Rathin Roy: Need for simple macro-framework

Economist Paul Krugman points to an interesting conundrum in the economics profession in a recent post. Young economists are encouraged to enter the theoretically seductive world of sophisticated macroeconomic theory to explain both aggregate demand and supply by postulating the microeconomic behaviour of agents and then making use of freely available data and cheap computing power to validate their postulations. Mr Krugman points out that these initiatives have not yielded much payback if one is looking for operational models of the macroeconomy. For example, the current favourite - output-gap modelling - recommends macroeconomic policy actions based on the gap between maximum potential output of an economy and the current level of output. But when reality confronts economists' "output-gap" models, what is adjusted is potential output, which is a judgement call that uses no more sophistication than old-fashioned calculations of an economy's production possibility frontier or equilibrium output derived from a simple IS-LM framework, that newly minted economics PhDs look at with contempt.

Policymakers, therefore, continue to do macroeconomics using simple constructs that make explicit the policy levers available for action. They, quite sensibly, use simple "old-fashioned" macroeconomic frameworks, while making the best of improvements in econometric technique. For example, the macro-econometric model used by the Bank of Japan is, in its own words, "a demand-oriented, traditional Keynesian model with IS-LM framework. It adopts recent developments in econometrics such as cointegration and error-correction procedures". So there is no great technical barrier to Indian policymakers doing the same, in the public domain. But does this happen?

Macroeconomic policy is crafted by the Government of India and the Reserve Bank of India (RBI). Historically, the Planning Commission took economic growth and sectoral composition as aspirations. They then took the savings and investment rate in the economy and the incremental capital output ratio as the key policy variables to determine how to achieve these aspirations, subject, additionally, to a foreign exchange constraint. With the transition to a market-based economy, the Commission adopted a multi-pronged approach - an in-house model that was a simple set of accounting equations that worked out the required savings investment rates to achieve the aspirational growth rate of GDP. Academic institutions were also contracted to provide general equilibrium, Vector auto-regression, macro-econometric, and IS-LM type structural models. All these models were in the public domain1.

RBI has been much more opaque. Staff have published models of the Indian macroeconomy in RBI and academic publications, but the institution has never owned a publicly available macroeconomic framework, unlike many other central banks.

With the demise of the Planning Commission, there is no policy node within the Government of India that is tasked with providing an analytical macroeconomic framework; this is not expected either of the NITI Aayog or the Ministry of Finance (MoF). The key MoF policy publication, the Economic Survey, is not (and was never intended to be) an analytical policy tool. Volume II is a "survey" - an exercise in economic geography. Volume I, a great public-policy document, uses analytics solely to establish the validity of certain stylised facts. There is no macroeconomics in the Survey. RBI publishes largely narrative commentary on the Indian macroeconomy, but not a public analytical framework which can be challenged or endorsed.

Macroeconomic analytics for policymaking cannot be an end in itself. It must be comprehensible to an audience wider than economics PhDs and focus clearly on the policy variables that can be used to influence economic outcomes. Institutionally owned models make explicit how the institution is thinking about the macroeconomy, especially relationships between economic variables that can be challenged by those who think differently. Two examples: Increasing government domestic borrowing can raise investment but can also raise the cost of capital for the private sector and crowd it out, or raise inflation by increasing aggregate demand. Policymakers need to take a view on this and a macroeconomic framework would make this view explicit together with the analytical and empirical rationale. RBI uses the repo rate as its instrument to target inflation. The effectiveness of that instrument depends on assumptions regarding the impact of the instrument on agricultural prices, and on the monetary transmission mechanism.

These things have been discussed elegantly and convincingly in papers authored by incoming and former governors,2 but these have not been owned by RBI.

An explicit, simple and public analytical macro-framework imposes accountability on macroeconomic policymakers. Conversely, the absence of such a framework does not allow for a common benchmark around which people can agree, disagree and utilise the collective wisdom of the policy community to the maximum extent. It disciplines the policy community by forcing them to articulate their vision, ideology and analytical habits rather than what happens now, which is the use and manipulation of data to establish the case for one's favourite stylised facts. Until we acquire this discipline, stakeholders will be sceptical of the credibility of India's economic policy community and macroeconomic policy institutions, and deservedly so.

1. Planning Commission (2009) Macro-modelling for the Eleventh Five Year Plan
2. Rangarajan, C and R R Arif (1990) "Money Output and Prices: A Macro-econometric Model" Economic and Political Weekly, 837-852; Urjit Patel and Amartya Lahiri (2016) "Challenges of Effective Monetary Policy in Emerging Economies", RBI Working Paper Series 01/2016

The writer is director, National Institute of Public Finance and Policy
rr1@nipfp.org.in