Nucifora's piece expresses concern that those who benefit most from the subsidies handed out by the Mozambican government (particularly fuel subsidies) are not poor. This is based on the share of the subsidy money that goes to each group. The reason it works out like this is because blanket subsidies are used. He suggests that targeted social protection policies would be better.
Absolutely. Targeted social protection is always more bang for your buck. But it also requires ongoing efficient administration, an excellent information base to work from in order to target effectively, very strong monitoring, expertise in social protection options and ongoing surveys and research to ensure that social protection packages keep pace with changes in real costs. I very much doubt Mozambique has that kind of capacity. I'm not sure South Africa has that kind of capacity. It's also not nearly as desirable across the board as it may seem from a purely economic perspective. For example, Nucifora says that:
On fuel, the government announced the intention to move to subsidizing urban transport which is clearly less costly and better targeted to the poor than the current blanket diesel subsidy.63% of the population is rural (2008). Many of them are poor. Mozambique is also a large country (in length if not width) with long, long stretches of bad road between scattered towns. Moving people between towns and around rural areas, especially, takes a long time, is uncomfortable and is often unpredictable. Plus, the roads get washed away every time it floods. Making it cost-effective (or simply cost-feasible) for people to move from one place to another sounds like a good idea. It especially sounds like a good idea in terms of access to health-care and education, even if it isn't high priority for the economy (although even that seems unlikely). Restricting this to the urban areas where the minority lives seems like a bad idea.
Nucifora goes on to argue that the current development model of the country - "public investment in infrastructure, attracting mega projects (which are capital intensive), and increasing access to public services" - isn't working because it is not producing jobs. For once, I agree with the more conservative side of this debate: Governments do not create jobs, the market does. That may well require the reforms mentioned: easing (some of) the regulations on economic activity, simplifying tax systems, removing (some) restrictions on land tenure. But are these incompatible with subsidies (targeted or not)?
The conclusion of the article seems to suggest that subsidies and economic reforms are mutually exclusive. The implication is that subsidies in some way prevent the government from implementing economic reforms on the regulatory side. Which in turns seems an awful lot like the old argument that the best way to help the poor is take away the handouts so that you can focus on economic growth so that you can create jobs.
If targeted subsidies are possible in Mozambique - and can be implemented effectively without leaving the 75% of the population living on $1.25 or less a day worse off- then they're a good idea. If not, blanket subsidies are a necessary evil. Either way, there is no reason this cannot happen at the same time as economic reform. There is also no reason economic reform should be used as an excuse for suggesting that the government stop doing what it is supposed to be doing, such as investing in public goods like infrastructure, improving (and improving access to) public services and providing social safety nets for the most vulnerable Mozambican people.