This version: 3 May 2002 Submission to…

                              This version: 3 May 2002


Submission to Victorian Parliament Public Accounts and
                 Estimates Committee

       Inquiry into Private Sector Investment in Public
                        Infrastructure



                                    John Quiggin

                Australian Research Council Senior Fellow

                                School of Economics
                      Faculty of Economics and Commerce
                           Australian National University




EMAIL John.Quiggin@anu.edu.au
FAX + 61 2 61255124
Phone + 61 2 61254853 (bh)
     + 61 2 62578992(ah)
http://ecocomm.anu.edu.au/quiggin
                            Summary & Recommendations


Public debt


     Private sector involvement in the provision of public infrastructure has grown

substantially over the last two decades. In some instances, this involvement has been

beneficial. In other cases, governments have entered into contractual arrangements yielding

smaller benefits and larger costs than would have arisen with traditional methods of public

procurement.

     Support for public-private partnerships in the provision of infrastructure has frequently

been motivated by a desire to reduce reported levels of public debt. This rationale for

private involvement is unsound and may lead to inappropriate allocation of investment

funds and misleading public accounts.


Recommendations on public debt


Medium-term fiscal policy should be developed in the accrual accounting

framework based on a generalised 'golden rule' of maintaining a stable ratio of

public sector net worth to GDP


Measures of debt and of financial net worth should not be a primary target of

fiscal policy. The adoption of a target of 'zero net debt' should be discouraged


 Further analysis of optimal ratios of debt to public asset ownership is required. A

level of debt equal to 50 per cent of assets is a reasonable interim target


Measures of public debt should be extended to include obligations under leases

and other long-term obligations, particularly where these arise from contracting

arrangements that replace public investments that would otherwise incur debt
                                             1
Risk allocation


      The fundamental principle stated in the Partnership Victoria documents, that risk

should be allocated to the party best able to manage it, is sound. However, the government's

preferred position on risk allocation is, in important respects, inconsistent with this principle


Recommendations on risk allocation


Where costs of operation are substantially influenced by decisions made in the

construction phase, risk should be allocated to the enterprise undertaking

construction through such mechanisms as guarantees. In other cases, risk should

be borne by the agency or enterprise providing the relevant service, which should

be separate from the construction enterprise.


Except where service specifications are stable and preferable, contracts for the

provision of services should be separate from contracts for the construction and

maintenance of physical infrastructure.


Public ownership is appropriate where the dominant risk arises from: network

risk (where the main network is publicly owned); market risk (where government

is the sole or main consumer of services); or regulatory risk


The principle of optimal risk allocation requires the availability of a range of

contracting arrangements. A single-contractor model will be appropriate only in a

minority of cases. For most infrastructure projects, standard public procurement

procedures, with subsequent public ownership of the asset will be preferable.




                                               2
Public-private comparisons


     The use of a public-sector comparator is artificial and meaningless if it is known

that, regardless of the outcome of analysis, the public sector comparator project will not

be implemented. Comparisons should not be distorted by the inclusion of budgetary

savings that represent transfers rather than efficiency gains, by unsound assumptions

about the superior efficiency of the private sector or by discount rates incorporating an

equity risk premium


Recommendations on Public-private comparisons


Evaluation of proposed partnerships should be undertake in two stages. If the

initial evaluation of the project indicates positive net benefits, it should be

evaluated against a public sector comparator. Assessment in the second stage

should be undertaken on the basis that, if the public sector comparator is found to

yield better value for money, it should be implemented.


The real pretax discount rate applicable to typical public sector comparator

projects should be 4 per cent.


While encouraging technical innovation, contracting arrangements with the

private sector should avoid financial innovation. Only well-established methods of

financing should be employed.




                                            3
    Private Sector Investment in Public Infrastructure in Victoria


1      Introduction

      Private sector involvement in the provision of public infrastructure has grown

substantially over the last two decades. In some instances, this involvement has been

beneficial. In other cases, governments have entered into contractual arrangements yielding

smaller benefits and larger costs than would have arisen with traditional methods of public

procurement.

      The object of this submission is to analyze private provision of public infrastructure

in a broad historical and policy context, to indicate both the benefits of appropriate

contractual arrangements and the danger in the pursuit of inappropriate goals, such as

spurious reductions in measured levels of public debt.

      The submission is organised as follows. Section 2 provides historical background

regarding the roles of the private and public sectors in Australia since European settlement,

with particular emphasis on the period since the 1970s. Section 3 deals with public debt.

The central point of the section is that it is not appropriate to use public-private partnerships

as a method of reducing reported levels of public debt. Rather, the contingent obligations

associated with such partnerships should be treated as having the same characteristics as

debt. Section 4 deals with the allocation of risk in contracting arrangements. The

fundamental principle stated in the Partnership Victoria documents, that risk should be

allocated to the party best able to manage it, is endorsed. However, it is argued that the

government's preferred position on risk allocation is, in important respects, inconsistent

with this principle. Section 5 deals with the concept of the public sector comparator. A

crucial observation is that this concept has failed in the UK because it was clear, in most

cases, that there was no prospect of actually implementing the public sector comparator.

                                               4
Section 6 concerns financial innovation and argues that, while technical innovation should

be encouraged, a conservative approach to financial innovation is necessary in the light

of past experience. Section 7 illustrates some points with reference to the CityLink

project.




                                            5
2        Historical background


2.1         The long view


      Almost any task undertaken by government can be, and has been, undertaken by

private enterprises. For example, the transport of convicts to Australia was undertaken

primarily by private contractors. However, the First Fleet was effectively a public venture,

being under the direct control of Governor Philip, while the Second Fleet was controlled

by the contractors, paid on a fixed rate per convict. As a result of the incentive to skimp

on food and medical attention, around a quarter of the convicts in the Second Fleet died,

and half were unfit for work when they arrived (Clark 1962) whereas the death rate for the

First Fleet was minimal.

      Similarly, both police and military services can be, and have been, privately provided.

Until the 19th century, reliance on mercenary troops was the rule rather than the exception

in European wars. However, the crushing defeats experienced by mercenaries at the

hands of French and American citizen armies led to the abandonment of this practice.

Recent experience, such as that of European mercenaries employed by the Mobutu gov-

ernment in Zaire (now Congo) reinforces the point that private provision of military

services is rarely satisfactory. More recently, private prisons have been revived, along

with many of the problems of abuse of power and lack of accountability that led to their

abandonment in the 19th century (Moyle 1994).

      Conversely, as Australian history shows, almost any good or service can be provided

by government. The first farm in Australia was the Government Farm on the site of the

present Botanic Gardens in Sydney. It was not a success. Another unsuccessful venture

was the Queensland government's establishment of publicly-owned butcher shops in the

1920s.    Public   provision   of   infrastructure   services   such   as   postal   services,

telecommunications and electricity was more successful.

                                              6
      As these examples illustrate, government enterprises perform poorly in some areas,

while private enterprises perform poorly in others. The existing distribution of activities

between the public and private sectors is, in large measure, the result of learning from

historical experience. Following the experiments of the early part of this century, govern-

ments have largely withdrawn from small-scale enterprises, such as butcher shops. On

the other hand, governments in Australia and most other countries took over the provision

of a range of infrastructure services such as rail transport, postal and telecommunications

services, where private operators had failed.

      Over the century from 1870 to 1970, the general tendency was towards expansion

in the role of government. The State took over the production and supply of goods and

services which had previously been provided, to the extent they were provided at all, by

the private sector. Australia, a country settled as an arm of the British government's

prison system, led the way in public provision of social welfare services, postal and

telecommunications services, railways and roads, universal public education and public

health services.

      Over the first half of the 20th century, other countries became more like Australia.

Whereas Australia's state-owned railways and public utilities were an exception in the

19th century, by the early postwar period it was the United States' insistence on retaining

such enterprises in private ownership that looked exceptional. After World War II the

early Australian experiments with social welfare systems were matched, and on most

measures surpassed, by European welfare states.

      Attention to this historical experience is important because it provides a great deal

of evidence regarding the relative performance of the private and public sectors in a

range of activities. The existing allocation of activity between the private and public

sectors is not purely haphazard, as much discussion has tended to imply. Rather, public

involvement in some areas of the economy, such as agriculture and retail trade has been

consistently unsuccessful. Conversely, the public sector's dominance in infrastructure

                                                7
activities and in the provision of human services such as health and education reflects the

inadequacy of earlier systems of private provision.


2.2        The postwar settlement


      The growth of the State from World War II to the 1970s was largely quantitative

rather than qualitative, since the boundary between the private and public sectors was

fairly stable. Public expenditure grew steadily as a proportion of gross domestic product

(GDP) partly because of the increasing importance of sectors such as health and education,

where public funding and provision played a large role, and partly because of demographic

changes, particularly increased life expectancy, which led to increased expenditure on age

pensions. By the 1970s, public expenditure, and the taxation needed to finance it, had

reached around 40 per cent of GDP in Australia and more than 50 per cent in many OECD

countries. Since much of this revenue was returned to households in the form of transfer

payments, the public share of output and employment was lower, but was still around 25

per cent in typical OECD economies by 1970.

      The growth of the State after 1945 was commonly discussed in terms of the `mixed

economy', consciously proposed as a `third way' between the unfettered capitalism of the

19th century and the comprehensive State socialism of the Communist bloc. The mixed

economy involved large-scale government involvement in an economy that was nevertheless

predominantly private. The achievements of the mixed economy were substantial. For

more than a quarter of a century, unemployment disappeared from the developed world.

Economic growth proceeded at rates never equalled before or since. The development of

extensive social welfare systems based on progressive taxation led to a reduction in

inequality in incomes and an even greater reduction in inequality in living standards.

      The longest sustained period of strong growth and full employment in the history of

the world economy coincided fairly closely with the period of maximum expansion of

government. Hence, simplistic claims that the private sector is, in general, more efficient
                                             8
and effective than the public sector are inconsistent with the lessons of history. The

central policy problem is to find the appropriate balance between the public and private

sectors.


2.3         The crisis of the 1970s


      The breakdown of Keynesian macroeconomic policies in the early 1970s reversed

the trend towards growth of government. The apparent success of free-market economists

such as Milton Friedman in predicting and explaining the failure of Keynesianism enhanced

the prestige of free-market views, though the gloss was taken off this achievement when

the monetarist policies proposed by Friedman proved no more successful than the Keynesian

policies they replaced. Higher unemployment implied increased expenditure on social

welfare benefits and therefore generated pressure to cut back other areas of public spending.

Finally, and perhaps most importantly, the loss of confidence in the capacity of governments

to control over the economy implied a greater need to cultivate `business confidence'.

      The incapacity of government revenue to meet demands for public expenditure has

been referred to as the `fiscal crisis of the state'. A variety of fiscal expedients, including

asset sales, reduction of capital expenditure and deficit financing have been advocated

and employed in response to the fiscal crisis of the state. Ultimately, however, there is no

easy solution - `what you pay for is what you get'.




2.4         Privatisation


      Privatisation was one of the first financial expedients adopted by governments in an

attempt to resolve the `fiscal crisis of the state'. It initially appeared that privatisation

offered an immediate source of cash that could appropriately be allocated to current

expenditure. Early privatisations such as those of Qantas and the Commonwealth Bank. It

                                              9
soon became clear, however, that privatisation involved the loss of a stream of income

(dividend payments and reinvested earnings) into the future, and was therefore similar, in

its fiscal effects, to taking on additional debt.

      Surveys of Australian experience with privatisation suggest that, in most cases the

interest savings from privatisation (assuming all proceeds are used to repay debt) have

been less than the earnings foregone as a result of privatisation. Hence, privatisation

      As will be discussed below, similar issues arise in relation to the projects for

private investment in infrastructure. Private investors must receive a return in the form of

payments from governments or infrastructure users. In either case, governments are

foregoing income that would accrue to them if the project were publicly owned. It is

impossible to obtain infrastructure services at no cost to the public. The only issue is to

determine the most cost-effective method of provision.


2.5         BOOT schemes


      A particularly popular way of packaging infrastructure projects in Australia has

been the Build, Own, Operate and Transfer (BOOT) system. Under this system, a private

enterprise constructs the project in return for access to a stream of user charges, such as

the revenue from a toll. After a period sufficient to cover the cost of construction, the user

charges are abolished and the asset is handed over to the public sector. From the viewpoint

of the cash system of public accounting, the government pays nothing during the period of

private ownership, and receives a free asset at the end. The biggest single Australian

example has been the CityLink road project in Melbourne, discussed below.

      Despite, or perhaps because of, their superficial appeal, BOOT schemes have not

been viewed favorably by Australian economists. The EPAC Infrastructure Taskforce,

the NSW Auditor-General, and the Industry Commission have all reported negatively on

such schemes

      First, the apparent reduction in public debt associated with projects of this kind is
                                               10
illusory. To provide a return to the operator, the public must either commit to a stream of

payments from general revenue or alienate a revenue source such as a toll. As the EPAC

Taskforce pointed out, the fiscal and macroeconomic impacts are essentially the same as

if the construction of a publicly-owned asset was financed by the issue of bonds.

      Second, when used to finance road construction project, BOOT schemes typically

involve a misallocation of risk, since the risk in revenue flows is usually related to the

planning of the transport network as a whole, rather than to the construction of a particular

project. Hence, in most cases, it is preferable for the construction and maintenance of the

project to be undertaken by competitive tendering with ownership passing to the network

owner (normally the relevant state government) on completion of the construction phase.

The gap of $2 billion between the construction cost of the CityLink project and the tolls

paid to the private consortium is, in part, compensation for the real costs of risk misallocation.

      Third, the set of road user charges associated with BOOT schemes is ad hoc and

arbitrary, being dictated by historical accident rather than economic considerations. On

average, the pricing system is perverse, raising the cost of using new, uncongested, roads,

then eliminating charges later, when roads are likely to be congested.

      Finally, in cases where private ownership is optimal, the commitment to transfer

the asset to public ownership must reduce welfare. There may, perhaps, be assets which

are optimally owned by the private sector at one point in their lives and by the public

sector in another, but the likelihood that such a crossover point will coincide with the

date at which the project is `paid off' is minuscule.


2.6         Competitive tendering and contracting


      The practice of contracting with private firms for the provision of public services is

a very old one. For example, the transport of convicts to Australia was undertaken primarily

by private contractors. However, the First Fleet was effectively a public venture, being

under the direct control of Governor Philip, while the Second Fleet was controlled by the
                                               11
contractors, paid on a fixed rate per convict. As a result of the incentive to skimp on food

and medical attention, around a quarter of the convicts in the Second Fleet died, and half

were unfit for work when they arrived (Clark 1962), whereas the death rate for the First

Fleet had been minimal. Subsequent tightening of contractual terms reduced death rates,

but also increased costs.

      In broad terms, the history of convict transportation has been repeated in more

recent experiments with competitive tendering and contracting. In the initial rounds of

contracting, private firms have offered to deliver public services at a price far below the

cost of public provision. As a range of hidden costs and problems have emerged, contractual

terms have been tightened. The results have included improvements in performance, but

also the loss of many of the financial savings that originally motivated the move to

contracting.

      The recent upsurge in private provision of public services began in the early 1980s

under the Thatcher government in the United Kingdom. The Thatcher government imposed

compulsory programs of competitive tendering and contracting on central government

agencies and local governments. A similar approach was adopted by the Kennett government

in Victoria and by the Howard government. Other governments have undertaken extensive

contracting out without adopting a comprehensive program of this kind.

      The increase in support for the policy of contracting out for the provision of public

services is closely related to the increasing popularity of the corresponding practice of

`outsourcing' in the private sector. In both the public and private sectors, policies of

contracting out or outsourcing have been adopted for a number of reasons.

      First, there has been a general shift towards the belief that organisations should

focus on the achievement of a single `core objective' or a small number of such objectives,

and should, as far as possible, avoid responsibility for peripheral activities. This belief

contrasts with the ideas of the 1960s and 1970s when `conglomerate' corporations, with

subsidiaries engaged in many different industries, were seen as a way of achieving

                                            12
diversification, and when government agencies typically sought to pursue very broad

definitions of `the public interest'.

      Second, improvements in understanding of the allocation of risk have led to a

desire to organise contractual relationships in a way that yields better management of

risk. Where specific operational risks can be distinguished from the general operations of

an organisation, contracting may provide an appropriate way of managing those risks.

      Last, but not least, there has been a desire to reduce the core workforce of public

and private sector organisations. In part, this reflects a change in fashions, as `downsizing'

rather than `empire-building' has come to be seen as the mark of a good manager. More

importantly, many organisations have found it difficult, because of legal restrictions and

concerns about morale, to reduce wages and conditions for core employees. Contracting

out or outsourcing has enabled corporations to replaced core employees with contract

employees who receive less favourable wages and conditions and to increase competitive

pressure on the remaining core employees.

      The primary motive for contracting out the provision of public services to the

private sector has been the desire to reduce public expenditure. In Australia, the most

widely-used estimate of the cost savings associated with contracting out has been that, on

average, the cost of providing public services will be reduced by 20 per cent as a result of

contracting out. This estimate is derived mainly from the work of Domberger and his

co-workers, and has been employed by the Industry Commission (1996) and other

government agencies.

      Other studies have suggested that, when the costs of tendering and contract

management are taken into account, and if there are no changes in wages and conditions

as a result of contracting out, the average cost saving from contracting out will be less

than 20 per cent in most cases (Paddon 1991, 1993). Paddon criticises the work of

Domberger and cites British estimates that the average cost saving was around 7 per cent.

      More importantly, the finding of superior private sector efficiency does not appear

                                             13
to extend, in general, to capital-intensive infrastructure activities. In the case of water, the

opposite finding, that the public sector is more efficient, has been more common

(Bhattacharya et al, Teeples and Glyer).


2.7          The PFI in the UK


      Having used the sale of public assets, typically at large discounts, to finance illusory

surpluses during the 1980s, the Thatcher government turned to the Private Finance Initiative

as a means of pursuing its principal fiscal objective, reduction of the Public Sector Borrowing

Requirement. The initiative had only modest success, largely because government

departments and agencies were unwilling to bear the higher costs associated with private

financing.

      On its election in 1997, the Blair government rejected the earlier version of the PFI,

but sought to persist a new and improved version, in which `value for money' would be a

crucial criterion. Critical reports from the Auditor-General and from Parliamentary

committees have found that this goal has not yet been achieved, in that, to date, PFI

initiatives have not demonstrably achieved improved value for money.

      One of the most significant defects in the operation of the modified PFI has related

to the use of `public sector comparators' as a basis for assessing whether projects offer

`value for money'. The idea is that if the analysis of the public sector comparator shows

that a given project could be undertaken at lower cost by the public sector, then the

project will not be considered for the PFI. In practice, however, it is frequently made

clear that, if the PFI option is not approved, the project will not proceed.

      In these circumstances, the analysis of the `public sector comparator' is not a

genuine comparison of options. Rather, it is an administrative hurdle that must be cleared

before a PFI project can proceed. In the absence of any prospect of public funding,

proponents of the project have a strong incentive to ensure that the comparison is favourable

to the private option. It is not surprising that the Auditor-General has concluded that, in
                                              14
many cases, it is impossible to determine whether PFI projects actually represent value

for money.

      Further doubt has been cast on the PFI by the poor performance of privatised

infrastructure providers. The operator of the rail infrastructure network rail track was

forced into administration in late 2001, following years of poor performance. The Labor

government blamed this failure on a botched privatisation undertaken in haste by the

previous Conservative administration. Early in 2002, however, the partially privatised air

traffic control system ran into similar difficulties and now faces the need for a government

bailout. This privatisation was undertaken by Labor, apparently against the advice of the

relevant safety authorities.

      In summary, advocacy of initiatives such as PFI represents, at this point, a triumph

of hope over experience. Previous experiments in private ownership of public infrastructure

have resulted in high costs and a misallocation of risk. The hope is that with improved

contracting procedures, these problems will be overcome and genuine cost savings will

be realised.




                                            15
3      Public debt

      The issue of debt has played a prominent role in Australian political debate for

many years. As in the accounts of a household or a private business, debt levels per se

should not be a central target of public fiscal policy. The appropriate target is the level of

net worth and, by implication, the sustainability of public expenditure with given revenues.

Net worth is the difference between the value of assets and the level of debt.


3.1         Budgetary cosmetics, Loan Council, PSBR


      As has already been noted, the primary motivation of the Thatcher government's

PFI was the desire to reduce the Public Sector Borrowing Requirement and, ultimately,

public debt. Similar concerns have been prominent in Australian jurisdictions, and

particularly in Victoria.

      Beginning in the 1980s, a number of Australian governments entered into

arrangements involving the sale and leaseback of public assets. Early deals of this kind in

the 1980s, notably some involving power stations, were often sham transactions designed

to evade controls on borrowing imposed by the Loan Council. From the 1990s onwards,

sale and leaseback has become standard operating procedure for some governments.

      The most egregious deals have been those undertaken by the Commonwealth

Department of Finance, which is willing to sell assets, then lease them back at rates of up

to 15 per cent, implying that the entire purchase price would be paid back in rent within

seven years. As has been noted by the Australian National Audit Office

      In effect, this approach ensures that the Commonwealth will hold no property. As

the Australian National Audit Office observes `By applying the hurdle rate of return of

15 per cent in the Commonwealth Property Principles to the selection of properties for

sale, it would be unusual for the Commonwealth to own property".

      The idea that problems with public debt can be resolved by encouraging the private
                                           16
sector to undertake infrastructure investment is superficially appealing. However, a more

rigorous economic analysis reveals two fundamental problems with this idea.

      The first is, that, in many cases, private infrastructure initiatives have been associated

with a series of guaranteed government payments which have exactly the same economic

and fiscal effects as the repayment of interest on a debt. In the case of the Sydney

Harbour Tunnel for example, the NSW Auditor-General concluded that the effect of the

contract was that the Tunnel was actually owned by the State government rather than the

nominal private owners, and that the obligatory payments to the owners were, in effect,

interest on a debt. Similar points arise where a `sale' is associated with a long-term

leaseback or `take-or-pay' arrangement.

      Related issues arise in relation to macroeconomic concerns. Efforts to restrict the

growth of public debt are sometimes motivated by concerns that rising interest rates will

`crowd out' private investment. The magnitude of this effect depends on the extent to

which Australian interest rates move independently of world rates. However, the effects

on interest rates of borrowing to finance a large infrastructure project are the same

whether the project is nominally owned by the government, a private provider, or some

combination of the two. In any of these cases, private investors outside the infrastructure

sector will feel the same effect.


3.2         Public debt and net worth


      As with a household or business, the crucial issue in assessing a government balance

sheet is not the level of debt but the government's net worth. Net worth is the difference

between the value of assets and the value of liabilities. The sale of assets invariably

reduces debt. However, asset sales are desirable only if the price received for the asset

exceeds its value in continued public ownership, that is, only if net worth is increased.

      Government balance sheets frequently distinguish between financial and non-

financial assets (the great majority of liabilities are financial). It is more useful, however,
                                              17
to distinguish between revenue-generating and service-generating assets. An asset that

generates revenue contributes directly to the capacity to service the associated debt. If

revenue flows exceed interest payments, such assets make a positive net contribution to

the government's current operating balance.

      By contrast, service-generating assets yield benefits to the public, but provide no

direct financial return to government. This does not imply that such assets lack value.

However, it is necessary to allocate either general taxation revenue or a hypothecated

revenue flow to meet the interest payments associated with financing the provision of

such assets.

      These points may be illustrated in relation to roads. All publicly-owned roads are

part of the government's stock of assets and therefore contribute to the net worth of the

public sector. The decision to construct a new publicly-owned road means that the

government incurs additional debt equal to the construction cost of the road. This debt

may be serviced by tolls, by specific increases in road user charges such as registration

fees and petrol taxes, or from general revenue.

      In the first case, the road is an income generating asset, and makes a positive net

contribution to the operating balance if toll revenue exceeds interest payments. In the

third case, it is a service-generating asset, and the cost of the flow of uncharged services

makes a negative contribution to the operating balance.

      In the current scheme, a toll road owned by a government department would be

regarded as a non-financial asset. If however, the road were operated by a government

business enterprise, the government's shares in the enterprise would be regarded as a

financial asset. This illustrates the arbitrary nature of the distinction between financial

and nonfinancial assets.

      The possibility of financing additional investment through road user charges suggests

that, like most classification schemes, there are intermediate cases in the distinction

between income-generating and service-generating assets. Although there is no charge

                                              18
for the services of individual roads, the road network as a whole may be regarded as an

income-generating asset for governments.

      It is important to note that service-generating assets are not, in any meaningful

sense, `inferior' to income-generating assets. In particular, as will be argued below, tolls

are, in most cases, an inefficient method of financing road projects. Nevertheless, the

fiscal implications of ownership of service-generating assets are different from those of

ownership of income-generating assets and this distinction should be reflected in public

accounts.


3.3          Leases, contingent liabilities and debt


      As has been noted above, the use of long-term leases has become particularly

popular as a means of reducing reported levels of public debt. Although the government's

obligations to make payments under such leases are effectively equivalent to the requirement

to make interest and principal repayments on public debt, the accounting treatment is

quite different.

      The use, or misuse, of leases to reduce reported debt levels has not been confined to

the public sector. A particularly notable example has arisen in the telecommunications

sector, where companies with excess capacity engaged in `swaps'. Two such companies

would lease each other's assets. Although the transaction had no economic effect, each

company was able to recognise the capitalised value of future rental receipts as current

income, while treating its own obligations as a contingent liability that did not need to be

declared as debt.

      There is no simple answer to the question of when a lease obligation should be

regarded as being equivalent to a debt. The crucial issues are the length of the lease and

the specificity of an asset. Obviously, the longer the lease on an asset the more the lease

is like a debt.

      The issue of asset specificity is more complex. A long-term lease on, say, an office
                                             19
block does not really bind governments to purchase the associated services, since there is

a well-established market for such services. If governments find that office space is no

longer needed, they can sublet it, or negotiate with the owner to terminate the lease on

commercial terms.

        By contrast, where governments lease a special-purpose asset such as a hospital,

there is no real option of subletting or termination. Once the contract is entered, the

government is effectively committed in the same way as if it had purchased the asset

using debt finance.


3.4           Accrual accounting


        Some of the issues discussed above have been addressed by the (still incomplete)

shift from cash to accrual accounting in government budgeting. Traditionally, the main

object of the Budget (and still an important one today) was to ensure that ministers were

accountable for public money, rather than to present an accurate picture of the government's

financial position. Hence, accounts were presented in cash flow terms without any distinction

between current and capital outlays.

        Cash accounting reports the flows of money payments and receipts. If money is

received during the accounting period it is counted as revenue, even though it might be

paid for services provided the previous or the following year. Similarly, payments are

included if they are made this year, no matter when the goods or services purchased are

used.

        As a result, the proceeds of asset sales were treated exactly like current revenue (or,

in some cases, as a reduction in expenditure) and, as far as the Budget was concerned,

available for spending in the year in which they are realised. It was gradually recognised

that a policy of selling assets to finance current expenditure was unsustainable. A simple

ad hoc response was to replace the cash measure of Budget balance with an `underlying'

measure, which excluded asset sales.
                                               20
      A more systematic response to the defects of the cash Budget balance measure has

been to adopt a system of accrual accounting. All Australian governments are moving

progressively to accrual accounting .

      The basic idea of accrual accounting is to separate current and capital expenditure

and to recognise revenue and expenditure as they accrue, rather than when they are

realised as cash payments. Under accrual accounting, the purchase price of capital assets

is amortised over the life of the asset, rather than being lumped in with current expenditure

in the year of purchase. Accrual accounting is not perfect, but it prevents the use of some

of the devices by which governments have fudged their accounts in the past.

      The crucial summary measures in an accrual system of accounting are the operating

result and the government's net worth. This is the difference between income derived

from taxation, grants and government business enterprises and current expenditure on

goods, services and interest payments, including an allowance for the depreciation of

physical assets. A surplus on the operating result implies that income has exceeded

current expenditure, so that net worth is increasing. Similarly, a deficit corresponds to a

reduction in net worth.


3.5         Credit ratings


      A common argument used to justify an strategy specifically focused on reducing

debt is that such a strategy will permit an improvement in credit ratings. Before considering

the effects of debt reductions on credit ratings, it is important to observe that the direct

benefits of a higher credit rating are quite small. For example, in 1993-94, Victoria, with a

Standard and Poors credit rating of AA, faced borrowing costs for 10-year bonds 10 to 40

basis points (0.1 to 0.4 percentage points) higher than those of New South Wales with a

credit rating two levels higher at AAA. (Moody's ratings were Aa3 for Victoria and Aaa

for NSW). On a debt of $10 billion, the associated interest difference is between $10

million and $40 million per year, a trivial amount in comparison with total revenue or
                                             21
expenditure

      In view of the small direct benefits of a credit upgrading, the emphasis placed by

many commentators on credit ratings must be attributed primarily to the view that credit

ratings represent an impartial judgement of the soundness or otherwise of government

fiscal strategies. In general, it is true that policies that tend to have a favourable (or

unfavourable) impact on the fiscal sustainability of government policy will also have a

favourable (or unfavourable) impact on credit ratings. For example, the introduction of

unfunded expenditure programs, or cuts in taxes that are not matched by expenditure

savings will tend to reduce credit ratings.

      However, this argument does not apply in all cases. Credit ratings are designed

specifically to inform and protect the holders of government debt. Policies that specifically

improve the position of holders of government debt will be viewed favourably by credit

rating agencies even if they are harmful to the state as a whole. In particular, reductions

in the level of debt will tend to improve credit ratings even if they are financed by

inefficient taxes and charges or by the sale of income-earning assets at inadequate prices.

The imposition of inefficient taxes and charges will tend to discourage investment and

employment while the sale of income-earning assets at inadequate prices will reduce the

net worth of the public sector and, ultimately, the capacity to provide public services,

even though both measures may improve credit ratings.

      It is paradoxical that many participants in the public policy debate have stressed

both the importance of credit ratings and the desirability of emulating the private sector.

Over the past three decades, private corporations have generally sought to reduce the cost

of capital by increasing debt levels and accepting correspondingly lower credit ratings.

Very few private nonfinancial corporations now aspire to a AAA credit rating. In particular,

private owners of infrastructure assets typically maintain Standard & Poors credit ratings

around BBB.

      Of course, there are important differences between the public and private sectors.

                                              22
Nevertheless, the decisions of private corporations reinforce the point that the pursuit of

high credit ratings is not, in itself, a sound basis for public policy.


3.6         The optimal level of public debt and gearing


      Fiscal policy should be focused on net worth rather than debt. This is not to say,

however, that the level of debt does not matter. It is generally agreed that excessive levels

of debt in relation to assets should be avoided, and also that, in most cases, some positive

level of debt finance is both financially prudent and more sensible than a `zero debt' or

`zero net debt' policy, whether for households, businesses or governments. Zero debt

policies typically imply that services, such as the transport services provided by improved

roads, must either be foregone or purchased at greater cost than would be associated with

debt-financed investments that could provide those services.

      While there is wide agreement that a `zero debt' policy makes little sense, there has

been little or no analysis of optimal levels of public debt. Developments in the private

sector provide some useful information. However, there are fundamental differences

between a corporation, financed by a mixture of debt and private equity and a government

asset, financed by a mixture of debt and accumulated taxation revenue. In particular, as

will be discussed below, the resulting allocation of risk is quite different.

      With these qualifications in mind, it may be observed that privately-owned

infrastructure assets are typically financed using between 50 and 60 per cent equity and

between 40 and 50 per cent debt. As a first approximation, this would seem to be a

reasonable rule for the public sector to follow.


3.7         Sustainability and the golden rule


      The most important issue in relation to public sector net worth is that of sustainability.

As a general principle, public policy with respect to income, expenditure, net worth and


                                               23
debt should be designed in such a way that existing policy settings can be maintained over

time without giving rise to an explosive growth in debt.

      This may be seen as an extension of the `golden rule' principle of maintaining

balance between income and expenditure over the course of the economic cycle. The

core of the 'golden rule' framework is that, as a general rule, policy should be designed to

maintain a stable allocation of public sector resources over the course of the business

cycle. Stability is defined in terms of the following ratios:

      (i) The ratio of public sector net worth to state product

      (ii) The ratio of current expenditure to state product

      (iii) The ratio of current income to state product

      If GDP is growing, and net worth is positive this rule implies that, on average there

should be net surplus of income over expenditure. However, this does not necessarily

imply that partial measures of net income, such as the cash or accrual balances for the

non-commercial sector should be in balance or surplus.

      The general principles underlying the golden rule may be modified to take account

of particular circumstances. For example, in a federal system, transfers of responsibility

between federal and state governments may imply changes in the optimal ratios of

federal expenditure to GDP and state expenditure to state product.

      More importantly, while a stable ratio of current expenditure to GDP is a reasonable

target in the medium term (a single business cycle of 8 to 10 years), long-term increases

in income are accompanied by increasing demands for services such as health and education,

which are predominantly supplied and funded by government. Thus, it is reasonable to

expect the ratio of current expenditure to GDP to increase gradually over time.


3.8         Summary and recommendations


      The analysis above shows that the desire to deliver services without affecting the

reported level of public debt has been the source of serious policy mistakes in the past.
                                              24
The following recommendations may help to prevent such outcomes in the future.


Recommendation: Medium-term fiscal policy should be developed in the accrual

accounting framework based on a generalised 'golden rule' of maintaining a stable

ratio of public sector net worth to GDP


Recommendation: Measures of debt and of financial net worth should not be a

primary target of fiscal policy. The adoption of a target of 'zero net debt' should be

discouraged


Recommendation: Further analysis of optimal ratios of debt to public asset

ownership is required. A level of debt equal to 50 per cent of assets is a reasonable

interim target


Recommendation: Measures of public debt should be extended to include

obligations under leases and other long-term obligations, particularly where these

arise from contracting arrangements that replace public investments that would

otherwise incur debt




                                          25
4      Risk and ownership


4.1         Fundamental principles


      The policy document, Partnership Victoria, sets out both principles governing risk

transfer, and a list of risk classes:

      The principle governing risk transfer is that risk will be allocated to whoever is

best able to manage it at least cost, taking into account public interest considerations.

This does not mean that all risk is transferred. If risk is transferred inappropriately, the

Government will pay a premium. The ability to secure risk transfer on worthwhile terms

requires the scope of the project to be drawn sufficiently widely. Because there will

always be a wide variety of risks associated with potential projects, the structure of a

partnership project needs to take account of which party is best able to take responsibility

for managing such risks as:

      Design and construct risk ­ to cost, quality and time;

       Commissioning and operating risk;

       Service under-performance risk;

       Industrial relations risk;

       Maintenance risk;

       Technology obsolescence risk;

       Regulation and legal change risk;

       Planning risk;

       Price risk;

       Taxation risk;

       Residual value risk; and (where appropriate)

       Demand (or volume/usage) risk.



                                            26
      Some subsidiary principles are also presented:

      Decisions on risk transfer will also recognise two general principles: Whoever is

allocated risk must have the freedom to choose how to handle and minimise it; and

Materiality must be considered. Where a department or agency is not the only user of an

asset, demand (or volume/usage) risk may also be transferred. The value of risks transferred

will be estimated and included in the Public Sector Comparator, to allow for a like-with-like

value for money assessment.

      The general principle of allocating risk to the party best able to bear it is sound.

However, the detailed treatment of risk is less satisfactory. The presentation of such a

long list of risks raises the danger of `not seeing the woods for the trees'.

      Moreover, the analysis in the Partnership Victoria documents is inadequate in crucial

respects. The treatment of market risk is not, in general, consistent with the core principle

of allocating risk to whoever is best able to manage it at least cost. The treatment of

network risk is adequate. Finally, one of the most important aspects of risk, the risk

associated with fluctuations in the aggregate economy is not explicitly recognised, and its

implicit treatment in terms of discount rates is incorrect.

      In this section, a summary of the main categories of risk is presented, with an

assessment of the optimal allocation of risk.


4.2         Construction


      Proposals to undertake a transport infrastructure project typically include an estimate

of the costs of construction. However, this estimate may be turn out to be an underestimate

because of increases in wages or the costs of other inputs, or because of unforeseen

technical difficulties, such as equipment breakdowns and adverse weather. In an economic

sense, failure to complete the project on time reduces the present value of the services

provided by the project and therefore increases the effective cost of the construction

phase. Less frequently, things may turn out better than expected, with the project being
                                              27
completed `on time and under budget'.

      In the past, it was common for public infrastructure projects to be constructed by

government departments using public sector employees. In general, this has proved less

satisfactory than the alternative of competitive tendering. In most cases, it is relatively

easy to ensure that the private constructor bears most of the risk associated with the

infrastructure projects , and therefore has incentives to overcome the agency problem. By

contrast, the incentives for individuals within a government department to minimise costs

are relatively weak and diffuse.

      In most cases, the optimal allocation of risk requires construction risk (including

site risk and design risk) to be borne by the enterprise undertaking construction. This is

consistent with the `government preferred position' presented in the Partnership Victoria

documents.


4.3          Operation


      Operational risk encompasses risks relating to industrial relations and maintenance

as well as commissioning and operating risk. After completion of the construction phase,

an infrastructure asset must be maintained. In addition, the operator may provide a range

of operational services using the asset. For some assets, such as roads, costs of operation

and maintenance are relatively stable and predictable and are small relative to initial costs

of construction. For other assets, such as airports, operations may be complex and subject

to substantial risk.

      Another important issue regarding risk and operational costs is the relationship

between the construction and operation phases. In some cases, decisions made in the

construction phase, for example regarding the quality of materials, may have a substantial

impact on subsequent costs of operation and maintenance. In such cases, contractual

arrangements in which the constructor is required to undertake maintenance may be

optimal.
                                            28
      In other cases, there is no such link, and the appropriate contractual relationship

involves a `turnkey' contract with payment on completion of the construction phase.

Recent public infrastructure projects have involved the creation of a consortium providing

a combination of construction, operation and financing, with which the government

contracts. In the absence of inherent links between these activities, such an approach

reduces the transparency of arrangements and increases the risk of adverse outcomes for

the public sector.

      In summary, no simple principle can be stated with respect to the optimal allocation

of operational risk. Broadly speaking, however, the following result can be stated:

      Recommendation: Where costs of operation are substantially influenced by

decisions made in the construction phase, risk should be allocated to the enterprise

undertaking construction through such mechanisms as guarantees. In other cases,

risk should be borne by the agency or enterprise providing the relevant service,

which should be separate from the construction enterprise.

      This recommendation differs from the `government preferred' approach presented

in the Partnership Victoria documents. The preferred approach involves government

contracting with a single party or consortium for both construction and operation. This

approach will be optimal only in cases where there is a close link between special design

features and subsequent operation.


4.4         Service specifications


      The principle that risk should be allocated to the party best able to bear it applies to

changes in service specifications. Where the services required from an infrastructure

project are subject to frequent and unpredictable change, the risk must be borne by the

service user, in this case, the government. The more the required risk, the stronger the

case for ownership of the relevant activity.

      As the costs of changes in service specifications have been recognised, construction
                                               29
contracts have increasingly relied on a clear preliminary specification of required standards

with little scope for changes in specification prior to completion of the project.

      In many cases, however, it is impossible to avoid changes in service specifications.

This is clearly true in relation to core public services such as health and education, and in

the medium term it also applies to less complex activities, such as ancillary services for

hospitals. Specifications for such services are inevitably subject to change in the medium

term. Among other things, this principle implies that governments, and for that matter

private corporations, should not enter into long-term contracts for the provision of even

moderately complex services,

      Since the optimal term for most service contracts of this kind is shorter than the life

of associated capital infrastructure such as schools and hospitals, this analysis reinforces

the point that the `government preferred' approach of contracting with a single party is

unlikely to be appropriate in such cases.

      Recommendation: Except where service specifications are stable and preferable,

contracts for the provision of services should be separate from contracts for the

construction and maintenance of physical infrastructure.


4.5         Demand or market risk


      Demand risk refers to the possibility of unforeseen variation in the demand for the

services generated by a project. Where there are many consumers, demand risk is

appropriately borne by the service provider. However, where there is a single major

consumer, that consumer should bear the risk associated with changes in their demand.

This situation applies to many public infrastructure projects. The analysis in the Partnership

Victoria documents, however, states a `government preferred' position that the private

partner should bear the risk in projects of this kind. This is inconsistent with the basic

principle of risk allocation.


                                             30
4.6        Regulatory risk


      All businesses are subject to regulation, and must bear the risk of possible regulatory

change. However, where regulatory risk is the dominant source of risk, the principles of

optimal risk allocation require that the government undertaking regulation should bear the

risk, either by insuring the returns to the asset owner or through public ownership. The

more significant and complex the regulatory risk, the stronger the case for public ownership.


4.7        Network risk


      The term `network risk' describes a class of risks applying to an individual asset that

is one part of a larger network, for example, an individual road in an urban road network.

Usage of a particular road will depend, to a large extent, on decisions made with respect

to other elements of the transport network. Hence, in many cases, it is inappropriate to

consider the risks associated with an individual asset in isolation from the larger network.

      In some cases, typically described as `interface risk', interaction with the larger

network is of relatively modest importance in relation to the services of the asset in

question. In such cases, a division of risk between the owners of the asset and the owners

of the network is appropriate.

      In other cases, however, the value of the asset is primarily determined by its

interaction with the network as a whole. Where network risk takes this form, the optimal

allocation of risk can only be achieved if the owner of the network also owns the asset. In

particular, this conclusion applies to most urban roads.


Recommendation: Public ownership is appropriate where the dominant risk arises

from either:


      Network risk (where the main network is publicly owned);

      Market risk (where government is the sole or main consumer of services); or

                                            31
      Regulatory risk




4.8         Systematic and idiosyncratic demand risk


      A crucial aspect of demand risk is the distinction between risk that is correlated with

movements in the general economy (often referred to as systematic risk) and risk that is

specific to a particular project (often referred to as idiosyncratic risk). Under plausible

conditions, idiosyncratic risk can be pooled and diversified in such a way that no individual

bears any significant risk. By contrast, because systematic risks are highly correlated,

pooling and diversification has little effect other than to redistribute a given risk within

the population.

      The Partnership Victoria documents do not explicitly address systematic risk. Rather,

systematic risk is reflected in the `cost of capital' or `discount rate' applied to projects,

which is typically substantially higher than the real rate of interest applicable to public

debt. This issue is discussed in Section 5


4.9         Summary recommendation


      The Partnership Victoria documents envisage a preferred position in which

governments contract with a single party which undertakes design, construction, financing

and operation of an infrastructure facility. Such a position will yield an optimal allocation

of risk only in special circumstances including the following conditions

      · Close integration between construction and operation phases

      · Simple and unchanging service specifications

      · Stable demand for services or multiple users in addition to government

      · Stand-alone projects or limited interface risk

      · Limited regulatory risk


                                             32
      If any of these conditions are not met, the single-contractor model is unlikely to be

appropriate. In general, the optimal allocations are likely to be a mixture of

      · Standard public procurement in the construction phase, with full public ownership

thereafter


      · Standard public procurement in the construction phase, with the constructor

retaining responsibility for maintenance

      · Standard public procurement in the construction phase, with separate contracting

for provision of associated services

      · A single-contractor model


Recommendation: The principle of optimal risk allocation requires the availability

of a range of contracting arrangements. A single-contractor model will be

appropriate only in a minority of cases. For most infrastructure projects, standard

public procurement procedures, with subsequent public ownership of the asset will

be preferable.




                                             33
34
5      Public-private comparisons


5.1         The need for real comparisons


      As was noted above, one of the most significant defects in the operation of the

British PFI has related to the use of `public sector comparators'. The problem is that, in

many cases, if the PFI option is not approved, the project will not proceed. In these

circumstances, the analysis of the `public sector comparator' is not a genuine comparison

of options. Rather, it is an administrative hurdle that must be cleared before a PFI project

can proceed.

      To overcome this problem, it is necessary that the analysis of the public sector

comparator should, as far as possible, be a comparison of alternative methods of

implementing a given project. This entails an expectation that, if the public sector comparator

proves more cost-effective it will, in general be implemented.

      The central implication is that private infrastructure projects should be considered

as part of an integrated planning process based on systematic benefit­cost analysis.

Proposed projects should be subject to a preliminary benefit-cost analysis, with the critical

ratio being the same as that required for approval of publicly-funded projects. If the

estimated benefit-cost ratio for a proposed project exceeds the critical value, it should be

evaluated against a public sector comparators, with the expectation that the more cost-

effective option will be implemented.


5.2         Real and spurious sources of cost difference


      Many assessments of contracting arrangements have been based on claims that on

average, the cost of providing public services will be reduced by 20 per cent as a result of

contracting out. This estimate is derived mainly from the work of Domberger and his

co-workers, and has been employed by the Industry Commission and other government

                                              35
agencies.

      However, other studies have suggested that, when the costs of tendering and contract

management are taken into account, and if there are no changes in wages and conditions

as a result of contracting out, the average cost saving from contracting out will be less

than 20 per cent in most cases. Arbitrary assumptions about cost savings have led to

failures in contracting, notably in the case of the (now-abandoned) system of centralised

contracting for IT services adopted by the Commonwealth government.

      In assessing the costs of private provision against a public sector comparator, it is

important to ensure that only genuine social cost savings arising from differences in

productivity and efficiency are taken into account. Sources of cost difference that should

be disregarded include:
        · Exemptions of state instrumentalities from taxes
        · Ability of private enterprises to avoid or minimise taxes
        · Differences in wages and conditions


5.3         The cost of capital


      The central principle on which the Partnership Victoria approach is allocated is that,

as far as possible, risks should be