One way to think about incentives is the following. Invite three individuals Amar, Akbar and Anthony for a drink. There is a simple rule. Amar will pay for the drink. Akbar will however choose the drink. But, in the end, Anthony will consume it. How will the situation play out?
What are the incentives when the three entities of Payer, Chooser and Beneficiary are completely different? The most efficient is when all three are one. You decide, pay and benefit from it. The worst is probably when all three are different.
Now think of the characteristics of spending by the government. Think education. The government spends for it, and you benefit (atleast, should), but your choice is not involved. Think charity. You choose and pay, but others benefit. Run through the various Acts of the government and you will invariably find that either choice has been circumscribed or the government has taken the mantle upon itself to provide choice. And in most cases, this “circumcision” of choice has no relation to the main function of a government, provision of public goods.
This is what Ajay Shah talks about in his blog post, A great new atlas of India is now available. Maps! The government pays for it, chooses how to deliver it, and you are supposed to benefit from it. Choice is limited because other players are severely restricted to provide them. Are the incentives to deliver there? How is restricting the provision of maps a public good? He goes on to talk about the public good mess of the government. I would highly recommend reading his blog if you are interested to know more about the Indian economy. For a not-too-short synopsis of his views read Moving Beyond State Capitalism. The author is Vijay Kelkar but I can see his collaborator Ajay Shah’s strand of thought through it. This was the question that he had prompted, and I know of only one organisation that is working on it. What percentage of the government’s actual spending is on public goods? How much is being spent by the state for ostensible private goods? (By now you know who has the choice of providing the required data!)
One could have theoretical arguments about what is a public good, but can an index be made that can customise government spending according to your theory of public goods. Why is it important? It will provide a sense of direction to a government and will hopefully spur competition between state governments to deliver more public goods. How they provide it is a different question!