Three changes to preference rules that would help LDCs keep more value

In the last post I argued that trade preferences should reward value as well as volume. Here I suggest three changes. They are aimed at the schemes that the EU, the UK and others run for least developed countries (LDCs) and for countries that have recently left the group.

I use Bangladesh as the example because it is by far the largest user of the EU's LDC scheme, Everything But Arms.

1. Tie the transition to progress, not only to the calendar

Today the EU gives a graduating LDC three more years of duty-free access. After that, Bangladesh would need to meet a two-stage production rule for clothing. Without GSP+ or a similar arrangement, its garments would face an average EU tariff of about 12 per cent.

For a sector where median markups are below 4 per cent on most products, that is a cliff.

Suggested change: keep the three years as a minimum, then extend duty-free access for sectors whose domestic value share is rising. A country that is investing in yarn and fabric capacity would have time to finish the job.

2. Count regional inputs and lighten the paperwork

In 2015, WTO members agreed in Nairobi that preference-giving countries should consider three things for LDCs: allowing imported materials of up to 75 per cent of a product's final value, widening cumulation, and simplifying documents, including self-certification. The text says "consider", so it binds nobody.

Suggested change: make these the default for LDCs and recent graduates. In particular, let inputs bought from other LDCs and regional partners count as local. This rewards regional supply chains. It does not force every poor country to build a full textile industry alone.

3. Stop penalising scale

Under the EU's proposed GSP rules, a safeguard on clothing can apply once a country's share of the relevant EU imports passes about 6 per cent. Bangladesh's share is above 20 per cent. A rule like this treats success in one product as a reason to withdraw the preference. It pushes a country to stay small or lose access.

Suggested change: for recent graduates, test safeguards against actual harm to producers in the importing market, not against a fixed import share, and phase them in.

A smaller fourth point: publish the data

LDCs have asked preference-giving countries to report how much eligible trade really uses each preference. Without those utilisation rates, nobody can tell whether a rule is helping.

What I am unsure about

• The EU's new GSP regulation has been delayed. I have not confirmed its final text, so the safeguard threshold may differ from the proposal I describe above.
• Bangladesh's graduation date is open. It is scheduled for 24 November 2026. The government has asked for a three-year deferral, to 24 November 2029, and the UN's Committee for Development Policy has supported the request. When I wrote this on 8 October 2026, I had not seen a final decision from the UN General Assembly.
• Each change has a cost for the preference-giving country. I have not estimated those costs.

The next post looks at why this matters so much for Bangladesh's garment sector.