Why infrastructure is the defining issue?

A single unified definition of infrastructure should not be attempted. It is neither desirable, nor useful, writes Vinayak Chatterjee.

Infrastructure needs to be defined to make policy interventions meaningful as we prepare for massive investments in the 12th Plan.

Why, you may well ask, do we need to define infrastructure at all? Well, a practical definition is required for the following seven reasons:

One, it is an umbrella word that is being loosely used to describe multifarious economic activities.

Two, serious financial interventions with public policy overtones are regularly being crafted. Examples include viability gap funding, easier norms for bank lending, insurance and pension funds being cajoled in this direction, infra-bonds, separate class of infra-NBFCs, and multi-billion dollar funds.

Three, taxation issues and tax breaks are designed to encourage activities.

Four, the Land Acquisition (Amendment) Bill, 2009, provides for the sovereign acquiring 100 per cent land under 'eminent domain' for the purpose of developing infrastructure.

Five, the nation measures its performance in terms of the GCFI (gross capital formation in infrastructure) as a per cent of GDP as well as in absolute numbers. We must know what it is that we are measuring.

Six, various bodies from the apex Prime Minister's Infrastructure Committee to state-level infrastructure  boards must know their playing field.

Seven, the move towards a new legislative architecture to create truly independent regulatory authorities for infrastructure supposes that we know what we desire to regulate.

Attempts to define 'infrastructure' use one or more combinations of the following characteristics:

(i) Essential inputs to the economic system

(ii) One-time lumpiness of investment; or high sunk cost

(iii) Natural monopoly

(iv) Non-tradability of output

(v) Economic versus financial return -- the "externalities" argument

(vi) High asset specificity

(vii) Non-rivalry (up to congestion limits) in consumption

(viii) Network character

(ix) Content versus carriage

(x) Large land agglomeration projects with trunk-infra requirements

(xi) Public versus private goods

Pure economic theory has the ability to confuse and obfuscate, and practitioners will have to slice through the warm fuzz of theory to arrive at practical, workable definitions.

Here are some suggestions for our policymakers to think about.

A single unified definition of infrastructure should not be attempted. It is neither desirable, nor useful. Rather, the Indian reality suggests having five clusters as below:

Core infrastructure

This includes transportation (roads, railways, airports, sea-ports, inland waterways) and energy (generation, transmission, distribution).

Urban infrastructure

Water, sanitation, sewerage, urban-transport, city-energy distribution, transport terminals, warehousing and logistics parks fall under this category.

Rural infrastructure

This would include irrigation, roads, energy, cold chains, mandis and drinking water.

Land-intensive

Special economic zones, industrial parks, new townships, industrial cluster development, IT parks, domestic economic zones, logistics and warehouse parks, and industrial corridors are a part of this kind of infrastructure.

Social infrastructure

Healthcare, education, leisure and entertainment, retail, tourism, housing, exhibition and convention centres, hospitality, IT, and telecom would fall under this category.

Public policy may be designed in such a way that each cluster is addressed in a different manner for providing support in line with specific requirements. Performance-monitoring cluster by cluster would also be more revealing and relevant.

Three clarifications are also in order.

First is universal intermediates. There is a school of economic thinkers who argue that infrastructure should be defined broadly to include 'essential inputs to the economic system'.

This argument would see plants and installations that produce coal, steel, cement, fertiliser, petroleum, natural gas, iron ore, (with possibly other minerals) included as infrastructure. Such 'universal intermediates' should be left out of any infrastructure definition.

Second is carriage versus content.  Water is content. But a water pipeline is carriage. Oil is content but an oil pipeline is carriage.

A port is carriage, but a ship is content.

A road is carriage, but trucks are content. There are many representations to the government where 'content' businesses want to be defined as infrastructure.

Here again, attempts to define infrastructure should stop at carriage and not include elements clearly in the content domain.

Finally, operations & maintenance (O&Ms) along with asset creation.

Currently, much of the nation's focus is on asset creation. But, infrastructure is as much about making assets work smoothly as it is about meeting expected service delivery standards.

Making infrastructure assets 'work' is the responsibility of operations and maintenance (O&M) companies.

Dredging regularly is as important as building a new port.

Maintaining stretches of roads and highways as well as collecting toll are as important as building a road. Running the water supply system of a city is as important as laying the pipelines and pumping stations.

Therefore, infrastructure O&M activities must also be added to the definition of infrastructure.

With the run-up to $1,000-billion investment in the 12th Plan, the defining moment is now!

The author is chairman of Feedback Ventures. Views expressed are personal