The article that Leela posted is a fascinating one.  AsifY bhai has been goading me, I see, to post on arbitrage.  Alas, there is very little time - a big paper is due tomorrow - so all I am going to do is post a few quick notes. These might be even less coherent than my usual babbling.

* My usual attitude towards subsidies on goods like fertilizer (or for that matter, petrol) is that given the porousness of borders with India, if the post-subsidy price is set lower than the (post-subsidy) price in India, there will be smuggling into India. Back in 2005, the word on the street was that a big chunk of petrol consumption in West Bengal was being subsidized by the Bangladeshi government, since petrol and diesel was being smuggled at will to India. (Talk about Bengali unity!)

*  Aside from arbitrage through smuggling, it’s hard to do subsidies really well to actually target the ones who need it most. The history of subsidies in East Pakistan and Bangladesh will reveal this to be true.

* More importantly (for the kinds of things that fundamentally get my juices flowing), subsidies encourage rent-seeking behavior. In our country, that has often meant abasement and servility to elites with access, and corruption.

*  Basic microeconomics tells you that it makes more sense (in a utility-enhancement sense) to transfer wealth to the individuals who will receive the subsidy directly than to do so indirectly through subsidies (this is even without taking into account the costs of the system…) I can see why you’d want wrinkles in the basic model though, so I won’t push this point too much.  Sample thoughts on what kind of wrinkles to add: Dynamic inconsistency. Information and knowledge created through subsidized programs. A misalignment between the incentives of the individual and the government (with the government, or “society” wanting to push certain ends over the other). Externalities.

*  How do you explain the smuggling into Bangladesh from India? There likely is a market for fertilizers (outside of the government channels), and the difficult supply of government fertilizer is driving the price up - beyond what fertilizer sellers can get in India + transport and transaction costs. Ergo: Smuggling into Bangladesh.

The presence of this market also has implications for how well these subsidies can be targetted in the first place - as I am sure readers can work out for themselves.

* The slippery slope argument that Tacit makes does not quite work. The simple answer to why give subsidies for fertilizer but not for electricity (though I do believe rural electricity is  subsidized in the ‘Desh) or say, cotton candy, might very well be because the returns from such a subsidy, taka for taka, are higher.

Bottom line: I can see why subsidies might seem attractive. But it’s hard to get subsidies right, and reach them to the right people.