When should a government agency have autonomy?

Financial Express, 23 July 2010

The ULIPs ordinance has been criticised on the grounds that it infringes on the autonomy of RBI or SEBI. Economic reasoning suggests a case for autonomy for two issues: Monetary policy (i.e. setting the short-term interest rate), and the treatment of individual transactions including licensing, enforcement, etc. There is no case for autonomy of financial regulators on questions of policy. The ULIPs ordinance is a mistake and should be criticised on its substantive policy errors. Infringing on the space of regulators is not one of them.

MoF can perform functions within itself, or it can place functions in external agencies. In either case, Parliament holds MoF accountable. And MPs are held accountable in elections. Where does `autonomy' or `independence' of regulators come in? How do economists or bureaucrats get to wield power?

From first principles, there are exactly two areas where there is a case for independence. The first concerns monetary policy. If a MoF has a say in setting the short rate, then interest rates tend to be pushed down prior to elections, which kicks off inflation after the election. Independence for the central bank was invented in order to prevent monetary policy from inducing this `election cycle'.

Doing this properly requires a singular bargain between legislators and the unelected economists of the central bank. On one hand, legislators give independence to the central bank. But no economist should be given power without accountability. Hence, the bargain that is struck involves two elements:

  1. The function of the central bank is pared down to the bare minimum: the setting of the short-term rate.
  2. The central bank has to submit to accountability mechanisms. The cleanest form of accountability is : inflation targeting. The central bank submits to the accountability process of achieving a quantitative inflation target in the medium run, in return for central bank independence.

Our Indian institutional arrangements on monetary policy are rooted in legislation which dates back to 1934. None of these concepts were understood then. Hence, terms like `independence', `accountability' and `inflation targeting' are not found in the RBI Act. RBI today has little independence on the one place where independence matters: the setting of the short rate. No RBI governor can set the short rate to a level which the PM and the FM do not accept.

RBI cannot be given independence in the present arrangement, where it performs an improbable array of extraneous functions, none of which require independence. As every manager knows, a worker with `multiple objectives' is someone with an ample supply of excuses for failure: he cannot be given autonomy and has to be micro-managed.

The second dimension of independence lies in specific transactions. This applies across both finance and infrastructure regulators. The argument is made that corporations are prone to bribe politicians and ask for favours on licensing or enforcement. In order to address this, the leadership of a regulator has to comprise top quality people in terms of both domain knowledge and ethics. This requires three things:

  1. An incorruptible recruitment process which brings in top quality leadership teams,
  2. A work culture at MoF where politicians and bureaucrats just do not discuss specific transactions with regulators, and
  3. A strong culture of the rule of law at regulators, to ensure that the bribes don't merely shift to the level of the regulator.

We are hence able to articulate two dimensions where independence is valuable: setting the short rate, and individual transactions. Policy issues are not covered under this umbrella. If there is a disagreement between (say) SEBI and MoF on policy issues, there is no independence trump card which asserts that SEBI must always have its way. Indeed, only MoF can lead the resolution of differences between SEBI and IRDA, either in an informal fashion or through the proposed FSDC.

The ULIPs ordinance is a mistake because IRDA has the wrong institutional culture. By handing over the ULIPs question to IRDA, we are likely to perpetuate sales practices through which insurance companies and agents exploit taxpayers (owing to tax subsidies for insurance products) and customers. This is the substantive criticism of the ULIPs ordinance. That the ULIPs ordinance `infringes on the autonomy of regulators' is a non-issue.

The world over, there is great criticism of finance on the issue of the power that financial firms exert over the policy process. The ULIPs scandal is India's third taste of this problem. There will be many others in the future. While we know a lot about how financial policy and regulation ought to work, policymaking in democracy is tugged by narrow corporate interests. Our challenge lies in harnessing the dynamism and genius of a private financial system, while avoiding the venality and theft which comes from regulatory capture. This will be achieved by focusing not on regulatory autonomy, but on the appointments process, in blocking meddling by government in individual transactions, and in demanding rule of law.

Back up to Ajay Shah's 2010 media page
Back up to Ajay Shah's home page