Things that the government should not do

The job of government is to address market failures: public goods, asymmetric information, market power, externalities. In addition, a certain focus on welfare programs is the inevitable consequence of universal suffrage. We can debate the appropriate design, and scale, of these programs.

That leaves a third class of issues where neither justification is present: where the government is doing something which is not grounded in market failure and it does not have political salience. The standard Indian example is: Air India. There is no case for government to be in that business.

The impending demise of the US Export-Import Bank [ht: Marginal Revolution] provides another example: Exim Bank [website, law].

Financing and risk-taking in connection with international trade or cross-border investment is the ordinary stuff that the financial industry does. The self interest of financial firms will produce these services. There are numerous anti-competitive provisions in the present policy framework, which are inhibiting entry of financial firms into a diverse array of businesses. Removing these barriers will enhance competition and create a diverse array of services that meet the requirements of customers.

There is no case for the government to subsidise exporters, but even if that were desired, there are more efficient ways to do this rather than subsidised credit.

The only role for government in finance is that of addressing market failures. The 9 components of the Indian Financial Code are:

1. Consumer protection: Left to themselves, financial firms will mistreat consumers. This is the heart of how we should think about financial economic policy.
2. Micro-prudential regulation: Left to themselves, financial firms will take on too much risk and fail too often, which will hurt unsophisticated consumers and impose externalities upon bystanders.
3. Resolution: The ordinary bankruptcy process does not work for financial firms, where failure can be disruptive. The government is required in the field of resolution, to identify financial firms that are not viable before they are insolvent, and gracefully handle the situation in ways that don't hurt innocent bystanders and protect unsophisticated consumers.
4. Systemic risk regulation: This is about seeing the woods and not the trees. Financial regulation induces pro-cyclicality; we need ways to combat this. We need ways to reduce the probability of systemic crises, and better ways to deal with them when they do come about.
5. Public debt management: We need an agency which will do investment banking for the government, and figure out the right ways to organise the market for government bonds.
6. Capital controls: We need to have the rule of law, and equal treatment of non-residents, in the working of capital controls.
7. Monetary policy: We need an accountable central bank which will safely produce fiat money, through a sound monetary policy process. 
8. Development and redistribution: Finance is being used as a tool for redistribution, and this requires sound foundations of governance. There are also some public goods of market infrastructure which require policy attention.
9. Contracts, trading and market abuse: These are adaptions of the standard law on property and contracts which are required for insurance and securities markets.

These nine areas are the only  roles required of the government in finance.

Once we decide that we want a financial agency (e.g. the Public Debt Management Agency, or the Resolution Corporation), full thinking is required on how to make the agency work properly. How to create checks and balances, and appropriate incentives, so as to produce high performance? A well drafted law is necessary but not sufficient in getting to a high performance agency. A badly drafted law can disrupt the organisation, and a well drafted law can nudge things along in the right direction, but constructing a high performance agency is a challenge in its own right, where a distinct set of issues are faced.