Great question!

1. Achieving a direct matching by the gvmt (even if it would be only with, say, a +25% match-factor or so, to keep it roughly in line with what tax-deductability means), instead of tax deductability, could indeed be more just, removing the bias you mention that unnecessarily favors the rich. Spot on imho.
2. That said, democracies seem to love "tax deductability": stealing from the state that feels a bit less like stealing. So, deductability can be the single mostly easily acceptable policy. If so, it might pragmatically be worthwhile to support it, despite the negative redistributional consequence just mentioned.
3. The best action might be to try to set up local EA charities that can easily get certified for tax deductability in Belgium, and use the money either directly to support the intl EA organizations, or, if that is difficult, then in the worst case simply to support similar work in parallel (?).
4. Mostly independently of 1. vs. 2., whether the (donation elasticity-adjusted) average donation is more effective than gvmt tax collection (or reducing standard gvmt tax burden*), feels difficult to say, and will depend a lot on how much you value different types of social goods. Atm, most donations will not be EA type donations, but one might expect people to give to causes often significantly better than for tax revenues, so I'd personally rather err towards the pro-deduct (or pro-match) side so far.

*It would be wrong to consider only 'tax revenue lost' for the gvmt as effect of the tax deductability. In expectation, in a simple model, gvmt will in the medium term partly respond with (i) lower expenditures, and (ii) increase standard tax rates in response to higher tax deductions. 

Btw, I personally would not worry about the €25 threshold. Avoiding to register/count too small sums seems a reasonable thing, even if you're right that it becomes less relevant in the digital world.

Agreeing with others that this is a good question - and it's not simple. (Because, of course, policy debates should not appear one-sided!)

Two of the key reasons I'm a fan of tax deductibility is because it's a clear signal about whether something is a charity, and because it's a behavioral incentive to donate - people feel like they are getting something from donating. (Never mind the fact that they are spending - it's the same cognitive effect when people feel like they "saved money" by buying something they don't need on sale.)

On the other hand, I think Rob Reich is right about this, and we'd be better off switching to a system that doesn't undermine our taxation system generally - though tax deductibility is far from the only culprit, and if this is a single change, the other loopholes are less publicly beneficial as side effects, so I would guess it's a net negative unless coupled with broader reform.  Note that I haven't read Rob's latest book, (he is an incredibly fast writer!) and maybe he talks about this. If not, I'd be interested in asking him for his take.

Given all of this, I don't have a strong take on this - and short of general reform, I'd at least be in favor of expanding tax credits for EA charities, so that they aren't relatively disadvantaged as places to give.

Personally, I have some small confidence that removing tax deductibility for all charities in all countries would be better than the system we currently have – charity could more easily be more international and people worry less about chasing deductions and more about impact. In countries where tax-deductibility is a small factor I think it's hard to justify work on expanding it (if most people don't care). In countries where it's a big deal then it's worth setting up charities in those countries or trying to expand the purpose (e.g. animal welfare charities in Australia excludes most things that would help farmed animals).

I prefer matching systems like Gift Aid in the UK over tax-deductibility, but still think that charity might be better off being just treated as any other purchase (probably without goods/services taxes though).

Thank you for your replies !

I also thought for a time that removing deductibilities could be good, but to me it all boils down to the price elasticity of donating. When it’s high enough, it does increase the overall amount donated, because people increase their donations. We just have to channel it to the efficient causes rather than remove it imho. It seems to me that we need answer to questions such as : 

1. Are there large crowding out effects (state disinvest because private sector invest) ? My impression is that even though private charities have advantages (innovation, niche focus), there are arguments for a system in which the states increases its budget (reducing deductibility, taxing more) in order to assume part of the missions currently assumed by the private. This would mean more control by citizens over what are the priorities (state budget, democracy), and also more fiscal justice (instead of rewarding the more wealthy with deductions, they would be taxed more for redistribution). Now I for one also think that the state is not the most "effective" of altruists, mostly because its focus is on its resident. 
2. Is the impact achieved through increased donations via deductibility greater than the good that this money can do if it remains in the state budget ? Certainly depends on the countries, priorities, governments, etc. If say all of a sudden the AMF grants deductibility, would you increase your donation ? To me a way forward could be to reach out to the charities that have become deductible in new countries and analyse their data to see the impact on donations, and measure the amount lost in state revenue. The worst case would be a charity with a large amount of donations from citizen of that country being granted deductibility (huge donations : huge spendings for the state) but seeing little increase in donation (few benefit for the cause).
3. Why is it that so few states use the matching mechanism ? It seems to have advantages over deductibility without much disadvantages, so certainly I’m missing something here.

Thank you all for your input on this topic, I appreciate it !