House of Commons - Treasury - Minutes of Evidence

Select Committee on Treasury Minutes of Evidence

Memorandum submitted by Professor Iain McLean and Alistair McMillan, Nuffield College, Oxford

REGIONAL FINANCE AND GDP IN EUROPE

The different methods of allocating finance between central and regional government depend on a variety of factors, including the taxation systems of central and regional governments, the number of government functions carried out at the different levels of administration, and the historical and political development of the regional governments. As noted in Iain McLean's evidence to the Treasury Select Committee (q 31), the issues surrounding redistribution of government resources in different countries are complex, and a much more in-depth study would be required to address these fully. This note presents some basic findings about the distribution of spending and the regional GDP of a number of European States.

One question from the Treasury Select Committee evidence that can be answered, is Mr Beard's enquiry about the differences in regional GDP across the regions of major European Countries (q 28). Eurostat have provided information on regional GDP across Europe, as well as in the European Candidate Countries (Eurostat news release no 13/2002, 29 January 2002). They note that Inner London has the highest regional per capita GDP (with results adjusted to control for exchange rate differences[5]), and that regional inequality appeared to be highest within the UK:

The NUTS-2 categorisation used in the Eurostat analysis are based on the administrative divisions that exist in EU member states. In this note we look at regions of the major countries, in line with the approach used in the Fiscal Crisis of the United Kingdom paper.

Table 1 presents an analysis of the range of GDP across the regions of some European Countries. This suggests that the broader pattern of regional inequality is not outstanding in the UK. France appears to have the most even distribution of GDP across regions (despite the inclusion of the exceptional De«partements d'Outre-Mer). The highest variation is in Belgium, caused by the small number of regions and the extremely high GDP of the Brussels-Capital Region. The case of Belgium shows that the choice of level of aggregation can influence any such analysis—in the UK the inner London effect noted above is diluted by its inclusion in the Greater London region. In Germany, the presence of rich city-States such as Hamburg, skew the distribution.

Table 1 Distribution of regional GDP in some European Countries

	Number of regions	Overall GDP per capita (PPS)	Standard deviation of per capita GDP for each region	Standard deviation as a percentage of average GDP
UK	12	21,395	3,864.4	18.1
Spain	18	17,480	3,620.6	20.7
Germany	16	22,579	6,509.2	28.8
Belgium	3	22,645	12,836.4	56.7
Italy	20	21,970	5,336.8	24.3
France	23	21,173	1,955.1	9.2

The following sections present the country specific data used to construct Table 1, and provide a brief outline of issues relating to revenue allocation to regional governments in the countries.

BELGIUM

	GDP per capita 1999 (Euro)	GDP per capita 1999 (PPS)
Brussels-Capital Region	46,989	46,179
Flanders	22,686	22,295
Wallonia	16,817	16,527
Overall	23,042	22,645

Under the Saint Michaels Agreement (September 1992) Belgium adopted a federal constitution, as a means of resolving the tensions between the Flemish and French population (Delmartino 1996). This saw the construction of a complicated regional and cultural administration, recognizing three regions (Flemish, Wallonian, and Brussels-Capital Regions) and three cultures (Flemish, French, and German speaking). Since the 1970s there has been a move towards regional fiscal autonomy, although the regions have had limited tax raising powers and the complexities of the over-lapping regional and cultural bases of administration limited the scope of implementation. Under the Saint Michaels Agreement greater tax raising gathering rights were given to the regions, along with the introduction of a mechanism linking revenue to actual GNP growth.

Delmartino, Frank (1996) "Belgium after the Fourth State Reform: Completed Federalism of Confederalism in the making?" in Gisela Farber and Murray Forsyth The Regions—Factors of Integration or Disintegration in Europe? Baden-Baden: Nomos Verlagsgesellschaft: 117-144.

FRANCE

	GDP per capita 1999 (Euro)	GDP per capita 1999 (PPS)
Ile de France	20,058	18,922
Champagne-Ardenne	21,247	20,044
Picardie	18,588	17,535
Haute-Normandie	20,880	19,697
Centre	20,328	19,176
Basse-Normandie	18,926	17,854
Bourgogne	20,439	19,281
Nord-Pas-de-Calais	17,937	16,921
Lorraine	18,745	17,683
Alsace	23,212	21,897
Franche-Comte«	19,479	18,376
Pays de La Loire	19,661	18,547
Bretagne	18,869	17,801
Poitou-Charentes	18,042	17,020
Aquitaine	20,580	19,415
Midi-Pyrenees	19,645	18,532
Limousin	18,216	17,184
Rho®ne-Alpes	23,096	21,788
Auvergne	18,946	17,873
Languedoc-Roussillon	17,576	16,580
Provence-Alpes-Co®te D'Azur	20,304	19,154
Corse	18,162	17,133
De«partements d'Outre-Mer	12,453	11,748
Overall	22,444	21,173

Whilst French regionalism has been seen as institutionally and nationalistically weak (Corsica and the overseas de«partement excepted), there is a functional regionalism associated with the implementation of public-policy since the Etablissements Publics Re«gionaux in 1972. These reforms gave powers over regional and economic development, education, culture, and vocational training to the regions; although they have been described by Richard Balme as remaining "quasi-specialised attribution competencies, which means that regional initiatives are in principle limited to fields of intervention defined by law" (Balme 1998: 183). However, the relative strengths of de«partement, city, and national governments have restricted the scope for the political development of the regions, which has been reflected in a limited budget and weak social identification. Despite this weakness, the regions have been the conduit through which capital financing for infrastructure and education have been channelled (Balme 1998: 187-8).

Balme, Richard (1998) "The French region as a space for public policy" in Patrick Le Gale«s and Christian Lequesne Regions in Europe. London: Routledge: 181-198.

GERMANY

	GDP per capita 1999 (Euro)	GDP per capita 1999 (PPS)
North Rhine-Westphalia	24,547	23,046
Bavaria	27,869	26,164
Baden-Wurttemberg	27,166	25,505
Lower Saxony	21,865	20,528
Hesse	29,177	27,393
Saxony	16,069	15,086
Rhineland-Palatinate	21,574	20,255
Saxony-Anhalt	15,376	14,436
Schleswig-Holstein	22,467	21,093
Thuringia	15,695	14,735
Brandenburg	15,903	14,931
Mecklenburg-Western Pomerania	15,755	14,791
Saarland	22,385	21,015
Berlin	22,495	21,119
Hamburg	41,381	38,850
Bremen	32,164	30,197
Overall	24,050	22,579

The post war West German constitution was constructed on a federal model, with the competencies of the States entrenched against the imposition of the federal government, although within certain administrative areas responsibilities were shared. The arrangement was reflected in the construction of the central legislature, with the popularly elected Bundestag counterposed with a Bundesrat, which was composed of members nominated from the State legislatures. The system has been described by Spahn and Franz (2000: 9) as one of "asymmetrical power sharing", whereby the federal government lays down the general framework of law, which is implemented by the States. Whilst taxation is determined by the central government, the revenues are disbursed amongst the States according to formulae with strong equalization components.

The major recent challenge to the German Federal structure has been the unification of the East and West, and the very different patterns of economic development in the two areas. This has added a new dimension to the debate over regional government spending, and a challenge to what had emerged as a consensual form of interest intermediation which, up to the early 1980s, had operated relatively smoothly. Whilst the Lander had shown clear differences in the rates of economic growth, Arthur Benz argues that this was not necessarily because of differences in regional policies, since the distinct economic regions do not coincide with the boundaries of the Lander, although this may change as the regions increasingly become the focus of policy making and infrastructure development (Benz 1998: 114). This new political identity became increasingly apparent in the 1980s, when the Constitutional Court was brought in to arbitrate in disagreements over the degree to which Lower Saxony's oil proceeds should be taken into account when fiscal equalisation was being considered; and again when the city-states of Hamburg and Bremen agitated for special treatment (Sturm 1992: 121-2). The unification led to the institutionalisation of an East-West divide, reflecting their different levels of income and wealth; although this was partly implemented through extra spending allocations specifically linked to the one-off event of unification (through the German Unity Fund).

Article 107 of the Constitution outlines the provisions for the financial equalization of government resource provision between States. Section (2) states that:

It has to be ensured by statute, that a reasonable equalization between financially strong and financially weak States is achieved; due consideration being given to financial capacity and financial requirements of communes or associations of communes. Such statute has to specify the conditions governing equalization claims of States entitled to equalization payments and equalization liabilities of States owing equalization payments as well as the criteria for determining the amounts of equalization payments. Such statute may also provide for grants to be made by the Federation from federal funds to financially weak States in order to complement the coverage of their general financial requirements (supplementary grants). (http://www.jurisprudentia.de/index.html).

The complexity of the system for financial equalization has, as noted above, caused controversy, which was exaserpated by the structural changes necessitated by unification. A ruling of the Federal Constitutional Court, or Bundesverfassungsgericht (1999), demanded clarification of the rules under which transfers between the centre and the regions was organized.

The operation of federal fiscal relations was criticised by an OECD Economic Survey in 1998. This report argued that the existing system required reform: 'the integration of the new states is producing greater financial strains than anticipated, and as it has become increasingly evident that convergence will take some time, the danger has emerged that the incentive structure of existing financial relationships may be perverse: the system of inter-governmental transfers aimed at equalising revenue levels leaves little return to a state's efforts to improve its finances' (OECD 1998: 70). The Report argues that the system, which is based on the equalization of a putative measure of revenue raising capacity distorted the incentives to implement optimal policies, since successful revenue raising by a State could lead to greater transfers away from that State, and vice versa. It also suggested that the system was biased towards States with dense populations, which had incorporated special dispensations into the criteria determining the re-allocation of resources which created biases within the system[7]. The Report recommended moving away from a system of co-financing, to one whereby the poorer States would receive block-grants from the federal government, and States would have greater freedom to control their taxing powers and revenues.

The ruling of the Constitutional Court on the Finanzausgleich called for clearer criteria for the redistribution of funds between States, and sought to break the link between tax income and redistribution. The German finance ministry responded by proposing that distribution of VAT revenues be adjusted to reflect financial equalization; replacing federal grants and complex co-financing arrangements.

Benz, Arthur (1998) "German regions in the European Union: from joint policy-making to multi-level governance" in Patrick Le Gale«s and Christian Lequesne Regions in Europe. London: Routledge: 111-129.

Sturm, Roland (1992) "The Changing Territorial Balance" in Gordon Smith, William E Paterson, Peter H Merkl, and Stephen Padgett (eds.) Developments in German Politics. London: Macmillan: 119-134.

Spahn, P. Bernd and Oliver Franz (2000) "Consensus Democracy and Inter-jurisdictional Fiscal Solidarity in Germany". Paper presented to the IMF Conference on Fiscal Decentralization, 20-21 November 2000 (http://www.imf.org/external/pubs/ft/seminar/2000/fiscal/index.htm, accessed 18 July 2002).

ITALY

	GDP per capita 1999 (Euro)	GDP per capita 1999 (PPS)
Piemonte	22,445	25,661
Valle D'Aosta	24,036	27,480
Liguria	20,232	23,131
Lombardia	25,330	28,959
Trentino-Alto Adige	25,234	28,849
Veneto	22,557	25,788
Friuli-Venezia Giulia	21,576	24,668
Emilia-Romagna	24,465	27,970
Toscana	21,112	24,137
Umbria	18,887	21,593
Marche	19,400	22,179
Lazio	21,633	24,732
Abruzzo	15,727	17,980
Molise	14,954	17,097
Campania	12,256	14,011
Puglia	12,511	14,304
Basilicata	13,881	15,869
Calabria	11,626	13,292
Sicilia	12,330	14,096
Sardegna	14,641	16,739
Overall	19,217	21,970

The Italian government restructured the allocation of government spending between regions after the institution of Ordinary Statute Regions (in 1970-72), which were to be financed mainly by central government transfers. In recent years the pressure for regional fiscal autonomy has been increased with the emergence of the Lega Nord, with its strong base in the (richer) northern regions. In the poorer regions (the Mezzogiorno) there has been an active debate about the ability of regional governments to deliver public services (Giarda 2000: 6). The divergence between the economic performance of the north and south has been highlighted by the OECD, which raised concerns about the effectiveness of reforms to regional government in addressing regional equality (OECD 1999: Ch 4).

The Italian Constitution (Article 119) sets out the basis of regional government finance:

(1) Municipalities, Provinces, Metropolitan Cities and Regions shall have financial autonomy with respect to revenues and expenditures.

(2) Municipalities, Provinces, Metropolitan Cities and Regions shall have autonomous resources. They shall establish and implement their own taxes and revenues, in harmony with the Constitution and in accordance with the principles of coordination of the public finances and the taxation system. They shall receive a share of the proceeds of State taxes related to their territory.

(3) The law of the State shall establish an equalization fund to the benefit of areas where the fiscal capacity per inhabitant is reduced, with no restrictions as to the allocation of its proceeds.

(4) The funds deriving from the sources mentioned in the previous paragraphs shall enable Municipalities, Provinces, Metropolitan Cities and Regions to finance in full the functions attributed to them.

(5) In order to promote economic development, social cohesion and solidarity, to remove economic and social inequalities, to foster the actual exercise of human rights, to pursue ends other than those pertaining to the exercise of their ordinary functions, the State may allocate additional resources or carry out special actions to the benefit of certain Municipalities, Provinces, Metropolitan Cities and Regions.

(6) Municipalities, Provinces, Metropolitan Cities and Regions shall have their own assets, assigned to them according to general principles established by State law. They may only contract loans in order to finance investment expenditure. Any State guarantee on such loans shall be excluded. (source: http://www.uni-wuerzburg.de/law/it00t.html).

The provisions of section stress the importance of regional "needs" in the allocation of finance.

Mediation between the state and regions are mediated through a State-Regions Conference, on which sit all Regions Presidents and certain members of the central Cabinet. Reforms suggested in 1999 led to each regional government presenting a programme of government spending, and bidding for government funds.

There are three different type of regions: 15 Ordinary Statute Regions, and five Special Statute Regions[8].

Giarda, Piero (2000) "Intergovernmental Fiscal Relations in Italy". Paper presented to the IMF Conference on Fiscal Decentralization, 20-21 November 2000 (http://www.imf.org/external/pubs/ft/seminar/2000/fiscal/index.htm, accessed 18 July 2002).

OECD Economic Surveys (1999) Italy Report.

SPAIN

	GDP Per Capita 1999 (Euro)	GDP per capita 1999 (PPS)
Andalucia	10,410	12,751
Canarias	14,035	17,191
Cataluna	17,461	21,389
Galicia	11,346	13,897
Valencia	13,786	16,886
Aragon	15,057	18,444
Asturias	12,317	15,087
Baleares	17,606	21,566
Cantabria	13,483	16,515
Castilla La Mancha	11,279	13,815
Castilla Leon	13,065	16,003
Extremadura	8,985	11,006
La Rioja	16,121	19,747
Madrid	19,363	23,718
Murcia	11,822	14,481
Ceuta Y Malilla	11,345	13,896
Pais Vasco	17,515	21,454
Navarra	18,160	22,244
Overall	14,270	17,480

Spain has undertaken a far-reaching decentralization of political and economic structures since 1978. As with the UK, this has been an asymmetric decentralization, motivated by long standing political imperatives; in particular the need to respond to separatist regional aspirations within a unified state. The pattern of reform has seen different fiscal and devolved institutions emerge in the regions, with certain regions (eg Basque Country and Navarra) given greater tax raising autonomy and control over government resources (see Guerra 1996: 154). As Vinuela notes, such asymmetrical constitutional principles can lead to problems:

"On the one hand, they introduce notorious differences in the political influence of the regional governments in national policy. On the other, they may affect the economic capacity of the less endowed regional governments to provide the public servicers at a level comparable to that of other better endowned regions. When these two problems emerge, they are a worrying source of resentment and division between regions, and therefore, of instability, unless corrective mechanisms are introduced in the system" (Vinuela 2000: 4)

Whilst the constitution does demand measures to promote the social and economic unity of the country, there are clearly conflicts between the political recognition of regional autonomy and the need to minimize inequalities between regions[9].

Revenue assignment by the central government, therefore, has to reflect both the levels and capabilities for regional revenue generation, and the extent to which governmental functions are shared between central and regional governments. The process by which revenue assignment is administered is complex, with the central government responsible for overall macro-economic policy and dealing directly with EU institutions; a system of bilateral committees representing regional and central government; and a mechanism, the Interterritorial Compensation Fund, for judging differences in the regional resource costs of public service provision[10].

The legal framework for the operation of the Interterritorial Compensation Fund is set out in an Organic Law on the Financing of the Autonomous Communities (LOFCA), which contains guidelines on the tax assignment and equalization of service provision. However, given that there are conflicting demands of autonomy and equalization built into the broader structure of the administration, there is wide scope for political manipulation of outcomes. LOFCA outlined certain parameters which would determine regional transfers, including population, "fiscal effort"[11], real income per capita, and the services which were devolved to regional governments; of which the key redistributive elements was income per capita. The "fiscal effort" criteria was imposed by the rich regions, to reflect better rates of tax collection (Vinuela 2000: 17). Additional criteria, such as insularity, surface, and population dispersion were added through negotiation. Each of these criteria were given weights, applied to aspects of government functions[12]. Weights were determined through negotiation, and varied over five yearly renegotiations. Health care and social security were excluded from the regional spending equalisation process; with resources allocated in terms of population and qualification, respectively. The system of resource equalization has had predictable political consequences:

The advanced regions are of the opinion that the present financing system is already very redistributive . . . Less advanced regions consider that the present redistribution is not enough' (Vinuela 2000: 19).

Guerra, Luis Lopez (1996) "Regions and Nationalisties in Spain: The Autonomous Communities" in Gisela Farber and Murray Forsyth The Regions—Factors of Integration or Disintegration in Europe? Baden-Baden: Nomos Verlagsgesellschaft: 145-155.

Toboso, Fernando (2001) "What degree of financial autonomy and tax responsibility do regional governments have in Spain since democratic transition in 1978". Paper presented to Conference on "Multi-level Governance: An Interdisciplinary Perspective", 28-30 June 2001.

Vinuela, Julio (2000) "Fiscal Decentralization in Spain". Paper presented to the IMF Conference on Fiscal Decentralization, 20-21 November 2000 (http://www.imf.org/external/pubs/ft/seminar/2000/fiscal/index.htm, accessed 18 July 2002).

NOTE ON CHANGES IN REGIONAL PUBLIC SPENDING AND THEIR EFFECTS ON INVERSE GDP CALCULATIONS

With the release of a new Public Expenditure Statistical Analysis (PESA) for 2002-03, it is possible to update the figures presented in the McLean and McMillan paper "The Fiscal Crisis of the United Kingdom".

Tables 1a and 1b shows the changes in spending on "devolved programmes" (ie total regional spending minus spending on Social Security).

Table 1a Per capita government expenditure on "devolved programmes" in the regions of the UK

	North East Region	North West	Yorkshire and Humberside	East Midlands	West Midlands	South West	Eastern Region	London	South East Region
1999-2000	2,803	2,704	2,480	2,416	2,517	2,405	2,389	3,330	2,346
2000-01	3,022	2,928	2,905	2,632	2,736	2,654	2,624	3,431	2,550
Change	219	224	425	216	219	249	235	101	204

Source: PESA 2002-03.

Table 1b Per capita government expenditure on "devolved programmes" in the territories of the UK

	England	Scotland	Wales	Northern Ireland	United Kingdom
1999-2000	2,612	3,412	3,095	3,881	2,740
2000-01	2,837	3,638	3,289	4,347	2,971
Change	225	226	194	466	231

Source: PESA 2002-03, Table 8.6b.

The changes are illustrated in Figure 1:

Unfortunately, we do not have updated figures for regional GDP (due from ONS in November 2002) or differences in regional price levels. However, using the 1999 figures as proxies, we can extend the inverse GDP model to the current levels of spending.

This gives us the updated table 2 and table 3, which replicate the analysis in tables 6 and 7 of the original paper, using the updated data. Since there are only marginal changes to the spending figures in table 1, and the GDP and price deflators used in the analysis are the same, there are no major changes in the pattern of distribution analysed here. A comparison of the effects of the changes in regional spending announced in PESA 2002-03 is shown in Figure 2, which transposes the original and new simulation of the effects of an inverse GDP formula (both unadjusted and adjusted for differences in regional price levels).

Table 2 Public spending under an inverse GDP formula (12 regions/territories)

Region	Actual Pub exp/head (A)	Index pub exp/head	GDP/head	Index GDP/head	InvGDP pub exp/head (B)	Residual (column A¸column B)
South-east	2,550	86	15,098	116	2,553	-3
East	2,624	88	15,094	116	2,553	71
Greater London	3,431	115	16,859	130	2,286	1,145
South West	2,654	89	11,782	91	3,271	-617
West Midlands	2,736	92	11,900	92	3,239	-503
East Midlands	2,632	89	12,146	94	3,173	-541
Yorkshire and Humberside	2,905	98	11,404	88	3,379	-474
North West	2,928	99	11,273	87	3,419	-491
North East	3,022	102	10,024	77	3,845	-823
Wales	3,289	111	10,449	81	3,688	-399
Scotland	3,638	122	12,512	96	3,080	558
NI	4,347	146	10,050	77	3,835	512
UK	2,971	100	12,972	100	2,971	0

Source: HM Treasury, Public Spending Statistical Analysis 2002-03, derived from Tables 8.6b and 8.12b. Office for National Statistics, Regional GDP 1999, summary table.

Table 3 Public spending under an inverse GDP formula, adjusted for regional price differentials (12 regions/territories)

Region	Price index compared to UK averagea	GDP/head at PPP	Index GDP/head	Index GDP/head at PPP	InvGDP pub exp/head	InvGDP pub exp/head at PPP	Unadjusted Residual (from Table 6)	Residual at PPP
South-east	3.1	14,730	116	114	2,553	2,408	-3	-66
East	1.5	14,871	116	115	2,553	2,385	71	32
Greater London	6.8	15,786	130	122	2,286	2,242	1,145	990
South West	-0.7	11,865	91	91	3,271	2,985	-617	-594
West Midlands	-1.2	12,045	92	93	3,239	2,945	-503	-464
East Midlands	-1.7	12,356	94	95	3,173	2,882	-541	-487
Yorkshire and Humberside	-3.4	11,805	88	91	3,379	3,006	-474	-360
North West	-2.2	11,527	87	89	3,419	3,070	-491	-416
North East	-4.7	10,518	77	81	3,845	3,381	-823	-642
Wales	-3.8	10,862	81	84	3,688	3,282	-399	-259
Scotland	-0.9	12,626	96	97	3,080	2,813	558	585
Northern Ireland	-0.2	10,070	77	78	3,835	3,506	512	520
UK		12,972	100	100	2,971	2,729	0	0

a Data for all units except Northern Ireland are inclusive of housing rents. For NI, relative housing rents are not available, so data are presented exclusive of housing rents. The effect of this is probably to exaggerate the positive residual for NI.

Source: HM Treasury, Public Spending Statistical Analysis 2002-03, derived from Tables 8.6b and 8.12b. Office for National Statistics, Regional GDP 1999, summary table. Price figures from Baran and O'Donoghue 2002: Table 2.

Figure 1: Government expenditure on `devolved programmes' in the regions and territories in the UK

Figure 2: Effect of changes in public expenditure on simulation of raw and PPP-adjusted residuals

5 Results are calculated in terms of a purchasing power standard (PPS) which is an artificial currency that eliminates the effects of exchange rates on national currency levels. Back

^{6   Nomenclature of Statistical Territorial Units. Back

^{7   City-States were weighted by a factor of 1.35 (compared to 1.0 for other States), and special dispensation was given to those States which included harbours. Back

^{8   One of these Trentino-Alto Adige divides into two Provinces, each with its own special statutes. Back

^{9   The Spanish Senate is designated as the Chamber of Territorial Representation, and oversees the governmental structures which determine the balance between the centre and the Regions. See http://www.senado.es/caract_territ/text_caract_territ_i.html for an overview. Back

^{10   This estimates the "effective cost" of service provision in each of the regions, in order to reflect direct costs (regional expenditure on personnel and purchasing goods and services), indirect costs (imputed share of central expenditures on "devolved" services), and investment costs (reflecting capital requirements) (Vinuela 2000: 8). Back

^{11   An indicator of the regional capability for revenue raising. Back

^{12   Weights varied according to the type of function, with different weights given to Educational functions, where population was given a much higher weighting. Back}}}}}}}



© Parliamentary copyright 2002	Prepared 4 November 2002