These small gains of $16bn will come at considerable costs. The losses in tariff revenues alone will be more than twice the benefit, or $32bn. Developing countries rely on tariffs revenue for 20 per cent of all government revenue, and therefore development policy.

I must be misinterpretting this.  Someone please set me straight.

The World Bank report that came out recently to bolster up the free-trade case at the Hong Kong talks is a book of 425pp, and it's not easy to pin clear statistics down since it is arguing in favour of an agenda and mostly presenting data in synthetic form (from what I've had time to see). What further complicates things is that it offers a whole string of scenarios, and the supposed results depend on which scenario is chosen.

But here are some points from the general presentation:

The potential gains from further global trade reform are huge. <snip>
Freeing all merchandise trade and eliminating agricultural subsidies are estimated
to boost global welfare by nearly $300 billion a year by 2015.
<snip>
Developing countries could gain disproportionately from further global trade
reform.
The developing countries would enjoy 45 percent of the global gain from
completely freeing all merchandise trade (table 1.3a), well above their current
share of one-fifth of global GDP. Their welfare would increase by 1.2 percent,
compared with an increase of just 0.6 percent for developed countries.

Note that the projected $300bn per annum gain is not between now and 2015, but from 2015 on. Second, that it has been arrived at counting completely freeing all merchandise trade and not just agricultural trade.

Poverty could be reduced under Doha.
Under the full merchandise trade liberalization
scenario, extreme poverty--those earning no more than $1 a day--
would drop by 32 million in developing countries in 2015 relative to the baseline
level of 622 million, a reduction of 5 percent. The majority of the poor by 2015
are projected to be in Sub-Saharan Africa, where the reduction would be 6 percent.

Under the Doha scenarios reported in table 1.8, the poverty impacts are far more
modest. The number of poor living on $1 a day or less would fall by 2.5 million
in the case of the core Doha scenario 7 (of which 0.5 million are in Sub-Saharan
Africa) and by 6.3 million in the case of Doha scenario 8 (of which 2.2 million
are in Sub-Saharan Africa). This corresponds to the relatively modest ambitions
of the merchandise trade reforms as captured in these Doha scenarios. If only
agriculture was reformed (Doha scenario 1), there would be much less poverty
alleviation globally and none at all in Sub-Saharan Africa. This shows the importance
for poverty of including manufactured products in the Doha negotiations.

Note the emphasized (by me) sentence. The obliteration of African poverty (in particular) does not depend on a simple opening-up of agricultural markets. It's the World Bank saying that.

I'm glad Kevin P Gallagher has written this, I hope he crunched the numbers right, and I hope there'll be further discussion of this in the FT and elsewhere.

There is trade between specific companies having physical presence in 'France' and 'Germany' and there is trade between companies only having a presence in 'France' and 'Germany.'  But if the trade is between an 'American' company in 'France' and a 'Polish' company in 'Germany' which has the goods shipped directly to a 'Japanese' company in England ... who is trading with whom?

Now let's say there is a 'French' company wholly owned by a trans-national headquartered in 'Japan' trading with a 'German' company also completely owned by the same 'Japanese' trans-national.  Is this trade or merely shuffling inventory and accounting records within the trans-national?

No one could have predicted