Infrastructure and Development: The present Challenges - PowerPoint Presentation

Slide1

Infrastructure and Development: The present Challenges

Ajitava Raychaudhuri

Department of Economics

Jadavpur University

Kolkata 700032 Slide2

Public Capital and Infrastructure –definitional issues

Does public capital mean a stock variable like private capital or does it mean the services which flow out of this government owned capital?

Is public capital a pure public good of Samuelson’s variety which is non-rival and non-excludable?

If we equate public capital with infrastructure, how do we measure its quantity?Slide3

Infrastructure and public capital is interchangeably used since contribution of private infrastructure is difficult to measure.Slide4

Growth Rates under planner’s and individual investor’s solution

Standard economic theory suggests if we compare the two growth rates the two growth rates, clearly:Slide5

The main reason is that competitive solution considers only the private marginal product of capital whereas the social planner considers the social marginal product which does not have the tax deduction from the marginal product. Is there any way to bring about pareto optimal growth in the competitive solution? It is possible if the government replaces the distortionary income tax by either a lump-sum income tax or simply a consumption tax which is by definition lump-sum since it does not vary with income.Slide6

Barro and Sala i Martin (1990) discuss a case where a public good is rival but not excludable. A common example is total highway mileage, which is the services from a publicly owned infrastructure denoted as

G

. The output is total highway traffic, and let this be denoted by

Y

. The problem is congestion which precludes the use of highway for some individuals at times of high traffic. Slide7

Here lump-sum tax produces the opposite result than the earlier model of Barro. The reason is that lump-sum tax is equivalent to zero user fee so that it does not discourage individuals to overuse the public capital’s service, hence is distortionary. However, the proportional tax is equivalent to user fee which internalises the congestion distortion Slide8

Public Private

Partnership

–

Emerging Issues in Modelling

The major issues in this emerging literature on PPP may be summarised as below:

What should be the conditions when government procurement through its own agency/department through hiring of different contractors is better than PPP which bundles construction and operation with a single contractor?

What are the implications of the mode of financing the private sector investment under PPP?Slide9

PPP versus conventional government procurement

Assume that an infrastructure project has an accounting (or revealed) cost given by

C =



- e

, where



is an intrinsic efficiency parameter of the firm, such that ‘high



’ firm has low productivity and vice versa, and

e

is an unobservable effort level. Here

C

is composed of two unobservable quantities, namely hidden information



and hidden action

e

. Slide10

The first is the case of adverse selection and the second is the case of moral hazard. Also suppose there is cost of effort denoted by



(e), having



(e)>0 and



(e)>0. Thus, any cost saving generates an additional surplus for the firm. If the firm is provided a fixed sum of money then for any cost saving surplus, the firm is the residual claimant.Slide11

If the firm minimizes its true cost with respect to e,

then productive efficiency requires the condition



(e)=1

. Such an outcome is always achievable through a fixed-price contract. However, a generous contract will attract a high intrinsic productivity firm who will outbid the low productivity ones but a low price contract will do exactly the opposite. So the government has the onerous task of choosing the optimal price for the contract

Slide12

To overcome this while guaranteeing the work is done, the government may have a contract which has a low fixed component and a variable component which is cost-plus if the firm declares that it has a

low



. Thus the

high



firm will still opt for a high fixed price scheme while those with

low



will go for a partial cost-plus scheme. Essentially the firms are offered a menu which reduces the informational rent. Slide13

Hart (2003) has the following scheme in his model – two firms, one called the ‘builder’ and the other called the ‘operator’ compete for an infrastructure project. The government can have a procurement policy which will utilise both the firms, while under PPP, the operation will be bundled and the builder will get the project for both building and maintaining it Slide14

Comparing the government procurement (unbundling) with PPP (bundling), we find that under unbundling the builder

internalises

neither Social Benefit

B

nor Cost

C

, so that there is underinvestment in both socially productive investment like use of best quality materials and techniques in building a

bridge

and

applicable unproductive investment

like promoting

a

design which reduces costs but increases traffic congestionSlide15

In case of bundling, he again fails to internalise

B

but internalises

C

. So in this case there is comparatively more productive investment (although less than first best) but at the same time there is more unproductive investment of

e

too (in fact, it is more than the first best solution). This shows that when quality of building can be well specified but not the consequent operations, government procurement is better, but when quality of operation and maintenance could be completely specified, PPP is better

.Slide16

PPP and Public Finance

There is a belief that once a PPP model (by implication, it is bundled) is adopted over the general procurement, the government is saved of that amount. This does not make sense once one remembers Ricardian Equivalence. The costs incurred by the private party is consequently recovered through user fees and government subsidies (alternatively known as minimum revenue guarantee).

Slide17

Engel, Fischer and Galetovic (2008) show that the form of payment to the private party is irrelevant as far as social planner’s optimum is concerned. What matters is the total payment made to him and not the division between user fee and subsidy. The intuition is that either way the government has to finance either the loss of user fee or provision of subsidy by distortionary tax defined as

(1+



)

for each rupee foregone or spent.Slide18

Empirical work on Public Infrastructure, Growth and Inequality

Munnell (1990) estimated an equation which is: ln Q - ln L = c0 + c1ln L + c2ln Kp + c3ln Ki + c4CU + u, where u is taken to be a first order autoregressive error term. Munnell used two alternative definitions of infrastructural capital- One is what she calls the core infrastructure capital, and it consists of highways, airports, mass transit facilities, electric and gas plants, water supply facilities and sewers; a second measure is more general, and it includes not only the core infrastructure capital, but also non-military public buildings such as schools, hospitals, police and fire stations, courthouses, garages and passenger terminals, and those used in conservation and development

. Slide19

As Munnell pointed out, the initial results indicated an implausibly high elasticity for the public capital or infrastructure, much higher than the private capital. The critics raised such issues as non-stationarity of data suffering from common long-term time trends, and simultaneity bias. Economists have tried to incorporate all of these but the debate has not subsided.

Slide20

Barro himself ran cross country regressions earlier and reported them at the end of his 1990 paper. In Barro’s theory, in case the countries are close to their respective maximal growth, gross public investment (

gi

) as a ratio to total investment including private and public (

i

) should not have significant association with growth of per capita income. Slide21

For 76 countries for which data on public investment is available, the estimated coefficient of regression of average annual growth rate of per capita GDP on

gi/i

, controlling for initial real per capita GDP in 1960 school enrollment rates and other indicators of political stability and market distortions, came out to be 0.014 and insignificant. This implies that the typical country was having public investment (in infrastructure) close to that maximizing respective growth rates. Slide22

There is not much empirical study on the PPP models presented earlier. One study by Hammami, Ruhashyankiko and Yehoue (2006) on PPP did not go into the contract theory issues so much discussed in the theoretical models, but has tried to identify the variables which influence PPP the most in a cross-country framework. The paper used ordered probit and logit analysis among other more standard techniques to ordinally scale the private sector participation in a project in a scale of 1 to 12 Slide23

Empirical work on India

Barro’s approach has influenced some of the studies on this. In a paper jointly written by the author of this paper (Roy and Raychaudhuri, 2009), Barro’s approach is modified to incorporate rent seeking motive of the public agents (like managers of government agencies) in charge of providing infrastructure services. Slide24

The rent seeking arises because it is enabled by law, the infrastructure is owned by government and as such there arises an institutional monopoly rent which the public agents equally divide among themselves

In this model, as already mentioned, it is assumed that the public sector enjoys monopoly power of supplying public capital. Government provides public capital through some decentralized public agents. Role of the government here is only to collect and transfer funds to the agents

. Slide25

As government does not know the actual cost of producing public capital government allows the agents to impose some user fees for the public capital. The user fee as defined above is equal to the value of marginal product of the public capital net of taxes. As a result they earn some monopoly rent over above their wages.Slide26

Now to discipline the public agents government transfers funds to the agents so that they allocate resources as efficiently as possible. Thus the premium is comprised of monopoly rent as well as the incentive given by the government. Monopoly rent is not taxable whereas incentive is taxed by the government. The premium function under decentralised agents’ model is defined as the difference between the revenue earned from user fee and incentive net of taxes and cost of providing public goods.

Slide27

In this paper the period considered is 1981 to 2005, using data on seven states of India

The results indicate that in all the seven states in India public capital affects growth of SDP significantly. The authors find that the elasticity of SDP with respect to public capital is positive and significantly different from zero in all the seven selected states. Different estimated elasticity values of the public capital over the states indicate that efficiency of public capital in raising output differs from state to state in India

Slide28

Comparison of actual allocation rates with the decentralised agents’ allocation rates as well as that of social planner’s allocation rates indicates that rent seeking motive exists in all the selected states in India. This also vindicates the fact that the rent seeking motive operates among the public agents. Slide29

The estimated elasticity coefficients and the optimum allocation rates differ from state to state but the paper does not say anything about why such a difference exists over the seven selected states. For this one needs to study the compositional difference in expenditure in different states. The compositional difference that is, how much a state spends on the maintenance of existing capacity might need to be focused as this expenditure is expected to affect growth of public capital significantly. Slide30

Lessons Learnt

The distinction between stock and flow regarding public capital, along with its monetary and real dimensions should be clearly highlighted in any work, either theoretical or empirical. Otherwise, there is always a possibility of

mis

-interpretation and

mis

-specification.Slide31

Understanding relation between private and public capital as well as private party’s participation in public capital building and maintenance needs a clear understanding of rent seeking nature of the activity. In addition, some micro-foundation based on contract theory and optimizing behavior of agents is necessary to portray some of the theoretical and empirical developmentsSlide32

One needs to extend some of the excellent empirical work done for cross-country studies or specific developed countries, for analysing Indian data. These works are based strongly on an underlying theory, trying to endogenise many of the variables which determine the investment in public capital.Slide33

India and PPP

Umbrella definition of PPPs in India

PPP means an arrangement between a government or statutory entity or government owned entity on one side and a private sector entity on the other, for the provision of

public assets and/ or related services for public benefit, through investments being made by and/or management undertaken by the private sector entity for a specified time

period, where there is

a substantial risk sharing with the private sector and the private

sector receives performance linked payments that conform (or are benchmarked) to specified, pre-determined and measurable performance standards.Slide34

Arrangement with Private Sector Entity: The asset and/or service under an arrangement will be provided by the Private Sector Entity to the public.

(a)

‘Public Services’ are those services that the State is obligated to provide to its citizens

(towards meeting the socio-economic objectives) or where the State has traditionally provided the services to its citizens. For example, provision of security, law and order, electricity, water, etc. to the citizens.

(

b) ‘Public Asset’ is that asset the use of which is inextricably linked to the delivery of a Public

Service. For example, public road which is linked to public transportation.

OR, those assets

that utilize or integrate sovereign assets to deliver Public Services. For example, right of way

on highways, shore-land of about 0.5 km abutting the ocean, or use of river / water bodies, etc. Note: Ownership by Government need not necessarily imply that it is a PPP. For example, a captive jetty is not a PPP even though it uses a sovereign asset, while a common user port is a PPP as in the latter case the service is provided for use by public.Slide35

Operations or management for a specified period

: Provides an element of time period after which the arrangement with the private sector entity comes to a closure. Hence, the arrangement is not in perpetuity.

Substantial risk sharing with the private sector

: It is typically specified to differentiate PPPs from mere outsourcing contracts. For example, a facility service contract is also an outcome based reward contract but not a PPP.

Performance linked payments

: It is to provide central focus on performance and not merely provision of facility or service. A mere deferred payment contract should not get qualified as a PPP.

Conformance to performance standards

: It is to provide a strong element of service delivery aspect and the concepts of quality and compliance to pre-determined and measurable standards to be specified by the sponsoring authority.Slide36
Slide37

Source: ORF Research Foundation Occasional Paper #49, 2014Slide38

PPP initiative in the Roads Sector has been largely on the BOT basis.

The policy framework for toll-based BOT projects was approved in 1997. In 2005 it was decided that all future

programmes

/projects under NHDP would be awarded only on BOT basis. Contracts based on BOT model are inherently considered superior to the traditional Engineering Procurement and Construction (EPC) contracts as BOT projects ensure higher quality of construction and maintenance of roads and completion of projects without cost and time overrun.

Also, to attract the private sector to projects that are not commercially viable but considered essential, the government has established a Viability Gap Funding (VGF) mechanism to provide a grant of up to 40% of the project costSlide39

Mainstreaming of PPP in the Roads Sector has been significant as shown below:

•

Requisite model documents, namely, Request for Qualification, Request for Proposal, Manuals for Specifications & Standards, Model Concession Agreements (for National Highways, State Highways and Operations and Maintenance of Highways), etc. have been standardized.

• Bidding procedures are systematic.

• Availability of certain financial schemes/guidelines – Viability Gap Funding Scheme and India Infrastructure Finance Company Limited (IIFC)Slide40

Some of the key constraints from the recent attempts in developing power projects on PPP mode include:

Lack of understanding on PPP structure

Land acquisition issues

Lack of standardized documents leading to diverse interpretation of issues at the states (lack of knowledge of available standardized documents)

Tighter access to liquidity

Arbitrary norms of pre-qualification primarily adopted by state entities

Inability to stick to set deadlines by the state entities conducting project development activitiesSlide41

Thank You