A comparative study on Spanish regions’ investment capacity in a budgetary discipline anticipated scenario, by means of multicriteria Promethee method

econstor Make Your Publications Visible. A Service of zbwLeibniz-Informationszentrum Wirtschaft Leibniz Information Centre for Economics Quijada, Maria Teresa Arevalo; Nuño, Maria Mercedes Castro; Yñiguez, Rocio Conference Paper A comparative study on Spanish regions’ investment capacity in a budgetary discipline anticipated scenario, by means of multicriteria Promethee method 44th Congress of the European Regional Science Association: "Regions and Fiscal Federalism", 25th - 29th August 2004, Porto, Portugal Provided in Cooperation with: European Regional Science Association (ERSA) Suggested Citation: Quijada, Maria Teresa Arevalo; Nuño, Maria Mercedes Castro; Yñiguez, Rocio (2004) : A comparative study on Spanish regions’ investment capacity in a budgetary discipline anticipated scenario, by means of multicriteria Promethee method, 44th Congress of the European Regional Science Association: "Regions and Fiscal Federalism", 25th - 29th August 2004, Porto, Portugal, European Regional Science Association (ERSA), Louvain-laNeuve This Version is available at: http://hdl.handle.net/10419/117057 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. www.econstor.eu 1 44TH EUROPEAN CONGRESS OF THE EUROPEAN REGIONAL SCIENCE Number: 227. “A COMPARATIVE STUDY ON SPANISH REGIONS’ INVESTMENT CAPACITY IN A BUDGETARY DISCIPLINE ANTICIPATED SCENARIO, BY MEANS OF MULTICRITERIA PROMETHEE METHOD”. Authors: ARÉVALO QUIJADA, M. T.: University of Seville and Fundación centrA:. Spain. CASTRO NUÑO, M. M.: University of Seville and Fundación centrA:. Spain. YÑIGUEZ OVANDO R.: University of Seville. Spain. ABSTRACT The principle of the budgetary discipline, compulsory for the Spanish regions by the Law 18/2001, December 12th [General Law of Budgetary Stability] and the Organic Law 5/2001, December 13th, complementary to the former one, established in the frame of the European Agreement for Stability and Growth, can generate conflicting situations with those Spanish regions which investment capacity depends on external borrowing. This paper deals with the corresponding relative position of the different regions, according to its investment capacity, using for that purpose a simulation exercise, in which we advance the budgetary stability constraint for the period 1997-2000. In this paper, the public financial activity is treated, for each region, through different public revenue and expenditure ratios per capita. This situation leads to consider a multicriteria Promethee method as the apropriate one to obtain a global ranking for all of them. In the opinion of Al-Shemmeri, Al-Kloub and Rearman (1997), this method is the most adequate one because of the following advantages: public authorities, as decision takers, can understand easily the results, regardless the knowledge they may have about it; the method uses understandable economic parameters; the method avoids distorting scale effects among different alternatives and, as well, makes possible the deviation evaluation between alternatives and, finally, allows for sensibility analysis. 2 1. INTRODUCTION The Stability and Growth Pact which was approved at the meeting of the European Union Council held in Amsterdam in June 1997, indicates fighting public deficit as the primordial economic policy objective of the states signing the pact. This commitment forms part of the Spanish internal legal system, through the enactment of Law 18/2001, dated 12th December 2001, on General Budgetary Stability and its complementary Organic Law, extending the scope of the stability objective to the Territorial Administration sector. Due to the fact that the Public Sector cannot incur public deficit, it must perform budgetary adjustments by either increasing taxes (which is not a particularly feasible option given the considerable existing tax pressure and the unpopular nature of this measure), or by reducing expenses, preferably capital expenses, since these are more flexible to reduce than current expenses1. As a result of the above, it seems likely there will be a conflict between the budgetary stability objective imposed by financial orthodoxy and investment potential in Public Administration, represented by capital expenses. The present study has been conducted on the budgetary stability – investment capacity binomial of the 17 Autonomous Communities (CCAA) and the Autonomous Cities of Ceuta and Melilla (CDAA), based on the assumption of preparing for the zero deficit condition during a 4-year period (1997-2000), using the multi-criteria Promethee method in order to provide information on the relative situation of each of these regions within the scenario described. To achieve the purpose indicated above, the study has been divided into the following sections: introduction, methodology, empirical analysis, conclusions, appendix including a net flows diagram and sensitivity analysis and references. 2. METHODOLOGY 2.1. Description of variables2. - PERSONNEL EXPENSES (PEX): Salaries paid by the Public Sector to civil servants and personnel recruited. - CURRENT EXPENSES IN GOODS AND SERVICES (CEX): Expenses arising from exercising current activities in the Public Sector. 1 De Haan et al (1996). 2 Aggregate values: 1997-2000. 3 - FINANCIAL EXPENSES (FEX): Credits which are necessary to satisfy the financial burden of public liabilities. - CURRENT TRANSFERS MADE (CTM): Non-compensated payments made for the purpose of financing current transactions. - REAL INVESTMENT (RINV): Expenses used to create or acquire capital assets. - CAPITAL TRANSFERS MADE (KTM): Non-compensated payments made for the purpose of financing capital transactions. - DIRECT TAXATION (DITAX): Basically income and capital gains tax. - INDIRECT TAXATION (INTAX): Basically sales tax. - FEES, PUBLIC PRICES AND OTHER INCOME (FPP): Monetary compensation and income arising basically from the sale of public goods and services. - CURRENT TRANSFERS RECEIVED (CTR): Non-compensated resources received by the Public Sector for financing current transactions. - CAPITAL INCOME (KI): Income arising from public estate or capital income. -DIVESTING OF EFFEECTIVE INVESTMENTS (DEINV): Income arising from the sale of public capital assets. - CAPITAL TRANSFERS RECEIVED (KTR): Non-compensated resources received by the Public Sector for financing capital transactions. -GROSS SAVINGS (GS): Current (Income – Expenses). - CAPITAL BALANCE (KS): Capital (Income – Expenses). - NON-FINANCIAL DEFICIT OR SURPLUS (DEF/SURP): GS + KS. - CAPITAL EXPENSES FINANCED BY NET INDEBTEDNESS (KEfind): Capital transactions financed through the reduction of assets or the increase of financial liabilities. - FINANCIAL SAVINGS (FS): Less financial expenses arising from budgetary balancing restrictions. -REDUCTION IN CAPITAL EXPENSES UNDER BUDGETARY STABILITY (∇KEbst): Reduction in capital transactions resulting from non-incurrence of debt. The budgetary stability alternatives are: ¾ Increasing taxes, and/or, 4 ¾ Reducing public expense, as defended by González-Páramo3. Basically, capital expenses are those most affected by cuts, due to reasons of political visibility4 and because they are more flexible5, compromising the productive capacity of the economic system. In comparing the situation in which Spanish regions would find themselves within this hypothetical scenario, we are raising the issue of putting into some kind of order a series of alternatives (CCAA, CDAA) in the face of multiple criteria (Public Accounting ratios as defined in paragraph 2.3). 2.2. The Promethee method: a multicriteria decision system. As we have said, in this work the public financial activity is treated, for each region, through different public revenue and expenditure ratios. This situation leads to consider a multicriteria Promethee method as the appropriated one to obtain a global ranking for all of them. In the opinion of Al-Shemmeri, Al-Kloub and Pearman6, this method is the most adequate one because of the following advantages: public authorities, as decision takers, can understand easily the results, regardless the knowledge they may have about it; the method uses understandable economic parameters; the method avoids distorting scale effects among different alternatives and, as well, makes possible the deviation evaluation between alternatives and, finally, allows for sensibility analysis. Actually, to decide in a multicriteria environment is difficult, because indeed most decision problems that arise in our daily life involve different often conflicting objectives that we try to satisfy simultaneously. In practice, this attempt is illusory and we have to consider best compromise solutions. So in general, we consider multicriteria decision problems of the following type: A is a set of n possible decisions or alternatives (finite set: Autonomous Communities and the Autonomous Cities of Ceuta and Melilla) which are evaluated 3 González-Páramo J. M. (2001). 4 Oxley, H. and Martín, M. (1991) 5 De Haan et al (1996). 6 Al-Shemmeri, T., Al-Kloub, B. and Pearman, A., (1997). {} MaagagagagOPTIMIZAT kj ∈ /)(),...,(),...,(),( 21 5 through k criteria g1, g2.., gk. The basic data for such a problem can be presented in a evaluation matrix that it gives the dominance relation, based on a unanimity principle, can be defined as follows: Aba ∈),( a dominates b iff .,...2,1,)()( khbgag hh = ∀ ≥ (with at least one >). The non-dominated alternatives are called efficient (or Pareto optimal) solutions. In practice, the dominance relation is often very poor and the number of efficient solutions can be rather large. Indeed, it is clear that such data do not generally induce a complete ranking on the set A of alternatives. The problem is not mathematically well stated and the notion of optimal solution does not exist. However the problem is most often economically well stated as it expresses the different and possibly conflicting objectives of the decision maker. In order to provide the decision maker with a good assistance a particular multicriteria methodology must be considerate, called PROMETHEE (means: Preference Ranking Organization Method for Enrichement Evaluations). The Promethee method (and its visual associated visual modelling: The GAIA plane), take into account all the necessary requisites for the most multicriteria models: ¾ The amplitude of the deviations between the evaluations of the alternatives are taken into account: dj (a,b) = gj (a) – gj (b). ¾ As the criteria are generally expressed in different units, the scaling effects are completely eliminated. ¾ When comparing a couple of alternatives (a,b), the multicriteria decision aid method, come to one of this conclusions: - a is preferred to b or b is preferred to a. - a and b are indifferent. - a and b are incomparable (this circumstance, allows the method to avoid to decide when insufficient information is available)7. ¾ All the parameters have economical significance. ¾ We can obtain different results depending on the additional information by the decision maker. ¾ This method analyzes the conflicting aspects between the criteria. It is very important to have the opportunity to speak to the decision maker, to appreciate 7 Incomparability between two alternatives appears when one alternative is good on some criteria and bad on others, while the opposite holds for the other alternative. 6 his/her preferences, and to have a clear interpretation of the weights of the criteria. Then this requisite set, the Promethee method, in order to consider the deviations and the scales or the criteria, associates a generalized criterion to each criterion g(.). For this objective, we define a preference function, which is obtained giving the degree or preference between alternatives for the decision maker. The generalized criterion associated, is defined by the following pair: {gj(.), Pj(.,.)}, where Pj (a,b) = P j {d j (a,b)} Mba ∈ ∀ , 1),(0 ≤ ≤ baPj To facilitate the association the generalized criterion to each criterion, in the classic literature8, there are six typical generalized criteria that are proposed to the decision maker. The choice is made interactively by the decision maker and the analyst according to their preference degrees. When a generalized criterion has been associated to each criterion, we define, with all the criteria, a multicriteria preference index of a over b, like this9: ),( ba π = ∑ = n ijj baPw 1 ),( , with ⎟ ⎠ ⎞ ⎜ ⎝ ⎛= ∑ = n ij w 1 1, Where: wj > 0 (j = 1,2,...,k), are weights associated to each criterion, according to its relative importance. If we consider how each alternative a, is facing the n-1 other ones, we can define the two following outranking flows: 1. The positive outranking flow: expresses how much each one is outranking all the others. The best alternative has the higher positive flow, because it represents its dominance power. () ∑∈ +− =Ab ba n a, 1 1 )( πφ 2. The negative outranking flow: expresses how much each alternative is outranked by all the others. The best alternative has the smaller negative flow, so represents its weakness. 8 Brans, J. P., (1984) (1986), Brans, J. P., and Vincke, P. H. (1985). 9 ),( ba π expresses how and which degree a is preferred to b, and ),( ab π how b is preferred to a, over all the criteria. 7 () ∑∈ −− =Ab ab n a, 1 1 )( πφ From the information about these positive and negative flows, we can deduce two natural rankings of the alternatives: 1. The PROMETHEE I PARTIAL RANKING: It is obtained from the pairwise comparisons and intersections: a is preferred to b ⇔ a and b are indifferent ⇔ )()()()( baandba −−++ = = φ φ φ φ a and b are incomparable ⇔ otherwise 2. The PROMETHEE II COMPLETE RANKING: It is the balance between the positive and negative outranking flows. The higher net flow is the better alternative. )()()( aaa −+ − = φ φ φ The complete ranking is defined by: a is preferred to b ⇔ )()( ba φ φ > a and b are indifferent ⇔ )()( ba φ φ = Let us notice there remain no incomparability but the resulting information is more disputable, because, a considerable part of the information is lost by considering the difference. As we have said, the Promethee method allows obtaining an important part with graphical information about the conflicting character of the criteria and the impact of the weights of the criteria on the final results. This is called GAIA visual modelling method (Geometrical Analysis for Interactive Assistance), and provides such information. It complements the rather approach of the Promethee procedure with a descriptive and graphically oriented analysis. The set of alternatives should be represented by n points in the k-dimensional space, but as the number of criteria is usually greater than two, it is impossible to have a )()()()( baandba −−++ <> φφφφ )()()()( baandba −−++ < = φ φ φ φ )()()()( baandba −−++ => φ φ φ φ 8 clear vision of these points. So, it is possible to define a plane in order to obtain a two dimensional representation of the alternatives. The GAIA plane is defined by vectors which represent the criteria in according by weights. As few information as possible get lost by projection, so a measure of the quantity of information being preserved, is given by δ parameter (it represents a percentage of the total information about the problem). About the GAIA plane interpretation, let us consider the projections of the unit vectors (of all the criteria) on the plane. These axes have different lengths and positions that mean a differentiation power of the criteria. The length of this vectors, is a measure of how much the criterion gj differentiates the alternatives (the longer vector belongs to the more criterion differentiates the alternatives). When two criteria expressing the same preferences, their vectors are oriented approximately in the same direction; while conflicting criteria are represented by axes having opposite directions. The projection on the plane of the different criteria in according to the assessed weights, allows a clear visualisation of the solution with the unit vector called The Promethee decision axis: π .If π is short, the Promethee decision axis has no strong decision power, so the unit vector is nearly orthogonal to the GAIA plane10. When this vector is long, the decision maker is invited to select the alternatives that are as far as possible in its direction. Moreover, each alternative has a projection in the GAIA plane, too. It is represented by a point that if it is located in the direction of a particular criterion axis, is generally a good alternative on this criterion. When the distance between two projected alternatives is small, is because they both are similar alternatives for the decision maker. The best alternatives are located in the direction of the Promethee decision axis π . The Promethee and GAIA methods have been implemented on personal computers, with several decision support systems. In this paper, we use the DECISION LAB 2000 program. This software allows obtaining a sensibility analysis about the results. A sensibility analysis is quite recommended before finalising the decision, because a modification or the assessed weights to the criteria can modify seriously the conclusions. 2.3. Criteria. We have established three different economic groups in per capita terms: 10 In this case, the criteria are conflicting and a good compromise should be selected near the origin. 15 MELILLA 0,3000 16 0,7000 16 0,4000 4 Prepared by the authors. TABLE 8 Extremadura and Ceuta are not comparable, since their partial orders alternate with each other. Madrid and the Community of Valencia are the worst-positioned, due to the considerable effort made by both regions in investment and their greater sacrifice in achieving budgetary stability Navarra is in first place (due to the fact that it is fully autonomous in financial terms), together with Andalucia. 4. CONCLUSIONS The GAIA figure shows the results of the analysis made in the previous paragraph. The vectors of criteria R15, R17 and R19 are overlapping, and thus have the same discriminatory effect on the alternatives. R6, R12 and R13 are more conflictive, since the angle formed by their vectors is the widest. 16 Axis π confirms the fact that the Community of Valencia is the worstpositioned (furthest away in the opposite direction) and Navarra is the best (as it is furthest away but in the same direction). APPENDIX The flow figures showing the analyses made and the sensitivity tests for each one are shown below, for the purpose of checking the reliability and stability of the solutions. SENSIVITY TEST FOR EXPENSE RATIOS RATIOS STABILITY INTERVAL Max. (%) Weight (%) Min.(%) R1 8,24 7,14 5,80 R2 15,15 14,29 13,64 R3 7,65 7,14 4,88 R4 15,37 14,29 13,25 R5 29,41 28,57 28,10 R6 29,58 28,57 27,08 SENSIVITY TEST FOR INCOME RATIOS RATIOS STABILITY INTERVAL Max. (%) Weight (%) Min.(%) R7 10,53 10,00 9,40 R8 10,69 10,00 9,09 R9 10,99 10,00 9,17 R10 11,21 10,00 9,65 R11 10,89 10,00 8,99 R12 26,34 25,00 24,24 R13 25,62 25,00 23,81 SENSIVITY TEST FOR BUDGETARY STABILITY RATIOS RATIOS STABILITY INTERVAL Max. (%) Weight (%) Min.(%) R14 14,50 12,50 11,71 R15 15,66 12,50 11,11 R16 13,48 12,50 9,90 R17 27,27 25,00 20,00 R18 13,27 12,50 10,64 R19 28,00 25,00 18,92 SENSIVITY TEST FOR RATIO SAMPLES RATIOS STABILITY INTERVAL Max. (%) Weight (%) Min.(%) R5 16,56 16,00 15,69 R6 16,33 16,00 15,74 17 R12 16,28 16,00 14,86 R13 16,37 16,00 15,63 R15 4,35 4,00 3,57 R17 16,42 16,00 15,63 R19 16,48 16,00 15,63 The solutions presented are stable, since the weightings assigned are at the intermediate point of the stability interval. REFERENCES - ALCAIDE INCHUSTI, J. y ALCAIDE GUINDO, P. (2003). “Avance de las magnitudes económicas españolas en 2002 y serie provisional del balance económico regional. Años 1995 a 2002”, Cuadernos de Información económica, nº 173, pp. 1-58. - AL-SHEMMERI, T; AL-KLOMB, B.; PEARMAN, A: (1997). “Model choice in multicriteria decision aid”, European Journal of Operational Research, 97, pp. 550-560. - ARÉVALO QUIJADA, M .T., GÓMEZ DOMÍNGUEZ, D., VÁZQUEZ CUETO, M.J. y ZAPATA REINA, A: (2002): “Un estudio de las Cajas de Ahorros Andaluzas mediante el método multicriterio Promethee”. Estudios de Economía Aplicada. Vol. 20I, pp. 5-27. - BRANS, J.P.; MARESCHAL, B.; VINCKE, Ph.: (1984). “PROMETHEE: a new family of outranking methods in multicriteria analysis”. Operational Research 84, Brans J.P. Ed. North-Holland, pp. 408-421. - BRANS, J.P.; VINKE, Ph.: (1985). “A preference ranking organization method, the PROMETHEE method”. Management Science, vol. 31, pp. 647-656. - BRANS, J.P.; MARESCHAL, B.; VINCKE, Ph.: (1986). “How to select and how rank proyects: the PROMETHEE method”. European Journal of Operational Research, vol. 24, pp. 228-238. - DE HAAN et al (1996): “Government capital formation: Explaining the decline “, Review of world Economics, 132, pp. 55-74. - GONZÁLEZ-PÁRAMO, J. M., (2.001): Costes y beneficios de la disciplina fiscal: la LGEP en perspectiva, Instituto de Estudios Fiscales, Madrid. - GORDO, L. y HERNÁNDEZ DE COS, P. (2000): “El sistema de financiación autonómica vigente para el periodo 1997-2001”, Documento de Trabajo, nº 0003, Banco de España. - http://www.estadief.minhac.es. - http://www.mineco.es/tesoro. 18 - LAGO PEÑAS, S., (2.001): “La dinámica de los gastos de capital de las Comunidades Autónomas: un análisis de sus determinantes”, Hacienda Pública Española, nº 157-2, pp. 65-82. - LÓPEZ LABORDA, J. y VALLÉS JIMÉNEZ, J. (2002): “Evolución del endeudamiento autonómico entre 1.985 y 1.997”, Papeles de Trabajo, Instituto de Estudios Fiscales, nº 2, Madrid. - MINISTERIO DE HACIENDA: Liquidación de presupuestos de las Comunidades y Ciudades Autónomas. Ejercicios 1.997-2.000. Secretaria de Estado de Presupuestos y Gastos, Madrid. - MINISTERIO DE HACIENDA: Presupuestos de las Comunidades y Ciudades Autónomas. Ejercicios 1.997-2.000. Secretaría de Estado de Presupuestos y Gastos, Madrid. - MONASTERIO ESCUDERO, C. (1996): “Los límites al endeudamiento de los gobiernos subcentrales. Teoría y evidencia para el caso español”, Papeles de Economía Española, nº 67, pp. 265-285. - ONRUBIA, J., (2.001): “De estabilidad e instituciones presupuestarias: necesidades y reformas en España”, Economistas, Colegio de Madrid, nº 87, Extra. - OXLEY, H. y MARTÍN, M. (1991): “Controlling government spending and deficits: Trends in the 80s and prospects for the 90s”, OECD Economic Studies, nº17, pp.145189. - SEVILLA SEGURA, J. V., (2.001): Las claves de la financiación autonómica. Ed. Crítica. Barcelona. 19 20 FIGURE OF EXPENSES OUTRANKING FLOWS FIGURE OF INCOME OUTRANKING FLOWS FIGURE OF BUDGETARY STABILITY OUTRANKING FLOWS FIGURE OF A RATIO SAMPLES OTURANKING FLOWS 21