La Economía de las Asociaciones Público-Privadas sebol 2013 11 08.pdf ·...

Motivation Highways Problems Promises Experience How Taking Stock

La Economía de las Asociaciones Público-Privadas

Eduardo Engel

Universidad de Chile and Yale University

5ta Conferencia Boliviana de Desarrollo EconomicoSanta Cruz, BoliviaNoviembre 15, 2013

1


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

2


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

3


Motivación

� Grandes de�cit de infraestructura pública en América Latina

� Cuello de botella para crecimiento y desarrollo

� Creciente importancia de las asociaciones público privadas (APPs)

� ¾Parte de la solución o profundizan los problemas?

� ¾Cómo hacerlo para que contribuyan a la solución?

4


Infraestructura: LAC vs. OECD

Indicador LAC OECD

Infraestructura de transportes 3.30 4.96

Calidad general 3.86 5.53Calidad de carreteras 3.58 5.19Calidad ferrocarriles 1.90 4.47Calidad puertos 3.93 5.21Calidad aeropuertos 4.44 5.58

Oferta eléctrica 4.24 6.13

Uso TIC 2.72 5.29Individuos usando internet, % 35.15 75.02Suscripciones banda ancha/100 pob 6.00 26.51Banda ancha internacional, kb/s por usuario 17.08 83.03Banda ancha móvil suscrip./100 pob 5.35 45.96Suscrip. telef. móvil/100 pob 112.49 118.16Telef. �jos/100 pob 17.02 41.46

Executive Opinion Survey, World Economic Forum, 2012. Puntajes van de 1.0 a 7.0.

5


Calidad General de Infraestructura

Pos. País Valor Pos. País Valor

1 Suiza 6.6 69 China 4.33 Finlandia 6.5 81 Uruguay 4.05 Francia 6.4 85 Ecuador 3.99 Alemania 6.2 95 Costa Rica 3.711 Portugal 6.2 96 Rep. Dominicana 3.718 España 5.8 97 Honduras 3.724 Reino Unido 5.6 106 Nicaragua 3.425 EE.UU. 5.6 107 Brasil 3.431 Chile 5.4 108 Colombia 3.436 Australia 5.2 111 Perú 3.439 Noruega 5.2 112 Argentina 3.444 Panamá 5.1 118 Bolivia 3.265 México 4.4 135 Venezuela 2.866 El Salvador 4.4 140 Paraguay 2.667 Guatemala 4.4 144 Haití 1.9

How would you assess general infrastructure (e.g., transport, telephony, and energy) in your country?[1=extremely underdeveloped; 7=extensive and e�cient] by international standards.WEF 2011-12 weighted average.

6


Inv. en APPs: países ingresos bajos y medios 1990�2009

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 20100

20

40

60

80

100

120

140

160

180

Year

PP

P in

vest

men

t com

mitm

ents

(bi

llion

US

dol

lars

)

7


Inv. en APPs: Unión Europea 1990�2009

1990 1995 2000 2005 20100

5

10

15

20

25

30

Year (financial closure)

PP

P in

vest

men

t (bi

llion

eur

os)

8


Conceptos básicos

Infraestructura pública:

� Inversión de larga duración e irreversible, utilizada para proveer unservicio público.

Tipos de infraestructura provistas vía APPs:

� Caminos, carreteras, puentes, túneles, ferrocarriles, aeropuertos,puertos, aguas, hospitales, escuelas, cárceles, estadios deportivos, ...

Tipo de APPs:

� BLT, BLTM, BOT, DBOT, DBFO, DBFO/M, ROT, JV,...

9


Motivación

Tres diseños institucionales para proveer infraestructura pública:

� provisión pública/método tradicional/servicio público

� privatización

� asociación público-privada/APP/concesión

Esta presentación:

� APPs vs. provisión pública

� carreteras

10


Motivación

Bajo los tres diseños institucionales: empresas privadas

Diferencias en:

� asignación de riesgo

� diseño y licitación de contratos

� fuentes de �nanciamiento

� contabilidad �scal

� economía política

11


Relación contractual bajo servicio público

� El sector público es dueño de los activos (gobierno nacional,estadual, municipal)

� El gobierno �nancia directamente el proyecto con deuda pública

� Presupuesto público es el demandante residual (`residual claimant')de los riesgos de

� construcción� mantenimiento� demanda

� El gobierno contrata por separado una empresa constructora y unoperador

12


Relación contractual bajo servicio público

13


Relaciones contractuales bajo APP

� La �rma tiene control (parcial) sobre los activos a lo más durante laduración de la concesión

� El gobierno y la �rma comparten (a veces de manera ambigua) losriesgos de:

� construcción� mantenimiento� demanda

� Plani�cación pública juega un rol importante

14



� Empaquetamiento (bundling) de construcción y operación en uncontrato de servicio de largo plazo entre la autoridad y una empresaprivada dedicada exclusivamente al proyecto (special purpose vehicleo SPV)

� Proyecto autocontenido, legal y económicamente:

� �ujo de caja prendado para pagar la deuda del proyecto

� Compensación del concesionario (SPV): alguna combinación detarifas y transferencias del gobierno

� Pagos de gobierno pueden ser contingentes a ciertos eventos: v.g.,garantías de ingreso mínimo

15



16


Motivación

� 25 años de experiencia con APPs:

� luces y sombras� hora de evaluar

� Argumentos a favor de APPs:

� válidos� inválidos� ¾magnitudes?

� Argumentos a favor de servicios públicos (o privatización):

� válidos� inválidos� ¾magnitudes?

� Preguntas a responder:

� ¾cuándo es una APP la mejor opción?� ¾cómo debiera hacerse una APP?

17


Una asociación de investigación:

Trabajos de Engel, Fischer y Galetovic

�Licitación de carreteras en Chile�, Estudios Públicos, 1996

�Highway Franchising: Pitfalls and Opportunities�, AER, 1997.

�Least-Present-Value-of-Revenue Auctions and Highway Franchising�, Journalof Political Economy, 2001.

�Privatizing Highways in Latin America: Fixing What Went Wrong,� Economia,2003.

�Soft Budgets and Renegotiations in PPPs,� working paper, 2009.

�Renegociación de concesiones en Chile,� Estudios Públicos, 2009.

�The Basic Public Finance of PPPs,� Journal of the European EconomicAssociation, 2013.

�Renegotiation and corruption�, work in progress.

�Is there a PPP interest rate premium,� work in progress.

The Economics of Public-Private Partnerships: A Basic Guide. CambridgeUniversity Press, forthcoming.

18


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

19


Highways

Main type of infrastructure with PPPs (by value)

Book considers other types as well

Top 10 Low-Middle Inc. Ctries. (MM USD 1990�2011)

China 47,449Brazil 32,142Mexico 25,374India 24,766Malaysia 16,552Argentina 14,094Chile 8,876Turkey 8,170South Africa 5,374Colombia 5,164

20


The case of Chilean highways: 1990

21


The case of Chilean highways: 2005

22


After 2010 Earthquake

23


Physical Characteristics of Highways

� Investments:

� large, sunk upfront, long lived asset� usually a natural monopoly (interurban) or part of a network (urban)

� Operation:

� excludable, rival (congestion an issue)

� Deterioration (and therefore maintenance):

� highly nonlinear in axle weight� proportional to usage� apparent long after it is optimal to restore the road

24


Physical Characteristics of Highways

� Quality of service is contractible:

� state of road can be veri�ed by independent parties� can measure quality of service (e.g.: time needed to remove brokencars)

� Demand:

� high (and mainly exogenous) uncertainty� example: Dulles Greenway

� Why public intervention?

� network planning� intensive use of public space and rights-of-way� monopoly requires toll regulation

25


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

26


Problems of Public Provision

Poor choice of projects

� Brazil, 1979�1984

� built 6,000 kms of new roads ... while 8,000 kms of existing roadswent from fair or good to bad quality

Enforcing projects that are built ful�ll service obligations

� insu�cient and untimely maintenance, too little, too late

� three times the cost

� lower quality of service on average

Excessive cost of projects chosen

� cost overruns

� delays

� capture and corruption

27


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

28


Promises of PPPs

Relieving strained budgets

� obviously not true if �nanced via government transfers� yet also not true when �nanced via user fees

E�ciency gains:

� advantages of bundling ... when service contractible� incentives for appropriate maintenance

Introducing competition

� Chadwick vs. Williamson

Charging appropriate user fees

� Indiana Toll Road example

Filtering white elephants

� market test ... if �nanced via user fees and no major governmentguarantees

29


Adam Smith and White Elephants

�When high roads are made and supported by the commerce that is carried on

by means of them, they can be made only where that commerce requires them.

[. . . ] A magni�cent road cannot be made merely because it happens to lead to

the country villa of the intendant of the province [. . . ]�

30


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

31


Typical contract

� Fixed term: e.g., 30 years

� Firm chosen via competitive auction

� Bidding variable: lowest toll, shortest concession term, highestannual payment to the government (cannon), lowest subsidy

� Minimum income guarantees

� Fiscal accounting: poor or totally absent

32


Governance

� Typically the same agency in charge of planning, adjudicating,monitoring and regulating the concession contracts

� Leads to poor monitoring and lax regulation

� Argentina's General Comptroller Report of 2003

33


Soft Budgets

PPPs allow o�-budget spending.

Useful for politicians/government.

In the UK, only 23% of capital cost of 599 PFI projects up to April 2009are on balance sheet.

�Cynics suspect that the government remains keen on PFI notbecause of the e�ciency it allegedly o�ers, but because it allowsministers to perform a useful accounting trick.�

The Economist, July 2nd, 2009.

34


Renegotiations and Spending Anticipation

While sometimes necessary, they are problematic

Often lead to additional works unrelated to original project

� circumventing budgetary controls

� paid by future administrations

� Santiago water collectors example

Guasch (2004), Guasch, La�ont and Straub (2007, 2008): analyze1000+ PPPs in Latin America

35


E�ciency Costs of Renegotiations

� Lack of competition for additional works may increase costsubstantially

� Adverse selection of ine�cient �rms good at lobbying

� Moral hazard problem: government becomes careless

� Bad project selection: white elephants more likely

36


Evidence from Chile

� 50 concessions (28 highways)

� 147 signi�cant renegotiations (avge.: every 2.5 years)

� Upfront investment: US$ 8.4 bn

� Renegotiations: US$ 2.8 bn

� How: Bilateral renegotiations: 83% (remainder by arbitration)

� When: 78% during construction phase � incomplete contracts?

� What: 84% involves additional investment.

� Who pays: 65% of bilateral paid by future administrations.

37


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

38


How should it be done

1. Avoiding bad faith renegotiations

2. Improving public accounting

3. Improving contract design

PPP legislation in various countries in Latin America have been reformedalong these lines.

39


How � Improving governance of renegotiations

Independent specialized agency reviews and approves projects, reducingspace for renegotiations.

Use service and not input standards in the PPP contract.

Additional works should be publicly tendered, if possible.

Independent agency ensures that contract value does not change afterrenegotiation:

� �lters �bad faith� renegotiations

� avoids adverse selection problem

� does not avoid anticipating spending

40


How � Improving budgetary accounting

Improve budgetary accounting of future capital costs, and of demandguarantees (tolled highways).

Including future liabilities in current budget not enough:

� increase in investment resulting from renegotiations must a�ectcurrent budget one-for-one

Intertemporal budget =⇒ PPP = public provision

41


How � Flexible Term Contracts

Flexible term contracts:

� reduce demand risk (beyond the control of the �rm)

� reduces need for guarantees and renegotiations

Particular case � PVR contract:

� government sets user fee and discount rate

� �rms bid on present value of toll revenues (PVR)

� contract lasts until winning bid collected:

42


How � Flexible Term Contracts

Properties of PVR:

� fair compensation is easy to calculate

� sizeable reduction in risk premium

� improves political economy of the contract

� easy to adjust tolls to demand: urban highways

� avoids winner's curse (cost-oriented bids)

First used in the UK: Queen Elizabeth II bridge at Dartford

Portugal's Litoral Centro: Euro�nance project of 2004

43


Experience with Flexible Term Contracts: Chile

44


Experience with Flexible Term Contracts: Chile

Month/year Winning bidProject auctioned (million USD)Ruta 68 (Stgo-Valparaíso-Viña) 02/1998 513Ruta 160, Coronel-Tres Pinos segment 04/2008 342Airport access road 07/2008 56Melipilla-Camino de la Fruta connection 08/2008 46Ruta 5, Vallenar-Caldera segment 11/2008 288Ruta 5, Puerto Montt-Pargua 05/2010 31Concepción-Cabrero highway 01/2011 318Alternative access road, Iquique 01/2011 167

45


How � Financing

� Project �nance commonly used

� Is there a PPP premium?

� PPPs: higher cost of �nancing than under public provision� public �nancing costs don't incorporate implicit governmentguarantee

� observed PPP premium may re�ect faulty contract design: �xedterm vs. �exible term

� incentives may be essential to realize e�ciency gains and ofteninvolve larger risk for the �rm

46


Motivation

Highways

Problems

Promises

Experience

How

Taking Stock

47


Summary � Conceptual

� Incentives and e�ciency:

� PPPs closer to privatization

� Public �nance � accounting for PPPs:

� PPPs closer to public provision

� Private �nance:

� PPPs closer to privatization

� Contract design and risk sharing:

� PPPs fundamentally di�erent from privatization and public provision:can use contract length when allocating risk

48


In favor of PPPs

Suspect:

� Saves public resources

Valid:

� Better and cheaper maintenance: bundling or political economy?

� Filter white elephants

� Easier to collect user fees and reduce distortionary taxes

� Avoid public agencies

Huge potential gains: 20 - 50% of upfront investment for three of theabove

49


In favor of public provision

Suspect:

� Lower �nancing costs

Valid:

� Expropriation risk less important

� Cannot be used to anticipate public spending

� Fewer opportunities to renegotiate

50


Conclusion

� Potentially large welfare gains under PPPs for highways

� Three out of four advantages of PPPs rely on user fees being amajor source of revenue for the concessionaire

� Contractible quality of service also important

� Case for PPPs less clear for other types of infrastructure (schools,hospitals)

� We now know the policies to handle some of the main pitfalls underPPPs

� avoid bad faith renegotiations� appropriate �scal accounting� �exible term contracts

51


Table of Contents

The Economics of Public-Private Partnerships: A Basic Guide

1. Introduction

2. Country Studies: UK, Chile, US, China

3. Highways

4. Incentives

5. Private Finance

6. Public Finance

7. Renegotiations

8. Governance

9. When and How to Implement PPPs

52


References � Engel, Fischer and Galetovic

�Licitación de carreteras en Chile�, Estudios Públicos, 1996

�Highway Franchising: Pitfalls and Opportunities�, AER, 1997.

�Least-Present-Value-of-Revenue Auctions and Highway Franchising�, JPE,2001.

�Privatizing Highways in Latin America: Fixing What Went Wrong,� Economia,2003.

�Soft Budgets and Renegotiations in PPPs,� working paper, 2009.

�Renegociación de concesiones en Chile,� Estudios Pï¿½blicos, 2009.

�The Basic Public Finance of PPPs,� JEEA, 2013.

�Renegotiation and corruption�, work in progress.

�Is there a PPP interest rate premium,� work in progress.

The Economics of Public-Private Partnerships: A Basic Guide. CambridgeUniversity Press, forthcoming.

53

La Economía de las Asociaciones Público-Privadas sebol 2013 11 08.pdf ·...

Documents

Transcript of La Economía de las Asociaciones Público-Privadas sebol 2013 11 08.pdf ·...