Model Building and Planning in India

Ever since the First Five Year Plan, we have utilised models in order to channel resources for achieving objectives of higher growth, establishing strong capital base, strengthening import substitution, reducing poverty, increasing foreign exchange resources and so on. The early plans made use of Harrod-Domar model and the Feldman-Mahalanobis models. The models used for planning purposes were largely taken from economic theory, which were then adapted (hopefully) by the Planning Commission. Owing to changes in the structure of the Indian economy, the nature of modelling has also undergone various alterations. This post highlights certain issues in the macro-modelling that was done for the 11th Five Year Plan, chiefly based on the publication by the Planning Commission Macro-Modelling for the Eleventh Five Year Plan of India edited by Kirit S Parikh.

It is incorrect to argue that planning in India has become redundant after the 1991 reforms. These reforms provided freedom to the firms with respect to what to produce and how to produce them. As the recent global financial crisis has shown, unregulated finance can lead to unfavourable outcomes for the financial as well as the real sectors of the economy. Also, significant divergences in income and wealth are being reported. This is the case, especially in India which is home to some of the richest and poorest people in the world. As Parikh writes, ‘As long as disparities in income, endowments and wealth persist, access to public goods and services is uneven and infrastructure paucity is there, we need active government policy. We need planning.’ [Planning Commission 2009, 16]

The 11th Five Year Plan’s goal is to have ‘Faster and More Inclusive Growth’. The basis of this goal is that the growth of GDP is treated as a necessary and almost sufficient condition for improving livelihoods. We know that markets exclude those without adequate purchasing power. And growth in GDP mainly results in an increase in the extent/size of the market. Unless appropriate systems are in place, ‘trickle down’ does not take place. Hence, an outline of how ‘inclusive growth’ can take place needs much greater attention. And it is disappointing to see that employment generation is not considered as a central objective. For the first time, the inputs of the 11th Five Year Plan have been provided by a ‘modelling forum’. The forum consists of researchers from NCAER, IGIDR, IEG and ISI Bangalore apart from the in-house team of the Planning Commission.

A formal model is constructed for the purposes of planning because it makes the assumptions transparent, ensures consistency and provides insights into the inter-relationship between various actors and sectors in the economy. These models in turn borrow concepts, categories, functional relationships and links from the paradigms in economics. The paradigms that have influenced modellers, according to Parikh are Input-Output, Walrasian, Neoclassical, Keynesian, Structuralist, Vector Auto Regression/Error Correction, New Neoclassical and Dynamic Stochastic General Equilibrium. The most obvious drawback of this classification is the mix-up of paradigms with tools of economics. For instance, IO framework, VAR models and DSGE models are only tools. For our later exposition, it would be helpful to point out the important characteristics of each of these paradigms/tools.

Input-Output: Usage of inputs in fixed proportion

Walrasian: Optimising behaviour of economic agents

Neoclassical: Pricing through supply and demand mechanism and full employment at prevailing wage rate

Keynesian: Underemployment equilibrium

Structuralist: Imperfect markets and incomplete monetization of the economy

Vector Auto Regression: All variables depend on lagged values of all variables and data speak for themselves

New Neoclassical: Microeconomic foundation of macroeconomics- importance of information, expectations and contracts

DSGE: Forward looking and optimising economic agents

In the modelling for the 11th Five Year Plan, six different models with different analytical approaches have been used. The models are a) Perspective Planning Division’s In-house Model, b) A VAR/VEC Model from ISI, Bangalore, c) A General Equilibrium Model from IGIDR, d) An Econometric Model from IEG and e) Macro-Econometric Model of NCAER. The various model scenarios show that the direction of policy shocks are similar ‘though the structure and philosophy of the models are different’. However, this is not such a shocking or an interesting revelation. For everybody knows that oil price shocks have a negative impact on the GDP growth rate and that the global slowdown affects the GDP adversely. The only result of interest is that of an increase in NREGs by 1% of GDP. This will result in an increase of around 0.35 to 0.5 GDP percentage points. However, it must be noted that this is based on ‘a general equilibrium model in which it is assumed that the adjustments to the new equilibrium are completed in one year. Thus, the impacts may be overstated as in reality this may not be the case’.

Overall, it seems that the assumptions of the various models are far from our Indian reality. There is no attempt at including the unorganised sector. And it is very clear, from the recent evidences from neuroeconomics, experimental economics and game theory that individuals are not rational optimizing machines. Instead, we are more moved by social concerns and we exhibit a pro-social behaviour which is norm-based. There is no explicit move to analyse employment generation. Nor is there the necessary focus on agriculture. It is stated that agriculture needs to grow by 2.4 % to 4 % so as to achieve 9% GDP growth rate. One cannot help but wonder whether the objective of economic planning in India is only about the GDP growth rate! And as to the use of VAR models, the generators of the model argue that since there is ‘no real basis to say which variable is endogenous and which is exogenous’, they adopt a ‘general equilibrium approach, where everything (except rainfall, of course) depends on everything else.’ [Planning Commission 2009, 88] An easy route indeed!

This publication by the Planning Commission is a must read for all those who are interested in understanding the Indian policy making. Also, it provides the crucial link between policy and theory. Hence, making the study of economic theory very significant, especially for policy makers. It will also be of interest to students and practitioners of time series methods using the VAR framework.


Planning Commission (2009), Macro-Modelling for the Eleventh Five Year Plan of India, edited by Kirit S Parikh, Academic Foundation: New Delhi.

Author: Alex M Thomas

A passionate student of economics!

2 thoughts on “Model Building and Planning in India”

  1. The United Nations Development Programme International Policy Center for Inclusive Growth (IPC-IG) will be hosting an Academic Forum on the 12th and 13th of April 2010 in correlation to the India-Brazil-South Africa (IBSA) summit to be held in Brasilia, Brazil on the 15th of April 2010.

    The theme will be social development strategies for inclusive growth.

    Topics at the Academic Forum include role of cash transfers, role of employment programs/policies, health innovation, intellectual property rights and access to essential drugs. Specifically:

    How do conditional cash transfer programs improve life of millions living in extreme poverty?
    What are the most relevant experiences from other parts of the world that can be effectively applied to IBSA countries?
    How IBSA countries can cooperate and share experiences if they adopt different approaches for CCT programs?
    What are the consequences of the creation of new groupings of emerging economies with a strong emphasis on South-South cooperation? How do we get civil society involved in high level political debates?
    We want to hear your opinion!

    Please join us at the discussion forum

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.