MUNICIPAL FINANCE
BACKGROUND
SERIES
Linkages Between Local Governments and Financial
Markets: A Tool Kit to Developing Sub-Sovereign
Credit Markets in Emerging Economies
John Petersen
with John B. Crihfield
APRIL 2000
The World Bank






Linkages Between Local Governments and Financial Markets:
A Tool Kit to Developing Subsovereign Credit Markets in
Emerging Economies'
1.     Introduction
Decentralization and devolution of government throughout the world have
brought new prerogatives and responsibilities to local governments as service providers
to their local constituents. At the same time, and a part of the same movement, has been
the desire to use markets as allocators of resources, including the allocation of credit in
the private capital markets. This development has gone hand in hand with an increasing
disposition towards private ownership and operation of many activities formerly carried
out by government, where possible. A third trend has been the increasing globalization of
the financial and real markets. Reduction in barriers to the mobility of capital and goods
has been a near universal objective. It has also placed increasing constraints on the fiscal
and monetary policies of individual national economies.
These three trends are tending to converge in the increased need for local
governments to be self-reliant. They are also pressed to be more efficient in their
operations, and increasingly to raise revenues from their own resources, and to access
domestic and international markets in the process. By the same token, the trends of
devolution and international competition are placing limitations on the ability of central
governments to provide for the needs of the local governments. Correspondingly, the
international lending and donor institutions are both limited in their resources and are
often constrained by tradition to dealing only through sovereign units.
An upshot of these movements is an increasing emphasis on the ability of sub-
national units to access directly the private capital markets to meet their development and
infrastructure needs.2 The purpose of this toolkit paper is to describe the linkages
between local government credit financing needs and the capital markets in the emerging
countries. The paper is presented in the form of a framework for decision making about
identifying, analyzing, and strengthening these linkages. No two countries are alike.
Therefore, it considers a host of connections between the operation of the credit markets
at various stages of development and the structure, powers, and practices of local
governments. Given the existing environment, of which numerous examples are
presented, the paper considers a number of policy options.
I This paper draws on the authors' experiences in analyzing and implementing capital markets in emerging economies.
In particular, we draw from GFG's contributions to M. Glasser, e. al. "Formulation of a Regulatory Framework for
Municipal Borrowing in South Africa," Research Triangle Institute (November 1998 draft) and various background
reports used in conjunction with our work in the Philippines, Poland, the United States, and in international credit
analyses in conjunction with projects for USAID, U.S. EPA, and the U.S. Treasury.
2 We use the terms local, municipal, subnational, and subsovereign often interchangeably throughout this paper,
unless we are dealing in a specific context. The terms can also encompass states, provinces, and other subnational
governments, depending on the context. Debt, loans, and bonds are used interchangeably, again depending on context.
I



It is important to note that local governments, whatever may be the desired goals
of greater autonomy and capacity at that level, are found in a variety of existing contexts
in terms of political power and authority. These structures will shape the desirability and
applicability of many of the market-oriented and local empowerment suggestions found
in this toolkit. There are generic differences between unitary states, hierarchical federal
states, and those systems that recognize separate spheres for each level, including that of
local government. These differences are embedded in constitutions and legal systems
that, although subject to change, will greatly condition the degree to which subnational
governments are free to act on their own.
The toolkit does not attempt to be prescriptive as to the best approach in all cases.
But it does start from the premise that local control and decision making are good things
that are to be cultivated and encouraged. It also argues for maximizing competition
among private-sector financial options at the local government level wherever that is
possible and prudent, reflecting the fundamental belief that numerous alternatives are
invariably better than a "one way" of doing things.
TIhe potential scope of the subject of subnational governments' access to financial
markets is very large. It involves not only the many policy choices and interconnections
within the government sector itself, but also a host of options regarding the organization
and operation of the financial sectors in each of the respective countries, as well as their
respective appetites for borrowing internationally. Aside from only a few exceptional
cases, local governments in most countries have traditionally been relatively minor
players in the financial markets; but, that situation is and should be subject to change.
The term "credit market" is used broadly in this paper since in many cases of
emerging economies the leading provider of credit will be the banking system. For
reasons discussed below, this reliance on banks as financiers may either be a substitute
for, or a precursor to, a functioning domestic securities market in local government
obligations. However, the view adopted here is that the development of a securities
market should be as a means of obtaining long-term capital. Such markets, appealing to
nonbank institutional investors, were in a phase of rapid development until the recent
crises in emerging bond and equities markets.
The view taken in this paper is that the objective is to broaden the scope of the
borrowing decision at the subnational level, and that the desired end result will be a
private capital securities market representing potential private providers of credit and
many governments competing for that credit. Questions of the possible scale of activity
and the stage of financial development may mean that domestic securities markets may
be limited in their growth potential and may be supplemented or even supplanted by
regional or international markets. It may also affect the types of instruments and
intermediaries that may make the most sense in rising capital in selected markets.
Several vantage points are possible when surveying the linkages of local
governments to the financial markets. One vantage point is that of "global" policy maker
2



determining how best to fit a municipal securities component into the emerging domestic
and existing international credit markets, given a host of other policy constraints and
objectives. The second is that of the would-be borrowing local government that is intent
on achieving as much flexibility as possible in financing decisions and desiring to secure
capital on the best possible terms. The third perspective is that of the lender/investor that
needs information to judge relative risk and rewards, remedies in the case of trouble, and
assurance that the rules of the game will not be violated or hastily changed. These three
views will not be consistent in all respects at all times. But the ultimate objective is
assumed to be the same: to improve subnational governments' access to credit in ways
that are consistent with the overall fiscal health of governments and the viability of the
domestic financial markets.
An organizing principle is needed. In this case, it is assumed that a local
government securities market is desired and that subnational borrowing will be dictated
largely by the operation of the market, working within a framework of rules necessary to
keep it a free and efficient allocator among competing uses. Many conditions need to be
fulfilled for markets to operate in an optimal fashion. With respect to subnational
governments, for markets that otherwise are operating efficiently, we might restrict
ourselves to the following three key conditions:
* Subnational governments are taken to be largely borrowing at their own
volition and relying upon their own sources of security.
* The market is free of undue restrictions and makes its allocation decisions on
the basis of risk and reward.
*  The market has full access to information that allows it to judge the financial
condition of the borrower and to judge risk and reward.
The desire and ability of governments and financial markets to achieve the above
conditions will rest on a bevy of underlying and intertwining policy and technical issues.
Thus, in performing a diagnostic analysis of the options and possibilities, the tool kit will
ask the following categories of topical questions:
Credit capacity, borrowing powers, and regulation within the governmental sector:
What are the types of debt security?
What debt instruments are to be used?
What types of subsovereign units are best candidates to borrow?
How is local debt to be authorized?
What limitations should be placed on borrowing?
What is the role of monitoring and oversight?
What are the remedies in case of fiscal problems?
3



Investor needs, regulation, and the operation offinancial markets:
What is the financial market structure?
Who are the potential investors, their investment objectives and constraints?
What is to be the regulatory framework of the marketplace itself?
What is the role of disclosure and how is it to be accomplished?
What is the role of credit analysis and credit ratings?
How might various risks be mitigated by the private sector?
How should credit assistance be provided to comport with the market?
Some might say that the above two series of questions represents a separation
between, first, the demand for loanable funds by the public sector and, second, the supply
of them as provided by private suppliers of credit. Such a separation is conceptually
useful; but, in practice things are more complicated. As will be evidenced in this paper,
the linkages between subnational governments and the credit markets, even in countries
with only nascent financial markets, are diffuse and complex. Few of the actors are
interested only in one financial relationship, and the relationships are often more than
only financial. The common thread is that subnational governments are increasingly
important economic actors, and the stage on which decisions are made is increasingly a
market.
4



2.    Subnational Governmental Entites as Borrowers
Subnational debt may be the obligation of a local, regional, provincial, or state
government, or of projects that are sponsored by them, including projects involving
the private sector through subsidies, partnerships, and concessions. The legal and
financial relationships that subnational governmnents may enter into, including the type
of debt they contract, can differ markedly among countries. In many places these
relationships are evolving, and even where they are established, they continue to be
dynamic. Thus, the policy maker and the analyst alike must be prepared to look at a
variety of factors and risk exposures when dealing with the debt transactions of
subnational governments.
Analysis of subnational government debt can have many elements in common
with other borrower types such as public utilities and private sector firms. But, there
are some special features relating to the powers, structure, and operation of the
subnational governments themselves. For example, most local govermments deal
exclusively in the domestic currency for revenues and expenditures. Thus, except for
certain types of facilities (electric power, ports, airports, telecommunications), they
will not have much access to foreign currency payments. Also, governments typically
have certain powers over the local market for services that approach monopoly status
that may be enforced by regulation. On the other hand, governments are site-specific
and unable to change the geographic locus of business or the fundamental nature of
the services they provide. And, by that same token, they do not go out of business.
A fundamental distinction in classifying debt is whether the obligor is the
subnational governrment itself or is debt to which it is a party. The obligor may be
the subnational government itself, relying primarily on its taxing power and other
general governmental revenues to form the credit backing of the loan (i.e., general
government borrower); or the obligation may be limited to a particular revenue
source of an enterprise to which the general governmental credit is not pledged (a
limited or non-guarantee obligation). This distinction is reasonably clear in the
United States where a revenue-generating project or enterprise that is financed with a
limited obligation is referred to as a revenue bond.
However, the distinctions between general and limited pledges can be blurry.
This can be the case in a project financing where there may be mixtures of public and
private funds, service and off-take contracts, profit-sharing arrangements, concessions
with guarantees of use, etc. This is especially so in those countries where
governments formerly carried on a variety of commercial and industrial activities that
are now being totally or partially privatized. A major distinction is found in those
"project finance" cases where the private sector is a direct investor in a project, is an
equity provider and is actively engaged in its operation and management. In these
cases, the analysis may become very complex and involve a blend of risk factors
involving both the public and private sectors.
5



2.1 Prototype Credit Structures
The accompanying charts present simplified schematic illustrations of three
prototype financing and credit structures involving subnational governments. For
ease of exposition, we have assumed that the financing involves a project, which is
typically the case. However, the borrowing might be used for other purposes,
including relending.
a. General Government Obligation (general revenue supported)
In this case the government uses its general revenues to support the debt service
payments and owns and operates the project itself. In most countries, this would be the
likely structure for public safety, public education, and health and welfare capital
expenditures, which activities themselves are not revenue producing. The government
issues the debt in its own name and pledges its general revenues. The proceeds of the
borrowing are spent on a project. However, neither the project itself so financed nor
earnings from it are specifically tied to the repayment of the debt. An important
variation on this theme is where the local government gets intergovernmental
assistance, such as shared taxes or grants, which are pledged as part of the security.
A. General Obligation
Revenues
(  Debt    )<            Government
Project
6



b. Government Limited Obligation (enterprise activity supported)
In this case, the facility is an enterprise that produces revenues through charges
and fees that are used to defray much or all of the costs of operation and debt service.
The debt is secured primarily or exclusively on the project earnings, general revenues
are typically not pledged directly, and there may be a prohibition against their use.
Common subnational enterprises are public utilities, such as water and sewer, electric
distribution, local toll facilities, public markets, harvest processing facilities, and local
ports and terminals, etc. The debt is issued either by the project itself (which may be a
limited-purpose special district) or on behalf of the project by the general governmental
sponsor.
B. Revenue Obligation
Reenue
|Govement
c. Project Financings (public/private undertakings)
In this third case, which is typically a "utility" type project, the government
enters into arrangements with the private sector in the form of concessions or
partnership agreements to build, own, and/or operate the project. The government,
either national or local or perhaps both, may provide any number of inputs into the
financial structure: equity interests, subsidies, and various forms of guarantees related
to the demand for outputs or for supplying needed inputs. Likewise, the private sector,
or international lending entities, may contribute debt, equity, and various enhancements
to the financial mix . The.obligations of, and returns to, the respective parties are
determined by contract. The debt is typically issued in the name of the project and
may be nonrecourse, looking only to project earnings, ownership, and/or assets for
security.
7



C. Project (Public/Private)
Rvnue
Govemment                                     Private
2.2    Classifying Potential Subnational Borrowers
Many subnational governments already have access to credit. That may be in the
form of access to government sponsored lending programs, bank lending, or sales of
bonds in the domestic or international capital markets. However, many more do not have
access and the odds and avenues for them gaining it run a gamut of possibilities.
A fundamental consideration has to do with the fiscal capacity and financial
acumen of subnational jurisdictions. These two factors, which relate to the ability and
willingness to pay, largely govern which units are candidates for debt issuance. They are
not always correlated with size, but larger jurisdictions typically are of greater interest to
private providers of credit for a number of reasons, including greater sophistication, the
ability to draw upon more resources, and the ability to spread the fixed costs of debt
transactions over larger volumes of borrowing. In most countries, a familiar triage of
possibilities occur. Three groups of jurisdictions can be identified as regards to the
likelihood for the issuance of subsovereign debt in private markets3. These groups are:
* Those units that because of size, and.financial and managerial resources already
have access to capital markets.
3 Later, there will be a discussion of concessionary finance and technical assistance. At this stage, the concern is with
identifying the likely "potential market" for private-sector capital access and under what conditions.
8



*  Those that have no or only limited access to capital markets, but can generate
revenues sufficient to their responsibilities and are otherwise capable of
borrowing private capital. This group consists of those units that (a) are large and
that have sufficient capabilities to attract private interest without direct central
government help, and those (b) that are too small or that lack the managerial
capability to attract private lending at present, but with assistance could gain
access.
*  Those that cannot generate sufficient revenue either to provide the services they
require or to build the needed infrastructure. Jurisdictions in this group, which
for all practical purposes are "financial wards" of higher levels of government, do
not have access to capital markets and most likely should not.
Jurisdictions in the first two groups have the potential to use private credit
resources under a regime where any central government assistance rendered to municipal
market development is essentially accommodative and indirect.4 Those local units in the
third group, the small and poor, most likely can not and should not borrow in the private
credit markets.
As handy as this triage might appear to be, it is one that will defy the drawing of
strict lines of demarcation in practice. The advocate of light-handed intervention and
believers in market solutions will say that the market will define, better than government
regulation, which jurisdictions fall into which category. Any classification scheme will
be dynamic over time and subject to revision. For example, it is conceivable that some
governments that are generally too small and too poor to gain access to markets using
their general revenue funds may latch on to a project financing that is creditworthy. Or, if
they choose to pledge (or can pledge) their intergovernmental payments, even those with
otherwise insufficient own-source revenues might qualify for private credit.
Government policies regarding intergovernmental finance and governing
technical and credit assistance to small and unsophisticated jurisdictions will affect how
the market assesses creditworthiness. As a general rule, countries that have a weakly
financed and unsophisticated local government sector will find themselves having to do
one of two things. Either they must largely forego the private securities market as a
medium for direct borrowing, or they will need to devise active policies to help move
local governments up the creditworthiness ladder.
4 By accommodative and indirect we mean that the assistance is not given to particular jurisdictions in terms of
concessionary credit assistance but rather is directed at creating laws and regulations that permit or engender private
capital flows to subnational govemments in a securities market context.
9



2.3    Making Distinctions Among Local Jurisdictions
Formulating the triage of local jurisdictions used above on the basis of existing
creditworthiness is useful for analytic purposes, such as describing potential demand for
credit access and the likely size and viability of a local government securities market.
But, should such distinctions be codified into law or regulation for purposes of
predetermining which units can access the markets and obtain credit? In developed
economies, freely operating credit markets effectively classify borrowers on their own,
and reflect their credit assessments in the prices charged for borrowing. Nonetheless,
even in these mature markets, regulatory classification is sometimes practiced by central
or state governments in order to provide certain privileges to some borrowers or to
impose restrictions on others. For example, in the United States most state govermnents
differentiate among local governments through various legal classification mechanisms,
and these differentiations can include differential borrowing powers. However, the credit
market itself further differentiates among govermnents, based on varying assessments of
creditworthiness. These assessments are based on perceived differences of the
jurisdictions' economic vitality, managerial efficiencies, financial condition, and the
necessity and viability of individual projects.
We believe that detailed regulatory prescriptions are best avoided. But, senior
levels of government have a legitimate interest in how their progeny and juniors behave
in the financial markets. If a significant amount of questionable subsovereign debt is
issued, two harmful things could happen. The severity of each depends on the political
makeup of the country. If the sovereign is providing either a dejure or defacto sovereign
guarantee, large amounts of failure can lead to national bailouts that increase the supply
of credit and lead to inflationary pressures. Domestically and internationally, in the
absence of national government bailouts, widespread defaults could create distrust about
such debt generally, and could cause general market instability.
To guard against imprudent behavior, most national governmental credit systems
use classifications. These classifications are not meant to prohibit debt for some classes,
but to differentiate among jurisdictions in terms of allowable maximum outstanding debt
or, more typically, the maximum debt outstanding in relation to some revenue source,
such as the property tax. However, in subsovereign financial systems being put into place
for the first time, or radically redesigned, there may be a tendency to overuse
classifications, which themselves may be poorly designed or not enforceable.
The strongest argument against rigid regulatory classification is that upward
mobility between categories of financial strength and managerial maturity should be
encouraged. Classifying a jurisdiction in a way that encourages it to be dependent on
external assistance and avoid responsible borrowing on its own is exactly the opposite of
the effect the government intends to have. Artificially limiting market access runs
counter to the basic policy goal of maximizing private sector investment.
10



2.4   Subnational Borrowers by Governmental Type
Within the local government sector, debt may also be incurred by municipal
enterprises and quasi-municipal entities. These types of borrowers can be created by
agreement of existing municipalities, or by legislation of the national or regional
government. There are at least three kinds of special-purpose arrangements. These are as
follows:
a)    Separate restricted funds, accounting arrangements, or special purpose entities
within a municipality, whose revenues and expenditures are restricted to specific
purposes and are separated from the general fund. These entities typically derive their
power from the municipalities, although they may be run with a good deal of
independence.
b)    Entities created by agreement by more than one municipality to accomplish a
special purpose (e.g., to provide fire protection efficiently across a broad area). Their
revenues and expenditures can be separated from those of the organizing municipalities.
Their powers can derive solely from the municipalities ("joint powers"), or though
legislation which can limit or extend such combining powers.
c)    Quasi-municipal entities created by state or national legislation. These entities
might provide municipal services (e.g., water development, disease control, or transport
services) where needs do not necessarily relate to municipal boundaries. They have such
powers as are described in authorizing legislation.
In theory, these three types of subnational special-purpose governmental entities
might issue limited obligation debt based on their own revenue sources and the ability to
borrow against them. That is, they do not necessarily have the guarantee of a parent
municipality.
As one might expect, there are two sides of the special-purpose, special entity
borrowing coin. The argument against such issuers is that they can lead to a proliferation
and fragmentation of local government and can cause diffusion of local revenue sources.
The argument in favor of such issuers is that they allow services to be delivered by an
appropriately crafted entity with targeted taxes, fees, and charges. Since geographic areas
of traditional general governments are typically inherited from a political and economic
logic that may be long past, the case for promoting special service districts along the lines
of economic service areas is often compelling.
The practical implication of the above discussion is that local government
borrowing powers should be kept widely available and flexible in the sense that they may
be harnessed to addressing not only common but also special infrastructure problems. A
special case in point, as we will examine below, is the special taxing and fee district
which may pernit a general unit of government to gear its taxing and charging powers to
11



the particular needs of subdivisions. The special taxing district has seen a good deal of
use in the United States.
One application of special districts is in the use of business improvement
districts. These are special taxing units that levy a tax in addition to the normal taxes and
have powers and personnel to address the special needs of downtown areas, especially
those that are subject to urban decay and distressed conditions, including extraordinary
sanitation and public safety needs. The concept has caught on in parts of Europe and is
beginning to spread to developing economies as well.'
Definitions and Cntrol f Public Debt
The defintion of public debt ca be an important policy issue in deciding the
boundaries of local government borrowing preroatives. In the European Union
legislation, which is subject to the deficit and debt limits contained in the Maastricht
Treaty, there are limits placed on public indebtedness which include the debt on the
central, regional, and local goverments including social security funds, but excluding
public enterprises. The limitation thus is not expressed in terms of the ownership of the
facility but rather the activity, the general govenment sector being defined as the
institutional units producing nonmarket services as their main activity. Effectuation of
the limitation has led to several concerns about how to coordinate debt incurred at the
sovereign level with that at the subsoveregn lvel.
The upshot Of the EU definition is that for local units, which are subject to the
above limitation, will have an inducement to place as much of their debt as possible on a
self-supporting, commercial basis in order to avoid the macro-level curbs on borrowing.
Evidently, however, the EU definitions would also include certain contingent obligations
that subnational govermnents might enter into in support of commercial debt, such as
obligations to purchase a commodity or service (an off-take guarantee) and pledges to
make up project operating deficits or debt service deficiencies from general flmds.
The curbs on general obligation tax-supported debt embodied in the EU limits are
akin to the individual state-based limitations of tax-supported debt that arose in the
United States. In the U.S., the restrictions on general debt hastened the rise of the
nonrecourse revenue-bond obligation that is used for enterprise activities and other forms
of non-recourse obligations such as the moral obligation bond.6 These limited
obligations, many of which are de facto supported by taxes and fees raised by the general
goverunent, once represented only a minor fraction of municipal borrowing. They now
typically make up around 60 to 70 percent of all bonds sold in the United States.
5 "Johannesburg's Changing Center," The Economist (November 7 - 14, 1998) p.50
6 See J. Petersen and R. Forbes, "Remarkable Rise of the Revenue Bond" in G. Kaufman (ed)
12



3.    Types of Debt Security
As noted in the preceding section, types of debt are defined by the kind of security
given by the borrower. Security (that is, where the money will come from to repay the
debt) typically runs to two questions (1) where the money is expected to come from and
(2) where it might have to come firom, which is the ultimate collateral and remedies in
case of failure. Before investing, creditors want to know not only where the money is
expected to come from, but also what are their remedies or security in the event of
default. Unless the remedies and security are deemed adequate, the markets may set risk
premiums that make credit unaffordable to many jurisdictions. Knowing the security on
debt is important: uncertainty as to remedy and security builds inefficiencies into the
market.
The question of what remedies should be available by law to creditors is critical.
Any framework for subsovereign borrowing needs to spell out what powers a jurisdiction
has to pledge assets, revenue streams, and to exercise its powers to set taxes, tariffs, and
other levies. It is also desirable to spell out how such security can be affected in the event
of default or other financial emergency.
3.1    General Debt
In most emerging countries, with the exception of the largest units, general
purpose subsovereign debt has in the past carried some form of sovereign government
backing. Increasingly, national governments are attempting to wean local governments
off of such backing and encouraging them to borrow on their own credit. The credit of
the subnational unit, without an explicit or implicit guarantee by the national government,
is often not clear. Accordingly, expressions such as "general obligation" or "balance
sheet" debt often mask an unresolved question of ultimate security: that is, just what
remedies are available to an investor in case the unit fails to pay on time and in full.
Questions of limited obligations and special security are dealt with below. Here we focus
on general obligations, incurred on behalf of a jurisdiction as a whole, to benefit the
population at large.
Aside from the good faith of a jurisdiction and the prospect of national assistance
if things get difficult, subsovereign general obligations have often been backed by the
ability of creditors to seize financial and physical assets. For a number of reasons, this
physical collateral system is not a sound approach to securing credits. Local governments
with physical assets that are unrelated to their municipal service responsibilities, e.g., a
commercial enterprise, might be best advised to divest themselves of the asset so as not to
divert scarce city management capacity to managing a potentially private activity.
Municipal assets that are used directly or indirectly in the provision of vital services
should not be risked as collateral. Although often desired by lenders, physical collateral
13



can divert the jurisdiction's attention from making sure its general revenues are sound
enough to support borrowing.
The terminology and meaning of the general obligation pledge is subject to
variation. The term "full faith and credit" originated in the United States and is generally
understood to mean more than a general, unsecured promise.7 Debt not backed by
specific revenue flows should be backed by a pledge of all general revenues as a source of
debt service payment. The jurisdiction could be specifically obligated to use any and all
of its general resources, including the raising of taxes and fees as required, to meet debt
service obligations. Stronger and more specific remedies for creditors are likely to
strengthen investor confidence in subsovereign debt in emerging economies, a subject we
return to in Section 10 of this paper.
3.2     Special Pledges and Limited Obligations
The above discussion considered the forns of security and remedies implicit in a
general obligation, promise to pay using general powers of taxation and other general
government assets, if needs be. We now consider the kinds of limited security that can
be given to secure subsovereign debt. Limited security arrangements include the
following:
o Pledging of physical or monetary assets.
O Pledging the right to operate a facility or provide a service.
O Pledging of selected revenues, such as:
from tariffs, fares, or rentals.
from particular taxes or special levies.
from grants or shared taxes (intergovernmental transfers).
- Pledging the power to set specific tax rates, utility tariffs, and other levies.
* Pledging by the executive to budget for and recommend payment of future debt
service, without an explicit binding pledge that those appropriations will be
made.8
* Pledging to assign the payment of future intergovernmental revenue.
7 In the United States the term full faith and power means the application of the general taxing power to the repayment
of the debt. That power in its traditional and strongest formulation has meant the imposing of taxes "unlimited as to
rate or amount" sufficient to repay the debt. Over the years various limitations and caps entered into the pledged
taxing powers, although by and large it remains as the ultimate in taxing power pledges. As a practical matter general
obligation defaults are almost unheard of and are promptly cured by a mandamus from a court to levy taxes and the
intervention of state governments to make sure that that happens. States in the U.S. are usually very sensitive that they
not get a bad reputation by the failure of a local govermment not to pay its debts for fear of the implications for other
local governments. Default by a general government means the loss of local governing powers, sometimes with the
appointment of a control board or a receiver to take over operations until the debt in resolved.
8 This type of security is known in the U.S. as appropriation or moral obligation debt. It recognizes that debt is not a
full faith and credit binding obligation but rather is subject to the will of successive legislatures. Its origins are lease
rental debt that holds that the obligation only runs from fiscal year to fiscal year and is subject to reconsideration of the
legislature each year.
14



These several types of local government security afrangements, and how they are
effectuated, are discussed in the following sections.
3.3   Pledges of Assets, Operating Rights, and Revenue Streams
A very common form of pledge backing debt in developed markets is one which
is restricted to a particular revenue stream or enumerated local government assets. Also
possible is for the creditor, in case of default, to step in and perform the activity and
receive the revenues. But, both the carving out of specific revenues and giving the
creditors rights to assets and operational powers raises a host of operational and policy
issues.
Local officials may be hesitant to pledge assets because the loss of such assets and
their benefits in the event of default would be dramatic. There may be assets or revenue
streams that are so vital to maintaining basic governance that they should be protected
from a debt service pledge. Examples of such vital assets and operations may include the
following:
I ntergovernmental share transfers to provide services to selected segments of
the population or earmarked purposes.
X Those physical facilities deemed essential to the public health, safety and
welfare, such as water supply, fire equipment, and hospitals.
Another consideration is that a government will have a limited number of
pledgable assets, especially non-vital properties. The extent to which they are forced to
pledge these assets will reduce their future ability to secure loans, and a loss of future
fmancial flexibility.
As a practical matter, private-sector lenders are of two minds when it comes to
asset pledges. They are anxious to achieve as much collateral as possible and to apply
leverage to reluctant debtors. However, lenders will be reluctant to take indispensable
physical assets as collateral on a loan. They may be willing to take marketable assets,
such as vacant land, office buildings, parking lots, or sports facilities, but they are not
likely to foreclose on essential service facilities, such as water or wastewater treatment
plants, town halls, city streets, or fire stations. If the legal system provides adequate
assurance, the lenders are more likely to secure local debts by pledges of actual or
potential revenue streams that are sufficient to cover the debt service.
Because of the practical and political problems of tying up essential facilities, a
prohibition on pledging properties deemed essential to the public health and safety could
be included in authorizing legislation with little impact on a market's development of
other useful security devices. If this restriction were included, minimum essential
services can be defined by law and then left to the underlying borrowers to decide what
15



fits the definition. These might include water, sewer, and refuse collection as is
necessary for human health and safety. Alternatively, assets relating to these services
could be pledged, but the law should require minimum essential services to be continued
at all times. These restrictions would not be barriers to borrowing, since lenders are not
interested in repossessing pipes in the ground, but rather desire revenue streams.
A pledge of revenues from public utilities is appropriate for financing related to
those utilities that benefit from the financed improvements. But, it raises concerns if the
pledge of revenue-producing properties is used to secure unrelated financing. Part of the
concern is simple economics. When a jurisdiction subsidizes general expenditures at the
expense of utility charges, it results in a misallocation of resources. In economic theory,
and usually in practice, the service that does the subsidizing is under-allocated, and the
service that is subsidized is over-allocated.
Importance of the Rate Setting Pledge
South Africa presents an interestin example of the importance of the rate setting
pledge in a revenue bond. and the potential problems when its application is uncertain.
Several South African cities are attempting to implement privatization plans that involve
what in fact are non-recourse revenue bonds. These are intended to rely exclusively on
the water tariffs of the privately operated water treatment plants to pay debt service. As is
typically the case, the tariffs will need to,be increased over time to meet increased
operating costs or to offset unforeseen expenditures. How , legisl    n the national
congress gives a national mrinister the discretion to set water tariff rates, in effect over-
riding local control and contract. If that hapens then the private concessionaires want
the local communities to agree to raise propert taxes to the degree needed to make up
any shortfall in revenues.
While the latter alternative improves the creditworthiness of the bonds, it has
raised other problems. First, the general tax base of the municipality is eroded by the
"standby" liability in case water rates are not raised and taxes must be. Second, the water
ratepayers and the property taxpayers are not one in the same. The multitude of water
users own little if any taxable property and they greatly outnumber the better off property
taxpayers. This opens up the potential that local elected officials might not regret, but
might indeed welcome the reneging on water rate increases by the national official since
it would shift the burden of the debt makeup to the wealthier, but fewer, constituents that
pay the property taxes.
3.4   Pledging to Set Tax Rates, Tariffs, and Other Levies.
16



Local revenue-raising powers may be limited in emerging economies. But, even
where the powers of local governments to set rates and make levies do exist, there can be
ambiguity as to whether local government can pledge to set tariffs, tax rates, or other
charges at a level sufficient to service a debt.9 There can be issues as to whether such
covenants unlawfully bind future administrations and legislators in the exercise of their
prerogatives. However, without such a forward-looking and binding-contract ability to
pledge, the pledge is much less meaningful to the point of being worthless. Where the
tariffs, rates, or charges can be increased or decreased at the discretion of the local
authorities, a covenant to set and maintain those tariffs, rates, or charges at adequate
levels to meet operating costs and pay debt service is a useful financing tool.
The experience in many emerging market countries is that the primary cause of
debt service default and payment arrears is failure to increase rates and charges that were
planned to be the source of revenues. Subnational jurisdictions will benefit from clear
legal authority to covenant future tariff or tax increases to secure debt. Jurisdictions may
choose whether or not to use this mechanism, but they should have the legal authority to
covenant, if possible.
3.5   Intergovernmental Revenue Intercepts
In many countries local jurisdictions can assign creditors their interest in specified
revenue streams, such as shared taxes and grants, received from the national or other
higher-level governments. Revenue intercepts are attractive to a creditor because they
promise stable and predictable revenue streams that can be easily accessed to pay debt
service. Intercepts can be designed in various ways either to assure adequate funds are
available to meet debt service payments prior to their coming due (an ex ante intercept) or
to be tapped only in the event of a local government's default (an ex post intercept).
Another variant is to have a bank credit facility standing by to advance money in case
funds are not on hand to meet debt service payments and then have the loan repaid out of
future intercept receipts.
In the case of pre-assignment of revenues to pay debt service, some have argued
that it may be tempting for jurisdictions to routinely not budget for or to pay debt service
and to force creditors to rely on intercepts. Accordingly, intercepted protected creditors
become lazy and do not care either about the underlying worth of the investment being
financed by the loan nor about the local government's financial condition or performance.
This, it is argued, neither encourages responsible priority-setting at the local budget level
nor promotes a rational market allocation of resources. If this "asleep at the switch"
problem is thought to be compelling, it can be addressed by assessing a significant
penalty or administrative fee to the borrower that routinely uses the intercept to pay its
debt service bills. This penalty ensures that the jurisdiction is always better off collecting
its own revenues and paying its own bills.
9 Such provisions are known as rate covenants.
17



There also may be a tendency to use withholding of transfers for all kinds of
purposes, not the least of which is ensuring debt repayment. In South Africa, for
example, there is a move afoot to use a share of the transfers for financial capacity
building and in the Philippines a mandatory 25% share of the internal revenue payments
to local governments must be used for developmental investment. Each instance of using
leverage over transfers to assure local activities may make sense individually. But, if the
mechanism is used for many different purposes, the overall intergovernmental transfer
system can become splintered into a set of constraining and competing mandates.
Intergvermetl Trasfer Payments As Collateral
Typically, in newly emerging economies, ;local governments are highly
dependent on transfers from the cental government. While these can be very volatile
and untested for sustained periods of time, they form a major portion of revenues and
are attractive for itercepton to cover debt service payments. Columbia offers an
exwnple of the widespread use of intercepts of central government payments to cover
debt service through the Findetere prgranm.
As a rule, if intergovernmental payments are used for pledging, there should be
a coverage factor so that the historic and/or expected level of transfer covers the debt
service payments by some fraction greater than one. In the Philippines, about 50% of
all revernes received by cities are  rough tergovernental transfers from the
national government, and for provinces the average figure is 75 %. The smaller and
more rural the local goverment, the higher the proportion of transfers to total
revenues. In that country, government owned banks (the de facto required depositories
for local governments) have gotten deeds of assignment of transfer payments to cover
bank loans. As aid is received, the banks have a right of offset against any loan
amounts owed the banks prior to dispersal for other purposes.
Intercepts can have a powerful impact on local borrowers, especially small and
remote units. The assignment to bondholders of state payments to local school districts
(which typically make up over 50 percent of revenues for the districts) is very common
in the United States. It is the basis for the high credit ratings enjoyed by local school
districts covered by such programs, the rating of which is usually only one full notch
below that on the state's general obligation debt.'I As a result of this wide-spread
appreciation of the impact of state assistance and other small-borrower preferences,
local schools are among the lowest cost borrowers in the U.S. municipal bond market.
3.6    Enterprise or "Self-supporting" Financing
10 One notch means if the state's general obligation is AAA, then the intercept protected debt will be rated AA.
18



A common use for special pledges of revenues and assets is for a self-supporting
"enterprise" that generates its own means of repayment and does not rely on recourse to
general revenues." This limited obligation involves the pledge of only specified system
or project revenues to repayment. In accounting terms, this implies the creation of a
special fund to receive the specified revenues, which are expended to meet costs
associated with the enterprise, including the payment of debt. This concept focuses credit
concerns on the viability of a particular project or system, rather than on the viability of
the jurisdiction. Poor or unsound general purpose jurisdictions can have viable
enterprises, and vice versa.
There are several advantages to this type of financing:
*  It establishes a relationship between the cost of service and service prices. This
promotes more efficient operations. The cost/price relationship need not be
absolute and can be modified in practice, but it has the benefit of assembling the
costs and making any subsidy paid transparent.
3  If the utility has been run at a surplus to subsidize other governmental functions,
then the added "tax" burden resulting to utility users becomes evident.
-  To the extent that general revenues have subsidized enterprise operations, the use
of dedicated revenue structures will free up general revenues to pay other costs.'2
-  Management and operation of revenue-producing facilities tend to be more
efficient and the facilities are better kept up since they need to be in shape to
produce revenues. This can be encouraged by contractual provisions protecting
income and value, paired with creditors' active interest in assets and their
operation.
*  When there are legitimate reasons to use general revenues as well as specific
revenues, it may be better to use general revenues to reduce the amount of debt
incurred. For example, this can be done by making a municipal "equity"
investment in the asset up front, and borrowing to build or acquire the rest of the
asset, pledging only revenues produced by the asset, or even a part of the asset's
operations, to meet debt service requirements. This practice is common in
Western European countries for many municipal utility operations.
There are also disadvantages to limited obligation, self-supporting financing.
11 This is the traditional notion of the enterprise revenue bond. There is a legion of variation on the idea, many of
which have been designed in the U.S. to circumvent restrictions on tax-supported debt. This approach allows and
encourages allocation of full costs of services to the beneficiaries. Economically, this is desirable because it leads to
efficient allocation of scarce resources.
12 In some places, such as in South Africa, the subsidy runs from the utility to the general fund, rather than the other
way.
19



*  The expressions "asset stripping" or "security dilution" convey the concern faced
by existing creditors ofjurisdictions that have relied on a utility to generate
subsidies for the general fund, when those revenues are instead pledged to a
utility-specific purpose.'3 In other words, where prior lenders have looked to the
overall revenues as a source of repayments, a subsequent sequestering or stripping
away of revenue streams weakens the credit.
*  Limited obligations may hinder redistribution of infrastructure and services
among population groups (say, fron rich sections to poor ones) by keeping
potentially redistributable revenues for the benefit of an already privileged area.
If there are to be preferential and redistributive policies, they typically must be
financed from the sponsoring unit's general funds.
*  Future financing options are restricted. Enterprise financing is a matter of
contract between the public-sector sponsor, acting on behalf of the enterprise, and
the investors."4 Investors typically will require restrictions that reduce the options
of borrowers in the future. For example, borrowers must meet certain conditions
before they are able to issue more debt secured on the enterprise earnings
(additional bond test), must conform to certain requirements regarding reserves
and insurance, and must abide by a rate covenant, as was discussed above.
In some countries the local governients have been able to use utility surpluses as
a means of subsidizing the general budget.'5 But the ideal of transparency implies that
utility surpluses used for cross subsidization should not be buried, but should be reflected
in a specific tax or surcharge that reflects the added cost of cross-subsidies. Without this
transparency, it is not possible to see if the utility is operating at an economic optimum in
getting the most delivered service per unit of input. Having reached that optimum, the
redistribution becomes a clear added cost for some and benefit for others.
The above discussion focuses on that enterprise financing associated with
traditional "natural monopolies," public utilities provided by local governments.
However, there are other potentially revenue-producing activities that are "semi-
commercial" in character, such as transportation terminals, public markets, abattoirs,
farm processing plants, and the like. In addition, industrial estates, tourism facilities
(including hotels), toll roads and bridges, and a number of other revenue-producing
facilities are candidates for complete or partial financing by the use of "revenue bonds.'
13 However, most economists would applaud this elimination of the cross-subsidy on efficiency grounds.
14 For example, there may be requirements that the borrower not pledge the same asset to another lender, except under
stated conditions (additional bonds test), that the revenues provide certain coverage of the debt service (rate covenant),
and that revenues be retained for use on the facility and to the benefit of bond holders (closed loop). Negotiation of
these restrictions and the associated tests is an integral part of the borrowing transaction.
15 Subsidies can be hard to detect. A government may fleet many of its costs off to the utility by allocations of
administrative and other costs to the utility or it may receive utility services below cost or for free. Where the utility is
part of the government, the allocation of costs can be highly judgmental.
20



Critical components of such proposals are the reliability and growth of revenues, the
technology used, the facilities' adequacy, firm construction dates and costs and future
operating costs being either fixed or controllable in the future.
Depending on the nature of the facility, determining the nature of and risks in
these technical and economic factors requires engineering studies and market demand
studies. The objective of the studies, from the financial perspective, is to obtain
objective estimates of what will be the net revenues available after operations to pay debt
service on the loans or bonds and, if not, what steps might be taken to ameliorate the
situation. Such studies are especially critical in new, free-standing projects where there is
no experience yet with operations.
21



;Impvortce of Feasibility Reports
Projects ftat are intended to be self-supporting should generate sufficient revenues
to pay for their operation and also tomeet some or all of their capital costs. In many
cases, projects represent m     ies either because of being essential in a technical sense
(water and electricity), or are exclusive in location (toll roads and bridges) or because the
sponsoring government exercises a high degree of market control by regulation (solid
waste and parking facilities) Other cprjects, however, such as sporting or cultural
venues, are not essential, suect to competition and fice greater market demand risks. In
any case, fiacilities may be subject to construction and technological risk, such as cost
overruns, strtup dela    or failure to pro e the eed   amounts or quality of output.
Assessingthesefacrs andasoct risks is at theheartofthe engineering and
marketing studies that underlie determ ing a project's feasibility. Emerging economies
are often short-handed of he technical sis needed to put together engineering and
market demand studies and the indepdence needed to provide objectivity in the
analysis.
The difficulties of separating proect promotion from technical analysis and the
unfortunate consequences of not doing so are reflected in the fate of the sports complex
built in the municipality of San Pedro Sula, Honduras. The complex was built to host the
Central American  iames. It was expecte to be able largelyto pay for itself. ;Some $15
million in bonds were sold to help offset the anticipated costs of $25 million (which were
expected to be offset by a varit of revenues). In fat4  the project ended up costing $36
million and net project revenues fell far belowexpectations. The city made a general
obligation pledge:in addition to the project revenues and it is now in serious difficulty.'6
The issuance of the bondsperse was not the problem with San Pedro.'7 Rather, it
was a lack of analysis that permitted the city to take on large and unfathomed risks.
Among the cited problems with feasibility study were:
*  no analysis of the market for "special seat" sales, although these were expected to
generate nearly 80% of the operating revenues
*  no detail of the assumptions that were used in the construction estimates
*  no project alternatives were considered, such as upgrading an existing stadium
*  no evaluation of using the private sector and more equity in the construction project
v  no description of the various risks or possible contingencies if the complex's revenues
were not realized.
16 San Pedro attempted to sell the bonds in its limited financial markets with no success (all the bonds went to the
contractors). The city was faced with a lack of bank credit, in part due to the weak and oligopolistic nature of the
domestic banking system.
17 Giovanni Giovannelli, Non Performing Municipal Borrowing in Central America - A Case Study First World Bank
Conference on Capital Market Development at the Subnational Level, Santander Spain (October, 1998)
22



In short, there never was a credible assessment of either the project's economics
and prospects nor of the impact on the guaranteeing municipality's finances.
ail ure to do engineering and feasibility studies or of having them performed
objectively by skilled professionals can lead to severe problems for the sponsoring local
government, especially if it pledges its own credit as part of the security. See the box on
the feasibility reports.
3.7    Special District Financing
A variant of enterprise financing is special district financing. A special district is
created to provide infrastructure and services to a subset of the general population and to
a geographic area that demands special types or levels of service. These districts have
been used to provide urban services (water, sewer, and roads) to areas that are developing
rapidly or that have special needs (such as downtown areas). They may be used to
provide infrastructure, services, or both.8 They are common in Western Europe and the
United States, and are beginning to emerge in developing countries.
There are many examples; but the common theme is that if there are special
benefits that can be ascribed to a particular area, then the special district provides a
mechanism for recovering costs that are associated with the benefits. Special districts can
provide a way of transcending political boundaries or uniting jurisdictions into a
single financing unit to provide a regional service. Of particular note is where new
development or extensive redevelopment is occurring and where property (as opposed to
individuals) is the long-term beneficiary of the improvements. For example, new public
utilities and transportation, storm drainage, parks, and similar improvements increase the
attractiveness of an area and enhance property values. As the area is improved by these
expenditures, the amenities themselves and access provided to them increase property
values, as well as other indices of economic activity and worth. If the costs of the capital
improvements are bome by the public sector, then the public sector should have a way of
capturing its investment. Such a value capture mechanism is possible with a special
district that gears its taxes or charges to pay for capital improvements, the benefits of
which were received by the improved properties.'9
A successful special tax relies on good and timely measurement of values and an
efficient collection system. In the United States, for example, some special taxing
districts are administered by private for-profit organizations that undertake the
18 A district may be "dependenf' and overseen by the governing body of the municipality, or "independent" with an
autonomous elected or appointed board. In many areas storeowners or homeowners form associations to manage the
district and levy charges. The key is the ability to levy taxes and charges, and to attach (i.e., to seize) properties that do
not pay.
19 In its most common form in the United States and a few other places in Europe, the tax district uses property taxes
(expressed as a percentage of taxable property value) or assessments (fixed dollar levies). However, the district can
use other bases to charge for the benefit or service, including square footage (or meters) or front footage (or meters),
number of vehicular trips (for roads), inpervious surface (for drainage), lumens of light (for lighting), residential
bedrooms (for educational facilities), square footage of space (for parking), and so forth. The key is that there be a
logical "nexus" between the improvement and the form of charge that is used.
23



calculations and do the tax billings, acting as agents for the host governments. The entire
revenue-raising mechanism is meant to support obtaining credit and is specified in the
loan or bond agreement. This improves the administration of taxes, and the integrity and
efficiency of the system becomes, in effect, a matter of contract with bondholders. In
Western Europe the special district or special authority covering all or parts of more than
one political jurisdiction has facilitated the subsequent privatization of services, such as
water utilities in the United Kingdom and France.
24



4.    Debt Structures, Instruments, and Methods of Sale
Debt structure and instruments have to do with the mechanics of how the principal
of debt is to be repaid, how long a debt will be outstanding, and how interest due on that
debt will be figured and paid. Method of sale has to do with the procedures by which
debt is offered to the final investors and debt obligations exchanged for the bond
proceeds. As a rule, these matters will largely be determined by the market. That is as it
should be, assuming that the market is competitive and that debtor, investor, and banker
are all equally well informed. The financial markets are fluid, and what might be
attractive one day can be unattractive the next. Inflexibility is costly.
A major concern at the national level is not to create regulation that interferes with
the flexibility of lenders and borrowers in structuring debt in ways that best suit both
parties. This section examines the case for maintaining flexibility and alternatives in the
structuring of local government debts. It describes debt structure and illustrates the range
of instruments available to suit the profiles of issuers and investors.
4.1    Maturity or Term of Debt
The maturity of a debt instrument should be no longer than and, better yet,
matched to the economic life of the asset it is financing. Ideally, amortizing the liability
on one side of the balance sheet is matched by the depreciation of the asset financed.
Thus, infrastructure assets, such as water systems, roads, or municipal buildings, which
typically have lives of 15 to 30 years, should, in principle, be financed with long-term
bonds of similar duration.20 In many emerging markets this matching is impossible to
accomplish as investors are unwilling to extend loans beyond a few years and even then,
the rates of interest may be exorbitant because either there is large perceived risk beyond
a few years and/or capital can be committed at short maturities at substantial returns.
Even if longer term capital is available, the shape of the yield curve may make
borrowers prefer the use of shorter term debt." Typically, the yield curve is upward
sloping, i.e., the longer the term of the debt, the higher the interest rate payable. The
upward slope is caused because investors want extra compensation for the lack of
liquidity of long-term lending. They have increasing uncertainty as to what economic
conditions, price levels and interest rates will be in the future. But that is not always the
case. Short-term rates may be temporarily driven up by liquidity shortages and efforts to
defend the currency. Moreover, if expectations are that the prevailing level of interest
rates is unsustainably high, and that rates are expected to fall, then the yield curve may be
inverted, with short-term rates higher than long-term rates. In such cases some borrowers
20 Matching asset life to debt term is also sound public policy because facilities can be paid for by those who use
them.
21 The yield curve is sometimes called the term structure of rates and relates the interest rate to the term of the debt
(the years it will be outstanding).
25



may borrow short, if they believe long-term rates will fall. Others may choose to lock in
the relatively lower long-term rates.
There is also a trade-off between the lower rates typical of short-term debt and
refinancing risk. If the debt is shorter in maturity than the life of the asset, the issuer is
exposed to refinancing risk, i.e., new debt may have to be raised during the life of the
asset at a higher rate than the original loan. If the borrower's credit risk has worsened, it
may not be possible to refinance. Refinancing can, of course, work in favor of the
borrower. That happens where the general level of interest rates declines or where the
issuer's credit has improved. In the case of general obligation bonds, the latter situation
could happen as a result of improved general creditworthiness of the jurisdiction. In the
case of project finance, the construction and initial phases of operation are riskier than
later phases in a mature project, and it may be possible to refinance at lower rates.
However, financiers are aware of this, and rely on the later phases to provide some
compensation for the additional risk taken at the outset, so would probably reserve the
right to refinance for themselves. All in all, maintaining an unhedged position is risky,
and not advisable with public funds.
4.2    Debt Service or Repayment Structures
The cash flow profile of debt refers to the way in which the borrower pays and the
lender receives interest and principal over the life of the liability. Several common
profiles are described below. A term bond typically has a fixed rate and term, and pays a
semiannual coupon with the principal due at the end. On the other hand, serial bonds
have the principal repaid in installments over the term of the debt. These repayments
may be variously structured. Principal may be repaid in equal increments, which is called
a level principal structure and leads to progressively smaller debt service payments
("front-end loaded"). This is a conservative approach that leads to rapid repayment of
debt and frees up future borrowing capacity quickly.
On the other hand, if the debt service payments are level, then the earlier year
payments contain proportionately more interest than principal. These are called a level
debt service structure. Alternatively, the debt service schedule may be structured to
increase over time. Use of the increasing debt service structure ("back-end loaded") in
municipal bonds is increasing in the United States and elsewhere in those cases where
growing revenues are believed very likely to occur.
Original discount bonds, called "zeros" when they fully discount future interest
payments, pay no or reduced interest coupons. The investor realizes interest return by
buying the bond substantially below its principal value. Such bonds can be originally
issued at discount or created synthetically by investment banks by stripping the coupon
off a standard term or serial bond. Zeros are attractive to parties who wish to secure a
fixed amount of capital in the future without being exposed to reinvestment risk. When
the zero bonds are created synthetically, the coupon stream from a stripped bond can be
sold to an investor who is primarily interested in an annuity flow.
26



Cash flow profiles can be engineered to match the cash flows generated by the
activity that is being financed. Liabilities can be index-linked, where the revenue flows
are expected to vary with an index, such as inflation or an input cost. In this case, interest
payments can go up or down, depending on the movement of the index. As noted,
amortizing payments can be structured with an escalating profile, with lower debt service
in the early years. This is common in commercial property finance, for example, where
there is a "ramp-up" period when rentals are expected to escalate, and can be appropriate
for certain municipal assets. Similarly, interest payments for initial periods can be
deferred by using bond proceeds to pay interest costs in early periods (capitalized
interest).
Bank loans or municipal bonds may be made at either fixed or floating rates of
interest, the latter being payment mechanisms where the interest rate is reset from time to
time. In emerging markets, the variable rate may be the only interest payment structure
available for obligations beyond a short maturity. The relative advantages of fixed versus
floating rate debt are similar to those regarding short term versus long term debt.
Floating rate debt implies continuous uncertainty about the cost of debt, but it can be
appropriate where the matching revenues are expected to vary with changes in interest
rates. However, this is not usually the case for municipalities. Financial flexibility and
access to liquidity are very important considerations for floating rate borrowers. If there is
limited ability to change taxes or rates to respond to rising interest rates, then over-
reliance on variable rate debt is worrisome. The rating company Standard and Poors
generally recommends that the combined short-term debt and variable rate debt not
exceed 20 % of total debt but it depends on the circumstances (degree of flexibility) and
matching of revenues versus debt service.22
Municipal bonds also may pay interest on a variety of "coupon" dates. Although
the semi-annual is the most common, structured loans can have varying coupon profiles
(semiannual, quarterly, even monthly) to suit the cash flows requirements of the borrower
and the capacities of the issuer.
There are several caveats to make in considering the cash flow of municipal
bonds. It is unlikely that the repayment structure of bonds should be regulated aside from
confining borrowing to capital expenditures and having maturities not exceed the useful
economic life. Most municipal bonds in emerging markets have had short maturities and
frequently have had term bond or bullet maturity structures. This, in effect, has meant
that the loans to local governments were for the construction and start-up period. Then,
implicit in the repayment structure, they have required the borrower to "roll over" the
loan into another one at maturity or to come up with alternative means of long-term
financing. Such an approach subjects the issuers and lenders to great uncertainty
22 In the United States, where there is an upward sloping yield curve from short-term to long-terrn tenor, there has
been a reward of between 100 and 200 basis points for using variable rate instead of fixed-rate debt. The structure
allows flexibility to restructure debt, since it typically can be called at the reset date to restructure debt.
27



regarding future debt service requirements and effectively holds borrowers hostage to
future changes that may be forced upon them when they again come to market to renew
the loan.
A final area of policy regarding the structure of instruments is that of the
restrictions that may be placed on interest rates or on the maximum maturity of bonds.
Interest rates may be capped by "usury rates" that set an absolute ceiling on the rates the
issuer will pay. While this was once common practice in the United States, the restriction
has disappeared for all practical purposes over the last twenty years. The effect of
limiting interest rates is to ration, capital away from governments during periods of high
interest rates. Such restrictions continue to be seen, however, as a matter of contract in the
case of variable rate instruments, where a cap is either introduced or a borrower may
purchase a rate cap contract from a commercial bank that will agree to pay the excess
interest for a fee. The other common restriction is on maximum maturity, which is often
specified in conjunction with the expected useful life of the improvement being financed.
Again, these are seldom effective and the market itself provides the limitation on how far
it will extend debt, especially at fixed interest rates.
4.3 Methods of Sale
Municipal securities can be sold to investors in a number of ways, ranging from
the competitive auctioning of bonds to the highest bidder to a placement of them with the
fimal investor, much as a direct loan is made from a bank. In most emerging markets, the
offering is done through a process called negotiation, in which the borrower procures the
services of a financial services firm (such as an investment banking firm) to sell its
bonds.23 The firm either will underwrite the issue, in which it agrees to buy the bonds at
a certain price, or act as an agent and make a "best effort" to sell the bonds for which it
receives a commission on the bonds sold.
The various methods of sale have their advantages and disadvantages. Generally
a competitive sale environment requires there be a very active market with a large
number of issuers offering fairly standardized securities and a large number of investors
interested in owning them. The large volume of activity results in a number of bankers
that follow the market, make bids, and place bonds to investors. It also means there are
other professionals that help with the design of the issues, preparing documents, and
running the auctions.24 The competitive auction, with several underwriters bidding on a
bond issue, as commonly seen in the United States municipal bond, is a rarity in other
markets.
23 For larger bond issues, a combination of firms, called a syndicate, may band together to underwrite and distribute
the issue.
24 Other professionals include financial advisors, legal counsel, auditing firms, and printers to produce the documents.
There may also be banks to handle the investment of proceeds and to oversee the payments under the debt contract
(trustees and paying agents).
28



Where bond markets are less homogenous and sales are irregular, issuers typically
require that an underwriter be hired to help prepare the issue and to seek out possible
investors. This technique is called a negotiated sale in which the underwriter is selected
before the sale of the issue to final investors. The negotiations can be made competitive
by injecting elements of competition among firms into the underwriting selection process
and subsequently by holding underwriters to the projected terms of the issue. To help
achieve competition, the issuer may employ .the services of a financial advisor that is
knowledgeable in the design of transactions and the process of marketing of securities.
The advisor can perform a variety of functions but in the negotiated transaction its
primary duty is to assure the issuer that it is being dealt with fairly by the underwriter and
is holding up its end of the transaction. The advisor usually helps the issuer in the
process of selecting the underwriters. See the box on Krakow Selects An Underwriter.
The underwriting process, as opposed to the use of a best effort marketing
arrangement, has the advantage of guaranteeing the issuer that sufficient funds will be
borrowed. However, the investment banker undertakes the risk of reselling the issue and
will demand more remuneration in the process of underwriting than simply acting as a
placement agent. To make their profit and to cover expenses, the underwriters buy the
bonds at a discount, which means that they buy the issue for less than the value at which
it is reoffered to the final investors.
Selecting an Underwriter by Competitive Negotiation
The gmina (city) of Krakow, Poland proposed a bond issue for 15 million Polish
zlotys in 1996. The city, assisted by a financial advisor, sent an RFP to a large number of
investment banking and commercial banking firms. The solicitation described the project
and needed funds, provided information about the city and asked for proposals from
interested firms. The RFP and selection process contained several elements designed to
make the choice of firms transparent and competitive. The RFP contained a tentative
maturity structure for the issue and asked respondents to price the bonds (provide interest
rates) and to indicate their gross profit, assuming that the bonds had been sold on a given
day. In addition, the respondents were asked to estimate an itemized list of costs and to
indicate which costs would be met from their profits and which would be paid by the
City.
Respondents were also asked to critique the structure and suggest alternatives of their
own. Firms were also asked about their experience and financial capacity. The
combination of factors were used in selecting the finalists, but all-in cost of borrowing
was the most important. All costs, including future interest payments and fees paid by
the city, were made comparable by using an all-in-cost internal rate of return calculation.
After the responses were analyzed by a committee, individual firms were
contacted to clear up any questions (all responses were kept confidential). Of the eight
firms (and syndicates of firms) that responded, the best three were invited to make
presentations and to make their best and final offer. A syndicate made up in part of the
29



local representative bank was selected. The final offer committed the underwriting
syndicate to price the proposed bonds on a par with Polish Treasury bonds of the same
maturity, a highly aggressive bid.
Subsequently, Krakow received an investment grade credit rating from Standard
& Poors and sold bonds (Deutsche mark denominated) in the Euromarket in late 1997,
the first Polish city to do so.
The mechanics of selling of the bonds and the setting of interest rates and other
terms differ among the various domestic securities markets. In some countries with
relatively small inactive markets, the terms of the bond offering may be set well in
advance of the sale date. The bonds may then be sold on a given day with a discount or
premium to make returns competitive with prevailing interest rate conditions. The
technique of fixing terms before the sale date tends to put the underwriter firms at most
risk and issuers pay an interest rate premium in fixing the rates. Another approach is to
commit to have the bonds underwritten at a certain mark-up or in relationship to some
regularly published interest-rate index, usually that on government bonds. Finally, the
terms can be determined by offering the bonds at a proposed structure and then changing
the terms to meet the effective demand from investors in what amounts to an "informal"
auction. The key point is that the terms and their acceptability to the issuer remain open
until the sales contract with the underwriter is signed (the bond purchase agreement).
Tlhe bond instruments or other evidence of ownership are delivered (physically or
electronically, depending on prevailing market practice) and money is exchanged for
them (settlement) from the day of sale or up to two weeks after the sale. Depending on
market conventions and the nature of the security, the borrower (or the underwriter) may
have selected a paying agent and/or a trustee that receives funds from the issuer and
makes the periodic payment of interest and principal. The trustee, if one is used in the
transaction, has the additional duty of overseeing the bond contract between the borrower
and ultimate buyers and is responsible for looking after the interests of the investors.
The important thing to remember is that beyond the procedures set down by
enabling laws at the national level, the specifics of the bond offering are a matter of
contract between the underwriter, the ultimate investor, and the issuer. This puts a
premium on the issuer getting good legal and financial advice independent from that
given by the underwriter and the final investor.
30



5.    Restrictions on the Issuance and Use of Subsovereign Debt
Most countries choose to place restrictions on the use of debt by local
governments. Restrictions on local governments involve substituting national policy for
local flexibility and the regulating effect of the market on municipal borrowing. Among
the restrictions that may be considered from a national perspective are restrictions on the
purpose of borrowing, restrictions on the amount of borrowing or debt outstanding,
restrictions on the use of proceeds, and restrictions on the type of security given or
recourse available to the lender.
A distinction that is commonly subject to regulation is between the use of short-
term debt and long-term debt. The difference between the two can be somewhat arbitrary
(and some times conflictive). Short-term debt is that which is due one year or less from
the date incurred; long-term debt is that which is due more than a year after it is incurred.
5.1   Restrictions on Short-term Debt
Short-term financing can be useful as a part of a local government's normal
operations, such as to pay operating expenses before annual tax revenues are received. It
can also be used in anticipation of nonrecurring revenue, such as from the sale of assets,
receipt of a grant, or issuance of long-term debt. Operating cash needs can be financed
from a municipality's working capital, or with the borrowed funds. The formner source is
usually cheaper and more reliable, but some local governments do not maintain adequate
working capital funds. Furthermore, events can happen that either slow the receipt of
revenues or cause unexpected surges in spending that lead to cash shortages. Ideally,
governments carry reserves of working capital, but such liquid reserves can be a source of
political bickering. Many believe that reserves are a target for political criticism or are a
temptation to politicians that any idle funds are fair game for their spending plans.
Whatever may be the pros and cons in a particular situation, "snowballing" short
term debt as a result of chronic operating deficits has been a leading cause of financial
emergencies when banks and other investors lose confidence in a government's ability to
run surpluses and repay its short-term debt. An outstanding case is that of New York
City in the 1970s, but more recent examples can be found.
As an example, South African municipalities may legally borrow to finance both
routine and unusual short-term needs. However, they are required by regulation to settle
their short-term debts by the end of the current fiscal year by and within twelve months
by the national Constitution. The usual form of borrowing is a bank overdraft, although
other kinds of short-term borrowing are also done. These short-term borrowing
techniques result in unsecured debt. Historically, the banking community in that country
allowed overdrafts as a matter of course. There is now an increase in the use of
overdrafts by municipalities, and in some cases, these overdrafts are not being settled as
required by law at the end of the fiscal year.
31



Recently the City of Johannesburg found itself caught out by having accumulated
a large amount of outstanding short-term debt for purposes of commencing a capital
spending program that was to have been funded by the sale of long-term debt. The
domestic markets closed to the City in late 1997, and it was embarrassed for sufficient
investor interest in rolling over its short-term debt. Ultimately, the Development Bank
for Southern Africa stepped in and made a loan secured on a specific tax source. The
local commercial banks, meanwhile, which had refused to roll over the short-term loan,
became irate that the DBSA had "peeled off' part of the general revenue base and had
done the deal on terms that were only marginally better than they were offering.
A major concern with short-term debt is assurance that recurring expenditures are
met from recurring revenues. Deficits that accumulate over time because of a failure to
match operating revenues with expenditures can reach levels that compromise a local
government's ability to deliver basic services. The objective is to allow reasonable
operational flexibility for local governments, while protecting the public against the
consequences of allowing operating deficit debt to swell to unreasonable levels. There
are major risks in snow-balling short-term debt. First, it can reach unreasonable levels,
requiring an unacceptably high proportion of revenues to be devoted to debt service at the
expense of public services. Second, if it accumulates over time and becomes a
permanent. fixture, the government finds itself at the mercy of the marketplace.
Eventually, when creditors perceive that the floating debt has reached excessive levels,
they can decide to withhold further credit extensions. Then, the pyramid of accumulated
short-term debt can collapse, as happened to New York City in the 1 970s.
To protect the public against unreasonable operating debt levels, short-term
borrowing could be limited in a number of ways, including the following:
- as a percentage of the municipal operating budget,
* to anticipated cash needs for a certain number of months, or
* to revenues or receipts to be received within a period of time.
To further protect against accumulations of debt, short-term borrowing could be
required to be paid off in fill at least once a year, with appropriate safeguards against
immediate re-borrowing, such as an interval when no short-term borrowing would be
permitted. However, it is also likely that either natural or financial emergencies may
make this a difficult area to enforce. Thus, a waiver procedure might be needed to permit
exceptions. Such waivers might be best viewed as part of a general intervention policy as
is discussed below.
5.2   Restrictions on Debt by Purpose
Debt allows local government to acquire or build capital improvements more
quickly than can be done on a pay-as-you-go basis. It allows more equitable payment
schemes, since users can be made to pay for the capital cost of facilities as these are used
32



over time. But, the use of long-term debt has costs and risks. It limits a local
government's budget flexibility in future years. Unwisely used, it can burden citizens
with high tax rates or service tariffs. A rule followed in many countries is that long-term
debt may only be used for investment (capital spending) purposes and not to cover
operating deficits. Even more restrictive is the rule that bonds may only be used for
"self-supporting" revenue generating activities, a requirement in the Philippines. In both
cases, the underlying conviction is that governments should only borrow long-term when
they use proceeds that will contribute some future capacity to repay.
Long-term debt is clearly appropriate for capital investment when the term of the
borrowing is related to the useful life of the capital asset being built or acquired. Should
multiyear debt be allowed for other purposes? Such purposes could include work-outs as
part of a fiscal recovery package, or extraordinary expenses related to the transition and
restructuring underway in a developing country.
At least two alternatives are available with regard to regulating the purpose of
long-term debt. First, a permissive general authorization for long-term subnational
borrowing would allow local government to borrow for any public purpose authorized to
it by law. It would be up to the local jurisdiction to decide what borrowing is wise and
appropriate and for the market to accept that purpose.
Alternatively, jurisdictions can be authorized to borrow only for specific public
purposes. Candidate purposes might include the following:
*  Building or acquiring a capital asset whose anticipated useful life will equal or
exceed the term of the borrowing;
*  Funding revenue-generating projects that are self-supporting,
*  Funding accumulated operating deficits as part of a legal or administrative
restructuring;
*  Funding extraordinary needs, including but not limited to, government
restructuring or retrenchment packages or programs to abate natural or man-
made disasters.
There are advantages to each alternative. Two factors favor a more liberal
authorization policy. First, local government finance is an evolving art, and there should
be room to adopt new forms and techniques of local finance. Second, if national policy
favors decentralization (which is the operating premise of this paper), then local managers
should be allowed decision flexibility. On the other hand, those arguing for limitations
could state, first, that a list of allowable uses provides clarity about what is permnissible,
which may help reassure young capital markets. Also, there may be a perception that the
public needs to be protected from politicians or managers who might otherwise use long-
term debt inappropriately for their own short-term gain.
33



5.3    Restrictions on Amount of Debt
A recent International Monetary Fund publication has argued for a rules-based
control of subnational borrowing, with limits on the debt of individual jurisdictions that
"mimic market discipline."25 Such control could be framed in terms of the maximum
annual debt service to a conservative projection of revenues that would be available to
26
pay debt service.
Debt Limitations
Subnational debt limitations are very common, although the  definitions used can
be arcane and open to confusion. Debt limits are typically of three sorts: (1) a limit on
the amount of indebtedness issued, usually expressed as a ratio of actual or potential
source of revenues, such as taxable property values; (2) a limit on annual debt service
as a percentage of not previously committed annual revenue; and (3) special limits on
short-term indebtedness (matrity of less than one fiscal period). In terms of annual
debt service, a restriction of its not being more than 15 % of current recurring revenues
is the most common. Limits on total debt issuance is expressed as a percentage of
revenues used in order to restrict the amount of total cash revenues attributable to
borrowing. Short term borrowing for cash flow purposes is often required to be paid
off by the end of the fiscal period, but this requirement is often violated.
Other restrictions commonly found relate to restricting long -term borrowing to
capital purposes (physical investments) and prohibitions on borrowing in foreign
currencies. Enforcement of restrictions can be uneven and frequently open to differing
interpretations. For example, in Poland the requirement that Regional Audit Agencies
(RIO) of the voivodships make sure that gmina (cities) have complied with budgeting
requirements and borrowing restrictions led to inconsistent interpretations of the
restrictions. It also led to political bickering between the gmina and vovoidshiops over
the desirability of the projects, although a substantive review by the latter of projects to
be financed was not called for in the borrowing laws.
The argument against such control over the amounts of long-term borrowing is
that it is an unnecessary interference with the market and that, in particular cases, it may
25 Teresa Ter-Minassian, editor, Fiscal Federalism in Theory and Practice, Washington DC, International Monetary
Fund, 1997, p. 171-172.
26 Although many government's debt ceilings are expressed in terms of the debt principal outstanding in relationship
to either a measure of tax base or revenue flows, the best prescriptions are likely to use annual debt service in
relationship to available revenues. Actually crafting such a restriction calls for considerable care in definitions. See
Charles Smith, "Measuring and Forecasting debt Capacity: The State of Oregon Experience," Government Finance
Review (December 1998) pp. 52-54. As Smith points out, the legal limit is usually much higher than the effective
market limit.
34



be desirable for local jurisdictions to temporarily exceed any given ratio as it invests for
the future or spends down accumulated reserves. The argument favoring such control is
that it protects the public from reckless borrowing by officials or elected representatives
who may not be sensitive to the long-term risks. However, there appears little in the way
of scientific bases for deciding on a particular limit.
A limiting ratio of outstanding debt or debt service to available resources for
repayment can probably be set that provides a reasonable degree of protection, and yet is
high enough not to interfere with sound management. An escape clause allowing the debt
limit to be exceeded either under financial emergency legislation or with special
permission from the senior level authorities (or, as in the U.S., by virtue of a local
referendum election) would provide for the exceptional case where a higher ratio makes
sense. Such prescriptions may be appropriate for general fund debt, but do not make
sense when applied to self-financing or enterprise projects. In these cases, if the pledge is
clearly limited to project revenues, this limit need not apply.
5.4   Authorizing Debt
There are several ways in which borrowing might be authorized in a local
jurisdiction. Loans can be approved by action of the local government executive, by
action of the local governing body, or by action of the community at large, such as
through a referendum. Local borrowing can also be approved by state or provincial
authorities, or by national authorities. Each of these mechanisms can be conditioned by a
variety of considerations, including the financial capacity of the borrower to pay debt, the
purpose of the borrowing, the form of borrowing, and consistency with national economic
policy. Most also act as a curb on the local officials' prerogatives.
Borrowing authorized by the executive seems appropriate for relatively small
amounts and short-term borrowing, where the future financial health of the local
jurisdiction is not at risk. However, large amounts or long-term borrowing should be
based on legislative act. Borrowing involves trading off future flexibility in exchange for
investment capital today. For large amounts of debt, the decision about what to borrow
for and how much are policy decisions properly made by a local governing body.
Experience in other countries suggests that without local council or legislative approval
(where elected bodies exist), the probability of future debt repudiation or refusal to enact
necessary tax or tariff changes to meet debt service obligations increases dramatically.
Also, public debate on debt policies and plans help keep the process open and visible.
(See box, Cebu Considers a Deal.)
Borrowing approval can also be made by the community (electorate) at large.
Borrowing authorized by referendum depends on the nature of the local political process.
It may encourage citizen participation in the decision making process and bind the
community for the long haul, notwithstanding changes in the elected council. At the same
time, voter approval for borrowing has disadvantages, since it adds to time and expense,
and can turn financial decisions into political battles.
35



Cebu City Considers a Deal
A worry is that unsophisticated local governments are subject to blandishments
and may lack the knowledge andvprocesses to fend for themselves as borrowers with
expanded opporuities. But that susceptibility has a lot to do with the local political
process and the role of the press. In fact, high-flying financial schemes are likely to catch
attention and, there are few who believe that when it comes to the private sector there will
be somethinig for nothing
TheCity of Ce*bu Philippines provides an eaample of how the process works.
The City and its Mayor have been activelyseeking financing for a ring road which is a
core to itsd       plan. In mid- 1998 a couple of firns, one based in Hong Kong
and another i Austria, proposed that the City enter into $500 million in loans and $75
million in a letter of credit. The pOce  are to be used for a ring road devel t that
is key to the City's development plan. The Mayor of Cebu liked the idea, and the firns,
eager to move things along, paid a couple of councilmen from the Mayor's party to go to
Europe to see, secretly it tumned out, examples of what the firm and its associates had
financed. The councilmen were treated in a first class way and, in order not to insult their
hosts, signed letters that the City would be entering in other agreement in three months.
Subsequently, things began to get sticky. Political opponents began to wonder if
this were a scam. The local reprentative of the investment firms, was arrested on 32
counts of passng bad checks and after posting bail, skipp town for Hong Kong.
Local bank officials indicated that three signatures had been forged on documents
shown in the transaction, evidently guaranteeing $150 million in city funds. The Mayor,
sensing problems, established a committee of knowledgeable citizens to meet with the
proponents; but the financiers refused to meet, saying it would be pointless. By this time
the Cebu scam saga was getting daily press coverage in Cebu and Manila.
The Mayor announced in late November that the fiacing consortium was
embarrassed by the episode and would give the City a grnt instead a loan, or at least put
in a few million Peso in the project for free. But, the next day, the consortia had
evidently changed its mind about any "grants instead of loans" and it was clear that the
"deal" would collapse.
There are two perspectives one may take on the ill-fated deal. One is that big
decisions are beyond the grasp of many local governments and should be kept out of their
reach. But, blandishments by over-energetic project proponents that stand to gain (or by
con artists that wish to defraud) are facts of life at all levels of governments. The other
perspective is that local party politics and news-hungry journalist are ready to shine
bright lights on shady deals. Having local banks and bond dealers around either to
compete openly for the deal or to help with the diligence is not a bad idea, either.
36



National or state borrowing approval could be predicated on several conditions.
These could include the financial capacity of the borrower to pay debt, as measured by
credit analysis, the application of a formula, or a ratio specified by law or regulation.
Other relevant considerations include consistency of local borrowing with national
economic policy (such as the timing of the borrowing), the purpose of the borrowing, and
the form of the borrowing. Such oversight and review might control irresponsible
borrowing at the local level. However, reviews introduce delays, require the development
of oversight capacity at the national or state level, and inject a point for political rather
than economic considerations to enter the decision making process. In general, the
marketplace, aided by appropriate disclosure rules and investor reviews (including those
of rating agencies if they are present) will do a better job of testing the financial capacity
of the borrower to pay debt than will national regulators.
Another question is whether state or national authorities should certify the
procedures used in the borrowing process.27 Such certifications can help build investor
confidence and relieves individual investors of some of the "due diligence" that would
otherwise be required. However, in most emerging countries there may be little if no
capacity at the senior levels of government to undertake this task when it comes to
securities offerings. Imposing a review requirement that cannot be done well slows the
development of the market more than leaving the due diligence inquiry up to the investor.
It also opens a zone for political second-guessing and bickering.
27 In the United Sates, the bond counsel (a legal counsel that specializes in municipal bond transactions) often drafts
the needed resolutions and contracts and provides opinions as to the transaction being in conformance with the
applicable laws and regulations. Some states in the United States also have procedural checks. Texas local
govemment issuers must obtain approval as to procedures (but not as to use of proceeds or other substantive matters
regarding purpose) from the Office of the Attomey General before issuing debt. This office reviews the proposed
bond issue's supporting documentation, certifies validity, and issues an opinion. This opinion is needed for the bonds
to be legally binding, and relieves individual purchasers of the necessity to inquire into the process by which the bonds
were issued. Sometimes softer approaches are used. Oregon requires that localities prepare bond documents following
recommended guidelines. Many states require that prospective sales be reported to a central office and placed on an
official calendar. The state of North Carolina requires a filing and a defacto approval before sale and that requirement
has its exceptions and is largely procedural.
37



6.    Financial Market Structure, Regulation and Operations
Any tool kit examining the options for a subsovereign credit system must look at
the "supply of funds" side of the equation, and where borrowers and lenders meet, that is
the financial marketplace. To what extent does there exist a "market" for subsovereign
obligations and how should would-be borrowers attempt to access it? Perhaps more
relevant in most emerging economies is the question of where subsovereign securities fit
into an overall strategy to develop domestic financial markets. Promotion of private
capital markets has been a primary objective or financial market regulators.
The financial market that we have in mind is understood to be one having some
level of effective competition in rates and terms and that involves the investment of
private capital, although various governmental entities may be suppliers of capital as
well. It is also, for our purposes, thought of as primarily "domestic," that is, where the
borrowers and lenders (or issuers and investors in the case of bonds) are subject to
domestic rules and deal in the local currency. Last, although issuance of equity capital by
governments is possible in some cases, the focus is on credit instruments, that is, financial
instruments that provide no voting rights or ownership but rather represent interest-
bearing obligations that are to be paid in a timely basis.
A key objective of many policy efforts in the past few years has been to create a
municipal bond market for subnational securities in their respective domestic markets. At
the same time this objective has been embraced, the domestic financial markets have
themselves been undergoing great change. Most of the liberalization and subsequent
growth of the domestic securities markets has focused in privatization and the desire to
promote private-sector equity ownership. It is in this context of reformulating the private
and public sectors that subnational borrowers must navigate.
6.1   Financial Market Structure
The focus on the early securities market growth in emerging and transitioning
economies has been on equity markets, with debt private markets coming later and
hesitantly. Most national debt markets are dominated by central government debt and
perhaps that of parastatals. The banking system, either government owned or privately
owned, also has tended to dominate investments in government bond markets along with
government-controlled investors. Corporate debt has traditionally been financed through
the banking system and there are relatively few corporate bond issues in most emerging
securities markets.
The topic of financial market structure and development far exceeds the relative
narrow scope of this paper, but it is vitally important to judging the various linkages that
local governments may forge with the capital markets. Many emerging governments are
at the earliest stages of developing securities markets, and those bond markets that do
exist likely are dominated by the national governments and the commercial banking
system, the latter of which may be very far along in terms of financial maturity.
38



One can gain a general idea of the relative size of the financial markets by
comparing figures that are collected on an international level. For example, the volume
of listed securities of exchanges or transactions on the exchanges to the overall GDP of
the country is a rough indicator of the relative role of the financial markets in the
economy.
More precisely in terms of credit markets, one would look at listed securities in
the debt market (such list including any exchange listings, as well as bonds in the OTC
market) in relationship to GDP. Relative size of the banking sector can be measured by
loans and investments to GDP and the size of securities markets to listed securities and
domestically-held debt to GDP. Similar measures of other financial institutions and
intermediaries provide indices as to the state of development of the domestic financial
markets.
The extent to which the central government is directly involved on the allocation
of credit is not always immediately apparent. Government ownership of banking system
and other financial institutions (retirement funds, insurance companies, etc.) can be very
influential in deciding which borrowers' needs are served and on what terms. Also,
governments in laying out prudential rules and reserve requirements for financial
institutions, for example, can either mandate or build in large incentives for them to
invest in certain classes of obligations. A favorite market support approach has been
either to require that reserves contain government bonds (both sovereign and
subsovereign) or to make the capital adequacy rules such as to favor investment in these
types of securities. Of course, such requirements are not entirely self-serving, since
implementation of monetary policy requires there be levers over bank portfolios. In
theory, open market operations can be carried on in any security. But, there is the
continuing problem of "adverse selection" where securities that have their market directly
manipulated by the monetary authorities are being either supported or subverted for
reasons unrelated to the subnational issuer.
The recent turmoil in the financial markets, especially that involving the Asian
countries, has focused attention on the relationship between the finance industry,
especially banking, and other industries. The degrees of interlocking ownership, control
of boards, and the extent of self-dealing between financial institutions and their non-bank
affiliates have been at issue. Although subsovereign governments have been a relatively
minor player in such concerns in Asia, in South America the relationship between
municipal and provincial governments and the banking system has come under
considerable scrutiny. Financial institutions named banks do not necessarily follow
prudential practices, just as regulators do not necessarily regulate nor are laws on the
books enforced.
39



___Reserve Requirements and Capital Rules
Prudential regulations can have a major impact on the market for various types
of obligations. Many sovereign governments have effectively built in markets for their
securities by requiring that fnancial institutions hold a certain amount of sovereign
direct or guaranteed obligations as part of their reserves. In other cases, banks may be
required to put up government debt as collateral if they wish to hold government
accounts. For example. prior to the availability of deposit insurance for large
denomation accounts, the collatral requirement on public deposits was a powerful
incentive for banks to hold U.S. municial bonds.
One source of useful information egding potential demand (as well as an
overall measure of perceived subsoerign risk) are the weights that banks must use to
calculate their capital adequacy. Although ftese have varied internationally, they are
increasingly coming into conformance with the BIS (Bank of international Settlement)
capital adequacy ratios. The rato refers to the ratio of bank capital to performing
loans (non-perforning loans carry special provisions). The BIS minimum is currently at
8%.
Under the BIS regime, loans to the sovereign government of the same country
as the bank are assigned a 0.0 sectoral risk weight (i.e.. they are assumed to be
domestically risk free) and those of private-sector finns are assigned a 1.0. The BIS
recognizes that the relationship between the central government and subnational
governments vary from cuntry to country and therefore allows the central bank in the
respective countries to assign the appropriate risk weight. Thus, the weightings
providei the central bank's opinion as to the relative risk of loans to the sub-national
govermnental sector in comparison to the sovereign and the private sector.'
In the U.S., the BIS credit factors range from 0.1 for general obligations to 1.0
for private activity (corporate) bonds. In foreign countries, subnational government
obligations that have explicit central govemnment guarantees have BIS ratios of 0.0
(which makes them tantamount to direct sovereign obligations) and those that do not,
have ratios that can range up to 1.0 or even higher. Ratios can be changed to
recogize overall changes in sectoral credit strength. This recently happened in South
Africa, where the ratio was increased from 0.1 to 1.0 for sub-national governmental
securities when the national government announced that it would no longer guarantee
municipal and provincial debt.
Prudential rules regarding other financial institutions such as insurance
companies and pension systems have similar impacts on various types of security. To
28 Higher capital adequacy ratios require banks to put up more capital per amount of loan, thereby raising the cost of
capital and increasing the interest rates demanded of the sector relative to the sovereign.
40



the degree that subnational securities have been lumped together with sovereign
| securities, they have often benefited by favorable treatment.
6.2   Market Development and Regulation
Since the 1980s, there has been a move worldwide toward decreasing direct
regulation of the financial markets and opening the markets up to greater domestic and
international competition. This has involved all aspects of the financial markets, from the
privatization of banking systems to feverishly creating stock exchanges to support the
privatization of formerly state-owned enterprises in many parts of the world. It has
brought with it a greater number of domestic firms in the securities business and more
openness to foreign firms doing business in the various domestic markets. The entrance
of foreign firms has been important also because they bring experience in financing
subsovereign obligations.
Not only have the financial markets become more open and larger, regulation of
them has also seen a move toward greater self-regulation by industry participants. The
shift has really been from regulation by fiat, with direct involvement by the government
in the investment decisions, to a greater reliance on market forces. These developments
present something of an enigmna. Less regulation by ministerial fiat and less official
involvement in individual transactions has given way to more in the way of general rule
writing and setting up rules of the road for functioning markets. Thus, the presence of
more firms and a greater variety and number of instruments in the market mean that there
is need for more regulation. But, the nature of regulation has changed. It is more in the
form of general rules of fair dealing, capital adequacy, and more by general rules rather
than by specific fiat. This changed regulatory mode depends on the operation of self-
regulatory bodies rather than by the central government agencies. This is not without its
costs and risks.
Where do subnational governmental borrowers fit into the emerging securities
market regulatory scheme? The question is just beginning to be asked in most places
since subnational government securities are still a rarity. Emerging markets have also
seen a mixed bag of motivations in the creation of regulatory schemes, including
proactive requirements to encourage sound business operations. For example, there is the
requirement in Chile that public issues of securities be rated by a licensed rating agency.
A similar requirement was being implemented by BAPEPAM, the securities regulatory
body in Indonesia, which was instrumental in creating a national rating agency, Perfindo.
The fundamental concept of regulation is to define the financial system and
establish rules by which it will operate. That is easier said than done. For example, what
are securities and what constitutes transactions in them? Different countries have
different legal traditions that can influence the nature of a market's operation, as is
discussed in the accompanying box on the Titulor Valor.
41



Another issue has to do with the regulation of the banking system and other
institutions that provide financial services. Here again countries enter with widely
varying traditions. There is the deep seated debate between advocates of the "universal"
banking systems and those that advocate the strict separation of the banking system and
the securities markets (such as found in the U.S.) Most emerging and transitioning
economies come out of a bank-oriented financial system, often with government-owned
or favored "universal" banks that have seen virtually every phase of domestic financial
commerce as fair game. As the financial markets broaden and mature, there is the need to
define the regulatory boundaries between banks, securities markets, and other financial
institutions. For example, in addition to the prudential regulation of traditional financial
institutions such as banks, insurance companies, and pensions, there are new entities to
regulate such as mutual funds, clearing and settlement operations, securities depositories,
and the use of new instruments such as derivatives and asset backed securities.
What is a Security?
Establishing the definition of security is important both from the legal point of
view as to what the investor can look to in support of the obligation and, more
mechanically, from the view of securities regulation in defining just what is the nature of
the transaction and the instrument involved. Efforts to regulate securities and to
harmonize laws among countries have been hampered by different concepts as to what is
a security.
For example, in the Spanish-speaking world (and, similarly, in the civil systems
of Eastern and Central Europe) the concept of security has differed from that which
evolved under English Common law and subsequently incorporated into securities
regulation. A security in Spanish-speaking countries is embodied in the concept of a
tftulo valor, which encompassed only a limited number of specific physical documents
that have the right of ownership embodied in the document itself Thus, the only
evidence of ownership for the security investor is the existence and possession of the
document itself. (The corollary to this is the "bearer" bond which is physical evidence of
indebtedness, but which nonetheless could be registered and immobilized.) The titulo
valor instrument is like money, since it can be transferred physically without re-
registration or even endorsement and is payable upon presentation.
As in the case of the bearer bond (the issuance of which became illegal for most
transactions after the early 1980s in the United States) the titulo valor proves to be
woefully inadequate as a concept for evidencing ownership given the nature of modern
transactions. Not only does it pose physical safekeeping and transfer problems, it does
not fit the needs of new financial instrument constructs.' New instruments necessarily
29 A major advantage (or disadvantage, depending on one's point of view) of the bearer form is that it presents major
problems for tax collecting authorities. In many countries, as a result, a presumptive tax is withheld at the source of
instrument payment if the authorities are aware of the debt (another reporting problem).
42



rely on book-entry and dematerialization, such as variable rate securities, derivatives, and
investment contacts. New definitions of security in Latin America now being enacted
into law rely on the economic basis of what constitutes the security, rather than the strict
definitions of what physical instruments qualify as itulor valor.
In emerging markets, financial markets regulation has a variety of roles to play
and the end results may not always be in harmony. Among the competing objective are:
* Market development: Regulators are first of all providing incentives to
market development, especially as part of the effort to convert formerly
government owned institutions into privately owned ones.
*  Market integrity: This requires prudential measures to minimize systemic
risk and protect the solvency of individual firms. Regulators want to foster
lively, creative markets yet protect the integrity of the payment system and
avoid excessive risk taking.
- Fairness: Protection of investor and prevention of monopoly power.
Regulators are keen to prevent fraud and manipulation. Asymmetry of
disclosure information (the issuers control it, the investors need it).
-  Efficiency: Markets are to be allocators of capital resources. The theoretical
benefits of markets as allocators of capital rests on competition among players
and fair price discovery mechanisms. But the participants must be limited to
those that have adequate capital and experience and pass standards of
behavior.
In most emerging countries, the majority of local governments will approach the
markets as largely untested and typically small borrowers. On the other hand, where they
have adequate revenue bases, they can often be viewed as potentially strong credits.
Even where banks have dominated the direct lending, the development of a securities
market benefits banks. It provides the banks themselves with more liquidity as investors,
even while promoting more competition with and among them by providing an
alternative source of funds to direct bank loans. And, a developing credit system allows
them other ways to earn fee incomes, such as acting as trustee.
6.3   The Securities Marketplace
Domestic securities markets are in various stages of development and reflect
differing philosophies regarding regulation, intensities of competition and technological
development. For example, how much trading occurs on the formal markets (exchanges)
versus Over-The-Counter (OTC) transactions. (This is a "telephone" market where there
are dealer to dealer trades in securities that are not listed on the exchange or that can be
43



traded off the exchange as well as on it.) How much of the off-market trading is reported
and how are such trades cleared (that is, the ownership of securities exchanged against
payments)? How integrated are the markets: is there a single market or are there
segmented markets by types of instruments?
There has been a tendency for regulators in emerging markets to force all
securities (equity and private debt) to trade on the exchanges in order to develop the
exchange. But formal exchange listing requirements, which typically are copied from
those in developed countries, and related fees can be burdensome. That burden can be
unfair and costly to new companies and to those that are small. One answer has been to
create a separate bracket for smaller, higher risk companies. (Japan, for example, has a
corner of its exchange set aside for small, higher-risk companies). Another approach is to
allow the development of an OTC dealer-to-dealer market or the restriction of certain
classes of offerings to sophisticated institutions and individuals. This approach provides
trading liquidity to otherwise less liquid shares without exposing the general public to
undue risk. New credits can season and then graduate to an exchange listing at a later
time.
However, rules that permit the creation of a number of separate markets can be
uneconomic and lead to an undesirable diffusion of resources. The most recent moves
have been toward fewer organized exchanges and towards a screen-based, fully reported
trading as opposed to the open-cry, single-place market.30 The more integrated and
transparent a market's operation, the better defined the market's participants and scope,
the more likely that market competition, on the basis of price and quality of service, will
discipline the market behavior without direct regulatory involvement.3' The more these
conditions are met, the more securities regulation can rely on self-regulation through self-
regulatory bodies as opposed to direct regulation for setting the detailed market rules and
enforcement.
30 A recent article on the future of the exchanges indicates that the number is likely to decline rapidly and a few large
international exchanges will dominate activity. See "Survey: Financial Centers" The Economist (May 9th -I 5th 1998)
pp. 54 et.seq. The exchanges may be fewer, but their physical location may be less relevant as trading can occur
wherever a computer can be plugged in.
31 A major factor in enlarging market competition is permiting the entry of foreign firms into the market.
44



7.     Disclosure and Market Regulation
Complete and timely disclosure of information about issuers is a necessary, if not
always sufficient, condition for the effective operation of any securities market. That
includes for municipal securities where the subsovereign borrowers themselves will
provide the ultimate security. Information--consistent, complete, timely and comparable-
-is at the heart of judging the risks and rewards of investments. While information will
not always answer all the questions (and bad information can give the wrong answers), an
absence of information makes even knowing the right questions to ask difficult.
Emerging and transitioning markets face particularly difficult problems with
disclosure. Many emerging economy countries are undergoing dramatic changes in their
fiscal structure, as are the financial markets structure and regulation themselves rapidly
changing. As we have noted, the vague basis for local security or the former direct
guarantee by the sovereign is being replaced by more specific pledges of assets and
revenues. Some countries, such as South Africa, have been and will continue to be highly
reliant for revenues on commercial public utility operations (water and electric) and other
emerging countries, for a variety of reasons, may wish to restrict long-term debt to self-
supporting commercial operations.
The ability of these governmental operations to generate resources to support
themselves, or to generate surplus for general revenue purposes, depends on efficient
technical and managerial operations. Even where there is primary reliance on local taxes
or on intergovernmental transfer payments, information on trends in these compared to
local expenses becomes vital to determining relative credit quality.32
Disclosures to the securities market are originated by the underlying borrowers
themselves, the local governments. The borrowers may be assisted (or even superseded)
by the central government or provincial government authorities in accumulating
information, but the idea is that the party financially responsible for timely and full
payment of debt service is the party responsible to make disclosures. A closely related
concept is that the party that is in control of decisions to honor obligations and that has
control over the relevant information is the one responsible for information.33 Except in
the case where the issuer is only a "conduit" borrowing on behalf of another entity that is
responsible for the debt, the issuer is the controlling party and obligated to report the
information.
32 Aside from the immediate needs of the bond market, it would appear that effective democratic government will
depend on reporting on a consistent format of the operating results of governments.
33 For example, a guaranty by a third party (such as the national government) has been seen as a reason to require less
disclosure on the part of the actual borrower. That concept, however, has been rejected in U.S. practice where a
guarantee (or insurance) does not obviate the need for full disclosure by the borrower. In South Africa (as elsewhere)
the custom evidently has been to lessen or relax requirements where the national government is the guarantor. As
noted above, the change is the fiscal climate will supersede that doctrine.
45



7.1    Disclosure Requirements and How They are Met
Disclosure can be required by central governmental fiat, by securities market
regulation, or as a byproduct of the operation of the market through contract and market
practice and convention. Market-dictated disclosure is realized through meeting investor
expectations, habitual procedure, and, most importantly, fear of antifraud provisions in
the securities law (or similar legal obligations related to fair-dealing and fraud).34
In the securities markets, disclosure is aimed at the investor for the purpose of
informing the investor so that it can make informed investment decisions. However, an
often overlooked but very practical byproduct of securities disclosure is that the
performance, condition, and prospects of borrowers are reported. These economic and
financial factors are of material interest to those other than investors in the market. Also,
the concept of disclosure reaches beyond investor "protection" (avoidance of fraudulent
behavior) to supporting the rationale allocation of resources and being able to evaluate
cost risk, whatever its level may be.35
Generally, formal disclosure requirements are met when the issuer sends
published reports to the marketplace. The timing and scope of reporting information are
important and technology is changing the process of publishing, as may be seen in the
box on the Internet. There often are recipients of the informnation who in turn "translate"
or re-communicate the disclosure and their opinions of it to existing and potential
investors. The most important of these are the rating agencies, and the agencies, as will be
discussed below, are often seen as a surrogate for disclosure.
The policy objective of securities market development argues that investor
protection needs to be counter-balanced with the need to reduce burdens on certain
classes of borrowers to accommodate their access to the securities market. Lessened
standards often are applied to smaller issuers or those classes of securities that are
believed to represent less risk.36
As noted, the content of disclosure can be determined in two ways: (1) by a
regulator's detailed listing of contents of documents and a sequence of presentation at
various times, or (2) by a "flexible" standard that depends in its application on what the
issuer and its advisors and agents perceive as information that investors should find
useful in reaching an investment decision. In practice there is usually a combination of
the two approaches, with a listing of generic types of information, leaving the particulars
34 The nature of the securities laws and effectiveness of anti-fraud provisions are critical to effectuating good
disclosure through a market-based mechanism.
35 This is not just an academic distinction but goes to the heart of market regulation. If the primary purpose is avoid
fraud and investor loss, then the emphasis should be on screening out high-risk securities that regulators feel might
cause loss to the investors. This substitutes a bureaucratic decision for that of the market place. The other approach,
and the one stressed in the U.S. philosophy, is to require full disclosure, and then to let the market decide on the
appropriate rate of return that is required to offset the level of risk, no matter what its magnitude.
36 Traditionally, government securities have belonged to this lower disclosure requirement, although that tradition has
been eroded in the United States and elsewhere and the exceptions are less likely.
46



to those responsible for the deal. Since the scope and detail of meaningful disclosure can
vary markedly both among and within markets, issuers and debt issues, the modem trend
is to rely on market forces and self-regulatory bodies to specify the details of content.
Disclosure documents can also be available from a central source (a depository) using
information received on a recurring or event-driven basis.37
Disclosure over the Intermet
Electronic transmission of information over the Internet has already begun to
change the processes of bond sale and disclosing infonnation in the securities markets.
Although the capability of electronic transmission of data has been known for few years,
it was not until the use of Irnemet became widespread ftat issuers were willing to move
their bond sales on to the Intemet, taking bids real time over the net. An early
experimenter with this procedure was the City of Pittsburgh which held its first Intemet
competitive bond auction in early 1997. That year, several large municipal bond issuers
permitted bidders for their bonds to either file bids by sealed envelope as is the
conventional practice or over the Internet just prior to the close of auction.
In 1998, municipal issuers began to publish preliminary official statements over
the Internet. Again the City of Pittsburgh led the way. Investors that wish hard copy
have the right to contact the City for a printed copy of the official statement. Normally
the City would print 750 copies of the official statement at a cost of $15,000. However,
only 4 requests for hard copy were received. In addition, many issuers are posting their
budgets and financial statements on web sites.
The economies of posting bond disclosure over the web are considerable to both
bond sales techniques and disclosure information. The access to a large number of
investors and underwriters at low cost will promote improved disclosure from a
timeliness and cost standpoint. Just as many of the emerging market exchanges are leap-
frogging the structure of their forebears in the developed countries, new information
technologies will swiftly change the flow of infonnation in the markets.
7.2 Financial Disclosures
Fundamental to continuing disclosure is the timely production of financial
statements that follow consistent accounting standards and that are readily available to
investors on a timely basis. Uniform accounting standards for local governments are
critical to disclosure. In many countries, accounting systems are in transition and under
37 In the U.S. that role is played in the municipal market by a limited number of officially sanctioned (but privately
owned) repositories as well as a central repository operated by an SRO (the Municipal Securities Rulemaking Board,
MSRB).
47



review for improving their timeliness, conceptual consistency and transparency. 38
International bodies are also active in trying to achieve cross-country comparability.
Strong accounting practices are central to improved financial management. And in that
regards, it should be noted that where the adoption of accounting standards required in
order for any borrower to sell bonds or undertake a loan (perhaps with the exception in
certain de minimus borrowers), then the adoption of the standards would be accelerated.39
Accounting standards and their applications vary greatly among countries and
between the private sector and public sector. Most governments come from an
orientation of controlling expenditures and revenues, stressing the legality of their actions
and reporting their conformance with adopted legislation. This has frequently led to the
use of cash accounting techniques in the recognition of receipts and outlays and
frequently obscured the economic purpose or life of the expenditures. On the other hand,
much of credit analysis does indeed focus on cash flows and particularly those that relate
to the availability of cash to pay debt service in full and on time.40
Another common distinction is where the government borrower is involved in an
enterprise activity. In those cases, there may be use of accrual accounting techniques that
conform with those used in the private sector. While the use of accrual accounting has
sound economic rationale for determining the worth and period income performance of
an activity, credit analysis will typically require conversion to a cash basis to assure that
the timing and magnitude of cash is available to meet debt service requirements.
Tlhe biggest concerns with cash accounting techniques are with its focus on short -
term assets and liabilities and the ability of the obligor to change results by accelerating
or delaying the timing of payments and receipts.
The frequency and independence of audits are important issues. Most
governments rely on audits performed by governmental auditors from senior levels of
38 One team of investigators reviewing the Latin American markets stress the problem of financial information: "The
first problem is the quality of municipal or subnational management and accounting, which is often poor and
incomplete." M. Freire, M. Huertas, B. Darche, Sub-national Access to the Capital Markets: The Latin American
Experience First World Bank Conference on Capital Markets Development at the Subnational Level (Santander,
Spain, October, 1998)
39 International Federation of Accounting (IFAC) Guidelinefor Governmental Financial Reporting. The IFAC is
attempting to develop widespread adoption of generally accepted accounting standards.
40 In credit analysis, it is customary for analysts to restate accounting reports to a cash basis to examine the availability
of cash to meet debt service payments. Revenue bond contract indentures are expressed in terms of minimums of
available current revenues after meeting expenses (cash outlays) in relationship to debt service needs.
41 Asset valuation techniques differ among countries. Those that use a historical basis can greatly understate the
replacement value of plant and equipment in periods of high inflation. For example, water utilities with much of their
investment in underground piping and reservoirs may have major assets that have expected useful lives of from 40 to
100 years. Utilities that use current market values for assets will appear to be much less leveraged in terms of debt as a
proportion of total assets, than those that do not. However, their current depreciation charges will likely be higher,
which makes them appear less profitable.
48



government. Such audits often check for compliance with various program requirements
as opposed to reflecting financial condition or assigning costs to activities. In some
countries, governmental financial records are not publicly available and bank secrecy
laws are an impediment to reporting various portions of the financial statement.
Unavailability of financial data for these or other reasons is a disclosure in and of itself,
and a warning flag that the one's ability to assess aveas financial risk is nil and that
political and legal risks are particularly important.
In addition to financial statements, and depending on the nature and
characteristics of the project being financed and the nature of the security pledged, added
information having to do with operations and characteristics of the service provided and
market served may be required in order to obtain appropriate disclosure.42 For example,
investors in limited obligation or enterprise security that looked to cash available after
operating expenditures to repay debt want to know about the operating characteristics of
the enterprise and the market it serves. They need this information in order to judge how
efficiently it was being operated and if there were any concerns about future strength of
demand, supplies, labor relations, environmental matters, lawsuits and the like. The list
of potential items worthy of disclosure can be long and the particulars will be dictated by
the nature of the operation and the security pledged.43 Thus, an important initial
disclosure will be the intention or contractual commitment of the issuer to provide
information on a recurring basis in the future.
An important caveat, and one that is adjunct to this subject, is where disclosure
requirements are made stringent and the regulators of the security markets then choose to
promote reliance on private sector advisory and information services to examine
disclosures and make informed judgments. An aspect of disclosure requirements is
recognition that not every investor is to read every document and understand every
nuance of every deal. Rather, a limited number of qualified professionals are doing the
examining and forming opinions for which they are paid by investors. These opinions are
published and become a "baseline" of assessment and themselves are part of the
disclosure process. The leading example of this role being played is the rating agencies
that post ratings on issuers and issues and continue to keep them under surveillance while
the debt is outstanding.' We now turn to the subject in the following section.
42 The word "appropriate" is used because once beyond a simple general government balance sheet pledge (and likely
even in that case), the information needed to assess risk will be situational to a particular local government. For
example, a government that relies heavily on utility operations such as electric and water systems, will find its ability
to pay debt heavily influenced by the operations of those utilities. If the cost of raw materials or labor costs are rising
rapidly and/or users are not paying their bills, then cash will be short and may lead to inability to make timely debt
service payments.
43 The listing of items to consider are found in various trade and professional publications. A good starting point for
generic items is continued in the GFOA Disclosure Guidelines for State and local securities (Chicago, 1991).
44 But, nothing is free. Regulators may not be doing the substantive reviews and forming opinions on the adequacy of
disclosure itself. They need to institute meaningful safeguards that those who do (such as financial advisors, rating
agencies, and other information providers) are professionally qualified, are operating in an ethical fashion and not
manipulating the market, and are free of conflicts.
49



Accounting for Accounting Differences
Accounting and financial record keeping differences can be an impediment to
analyzing the perfomiance and condition of governments and their enterprises. Because
most subnational governments do not pay taxes and do not have their securities listed on
the stock exchanges, the pressure to have prompt reporting and uniform accounting has
been lacking.
Disclosure of information is meant to sot analysis of the risk and reward
relationship. Appraisig :"economic" risk, that is, that the borrower will be able to pay
interest and principal as promised depends on recurring knowledge of its financial
performance (operating statement) and condition (balance sheet). Since most problems
involving "willingness to pay" are occasioned by fiscal stess, strong financial reporting
practices supports assessment of this risk determination, as well.
From a disclosure standpoint, the immediate objectives are to get financial data on a
comparable basis; to measure the availability of dependable, recuring revenue streams to
make debt service payments; and to measure liquid reserves available to continue meeting
debt service requirements in case the recurimng revenues are interrupted. With proper
reporting, other items, such as the strength and stability of the underling economy, other
indebtedness and the mix and costs of various inputs used by the borrower, are also
disclosed by or can be calculated from the financial statements and their footnotes.
50



8.    Credit Analysis and Credit Ratings
Credit analysis is an activity found on the "buy" side of the market. It is a process
where investors and their advisors examine available information regarding issuers and
their obligations and seek to make judgments regarding the relative rewards from a
particular investments versus the risks those investments entail. Credit risk is typically
taken to mean the economic, legal and political risk inherent in a particular obligation, or
more succinctly, the default risk it embodies. That is distinct from the concept of market
risk or interest rate risk which usually pertains to how the entire debt market (interest
rates and exchange rates, in the case of foreign currency denominated debt) will perform.
The preceding section on disclosure discussed the types of information that analysts
require of the issuers themselves in judging credit risk; but, information used in credit
analysis can be garnered from a variety of sources, such as a governmental statistical
agency or even the local newspapers. The important concept is that there are multiple
and objective sources of information which is timely and accurate.
Credit analysis demands resources and analytical skills that many investors,
especially smaller institutions and individual investors, do not have enough of to justify a
commitment to understanding a particular variety of credit. Thus, they prefer to rely on
the opinions of experts. The role of the expert opinion is best reflected in the function of
the commercial credit rating companies, or rating agencies as they are often called. An
independent, objective system of credit ratings of high quality is seen as an essential
component to the development of a vibrant, private sector capital market. It is especially
important in the case of securities market investments where the investors are both
numerous and "passive" and must rely on provision of information from the issuers and
others in arriving at investment decisions. If the ratings are respected and used, the
rating companies have the clout to demand disclosures and the skills to interpret the data.
To the degree they are successful in obtaining data and their use of it reflects legitimate
risk indices, then the entire market is aided by the categorization of debt and the
monitoring of performance.
8.1 Subsovereign ratings
The appeal of credit ratings is clear: They provide a third-party opinion by
experts that inform investors that themselves do not have the skill or resources to carry on
their own investigations about the relative creditworthiness of competing investment
opportunities. This is particularly so if investors have a diverse portfolio of securities,
each of which represents only a small part of the holdings.
The concept of creditworthiness is important in this context. It has to do with the
comparative risk of there being "payments difficulties" experienced by the borrower. The
rating agencies uniformly do not rate the comparative market values of securities or
general market risks per se. Each rating agency has its own formulation as to how
various factors are to be weighted in arriving at an opine, but they typically look at pretty
much the same factors when it comes to subsovereign credits, as is discussed below.
51



Aside from the United States, ratings of subsovereign governments is very much
in its infancy. An increasing number of developed countries now have their subdivisions
rated, especially those that have entered in the international bond market. For emerging
and transitioning countries, the number of bond ratings by recognized international rating
agencies, while growing, is still sparse. Nonetheless, the rating agencies have been
staking out the subsovereign government area and many observers believe that progress
in the development of subsovereign securities markets will depend on establishing
ratings.
8.2 Subsoverign Rating Factors
Each of the rating agencies has its own recipe for figuring the alphabet rating
categories they assign to respective credits. Ratings range from AAA for the highest
category which is usually only conferred on sovereign credits down to C or D categories
which are for those bonds in default, with varying hopes for ultimate or partial
repayment. While the major agencies have different ways of weighting the factors, they
do agree on the major analytical ingredients they consider in judging the creditworthiness
of subsovereign credits. These can be summarized as follows, with some indications as
to how various factors help or hurt a credit rating:
* Sovereign Rating Ceiling:
The rating of the national government usually sets the top limit on the rating that a
subsovereign unit can enjoy. National govermments set monetary and fiscal policy and
usually have first claim on foreign exchange and can change the rules of the game for its
junior units of government. Exceptions to this rule can be found if the debt is secured by
offshore assets or revenue streams.
D Economy:
Fiscal health is usually closely linked to the health of the local economy and the
diversification in activity (which often comes with size) helps balance the economy's
performance. Demographics are important. A high dependency population (the very
young and very old are negatives) and too rapid growth in population are negatives.
Higher-income and more educated population is a plus, as is an acceptable distribution
and rate of growth in income.
* Structure and Management:
An assignment of functional spending responsibilities consistent with revenue
resources is a plus. Intergovernmental transfers are looked at for their size and
predictability. The willingness and ability of the national government to detect and stem
financial emergencies is a positive. The rigor and timeliness of budgetary and financial
laws are examined and can be either a plus or negative depending on the flexibility they
provide localities. Past perfornance in achieving budgetary balance are important.
Timeliness and comprehensive of financial reporting and following consistent standards
are a plus.
52



* Fiscal Performance:
Revenue composition and trends are considered with ability to set rates at the
local level seen as a plus. Tax burdens should be acceptable in comparison to
neighboring regions. Effective use of charges and fees are viewed favorably, large
transfers of general funds to local enterprises are not. Composition and trends in
expenditure. Capital spending and maintenance spending are a plus; a large wage bill is a
negative. The ability to budget and to accurately realize budgets are a plus. Positive
balances (surpluses) in current operating budget is a strong positive. Capital budget
planning and paying for large amounts with current revenues a plus.
* Financial position:
Liquid assets and marketable real assets are favorable factors, as are healthy
reserves in relationship to annual expenditures. Outstanding debt is considered. Short-
term debt is a concern if not periodically retired. Long-term debt and contingent debt
(where there are guarantees to others) is generally a negative unless used in support of
productive (self-supporting) activities. Short maturity debt with principal due at the term
(bullet maturity) is a negative because of continuing pressure to refinance and potential
burden on current revenues. Overlapping debt of other governments that relies on same
economic base is considered.
* Legal framework:
The lack of clear laws, legal precedent or effective judicial system are major
impediments, especially where there are restricted revenue or enterprise-based pledges. A
history of repudiations or insolvencies is a large negative. Approval of borrowings by
senior units and other restrictions on local borrowing may be a positive if efficient and
nonpolitical, but can be a negative if complex, difficult and political.
* Accounting and financial reporting:
The basis and quality of financial records is examined, and prompt, consistent
reports are a plus. Timely and independent audits are a positive. Cash flow information
or cash basis accounting that provide reliable information on cash available to pay debt
service is a positive. Evaluation of liquid assets and accounts receivable can be issues in
that required investments in government bonds can be risky and accounts may be in
arrears.
Opinions on credit quality are not static and the relative importance of factors can
change over time. National policies having to do with items other than local debt per se
can change the mix and weighting of credit factors. Laws governing purchasing policies,
public employee retirement benefits or wages, or the reassignment of functions and
revenue sources can all shift the focus of analysts.
8.3   Role of the rating agencies
53



The role of credit ratings is not without controversy. On the one hand, emerging
markets have a shortage of trained analytical staff and the existence of the rating agency
provides for a pooling of skills that can opine on the quality of a credit on behalf of all
investors, using a standard methodology (at least standard to each agency). On the other
hand, the concentration of opinion in a few hands using methods that are proprietary and
not fully disclosed can lead to a dangerous dependence on a handful of "experts " that
influence the markets without an effective check. Another urge, as noted, is to have a
regulatory requirement that bonds must have a rating before they can be listed in the
exchanges or sold to the investing public. The international rating agencies themselves
have considerable difficulty with the regulatory rating indicating that it leads to the
"shopping for ratings" to obtain the highest rating or an acceptable one at the lowest cost.
Requirements for ratings can also lead to the creation of national agencies that are not
technically competent and that can be politically influenced. The major agencies
themselves prefer to have the market for their opinions free and for the investors to be the
final arbiters as to which agencies' opinions are worthwhile.
The development of credit ratings in emerging markets have followed two tracks
that have tended to overlap. The first is for various market participants in the emerging
markets to create a domestic rating agency de novo, sometimes in alliance with an
established international rating agency. The focus of these homegrown agencies has been
on domestic credits and to meet regulatory requirements. Generally, their opinions have
carried little weight internationally. The second track, which appears to be the prevailing
path, has been for the major international rating agencies to open offices or to acquire
local firns that fashion ratings.
8.5   Expanding the market for ratings
The international rating agencies have been establishing beachheads in
subnational markets, both to cover the changing circumstances of subnational borrowers
and in anticipation of a future new issue market. This process is illustrated in South
Africa. CA Ratings (now affiliated with Standard and Poor's), Fitch-IBCA, and Duff&
Phelps are actively promoting their products, even though the South African local
government bond market is now moribund. For the present, the primary market for rating
services appears to be tracking the fortunes of approximately five billion Rand in South
African municipal bonds that no longer trade. Whatever may be the size and cloudy
prospects of the new issue municipal bond market in that country, the agencies continue
to show substantial commitment to following municipal debt.
A particular role for the agencies in South Africa is in monitoring outstanding
debt for banks, insurance companies and other institutional investors that neither have
any analytical capacity of their own, nor much desire to invest in any. What happened
was that prior to 1994 South African municipal bonds carried an implicit sovereign
guaranty. Once that was revoked by the new regime, investors found themselves
suddenly having to distinguish among municipal credits that had been for all purposes
homogenous. Furthennore, the transfornation into a new government structure presented
54



new elements of risk. Rating agencies supply the analysis by pooling the credit research
for their subscribers, few of which want to devote the resources needed to follow
individual credits.
55



Emerging Market Ratings and 0BondInsurance
International credit ratings are a relatively recent thing and started first with the
primarly Western European countries and corporations that were active in the
Euromarket in the 1980s. There were very few subsovereign credit to rate given the
dominance of bank lending to the sector and te propensity of sovereigns to guaratee
debt of theiriuniors.         ;                         o s    X         grt
Subsequently, the rating agencies inched into the emerging markets by way of
first raig the sovereigns borrowing in hard currencies and then those public or private
operations that gave promise of providing hard currency to pay back international
bondholders. The next instrument to come on the scene was the asset backed security (the
ABS). These are issues secured by pols of underlying loans, starting off with car loans,
credit card accounts, and mortgages (NOS or mortgage backed securities). The higher
end of these markets (representing United States and Westem European issuers) has
became flooded and the margins are very thin. Attention next turned to the emerging
markets, especially with the advent of the Brady bonds.
T'he assets that back the ABS configuration are tYpically dollar-denominated
securities consisting of export receivable, credit cards, and telephone receivable. Using
the ABS approach, the borrowers have been able to borrow at much lower rates than in
the domestic markets. However, to gain access, these entities have had to gain ratings
and in order to get ratings have typically had to obtain credit enhancements. That in turn
stimulated the growth of bond insurance.
The secret to this has been to structure the debt through an offshore origination
and securing the debt by receivable through a trust. The futue receivables are held by an
offshore trust and receivable obligors are required to make payment to that trust.
Payments never enter the country of the issuer. This avoids problems of convertibility
and should mitigate most if not all of the sovereign risk. These obligations are therefore
not constrained by the sovereign rating of the borrower's country.
More insurance companies are being formed to handle nontraditional, including
emerging market, business. These insurers will handle non-investment grade paper and
no longer price under the assumption of zero loss.45 They require higher reserves and
may have less than the highest bond rating. Insurers make money where the perception
of risk exceeds the actual risk and where by close monitoring and direct involvement they
can alter the actual risk. Risk perceptions may be institutionalized in various prudential
restrictions placed on lending institutions and investors. These perceptions and
restrictions cause credit spreads among classes of debt that the insurer, by superior access
45 Investment grade paper is understood to be that which is rated in the fourth tier of ratings (BBB or Baa) or higher.
By convention this is paper the value of which can be carried on the books at purchase price by financial institutions.
With the emphasis on marking all securities "to market" (current prices) that convention has slipped.
56



to information, deeper analysis, and ability to diversify risk, can narrow by "renting out"
the use of its credit rating charge premiums and enjoy a return on capital.
But there continues to be some pockets of interest in new debt issues. Fitch-IBCA
provides ratings for the privately financed South African bond bank, INCA, on its
municipal investments and holdings. With the rapid change in the South African local
government structure, there is a desire by existing investors to stay current, and there is
the continuing prospect that once the governmental structure settles down there will be a
flood of new issues. Borrowers, anxious to position themselves are keeping ratings up to
date. At least two of the rating agencies have developed rating contracts that provide for
continuing rating of the borrower. Each agency has compiled data for a much larger
number of municipalities than they have been called upon to rate, and each makes an
effort to recast data reported to them into standard formats. At the same time, it appears
that not all segments of the investor community are either familiar with, or convinced by,
rating resources and opinions. Some investors express fundamental reservations about the
value of credit ratings in general. But once the ratings are published, all investors must
be aware of them and calculate the effects into their pricing decisions. It seldom pays to
bet against the rating of a respected agency.
Rating agencies suffer from inherent difficulties that go with being both
fmancially viable and having a powerful effect on market behavior. First, their
methodologies are necessarily proprietary. If everyone could emulate the rating through
the use of a known formula, then why pay for a rating? Second, important factors used in
ratings can be largely subjective. What is the risk of political instability including either
visiting bad times on the nation or even repudiating debt? (Even when the "good guys" as
measured by other standards win control in a country, bondholders and creditors can lose
if the terms of outstanding debt are unilaterally changed.) Third, in publishing opinions
the rating organizations assume that certain conditions and relationships will prevail. In a
rapidly changing world, such assumptions may be erroneous.
8.6    Credibility in Ratings
The problem of credibility is borne of a distrust of the rating process in many
places, rooted in instances where the rating agencies' response to changes in credit
quality have either failed to foresee financial disruptions or have lagged behind rapidly
moving events. A present example of the fallibility of ratings is found in the precipitous
downgrading of several sovereign credits in Asia during the on-going financial turmoil.
The accompanying box shows where the ratings by the credit rating company of
Standard & Poors for six Asian credits stood as of the December 1996. Generally, the
dividing line between "investment grade" and "noninvestinent grade" is drawn between
the BBB and BB categories, using the S&P nomenclature.46 The precipitous declines in
the ratings of Indonesia and Korea, and the serious slides of Thailand and Malaysia
caused havoc for them in the markets. It is interesting to note that in December of 1996,
46 The equivalent dividing line for Moody's is between Baa and Ba. Duff& Phelps and Fitch IBCA use the same
symbols and demarcations points between investment grade and noninvestment grade as does S&P.
57



all of the countries were listed as having either stable or positive credit outlooks.4' Not
only were the ratings reduced over the next two years, but the countries also went through
a continuing period of "negative outlook" on S&P's Creditwatch, which exacerbated the
uncertainty as to how far they would fall.
Credit Rating Volatility in Asia*
Decemberj196  September 197Dcnber 1997  September 1998
India            BB+ BB+                            BB+              BB+
Indonesia        BBB    BBB                         BB+              CCC
Korea             AA-              AA-              B+               BB+
Malaysia          A+               A*               A-              BBB-
Philippines      BB+              BB+               BB+              BB+
Thailand           A               A-              BBB              BBB-
$ Selected Standard and Poor's long-term foreign curency sovereign ratings
Several other emerging market sovereign ratings have been downgraded in the
wake of recent events. The drops were especially sharp following the Russian
devaluation and default in the summer of 1998, which sent all the emerging markets into
a tailspin. (Russia, prior to its currency and credit crash, actually had investment grade
sovereign ratings from both Standard & Poors and Moody's on some of its Euromarket
obligations.) Governments tried to protect currencies and depleted foreign reserves.
Depletions were followed by devaluations, flights of capital, and widespread concerns
over domestic firms and banks making payments in foreign currencies and, ultimately,
domestic currency as well as credit quality and systems deteriorating.
Subsovereign government credit ratings also were lowered, but selectively.
Typically because the sovereign rating, the effective estimate of "macro" creditworthiness
and the cap on the subsovereign ratings fell. Again looking at Standard and Poors, in the
year between October 1997 and October 1998, out of the 18 ratings on subsovereign
governments, 7 were lowered (two in Korea and five in Russia). At that time neither the
central Europe or South American subnational government ratings were affected.
Whether changes in credit ratings anticipate, coincided with, or themselves
stimulated financial turmoil in the financial markets is an important question and is being
asked with increasing frequency. Once rated, issuers run the risk that the agencies may
47 Stable means hat there is nothing on the horizon to suggest a downgrading. Positive outlook means that there are
indications that the rating may be upgraded, with a negative outlook implying that the rating may be reduced.
58



change their minds as economic and political conditions change. Relatively well-rated
Malaysia was shocked to have its rating dropped from the single A category to that of
triple-BBB just days before a large international bond offering, a move sure to cost the
country higher interest rates. The nation's Prime Minister called for controls over the
market power exerted by the rating companies. Interestingly, the credit ratings for some
of the lower-rated Asian borrowers were not changed amid the recent market tumult.
Evidently the rating agencies got it right for India and the Philippines in the first place. It
is also interesting to note that both of these on-the-fringe of creditworthiness countries
had lagged behind the formerly high-rated "tigers" when it came to economic growth and
the pace of capital market development.
8.7 Private Bond Insurance
Allied with the development of international credit ratings has been the
development of commercial bond insurance. Bond insurance acts as a third party
guarantee that debt service will be paid in timely fashion. The attraction is that the
insurer carries a high (best of all, the highest) credit rating from the internationally
recognized rating agencies. This third-party guarantee of debt with a high credit rating
lowers the cost of borrowing more to the borrower than the premium that it pays for the
insurance.
Bond insurance had its origins in the United States and has been tremendously
successful and pervasive in that country's municipal securities market.4" For bond
insurance to catch on and be successful, two things need to take place: (1) investors find
value in the promise of these insurers to meet the debt service payments and (2) investors
perceive differences in credit quality among issuers, usually expressed as a difference in
rates of interest that they demand. Differences in credit quality are typically encapsulated
in the rating symbols that are conferred on the debts. The commercial insurer has a high
rating from the recognized rating agencies that carries with it the promise of a lower
interest rate for the insured borrower. While these are accepted notions in the highly
developed subsovereign markets in the United States, the notions are still novel ideas in
emerging markets. Not surprisingly, the idea of bond insurance has been most
successfully applied to sales in international currency markets.
In recent years, bond insurers have transformed themselves into taking a much
broader approach than was traditional. Much as in the case of credit ratings, commercial
bond insurance is a product of the U.S. municipal bond market. It became an
international commodity as both the U.S. bond insurers found the original market
becoming surfeit and as international markets became larger and more complicated.
While all major insurers still have the AAA rating, several smaller insurance firms have
emerged that will be of less than prime grade and will cover credits that are less than
investment grade.
48 The United States routinely has 50 per cent of the dollar volume of municipal bonds sold carrying bond insurance.
59



The international bond insurance market was beckoning until 1997. In 1996
Standard and Poors asked chief executives of the international insurance industry what
they saw in terms of future international expansion. At that time, international business
made up about 2% of the bond insurance companies' "book.""9 They responded that
insurance was estimated to grow rapidly to 9% of the outstanding business in 2000 and
17% by 2005. Although most of the early bond insurance was sold in Western Europe, it
was clear that the rapid expansion was likely to come in Asia. In 1996, a consortium of
firms started up ASIA limited, which was slated to be a nonprime grade competitor for
Asia business, in particular. Also, the relatively small-sized insurer Capital Markets
Assurance Company (CapMAC) reached heavily into the international markets in hopes
of opening up new frontiers of profits. The company's excursion out in front of the
majors did not fare well. The Asia turmoil laid both ASIA and CapMAC low, the latter
being subsequently absorbed by the bond insurance giant MBIA.
The international financial turmoil of 1997 sent a strong warning to the industry
that perhaps the risks of the new emerging market frontier were not adequately
understood. On the other hand, it can be argued that the slow entry of the majors was
well rewarded since they avoided large capital charges and the downgrading that crippled
ASIA. Standard and Poors has opined that future growth in the emerging markets was
likely to be more cautious than was contemplated in 1996. The insurance industry had
had a bad experience once before by entering the real estate market. While the growth of
private insurance can be expected to continue, it will likely be much slower in the
emerging market area than was originally thought.
"[T]he primaries have redefined their strategy as a result of the recent global
crisis placing less focus on emerging markets and targeting less risky issuers and
transactions that characterized the early years of global expansion."50
None of the primary bond insurers were affected adversely by the 1997 and 1998
plunges. The primaries have only 3.1% of their par exposures in foreign based policies.
Municipal type-international business is about two and half times as profitable as
domestic work and has been largely restricted to superior, investment-grade issuers. With
less competition, now that CapMAC has been taken over by MBIA, the possibility of
higher premiums has improved.
The crises in the Asian bond markets in 1997, was followed by the broad-scale
emerging markets crisis of the summer of 1998, which was caused by the Russian
devaluation and default. The major insurers were spared the fallout because they had
been slow to add Asia credits to their risk portfolios. However, a new specialty insurer,
ASIA, was caught in the down-draft because of its regional orientation and, hence, lack of
diversification. Although upon its creation in 1996 it was given a respectable "A" rating
49 See Richard Smith " 1997 May Prove to be Pivotal Year, "Standard and Poor's Bond Insurance Book 1998
(Standard & Poors, New York) p.5.
50 D. Veno and R. Smith "Bond Insurers International Exposure; Opportunity Reward and Manageable Risk"
Standard & Poor's Credit Week Municipal (September 28, 1998) p.21
60



by Standard and Poors, the very next year this was lowered to BA as rating downgrades
of the policies in its portfolio caused a major erosion in its capital position.5"
9.     Monitoring and Oversight of Local Government Financial Condition
A senior government has a justifiable interest in subsovereign indebtedness and
subsovereign finances more generally. The kinds of information required to understand
the nature of subsovereign debt and the financial condition of local governments that are
debtors are much the same for governments and investors. Whether conditions are
improving or deteriorating, and actual or anticipated failure to meet expenditures,
including debt service, are vital concerns to the investors and should be to senior level
governments, as well. As a result of this common interest, an active securities market is
an important way to stimulate continuing interest in local financial condition.
Furthermore, what the central government is willing and able to do to avoid and cure
financial problems are of fundamental concern to the investor and, by the same token,
reflect how active or passive the sovereign will be regarding the financial health and
service capabilities of its progeny.
Financial monitoring may focus only on borrowing localities, with specific
information needs tied to disclosure as mentioned above. Or, it may be of a more general
nature pertaining to all localities and include such things as annual budget and
expenditure reviews. As indicated, much of what an emerging country needs for local
debt monitoring can be achieved by an active municipal securities market that insists
upon continuing disclosure requirements, and by the availability of audited, standardized
financial statements. The evolution of the credit market may be the major factor in the
evolution of the relationship between the central government and its junior partners.
Once market dictated transparency and regularity of reporting is achieved, there should
be a lessening need for on-going supervision or regulation of subnational jurisdictions.
The political and financial relationships between sovereigns and subsovereign
units are rich and varied. They are evolving along new lines, many of which will be
unique to the country's tradition and depending on where it is along the devolutionary
scale. Rather than prescribe a single approach to monitoring and oversight of
subsovereign conditions, it might be best to review what has been international
experience in both developed and emerging countries.
9.1    International Experience
51 The involvement of the Asian Development Bank and other owners of ASIA Ltd. was hoped to provide a certain
degree of insulation because of the "management insights" and one would suppose the political clutch that the owners
represented. The Asian Development Bank was a significant owner of ASIA and it was hoped that as a vital source of
credit, default on insured obligations would represent a "last option." Be that as it may, the tumble in Asian ratings
had terrible consequences for ASIA's insured portfolio. Short of widespread defaults, a massive systemic downgrading
of credits is the worst thing that can happen to an emerging market insurer.
61



The nature of, and justification for, national government oversight and
intervention in local government financial affairs varies fundamentally between those
nations that are unitary governments with a strong sovereign center versus those that are
federal systems where certain prerogatives are left with the affiliating states and their
local governments. For example, the United States, Canada, India, and several Latin
American countries have the federal system of government with specific powers and
prerogatives reserved to each of the layers. Within these federal systems, local
governments are typically subordinate to the states or provinces. However, local
governments within the states may themselves posses varying degrees of independence
for exercising self-governance. Generally speaking, in the federal systems, oversight of
the local governments finances will be largely left to the states. In the unitary nations, the
oversight will be by the central government from which all powers of the state are
derived.
9.2    United States Experience
Oversight and intervention devices vary greatly in the United States. As a general
rule, the older states in the Eastern part of the country (the original colonies) have tighter
controls and oversight over the local governments. These are called "Dillon Rule" states
wherein the local governments are solely progeny of the parent states and have only those
powers that are expressly given them in the constitutions and by the legislatures.'2 These
states , since the local governments are seen as accountable to the parent, often have strict
reporting requirements and if the local governments gets into trouble, the state is typically
in a position, if it chooses, to step in and take over the operation of the government,
including removing local elected and appointed officials. Because the administration and
finances of local governments in the U.S. have been at a high level for many years (since
the great depression of the 1930s), there are only a few examples of direct intervention;
but, it can be very sweeping when used.
In New York in the mid-1970s, the State of New York stepped in at the time of
the New York City crises and established a control board that, by virtue of its required
approval powers over all financial decisions taken by the City, effectively dictated city
finances. The control board stayed in effect for a period of five years until the City had
enjoyed two years of budgetary balance. At the time of the takeover, the State took back
the City sales tax and used it to secure indebtedness which was funded out the City's
accumulated deficit. A new financing vehicle, the Municipal Assistance Corporation was
created to sell bonds backed by the special sales tax and to refund outstanding City notes
as they came due. The City had use of the sales tax revenues only to the extent that the
debt service was first paid on the refunding bonds. The federal government initially
refused to provide special aid, however it did accommodate the workout by providing a
liquidity facility to the City. It also sponsored federal legislation that permitted the City's
pension system to invest in the City's and Assistance Corporation's securities without
52 Dillon was a state of Kansas judge that in the late 19th century laid out the theory of expressed and implied powers
for local govermments under the constitution of the states.
62



violating various federal prudential standards for pension investments. The pension
systems actually financed most of the recovery and bought something on the order of $4
billion in Municipal Assistance Corporation bonds.
Another example of strong state intervention occurred in Massachusetts with the
City of Chelsea. When the City was on the brink of insolvency (it had little debt
outstanding but was defaulting on payroll and vendor payments and there was widespread
corruption) the state governor removed all the elected officials from office and appointed
a receiver. That person reported only to the governor and ran all aspects of the City,
approving all contracts, tax levies and the like. The state also created a special guarantee
program to back the city's bonds which were sold to fund several improvements. After a
period of three years during which the city charter was rewritten and approved by the
state legislature, elections were held and the City was given back to newly-elected elected
officials. The legislation provided that the City had to meet certain tests including
financial operations or it would again revert to receivership.
The City of Bridgeport, Connecticut provides another example of state
involvement in the case of local government financial emergency. Originally, the State
of Connecticut attempted to use a limited control board approach that had the board
approve the City's budget, but with that body not having the power to oversee or enforce
the implementation of the budget. The City overspent its budget and, having various
political differences with the state, attempted to go into bankruptcy under the federal
bankruptcy code, which has provisions in Chapter Nine for defaults by local
governmental units." The State of Connecticut opposed the City's bankruptcy petition
and the bankruptcy court held that the City was not technically insolvent54 The state
subsequently stiffened the powers of the control board and provided transitional aid and
the City did not default on its debt.
Tlhe City of Philadelphia when faced with financial emergency in the 1980s came
under the oversight of a New York City style state control board that had oversight over
all spending decisions. The Washington DC model is much the same, where there is a
body that must approve budgets and approve expenditures and which, in this case, in
effect has taken day to day control of the operation of key city services. While elected
officials are left in place, they effectively lose control over the spending decisions.
The above cases of municipal intervention in the U.S. relate to states where the
state has traditionally taken a strong role in oversight of local government, or in the case
of Washington DC, the U.S. Congress has exercised influence over the capital city's
53 The municipal bankruptcy chapter was permissive in that a state could forbid a subdivision from filing under the
chapter. The state of Connecticut, however, did not legislate such a provision until after the city had filed for
protection. Since the city of Bridgeport was found not to be technically bankrupt, the issue of whether or not a state
could prohibit filing after the filing had been made was not decided.
54 John E. Petersen testified in the case and pointed out that the city had $400,000 in cash balances and had not
demonstrated that it could not get more by simply raising taxes or cutting expenditures. The judge accepted his
argument.
63



operation." There is, however, another salient tradition of much less oversight and non-
intervention in the Untied States, which follows from a tradition of "home-rule" whereby
the local governments have much more autonomy. This appears to be especially
prevalent in states west of the Mississippi River. The most important recent example is
that of Orange County California. In this case, the State of California had not specifically
denied local govermnents the use of the federal bankruptcy code. Undeterred, the County
when it had insufficient funds to pay debt on time in December 1995, entered into
bankruptcy (providing it with inimediate protection from its creditors) and defaulted on
$200 million in short term debt. The State of California refused to get involved and the
County entered into extensive litigation and subsequent settlements on its own without
state intervention or oversight.
In a somewhat similar case in the 1980s, the Washington State Power Supply
System, which was a large regional utility ( a regional authority) owned by a number of
local governments (special districts and municipalities) in three states. The Washington
Power Supply System (WSPPS) defaulted on revenue bonds. The State of Washington's
Supreme Court held that the basic contract on which the borrowing had been secured was
invalid and the borrowing itself was thus ultra vires (invalid).56 Because of the limited
obligation nature of the pledge, the bondholders were simply out of luck. The new
construction on nuclear plants ceased. No action was taken against the underlying
jurisdictions (except for securities fraud litigation). None of the respective state
governments got involved in attempts to bail out the local utilities.
9.3    Monitoring and Oversight in Other Countries
(a)    Canada
In the Canada federal system, the provinces have parental powers over the local
governments and effectively control their finances. The Canadian system is highly
decentralized when it comes to the national-provinces relationship but highly centralized
when it comes to provincial-local relationships. Quebec is a special case and sets its own
taxes. Localities are reliant on the property tax (which is set by the province) and
transfers from the provinces. Local government capital spending and borrowing are
generally subject to provincial approval and most borrowing is done through provincial
intermediaries (bond banks) that provided additional security by provincial pledges.
Local governments have limited powers and are essentially fiscal wards of the provinces.
55 Although the District of Columbia has its own elected officials, the U.S. Congress has what amounts to line item
control over the budget and can mandate or set aside other laws in the District.
56 The court reasoned that the utility only had the ability to charge for electricity actually produced and distributed. It
did not have the legal ability to levy charges and pay for electricity not produced nor received. This pledge of payment
even in the event electricity is not produced or received (a "hell or high water" provision of payment) was necessary to
meet debt service in the case of delays in completing construction, which there were because of massive engineering
and construction problems and environmental concerns.
64



(b)   South Africa
South Africa represents an important example of an emerging country that reflects
the pressure of a changing govermmental structure on intergovernmental fiscal
relationships. The country has moved from a highly centralized system of government to
one that in its constitution now recognizes three spheres of government, national,
provincial and local. South Africa has gone through a reorganization that has combined
the formerly white municipalities with the less affluent black townships. The rapid
amalgamation has led to a variety of problems, including the nonpayment of property
taxes and utility bills by the newly absorbed area. Since the public sector financial
structure of South Africa has placed and continues to place much of the fiscal
responsibility on local units, the nonpayment of taxes and charges has caused widespread
fiscal stress. As of now, insolvent local units come under the control of the provincial
level. Unfortunately, that level is on even more tenuous grounds than the local
governments themselves.
Responding to the fiscal problems at the local level, the national government has
instituted "project viability" which requires quarterly reports from municipalities as to
their financial position. These positions are subject to monitoring and, if needs be,
supervision of the distressed units. The metes and bounds of supervision are yet untested,
but the quarterly current financial position monitoring is probably the most continuous
and timely found in the world.
(c)   Argentina
Argentina is a historically highly decentralized governmental system, with
significant powers given to the provincial governments. The provincial governments are
the parents of the local governments, much like the U.S. and Canadian systems. The
regions vary greatly in income and level of development. Essentially the central
government raises taxes, the majority of which are then transferred to and spent at the
provincial level in providing services. This financial structure obviously places great
importance on the intergovernmental distribution mechanisms. All three levels of
government are permitted relatively free reign to borrow, much of which has been done to
cover operating deficits. This gave rise to large amounts of unsustainable debt. Another
problem was the influence that local governments had over the investors. Most of the
financing was done through provincial owned banks that were highly influenced in their
investment decisions by the needs of state and provincial governments.
After large increases in debt both domestically and internationally in the 1980s,
the central government stepped in to bail provinces and cities out by replacing local debt
with national debt. In the 1990s the national government essentially closed the window
of the provincial banks to lend to their parent governments. However, provinces have
continued to borrow by pledging future intergovernmental revenues and borrowing from
private banks. Generally, in Argentina there has been a recurring problem of a lack of
discipline where provinces and localities can borrow to cover current deficits. Since the
65



provinces and municipalities have a high degree of independence, the central
government's ability to mandate the former's behavior is limited. The federal
government and provinces have entered pacts to try to control provincial spending and
borrowing.
(d)   Brazil
Like South Africa, Brazil's constitution provides nominally "equal status" to all
three levels of government. Throughout the 1990s, the country has embarked on a policy
of devolution. As in the case of Argentina, lack of effective control by the central
government led to the running up of high levels of indebtedness by the provinces and
localities. The provinces and localities experienced widespread defaults in the 1980s.
Subsequently, the debts were rescheduled in the 1980s by the central government in order
to convert short-term debt into long-termn debt. A major problem in Brazil was that the
national government had no effective control over the amount of debt incurred by the
local government and the provinces. In the final analysis, Brazil was unwilling to allow
for massive defaults. As in Argentina, negotiations among the levels of government are
ongoing.
(e)   Transitioning Economies in Europe
All of the transitioning economies of Eastern and Central Europe emerged from
highly centralized systems where the local government subdivisions were service delivery
points for the state and highly dependent on the state for fiscal transfers. In addition,
local governments might own various enterprises that generate revenues, but that often
operated at a deficit. Financial reporting systems were often designed for measuring
levels of inputs or for tax purposes.
Economies tended to be on a cash basis with a small and highly centralized
banking sector and no functioning capital markets. Major capital spending was financed
by grants or soft loans and was directed by the central government or, in the case of
smaller routine projects, financed on a pay-as-you go basis by the locality. As a result of
the format and purpose, financial reports often gave little insight into the financial
condition of the local governments. Since the local governments had no existence
beyond the state, there was little consideration of monitoring or interventions aside from
the removal of officials by the central government if they failed to perform as wished.
There evidently has been little thought given to formally coping with financial
emergencies of local governments, although Hungary reportedly does have legislation in
place for municipal bankruptcies.
More recently, these transitioning governments have been moving to greater local
autonomy and various financial reporting systems have been instituted to provide more
usefil information about local conditions. These systems tend to follow the European
model of full accrual accounting and the balance sheets are often spotty and suspect
because of questions of ownership, value of real assets and accounts receivable. Largely,
66



capital financing has been done through specialized loan funds or commercial banks
(themselves often undergoing the process of de-nationalization and carrying suspect
balance sheets) that have traditional relationships with the local govermments.
9.4   Formulating and Enforcing an Intervention Process
Intervention may be needed when a local government is in fiscal stress. In the
credit market context that means that the governmental obligor is threatening or has in
fact defaulted on its obligations. There, of course, can be other indices of stress and
default may be the last step in a long journey of deteriorating service levels and finances.
The question then becomes, what steps do senior level governments (or others) take to
protect citizens and creditors and to correct whatever is causing the financial malaise?
Available remedies to financial emergency may be of most immediate interest to
investors, but how they are enacted and enforced is very much a question of national
policy interest since they affect issues of self-governance, the delivery of essential
services and the health of the financial markets.
An intervention process can be defined by generic national law, or be a matter of
individual contracts between debtors and creditors. A viable municipal borrowing market
need not have a detailed statutory intervention process. Rather, the parties can define the
intervention and feceivership processes contractually, and these processes can be
customized for a particular deal. However, there may be constitutional restrictions on the
ability of a local government to contract for intervention and practical problems of having
courts enforce the contract. Therefore, it is likely best that national policy makers
develop an intervention process by law or regulation for reasons both of clarifying
creditor's and debtors' rights and of having a clear framework for dealing with
subnational financial emergencies more generally.
As we have commented before, the best approach to financial oversight is through
market forces that demand the timely provision of information, that, in turn, serve to control
access to mnarket, and thereby force financial discipline. But, information demands from
rating agencies and the widespread use of uniform accounting and financial reporting
standards are essential in this scenario. It may be that where the institutions and market
players are in the formative and untested stages, sole reliance on the market may be
premature. Nonetheless, it is important not to discourage market initiatives nor to weaken
market incentives. Legal requirements that bond market participants must disclose and
send information to a central point help the markets work more efficiently and prod
governments to assume and fulfill reporting responsibilities.
While individual local jurisdiction should be free to negotiate monitoring and
intervention provisions with creditors, a codified national approach helps to demarcate the
relationship between subsovereigns and the financial markets. A commonly used law or
regulation in this respect lends certainty and avoids the fallout that an individual default
might have on other local jurisdictions.
67



A legislated intervention and default cure process should include the ranking of
and remedies for respective creditors. A number of options are available as to the
priority of claims. In some countries, local governments are able to put owners of
bonded debt at the head of the line. In others, the depository bank or the senior level of
government gets first fruiits. In still others, domestic creditors come before foreign ones,
a position that is likely to discourage foreign lending. In the case of security, the first to
take physical possession may have the advantage.
Options are numerous in terms of remedies available to creditors. Creditors could
be given the right to intercept funds due to a jurisdiction from other spheres of
government. They could also have a right to trigger imposition of an additional tax
within the defaulting jurisdiction, or to trigger the appointment of a receiver to control
expenditures or the operations of a jurisdiction. At the same time, the citizens need
protection for "minimum essential" services, such as water needed for health and safety,
sewage operations, and refuse collection. Creditors should have the right to apply to
court for execution on their security interests and for judicial intervention. Courts should
be empowered to deal with insolvency and the priority of claims among creditors, and to
discharge debt where the local jurisdiction could not otherwise be made solvent.
The particulars of an intervention process will reflect the structure of govermnent
and varying concerns about preserving local democratic prerogatives and protecting
essential services. However, it is essential that it be as clear and comprehensive as
possible and that it provide substantive remedies and protections for private investors.
68



10.    Designing and Implementing Credit Assistance
In the initial sections of this paper three groups of subsovereign jurisdictions were
identified based on their "readiness" to access private credit markets. Financial
condition, managerial skills, and (to a certain extent) size each contribute to the
estimation of credit quality and possible private market access. The first group includes
those jurisdictions that already have access, but could enjoy more and better options
given a supportive policy and regulatory environment. The second group potentially can
have access, with various forms of help, including credit assistance that is complementary
with the operation of credit markets. And a third group consists of jurisdictions that
cannot access the market, even through market-oriented intermediaries (or the effort to do-
so is likely to be counterproductive from a policy standpoint)."
A central question is how to assist jurisdictions that do not now have the resources
to be self-financing, possibly because they do not have an adequate tax base. If the
central governments chooses to assist these jurisdictions through a predictable and stable
system of intergovernmental transfers, even smaller units can have adequate local
revenues. Such revenue streams from both local sources and intergovernmental transfers
can be used for capital investment (with or without borrowing).
If a community has reliable revenue streams, then it belongs to the second group.
Access to borrowed capital should, in principle, be available. Private markets may still
not serve these jurisdictions because of the small size of their financing needs, their
inability to do analysis and planning, or their inability to deal with the capital markets
concepts and practices. For this group, market intermediaries and technical assistance
could be made available to help bridge these gaps. This middle group of potential
borrowers is the major focus of this section.
National governments can provide an environment to promote marketability of
local debt. Macroeconomic and regulatory policies that are right, and in place, are a large
part of what needs to be done to help municipalities have access to borrowing. Several
questions arise when considering assistance for local jurisdictions in the two groups that
do not have access to the market. Should assistance of any kind be rendered to help local
jurisdictions gain access to credit? If so, what forms should such assistance take in order
to encourage private capital market participation and to minimize "crowding out" of
private capital providers? How should developmental resources be allocated? Finally,
what administrative and technical assistance should be used to deliver assistance?
57 It is useful to understand why particular jurisdictions cannot access the capital markets. Distinctions in this regard
are what differentiate the second and third groups. The second group is made up of jurisdictions that have or could
have predictable, stable, and sufficient revenues to meet their service delivery responsibilities, but whose capital
requirements are relatively small, or who lack the skills and capacity to develop and pursue borrowing options. The
third group includes jurisdictions that do not have, and cannot develop, adequate financial resources to meet their
responsibilities. Borrowing will not solve this problem, and could even exacerbate it. One should not create capital
borrowing programs for municipalities with inadequate revenue sources.
69



10.1   Types of Credit Assistance
Assistance in accessing capital markets can take several forms. These range from
technical assistance and credit enhancement, to direct lending positions and interest rate
subsidies that induce private market participants to join a transaction. At least three basic
questions should be asked to judge whether to use a given tool in a given instance:
* Does the assistance technique leverage private sector investment?
* How likely will this tool crowd out private-sector capital?
* Does the tool increase the risk of moral hazard; that is, how likely will it be
misinterpreted as a form of state guarantee?
The above questions are considered next in the context of several forms of assistance that
might be rendered to promote private capital market development.
10.2 Technical Assistance
Technical assistance that helps familiarize jurisdictions with credit markets
practices and helps them become more "creditworthy" will be most likely to leverage
private-sector capital. It is least likely to crowd out private capital and the most likely to
attract its interest. It also is least likely to raise moral hazard risk. Technical assistance
and training provided in accounting and budgeting, identifying and analyzing capital
investment projects, operating and managing facilities, and the like, expand managerial
skills and encourage more efficient financial practices.
As a practical matter, technical assistance is difficult to do in the abstract and it
goes much better with practical applications. Technical assistance and training should
maximize the likelihood that a borrower will eventually be able to do some or all of these
functions for itself. Wherever possible, the bias should be toward creating local
institutional and technical capacity. Technical assistance and training that is most
supportive to capital market access are found in capital planning, cash flow projections,
and project management. These disciplines allow the municipality to work with budget
constraints; to practice matching revenues to expenditures; to consider how much to
borrow, for what purposes; and how quickly it can and should pay back loans. In the
technical realms of debt finance, there are a variety of choices. As is discussed below,
either public or private lending entities can help provide access to markets, especially if
standardized documentation and processes are developed. The standardization of the
documentation and of the components of the transaction helps to resolve questions of
security and keep costs down.8
58 The costs of developing pioneer bond issues are considerable since they represent for public and private parties
alike a heavy investment in learning skills, developing documentation, and charting new procedures. In the case of
the Philippines, these costs for four relatively small bond issues ranged from 4 to 5 per cent of the total issue proceeds.
Bond issues are very much subject to economies of scale since the novelty and complexity of a deal may have little to
do with the size of the issue.
70



10.3 Financial Assistance
Financial assistance that is designed to help local jurisdictions gain access to
private credit can take several forms. When contemplating linkages to the private credit
markets it should be recognized at the outset that direct financial assistance that is
insulated from a market test carries significant drawbacks and risks, since it always
embodies the problems of adverse selection and moral hazard.59 And to the degree it is
institutionalized, it can foster long-term dependency.
The lure of cheap credit provides an incentive for local governments to be or
appear to be needy, rather than to be self-sufficient. Direct assistance also creates hidden
subsidies in the form of contingent guarantees and enhancements. It also can crowd out
the private sector, which will typically both set higher credit standards and charge more
for lending. Finally, direct assistance is usually less efficient at leveraging private sector
resources than technical assistance. But concessional finance continues to have a role in
most economies, either to encourage sought-after activities or to surmount barriers. And,
careful design can reduce the above drawbacks and risks, even if it cannot eliminate
them.
In addition to the above problems, concessional financing can distort choices.
The burdens in future facility operating costs can be overshadowed by the sudden
availability of capital credit. Financial assistance reduces only the cost of capital
facilities to the borrower. By helping the unit access debt, the assistance provider is also
helping the borrower take on increased facility operating costs. Borrowers with no access
to capital from hard credit sources may not have been required to fully investigate these
costs nor to build them into their budget planning. There may be little or no capacity to
properly operate, maintain, and ultimate replace the facility, which rapidly slips into
disuse.
A fundamental principle governing the design of direct financial assistance is that
it should always be looking for an exit strategy and a shifting of obligations to the
commercial credit markets. This means that the assisting entity might absorb some of the
risks that are at present unacceptable to the private credit markets. 6
10.4 Direct lending
59 Adverse selection refers to the phenomenon of having only high-risk and otherwise unmarketable credits use the
credit preference and receive the advantages. Moral hazard refers to an incentive for users of the preference to shirk on
their obligations (default).
60 This might mean finding a way to eliminate a narrow risk (e.g., environmental risk) through provision of risk
insurance. Or, it might mean taking a junior lien in order to comfort potential private lenders. Or, it might mean a
guarantee on the "long end" of a debt structure if commercial lenders are able to provide short and medium principal
maturities.
71



Direct lending is an inefficient form of financial assistance, and one that is likely
to crowd out private lenders and to invite moral hazard. Unfortunately, many of the
direct lending programs aimed at the local level have been directed from the center, often
making loans to unwilling and inattentive local units, that ended up treating them as
grants. However, there can be constructive direct lending roles. The International
Finance Corporation's A/B Loan syndication and certification structures have
demonstrated leverage efficiencies can be achieved with such instruments, if they are well
designed. In order to increase leverage and reduce crowding out and moral hazard, direct
lending should be designed to induce co-financing by commercial lenders. The smallest
possible direct lending role required to achieve this objective will minimize the risk of
crowding out and maximize the efficiency of the assistance rendered. Thus, for example,
if only a five percentjunior lien position will induce the private market to join in a co-
financing, the provider of this form of assistance must be prepared to forego a larger loan
program.
Although past direct lending programs have had a poor record for loan repayment,
the tide appears to be turning as of late in some countries (See box on Development Loan
Collection Enforcement). Credit discipline, if it is instilled into direct lending programs,
can help prepare borrowers for the realities of the private market so long as sufficient
economic inducements can be designed to enable borrowers "to graduate" to private
market access.
10.5   Debt service subsidies
Debt service (either interest rate or principal payments) subsidies resemble direct
lending in that they constitute ongoing payment streams to support local borrowing that
can be inefficient. They are more likely to lead to moral hazard than is more indirect or
softer forms of financial assistance, such as insurance, partial guarantees or technical
assistance. Nonetheless, they can be useful tools, if they are well designed to insure two
things: (1) that the smallest subsidy necessary to induce private capital market
participation is used, and (2) that they are used only where the assistance provider can be
reasonably assured that this is the only tool that will make the borrower creditworthy.
Among the devices used to subsidize interest costs through the private credit
system are so-called linked deposits and co-lending programs. In the former case, a
commercial bank may receive a deposit from a government intermediary that agrees to a
reduced rate of interest if the bank agrees to use the deposited funds to make a loan for a
particular purpose to a local government. In other words, the deposit is "linked" to a
particular use of loanable funds. Advantages of this approach are that the private
institution still takes the credit risk, must do the credit analysis, and administers the loan.
A similar approach is where the government intermediary makes a co-loan for a portion
of the principal amount and does so at a reduced rate of interest. Here, the borrower gets
the advantage of the "blended rate" on its loan. The private lender, however, still has its
principal at risk and administers the loan, while having the intermediary as a partner in
the transaction.
72



A well recognized problem has been the moving of local governments from a soft
loan environment into a atmosphere of harder credit demands. Development banks have
generally had a very poor loan repayment record which has made many observers
skeptical about the ability of local govemments to make the transiton into a securities
market atmosphere. However, there are efforts to core the situation in some countries
by calling delinquent borrowers on the carpet. South Africa demonstrates one way of
doing this.
In January 1996, the Development Bank of South Africa (DBSA) inherited the
development loan portfolio of the Ministry of Finance for local governments. The
portfolio, known as the LALF portfolio, consisted of approximately 390 loans
representing about 900 million Rand ($50 million) that had been made to local
gove.nments primanly under the old regime that existed prior to 1994. At the time of the
transfer, most borrowers were timely in their payments. Amid the trmoil of the
transition into the new govenmnental structure in South Africa, many of the local
government obligors for whatever reason began to go into default. The DBSA in 1998
undertook a program of enforcing payment of the loans. The DBSA saw its role as that
of acting as a bank with commercial incentives and a capital position to protect. Thus,
while the original terms of the loans might be concessionary, its role was to keep the
payments on schedule and to instill discipline into the borrowers. DBSA is not expected
to lose money and erode its capital base; its goal is to make some money and make
reasonable returns to capital, while promoting longer-term, socially useful development
DBSA recoiled at the growing delinquency rate of the loans and moved to deal
with the local authorities in a manner to re-institute timely payment of the loans. Loan
officers were assigned to the respective regions and given procedures for going after
over-due loans. In three provinces, 32 of the 40 loans that were defaulted were put back
on a timely basis using technical assistance and the threat of closing off future credit from
the bank as means of restoring the credits. In South Africa, both govermnental and private
lenders have the power to seize assets of the borrower.
10.6 Guarantees and Insurance
Guarantees are a traditional and important form of financial assistance. Due to
their contingent nature, their cost is not easy to measure at the time the guarantee is given.
They can lead to lax lending practices and hinder the creation of effective private
markets. However, guaranteeing specific risks or specific maturities may be worth
consideration. The ability of a credit assistance provider to reduce or eliminate specific
risks in a transaction (e.g., specified environmental hazards or the repudiation of certain
contractual obligations) or to back maturities that the private sector is unwilling to
provide can leverage private capital investment. Properly designed and implemented, a
73



"surgical" use of guarantees can reduce the risks of crowding out and moral hazard.
Some of the World Bank's guarantee operations have begun to demonstrate the utility of
guaranteeing specific risks or maturities as a means of inducing private capital providers
to participate.
One way to reduce the risk that these and other credit enhancements will crowd
out commercial lenders is to price them according to the respective degrees of risk
presented by different borrowers. Under such a regime, local borrowers and commercial
lenders see the relative costs involved in securing guarantees and will treat them as
having a cost. The idea of charging needier borrowers more in terms of a premium than
those that are better off may be offensive to some observers. But, the incentive must be
present to improve financial operations if the governments are ever to stand on their own
in the credit markets. Buying down part of the costs with grants but making the issuers
borrow at risk adjusted rates on the margin may be one approach to forcing units to pay
attention to market interest rates and to scale projects accordingly. Yet another option is
to price enhancements with "seasoning" premiums that are in part subject to rebate as
borrowers' live up to their obligations and see their circumstances improve.
10.7 Intergovernmental Payment Intercepts
As discussed previously under forms of security, intercepts are a form of financial
assistance, and one that need not have any significant cost to the national government.61
Intercepts can be a powerful credit enhancement. They are more effective, the more
dependent the jurisdiction is on transfer payments from other spheres of government.62 A
stream of stable, predictable intergovernmental transfers can be made pledgeable and
interceptable, which would enhance creditworthiness. Significant penalties or
administrative fees when the intercept is exercised could discourage jurisdictions and
creditors from relying on the intercept mechanism, and encourage them both to manage
the debt payments in a businesslike fashion. Such a mechanism can leverage private
sector funding, rather than crowd it out. In the Philippines, the intercept mechanism is
being combined with a guarantee scheme to enhance loans provided by private banks to
local governments. See box on the Local Government Unit Loan Guarantee Corporation.
61 If the aid is to be provided anyway, making it pledgeable and interceptable does not add to the cost. Any
administrative costs could be borne by the borrower.
62 Traditional rating agency analysis has given state aid intercepts only modest credit enhancing power in the United
States. IHowever, the power of a state aid intercept can be substantially increased if the state aid flow goes through a
trustee-administered "lock box" arrangement in which debt-holders have first access to the revenue as a matter of
course.
74



Local Government Unit Guaranty Corporation
Under the sponsorship of the Philippine Bankers Association, a banking
consortiun of 22 Philippine and foreign banks has created the Local Govenmment Unit
Guaranty Corporation (LGUGC). The LGUGC will provide guatees (ins ce) on
loans made by participating financial institutions to local goverments (LGUs). Some
230 million Pesos were raised by subscription from the participating banks, deposited
into a special account, and are available for backstopping the guarantees of periodic debt
service payments.63
The premise of the LGUGC progran is that the guarantee will stimulate private
commercial bank interest in local govermnent credits. For institutional and regulatory
reasons, these have been borrowing only from GFI s (Government Financial
institutions). The GFIs are being privatized and are sometimes "loaned-up" on LGU
credits. In.any event, the GFIs and the goverrnent are under pressure to open up the
LGU market to greater competition and to develop a munticipal bond market The
guarantee program gives comfort to the private banks as they start lending to the sector.
It is also anticipated that it can serve as an enhancemuent for bond issues. The guarantee
depends in large part on a pre-assignment of the LGU's intergovenmmental payments to
the corporation which can be tapped in the event of default. The initial program is
geared to the largest 120 LGUs at the outset. Once the guarantee system is in place, it
can reach down to smaller governmental units.
10.8  Intermediaries for small borrowers
The policy issue is whether a special intermediary should be created for
jurisdictions that cannot access credit markets through existing market mechanisms.
Special intermediaries should not replace existing commercial lending and underwriting
institutions, but should complement them. In some countries, the private sector may be
able to provide such intermediation, without the need to create a new agency or function
for government, and, in principle, this is desirable. However, a small issue may not
attract the market's attention and even if it does, it may not be economical to finance
them in the formal securities markets.
Many kinds of intermediary models are possible, including bond banks (see
below), bond pools, revolving loan funds, and municipal lending institutions. Such an
institution might borrow in its own name and use the proceeds to purchase debt
instruments of municipal borrowers (banking), or it might assemble and repackage
63 Technically, the program works like bond insurance programs as opposed to a letter of credit or conventional bank
guarantee because-there is no acceleration of debt in case of default. Rather the insurer "steps in" and assumes debt
service payments and keeps the original term of the loan going. This means that the annual liability is much less for the
insurer and it can work at a higher leverage factor.
75



municipal debt instruments and make them available to the market (pooling). A major
attraction of such pooling and banking structures is that they can provide economies of
scale in issuance and because of their larger size of issuance have a better chance of
attracting a secondary market interest.
But, any intermediary model has costs. These costs may include administrative
costs, perhaps subsidized relending rates, credit enhancement costs, or a combination of
all three. However, with a properly designed and efficiently run intermediary, the costs
will likely be less than those involved in outright capital grants, and will have the virtue
of helping local officials understand and practice the trade-offs involved in debt finance.
Special intermediaries can be designed to provide several services to local
governments. These include access to capital markets for jurisdictions that otherwise
would not have access; savings on the sundry fixed costs of debt issuance; streamlined
and standardizing borrowing procedures and documentation; assistance with capital
planning and cash-flow projections; and pre-structuring loan packages. The senior
government may also decide whether to offer direct financial assistance, such as credit
enhancement (e.g., in debt service guarantees or itself procuring commercial
enhancements) or by re-lending the intermediaries' funds at subsidized interest rates.
The more "passive" the financial assistance and the more it is used in tandem with normal
credit channels, the better. As discussed above, there are moral hazard risks associated
with direct financial assistance that is insulated from a market test. In our opinion it is
better to expose the novice borrower to the actual costs of capital and the discipline of the
market, at least on the margin.
If the objective is to promote local self-sufficiency, it is generally advisable to
avoid enhancement methods that reward dysfunctional units, that are non-transparent, or
that crowd out private investment. If there is a stream of stable, predictable
intergovernmental transfers for jurisdictions that themselves at present lack the resources
to be self-sufficient, these transfers could be made pledgeable and interceptable. This
would enhance creditworthiness and leverage private sector funding at little or no cost to
the national government. However, it is a judgment call as to the extent otherwise
impecunious units should be encouraged to borrow. For those local units that have little
prospect for financial self-sufficiency, it may be simply a way for the senior level that is
providing the transfers to pass the buck of indebtedness.
76



Assisting Small Bond Issuers: Is the U.S. Experience Rtelevant?
The U.S. is often perceived as having a highly sophisticated financial market, with
knowledgeable and skilled investors and issuers. But that is not true with many of the
estimated 40,000 subnational governmental issuers in the U.S. bond market. Many are
small and unsophisticated. However, the well-established legal -and regulatory processes,
the availability of skilled advisors, and competition among potential lenders greatly
improves the access of smaller units to the markets. A combination of state-backed
financial intermediaries such as bond banks, the availability of private bond insurance,
and preferential federal tax policy keep competitive pressure on dealers and banks in*
providing services to small issuers.' As a result, the typical U.S. small local government
credit has been elevated to being very competitive in the markets.
Encouraging this largely market-driven process, a good deal of attention has been
given to upgrading local govenmuent financial management practices and reporting, an
oversight function that states have long had as regards their progeny but have worked at
with increasing vigor during the past few decades. Bond banks, bond insurance and other
organized lending and credit enhancing programs have placed requirements that local
govenmments report their financial condition on a regular basis along lines of generally
accepted accounting principles. These developments, along with the widespread use of
credit ratings and recently adopted securities-related reporting requirements of the U.S.
Securities Commission, have also worked to standardize and regulanze financial reports.
10.9 Bond Banks
A recent survey of state bond banks in the United States shows that these have a
variety of administrative and program structures and financing experience that can be
useful to emerging municipal markets.65 Initially, the mission of the bond banks in that
country was to improve access to the financial markets for small, rural governments.
Instructively, the approaches used in operating the banks and the scope of their activities
have varied, depending on the relative priorities among various of financing needs and
their legal and political environment. The earliest bond banks were seen as general
extensions of state government creditworthiness to help close the marketing gap for
small, isolated general-purpose municipal issuers. As experience was gained, the banks
frequently have taken on specialized areas of activity, such as financing environmental
activities, local schools, short-term borrowing, and equipment leasing. They have also
64 Tax policies can be very important along these lines. Small issuers (those that issue less than $10 million annually)
have a niche market because of special tax provisions regarding the holding of their bonds by banks. The provision
cannot be "passed through" the pooled borrowings by bond banks, which encourages the small issues to be sold
separately and has made bond banks somewhat less competitive for loans to the smallest issues.
65 John Petersen, A Primer on State Bond Banks in the United States, paper at the First Conference on Capital Market
Development for Subsovereign Governments, World Bank, Santander Spain (October, 1998)
77



moved to taking up limited obligations, participation in structured transactions, and
providing enhancements."
Because they are competitive with private lenders and dealers and many times can
finance at lower costs or on better terms than are otherwise available in the market, the
creation of bond banks has been resisted by both commercial banks and securities deal in
many states.6' After early adoptions in several states, the bond bank movement in the
U.S. slowed down as opposition from such competing interests and concerns about
stretching state credit enhancements too thin began to crystallize. More recently, the
interest at the national level in creating revolving funds as a means of replacing recurring
capital grants from the central government has re-activated interest in state-based
financial intermediaries, including the traditional bond banks.
66 Not surprisingly, most U.S. bond banks have been created in largely rural states in which there were not strong
regional financial markets. For example, the creation of bond banks in New Hampshire, Vermont, and Maine was to
an extent a reaction to those states being in the backwater of the New England regional financial markets that were
dominated by large Boston banks.
67 Because they aggregate small issues into a large one, they can provide economies of scale. But that process reduces
the amount of business available to regional dealers and banks. On the other hand, large money center dealers may
support their creation if they think they will get the underwriting business. The latter, however, have little political
influence in the state compared to those investment firms that are located there.
78



11.   Concluding Observations and Policy Guides
The foregoing sections of the paper have covered a number of aspects regarding
how subnational governments link into the overall framework of the private credit
markets. The linkages are multifaceted, even in relatively immature systems. And, they
are dynamic, changing as the structure of governments and the financial markets change.
We chose to concentrate in the activities and constraints of borrowing by subnational
governments, which served as the focal point for discussing relationships, practices and
policies. We commenced by determining which governments are likely to be candidates
to borrow, what security they might pledge, how they should or should not be constrained
in their decisions, and the various instruments they might choose to use and means by
which they would place them in the market.
Linkages are approached from various angles: the potential investor, the
organization of the securities market and its regulation, likely investor groups and their
regulation, the need for inforrnation and its use to analyze credit, and the rating and
private insuring of securities. We concluded with an examination of two of the specific
policy arenas, credit assistance and financial emergencies. We saw what tools senior
governments might choose to wield in developing markets, looking at forms of credit
assistance and methods by which senior governments monitor and, when necessary,
intervene in the affairs of local governments.
Throughout the toolkit, we have made various suggestions and drawn conclusions
regarding how national policy regarding linkages between subnational governments and
the private financial markets should be fashioned. These suggestions and
recommendations, of course, are constrained by two things:
* own experience with and our reading of others' experiences in several emerging
countries, and
* our objective of promoting competitive and sustainable private-sector credit markets
that deal in subnational securities and loans.
There is no desire to say here that "one size fits all," nor to minimize different
structures of government and levels of credit market development. Nonetheless, we
believe that the following observations regarding policies and practices form a point of
departure in appraising a country's readiness to promote markets for subnational
government securities. They also can stimulate debate over the substance of existing
markets and how access to them might be improved and new markets best engendered.
79



11.1  Policy Guides Regarding Security Pledges, Instruments, and Methods of Sale
*  It is better not to make distinctions among local governments as to their powers and
the procedures they are to follow with respect to borrowing. Generic laws that
contain generally broad formulations and simple parameters based on objective
criteria are better.
*  The legal status and remedies for various types of obligations of subnational
governments should be clear. That is frequently not the case where such obligations
have formerly carried explicit or implicit sovereign guarantees. The ultimate security
and the process for enforcement of the obligation by creditors should be explicit and
easy to effectuate.
*  Parties to debt transactions should be empowered to design security provisions to
meet circumstantial and specific needs, as well as general requirements. Essential
services, for example, can be defined and minimum levels protected in the case of
asset and intergovernmental transfer pledges.
- In addition to tax-supported debt, jurisdictions should be able to offer limited
obligation security arrangements (e.g., revenue bonds) that do not involve the pledge
of general revenues.
*  Local governments should be able to enter into tariff-setting and other covenants with
respect to limited obligation debt. If senior level governments retain rate-making
approval, there should be provision for their giving prior approval to rate adjustments
for purposes of debt contracts or otherwise providing indemnification against default
for want of the ability to adjust rates.
*  Local governments should be able to make assignmnents of their revenue (i.e.,
revenue intercepts). An exception can be made for those revenues as are determined
to be necessary to provide minimum essential services. This might also be done by a
regulation that only up to a specific fraction of transfers may be pledged to debt
service payments."
* Local governments should have the ability both to create and to join with others in the
creation of special districts to address service needs related to specific areas or
activities. They should have revenue powers that allow them to capture value created
by their activities and investments within or on behalf of those districts.
*  The marketplace should be free to decide on specific types of instruments and
associated payment mechanisms to employ so long as those types are generally
pernitted of borrowers. Unless there are compelling reasons to make restrictions of
68 This is akin to a minimum coverage requirement often found in limited obligation bonds.
80



all borrowers, there is no basis to treat subnational units differently so long as there is
full disclosure and competitive norms are met.
* Wherever possible, it is best to introduce competition into the marketing of debt.
This can be accomplished by formal solicitation and bidding procedures to procure
services of underwriters, advisors, banking services, and other professional
specialties.
11.2. Policy Guides Regarding Borrowing Powers
* Restrictions on borrowing powers may be appropriate in many emerging market
situations where the objective is to balance local self-determination with limited
experience and capacity. However, the overall object is to link self-determination
with fiscal self-sufficiency and local accountability. This is best accomplished by
observing market driven standards and using market-driven incentives.
- Short-term debt should be limited to only providing for cash flow shortfalls in
anticipation of realistic income streams to be realized within the fiscal period. Short-
term debt should be paid off annually by the close of the fiscal year, and remain paid
off for a reasonable interval of time. Exceptions may be in the event of financial
emergencies and part of the emergency or intervention plan.
• Long-term debt should be limited to capital investment in property, plant, and
equipment. It should not be used to finance operating deficits except as part of a
financial emergency recovery plan as defined by statue and regulation.
* Limitations in debt outstanding should be related to either the available tax base or
some measure of recurring revenues and where possible, should be expressed in terms
of annual or maximum debt service. For example, total debt service (principal and
interest) on general obligation long-term debt should be limited to a percentage of
projected available annual revenues. The market should determine acceptable ratios
of debt service coverage in the case of self-supporting debt, which will vary by the
technical and economic aspects of the improvement and the security pledged.
* Approval of borrowing by a jurisdiction's legislative body in most cases is sufficient,
so long as the debt outstanding after borrowing falls within certain parameters.
Routine approval by senior levels of govenmment diminishes local flexibility and
responsibility and opens the door to political manipulation. However, waivers of
limitations in unusual cases by cognizant state or national authorities may help
flexibility.
81



11.3   Policy Guides Regarding Financial Market Regulation and Disclosure
* Local governments as borrowers should fit within the overall regulatory framework
for banking and securities markets. Balancing the competitive norms of market
efficiency and development while preserving the integrity of the payment system and
protection of investors is necessary. Generally, subnational governments should enter
financial markets on an equal footing to private firms, while recognizing the
distinctions that flow from their taxing and governance powers.
* A secondary market for securities is important to investor liquidity. But, formal
listing on exchanges should only be required where the potential size of secondary
activity justifies the time and expense. Neither local governments nor smaller private
firms should face excessive requirements that may accompany listing on security
exchanges.
* A secondary market for subnational can be developed among the Over-The-Counter
markets, which may prove to be more efficient for smaller issuers, the bonds of which
are traded infrequently. Investor protection needs to be balanced with permitting
economical access by smaller issuers, which should be a fundamental tenet of
disclosure requirements.
* A key concern in market regulation is that of disclosure. Subsovereign securities, as
those of other issuers, should be subject to disclosure standards that require both
information at the time of the initial offering and subsequent regular reporting to
investors. The focus should be on the process and generic needs. The substantive
data needs in meeting such standards may best be left to self-regulatory bodies in the
market and to participants in individual transactions.
* Financial information needs to be reported in clear, common formats and promptly
after the close of the fiscal period. For debt monitoring purposes, reporting on a cash
or modified accrual basis are especially useful, as are cash flow statements. Audits
should be independent, recurring, and punctual.
* A central repository of financial information regarding governmental borrowers is a
useful tool in promoting efficient disclosure. Such a repository should also contain
information regarding the security pledged and liens against real and personal
property if those are not recorded elsewhere.
11.4 Policy Guides Regarding Credit Analysis, Credit Ratings, and Bond Insurance
* Credit analysis is a product of the credit market's operation and only becomes viable
when and where there are a variety of competing investments with differing risk and
reward characteristics. Where there are large numbers of passive investors and
82



transactions are diverse and numerous, individual investors will become reliant on the
opinions of specialists.
* Credit ratings are the leading form of institutionalized credit analysis and assist in
developing an active securities market by pooling skills to develop opinions. Ratings
play an important role in that they focus on credit risk (risk of payment delay or
default) which is then used to help judge risk and reward.
* Ratings have a positive side benefit of ranking governments on their perceived ability
and willingness to pay. The symbols are easily understood and lend popular appeal to
the otherwise lackluster but essential task of improving finances (as a means to
upgrading one's rating).
* Credit ratings can tend to centralize and dominate credit analysis and as proprietary
operations, the precise basis for ratings is not always clear. Reputations for
objectivity and freedom from influence (as well as being correct most of the time) are
necessary for credibility. The demand for ratings should derive from the market
itself. Ratings companies should compete. In domestic markets, it is not a good idea
to demand ratings of issuers as a matter regulation, to have "official rating agencies"
nor to set "standards" for ratings. In foreign market borrowings, aside from private
placements, international ratings are likely required by the market.
* Private sector bond insurance is of potential interest to subsovereign credits. The
insurance companies are somewhat active in emerging economies, although recent
events have dulled interest. With the possibility for diversification limited within
countries, commercial insurance may be slow to take hold. However, credit
assistance programs that have sufficient capital, that are market oriented and that use
insurance principles in determining charges may hasten development.
11.5 Policy Guides Regarding Credit Assistance and Financial Interventions
* Credit assistance, where provided, should be provided only to that level necessary to
permit a government to access the private credit markets. Governments should be
subject, to the extent that they can be, to the costs and demands of private credit at
least at the margin. Those that are too poor and too small to borrow in the market
should not be encouraged to borrow unless and until their underlying financial
situation makes that feasible.
* Direct lending, interest subsidies, guarantees, and insurance and other financial
assistance should be designed to provide incentives for governments to access the
markets on their own. Such programs should recognize differences in
creditworthiness and have those differences reflected in the rates of interest or the
amount of loans extended.
83



* Financial intermediaries that pool smaller loans into larger offerings such as found in
bond banks can provide economies of scale and give the opportunity for greater
liquidity in the secondary market. Such bond banks or pools can either be public
sector or private sector in sponsorship but should be essentially financial institutions
and subject to credit market discipline.
* Forms of intervention by senior levels in the case where there is a local government
financial emergency should be comprehensive and complete. They should provide for
creditor rights, remedies, and workouts, as well as financial recovery (or dissolution)
of the debtor unit itself. There should be the added flexibility in terms of making
specific pledges of security and remedies a matter of contract. But, it is best to have
in place a statutory framework to define rights, essential services, and the procedure
for recomposition of debt. Interventions should be rare and not be used as a backdoor
for the senior level of government to bail out local governments and their creditors.
84






The World Bank
Urban Development
1818 H Street. NW
Washington, DC 20433
USA
Internet: www.worldbank.org/urban