POLICY RESEARCH WORKING PAPER 1289
A FiscalNeeds Approach A simple framework for
to EqualizationTransfersin aggregate and sectoral public
Decentralized Federation assessing fi,cal
a DsecentralizedFederation needs
Anwar Shah determining fiscal
and regional fiscalinequities
in a decentralized federation.
P Lcy RESEARCH WORKING PAPER 1289
Shah reviews the conceptual basis for fiscal equalization Fisrai equalization transfers that reduce or eliminate
transfers, analyzes the theoretical implications for differentials in net fiscal benefits create a rare instance in
optimal design of equalization transfers, and suggests economics when considerations of equity and efficiency
quantitative approaches for assessing the fisral needs of coincide. These transfers must allow for differences in
subnational governments and determining their the spending needs and revenue-raising abilities of the
entitlement to equalization transfers. various subnational governments.
Shah illustrates proposed methods using data for local Shah argues for a two-tiered approach to equalizati
and provincial Canadian governments. The proposed The first tier would be a federal responsibility to equalize
methods could be useful tools, he says, for undertaking the burden of federal taxes.
systematic objective reviews of aggregate and sectoral The second ti - would be an interprovincial
public spending in developing countries, equalization fund to be administered by the Council of
Shah argues that in a decentralized federation, fiscal Provincial Finance Ministers. It wou!d entail a
inefficiencies and inequities arise because of subnational comprehensive equalization system that takes into
governments' differing levels of ability to provide account provincial fiscal capacities as well as provincial
comparable public services at comparable tax rates. spending needs. The standard of equalization would be
This paper - a product of the Public Economics Division, Policy Research Department - is part of a larger effort in
department to develop tools for analyzingpublic expenditures and policies to reform fiscal systems in developing countries.
Copies of the paper are available free frorr :.ie World Bank, 1818 H Street NW, Washington, DC 20433. Please contact
Carlina Jones, room N10-063, extension 37754 (35 pages). April 1994.
the to the
The Policy Research Working Paper Seriesdisseminates findings of vork in progress encourage exchange ideasabout of
development As, is are
issues. objectiveof the series to get thefindingsout quickly, evenif thepresentations lessthan fully polished.The
paperscarrythe namesofjhe authorsand should beusedand citedaccordingly.Thefindings,interpretations, are
and conclusions the
authors' oumand should jot beattributed to the World Bank,its ExecutiveBoardof Directors,or any of its membercountries.
Produced by the Policy Research Dissemination Center
A Fiscal Needs Approach
to Equalization Transfers
in a Decentralized Federation
Table of Contents
1.0 Introduction ............. 1
2.0 Conceptual Basis for Equalization .2
3.0 Implications of the Theory and the Constitutional Mandate
for an Optimal Design of Equalization Transfers. 3
4.0 Alternative Approaches to the Measurement of the Expenditure
Needs of Subnational Governments. 8
5.0 The Representative Expenditure System .13
6.0 An Alternative Approach to the Measurement of Expenditure Needs.16
Concluding Remarks .............. ....... ......................... 17
References .............. 32
An earlier version of the paper was presented at the Canadian EconomicAssociationMeetings in
Kingston, Ontario, Canada in June 1991. The author is grateful to Shabid Yusuf, Boris Plescovic,
(World Bank), Professor Robin Boadway (Queens University) and Mr. Douglas H. Clark,
(Ministry/Department of Finance, Govermnent of Canada) for helpful comments.
The existing structure of intereovernmentalfiscal relations in Canada has occasionallybeen
portrayed as a textbook exampleof an almost optimaldesign. The division of powers and transfers in
Canada are broadlyconsistentwith the prevalentviews on fiscal federalismand many countries around
the world including Russia, China and europeaneconomiesin transition are interestedin lessons from
the Canadianexperiencewith federalism.An area of foremostinterest in this respect is its equalization
program which has received rave reviews and has been described as a "glue that holds the federation
together" (Courchene, 1984)and advocatedas model for other countriesto adaptand follow (see Shah,
1991a, 1991c). The existing fiscal equalizationprogram, however, attempts to equalize per capita tax
burden alone and completelyignores the expenditureside. This paperexaminesthe consequences this
neglectboth conceptuallyas well as empiricallv.
The paper concludesthat a programof equalizationthat ignoresthe expenditureside cannot be
defended on economic efficiencyand equity grounds. The paper suggeststhat the present system of
equalizationin Canada couldbe improvedby explicitlyincorporatingrelative expenditureneeds of the
Canadianprovincesinto the formula. The paper further proposestwo alternate objectivemethodologies
to measure expenditureneeds of the Canadianprovincesand provides tentativecalculationsof the fiscal
needs of the Canadian provinces under a comprehensiveprogram of equalization.These calculations
amongprovinceswhichis significantly variancefrom
understandablyresultin an allocationof Lransfers at
the existing equalizationprogram. The paper further suggestsa two-tiered approachto equalizationin
Canada. The first tier would be a federal responsibilityand it would attempt to equalizethe burden of
federal income and commoditytaxesonly. The secondtier would be an interprovincial
to be administeredby the Councilof ProvincialFinanceMinistersand wouldcomprisea comprehensive
equalizationsystem that takes into account provincialfiscal capacitiesas well as expenditureneeds. The
standardof equalizationwould be dietermined negotiations.
2.0 Conceptual Basis For Equalization
A theoreticalcase for equalizationon equity and efficiencyground continuesto be clouded in
controversy. Proponents (see Buchanan 1950, 1952, Boadway 1980, 1990 and Boadway and Flatters
1991)have argued that equalizationtransfersby p .ttingequal fiscal treatmentof identicalpersons in
a federationpromotesequity. Suchtransfers by discouraoingfiscallyinducedmigrationand by enabling
Canadian provincesto provide certain minimumstandards of public services reduce barriers to factor
mobilityand therefore, enhance economicefficiency. Boadway,Roberts and Shah (1993) have argued
that equalizationtransfers that reduce net fiscal benefits differentialscreate one of those rare instances
in economicswhen equity and efficiencyconsiderationscoincide.
An opposingviewpointquestionsthe efficiencyand equitybasisof equalization.Scott(1952)and
Courchene(1978) have arguedthat equalizationpaymentsto correct fiscal inequitiesinduceinefficiency
in the regional allocationof resources. This happensbecausethe grants discouragethe outmigrationof
labor to high incomeregions where it would be more productive.Oates (1984) and Shah (1988, 1989b)
have argued that since capitalization taxes and expendituresis a pervasivephenomenon,the case for
equalizationtransfers to retard fiscally inducedmigrationis extremelytenuous. In the presence of full
capitalization,there may not be any efficiencyand equity basis for fiscal equalizationtransfers of a
general non-matchingvariety, because people in regions with fiscal surplusespay relatively more for
private services and less for public services, and vice versa for regions with fiscal deficiencies.
further ensuresthat the existinghome-owners
cannotavoidthe consequences local public
policies by movingout of the area. Since net benefits are capitalizedinto propertyvalues, a capital gain
or loss on accountof the local public sector is realizedat the time of the sale of property. This suggests
that in the presence of capitalization,Tiebout's prescriptionthat a system of local governmentswould
ensure optimal levels of local public servicesis not guaranteed(Shah, 1988, p.211).
Thus the role of equalizationto promoteefficiencyand equity in a federationand to replicatethe
advantages of a unitary form of government and at the same time permitting a fully decentralized
provision-:f public servicesis well recognizedin the literature. Yet as the aboveargumentssuggest,such
a case is by no means universallyacknowledgedeither in theory or in practice. For example, inspite of
widely divergent fiscal capacities of US states, there is no federal or interstate equalizationprogram
currently in place in the USA. Thus equalizationmust be recognizedas a matter of political taste. In
Canada, indeed, there is a strong political preferencefor such a program as indicatedby the enshrining
of a special equalizationprinciple in the 1982ConstitutionAct. Section36.(2) of the Act states:
"Parliamentand the Governmentof Canada are comiitted to the principle of
makingequalization goverrnents have sufficienst
pay.-ents to ensurethat provincial
revenuesto providereasonablycomparablelevels of publicservices at reasonably
comparablelevels of taxation."
The following sections reflect upon the design of an equalizationprogram that would be
consistentwith the theoreticalcase for equalization.
3.0 Implications of the Theory and the Constitutional MIandate For An Optimal
Design of Equalization Transfers
If one ignoresthe conceptualobjectionsto equalizationtransfersand insteadembracesthe prevalent
view that equalizationtransfers are justified on efficiencyand equity grounds,what design implications
EfficiencyBasis of Equalization: The literature on the implicationsof economictheory for an
optimaldesign of equalizationtransfers is quite limitedand this literatureis heavily influencedby Robin
Boadway'sviewson this subject.Boadway(1980),Boadwayand Flatters(1982, 1991),Boadway,Flatters
and LeBlanc(1983) and Auld and Eden (1984) among others have givensome thoughts to devisingan
equalizationprogram based on economictheory. Boadwayand Flatters (1982) on efficiencygrounds
advocatecompleteeliminationof differencesin net fiscal benefitsacross provinces.This they state calls
for an equalizationschemethat fully equalizesdifferences in both residence(income taxes) and source
(resource taxes, property taxes, etc.) based tax revenues.Au!d and Eden (1984) also conclude that a
revenue equalizationprogram would be consistentwith economictheory. These prescriptionsare open
to debate. For example, Boadwayand Flatters base their efficiencyargumentson differential net fiscal
benefits but the revenue equalizationschemJs they recommendwould tend to equalize per capita tax
burden alone. Auld and Eden also implicitlyassume that equal tax treatmentof equals is equivalentto
equal fiscal treatmentof equals. Courcheneand Copplestone(1980) rightly point out that:
"..if efficiencyis utilized as the guidingprinciplein reallocatingrevenues, then should it not be the
case that the efficiencycriterion be carried over to the productionor expenditureside as well?"(p.1JO)
It is refreshing to note that as early as 1959, Musgravehad a vivid perception of the problem.
Though not unalterablyopposedto a pure revenue equalizationformula, he argued that such a formula
would enable a provincial government to vote additional services while assuming a fraction of the
increasedcost, thereby encouragingfiscal irresponsibility. rectifythis, he suggests, "thatthe principle
of equal tax treatmentof equalsbe replacedby a new rule accordingto which people with equal income
shouldexperiencethe same fiscal residue(imputedbenefitsminus tax costs) or net benefit derived from
budgetoperations(1959, p. 182).Only an equalizationschemethat considersboth the differentialrevenue
capacities and expenditureneeds would be consistentwith this view. Subsequenttheoretical work of
Musgrave(1961), Le Grand (1975)and McMillan(1981)is also supportiveof this view. Australiais the
only federalcountrywhic'n adoptedsucha comprehensive seriouslyflawedin its implementation)
approachto fiscal equalization.
Horizontal Fiscal Imbalance and Equalization: This view recognizes the imbalance between
expenditureneeds and revenuemeans amongprovincialgovernmentsof Canadaand calls for the tederal
governmentto providetransfersto the provinceswith relativelylow fiscal capacityand/or relativelyhigh
fiscal needs. The idea is to ensure that every Canadianhds access to reasonablycomparable levels of
public serviceswithin his chosen localityat a cost in line with what he would pay elsewhere.This is the
so-called 'fiscal need' principlewhich permeatesthe writings of the Rowel-SiroisCommission(1939),
Hanson(1961), Graham (1964, 1980), Sharp(1966), Clark (1983), Courcheneand Copplestone(1980),
Courchen! (1983) and Shah (1983a, 1983b, 1984a, 1984c). This is also the view enshrined in the
These authors have argued that equalizationtransfers
Canadian ConsticutionAct 1982, Section 111.36.
should considerboth the expenditurcneeds and the revenuemeans of the provincesin determiningtheir
equalizationentitlements. Thus the existing representativetax system should be complementedby a
Horizontal Equity and Equalization: Horizont:' equity refers to the principle that equals should
be treated equally. Buchanan(1950) extendedthis conce.t to fiscal federalismto mean that the federal
governmentshould ensure that all its citizensare treated equallyunder the total fisc regardlessof their
place of residence. This is termed as the "broad based" view of fiscal equity by Boadwayand Flatters
(1982) and the EconomicCouncil of Canada (1982). Accordingto this view the fiscal system should be
horizontallvequitablenationwidein termsof the actionsof all governments,federal, provincialand local.
rhus individualswith the samemarket incomein the absenceof the public sector should haveequal real
incomes afterwards. In other words, the fiscal system should be locationallyneutral. The Canadian
residents with similar incomes across provinces should receive the same net benefits (imputed public
servicebenefitsless tax costs). Onlyan equalizationprogramwhichtakesaccount of both the expenditure
needs and revenue means of provinceswouldbe consistentwith this broad based view.
A somewhatdifferent view of fiscal equity is called the narrow-based"view by Boadway and
Flatters (1982). This view proposesthat the federal go:.wnmentshould take as a starting point the level
of real incomeattainedby personsafter provincial(and perhapslocal) fiscal impactshavebeen accounted
for. The federal government,accordingto this view, shouldonly be concernedwith equal fiscal treatment
of Canadian residents with cespect to its own actions alon;. Boadway and Flatters argue that in the
Canadian context, a "narrow based" view of horizontalfiscal equity is more appropriate because the
CanadianConstitutionvests ownershiprights of natural resources with the provinces.Accordingto this
view, fiscal inequityarises only becausethe federalgovernmentdefinespersonalincomefor tax purposes
differently'froma household'scorn-rehensivereal (or true) income.To correct for this inequity,Boadway
ar.d Flatters (1982) recommendthat only differences in net fiscal benefits multiplied by the federal
average marglnalincometax rate be equlalized.
This accordingto Boadway,Flatters and LeBlanc(1983,
p.178) would require (partial)equalizing22.5% and 45.75% of source based taxes and full equalization
of other federal taxe, The lower limiton resourcetax equalization obtained if the federal government
is assumedto have no propertyrights and the upper limitif it claims30% -roperty rights. Auld and Eden
(1984) also agree with this narrow view of horizontalequity but concludethat it implies that there be
completeequalizationof all provincial-localrevenuesbut only partial equalizationof natural resource
rents. They, however, perceive long term adverseconsequencesfor nationalunity of such a program
of partial equalization.
Boadway and Flatters (1982) and Auld and Eden (1984) implicitlyassume that public sector
benefits per household are equal across all provinces - an untenableassumption.If this assumptionis
relaxed, it does not follow that partial revenueequalizationalone would sufficeto ensure fiscal equity
consistent with the "narrow-based"view. Partial expenditureequalizationalso would be required to
partially equalize net fiscal benefits.
In conclusion,economictheory suggeststhat an equalizationprogramshould attempt to equalize
net fiscal benefits across provinces.A pure revenueequalizationsystemalone is not likely to accomplish
such an objectiveor even a more modestone of ensur.ng"reasonablycomparablelevelsof publicservices
at reasonablycomparablelevelsof taxation"to Canadianresidentsat large. Optimalityrequiresth._tthe
equalizationformula should take into account both the revenueand the expendituresides of provincial
local budgetaryoperations.The existingprogramof equalizationin C. , couldtherefore, be improved
by explicitly incorporatingexpenditureneed into the formula. Expenditureneed equalizationwould also
help overcomesomeserious limitations the existingprogramnotedby leadingacademics.For example,
Professor Courchene(1984, pp.405-406)notes that:
"...while the potential exists for equalization to enhance economic efficiency, it seems highly
unlikely that any of Canada's recent equalizationprograms could be defended on efficiency
grounds.Indeedit wouldappearthatRFPS (Representative Five Provim.e Standard)likethe RNAS
(RepresentativeNational AverageStandard)before it, inhibitstiheoptimaloutmigrationfrom the
eastern provincesto Ontarioand encouragesexcess migrationfrom Ontarioto energy producing
Finally, under a fiscalneed approach,equalizationwouldassume an overarchingrole in federal--
provincialfiscal arrangements.It would be a residualprogram incorporatingany changesin the funding
of other federal programs.
4.0 Alternative Approaches to the Measurement of the Expenditure Needs oT
Expenditureneed have craditionally
beenthoughtto be muchmoredifficultto defineand measure
than is its revenue equivalent, fiscal capacity. There is no doubt that in Canada the difficulties oi
measurementare greater for expenditureneeds than for fiscalcapacity.Thesedifficulties,however, may
be overstated. Break (1980) referring to the United States, writes that the difficulties involved in
measuringexpenditureneedare aboutthe sameas those encounteredin using a represfntativetax system
to measure fiscal capacity. BreaKstates that in order to measure expenditureneed, it is necessary to
define an equalization standard; to determine differential costs due to differences in input-output
relationships,nrtlre of serviceareas and the compositionof population;and to distinguishbetweenthose
need/costdifferentialsthat are due to differentialtastesor inherentcost disabilitiesand those .hat are due
Lopolicy decisions.It may also be necessaryto identifyand measuredifferentialsthat are attributableto
strategic behavioron the part of provincesin respect of federal transfer payments. These steps will, in
some cases, involve formidabledifficulties.However, other federations, includingAustralia, Germany
and Switzerland are attemptingto address at least some of these problems in their own equalization
programs. The need to do so in Canada is increasedby the explicitreference to "reasonablycomparable
levels of public services" that is containedin subsection36(2) of the ConstitutionAct, 1982which sets
out the commitmentof the Governmentof Canadato the principleof makingequalizationpayments.
A number of Canadianwriters have concernedthemselveswith the measurementof expenditure
needs. These includeHanson(1961), Clark (1969,1983),and Courchene(1984). The variousapproaches
proposed by these authors are summarizedin Shah (1984a). These authors, in general, and Courchene
in particular, showa vividperceptionof variousconceptualstepsinvolvedvet theyfail to developfeasible
and objectiveoperationalapproaches.Similarly,fiscal need approachesused by the Canadianprovinces,
Australia, Germany and Switzerlandwhile interestingdo not hold any promise of satisfyingthe fiscal
manner. These approachesare briefly reviewedhere and
need criteria in an objectiveyet cotmprehensive
a reader is invited to refer to Shah (1983c, 1984a) for formulae and technical details and critical
The Australian Apporoach: Of all federal countries, Australia is best noted for its balanced
emphasis on expenditure need and revenue means factors in determining state relativities for the
distribution of lnconditional equalization transfers. Shah (1983c, 1984a) provides details of the
methodologyadoptedby the AustralianGrantsCommissionin assessingneeds of the memberstates. Very
briefly, the Commissionmeasureseach state's expenditureneeds for a serviceor category of expenditure
by calculatingthe differentialcost, for the state whose needs are being assessed, of providingservices
of a standardlevel, range and quality as determinedby the Commissioners examiningrelevant data
composition,populationdensity, urbanizationand physical
and making field visits. Socio-demographic
environmentfigure prominentlyin assessingdifferentialcosts. A state's expenditureneed as measured
by these procedurescould be either positiveor negative.
The Australianapproach is commendablefor its comprehensiveness the proceduresused for
the assessmentof both revenue and expenditureneed appear to be somewhatcrude, imprecise and
subjective. Expenditure equalization appear to place too much reliance on demographic factors.
Determinants expenditureneed are arr-vedat usingbroad judgementrather than any hard quantitative
analysis. The procedure involvesa detailedanalysisof budgetarydata and then subjectiveassessmentof
relative need, following written and oral arguments about principles and methods in adversary
proceedings. The process adopted is unnecessarilycumbersome,unduly time consumingand places too
much reliance on broad judgement. Such a process could only work if an atmosphereof exceptionally
high degree of compromise,cooperationand accomrnodation
- 10 -
Therefore, while the Australianshave done a great deal of work over a period of many years it seems
unlikely that their approachis suitablefor applicationin Canada.
Germanyand Switzerland:Germanyand Switzerlandalso incorporateexpenditureneeds in fiscal
equalizationbut their approachesare not nearly as comprehensive the Australianmodel. In Germany,
average nation-widetax revenue per capita is used as the proxy for expenditureneed for each lander.
Adjustmentto this measure are then made to reflect the presence of "specialburdens". In local fiscal
equalization, many types of burdeps are taken into considerationbut at the lander level only two
modifications of the average per capita figure are used, both of which, however, relate to local
populationsize adjustmentis done to recognizethe higher per capita
government. A local governrment
cost for larger communities(local governments).A graduated population density adjustment is also
havebeen applied, "tax-need"figures are calculatedseparatelyfor
applied. After these two modifications
lander taxes, and for the local taxesto the extent they are included.Taken together they form the lander
tax-need indicator.This indicatoris then comparedto tax potentialindicatorto determinefiscal surplus
or deficiencyof a state. Actualequalizationentitlementsare then determinedwith referenceto a standard
equal to a fraction (92 percent in recent years) of the nationalaverage per capita.
Swissformulafor expenditureneedconsiderspopulationdensity,cultivatablesurfacearea, surface
area cultivatablein mountainousregions, surface area economicallyproductive.The formula recognizes
regions. It assumesthat the costs
specialneeds due to higherservicingcosts of cantonswith mountainous
of providingminimum standardsof public services is higher the larger the mountainouszone and the
lower the populationdensity, hence it provides compensationfor these two factors. As fiscal needs are
assumedto vary inverselywith the compositeindex of density, the indexis constructed in such a way
that it stays constantfor densitylevels higher than the nationalaverage.
Canadian Provinces: The Canadianprovincesuse simple measuresof expenditureneed in their
general purposetransfers to municipalities Table 1). The most sophisticated these approachesis
- 11 -
the one taken by Saskatchewan.In that province, the standard municipal expenditure of a class of
municipalitiesis assumedto be a functionof total populationof the class and regressionanalysisis used
to derive a graduatedstandardper capitaexpendituretable for municipalgovernmentsby populationclass.
developedby the Canadianprovincesfor measuringexpenditureneeds do
In general, the methodologies
not take detailed account of differencesamong municipalities the cost of and need for such services.
However, they do indicate an emphasison the expenditureside of public finance, which is at present,
lackingin the federal government'sfiscal equalizationprogram. It is doubtfulthat the expenditureneed
criteria used by the provincescould be used in the federal programbecausethey relate to only a subset
of public servicesand becausethey relate to such a large numberof governmentalunits (as comparedto
ten units in the federal prograrn) that the task of developingrealisticmeasures is much more difficult.
It may, neverthelessbe instructiveto list the variousexpenditureneed elementsused by variousprovinces
grants among municipalities.These include;
in order to allocate their general purpose (unconditional)
population size, populationdensity, population growth factors, road length, dwelling units, location
factors(e.g. northernlocation),urbanizationfactors(primaryurban populationand urban/ruralclass) and
social assistancepayments(see Shah 1983c).
Courchene (1984): Professor Courchene has over the years made many eloquent pleas for
introducinga fiscal need approach to equalizationin Canada. He has argued that:(a) It costs more per
capitato financt some programs in certain provincesthan in others. Hencea pure equalizationprogram
would not ensure comparablelevels of services across provinces; (b) A constitutionalchallengeto the
existingprogram could be mountedon the groundsthat it does not take expenditureneeds into account
and, therefore,does not ensure that provinceshavesufficientrevenuesto providereasonablycomparable
levels of public services; and (c) A fiscal need approach would automaticallytake into account any
changes in funding of other federal programs. This is because under an expenditure approach,
- 12 -
equalizationwouldbe a residualprogramincorporating alterationelsewherein the federal-provincial
finance area and ensuring that provincialfinances come up to the agreed upon level. The following
expressionfor an expenditurerelated equalizationprogram by Courchene(1984) illustratesthis point:
Per Capita EqualizationFor Provincei
Proxy for per capita expenditureneed
Proxy for province i's ability in per capita terms to financepublic services
All federal transfers to province i
Courchene has argued that an equalizationprogram designed along the lines of the above
fiscal arrangements.This he stated
expressionwouldbecomethe overarchingaspectof federal-provincial
was not true of the existing program. Under the existingprogram, Ottawa can increase or decrease its
share of funding for the establishedprograms or the Canada AssistancePlan and this will not have an
impact oniequalizationentitlements.He said:
"In my views, the expenditureapproach,incorporating it does the rest of the interprovincial
financial interface, might be preferred to the existingapproaches, which treat equalizationin
considerable,or even complete,isolationfrom other federal-provincial transfers" (p.273)
The abovediscussionindicatesthat whilethe felt needfor a fiscal needapproachto equalization
is quite strong, only limitedoperationalguidanceis availableto Canada if at sometimesin the future it
wishedto complementexistingrepresentative system(RTS)with a systemthat consideredexpenditure
need as well. Clark (1983) and Courchene (1984) have advocatedthe desirability of developing a
representativeexpendituresystemthat wouldcomplementthe existingRTS yet they did not implementor
- 13 -
evenfully elaboratehow such a systemwouldwork. Shah(1984a, 1984c)sketchedout details of possible
approachesto this issue as well as presentedsome preliminaryresults. The followingsectionscarry the
work initiated by Shah further and elaborate and implementtwo alternate empirical approachesto the
measurementof expenditureneedsof the Canadianprovinces.It must be emphasized the calculations
presented here are extremelytentativeas these are based on data that is dated yet they serve important
objectives.They demonstrate technologydoesexistto measureexpenditureneedsobjectivelyand that
the redistributivethrust of a comprehensiveequalizationsystem is likely to be very different from the
5.0 The Representative Expenditure System
A representative expenditure system (hereafter cal!ed RES) takes the consolidated
provincial-localexpendituresdisaggregatedinto major functionsas the basis for calculatingdifferential
expendituie needs. More specifically, the following steps are involved in making expenditure need
(1) functionsinto major categories
e.g. transportationand communications;
(2) Identificationof significantdeterminants(Need/costfactors)
(3) Quantificationof differential contributionof these factors in explainingthe expenditure
(1) Use results in Step I to determinerelative factor weights.
(2) Compute an index using these weights to redistributea proportion of total expenditures.
The proportionis determinedbased on a percentageof total variance explainedby given
- 14 -
Derivationof hypotheticalexpenditureby province for each categoryof servicesbased on
the indicescomputed in Step II.
HypotheticalExpenditureon a subfunctionobtained in Step III
National Average WeightedExpenditurePer Capita For That Subfunctionfor the Year
Empirical Implementation the RepresentativeExpenditureSystem: The RES is implemented
here by examiningeight categoriesof consolidated expendituresfor the period
1971-1981for each of the ten provinces.The followingbreakdownof these expenditureswas used:
1. Transportationand Communications
2. Social Services
6. Elementaryand SecondaryEducation
7. General Services
8. Other Expenditures
For each of these categoriesof public services,time series and cross-sectiondata on relevant
need/cost factors and other determinantswere assembled.Regressionanalysiswas then used to identify
significantfactors. For this purposes,Kmenta'spoolingregressionswere estimatedusing iterativesystems
- 15 -
of simultaneous equations estimationmethod. The regression results are reported in Table 2. This
procedure resulted in an extremelygood overall fit for the system of equationsas a whole (Adjusted R-
Square=0.9997, Log-likelihoodFunction=-4169.3 and Chi-Square = 307.9 With 63 D.F.) as well as
good fit for individualregressions.These regression results are then used to assign relative weights to
various identifiedfactors (see Table 3) and also to computea compositeindexof relativeneed to be used
for developing standardizedexpendituresfor each province for the categoriesof public expenditures
examinedhere. These indicesalong with relativeweights assignedby regressionanalysisto factors that
comprise such indices are reported in Table 3. It is interestingto note that for transportation and
educationservices regionallydifferentiatedprice indices receive major weights whereas for health and
police protection, proportion of people in urban and/or metropolitan areas explain much of the
interprovincialvariation in expenditures. The factor weights specified in Table 3 are then used to
redistributethe proportionof total all provinces expendituresthat are explainedby these factors. Per
capita hypothetical standardized expenditure so derived is then compared to national per capita
expendituresto determineper capita needs for the specifiedservice. The samne
for all the categoriesof expendituresexaminedand aggregateneeds for each province is derived from
a summationof thesepositiveand negativeentitlements multiplyingby the relevantpopulationfigures
(see Table 4). In Table 5, expenditureneeds are updated for the fiscal year 1991-92by assuming a
growth rate for total entitlements
consistentwith the growthrate of paymentsfrom the existingprogram.
This table suggeststhat if a comprehensiveprogram of equalizationto be administeredby the federal
governmentwas institutedin the Year 1991-92,its redistributiveimpactwouldbe in favor of Quebecand
againstthe maritime provincesand Manitobaand Saskatchewan.The total equalizationpaymentsby the
federal governmentunder such a comprehensivesystem are estimatedto be smaller by almost a billion
dollar than the paymentsassociatedwith the current representativetax system.
- 16 -
6.0 An Alternate Approach To the Measurement of Expenditure Needs
The RES approach implementedabove presents itself as a reasonablyobjectiveprocedure to
determinefiscal needs of the Canadianprovinces.An econometricapproachbeing proposedhere presents
even a rmorestraightforwardand objectivealternative.The calculationof expenditureneeds based on this
approaih entails the followingsteps:
Step I (Sameas the R.E.S)
(1) Identification significantfactors (Need/cost)
Differentialcontributionof these factors in explainingprovincial-local
Evaluationof regression results for each provinceby holdingthe fiscal capacity and non--
need factors at national average values and substitutingactual values of the need/cost
variables. The resultingfigures would be the hypotheticalper capita expenditurefor each
Evaluateregression results at mean values for all variables(need as well as fiscal capacity
etc.). This wouldbe the standardizedexpenditurefor each subfunction.
Hypotheticalper capitaexpenditure(Step II)
Standardized capita expenditure(Step III)
Table 6 reports these resultsfor selectedcategoriesof expenditurein per capitaterms for the fiscal
year 1981-82. The same results are projected for 1991-92in Table 7 and combinedwith the RTS to
determineoverall entitlementsfor each of the provinces.Note that Quebec, Ontario,Alberta and British
Columbia qualifyfor pos.iive expenditureneed entitlementsand all other provincesqualifyfor negative
under the RTS and the expenditureneeds
expenditureneed entitlements.The summationof entitlements
- 17 -
under the RTS and the expenditureneeds
expenditureneed entitlements.The summationof entitlements
results in net negativeentitlementsfor Ontario, Alberta and B.C. only. These negativeentitlementsfor
a federal program would be ignored. Overall, the proposed system redistributestransfers in favor of
Quebecand againstother equalization for
receivingprovinces.Overallentitlements a federalprogramare
estimatedto be higher than the existing program. Finally, Table 8 presentscalculationsof relative costs
(costsbased on need as opposedto actualcosts)of publicservicesbased on the assumptionthat each and
every provinceis equalizedeither upwardsand downwardsto nationalaveragefiscal capacity. The table
suggeststhat the need related required cost of provisionof public servicesin the equalizationreceiving
provinces except Quebec are uniformly lower than national average. This again reconfirms earlier
conclusionsthat a pure revenue equalizationprogram would be more favorableto these provincesthan
A word of caution is in order here. The calculationspresentedhere are based on a comprehensive
yet a dated data set. Quite conceivablywhen the samedata set is extendedto 1991-92,there might well
emerge significantchangesin regressioncoefficientsand the resultingentitlements.
The paper has presented two simple yet objectiveapproachesto the measurementof the fiscal
the ease of such calculations.In
needs of the Canadianprovinces and also demonstrated computational
doing so, the paper has made a case for a re-examination the existingequalizationsystem, a case also
supportedby Courchene(1984) as follows:
"Even if a fiscalneed approachis a non-starterfor political,conceptualor computational
the notion that equalizationcould play an overarchingrole in terms of balancingthe provinces'
expenditureneeds and revenuemeans merits some consideration"(p.273).
- 18 -
If a serious re-examination the existingprogram of the type discussedin this paper takes place
in future, it should also address whether or not a fiscal equalizationprogram should remain a federal
responsibility or instead become a shared responsibility. An ideal equalization program in most
federationswould be an interstatefiscal needequalizationfund whichassessesboth negativeand positive
equalizationgrants to memberstatessuch thatnet transfersequalzero. Thus the programby d^^.gnwould
be self-financing.In Canada, though due to a long tradition of federal involvementin equalization,
perhaps a preferredalternativewouldbe a two-tieredprogram- a federaland an interprovincial
also Courchene 1984). The first tier would consist of a federal program based on an acceptance of a
narrower than the "narrow-based"view of equity as proposedby Boadwayet al.(1982, 1983). Such a
program would limit federal role to only an area of its strict responsibilityi.e. to equalizethe burden of
federal income and commoditytaxes only either through tax credits to individuals or through direct
equalizationtransfers to provincialgovernments.The secondtier wouldbe an inter-provincial
equalizationfund, to be administeredby the provincesthemselves,and would assess both negativeand
positive equalizationtransfers to provinces such that net transfers equal zero. The latter program is
expectedto bring a greater sense of participationamongprovincesin the federationas the contributions
of have-provincesto have-notprovinceswould be transparent. Such a program would be self financing
and would also eliminatepressures for increasesin equalizationpaymentswhich a federal program by
its very nature cannot avoid. The proposedtwo tier structure would be consistent with the conceptual
basis of equalizationand would afford a built-in mechanismfor enforcinga disciplineon equalization
payments as two opposing political influences would work to determine the level of equalization
payments.The proposedprogramwouldfurther reducethe financialsqueezeon the federal budgetplaced
by the current system of equalizationpayments.
- 19 -
The existingequalizationsystem separatestaxingand spendingdecisionsin maritimeprovincesin
a major way and therefore eliminatesa measureof local accountabilityand might even create incentives
for fiscal mismanagement.The empirica; evidencein Shah (1984c) indicatesthat the relative costs of
public employmentin the equalizationreceivingprovinceswas nine percentagepoints higher than the
same in the "have" provincesin 1982(average indexof 1.12 in the "have not" provincesvs 1.03 in the
wealthy provinces (Shah, 1984c, Table 5.3)). Could it be that part of the equalizationfunds are being
used by the recipientprovinces to provide higher wages for the public service employeesrather than
improvedprovision of public services?This questionmerits further study.
- 20 -
Table 1: Canada: Basis of Provincial Unconditional Assistance to Local
Capacity Tax Effort Expenditure
Province Factors Factors Needs Factors
Newfoundland Loss of revenue with property tax - population
respect to exemptions revenues and - per capita
provided to old age water and assistance
pensioners sewers rates - road mileage
Prince Edward Island ---------------------------------see New Brunswick-------------------------------------
Nova Scotia property assessment -- - dwelling units
New Brunswick property assessment -- - shareable
(per capita and per expenditures
Quebec property assessment taxes from
Ontario property assessment previous year's - population
net levy - population
- municipal grouping
Manitoba -- -- - population
- urban population
Saskatchewan property assessment -- - population
- expenditure by
Alberta property assessment total tax - population
revenues growth in
excess of 5 %
British property assessment - population
Columbia - expenditure
Source: Shah (1983a)
- 21 -
Table 2: Pooling Regressions
(110 Observations - t ratios are given in parenthesis)
(1) TC = -397.82 + 7488.7 RSPC + 267.1 NCAR + 0.264 SNOW
(-4.5) (5.5) (4.3) (4.2)
+ .00399 PMA + 0.195 AWW + 410.81 TRUCK
(1.2) (1.0) (3.2)
+ .10733 TT + .0815 OR - 13.3 TIME
(3.7) (7.4) (-2.8)
+ .00417 GDP
(1.4) = 0.88
(2) SS = - 246.6 + 2.558 FLFPR + 5.8 IBR + .0004 SPF
(-4.2) (2.2) (5.2) (5.3)
+ .04 BP + 1953.8 DD + 965.37 GIS
(2.4) (0.4) (2.2)
- .003 OR - .156 UT + 1.8 WELT
(-3.4) (-6.4) (10.0)
+ 1.3 PGIP + 21.2 TIME
(.20) (7.8) R = 0.92
(3) HE = -21.3 + .05 P65 + 2.6 BIRTHS + 37.6 HSEP
(-0.6) (0.4) (2.7) (2.4)
+ 2318.0 TA + 0.228 HT + 0.019 OR
(0.8) (1.3) (1.5)
+ 34.0 TIME + 0.009 GDP - 45.5 NB
(10.7) (2.7) (-4.1)
- 169.5 QUE - 248.7 ONT - 46.4 SASK
(-2.4) (-2.5) (-3.5)
- 64.2 ALTA - 43.4 BC
(-2.7) (-1.9) k2 = 0.95
- 22 -
Table 2: (Cont'd)PoolingRegressions
(110 Observaticns- t ratiosare given in parenthesis)
(4) PPP = -2.2 + 0.03 POPN + 2751.1 TA + .003 GNP
(-0.3) (5.7) (3.1) (2.9)
0.015 TT + .013 OR + 0.14 TIME
(-1.7) (3.6) (6.9) 94
2 = 0.
(5) PSE = -7.3 + 3.7 MDNAS + 0.7 EPI + 0.013 OR
(-1.1 (4.6) (8.9) (4.2)
- 0.104 PSAT - 14.4 NB - 22.2 QUE
(-1.1) (-3.2) (-2.0)
- 81.9 ONT - 19.5 SAS + 27.2 ALTA
(-4.0) (-3.8) (3.4)
- 26.3 BC
(4.9) R2 = 0.91
(6) ESE = 6.64 + 0.015 P517 + 1.5 EPI + .04 OR
(0.3) (2.5) (4.4) (5.0)
+ 0.051 UT + 5.7 TIME
(1.7) (1.2) R2 = 0.92
(7) GS = -22.7 + 0.04 AWW + 0.18 PD + 0.07 TT
(-1.9) (4.0) (3.0) (5.6)
+ 0.06 OR + 63.8 QUE
(9.5) (6.7) k 2 = 0.79
(8) OE = 63.5 + 0.125 OR + 47.9 TIME
(2.6) (7.6) (10.2) k2 = 0.81
- 23 -
Table 2 (Cont'd): VariableDefinitions
AWW Averagewee!4ywages - industrialcomposite($ per person)
BIRTHS Births (000)
DD Persons claimingdisabilitydeduction(Tax Returns Data)
EPI EducationPrice Index (1981 - 1.00)
expenditureson elementaryand secondaryeducation($ per capita).
FLFPR Female Labor Force ParticipationRate (%)
GDP Provincialgross domesticproduct($ per capita)
- a proxy for property values
GIS G.I.S. recipients
expenditureson GeneralServices($ per capita).
HE Provincial-localExpenditureson Health Care ($ per capita).
HSEP Number of days of hospitalseparationsfrom all causes (000)
HT Health transfers ($ per person)
IBR Number of illegitimatebirths per thousandsunmarriedfemales of child-bearingage
MDNAS Full-timeenrollmentIn medicine,dentistry, nursing and applied sciences.
NB New Brunswick
NCAR Proportionof area that is non-cultivatable
OR revenuesfrom own sources ($ per capita)
- 24 -
Table 2 Cont'd: VariableDefinitions(Cont'd).
PGIP Provincialgovernmentin power by political affiliation
(Dummy = 0 for PCS, SOCREDS
I for liberal, NDP)
P517 Population5-17 years (000)
PMA Populationin metropolitanareas (000)
POPM Male populationin age group 16-24 (000)
PPP expenditureson protectionof persons and property($ per capita).
PSE Provincial-localExpenditureson post-secondaryeducation($ per capita).
PSET cash and tax transfers for post-secondary
Federal-provincial education($ per capita)
P65 Population65 years and over (000)
Dummy for the provinceof Quebec
RSPC Paved roads and streets kilometerageper capita
SPF Singleparent families(000)
SNOW Snowfall(in centimeters)
SS expenditureson social servicesand welfare($ per capita).
TA Number of traffic accidentsinvolvinginjury or death
TC expenditureson transportationand communication per capita).
TIME Time trend
TRUCK vehicles registrations(per capita)
TT transfers ($ per capita)
UT transfers ($ per capita)
WELT welfare transfers ($ per capita)
- 25 -
Table 3: Factor Weights for the R.E.S.
& Snowfall (Annual- in centimeters)SNOW 0.1020
HighwayConstruction Price Index(HCPI) 0.6580
Pavedroads and streetsper square
kilometer area (RSPR)
Non-cultivatable as a proportion
of totalarea (NCAR) 0.2357
Index=(0.10*ISNOW+0.66*IHCPI+ 0.0005*IRSPR 0.24*INCAR)*ISRP
Post-Secondary Full time enrollment grade 13+(000)(PSS)
Percentageof Population havinga minority
languageas mothertongue(ML) 0.190
Unemployment (UR) 0.018
Education Price Index(EPI) 0.717
Help Wanted Index(HWI) 0.010
ForeignPost-Secondary Students(FPS) 0.017
Index=(0.18*IPSS .70*IML+ .08*IUR+ .04*IFPS)*IHWI*IEPI
and Populationunder 18(PO17) 0.014
Education(ESE) PopulationDensity(PD) 0.017
EducationPrice Index(EPI) 0.969
Index = (.02*IPD+ .98*IEPI)*IP017
Health (HE) Alcoholism for
Index=(0.123*IALCO + 0.877*IPU)
- 26 -
Table 3 Cont'd: Factor Weights for the R.E.S.
Expenditure Category Need/Cost Factors Relative Weights
Social Services (SS) Single Parent Families (SPF) 1.00
Police Protection Criminal Code Offenses (CCO) 0.39
Proportion of Population in Metropolitan 0.61
Index=(.39*ICCO + .61*IPMAR)
General Services (GS) Private sector wages (Industrial 0.769
Percentage of population having a minority 0.001
language as mother tongue (ML)
Population Density (PD) 0.023
Population (POPF) 0.039
Snowfall (Annual - in centimeters) (SNOW) 0.168
Index=(.001*ML + 0.175*ISNOW + 0.80*IAMW + .024*IPD)*IPOPF
Note: Calculations based on regression coefficients.The use of a variable prefixed by I means that a relative index of the variable
- 27 -
Table 4: Expenditure
For the 1981-82Fiscal Year
NFLD P.E.I. N.S. N.B. QUE. ONT. MAN. SASK. ALTA. B.C.
Transportation 26460 4264 23192 30523 -33386 -81929 27956 35075 -24094 17117
Social Services -10420 -2698 -23879 -33674 124746 112291 -7777 5773 -67251 -97199
Health Services -104572 -23909 -138270 -121823 299878 812901 -154874 -141592 -210900 -216861
Protection of Persons
and Property -28375 -5966 -35036 -25755 -33738 111284 -18075 -12932 29100 19492
Education -28407 -6570 -37575 -32907 30707 348366 -46082 45323 -74544 - i07576
Elementary and Secondary
Education 6885 360 -8900 -6720 172915 -106613 -19465 -6679 -28220 -3032
General Services -1084 -148 -299 -598 -2331 1974 -2508 -1476 2297 4222
Total -160560 -39406 -246285 -212822 593500 1398305 -277572 -199288 419668 436142
- 28 -
Entitlementsfor the 1991-92Fiscal Year (Millions$)
Table 5: Equalization
BasedUpon a Comprehensive EqualizationSystem I
EntitlementsUnder EntitlementsUnder EntitlementsUnder a
the ExistingRepresentative the ProposedRepresentative Fiscal Need ApproachI
Tax System (RTS) ExpenditureSystem(RES) (RTS + RES)
(a) (b) (c) = (a) + (b)
Newfoundland 972 -311 661
Prince Edward Island 211 -75 136
Nova Scotia 970 -475 494
NewBrunswick 939 -409 529
Quebec 3899 1145 5044
Ontario -5022 2701 -2321
Manitoba 958 -537 421
Saskatchewan 517 -384 133
Alberta -4321 -811 -5132
BritishColumbia -890 -842 -1732
Total 8466 0 7418
Source: Author's calculations.
- 29 -
Table 6: Expenditure Neeos Based on Econometric Analysis
Elementary Post- Transportations Total
Police General & Secondary Secondary and Social Health per
Protection Services Education Education Communications Services Services Capita
Newfoundland -22 5 12 23 -16 -43 -23 -63
Price Edwards Islands -28 -9 26 13 -2 -47 -54 -100
Nova Scotia -5 -10 1 2 0 -38 -27 -76
New Brunswick -11 -10 -6 -2 13 -40 -33 -87
Quebec 20 3 28 14 11 35 33 145
Ontario 15 -5 -22 -16 -3 52 54 75
Manitoba 13 -23 -29 -4 -3 20 -2 -29
Saskatchewan -11 -19 -13 -19 12 21 -21 -50
Alberta 3 1 -19 26 13 -25 11 11
British Columbia 9 -6 -6 -26 -8 44 22 30
Table 7: Equalization Entitlements for the 1991-92Fiscal Year (Milion $)
Based Upon a ComprehensiveEqualization System II
Entitlements ExpenditureNeed EntitlementsUnder
the ExistingRepresentative EntitlementsUsingEconometric a Fiscal Need
Tax System(RTS) Analysis(EA) (RTS+EA) ApproachII
(a) (b) (c) = (a) + (b)
Newfoundland 972 -69 903
Prince EdwardsIslands 211 -25 186
Nova Scotia 970 -130 840
New Brunswick 939 -120 818
Quebec 3899 1888 5186
Ontario -5022 1436 -3586
Manitoba 958 -61 897
Saskatchewan 517 -94 423
Alberta -4321 53 -4268
BritishColumbia -890 183 -706
TOTAL* 8466 0 9253
*Ignoresnegativeentitlements (a) and (b).
Source: Author's caiculations.
- 31 -
Table 8: Indexes of Relative Costs (Needs) for Public Services for
If Each and Every Province Had the National Average Fiscal Capacity
For the 1981-82 Fiscal Year
NFLD P.E.I N.S. N.B. QUE. ONT MAN. SASK. ALTA. B.C.
Transportation 109.15 105.48 103.16 108.29 93.02 91.67 95.48 105.92 91.28 96.47
Social Services and
Welfare 99.24 98.36 96.89 92.07 108.25 106.73 101.81 105.05 96.44 95.16
Health Services 86.61 84.76 90.15 88.17 125.49 133.52 92.21 93.00 101.76 104.33
Protection of Persons
and Property 81.76 82.49 86.69 89.18 107.30 117.65 100.23 102.66 117.71 114.34
Education 89.70 87.71 92.86 91.24 120.03 139.75 92.54 91.49 98.96 95.70
Elementary and Secondary
Education 102.43 100.95 98.80 98.94 104.79 98.50 97.44 99.38 98.46 100.31
General Services 99.48 99.75 100.09 99.89 100.09 100.32 99.27 99.63 100.63 100.84
Total 95.84 94.31 95.53 94.94 110.57 113.32 96.32 98.86 99.58 100.71
- 32 -
Auld, Douglasand Eden, L.B. (1984). Equalizationand the CanadianConstitution. Governmentand
Baffes,John and AnwarShah(1993). Productivityof PublicSpending,SectoralAllocationChoices,and
EconomicGrowth, Policy ResearchWorkingPaper No. 1178, The World Bank, Washington,D.C.
Bird, Richard (1986). Federal Finances in Comparative Perspective. Canadian Tax Foundation,
Bird, Richard M. (ed.) (1980). Fiscal Dimensionsof CanadianFederalism.Toronto: Canadian Tax
Boadway,Robinand AnwarShah(1993). FiscalFederalismin Developing/Transition Economies. Some
Lessonsfrom IndustrializedCountries. Forthcomingin the NationalTax Journal
Boadway,Robin, Sandra Robertsand AnwarShah. 1993. "The Reformof Fiscal Systemsin Developing
Countries: A Federalism Perspective." Paper presented at the Conference on Fiscal Reform and
Structural Change sponsored by the InternationalDevelopmentResearch Centre and the Indian
StatisticalInstituteat New Delhi, August25-27, 1993.
Boadway,R., F. Flatters and A. LeBlanc(1983a). RevenueSharing and the Equalizationof Natural
ResourceRevenues.CanadianPublicPolicy IX: 2: 174-180.
_and F. Flatters (1983b). Efficiency, Equity and the Allocation of Resource Rents. In
McLureand Miezkwski(1983), pp. 99-125.
and F. Flatters (1983a). Equalization in a Federal State: An Economic Analvsis.
EconomicCouncilof Canada, Ottawa.
and F. Flatters (1983b). "Efficiency and Equalization Payments in a Federal System of
Government: A Synthesisand Extensionof Recent Results", The CanadianJournal of Economics.
XV. 4, 613-33.
Fiscal Transfers in Canada. Canadian Tax Foundation,
Boadway,Robin (1980). Intergovernmental
Boadway,Robin and Frank Flatters (1991). Federal-Provincial
Recent Constitutionaland Policy Developments.In McMillan(1991), pp. 87-121.
Break, George C. (1980). FinancingGovernmentin a Federal System. Washington,D.C.: The
Buchanan,J.M. (1950). Federalismand Fiscal Equity. AmericanEconomicReview40 (September):
- 33 -
(1952). Federal Grants and Resource Allocation. Journal of Political Economy 60
Clark, Douglas H. (1983). Canadian Experiencewith the RepresentativeTax System as a Means of
Measuringthe RelativeFiscalCapacitiesof Provincialand Local Governments.Testimonybeforethe
United States Senate Committee on GovernmentalAffairs Subcommitteeon Intergovernmental
Relations, April 6, 1983.
(1969). Fiscal Need and Revenue Equalization Grants. Toronto: Canadian Tax
Courchene,T.J. (1978). Avenuesof Adjustment:The TransferSystem and Regional Disparities.In M.
Walker, Editor, Canadian Confederationat the Crossroads: The Search For Federal-Provincial
Balance.Vancouver:The Fraser Institute, pp. 145-86.
Courchene,T.J. and G.H. Copplestone(1980). AlternateEqualizationPrograms:Two Tier Systems.In
FiscalDimensionsof CanadianFederalism,editedby R.M. Bird. Toronto: CanadianTax Formation:
Courchene, T.J. (1984). EqualizationPayments.Ontario EconomicCouncil.
Courchene,T.J. (1986). EconomicManagementand the Division of Powers. Toronto: Universityof
Today and Tomorrow. Ottawa: Supply
EconomicCouncilof Canada (1982a). FinancingConfederation.
EconomicCouncil of Canada (1982b). MigrationInfluencedby FiscalPrograms, News Release.
December 15, 1982.
Gramlich, Edward (1991). CanadianFiscal Federalism:An Outsider's View. In MelvilleMcMillan
Hanson, Eric J. (1961). Fiscal Needs of the CanadianProvinces. Toronto: CanadianTax Foundation.
Lynn, J.H. (1968). ComparingProvincialRevenueYields. The Tax IndicatorApproach. Toronto:
(1964). Federal-Provincial Financial Relations. Studies of the Royal Commission of
Taxation, No. 23, Ottawa: Queen's Printer.
McLure, C.E. (Ed.) (1983). Tax Assignmentin FederalCountries. ANU Press, Canberra, Australia.
McMillan,M., Anwar (Chaudry)Shah and David Gillen (1980). The Impact of Provincial-Municipal
TransportationSubsidies.A Report preparedfor the AlbertaTransportationMinister. August 1980.
McMillan,Melville(1975). Towardsthe More OptimalProvisionof Local Public Goods:Internalization
of Benefitsand Grants. PublicFinanceOuarterly. Vol. 3, July 1975, 229-261.
- 34 -
Melville, McMillan(1991). ProvincialPublicFinances:Vol. I- ProvincialSurveys; Vol.2- Plaudits,
Problems and Prospects. CanadianTax Paper No. 91. Toronto: CanadianTax Foundation.
Musgrave, R.A. (1961). Approachesto a Fiscal Theoryof PoliticalFederalism. In Public Finance.
Needs. Sourcesand Utilization.PrincetonUniversityPress, pp. 97-122.
(1959). The Theory of Public Finance. New York: McGraw Hill and Polinsky, A.M.
(1970). RevenueSharing- A CriticalView. In FinancingState and LocalGovernments.Proceedings
of the Monetary ConferenceNantucketIsland, Massachusetts,June 14-16, 1970 sponsoredby the
Federal Reserve Bank of Boston.
Oates, W.E. (ed.) (1977). The Political Economyof Fiscal Federalism.Lexington, Mass: D.C. Heath.
(1975). Financing the New Federalism. Washington, D.C.: Resources for the Future,
(1972). Fiscal Federalism. New York: Harcourt, Brace, Jovanovich.
Scott, A.D. (1977). An EconomicApproachto the Federal Structure.Centre for Researchon Federal
Financial Relations, The Australian National University, Reprint Series: Number 27 Canberra:
(1952). Federal Grants and Resource Allocation. Journal of Political Economy 60.
Shah, Anwar, and Zia Qureshi, et al. (1994). IntergovernmentalFiscal Relations in Indonesia.
DiscussionPaper, The World Bank, Washington,D.C. Processed.
(1994). "Perspectives on the Design of Intergovernmental Fiscal Relations. Policy Research
Paper, World Bank, Washington,D.C.
(1993). The Theory and Practice of Intergovernmental Transfers." Paper presented
Fiscal Relations, Shanghai,China, October25-28, 1993.
at the Conferenceon Intergovernmental
(1992a). "Empirical Tests for Allocative Efficiency in the Local Public Sector." Public
(1992b). "Dynamics of Public Infrastructure and Private Sector Productivity and
Profitability." The Review of Economicsand StatisticsLXXIV(1):28-36.
(1991 s. "Perspectives on the Desir,n of Intergovernmental Fiscal Relations." Policy,
Research and ExternalAffairs, WorkingPaper Series No. 726. World Bank, WashingtonD.C.
(1991b). "A Fiscal Need Approach to Equalization." Paper presented to the Canadian
EconomicAssociationMeetings,Kingston,Ontario, Canada, June 1991.
(1991c). The New Fiscal Federalism in Brazil. The World Bank Discussion Paper Series No.
124. World Bank, Washington,D.C.
- 35 -
(1990). "Intergovernmenrtal Fiscal Relations in Sri Lanka." Paper presented at the 46th
Congress of the InternationalInstituteof Public Finance, Brussels,August27-30, 1990.
(1989a). "A Linear Expenditure System Estimationof Local Fiscal Response to Provincial
TransportationGrants." KentuckyJournal of Economicsand Business: 150-168.
(1989b). "A Capitalization Approach to Fiscal Incidence at the Local Level." Land
(1988). "Capitalization and The Theory of Local Public Finance: An Interpretive Essay."
Journal of EconomicSurveys2(3): 209-243.
(1984a). "Alternative Approaches to The Measurement of Expenditure Needs of Canadian
Provinces." Finance Canada DiscussionPaper, March 1984.
(1984b). "An Empirical Analysis of Public Transit Subsidies in Canada." Finance Canada
DiscussionPaper, June 1984. Also publishedIn Roy Brown, ed., Quantityand Qualityin Economic
Analysis,Vol. 111986, UniversityPress of America.
(1984c). "A Preliminary Analysis of Expenditure Need Compensation in Fiscal
Equalization." FinanceCanada DiscussionPaper, November.
(1983a). "Municipal Fiscal Needs and Provincial-Municipal Unconditional Transfers in
Canada." DiscussionPaper, FinanceCanada, Ottawa, March.
(1983b). "Efficiency, Equity and Fiscal Equalization Grants: Issues and Alternatives."
Finance Canada, DiscussionPaper, October.
(1983c). "The New Federalism in Australia." Finance Canada Discussion Paper, February
(1981). "The Reform of Provincial-Municipal Transfers in Alberta." A Report prepared for
the Ministerof MunicipalAffairs, Alberta.
_________ (1979). "Provincial Transportation Grants to Alberta Cities: Structure, Evaluation and a
Proposalfor An alternateDesign." AlbertaTransportation,Edmonton,Canada. An abridgedversion
publishedin Roy C. Brown, ed., Quantityand Qualityin EconomicResearch,Volume1. 1985. New
York and London: UniversityPress of America.
Policy Research Working Paper Series
Title Author Date for paper
WPS1264 A Rockand a HardPlace:The Two J. MichaelFinger March1994 M. Patefha
Facesof U.S.TradePolicyTowardKorea 37947
WPS1265 ParallelExchangeRatesin MiguelA. Kiguel March1994 R. Luz
Countries:Lessonsfrom Stephen O'Connell 34303
WPS1266An EfficientFrontierfor International SudhakarSatyanarayan March1994 D. Gustafson
Portfolioswith CommodityAssets PanosVarangis 33732
The Case Zeljko Bogetic
WPS1267The Tax BaseinTransition: March1994 F. Smith
of Bulgaria Arye L. Hillman 36072
WPS1268The Reformof Mechanisms for ElianaLa Ferrara March1994 N. Artis
ForeignExchange Theory GabrielCastillo
and Lessonsfrom Sub-Saharan John Nash
WPS1269 Union-Nonunion WageDifferentials Alexis Panagides March1994 I. Conachy
in the Developing Harry AnthonyPatrinos 33669
Study of Mexico
WPS1270 HowLand-BasedTargetingAffects MartinRavallion March1994 P. Cook
RuralPoverty BinayakSen 33902
WPS1271 Measuring Effect of External F. DesmondMcCarthy March1994 M. Divino
Shocksand the Policy Responseto J. Peter Neary 33739
Them: Empirical Applied GiovanniZanalda
to the Philippines
WPS1272The Valueof Superfund Cleanups: ShreekantGupta March1994 A. Maranon
Evidence from U.S. Environmental GeorgeVanHoutven 39074
ProtectionAgency Decisions L.
WPS1273DesiredFertilityand the Impactof Lant H. Pritchett March1994 P. Cook
WPS1274The NewTradeTheoryand Its AsadAlam March1994 A. Alam
Relevance DevelopingCountries 87380
WPS1275Female-Headed Households, RicardoBarros March1994 K. Binkley
Poverty,and the Welfareof Children LouiseFox 81143
WPS1276Is There Persistence the Growth AshokaMody March1994 M. Patena
of Manufactured Exports?Evidence KamilYilmaz 37947
WPS1277 PrivateTraderResponse Market Steven Jaffee March1994 C. Spooner
Liberalization Tanzania'sCashew 32116
Policy Research Working Paper Series
Title Author Date for paper
VPS1278 Regulation Commitment the AhmedGalal March1994 B. Moore
Development Telecommunications 38526
WPS1279OptimalHedgingStrategyRevisited: YingQian March1994 S. Lipscomb
Acknowledging Existenceof RonaldDuncan 33718
Nonstationary Time Series
WPS1280The EconomicImpactof Export WendyE. Takacs March1994 M. Pateha
Controls:An Applicationto Mongolian 37947
Cashmera Romanian Wood Products
WPS1281 Humanand PhysicalInfrastructure: Emmanuel April 1994 L. Longo
and Pricing Policies
Public Investment 37786
WPS1282 Copperand the NegativePriceof Larson
DonaldFrederick April 1994 A. Kim
WPS1283InterestRates in Open Economies: DipakDasGupta April 1994 B. Kim
Real InterestRate Parity,Exchange BejoyDasGupta 82467
Rates,and CountryRisk in Industrial
and Developing Countries
WPS1284The SovietEconomicDecline: WilliamEasterly April 1994 R. Martin
Historicaland Republican Stanley Fischer 31320
Economic RobertG. King
WPS1285CapitalFundamentalism, April 1994 P. Sintim-Aboagye
Development, Economic Growth Ross Levine 38526
WPS1286 Economic and
Transformation the Luca Barbone April 1994 S. Hatbi
FiscalCrsis: A CriticalLook at the jr.
Domenico Marchetti 37143
CentralEuropeanExperience the 1990s
WPS1287 UnstableInflationand Seignorage Jacques Morisset April 1994 D. Jenkins
Revenues LatinAmerica: How Many 37890
limes Canthe Govemrnent FoolPeople?
WPS1288 The Public Financeof Infrastructure: VinayaSwaroop April 1994 C. Jones
Issuesand Options 37699
WPS1289A FiscalNeedsApproach to AnwarShah April 1994 C. Jones
Transfersin a Decentralized
and an Stijn Claessens
WPS1290Oil PriceInstability,Hedging, April 1994 D. Gustafson
Oil Stabilization PanosVarangis 33714