, it cut the projection for this year to 3.2%, the weakest since 1988-89, and raised inflation to 10.9%.
The World Bank expects the weakness to persist through FY28, two of the three years we have been given, and private sector credit growth fell to 4.5% in June, a 33-year low.
A diversified export base is built out of investment in laboratories, certification, cold chain, and reliable power, and none of that is financed out of 3.2% growth.
Why graduate at all?
First, the choice is not ours. Graduation follows from meeting the thresholds at two consecutive reviews, and at the 2024 review our income per head stood at $2,684 against a threshold of $1,306, the Human Assets Index at 77.5 against 66, and the vulnerability index at 21.9 where 32 or below qualifies. A government can ask for time, which we have now done, and after that the question is closed.
Second, the preferences are worth less every year even if we keep them, as a margin of preference is only the gap between our duty and our competitors'. Vietnam already sells into the European Union under a trade agreement and India is negotiating one, so that gap narrows on its own.
Third, the preferences shape what we make. They apply to the goods we already produce, so they reward more of the same, and a quarter-century of duty-free access to Europe has left one sector supplying four-fifths of exports. A country off the list negotiates agreements instead of receiving concessions, and an agreement can be written for a product we do not yet sell. The loss above is the price of the transition rather than a permanent cut in what we can earn.
What the three years must deliver
First, the Ministry of Commerce should obtain a written position from the European Commission, before the 2027 review, on whether our apparel qualifies for GSP+ or standard preferences. On Mohiuddin Rubel's reading of the safeguard threshold, it qualifies for neither, and every other decision depends on the answer.
Second, the government should ratify the outstanding conventions required for GSP+ and file the first implementation reports by December 2027, as the scheme rests on a record of enforcement.
Third, the Ministry of Textiles should publish an annual target for domestically produced woven fabric, beginning this fiscal year, as the origin rule applies to the fabric rather than to the duty.
Fourth, it should bring the agreements with Japan and South Korea into force by December 2027 and conclude one more by the end of 2028.
Fifth, the national accreditation board should secure European recognition of its certificates by 2028, as a buyer who will not accept our test report needs no tariff to refuse our goods.
Sixth, the Directorate General of Drug Administration should settle licensing before November 2029 for the fifth to a quarter of locally made medicines patented abroad, as a 4% royalty adds 2% to 3% to the price of a generic.
Seventh, the Ministry of Finance should release the annual progress report at home when it submits it to the Committee, stating the target, the deadline and the responsible ministry for each item above.
Every one of the seven can be completed before November 2029, provided each has a named ministry and a date set against it in writing. Otherwise the three years will pass as the past year has.
Md Rubaiyath Sarwar is Managing Director of Innovision Consulting, a Trustee of Panam Institute.