Vol 8 | No 1 | SPRING 2010 ARTICLES Institutions Matter: Financial Supervision Architecture, Central Bank and Path-Dependence

South-Eastern Europe Journal of Economics Vol 8 | No 1 SPRING 2010

Vol 8 | No 1 | SPRING 2010 ARTICLES Institutions Matter: Financial Supervision Architecture, Central Bank and Path-Dependence. General Trends and the South Eastern European Countries DonATo MASCIAnDAro, MArC QuInTyn Prospects of Changes in regional Economic Structures Since Eu Accession K. MATTAS, C. CIoBAnu, D. PSAlToPouloS Economics Education: Can liberal Arts Make Economics More Appealing? SErDAr onGAn Multidimensional Poverty in Greece: A Deep, Persistent Grey? A. lyBErAKI, P. TInIoS, T. GEorGIADIS Identifying Spatial labor Markets in Greece from the 2001 Travel-to-Work Patterns ProDroMoS-IoAnnIS K. ProDroMIDIS

BOOK REVIEW Wealth, Welfare and the Global Free Market A Social Audit of Capitalist Economics by Ibrahim ozer Ertuna reviewed by ThEoDorE P. lIAnoS Human Development in the Era of Globalization Essays in Honor of Keith B. Griffin edited by J.K. Boyce, St. Cullenberg, Pr. K. Pattanaik and r. Pollin reviewed by ThEoDorE P. lIAnoS

FUNDING INSTITUTIONS unIVErSITy oF MACEDonIA web site: http://www.asecu.gr unIVErSITy oF IoAnnInA ISSn 1792-3115 unIVErSITy oF WESTErn MACEDonIA editorialboard Theodoros Lianos, Editor Athens University of Economics and Business, Athens, GREECE Stylianos Fountas, Associate Editor University of Macedonia, Thessaloniki, GREECE Michael Chletsos, Associate Editor University of Ioannina, Ioannina, GREECE Katerina Sarri, Associate Editor University of Western Macedonia, Kozani, GREECE advisoryboard Anders Aslund Institute for International Economics, USA Robert Fearn North Carolina State University, USA Ben Fine University of London, SOAS, UK Ulrich Koester University of Kiel, GERMANY Janos Kornai Harvard University, USA Philip Martin University of California, Davis, USA Jeffrey Sachs Columbia University, USA Amartya Sen Cambridge University, USA Kenneth Thomson University of Aberdeen, UK Richard Wolff University of Massachusetts, Amherst, USA THE OFFICIAL JOURNAL OF THE ASSOCIATION OF ECONOMIC UNIVERSITIES OF SOUTH AND EASTERN EUROPE AND THE BLACK SEA REGION

Vol 8 | No 1 | SPRING 2010

FUNDING INSTITUTIONS

UNIVERSITY OF MACEDONIA UNIVERSITY OF IOANNINA UNIVERSITY OF WESTERN MACEDONIA

web site: http://www.asecu.gr UNIVERSITY OF MACEDONIA PRESS ISSN 1792-3115 THESSALONIKI SEEJE Editorial Office: ASECU, University of Macedonia 156 Egnatia str., 540 06 Thessaloniki, Greece tel: +30 2310 891 793, fax: +30 2310 891 748 e-mail: [email protected] web site: http://www.asecu.gr

university of macedoniaPRESS 156 Egnatia str., 540 06 Thessaloniki, Greece, tel: +30 2310 891 741, fax: +30 2310 891 731 e-mail: [email protected], http://www.uom.gr/uompress contents

Articles Institutions Matter: Financial Supervision Architecture, Central Bank and Path-Dependence. General Trends and the South Eastern European Countries 7 Donato masciandaro, Marc Quintyn Prospects of Changes in Regional Economic Structures Since EU Accession 55 k. mattas, c. ciobanu, d. psaltopoulos Economics Education: Can Liberal Arts Make Economics More Appealing? 73 Serdar Ongan Multidimensional Poverty in Greece: A Deep, Persistent Grey? 87 A. Lyberaki, p. tinios, T. Georgiadis Identifying Spatial Labor Markets in Greece from the 2001 Travel-to-Work Patterns 111 Prodromos-Ioannis K. Prodromidis

Book Review Wealth, Welfare and the Global Free Market A Social Audit of Capitalist Economics 129 by Ibrahim Ozer Ertuna reviewed by Theodore P. LianOS Human Development in the Era of Globalization Essays in Honor of Keith B. Griffin 131 edited by J.K. Boyce, St. Cullenberg, Pr. K. Pattanaik and R. Pollin reviewed by Theodore P. LianOS aim and scope of

ASECU was founded in 1996 as Association of South-Eastern Eu- rope Economic Universities with the general aim of promoting the interests of those economic universities in South-Eastern Europe which are public, recognized or financed by the state of origin. At the General Assembly of 2007 it was decided to enlarge ASECU and to take in the countries of eastern Europe and the Black Sea region and associated members from the Middle East and north-eastern Africa. ASECU was modified inAssociation of Economic Universi- ties of South and Eastern Europe and Black Sea Region and pres- ently counts forty five (45) members. Forty one (41) of them are full members from the countries Albania, Armenia, Bosnia-Herzegovina, Bulgaria, FYROM, Greece, Montenegro, Poland, Russia, Romania, Slovakia, Serbia, Turkey and Ukraine. As associated members have been accepted four (4) Universities from Lebanon and Egypt. The specific aims of the Association are: To promote cooperation between Economic Universities, Faculties, Departments; i.e., especially: a) to exchange views and information about syllabi, b) to exchange undergraduate and postgraduate students and c) to exchange teaching and research staff. To provide members with the opportunity exchange information, opinions etc. by publishing a relevant scientific journal or by cooperation in elaborating scientific studies in relation to the future development of higher education and research as well as to improve their quality in the field of economic studies and business administration. To undertake initiatives for the protection of the interests of members and their institutions, so as to be supported by international organizations and in particular by the higher education institutions of the European Union. To encourage cooperation between universities inside and outside the countries referred to in the Association. To pursue cooperation in the field of higher education with the consolidation of close relations with other organisations having similar aims, e.g. E.U.A.; To provide opportunities for harmonising the degrees of faculties and departments of the universities participating in the Association; To promote cooperation between economic universities, faculties, departments in the field of research for the benefit of the economy, the society, peace and the cultural development of the countries referred to the Association.

Presidency and Board Members The following Board was elected during the fourth General Assembly of ASECU, on the 2nd of March 2007, at the University of Macedonia in Thessaloniki, Greece.

Yannis Tsekouras, President University of Macedonia, Thessaloniki, Greece Atanas Damyanov, Vice President University D. Tsenov, Svishtov, Bulgaria Bozidar Cerović, General Secretary University of Belgrade, Belgrade, Serbia Dhori Kule, Member University of Tirana, Tirana, Albania Bobek Suklev, Member University St. Cyril and Methodius, Skopje, Fyrom Vesna Karadjić, Member University of Montenegro, Podgorica, Montenegro Miruna Marinescu, Member Academy of Economic Studies, Bucharest, Romania

South-Eastern Europe Journal of Economics 1 (2010) 7-53

Institutions Matter: Financial Supervision Architecture, Central Bank and Path-Dependence. General Trends and the South Eastern European Countries

Donato MasciandarO* Department of Economics and Paolo Baffi Centre, Bocconi University Marc Quintyn IMF Institute, International Monetary Fund, Washington

Abstract We propose a path-dependence approach to analyzing the evolution of the financial supervisory architecture, focusing on the institutional role of the central bank, and then apply our framework to describing the institutional settings in a selected sample of countries. The policymaker who decides to maintain or reform the supervisory architecture is influenced by the existing institutional setting in a systematic way: the more the central bank is actually involved in supervision, the less likely a more concentrated supervisory regime will emerge, and vice versa (path-dependence effect). We test the path-dependence effect describing and evaluating the evolution and the present state of the architecture of six national supervisory regimes in South Eastern Europe (SEE): Albania, Bulgaria, Greece, Romania, Serbia, and Turkey. The study of the SEE countries confirms the postu- lated role of the central bank in the institutional setting. In five cases the high involvement of the central bank in supervision is correlated with a multi-authority regime, while in one case a high degree of financial supervision unification is related with low central bank involvement.

JEL Classification: G18, G28, E 58. Key words. Financial Supervision, Central Banks, Path-Dependence, Political Economy, South Eastern Europe

* Corresponding Author: Paolo Baffi Centre,B occoni University, via Sarfatti 25, 20136 Milan, Italy e-mail: [email protected]. Special thanks also go to Rosaria Vega Pansini for extensive, skilful and patient research assistance while compiling the database and the country profiles. 8 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

1. Introduction Over the last few years the financial supervision landscape has been radically transformed. Many countries have made deep reforms of the architecture of financial supervision, and more are contemplating changes. In the last twenty years (1986-2006) 94% of the countries included in a large and heterogeneous sample of 102 nations chose to reform their financial supervisory setting (Figure 1). The restructuring wave is making the supervisory regimes less uniform than in the past (Masciandaro and Quintyn 2009). In several cases the architecture still reflects the classic structure, with separate agencies for banking, securities and insurance supervision. However, an increasing number of countries show a trend towards a certain degree of consolidation of the supervisory responsibilities, which in several cases has resulted in the establishment of unified regulators, that are different from the national central banks.1 Various studies (Barth, Nolle, Phumiwasana and Yago 2002, Arnone and Gam- bini 2007, Čihák and Podpiera 2007) claim that the key issues for supervision are (i)) whether there should be one or multiple supervisory authorities and (ii) whether and how the central bank should be involved in supervision. More importantly, these two crucial features of a supervisory regime seem to be related. The literature has tried to go in depth into the analysis of the supervisory reforms measuring these key institutional variables (Masciandaro 2004, 2006, 2007 and 2008), i.e. the degree of consolidation in the actual supervisory regimes, as well as the central bank involvement in supervision itself. The descriptive analysis (Masciandaro 2004) signalled an intriguing result: the national choices on how many agencies should be involved in supervision seems to be strictly correlated with the existing institutional position of the central bank. The degree of supervisory unification seems to be inversely related with the central bank’s involvement in supervision. The trade-off – and the related, so called central bank fragmentation effect – was confirmed first using a cross-country analysis of the reforms in the supervisory regimes (Masciandaro 2006) and then by going more deeply into the economics of the central bank fragmentation effect (Masciandaro 2007 and 2008, Masciandaro and Quintyn 2008, Dalla Pellegrina and Masciandaro 2008). From a political economy point of view, the central bank fragmentation effect can be explained as a peculiar case of path-dependence effect (PDE): the incumbent policymaker, in choosing the level of financial supervision consolidation, is influenced by the characteristics that already exist in terms of the central bank position. The

1. For a survey see e.g. De Luna Martinez 2003, Masciandaro 2005, and Cihak and Podpiera 2007b. The legal issues are described in Mwenda 2006. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 9 policymaker’s choices are viewed as a sequential process in which the institutional position of the central bank matters. Notwithstanding the evidence, different questions still remain unanswered. Among others: Is the PDE able to explain the features of the financial supervisory regimes in a narrow and well defined set of countries? Can the same methodology be a useful instrument to shed light on specific case studies? The aim of this paper is to test the PDE in describing the current features of the supervisory regime in six countries of South Eastern Europe (SEE). The paper is organised as follows. Section two presents the theoretical setup. In section three we review the institutional and empirical background. Section four describes the supervisory architectures in the six countries of South Eastern Europe. Section five applies the architecture indicators to test the robustness of the PDE in the SEE sample, providing also a comparison between their actual regimes and the supervisory setting implemented in the other European countries. Section six will put forward some conclusions as well as possible directions for future research.

2. Theoretical Background: Explaining the Path-Dependence Effect Our theoretical framework is based on three hypotheses. First of all, gains and losses of a supervisory regime are variables computed by the incumbent policymaker, who maintains or reforms the supervisory regime, following his preferences. Secondly, the policymakers are politicians: politicians are held accountable at the elections for how they have pleased the voters. All politicians are career oriented agents, motivated by the goal of pleasing the voters in order to win elections. The main difference among the various types of politicians concerns which voters they wish to please in the first place. Thirdly, the policymakers are influenced by the institutional setting in which they operate. The relationship among the political choices on the future of the supervisory architecture and the actual institutional position of the central bank can be highlighted using a simple model (Masciandaro 2008), which applied a general framework of political choices (Alesina and Tabellini 2003). Consider a society that wishes to assign to an elected policymaker the task of designing the optimal shape of the financial supervisory architecture, focusing on the level of consolidation of the institutional regime that guarantees the effectiveness of the financial supervision policy (thereafter the effective level of unification). The effective levely of unification is determined by the policymaker’s effort a and by his ability:

y = a + Ω (1)

Ability is a random variable; for the sake of simplicity, let us suppose that Ω can assume two values only. The policymaker can be outstanding or not. Therefore the parameter can be ΩL or ΩH with ΩL < ΩH and 10 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

ΩL with probability p

ΩH with probability (1 – p) Ω = ¿{}

Our aim is to show that taking into account the social preferences is not sufficient to explain the shape of the supervisory regime if the incumbent policymaker is career concerned. Therefore let us assume that the citizens care about the effectiveness of the supervisory regime according to a classic well-behaved concave function u = U (y): the social welfare increases with the level of unification. Linear preferences are used: U(y) = y (2)

The policymaker will take the decision whether or not to reform the supervisory setting, taking into consideration his own personal objective function. The policymaker’s effort is costly, and the convex and increasing cost function is defined as C = c (a). The reward for the policymaker is labelled R (a). The two functions are traditionally well-behaved. The policymaker’s utility function is defined as:

R(a) – c(a) (3)

Now we can introduce the role of the institutional position of the central bank. Let us assume that the costs of implementing a higher level of financial supervision concentration can depend on the existing institutional position of the central bank. If a high level of central bank involvement in supervision is the status quo, under specific conditions unified supervision is more difficult to implement, and this means that the politician’s task is, ceteris paribus, more costly. In order to identify these conditions, let us consider that a policymaker aiming to consolidate supervision faces two alternative paths: to create a central bank, monopolist in supervision; or to establish a single financial authority, different from the central bank. The creation of a monopolist central bank can produce information gains, but can be costly for different reasons. First of all, the policymaker may dislike the implementation of a monopolistic central bank if the consequent extension of the classic moral hazard risks – which can occur when monetary policy and supervision policy are delegated to the central bank – are high (Goodhart and Schoenmaker 1995, Llewellyn 2005) (moral hazard risk). Secondly, implementing a monopolistic central bank regime can also be costly when the policymaker also delegates the conduct of business controls to the central bank, an area in which central banks have traditionally sought not to be involved. Instead they prefer to focus more on stability issues (Goodhart 2007 and Bini Smaghi 2007) (conflict of interests risk). Thirdly, the policymaker has to take into account D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 11 the risks of increasing the bureaucratic powers of the central bank (bureaucracy risk). Table 1 documents possible country cases where the political wish to avoid excessive concentration of bureaucratic power can explain the recent evolution of the supervisory setting. In the perception of the HH policymaker the overall evaluation of the bureaucracy risk can take into account different factors. For example, the risk that the central bank, given its bureaucratic power, will please the banking and financial industry (captured central bank), or the possibility that the central bank abuses its degree of institutional independence. But also the alternative solution – establishing a unified supervisor outside the central bank – can face difficulties caused by the central bank position. In fact, the policymaker may face costs in establishing a single financial authority – and thus reducing the central bank’s involvement in supervision – if the central bank’s reputation is high (reputation risk). At the same time, however, if the reputation of the central bank is low, or decreasing, the establishment of a single financial authority is more likely to occur. The role of reputation can work in both directions. Historical cases are described in Table 1. Therefore, we identified four different potential reasons to explain what we called the PDE: the more the central banker is involved in supervision, the less likely a unified supervisor will be established. Whatever the effort of the policymaker, the central bank involvement in supervision can increase the costs of implementing a financial supervision unification. But under which conditions does the PDE become relevant in explaining the policymaker’s decisions on the shape of the supervisory architecture? The sequence of events is as follows. Society chooses to delegate to the policymaker the task of designing the level of supervisory unification. Next, the policymaker decides to maintain or to reform the supervisory regime, choosing effort a, before knowing his ability Ω in implementing this particular policy task. Finally, nature chooses Ω, outcomes are observed and the reward is paid. The incumbent policymaker wishes to be re-elected. Now we can take into account the possibility that two different types of policymakers exist (Masciandaro and Quintyn 2008). On the one hand, one can adopt a helping hand (HH) view (Pig- ou 1938) of the policymaker: he is motivated to improve general welfare. The HH policymaker chooses to maintain or reform his country’s supervisory structure in an attempt to improve the efficiency of overall resource allocation. From the policymaker’s point of view implementing this task is convenient if his re-election is more likely to occur if the citizens’ utility exceeds a threshold W. Denoting by β the value of office and by α1 the effort, the reward function – given (1) and (2) – for the HH policymaker is:

R(α1) = β Pr(U ≥ W)

R(α1) = β Pr(α1 + Ω ≥ W) = β Pr(Ω ≥ W – α1) 12 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

0 ≤ R(α1) ≤ β Given the citizens’ threshold, we consider here the more general case2, when every policymaker – outstanding or not – can be potentially able to extract benefits fulfilling the mandate of reforming supervision:

ΩL ≥ W – α1; then ΩH > ΩL ≥ W – α1 and therefore Pr(Ω ≥ W – α1) = 1 and R(α1) = β Voters are rational. They realise that the alternative to re-electing the incumbent is to get another politician with average ability. It follows that:

e W = α + ΩAV with

Where αe are the voters’ expectations. The HH policymaker chooses effort before observing his talent in implementing the supervisory regime reform, taking the expectations as given. The utility function of the politician is:

The function can assume the following value:

e where ΩL + α1 = Ω(α) — given the expectations α and the skills of the policymaker

ΩL and ΩH — can be considered the re-election condition (Figure 1, second graph). The re-election condition depends on the policymaker’s effort only. Which is the optimal effort? Given the re-election condition, and the value of the office β, the policymaker will decide if and how to implement the supervisory reform — i.e. the optimal effort level α1 — taking into account the marginal cost of implementation (Figure 1, first graph). It is evident (Figure 1) that the interest of the HH policymaker in implementing a greater level of supervisory consolidation will depend, ceteris paribus, on the cost level. In particular, if the PDE holds, for any level of effort, the more the central bank is involved in supervision, the greater the shift up of the costs level will be, and consequently the smaller the likelihood of a supervision consolidation (the optimal level of effort becomes progressively smaller). If the welfare costs become greater than the value of the office, for the HH policymaker it is not convenient at all to use the su- * pervisory reform to increase his probability of re-election (α 1 = 0). The citizens will

2. See Masciandaro 2008 for a general discussion. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 13

Figure 1: The HH Policymaker decision on the optimal level of effort in implementing the supervisory reform not appreciate a supervisory reform whose social losses are greater than the social benefits. The status quo – low consolidation in supervision with high involvement of the central bank – will be confirmed. Obviously, if theHH policymaker evaluated that the increasing involvement of the central bank in the supervision does not imply greater costs – i.e. the risks of moral hazard, bureaucratic excessive power, conflict of interests, reputational losses are negligible, while the information gains in having the central bank deeply involved in supervision are potentially high – the reform will be more likely to occur, producing an “inverse” PDE: the high involvement of the central bank in the supervision will be consistent with high level of supervisory consolidation. Returning to the features of the policymaker, we can use alternatively a grabbing hand (GH) view of the political process (Shleifer and Vishny 1988). According to the GH approach, the policymakers are motivated by the aim to please the interest of specific, well-defined voters. In our case, the financial industry may be considered a highly organised and powerful interest group. The GH policymaker, in defining the supervisory setting, depends on the market view of supervision, if this univocally determines his re-election. The preferences of the financial constituency can be written as: 14 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

The parameter δ represents the importance of the supervisory consolidation goal for the financial constituency. Using δ we can study explicitly the possibility that the central bank should be a captured institution. In fact if we consider the existence of a financial lobby, we have to take into account the possibility that also the central bank should please the financial constituency. If the central is a captured agency, δ is equivalent to the degree of central bank involvement in supervision: the more the central bank is involved in supervision, the greater the financial constituency’s interest in implementing a more consolidated supervision (δ ≥ 0). Alternatively, if the central bank is an independent agency, the more the central banker is involved in supervision, the smaller the financial industry’s preference toward more consolidation in supervision (δ ≤ 0), in order to avoid the risk of giving more power to a non- captured central bank. The parameter f represents the campaign contributions; their purpose is to determine the incumbent’s chances of winning the elections. Let us assume that the policymaker’s effort devoted to implementing the supervisory regime is observable by the financial constituency; the financial professionals can be considered insider agents with respect to the other citizens. Therefore the campaign contributions can be contingent upon the policymaker’s effort: f(a2); for simplicity f = ka2. The GH policymaker chooses effort, taking into account the lobby goal function, as well as the usual social potential costs of implementing the reform, linked to the central bank’s involvement in supervision. The utility function of the GH policymaker is:

Other things being equal, the level of revenues R(y2) and the level of costs C(α2) of the GH policymaker depend both on the effort α2 in implementing the supervisory * reform (Figure 2). In equilibrium the optimal effort α 2 will equate marginally costs and benefits (provided that β| (1 + δ) < (βk + c)|). The interest of the GH policymaker in implementing a greater level of supervisory consolidation will depend, ceteris paribus, on how captured the central bank is — i.e. effect on ∿ of the central bank involvement in supervision — and notwithstanding the costs in increasing the consolidation — i.e. effect on c of the central bank involvement in supervision. If the central banker is captured (∿ > 0), it D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 15

Figure 2: The GH Policymaker decision on the optimal level of effort in implementing the supervisory reform is more likely that the financial constituency likes the supervisory consolidation and the reform will be implemented: the costs level will determine the optimal level of * effort α 2 of the GH policymaker. Under these conditions a central bank unification effect is more likely to occur: we will have the “inverse” PDE: the more the central banker is involved in supervision, the more likely a unified supervisor will be. Oth- erwise, the more the central bank is independent from the financial constituency (d < 0) the smaller the policymaker’s effort in implementing the supervisory reform and a supervisor different from the central bank is more likely to be established. If the * central bank is a strong, independent one (d < –1), the GH policymaker (α 2 = 0) will prefer the status quo — low consolidation in supervision with high involvement of the central bank — and we will have again our PDE. Let us summarise the main findings. If the policymaker acts as an HH type the central bank involvement in supervision can be viewed as an obstacle in the supervision consolidation if at least one of four reasons — moral hazard, conflict of interest, bureaucracy power and reputational losses — is present. The PDE is likely to occur. If the policymaker chooses to please the financial community acting as a GH type, the PDE is less likely to occur, provided that the financial community likes a more consolidated supervision, and the central bank is a captured one. If and only if these assumptions hold we can disentangle the effect of different types of policymakers on the relationship between financial supervision unification and central bank involvement. Otherwise a signal extraction problem occurs. For example, other things being equal, if the central bank is not a captured one and the policymaker acts as a GH type, the PDE is more likely to occur again. 16 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

3. Institutional and Empirical Background: Measuring and Testing the Path- Dependence Effect Our theoretical framework predicts the possibility of different degrees of unification in the design of the supervisory structure, depending on the type of policymaker involved, and on the features of the parameters of the model; in particular our attention has been focused on the institutional role of the central bank in supervision, in order to shed light on the PDE. In the real world, the type of policymaker — as well as all the structural and institutional channels which influence his behaviour — is a hidden variable. At each point in time, we can only observe the politicians’ decision to maintain or reform the supervisory structure, in particular its level of unification. Therefore the next step is to measure the degree of unification in the actual supervisory regimes, as well as the central bank involvement, which represents our key explanatory variable. How can the degree of unification of financial supervision be measured? This is where the financial supervision unification index (FSU Index) proposed in Mas- ciandaro 2004 and used in Masciandaro 2007 and 2008 comes in (description in Table 2). This index was created through an analysis of which and how many authorities in the 102 countries examined are empowered to supervise the three traditional sectors of financial activity: banking, securities markets and insurance3. The country sample depends on the availability of institutional data4. To transform the qualitative information into quantitative indicators, a numerical value has been assigned to each type of regime, in order to highlight the number of the agencies involved. The rationale by which the values have been assigned simply considers the concept of unification of supervisory powers: the greater the unification, the higher the index value5. Figure 2 shows the distribution of the FSU Index. On the one hand there are 42 countries (41 percent of the sample) with a low consolidation of supervision (the Index is equal to 0 or 1). On the other, there are 31 countries (30 percent) that established a unified supervisor or that adopted the peaks model, with a high level of supervisory consolidation (the index takes the value 6 or 7).

3. Sources: for all countries, official documents and websites of the central banks and the other financial authorities. The information is updated to 2006. See Table 2. 4. In the empirical analysis we do not include the very small countries and territories (Bahrain, Bermuda, Cayman Islands, Gibraltar, Hong Kong Maldives, Netherlands Antilles, Singapore and United Arab Emirates) with a single financial authority so as to avoid an evident bias in the empirical analysis. 5. For more information see Masciandaro 2004 and 2008. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 17

Now we will consider what role the central bank plays in the various national supervisory regimes. We use the index of the central bank’s involvement in financial supervision: the Central Bank as Financial Authority Index (CBFA) (description in Table 2). Figure 7 shows the distribution of the CBFA Index. In the majority of countries in our sample (45) the central bank is the main bank supervisor (the Index is equal to 2), while in few countries (10) the central bank is involved in the overall financial supervision (the Index is equal to 4). It is interesting to note that, in general, the present degree of central bank involvement has been established in previous years, and then confirmed in subsequent reforms; this observation is consistent with our path-dependence approach. In fact, for each country we compare the year in which the present degree of central bank involvement in supervision was established (i.e. definition of the CBFA Index, blue line), with the year of the most recent reform of the supervisory architecture (i.e. definition of the FSU Index, red line) (Figure 4). Given the data of 88 national reforms of the supervisory architecture, the central bank involvement was confirmed in 67 cases (76%), decreased in 16 cases (18%), increased in 5 cases (6%). The natural next step in our analysis of the supervisory regimes is to bring both indexes together. The result is shown in Figure 8 where we see that the two most frequent regimes are polarised: on the one hand, the Unified Supervisor regime (18 cases, red ball); on the other, the Central Bank Dominated Multiple Supervisors regime (31 cases, yellow ball). The figure seems to depict a trade off between supervisory unification (or consolidation) and central bank involvement, notwithstanding eight outliers (green balls). Now, considering both indexes for the countries in our sample (Figure 5), the analysis shows that the two most frequently appearing regimes are the extremes: on the one hand, Unified Supervisory regime (13 cases, red ball); on the other, Cen- tral Bank Dominated Multiple Supervisors regime (27 cases, white ball). The Figure seems to depict a trade off between supervision unification and central bank involvement, with two outliers (green ball). Is the PDE valid in our expanded sample? Looking at Figure 5 the answer to this question seems to be “yes”, although it needs to be a cautious one. In fact, while it seems to be true that the more common supervisory regimes are the two polarized ones, we also need to recognize that the number of outliers — i.e. regimes where both the consolidation and the central bank involvement increase — is greater compared with the previous studies on the issue. To perform a closer inspection of the data, we compare the features of the supervisory regimes across time, maintaining the country sample constant. We consider that the existence of the PDE has been confirmed using a sample of 88 countries, with information updated during 2006 (Masciandaro 2008). Using the same country sample, from that time to today other reforms were implemented, producing changes both in 18 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 the supervision consolidation and in the central bank involvement, and consequently in the overall shape of the regimes. We calculated the average levels of the FSU index and of the CBFA index in 2006 and in 2009: the level of consolidation is greater — from 2.88 to 3.34 — but the same is true for the degree of central bank involvement in supervision — from 1.76 to 1.84. Then we analysed how many countries adopted each supervisory regime in 2006 and in 2009, comparing the two situations. The number of countries which adopted the unified regime outside the central bank increased — from 13 to 18 — while the number of countries with central bank dominated regimes went down — from 27 to 25 — but also the number of outliers — unified regimes inside the central bank — is larger — it went from 2 to 4. These figures seem to indicate that only the consolidation process continued for sure, irrespective of the location of the unified powers. In any case we have to devote more attention to the future evolution of the PDE. Finally, it is possible to empirically investigate the robustness of the PDE (Mas- ciandaro 2007 and 2008, Masciandaro and Quintyn 2008, Dalla Pellegrina and Mas- ciandaro 2008). Following our theoretical setup, each policymaker — either the HH type or the GH type — maximises his objective function and determines his optimal level of supervision unification, given the features of the structural variables. The result that emerges is a significant inverse relationship between supervision unification and central bank involvement. So far the PDE matters.

4. Supervisory Architectures and Central Bank Role in SEE countries The preceding sections pointed out that the existing institutional role of the central bank — i.e. the status quo — can explain the future evolution of the supervisory architecture. Both the theoretical model and the econometric analysis claim the robustness of the PDE. Now we wonder if the path-dependence approach can be useful in exploring the features of the supervisory architectures in a selected sample of countries. It might be interesting to ask if the PDE is evident in the six SEE countries. Finding a response would help us not only to interpret what has happened in the past but also to project scenarios of change for the future, with a particular focus on prospects within the European Union framework. In order to analyse the supervisory architecture, we first describe the six supervisory regimes.

4.1 Albania During the last fifteen years, Albanian financial services have undergone major changes in the process of transition from a centralized economy, starting at the beginning of the 1990s. Albania inherited a underdeveloped financial sector from the communist regime. The weakness of financial institutions was the major factor responsible for the development of pyramid schemes. The collapse of these schemes dragged D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 19 the Albanian financial system into a crisis in 1997. The Government tried to resolve the crisis by issuing new sets of rules and regulations. As part of the IMF-supported emergency program, a new banking law was approved in July 1998 establishing a two-tier banking system in which the Bank of Albania (BoA) was given the role of supervisor of the banking system. Amendments to the banking law further strengthened the role of the BoA and the financial system. Besides the BoA, there is also the Albania Securities Commission and the Insurance Supervisory Authority (table 6).

Bank of Albania First established in 1913, the central bank did not survive during the First World War. The period between 1925 and 1944 was characterized by attempts to establish a National Bank of Albania and then, during the period 1944-1992, by the establishment of the State Bank of Albania with both the function of a central bank and a commercial one. Finally with Law No. 7559 ‘On the Bank of Albania’ of 1998 a two-tier system was created with the Bank of Albania performing the function of a central bank. Moreover, the law gives the BoAa the function of supervisor of the banking system. According to the above mentioned law, the Banking and Supervision Department was created to perform supervisory functions and prepare the regulatory framework of the Bank of Albania. New regulations were, in fact, drafted during the succeeding years by the Supervision Department, also to integrate into Albanian banking legislation the provisions of the Basel Committee Principles. The year 1998 can be considered as the year when further steps in the area of financial supervision were taken, starting from the amendment to the regulation “On the Bank of Albania” (law no. 8269/1997) and regulation on “On banks in the Republic of Albania” (law no. 8365/1998) which stipulates that the Bank of Albania is the sole authority to issue licenses and to regulate and supervise all the banks in the Republic of Albania. These amendments were made urgent by the fact that the banking system in Albania was gradually changing, and by the crisis in 1997. Following the provisions contained in the ‘On the Bank of Albania’ law as amended in 1997, the BoA is an entirely independent institution, accountable to the Alba- nian Parliament (People’s Assembly) and responsible for the implementation of monetary policy and supervision of the banking system. As stated in art. 41 of the statutes of the BoA, the management of the Bank is represented by the Supervisory Council, consisting of a Governor and of two Deputy governors. The Supervision Department reports to the First Deputy Governor. The President of the Republic of Albania appoints the Governor of the Bank, at the Prime Minister’s recommendation. The Minister of Finance has the right to attend meetings of the Supervisory Council of the BoA, even though he is not entitled to vote. 20 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Banking legislation received a recent contribution in the form of the Law ‘On Banks of the Republic of Albania’, No. 9662/2006 which contains in its Chapter VI detailed provisions about the process of issuing and revoking licenses by the Bank of Albania. If the Commission — see below — is the supervisory authority for issuing licenses into the securities market, the Bank of Albania has the authority to establish a securities market. Moreover, the Bank established also a Supervisory Board for the Stock Market in order to supervise the activity of the stock exchange and to change its regulations. The BoA may, in fact, issue directives and also demand amendments to the regulations of the securities market.

Albanian Securities Commission It was initially established by Law 8080/1996 as an independent authority and its main aim is to regulate and supervise the securities market. It is composed of five members that are proposed by the President and appointed for five years. The main function of the Commission is to issue licenses for securities’ traders and supervise the management and the operation of all the licensed companies. According to the recent IMF FSAP report (2005), the securities market is at a very early stage of development, with little market activity. Fees for licensing, approvals and all other activities should be the main funding sources allowing it to operate financially independently. The Commission is in fact still dependent on funding from the Government, because of the low level of activity in the market.

Insurance Supervisory Authority The legal framework regulating the supervision of the insurance sector is contained in Law 7506/1991 and Law 8081/1996.The ISC mainly concentrates on the supervision of the financial status of insurance companies. Operating in the insurance sector is subject to approval from the ISC, which also issues decision and regulations relevant to the operation of insurance companies.

4.2 Bulgaria In Bulgaria two financial supervisory authorities can be identified (Table 7): theB ul- garian National Bank (BNB) and the Financial Supervision Commission (FSC).

Bulgarian National Bank The BNB was established in 1879 even though it acquired the status of central bank only with the Law on the Establishment of the Bank in 1985. Since then, the legal framework has further defined and extended the competencies of the BNB. In 1926, after a period of crisis during which the Bank operated under the direction of the D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 21

Ministry of Finance, the BNB, according to a new BNB Law, gained the status of a real central bank. It was also the first time that the regulatory power of theB ank over the banking system was strengthened together with its independence, especially with respect to decision of government lending. The period between 1947 and 1989 was the time when a state monopoly of banking was established by the Communist Party and the BNB lost most of its independence. After 1980 a two-tier banking system was created with the BNB playing the role of a central bank issuing institution. At the beginning of the 1990s, BNB regained part of its lost independence: in 1992 a new Law on Banks and Lending restored the existence of central banking and commercial banking. The year 1997 represents an important year for the general reorganization of the functions and status of the BNB. The new BNB Law reorganised the monetary system creating a Currency Board Arrangement and three new departments: the Issu- ing, the Banking and the Banking Supervision Department. Successive amendments to the Law of the BNB, for example in 2005 and 2006, strengthened the financial and institutional independence of the BNB and its duties in the design of monetary policy and the stability of the financial sector. Specific supervisory provisions of theB NB on the banking system are contained in the Law on Credit Institutions, recently approved by the Bulgarian Parliament (July 2006 and amended in 2007). As stated in art.1 ch.I, this law contains terms and procedures for granting licenses and conducting activities and supervising credit institutions in Bulgaria. The BNB is the only financial authority that can grant and revoke licenses to banks.6 In operating its supervisory role, the BNB is independent (art. 44, ch.7 of the Law on the BNB) and accountable to the National Assembly to which it has to present an annual report. The managing board is constituted by a Governing Council, a Gov- ernor and three Deputy Governors. The Governors are elected by the National As- sembly, which elects also the Deputy Governors. The other three members of the Governing Council are appointed by the President of the Republic.

Financial Supervision Commission The FSC was created in 2003 through the merger of the former Insurance Supervi- sory Agency, the State Insurance Supervision Agency and the National Securities Agency, which became Departments of the new FSC (FSC Act, promulgated in Janu- ary 2003, art.10). The FSC is now the only financial supervisory authority for the security and insurance sector. It is an independent authority accountable to Parliament and it is composed of seven members, the Chairman, the Deputy Chairmen and the other Commissioners, all elected by the National Assembly.

6. However, Article 14 of the law also stipulates that the BNB shall first consider the written state- ment of the Financial Supervision Commission before licensing a bank. 22 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

4.3 Greece The legal framework regulating the financial sector in Greece is fragmented in different codes, laws and regulations because there has not been a systematic codifica- tion of all the laws. In particular, the Ministry of Economy and Finance has issued Ministerial Decisions to regulate specific issues related to the financial sector while the Ministry of Development has regulated the private insurance sector. The three authorities of financial supervision are the Bank of Greece (BoG) for the banking sector, the Hellenic Capital Market Commission (HCMC) for the capital market and the Financial Supervisory Committee for Private Insurance (Table 8).

Bank of Greece As stated in Art. 2 of its statutes, it is the supervisory body for all the banks and credit institutions operating in Greece. It was founded in 1927 and started its operation in 1928. The legal framework for the operation of the BoG is represented by its statutes, first ratified byL aw 3424/1927 and amended several times. The statute does not state explicitly that the BoG is an independent institution but only that the BoG should not take instruction form the government or any organization and that the government or any other political authority should not try to influence the operations of the BoG. The General Meetings of Shareholders were given a wide range of powers, ranging from the approval of the annual report to the election or removal of members of the General Council, and the proposal to amend the Statutes (these proposals should be submitted to the Parliament through the Government). The Minister of Finance may nominate a Government Commissioner with the right to attend the General Meeting of the Shareholders without having the right to vote. The General Council is entitled to make all the other decisions and powers not specifically reserved to the General Meeting. It is composed of the Governor and two Deputy Governors, two members of the Monetary Policy Council and six Councillors. The Governor and the Deputy Governors are appointed for a 6-year term at the proposal of the Council of Ministers. The General Council should approve an annual report to be then submitted to the An- nual General Meeting. Moreover the Annual General Meeting shall elect 3 Auditors to examine the balance sheet of the bank. The statutes also clearly indicate the cir- cumstances in which the Governor and other members of the General Council may be relieved of their office. The dispositions about prudential supervision of theB oG are contained in Art. 55A of the Statutes. The BoG has regulatory and supervisory power over all credit institutions, including the right to impose sanctions and penalties. The main changes to the legal framework were introduced during the last decade, mainly due to the adoption of EU directives. In fact, the pursuit of business by credit institutions is mainly governed by Law 2076/1992 which incorporated into Greek banking legislation the 2nd Banking Directive (89/646/EEC, as codified by Directive D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 23

2000/12/EC). This law contains dispositions about the procedures for granting and withdrawing licenses. The two most important amendments were intended to modernize the operation of the BoG, in line with the provisions of the Treaty of the EU. The new Statutes explicitly state that the Bank’s primary objective is to ensure price stability; they safeguard the Bank’s independence and establish its accountability to Parliament, create a new body at the Bank of Greece, the Monetary Policy Council, and recognize the Bank’s legal integration into the Euro system.

Hellenic Capital Market Commission The HCMC was initially created by Law 148/1967 as a special Committee of the Ministry of Economy and Finance without legal personality. It obtained legal personality in 1991. In fact, as stated in Law 1969/1991, the HCMC is a self-governing institution operating under the jurisdiction of the Ministry of National Economy with the aim of supporting the stability of the capital market and supervising the operation of market participants, ensuring the protection of investors and the sound operation of the stock market. The amendments contained in Law 2324/95 introduce a legal framework that establishes the HCMC as an independent authority. In accordance with Law 1969/91 and following amendments, in particular those contained in Law 2396/96, the CMC can issue regulatory provisions, grant and revoke licenses, impose sanctions, enforce the applicable legislation and draft the annual budget. It is composed of a seven-member Board of Directors and a three-member Executive Committee. The Minister of the National Economy appoints the Chairman and the two Vice-Chairmen of the Board for a five-year period. The remaining four board members are selected by the Minister for the National Economy among the candi- dates proposed by the BoG, the Board of Directors of the Athens Stock Exchange, the Union of Institutional Investors and the Federation of Greek Industries. The CMC is financed by fees and contributions paid by the supervised entities and it also receives funding from the government. The Law also establishes accountability arrangements for the HCMC.

Financial Supervisory Committee for Private Insurance Until recently, the Directorate of Insurance Enterprises and Actuaries of the Minis- try of Development (that was previously under the Minister of Commerce) was the competent supervisory authority for the supervision of the insurance sector. It was responsible for granting authorization for the establishment and operation of insurance companies as well as for exercising administrative and financial supervision of these companies. The Ministry exercised prudential supervision over the solvency and accounting of insurance companies, as well as conduct-of business regulation. Moreo- ver, the Ministry of Development was responsible for imposing monetary sanctions 24 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 in cases of non-compliance with the legal framework. The supervision of insurance is currently in transition. Law 3229/2004 establishes the Financial Supervisory Com- mittee for Private Insurance as the authority responsible for the regulation and supervision of the insurance sector under the supervision of the Ministry of Finance. The new authority will be administered by a nine-member Board of Directors composed as follows: the Deputy Chairman, who will be the President of the Associa- tion of Greek Insurance Companies and representatives of the Consumer Protection Directorate, the Hellenic Actuarial Society, a representative of the Insurance Brokers Association, as well as representatives of the Ministry of Finance and the Ministry of Development, two assessors and a representative of the Guarantee Fund. The Au- thority will be funded by contributions from supervised insurance companies and the State budget. Together with this new supervisory authority, a Coordination Board will be created consisting of the Director of the Bank of Greece and the Chairmen of the Capital Markets Committee and the Insurance Committee The aim of this Board is to promote cooperation between the three existing supervisory authorities and to formulate proposals for the introduction of supervisory regulations for financial institutions and to promote the unification of the three supervisory bodies.

4.4 Romania Romanian financial services have undergone a general process of reshaping and re- definition in order to meet the EU criteria for accession. This has been accompanied by a progressive strengthening of the legal framework for financial operations. Ro- mania has three financial supervisory authorities (Table 9): the National Bank of Romania (NBR), the National Securities Commission (NSC) and the Insurance Su- pervision Commission (ISC).

National Bank of Romania The NBR, established in 1880, is — through its Supervisory Committee — the only supervisory authority for the banking system. Between the late 1940s and the end of the 1980s, the NBR lost its identity as a central bank because it was reduced to a mere financial instrument in the hands of the Government. By the end of the Com- munist era, in 1991, with Law 34/1991 it regained the status and functions of a central bank. But we have to wait until 1998 with Law 101/1998 (the Banks Act) to have a ruling status for the NBR which represents the legal framework of the Bank, establishing that the NBR has legal identity as central bank, and extends its functioning in the directions of monetary policy and prudential supervision. This law has been recently replaced by the NBR Act contained in Law 312/28 of June 2004 which clarifies primary objectives and tasks of the central bank, from the implementation of monetary policy to the authorisation, regulation and prudential supervision of credit D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 25 institutions. In particular, art. 25 of chapter V of the Law recalls also the provision contained in Law 58/1998 (the Banking Act), which established that the NBR must ensure the sound functioning of credit institutions, may issue licenses and regulations, norms, orders and circulars, take measures and apply sanctions to those institutions that do not respect the criteria promulgated by the central bank. In operating its statutory functions, the NBR is managed by a Board of Directors composed of nine members appointed by the Parliament for a period of five years. The NBR is accountable to the Parliament, to which it is obliged to present an annual report.

National Securities Commission The NSC succeeded the Securities Agency established by a government ordinance 18/1993 as a general department of the Ministry of Public Finance. It was established by Law 52/1994 as the autonomous authority responsible for securities and the stock exchange, with legal personality. Its functions in terms of regulation and supervision have been strengthened by Law 129/2000 and successively by a new law approved in 2002 (the Statute of the NSC, Law 514/2002). The NSC is composed of seven members appointed by the Parliament, who can be dismissed according to the legal provisions. It is accountable to the Parliament, to which it has the obligation to submit an annual report. Its members may be dismissed by the Parliament only according to the conditions expressed by the Statute of the NSC. The NSC’s revenues come from fees charged to the regulated entities for the supervisory services carried out. The Com- mission is the only legal authority with powers of licensing, authorizing and issuing norms and regulations regarding supervision in the security market.

Insurance Supervisory Commission Like the NSC, the ISC was created to replace the former authority responsible for the insurance market, the Supervisory Office of Insurance and Reinsurance Activity of the Ministry of Public Finance. It is an administratively and financially autonomous authority set up through Law 32/2000 (the Insurance Regulatory Act). It is managed by a Council composed of five members, all appointed by Parliament. Its organization has been constantly improved and changed: the current organization includes specialized directorates The ISC now represents the only authority responsible for the authorization, supervision and regulation of the insurance sector.

4.5 Serbia In Serbia two financial supervisory authorities can be identified (Table 10): the Na- tional Bank of Serbia (NBS) and the Securities and Exchange Commission (SEC). The process of reorganization and development of financial supervision in Serbia started in the beginning of the 1990s with a general transition toward a market econ- 26 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 omy. During the early stages, the financial sector was affected by political difficulties in the Balkan area and the economic downturn caused by the war. The end of the war and the ensuing further growth increased the necessity to provide a legal framework for the changes in the financial system. The recent developments demonstrate that the country is indeed constructing a legislative framework that a growing financial sector requires. New Laws were passed in order to organize some sectors, like the securities market, that were still not well established. From the point of view of financial supervision and regulation, the country is also further developing and refining its framework in response to emerging needs.

National Bank of Serbia The NBS was established in 1884 under the name Privileged National Bank of the Kingdom of Serbia. Right after WWII, the Bank was nationalized under the name National Bank of Yugoslavia. In 2003, following the break-up of Yugoslavia after the war, the central bank was named the National Bank of Serbia, as an autonomous and independent institution of the Republic of Serbia. Functions, objectives and the organization of the NBS are governed by the Law on the National Bank of Serbia No. 72 of July 2003. It is the institution responsible for the conduct of monetary policy and for the supervision and the licensing of credit institutions. The main bodies of the NBS are the Monetary Board, the Governor and the Council. Management of the operations of the bank is in the hands of the Governor who is appointed by the Parliament for a five-year term. The five members of the Council are also appointed by the Parliament. The NBS’ powers concerning the supervision and the regulations of the financial sector involve the issuing and withdrawal of licenses for bank and credit institutions, the adoptions of rules governing prudential standards for banking operations, the supervision of banks’ adherence to prudential standards and the provision of regulations governing the licensing process. The Law on the NBS further stipulates that in carrying out its responsibilities the central bank acts as an autonomous and independent institution. The recent Supplement to the Law on the NBS No. 55/2004 extends the supervisory role of the NBS to the insurance sector (issuing of insurance, reinsurance, brokerage licenses) which was previously one of the roles of the Ministry of Finance.

Securities and Exchange Commission The SEC is a separate independent legal agency established in 1990 for regulating and supervising the securities market in Serbia and protecting investors. It consists of a chairman and four other members appointed by the Government. Its function is to ensure financial discipline in security trading, to issue licenses for financial exchange and supervise the operation of various actors in the financial market. The Commis- sion also has regulatory powers. In order to carry out its functions the Commission D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 27 is financed through fees from the supervised entities. When fees do not cover the operating costs, the state budget will cover the difference.

4.6 Turkey Two financial supervisory authorities operate in Turkey (Table 11): the Banking Regulation and Supervisory Agency (BRSA) and the Insurance Supervisory Office (ISO). Turkey started to modernize and liberalize its financial sector in the late 1980s. During the past two decades, in fact, the Turkish financial system has undergone a transformation toward liberalization in term of corporate governance of the participants and in terms of legislative provisions governing the financial services. The Turkish economy has been hit by three different crises during the last 15 years (1994, 2000 and 2001). The last one has driven the banking and financial sector into a severe downturn. The Turkish authorities, recognizing the structural problems of the economy, made important changes in order to put it on the right path toward development. The most significant program adopted by the Government to restructure the economy was the one for the ‘Transition to a Strong Economy’ in 2001 designed to help especially the weak banking sector.. New regulations were adopted regarding internal control, corporate governance and risk management. Some of the objectives of the recovering program decided by the government are slowly putting the Turkish banking and financial system on the right path to development.

Banking Regulation and Supervision Agency The BRSA was established in 1999 by the Banks Act No. 4389 art.3 and started its operations in 2000 as a legal entity with financial and administrative autonomy. Prior to the Bank Act, the Treasury Under-Secretariat and the Central Bank had been the two main regulatory and supervisory bodies in the banking sector. Starting in a period when the banking sector, under the threat of a new crisis and in general, in a period when the banking system was very vulnerable, the Agency had to face the difficult task of restructuring the entire banking system. Among theB RSA’ s duties are: the implementation of banking legislation; monitoring and supervising the banking system and creating a proper environment for banking and financial actors to operate in. With the creation of the BRSA, the Savings and Deposits Insurance Fund (SDIF) whose task is among others to administrate and supervise banks whose license was withdrawn by the BRSA (a Fund that was previously under the authority of the Cen- tral Bank), started to operate under the administration of the BRSA. Under amending Act No. 5020/2003, the management was separated from the BRSA. The decision- making and managing body of the Agency is the Banking Regulation and Supervi- sion Board (BRSB), which is appointed by the Council of Ministers (Cabinet) and has seven members, all appointed by the Council of Ministers. They are appointed for a six-year period and cannot be dismissed. The BRSA is accountable to the Prime 28 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Minister, who requires annual accounts of the BRSA to be audited by a committee formed by an auditor from the Supreme Court of Public Accounts, an inspector from the Prime Minister’s office and one from the Minister for Finance’s office. The supervisory system has recently been further strengthened to bring it closer to international standards of prudent regulation.7 In order to make the Turkish financial market more competitive, a new Banking Act (No.5411) was passed at the end of 2005. It contains new provisions in terms of supervision by the BRSA relating to new companies whose operations go under the supervision of the BRSA (such as financial holding companies, leasing, factoring and consumer finance companies); activities that banks may engage in are listed clearly and in compatibility with the directives of the EU; it establishes an audit committee with all members chosen from non-executive members of the board of directors, to assist board of directors for on-side and off-side supervision activities.

Insurance Supervisory Office The ISO was first established as an auditing board for the supervision of insurance companies by the law on the Auditing of Insurance Companies of 1959. It was then placed by Law 7397/1963 under the organization of the Ministry of Commerce. In 1994 it was moved to the Under-Secretariat of the Treasury. Its main objectives and responsibilities are to supervise, audit and investigate the activities of the insurance companies. Supervisors of the Office are authorized periodically to check all books, records, statements and accounting documents at the premises of insurance and reinsurance companies. The supervisory framework is life and non life (general) insurance.

5. Evaluating the Supervisory Architectures and Central Bank Role in SEE countries: a Comparative Analysis To evaluate the supervisory regimes in the SEE countries we can perform a comparative analysis using our institutional indicators on a sample of 30 countries (the sum of the SEE countries and the EU countries). Let us consider first the degree of unification of financial supervision (Figure 6). Five SEE countries out of six (except Turkey) reach the minimum level of the FSU index, while the average level of this index is 4.1. The standard deviation of the SEE countries is 1.51, while the overall sample shows a standard deviation equal to 2.71. Therefore the SEE countries show a lower and more homogeneous level of concentration.

7. They cover aspects such as capital adequacy, control and risk management, consolidated and cross-border supervision of banks, etc D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 29

Focusing on the degree of central bank involvement in supervision (Figure 7), five SEE countries out of six (except Turkey) present a level of CBFA greater than 1.75, which is the average level. The standard deviation of the SEE countries is 0.75, while the overall sample shows a standard deviation of 0.94. Thus, the SEE countries show a higher and more homogeneous level of central bank involvement in supervision. Finally each SEE institutional structure can be identified using the two indices (Figure 8). Considering the overall sample, the analysis confirms that the two most frequent regimes are polarised: on the one hand, Unified Supervisor regime (8 cases, red ball); on the other, Central Bank Dominated Multiple Supervisors regime (6 cases, white ball). The Figure depicts again the trade-off between supervision unification and central bank involvement, with one outlier (green ball). The SEE countries (yellow stars) are all characterized by Central Bank Dominated Multiple Supervisors regimes, with the exception of Turkey. The central bank role seems to matter. In the descriptive analysis we can also adopt alternative indicators of the central bank role. We use two different proxies of central bank importance: central bank independence and central bank age, assuming that an old central bank is more influential. Considering central bank independence (Figure 9), four SEE countries have a degree of central bank independence clearly greater than the average level (0.73), while for two countries – Serbia and Turkey – the independence is quite close to the average. At the same time, the central banks of three SEE countries are relatively old (Figure 10). Finally, just to add more information, let us remember that in the econometric analysis the multi-authority regime is more likely to occur when the political governance index is low and when, other things being equal, the country’s population is relatively big. All the five SEE countries with multi-authority regimes show low political governance performances (Figure 11), but also Turkey’s ranking is in the same category. Considering the population, only one SEE country with a multi-authority regime (Romania) has a population bigger than the average level. In conclusion, the study of the SEE countries seems to confirm the possible role of the institutional position of the central bank in influencing the policymaker’s choices in reshaping the financial supervision architecture. Given that the type of policymaker is unknown, the story of the PDE goes as follows. Each policymaker, in determining the future level of unified supervision, could be influenced by the actual involvement of the central bank, but under different conditions. If the policymaker is of an HH type, it should care about the effectiveness of supervision, in order to please the citizens. If the policymaker sees supervision consolidation as a welfare improvement, then the central bank involvement could be viewed as an obstacle, but only if at least one of the four reasons described above — i.e. moral hazard, conflict of interest, bureaucracy and reputation — holds, and the risks of welfare costs overcome the potential benefits in terms of informational gains. 30 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

If the policymaker is of a GH type, it wishes to please the financial constituency. In this case the PDE effect holds if, and only if, the financial constituency dislikes the unified supervision and this condition is more likely to occur if the central bank is an independent agency. What are the policy implications of our analysis? At the national level our path- dependency approach can explain both the emerging trend towards the single financial authorities, and the (so far few) cases of supervisory consolidation in favour of the central banks. Let us consider for example the three European cases in which the reforms of the supervision architecture increased (Ireland and the Netherlands) or could increase (Italy) the power of the national central bank. Ireland, the Netherlands and Italy are members of the Economic and Monetary Union, and their central banks no longer have full responsibility for national monetary policies. Therefore we can interpret these reforms as cases in which HH policymakers can increase the central bank role in supervision, given that the risks of moral hazard, conflict of interest risks, excessive bureaucratic power are likely to be low, while the information and reputational gains are likely to be high. Alternatively, we can study each of these reforms as possible cases of GH policymakers that implement the supervision regime favoured by the financial industry. In general, disentangling the two interpretations in specific country cases can be a welcome extension of the research agenda. At the European level our model says that the establishment of a single financial authority is less likely to occur with the presence of a European central bank deeply involved in supervision. Conversely, the less the European Central Bank is involved in the financial supervision architecture, the more likely the establishment of a Euro- pean Single Financial Authority will be.

6. Conclusions: Research Agenda The current worldwide wave of reforms in supervisory architectures leaves the interested bystander with a large number of questions regarding the true determinants of, and motivations behind, these changes. These questions are all the more justified because the emerging institutional structures are certainly not homogeneous across countries. An answer to these questions requires a political economy approach. Indeed, financial supervisory reform is a political process which involves many stakeholders: the political class, the central bank, the supervised entities, as well as the customers of the financial services. So the all-encompassing question is: which considerations and views prevail in the end in the decision making process, and to what extent are the decision-makers taking into account the views of these different classes of stakeholders when deciding on a reform of the supervisory structures. This paper tries to offer more analysis on these questions by looking specifically at the impact of the central bank factor in a selected sample of countries. Considering D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 31 the six SEE countries, the high involvement of the central bank in supervision seems to be correlated with a multi-authorities regime in five cases, while in one case the same relationship holds in the opposite direction: a high level of supervisory consolidation is related to a low level of central bank involvement. However, proper historical research is warranted. More specifically, in order to shed light on the determinants of the evolution of the supervisory structures it would be necessary to find out to what extent the different political explanations ofthe central bank fragmentation effect were in action in the past, country by country. The multi-authority model dominated by a central bank can be the final result of a political environment which fears a monopolist central bank as a potential source of inef- ficiency — moral hazard, conflict of interest, bureaucratic power — and at the same time can face difficulties in changing the central bank involvement in supervision, given its reputation endowment. At the same time this supervisory model can be the final result of a political class that wishes to please the banking and financial community, when the financial community dislikes the unified supervision and the central bank is a non-captured one. 32 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Tables and Figures

Figure 1: Reforms of the Supervisory Architectures per Year (1998-2009)

Figure 2: The Financial Supervision Unification Index D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 33

Figure 3: The Central Bank as Financial Supervisor Index

Figure 4: FSU and CBFA Indexes: Time Lags 34 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Figure 5: FSU and CBFA Indexes: The Trade Off (Financial Supervision Regimes)

Figure 6: Financial Supervision Unification Index: EES and EU Countries D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 35

Figure 7: Central Bank as Financial Authority Index: EES and EU Countries

Figure 8: Architectures of Financial Supervision: EES and EU Countries 36 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Figure 9: Central Bank Independence: EES and EU Countries

Figure 10: Central Bank Age: EES and EU Countries D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 37

Figure 11: Political Governance: EES and EU Countries

Figure 12: Population: EES and EU Countries 38 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Table 1. Politicians and Supervisory Reforms: When the Central Bank Role Matters

EFFECTS EPISODES BUREAUCRACY In the UK case, Goodhart 2004 and Westrup 2006 stressed that, among all EFFECT the arguments that led the Government in 1997 to establish the Financial Services Authority (FSA), removing supervision from the Bank of England could have been a quid pro quo for giving it monetary independence, on the grounds that a central bank with too many functions could be too much of a power centre within the democratic system. In Norway, due to the banking crisis in the early 1990s, the possibility of merging the BISC with the central bank was considered by a committee appointed by the Ministry of Finance. But the Parliament, in order to avoid an excessive concentration of power, ruled that the BISC continue as a separate and independent agency (Skogstad Aamo 2005). REPUTATION The difficulties in implementing reforms that reduce the central bank in- EFFECT volvement in supervision when its reputation is high are documented in several case studies. In France a reform was recently implemented, merging into one regulatory authority — Autorité des Marchés Financiers (AMF) — different financial supervision responsibilities, but the Banque de France preroga- tives remained unchanged. In 2004, after the Parmalat scandal, the Italian Government proposed a draft text of a bill, concerning a general reform of the supervisory architectures, based on the establishment of a single financial authority. The proposed reform encountered strong opposition from a bi-partisan coalition, defending the role of the Bank of Italy in promoting financial stability. The reform was rejected. Finland has opted not to adopt the unified approach in financial supervision, in contrast with the other Scandinavian countries. Taylor and Fleming 1999 claimed that the Bank of Finland involvement in supervision has to be considered in explaining this choice. In Iceland, prior to the establishment of the single financial agency, banking supervision was conducted by the central bank. In 1996, a committee was set up by the Minister for Commerce, to look at prospects of moving toward unified supervision. Mwenda and Fleming 2001 reported that only one member on the committee — the central bank official — voted against the introduction of unified financial supervision. However, the central bank obtained the ability to appoint one of the three members of the single financial authority board. Also in Germany — as Westrup 2007 reported — the Bundesbank staged a public campaign to lobby against the creation of Ba- Fin. By contrast, if the reputation of the central bank is low, or decreasing, the establishment of a single financial authority could be more likely to occur, despite its involvement in supervision. The supervisory failure of the UK central bank is well documented in Westrup 2006. The link between banking instability, central bank reputation failure and single financial authority establishment is also evident in theB al- tic unified supervisory architectures and in the case ofK orea. Estonia experienced a severe banking crisis in 1998 and 1999. In May 2001, the Estonian D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 39

Parliament adopted the Financial Supervisory Authority. Before the Act, the supervision was split into the three traditional sets of institutions. The Bank of Estonia was responsible for state supervision of banking (Live 2005). Latvia experienced banking and financial crises in 1995 and in 1998. In July 2001, the Financial and Capital Market Commission was established, as a consolidated institution. In Korea, until 1997, the central bank was responsible for banking supervision; however — as Lee noted — the Ministry of Finance dominated the central bank. Following the 1997 financial crisis, a presidential committee recommended a drastic overhaul of the organization of the central bank and the country’s supervisory structure. As a result, the former four financial supervisory authorities were combined into one integrated financial supervisory body, the Financial Supervisory Committee. Also China — Quintyn et al. 2006 — moved supervision out of the central bank in the wake of a period of financial sector distress. It is interesting to note that the reputation failure effect can hold regardless of the nature of the agency involved. In Norway — as we noted above — after the 1990s banking crisis the Ministry of Finance considered the possibility of merging the single financial authority with the central bank. 40 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Table 2. Supervisory Authorities in 102 countries: FSU Index and CBFA Index (year:2008)

Banking Securities Insurance FSU CBFA Countries Rating Weight Sector (b) Sector (s) Sector (i) INDEX INDEX Albania CB SI SI 3 1 4 2 Algeria CB,B1,B2 S - 1 -1 0 2 Argentina CB S I 1 0 1 2 Australia BI,S BI,S BI,S 7 -1 6 1 Austria U, CB U U 7 -1 6 1 Bahamas CB S I 1 0 1 2 Bahrain CB CB CB 7 0 7 4 Belarus CB S I 1 0 1 2 Belgium U U U 7 0 7 1 Bolivia B SI SI 3 0 3 1 Bosnia CB,B1,B2 S I 1 -1 0 2 Botswana CB S I 1 0 1 2 Brazil CB S CB,I 1 1 2 3 Bulgaria CB S I 1 0 1 2 Cameroon B S I 1 0 1 1 Canada BI Ss(**) BI 3 0 3 1 Chile CB,B CB,S I 1 -1 0 3 China B S I 1 0 1 1 Colombia BI CB,S BI 3 -1+1 3 1 Costa Rica B S I 1 0 1 1 Croatia CB SI SI 3 1 4 2 Cyprus CB S I 1 0 1 2 Czech Republic CB CB CB 7 0 7 4 Denmark U U U 7 0 7 1 Ecuador BI S BI 3 0 3 1 Egypt CB S I 1 0 1 2 El Salvador BI S BI 3 0 3 1 Estonia U U U 7 0 7 1 Finland BS BS I 5 0 5 1 France CB,B1,B2 CB,S I 1 -1+1 1 3 Georgia U U U 7 0 7 1 Germany U,CB U U 7 -1 6 1 Ghana CB S I 1 0 1 2 Greece CB S I 1 0 1 2 Guatemala BI S BI 3 0 3 1 Haiti CB - Is (**) 1 0 1 2 Hungary U U U 7 0 7 1 Iceland U U U 7 0 7 1 India CB,B S I 1 -1 0 2 Iran CB CB I 5 0 5 3 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 41

Ireland CB CB CB 7 0 7 4 Israel CB SI I 1 1 2 2 Italy CB,S CB,S I 1 1 2 3 Jamaica CB S1,S2 S1,S2 3 -1 2 2 Japan U,CB U U 7 -1 6 1 Jordan CB S I 1 0 1 2 Kazakhstan U,CB U U 7 -1 6 1SICB,S Kenya CB S1, S2 I 1 -1 0 2 Korea U U U 7 0 7 1 Kyrgyzstan CB S I 1 0 1 2 Latvia U U U 7 0 7 1 Lebanon B,CB CB I 1 1 2 3 Libya CB SI SI 3 0 3 2 Lithuania CB S I 1 0 1 2 Luxembourg BS BS I 5 0 5 1 Macedonia CB S I 1 0 1 2 Madagascar BS BS I 5 0 5 1 Malaysia CB S CB 3 0 3 3 Malta U U U 7 0 7 1 Mauritius CB SI SI 3 0 3 2 Mexico CB,B CB,S I 1 1 2 3 Moldova CB S I 1 0 1 2 Montenegro CB S I 1 0 1 2 Morocco CB,B S,B I,B 1 -1+1 1 2 Namibia CB SI SI 3 0 3 2 Netherlands CB,S CB,S CB,S 7 -1 6 4 New Zealand CB S I 1 0 1 2 Nicaragua U U U 7 0 7 1 Norway U U U 7 0 7 1 Pakistan CB SI,Ss** SI 3 -1 2 2 Panama B S I 1 0 1 1 Peru BI S BI 3 0 3 1 Philippines CB CB,S,Ss** I 1 1 2 3 Poland U U U 7 0 7 1 Portugal CB S I 1 0 1 2 Romania CB S I 1 0 1 2 Russia CB S I 1 0 1 2 Rwanda CB CB I 5 0 5 3 Saudi Arabia CB S CB 3 0 3 3 Singapore CB CB CB 0 0 7 4 Slovak Republic CB CB CB 7 0 7 4 Slovenia CB S I 1 0 1 2 South Africa CB,B SI SI 3 -1 2 2 Spain CB.Bs(**) CB,S I 1 1-1 1 3 42 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Sri Lanka CB S I 1 0 1 2 Sweden U U U 7 0 7 1 Switzerland U U U 7 0 7 1 Tajikistan CB CB I 5 0 5 3 Tanzania CB S I 1 0 1 2 Thailand CB S I 1 0 1 2 Trinidad Tobago CB S,CB I,CB 1 2 2 4 Tunisia CB S I 1 0 1 2 Turkey B S I 1 0 1 1 Ukraine CB SI SI 3 0 3 2 UAE CB S I 1 0 1 2 Uganda CB S I 1 0 1 2 UK U U U 7 0 7 1 USA CB,B S,Ss** I,Is(**) 1 -1 0 2 Uruguay BC,BS BC,BS I, BC 5 1 6 4 Venezuela B S I 1 0 1 1 Vietnam CB S I 1 0 1 2 Zimbabwe CB S I 1 0 1 2

The initials have the following meaning: B = authority specialized in the banking sector; BI = authority specialized in the banking sector and insurance sector; CB = central bank; G= government; I = authority specialized in the insurance sector; S = authority specialized in the securities markets; U = single authority for all sectors; BS = authority specialized in the banking sector and securities markets; SI = authority specialized in the insurance sector and securities markets.

(*) (b) = banking or central banking law; (s) = security markets law; (i) = insurance law (**) = state or regional agencies, or self-regulated agencies Source: Masciandaro (2004 and 2006) and our calculation

* FSU INDEX The index was built on the following scale: 7 = Single authority for all three sectors (total number of supervisors=1); 5 = Single authority for the banking sector and securities markets (total number of supervisors=2); 3 = Single authority for the insurance sector and the securities markets, or for the insurance sector and the banking sector (total number of supervisors=2); 1 = Specialized authority for each sector (total number of supervisors=3). We assigned a value of 5 to the single supervisor for the banking sector and securities markets because of the predominant importance of banking intermediation and securities markets over insurance in every national financial industry. It also interesting to note that, in the group of integrated supervisory agency countries, there seems to be a higher degree of integration between banking and securities supervision than between banking and insurance supervision; therefore, the degree of concentration of powers, ceteris paribus, is greater. These observations do not, however, weigh another qualitative characteristic: There are countries in which one sector is supervised by more than one authority. It is likely that the degree of concentration rises when there are two authorities in a given sector, one of which has other powers in a second sector. On the other hand, the degree of concentration falls when there are two authorities in a given sector, neither of which has other powers in a second sector. It would therefore seem advisable to include these aspects in evaluating D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 43 the various national supervisory structures by modifying the index as follows: adding 1 if there is at least one sector in the country with two authorities, and one of these authorities is also responsible for at least one other sector; subtracting 1 if there is at least one sector in the country with two authorities assigned to supervision, but neither of these authorities has responsibility for another sector; 0 elsewhere.

** CBFA INDEX For each country, and given the three traditional financial sectors (banking, securities and insurance), the CBFA index is equal to: 1 if the central bank is not assigned the main responsibility for banking supervision; 2 if the central bank has the main (or sole) responsibility for banking supervision; 3 if the central bank has responsibility in any two sectors; 4 if the central bank has responsibility in all three sectors (Table 1). In evaluating the role of the central bank in banking supervision, we considered the fact that, whatever the supervision regime, the monetary authority has responsibility in pursuing macro-financial stability. Therefore, we chose the relative role of the central bank as a rule of thumb: we assigned a greater value (2 instead of 1) if the central bank is the sole or the main authority responsible for banking supervision. 44 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Table 3. Financial Supervision in Albania

Authorities Insurance Securities for Financial Bank of Albania Supervisory Commission Supervision Committee Legal Framework Law 8269/1997 “On the Bank of Law 8080/1996 Law 8081/1996 Albania” Supervised BoA is the regulatory body SC is the ISC is the Institutions supervising all the credit supervisory body competent body for institutions for securities supervision in the market. insurance market. 1925 Establishment of the National Bank of Albania. 1990s • Law 7559/1992: establishment • Law 8080/1996: of a two-tier system with the establishes the Bank of Albania performing all SC as the sole the functions of a central bank; independent • Financial and banking crisis entity responsible in 1997 with negative effects for the regulation especially on state-owned banks. of the securities • Law 8269/1997 “On the Bank market in of Albania” and Law 8365/1998 Albania. “On banks in the Republic of Albania”: Bank of Albania is recognized as the sole authority that issues license and supervises all the banks in the Republic of Albania. • Law 8384/1998: adoption of a new banking regulation 2000s • Law 9992/2006: The new banking law stipulates the core principles in conducting banking activity in the Republic of Albania (license requirements) as well as rules and procedures to ensure a sound banking system, management and administration methods for banks, branches of foreign banks and their decision-making authorities, as well as the bank’s legal obligations consistent with the powers granted by the law D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 45

Current BoA operates as an independent SC is established ISC concentrates settings institution accountable to the as the licensing its activities on Parliament. Within its duties, BoA and supervising monitoring the licenses, supervises and regulates authority for the financial status the activities of banks and other securities market of the supervised financial institutions. and the securities undertakings. traders’ activities. It licenses and supervises the insurance and reinsurance sector. 46 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Table 4. Financial Supervision in Bulgaria

Authorities for Financial Supervision National Bank of Bulgaria Financial Supervision Commission

Legal Framework Law on the Establishment of the FSC Act of 2003 BNB 1985

Supervised Institutions BNB is the regulatory body super- Bulgarian FSC is the supervisory vising all the credit institutions body for securities and the insurance markets

1947-1989 The BNB loses most of its independence because a monopoly of state banking was created under the direction of the Communist Party

1990s • 1992: a new Law on BNB restored the distinction between commercial banks and the central bank • 1997: amendments to the Law on the Bulgarian National Bank creating 3 distinct new departments: the Issuing, the Banking and the Banking Supervision Department:

2000s • 2006: a new Law on Credit • Financial Supervision Act Institutions contains specific established the FSC merging the provisions on licensing and re- Insurance Supervisory Agency, voking licenses to the supervised the State Insurance Supervision institutions Agency and the National Securi- ties Agency Current settings The BNB is independent and The FSC is supervisory authority accountable to the National As- for the security and the insurance sembly. The managing board sector. It is an independent author- consists of a Governing Council, a ity accountable to the Parliament. Governor and three Deputy Gov- It is organized in three depart- ernors. The Governors are elected ments (the Insurance Supervi- by the National Assembly. The sory Agency, the State Insurance other three members are appointed Supervision Agency and the Na- by the President of the Republic. tional Securities Agency) and it is The BNB can issue indications and composed of seven members, the rules about financial supervision in Chairman, the Deputy Chairmen the banking system and is respon- and the other Commissioner, all sible for the issuing of licenses to elected by the National Assembly. operate in the country. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 47

Table 5. Financial Supervision in Greece

Private Authorities Capital Market Insurance for Financial Bank of Greece Commission Supervisory Supervision Committee

Legal Statute BoG (first ratified in Law 1969/91 Law 3229/2004 Framework 1927)

Supervised BoG is the regulatory body Hellenic CMC is the super- ISC is the super- Institutions supervising all the credit visory body for financial visory authority institutions intermediaries operating in for the insurance the capital market: invest- companies. ment firms, mutual fund management firms, portfolio investment companies and the securities clearing and settlement systems. 1960s CMC established as a special committee of the Ministry of Economy and Finance

1990s • Law 2076/1992: licensing • Law 1969/1991: establish- • The competent requirements for banking ment of the CMC with legal authority for services. It incorporates personality regulating listed the supervision into the Greek banking companies and other par- of insurance legislation the 2nd Banking ticipants in all the markets; market and Directive (89/646/EEC, • Law 2324/1995: legal companies was as codified by Directive framework introducing the the Ministry of 2000/12/EC). Hellenic CMC as an inde- Development • Law 2609/1998: bringing pendent authority; the operational framework • Law 2396/1996: admin- of the BoG in line with the istrative and enforcement provision of the EU Treaty power concerning super- and the ESCBs and ratify- vision and regulation of ing Art. 55A on prudent financial institutions and supervision of the Statute of markets. the BoG.

2000s • Law 2832/2000: bringing • Decision No. 5/204/2000 • Law 3229/2004 the operational framework enacts corporate gover- establishes the of the BoG in line with nance regulation for listed Private Insur- the provision of the EU companies; ance Supervi- Treaty and the ESCBs and • Law 3158/2003 enhances sory Committee ratifying Art. 55B which CMC as the competent supervised by establishes a framework for authority in relation to the the Ministry of administrative sanctions by authorization of secondary Finance. BoG. markets 48 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Current Prudent supervision of credit CMC consists of a 7-member A new authority settings and financial institutions is Board of Directors, whose is expected to carried out by the Supervi- Chairman and Vice Chairman assume its duties sion of Credit and Related are appointed by the Ministry in the beginning Financial Institutions Depart- of Finance for 5 years, and a of 2008. ment (SCRFID). 3-member Executive Com- mittee. Objectives: promote the establishment of sound conditions for the operation of the capital market. It intro- duces rules and regulations and supervises compliance with them by financial intermediaries. It is accountable to the Parliament and twice a year the chairman is obliged to appear in front of a special committee of the parliament. The Ministry of Economy and Finance also supervises the CMC. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 49

Table 6. Financial Supervision in Romania

Authorities Insurance National Securities for Financial National Bank of Romania Supervisory Commission Supervision Commission Legal Law 101/1998 (NBR’s Act) Law 52/1994 Law 32/2000 Framework replaced by Law 312/28 of June 2004 Supervised National Bank of Romania is NSC is the supervisory ISC is the super- Institutions the regulatory body supervis- commission for securities visory authority ing all the credit institutions and stock market for the insurance companies. 1940s to late • Loss of legal identity. The 1980s NBR became a mere instrument of the state. 1990s • Law 34/1991: the NBR • 1993 the NSC established regains the status and func- as a special committee of tions of a central bank. the Ministry of Economy • Law 101/1998: legal frame- and Finance by govern- work of the Bank, establishment ordinance 18/1993. ing the legal identity of the • Law 52/1994: establish- NBR. ment of the NSC as the autonomous authority responsible for securities and stock exchange. 2000s • Law 312/28 of June 2004: • Law 129/2000 and succes- • Law 32/2000 contains primary objectives sive Laws (the Statute of (the Insurance and tasks of the central bank the NSC, Law 514/2002): Regulatory Act): in terms of monetary policy revision of the legal establishment and prudent supervision. framework of the Com- of the ISC as mission clarifying objec- an autonomous tives and functions. regulatory entity. Current • NBR may issue licenses and The NSC is composed of It is managed by a settings regulations, norms, orders seven members appointed Council composed and circulars, take mea- by the Parliament to which of five members sures and apply sanctions it is also accountable. It is all appointed by to those institutions that do financed by fees charged to the Parliament. not respect operating criteria the regulated entities. It is the authority set by the central bank, It has powers of licensing, responsible for draw supervisory reports authorizing and issuing the authorization, on inspections. The NBR norms and regulations supervision and is managed by a Board of regarding supervision in the regulation of the Directors composed of nine security market. insurance sector. members appointed by the Parliament to which it is also accountable. 50 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

Table 7. Financial Supervision in Serbia

Authorities National Securities for Financial National Bank of Serbia Ministry of Finance Commission Supervision Legal Law on the National Bank of Insurance Law of Framework Serbia No. 72 of July 2003 1996 and Law on the Supplement to the Law on the National Bank of Serbia No. 55/2004 Supervised National Bank of Serbia is the SEC is the supervi- MoF was the supervi- Institutions supervisory body for all the sory commission for sory authority respon- credit and more recently for all securities and stock sible for regulating the insurance institutions. exchange. and monitoring the Insurance sector. 1990s • Law 34/1991: the NBR re- Established in 1990 gains the status and functions for regulating and of a central bank. supervising the securi- • Law 101/1998: legal frame- ties market in Serbia. work of the Bank, establishing the legal identity of the NBS. 2000s • Law on the National Bank of Serbia No. 72 of July 2003: set out the functions, the objectives and the organisation of the NBS. • Law on the Supplement to the Law on the National Bank of Serbia No. 55/2004: it contains some extensions of the supervisory power of the NBS to include also supervision of the insurance sector. Current It is the institution responsible The NSC is composed The Ministry has settings for the conduct of monetary of seven members the power to issue policy and for the supervision appointed by the licenses, regulations and the licensing of the credit Parliament to which and to carry out super- institutions. The main bodies it is also accountable. visory powers over the of the NBS are the Monetary It is financed by fees insurance companies. Board, the Governor and the charged to the regu- All these tasks passed Council. The managing of the lated entities. under the competence operations of the bank is in the It has powers of of the National Bank hands of the Governor who is licensing, authorizing of Serbia by Law on appointed by the Parliament. and issuing norms and the Supplement to the regulations regard- Law on the National ing supervision in the Bank of Serbia No. security market. 55/2004 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 51

Table 8. Financial Supervision in Turkey

Authorities for Banking Regulation and Insurance Supervisory Office Financial Supervision Supervision Agency Insurance Supervision Law Legal Framework Banks Act No. 4389/1999 7397/1994 Supervised Institutions Banking Regulation and Supervi- ISO is the supervisory authority sion Agency is the authority for the insurance sector. responsible for the issuing of regulations and supervision of the banking sector. 2000s Established by the Banks Act, it started its operations in 2000 as a legal entity with financial and administrative autonomy. Amendment Act No. 5020/2003 with which the management of the Savings and Insurance Funds has been separated by the BRSA. A new Banking Act (No.5411/2005) containing new provisions in terms of supervision by the BRSA. Current settings The decision-making body of the The ISO has one president and a Agency is the Banking Regula- number of insurance supervisory tion and Supervision Board experts and actuaries. (appointed by the Council of Ministers. It has seven members all appointed by the Council of Ministers for a six-year period. The BRSA is accountable to the Prime Minister. 52 D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53

References Abrams, R.K., Taylor, M.W., 2002. Assessing the Case for Unified Sector Supervision. FMG Spe- cial Papers n.134, Financial Markets Group, LSE, London. Alesina, A., Tabellini, G., 2003. Bureaucrats or Politicians? Part II: Multiple Policy Task. Discus- sion Paper n.2009, Harvard Institute of Economic Research, Harvard University, MA. Arnone, M., Gambini, A., 2007. Architecture of Supervisory Authorities and Banking Supervision. In: Masciandaro D. and Quintyn M. (Eds.), Designing Financial Supervision Institutions: Inde- pendence, Accountability and Governance, Edward Elgar, Cheltenham, 262-308. Barth, J.R., Nolle, D.E., Phumiwasana, T. and Yago, G. 2002. A Cross Country Analysis of the Bank Supervisory Framework and Bank Performance. Financial Markets, Institutions & Instruments, vol. 12, n.2, 67-120. Beck, T., Demirguc-Kunt, A., and Levine R., 2001. Law, Politics and Finance, World Bank Policy Research, Working Paper n. 2585, Washington D.C. Bini Smagh,i L., 2007. Independence and Accountability in Supervision. In: Masciandaro D. and Quintyn M. (Eds.), Designing Financial Supervision Institutions: Independence, Accountability and Governance, Edward Elgar, Cheltenham, 41-62. Cîhak, M., Podpiera, R., 2007a. Experience with Integrated Supervisors: Governance and Quality of Supervision. In: Masciandaro D. and Quintyn M. (Eds.), Designing Financial Supervision Institutions: Independence, Accountability and Governance, Edward Elgar, Cheltenham, 309- 341. Cîhàk M. and Podpiera R., 2007b. Does More Integrated Supervision mean Better Supervision? Finlawmetrics 2007, Bocconi University, mimeo. Dalla Pellegrina L. and Masciandaro D., 2008, Politicians, Central Banks and the Shape of Finan- cial Supervision Architectures, Journal of Financial Regulation and Compliance, forthcoming. De Luna Martinez, J., Rose, T.A., 2003. International Survey of Integrated Financial Sector Super- vision. Policy Research Working Paper Series, n. 3096, World Bank, Washington D.C. Garcia Herrero, A., and Del Rio P., 2003. Implication of the Design of Monetary Policy for Finan- cial Stability. 24th SUERF Colloquium, Tallin. Société Universitaire Européenne de Recherches Financières, Wien. Goodhart, C.A.E, Schoenmaker, D. 1995. Should the Functions of Monetary Policy and Banking Supervision be Separated? Oxford Economic Papers, vol. 47, n. 4, 539-560. Goodhart, C.A.E., 2004. Financial Supervision from an Historical Perspective: Was the Develop- ment of Such Supervision Designed, or largely Accidental?, Conference on the Structure of Financial Regulation, Bank of Finland, mimeo. Goodhart, C.A.E., 2007. Introduction. In: Masciandaro D. and Quintyn M. (Eds.), Designing Fi- nancial Supervision Institutions: Independence, Accountability and Governance, Edward Elgar, Cheltenham, 12-26. Honohan, P., Laeven, L., (Eds.), 2005. Systemic Financial Crises: Containment and Resolution, Cambridge University Press, Cambridge UK. La Porta, R., Lopez de Silanes, F., Shleifer, A., Vishny, R.W., 1997. Legal Determinants of External Finance. Journal of Finance, vol. 52, n.3, 1131-1150. D. MASCIANDARO, M. QUINTYN, South-Eastern Europe Journal of Economics 1 (2010) 7-53 53

Llewellyn, D.T., 2005. Institutional Structure of Financial Regulation and Supervision: the Basic Issues. In: Fleming A., Llewellyn D.T. and Carmichael J. (Eds.), Aligning Financial Supervi- sion Structures with Country Needs, World Bank Publications, Washington D.C, 19-85. Masciandaro, D., 2004. Unification in Financial Sector Supervision: the Trade off between Central Bank and Single Authority. Journal of Financial Regulation and Compliance, vol. 12, n. 2, 151-169. Masciandaro, D., (Ed.). 2005. Handbook of Central Banking and Financial Supervision in Europe, Edward Elgar, Cheltenham. Masciandaro, D., 2006. E Pluribus Unum? Authorities Design in Financial Supervision: Trends and Determinants. Open Economies Review, vol. 17, n. 1, 73-102. Masciandaro, D., 2007. Divide et Impera: Financial Supervision Unification and the Central Bank Fragmentation Effect, European Journal of Political Economy, 285-315. Masciandaro D., 2008, Politicians and Financial Supervision Outside the Central Bank: Why Do They Do It?, Journal of Financial Stability, forthcoming. Masciandaro, D., Quintyn, M., (Eds.), 2007. Designing Financial Supervision Institutions: Inde- pendence, Accountability and Governance, Edward Elgar, Cheltenham. Masciandaro, D. Quintyn M., 2008, Helping Hand or Grabbing Hand? Politicians, Supervision Regime, Financial Structure and Market View, North American Journal of Economics and Fi- nance, forthcoming. Masciandaro, D., Quintyn, M., 2009, Reforming Financial Supervision and the Role of the Central Banks: a Review of Global Trends, Causes and Effects (1998-2008), CEPR Policy Insight, n.30, 1-11. Mwenda, K.K., Fleming, A.,2001. International Development in the Organizational Structure of Financial Services Supervision, Conference on the Challenges of Unified Financial Supervi- sion, Central Bank of Estonia, Tallinn. Mwenda, K.K., 2006. Legal Aspects of Financial Services Regulation and the Concept of a Uni- fied Regulator, The International Bank for Reconstruction and Development/The World Bank, Washington. Pigou, A., 1938. The Economics of Welfare, London, Macmillan & Co. Shleifer, A., Vishny, R., 1998. The Grabbing Hand, Cambridge, Harvard University Press. Skogstad Aamo, B., 2005. Norway. In Masciandaro D. (Ed.), The Handbook of Central Banking and Financial Authorities in Europe, Edward Elgar, Cheltenham, 200-225. Taylor, M.W., Fleming, A., 1999. Integrated Financial Supervision: Lessons from Northern Euro- pean experience. Policy Research Working Paper n.2223, The World Bank. Westrup J., 2007. Independence and Accountability: Why Politics Matters. In: Masciandaro D. and Quintyn M. (Eds.), Designing Financial Supervision Institutions: Independence, Accountability and Governance, Edward Elgar, Cheltenham, 117-150.

South-Eastern Europe Journal of Economics 1 (2010) 55-72

Prospects of changes in regional economic structures since eu accession

K. MATTASa*, C. CIOBANUb, D. PSALTOPOULOSc a,b Aristotle University of Thessaloniki, Greece c University of Patras, Greece

Abstract In recent years, the European Union (EU) has expanded by including two additional “less developed” new members, Bulgaria and Romania, and in the near future several other neighboring countries are on the list. The case of the region of East Macedonia and Thrace, a remote region that neighbors Bulgaria, might offer some insights into the changes which will occur in the economy of these new EU members. Changes in the structure of the regional economy are traced by estimated various indices of structural changes using two input-output tables, the 1980 I-O, a year before Greece’s accession to EU, and the 1997 I-O, one and a half decade after the implementation of several EU supported programmes. The cause of structural changes cannot be identified by applying this methodology, only the final outcome in terms of sectoral structure. Results reveal that significant transformations took place in this regional economy altering the interdependence between producing and consuming sectors. It is not clear that this change has moved the whole regional economy to a more competitive level as highly supported sectors grew substantially.

JEL Classification: R100, R150, R580. Key words. Greece, East Macedonia and Thrace, input-output analysis, structural change, regional development

* Corresponding Author: Aristotle University of Thessaloniki, Department of Agricultural Eco- nomics, School of Agriculture, P.O. Box 225, 54124 Thessaloniki, Greece e-mail: [email protected] 56 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

Introduction The 1981 accession of Greece into the EU inflated the future prospects for spectacu- lar growth and transformation for remote and less developed regions in this country. These expectations are paralleled in today’s expectations of new EU members, Bulgaria and Romania, and several other Balkan countries on the waiting list. These aspirations witnessed throughout the Balkan region provided the impulse to investigate the structural changes occurring in a Greek region, East Macedonia and Thrace, neighboring Bulgaria, fifteen years after the area’s inclusion in the EU. Structural changes are viewed here under an Input-Output (IO) framework. The terms “structural” and “technological” change overlap to some extent in the IO literature. Probably the best resolution of the ambiguity induced by this term is owed to Carter (1970), who refers to “technological” change as replacement of one production process by another and “structural” change as a change in input requirements, new products, and the relative size of sectors within an economy. Neverthe- less, the identification of methods that measure sectoral interconnectedness is crucial as development planners prefer to expand sectors with extensive interindustry ties, rather than those with weak interindustry connections (Diamond 1974). The IO approach is particularly well-suited to the analysis of structural changes, given its disaggregated nature and its attention to tracing intersectoral connections (Rose and Miernyk 1989). In this context, the objective of this paper is the intertemporal analysis of structural changes in the regional economy of East Macedonia and Thrace, a NUTS 2 less developed area in Greece. This is accomplished by a comparison of the 1980 and 1997 regional IO tables and the computation of various IO indicators which reflect structural change. A widely used hybrid technique, the so-called Generation of Re- gional Input-Output Tables (GRIT) procedure (Jensen, Mandeville and Karunarante 1979), is employed to generate regional IO information. The choice of the East Mac- edonia and Thrace study region for this analysis is justified by the fact that several of its structural and developmental characteristics in the early 1980s resemble the current situation in the adjacent Bulgaria; thus, an investigation of structural changes in this region might offer useful insights into the changes which could occur in the economy of Bulgaria after its accession to the EU. Consequently, the choice of 1980 as the base-year for this analysis is justified by the need to account for study-region structural characteristics before the country’s accession into the EU (i.e. 1980); in a rather similar manner, the choice of 1997 as the “second” base-year (and not of the more recent 2000 and 2005, for which Greek national IO tables are available) is justified as follows: first, the choice of 1997 ensures that structural change investigated here is associated with a rather satisfactory (in terms of its length) period of 17 years, during which all EU institutional and policy “environment” has been fully applied in Greece; second, more recent Greek IO tables were constructed through the use k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 57 of ESA 95 and thus a comparison of tables built under different accounting systems would have given rise to an incompatibility problem; finally, unlike the 1980 Greek IO table which is expressed in Greek Drachmas, more recent ones (i.e. 2000, 2005) are expressed in Euros, something that would complicate comparisons. The paper is organized as follows: the next section describes the GRIT technique adopted in the construction of regional tables, followed by a presentation of the several measures and indicators, employed here to investigate structural changes. Then the results of the analysis are presented, ending with the main conclusions and policy implications.

Theoretical framework Regional input-output modelling Isard (1951, 1953) and Leontief (1953) were the first scholars who sought regional extensions to the IO model. From the early days of these efforts, the high cost of obtaining the necessary regional data through survey methods forced researchers to develop short-cut or non-survey methods which facilitate the construction and use of regional IO tables without incurring prohibitive costs. Applications of regional IO analysis are provided among others by Miernyk et al. (1967), Schaffer (1976), Sawer and Miller (1983), Karunarante (1989), Midmore (1993), and Tzouvelekas and Mat- tas (1995). The terms “survey” and “non-survey” suggest the existence of two well- defined and mutually exclusive groups, but in practice, most of the IO tables are “hybrid” ones, constructed by semi-survey techniques, employing primary and secondary sources, to a greater or lesser extent (Round 1983). One of those techniques, extensively used, is the Generation of Regional Input-Output Tables technique (Johns and Leat 1987). The GRIT technique, developed by Jensen, Mandeville and Karunarante (1979), was originally applied to the production of the IO tables for the regions of Queens- land, Australia, from both national IO tables and other sources. It is based on a com- bination of non-survey methods, but allows modifications of mechanically produced tables at the discretion of the analyst, to produce more accurate regional tables. The GRIT method estimates the flows of regional intermediate demand by applying the employment-based Cross Industrial Location Quotient (CILQ) to corresponding elements of the national matrix. After deriving initial estimates of regional technical coefficients, the GRIT procedure permits the insertion of superior data, where appropriate (an issue judged by the discretion of the analyst), to replace mechanically derived estimates. The superior data may come from survey data, published statistics and other sources (Johns and Leat 1987, Psaltopoulos and Thomson 1993, Psaltopou- los 1995). As noted by Tohmo (2004) other adjustment formulas such as the FLQ (Flegg et al., 1995) can also be applied for the mechanical regionalization of national 58 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

IO tables. However, as in several other studies which used GRIT (indicatively, Johns and Leat 1987; Psaltopoulos et al. 2004), the CILQ method was chosen here.

Measures of structural changes IO analysis has been extensively employed to compare the structure of production over time and across countries. Rasmussen (1956) used an IO model in measuring changes in the structure of production in Denmark between 1947-1949. In this semi- nal study he proposed a method for the measurement of sectoral linkages using the open static IO model. Chenery and Watanabe (1958) used IO tables for Finland, Italy, Japan and the USA, to compare the structure of production in these countries, and revealed the existence of similar structural patterns. Simpson and Tsukui (1965) used the US 1947 and Japanese 1955 IO tables in comparing production structures. Yan and Ames (1965) developed a new method of measuring structural change in the US economy between 1919-1929. Carter (1970) studied structural changes in the US economy between 1939-1961 by measuring changes in the input coefficients and Staglin and Wessels (1972) examined intertemporal structural changes in the German economy. After a short lull in research interest in this area during the 1970s, interest in the study of structural change re-emerged in the 1980s. Several researchers focused their attention on the comparative analysis of the structure of production on a disaggregated sectoral level (Kubo 1985, Kubo et al. 1986a, 1986b). Relevant individual country studies include those of Skountzos (1980), who examined structural changes in the Greek economy between 1958-1970, Forsell (1988) who measured the structural changes in the Finish economy between the 1960s and the 1970s, and Urata (1988) who investigated intertemporal variations in the Soviet economy for the period 1959-1972. Also, Skolka (1989) examined structural variations in the Austrian economy and Lee (1990) studied structural changes in the US agricultural sectors during 1972-1982. More recently, Sonis et al. (1996), Sonis and Hewings (1998), Cho, Sohn and Hewings (1999), Guo and Planting (2000) used new decomposition approaches (such as field of influence) to visually display structural changes, and thus provide a more comprehensive view of changes in economy over time. Finally, several structural change indices have been used by Skountzos et al. (2007) in the context of a study on the Greek economy. Among a wide range of indices employed in the above-mentioned studies, we here briefly described those applied in this study. k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 59

Linkage indices. A measure of structural change can be found by considering changes in the elements of the Leontief inverse matrix.1 Thus, two indices, which can be used to describe an inverse matrix and in turn the changes in its elements, are defined as follows (Rasmussen, 1956):

(1)

and

(2)

where U.j is the index of power of dispersion and Ui. is the index of sensitivity of dispersion. U.j and Ui. are also measures of backward linkages and forward linkages, respectively.2 The index of power of dispersion describes the relative extent to which an increase in final demand of industry j is dispersed through the system of sectors. The meaning of U.j may also be explained by mentioning that the index shows the magni-

1. Throughout this study the following notation is used. is the direct require-

ments coefficient matrix which is also called the technical coefficients matrix, where Χij is sector j’s direct input from sector i, and Xj is total output of sector j; is the total requirements matrix which is often referred to as Leontief inverse matrix.

2. The backward linkage of sector j would be given by the sum of the elements in the jth column of the total requirements matrix. Hence, , j = 1, 2, …, n, where b.j is the backward linkage of sector j and zij is the element of total requirements matrix. Similarly with the backward linkages the forward linkage for sector i is given by the sum of the elements in the ith row of the total requirements matrix. Hence, , i = 1, 2, …, n, where bi. is the forward linkage of sector i and zij was defined above. 60 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

tude of changes in the system of sectors, caused by a change in industry j. If U.j > 1, then the impact of a unitary increase in the final demand of sector j draws heavily on the system of sectors. If U.j < 1, then a unitary increase in the final demand of sector j will have relatively small impact on the system of sectors. The index of sensitivity of dispersion indicates the extent to which a change in the system of sectors will affect sector i. If Ui. > 1, then a unitary increase in the final demand of the system of sectors will have bigger impact on sector i than on the other sectors (and vice versa in case of Ui. < 1). There has been an extensive literature on the use of linkage indices to examine the structure and functioning of an economy; some authors have attempted to modify these indices, while others have been very critical of the whole approach (Cella 1984, Hewings et al. 1989, Soofi 1992).

The U.j and Ui. indices are based on averages. Averages are sensitive to extreme values and may give misleading results. Consequently, these indices do not fully describe the structure of a particular sector. For instance, it is possible that an increase in final demand for the product of a particular sector, characterized by a high index of power of dispersion may not affect other sectors. Such a situation would arise if a particular sector draws heavily on only one or few industries. To overcome this difficulty, the indices of coefficient of variation are usedby 3 Rasmussen (1956) as additional indices. A high V.j can be interpreted as an index showing that a particular sector draws heavily on one or a few sectors and a low V.j shows that a sector draws evenly from the other sectors. The Vi.’s can be interpreted similarly. Indices of concentration. To compare changes in the structure of production of the regional economy at two distinct points of time, the indices of concentration for sectoral transactions was proposed (Soofi 1992):

(j = 1, 2, …, n)

3. The indices of coefficient of variation are defined as follows: ,

j = 1, 2, …, n and , i = 1, 2, …, n where Vij is the backward index of

variation, Vi. is the forward index of variation and Z was defined before. k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 61

where G.j is the backward index of concentration, V.j is the backward index of variation and cij = zij/b.j.

(i = 1, 2, …, n)

where Gi. is the forward index of concentration, Vi. the forward index of variation and dij = zij/bi.. Since these indices are increasing functions of the size of the inverse matrix, they can be expressed as percentages of the theoretical maximum indices. Hence, the percentage concentration indices G* are calculated as follows:

(3)

G.j and Gi. are measures of variation in intersectoral transactions. When there is no variation in a sector’s sales to (purchases from) other sectors, then the sum of sector sales (purchases) will determine the number of sector connections. Generally, given the sum of the ith sector’s sales (purchases), a large value for G implies more sector connections. In contrast, a small value for G implies fewer intersectoral sales (purchases). In the extreme case where G = 0, total skewness in sectoral transactions predominates, implying maximum concentration. When G is expressed as percentage, G* of 100 indicates complete uniformity of sectoral transactions. In addition, Soofi (1992) constructed a general index (GI) representing the combined effects of the ranks of U and G as follows:

GI = α(RG – RU) + RU (4) where RG represents the ranks of concentration indices and RU represents the ranks of linkage indices. α is the weight to be attached to the concentration index; this parameter reflects the planners’ preference for the sectors with uniform sectoral sales and purchases.4 If RG = RU, then the ranking of G or U alone should be sufficient in decision- making. For α = 0,5 the GI index value, for a sector which is ranked number one by both U and G indices, is equal to 1. If RG > RU, the sectors with a lower measure of concentration and high linkages are ranked lower than sectors with the same linkage value but higher measure of concentration. If RG < RU, given two sectors with equal linkage index but different concentration measures, the GI will rank the sector with the larger concentration measure higher. Note that the GI modifies the ranking sectors

4. GI, G and U are used as generic terms to include all general indices, concentration indices and linkage indices, respectively, regardless the data used in their calculation. 62 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 with wide differences in values for G and U. Moreover, the GI will have a small effect in the ranking of sectors with small differences between G and U rankings.

Empirical results The regional economy East Macedonia and Thrace is a EU region (NUTS 2 level) that covers the north- eastern part of Greece. The geopolitical location of East Macedonia and Thrace is a factor that might have contributed to its isolation and marginalisation. However, the recent significant changes to the map of Europe, EU enhancement, have changed the region’s prospects for growth (Regional innovation and technology transfer strategies 2001). The particularly favorable complex of development incentives in East Macedonia and Thrace, supported by the Community Support Framework and the Regional Development Plans, offer significant investment opportunities mainly for the establishment of new technology-based firms. Briefly, the regional economy is dominated by relatively few sectors, agriculture, trade and construction being the most important. These sectors account for 74% of output and 82% of employment in 1980 and 62% of output and 72% of employment in 1997. In 1997, agriculture contributed 12% to the regional output and 39% to the regional employment, relative to 1980 figures, 21% and 51%, respectively. On the other hand, the contribution of services sector to the regional output and employment was increased. Construction contributed approximately 9% to the regional output in 1997 and accounted for 5% in employment. The corresponding numbers for 1980 were 13% and 10%, respectively. The main manufacturing sector, food and beverages, which is considered a traditional sector in the region, contributed 8% to the regional output in 1980 and 1% to the local employment (the corresponding figures for 1997 were 6% and 2%).

Indices of structural changes The indices of structural changes are estimated on the base of two 34 sector (Ap- pendix 1) IO tables, the 1980 I-O, a year before Greece’s accession to EU, and the 1997 I-O, and sixteen years of interventions to change the structure of the regional economy. The GRIT technique was used to generate both tables (Jensen et al., 1979), involving the mechanical regionalization of the national IO tables as well as the insertion of superior data obtained from secondary sources and selected interviews with local policy makers and stakeholders. Table 1 presents two indices of structural changes: the index of power of dispersion (backward linkage index) and the index of sensitivity of dispersion (forward linkage index). These two indices provide a quantitative description of the structure of the economy for the period 1980-1997, concerning the 34 local production sectors. k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 63

From the backward linkage indices for 1980, it is obvious that trade, food and beverages, hotels and restaurants and livestock are the sectors with the most backward connections with the other sectors. For 1997 the backward linkage indices, column (2) of Table 1, identify food and beverages, leather industry, metal products and tobacco products as sectors having the largest number of backward transactions with the rest of the economy. Regarding the rate of change between the two time points, column (3) of Table 1, the agricultural sectors (cereals, vegetables and fruits) show an important increase in backward linkages. In other words these sectors, not only together but separately too, affected more heavily the whole economy in 1997 than in 1980. By contrast, for the livestock sector the changes are reversed. Comparing 1980 and 1997, it seems that a substantial increase (over 10%) has occurred in the backward linkages of sectors such as clothing, leather industry and chemicals. The other manufacturing sectors seem to influence the system of industries to a relatively small extent, recording a smaller increase in their backward linkages. A substantial decrease (over 10%) has occurred in the backward linkages of the transport equipment sector. Among the non-manufacturing industrial sectors, only construction showed a considerable increase in backward linkages during the time span. Looking at the service sectors, substantial increases have occurred in financial intermediation and the other service sectors. The indices of sensitivity of dispersion, column (4) of Table 1, for 1980 show as sectors with highest forward linkages trade, real estate, renting and business activities, financial intermediation, transport and communication and cereals. For 1997 the forward linkages, column (5) of Table 1, identify trade, real estate, renting and business activities, metal products, chemicals and livestock as sectors having the most forward connections with the other sectors. In other words these sectors appeared to be more strongly influenced by a general increase in final demand than other sectors. Regarding the indices of sensitivity of dispersion and their changes between 1980 and 1997, column (6) of Table 1, it can be seen that sectors with increases of over 10% in forward linkages include fruits, extraction of crude oil and natural gas, textiles, leather, wood, rubber and plastic products, chemicals, metal products and real estate, and renting and business activities. This means that the impact of a unit increase in the final demand in the whole system of sectors on each of the above sectors was bigger in 1997 than in 1980. Sectors with a decrease of over 10% in their forward linkages are forestry, mining and quarrying, transport equipment, construction, trade, financial intermediation, public administration and defense, and health. Furthermore, Table 2 presents the indices of concentration and the general indices for 1980 and 1997. For 1980, the measures of concentration identify trade, food and beverages, hotels and restaurants, livestock and health sectors as having the most backward connections with other industries. Trade, financial intermediation, real es- 64 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

Table 1. Linkage indices for 1980 and 1997

Indices of power of dispersion (U.j) Indices of sensitivity of dispersion (Ui.) 1980 1997 (2):(1) 1980 1997 (5):(4) 1980 1997 (2):(1) 1980 1997 (5):(4) Sectors (1) (2) (3) (4) (5) (6) 1a 0.830 (28) ** 0.931 (22) 1.122 1.326 (5) 1.265 (6) 0.954 2 0.843 (27) 1.010 (12) 1.199 0.935 (14) 1.008 (14) 1.078 3 0.824 (30) 0.980 (14) 1.189 0.857 (19) 1.176 (8) 1.372 4 1.236 (4) 0.966 (16) 0.782 1.171 (6) 1.267 (5) 1.082 5 0.797 (31) 0.750 (33) 0.941 0.887 (18) 0.787 (27) 0.887 6 0.950 (17) 0.914 (25) 0.963 0.817 (30) 0.742 (30) 0.909 7 0.795 (34) 0.726 (34) 0.914 0.795 (34) 0.955 (16) 1.201 8 0.826 (29) 0.791 (30) 0.957 0.937 (13) 0.797 (26) 0.850 9 1.546 (2) 1.421 (1) 0.919 1.093 (8) 1.156 (9) 1.058 10 1.079 (9) 1.150 (4) 1.065 0.843 (23) 0.840 (23) 0.995 11 1.029 (12) 1.081 (9) 1.051 0.857 (20) 1.028 (12) 1.200 12 0.850 (26) 0.988 (13) 1.163 0.838 (25) 0.767 (29) 0.916 13 0.987 (15) 1.175 (2) 1.190 0.799 (32) 0.880 (20) 1.101 14 1.109 (6) 1.052 (11) 0.949 0.893 (16) 1.015 (13) 1.136 15 1.111 (5) 1.053 (10) 0.948 1.164 (7) 1.065 (10) 0.915 16 0.796 (32) 1.088 (8) 1.367 0.818 (29) 1.299 (4) 1.588 17 1.032 (11) 0.941 (21) 0.911 0.854 (21) 0.942 (17) 1.103 18 0.948 (18) 0.963 (17) 1.016 0.916 (15) 0.896 (19) 0.978 19 1.101 (7) 1.171 (3) 1.063 0.948 (12) 1.372 (3) 1.448 20 0.910 (20) 0.950 (20) 1.044 0.821 (28) 0.846 (22) 1.031 21 1.023 (13) 0.807 (28) 0.789 1.010 (10) 0.834 (24) 0.825 22 0.904 (22) 0.961 (18) 1.063 0.846 (22) 0.828 (25) 0.979 23 1.016 (14) 0.852 (27) 0.839 1.040 (9) 1.036 (11) 0.997 24 1.051 (10) 1.102 (7) 1.048 1.007 (11) 0.874 (21) 0.868 25 1.837 (1) 0.957 (19) 0.521 2.277 (1) 1.780 (1) 0.782 26 1.285 (3) 1.121 (6) 0.873 0.835 (26) 0.785 (28) 0.939 27 0.945 (19) 0.926 (23) 0.979 1.328 (4) 1.232 (7) 0.928 28 0.850 (25) 0.973 (15) 1.144 1.453 (3) 0.933 (18) 0.642 29 0.909 (21) 0.870 (26) 0.957 1.480 (2) 1.689 (2) 1.141 30 0.867 (23) 0.920 (24) 1.061 0.843 (24) 0.726 (34) 0.861 31 0.865 (24) 0.788 (31) 0.911 0.801 (31) 0.739 (31) 0.922 32 1.080 (8) 0.777 (32) 0.719 0.827 (27) 0.737 (32) 0.891 33 0.976 (16) 1.141 (5) 1.169 0.888 (17) 0.972 (15) 1.094 34 0.795 (33) 0.800 (29) 1.006 0.796 (33) 0.733 (33) 0.921

Column (1) and column (2) have been calculated according to formula (1). Column (4) and column (5) have been calculated according to formula (2). ** Figures in parentheses indicate the sectoral ranking in terms of linkage indices. a Sector’s classification may be found in Appendix 1. k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 65

Table 2. Concentration indices and general indices for 1980 and 1997

Backward Forward Concentration indices General indices Concentration indices General indices * b * (G .j) (GI.j) (G i.) (GIi.) Sectors 1980 1997 1980 1997 1980 1997 1980 1997

(1) (2) (3) (4) (5) (6) (7) (8)

1 15.99 (30) *** 26.69 (30) 29 26 77.03 (4) 69.96 (7) 4.5 6.5 2 18.56 (28) 46.13 (24) 27.5 18 46.49 (12) 45.65 (20) 13 17 3 26.64 (27) 39.96 (25) 28.5 19,5 37.76 (15) 64.33 (10) 17 9 4 73.90 (4) 36.71 (28) 4 22 69.08 (6) 70.06 (6) 6 5.5 5 6.42 (31) 25.23 (31) 31 32 43.95 (13) 38.6 (24) 15.5 25.5 6 53.87 (14) 60.86 (12) 15.5 18,5 19.96 (28) 19.11 (27) 29 28.5 7 0 (34) 2.85 (34) 34 34 0 (34) 64.09 (11) 34 13.5 8 27.54 (26) 39.21 (26) 27.5 28 52.82 (10) 40.43 (23) 11.5 24.5 9 79.56 (2) 80.19 (1) 2 1 57.36 (9) 68.51 (9) 8.5 9 10 61.35 (8) 67.84 (5) 8.5 4,5 1.12 (33) 0.57 (33) 28 28 11 59.33 (11) 66.53 (7) 11.5 8 26.96 (23) 61.35 (12) 21.5 12 12 35.39 (24) 64.18 (9) 25 11 31.69 (19) 15.91 (30) 22 29.5 13 58.42 (12) 66.94 (6) 13.5 4 2.22 (32) 7.28 (32) 32 26 14 62.20 (7) 59.46 (16) 6.5 13,5 24.21 (25) 54.97 (18) 20.5 15.5 15 38.42 (22) 56.00 (20) 13.5 15 47.64 (11) 57.60 (14) 9 12 16 4.08 (32) 65.94 (8) 32 8 24.06 (26) 78.60 (3) 27.5 3.5 17 60.98 (9) 60.29 (14) 10 17,5 29.39 (22) 60.69 (13) 21.5 15 18 48.63 (16) 56.85 (19) 17 18 42.45 (14) 46.13 (19) 14.5 19 19 60.01 (10) 53.64 (21) 8.5 12 36.69 (16) 69.42 (8) 14 5.5 20 47.75 (17) 58.16 (17) 18.5 18,5 22.08 (27) 40.64 (22) 27.5 22 21 16.88 (29) 25.05 (32) 21 30 5.59 (30) 35.20 (26) 20 25 22 46.15 (18) 62.68 (10) 20 14 31.50 (20) 42.27 (21) 21 23 23 55.85 (13) 51.77 (22) 13.5 24,5 59.75 (8) 71.35 (5) 8.5 8 24 65.19 (6) 74.79 (3) 8 5 61.31 (7) 55.57 (16) 9 18.5 25 82.74 (1) 60.24 (15) 1 17 90.96 (1) 90.81 (1) 1 1 26 78.52 (3) 76.05 (2) 3 4 30.72 (21) 38.17 (25) 23.5 26.5 27 37.49 (23) 57.46 (18) 21 20,5 75.21 (5) 78.34 (4) 4.5 5.5 28 35.24 (25) 60.60 (13) 25 14 82.70 (2) 56.18 (15) 2.5 16.5 29 44.91 (19) 51.24 (23) 20 24,5 79.75 (3) 88.05 (2) 2.5 2 30 40.25 (20) 62.01 (11) 21.5 17,5 33.55 (18) 0 (34) 21 34 31 39.25 (21) 38.83 (27) 22.5 29 9.93 (29) 17.63 (28) 30 29.5 32 67.61 (5) 35.77 (29) 6.5 30,5 26.18 (24) 17.05 (29) 25.5 30.5 33 52.91 (15) 70.50 (4) 15.5 4,5 35.58 (17) 55.21 (17) 17 16 34 0 (33) 10.27 (33) 33 31 4.17 (31) 13.62 (31) 32 32

The numbering of sectors is in accordance with table 1. Columns (1), (2), (5) and (6) have been calculated according to formula (3). Columns (3), (4), (7) and (8) have been calculated according to formula (4). *** Figures in parentheses indicate the sectoral ranking in terms of concentration indices. b α = 0,5 66 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 tate, renting and business activities, cereals and transport and communications are the sectors with the largest number of forward connections. The backward general index identifies trade, food and beverages, hotels and restaurants, livestock, wood industry and health sectors as having the closest value to one. The forward general index identifies trade, financial intermediation, real estate, renting and business activities, cereals, transport and communications and livestock as the leading sectors. For 1997, the ranking of measures of concentration identifies the following sectors as having the largest number of backward transactions with the rest of the economy: food and beverages, hotels and restaurants, construction, other services and tobacco products. Trade, real estate, renting and business activities, chemicals, transport and communications, electricity and water and livestock are identified as sectors with the largest number of forward connections with other sectors. The backward general indices list the following sectors as having the smallest values: food and beverages, leather industry, hotels and restaurants, tobacco products, other services and construction. The forward general indices identify trade, real estate, renting and business activities, chemicals, livestock, metal products and transport and communications as the leading sectors. The concentration indices reveal that trade is the sector with the most backward and forward connections with other industries both in 1980 and 1997. Food and beverages appears to have the first and the second strongest backward connections with the rest of the economy in 1997 and 1980, respectively. Real estate, renting and business activities sector appears in 1980 and 1997 as having the third and the second, respectively, forward connections with other sectors. Correlation coefficients representing the correspondence among the ranking of the above computed indices are reported in Table 3. Results suggest that the coefficients of rank correlation between various indices for 1980 and 1997 are either quite low or quite high. These relationships indicate that during this period continuous transformations took place in the economic structure of the region, with several sectoral activities maintaining their position in the economy and others maximizing or minimizing their importance to the economy of the region. Between 1980-1997 coefficients of rank correlation vary from 0.316 among linkage indices and 0.637 among general indices. Generally, it can be observed that both row and column measures show weak or negative correlation among the regional indices during the period under study. Also, a moderate correlation is observed between forward concentration index and general index, (0.650), the forward general index and linkage index, (0.641), and the forward general indices, (0.637). The correlation coefficients between the ranking of backward linkage index and forward linkage index, backward linkage index and row concentration index, backward linkage index k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 67

Table 3. Rank correlation coefficients among various indices for 1980 and 1997

80 97 80 97 80 97 80 97 80 97 80 97 U.j U.j Ui. Ui. G.j G.j Gi. Gi. GI.j GI.j GIi. GIi. 80 U.j 1 0.316 0.334 0.231 0.893 0.246 0.138 0.136 0.970 0.311 0.234 0.186 97 U.j 1 -0.038 0.198 0.304 0.760 -0.076 -0.018 0.334 0.927 -0.060 0.081 80 Ui. 1 0.635 0.160 -0.229 0.873 0.632 0.232 -0.143 0.973 0.641 97 Ui. 1 0.121 -0.152 0.529 0.915 0.174 0.020 0.617 0.976 80 G.j 1 0.371 0.115 0.071 0.971 0.363 0.121 0.095 97 G.j 1 -0.155 -0.204 0.328 0.929 -0.225 -0.191 80 Gi. 1 0.615 0.103 -0.140 0.954 0.568 97 Gi. 1 0.092 -0.133 0.650 0.975 80 GI.j 1 0.362 0.158 0.134 97 GI.j 1 -0.159 -0.067 80 GIi. 1 0.637 97 GIi. 1 and row general index are –0.038, -0.076 and –0.060, respectively, showing that a negative correlation exists between these pairs of indices. The weak and negative correlation among most of the ranked indices for linkage and concentration is an indication that the structure of production of the regional economy during 1980-1997 recorded significant dissimilarities.

Conclusions When Greece joined the European Community in 1981, East Macedonia and Thrace, faced threatening structural problems, and EU funding was utilized to revamp the whole structure of the region. The construction of the IO regional model and the computation of several indices of structural changes provide useful insights into the structure of the regional economy and the embodied changes between 1980 and 1997, changes that to a large extent would be attributed to initiation of an array of supporting schemes over this long period. Recommended IO indices by a large number of IO scholars have been estimated and then the changes recorded in this time stretch are compared. Estimations based on the linkage indices have revealed that some sectors recorded an increase and others a decrease in their importance as stimulators of output changes in the economy between 1980 and 1997. The degree to which different sectors affect the whole system of sectors through the demand for intermediate inputs has changed considerably during this period (agricultural sectors, clothing, leather industry, chemicals, construction, financial intermediation and other services affected more heavily the system of sectors in 1997 than in 1980). Likewise, observations can be drawn on the degree to which the activity in the system of sectors affected the activity in a particular sector. Some sectors noticed an expansion (fruits, extraction of crude oil and natural gas, textiles, leather, wood, rubber and plastic products, chemicals, metal 68 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 products and real estate, renting and business activities) and others a contraction (forestry, mining and quarrying, transport equipment, construction, trade, financial intermediation, public administration and defense, health) in their dependence on the activity of the whole system of sectors. The measurements of concentration have shown that decision-makers should attach expansion priorities (especially) to sectors such as trade, food and beverages, real estate, and renting and business activities. Each of the measures and indices employed in this study cumulatively provides a comprehensive view of the structure of the regional economy of East Macedonia and Thrace and the changes recorded between 1980-1997. According to these results, all of the regional sectors have experienced significant structural changes during the period under study. The analysis shows that the economy of East Macedonia and Thrace depends heavily on agricultural activities. Agriculture continues to be an important and indispensable part of the regional economy, strongly linked with the rest of the economy. Moreover, there are several service sectors as well as construction and manufacturing sectors that have fairly strong linkages within the regional economy. A comparison of the ranking of the economic sectors based on linkage indices, concentration indices and general indices for 1980 and 1997, indicates that a weak or negative correlation exists between these measurements. This enables us to conclude that severe changes in the structural characteristics of the regional economy have occurred during the period under study, though there are no indications that the economy was radically transformed into a more competitive one, as sectors highly subsi- dized by the EU prevailed in the region. Agriculture is still an important sector for the regional economy, especially if the links with food processing are taken into account. However, the comparatively low competitiveness of several agricultural sub-sectors could also undermine a (rather) competitive sector such as food processing. In terms of other sectors, it can also be noted that economic activities which have seen an expansion (such as textiles and wood products) are currently (for some years now) facing increased competition from abroad, and hence, rather uncertain prospects. On the other hand, linkages seem to have declined for sectors that have gained ground in terms of competitiveness (i.e. financial intermediation, construction). The overall picture of this analysis, though it cannot be claimed to be a complete and comprehensive one, still offers valuable insights into future prospects, but also a warning signal that beyond any EU funding schemes local policy-making remains a determinant factor for permanent and promising change in a region. k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 69

References Carter, A. 1970. Structural change in the American economy. Harvard University Press, Cam- bridge, Massachusetts. Cella, G. 1984. The IO measurement of interindustry linkages. Oxford Bull Econ Statist 46:73-84. Chenery, H.B., and T. Watanabe. 1958. International comparisons of the structure of production. Econometrica 26:487-521. Chletsos, M., Chr. Kollias, and G. Manolas. 1997. Structural changes and their impact on the relationship between wages, productivity and labor demand in Greece. Center of Planning and Economic Research Athens Discussion Paper No. 64. Cho, B., J. Sohn, and J.D. Hewings. 1999. Structural change in the Korean economy between 1975 and 1995. Regional Economics Laboratory University of Illinois Discussion Paper 99-T-6. Credit Bank 1999. The Counties of Greece. Credit Bank, Athens Diamond, J. 1974. The analysis of structural constraints in developing economies: a case study. Oxford Bull Econ Statist 36:95-108. Flegg, A.T, C.D. Weber and M.V. Elliott. 1995. On the appropriate use of location quotients in generating regional input-output tables. Reg Studies 29: 547-561. Forsell, O. 1988, Growth and changes in the structure of the Finish economy in the 1960s and 1970s. In: Ciaschini M. (eds.) Input-output analysis: current development, Chapman and Hall, London. Guo, J., and MA. Planting. 2000. Using input-output analysis to measure US economic structural change over a 24 year period. Macerata, Italy Paper presented at the 13th International conference on input-output techniques. Hewings. GJ., DM. Fonseca, J. Guilhoto, and M. Sonis. 1989. Key sectors and structural change in the Brazilian economy: a comparison of alternative approaches and their implications. J Pol Modeling 11:67-90. Isard, W. 1951. Interregional and regional input-output analysis: a model of a space economy. Rev Econ Statist 33:318-328. Isard, W. 1953. Some empirical results and problems of regional input-output analysis. In: Leontief W (eds.) Studies in the structure of the American economy, Oxford University Press, London. Jensen, R.C., T.D. Mandeville, and N.D. Karunarante. 1979. Regional economic planning: generation of regional input-output analysis. Croom Helm, London. Johns, P.M., and P.M.K. Leat. 1987. The application of modified GRIT input-output procedures to rural development analysis in Grampian region. J Agr Econ 32:243-256. Karunarante, N.D. 1989. Australian development issues: an input-output analysis. Newcastle upon Tyne, Avebury. Kubo, Y. 1985. A cross-country comparison of interindustry linkages and the role of imported intermediate inputs. World Devel 13:1278-1298. Kubo, Y., J.De. Melo, S. Robinson, and M. Syrquin. 1986a. Interdependence and industrial structure. In: Chenery H., Robinson S., Syrquin M. (eds.) Industrialization and growth: a comparative study, Oxford University Press, New York. 70 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

Kubo, Y., S. Robinson, and M. Syrquin. 1986b. The methodology of multisector comparative analysis. In: Chenery H., Robinson S., Syrquin M. (eds.), Industrialization and growth: a comparative study, Oxford University Press, New York. Lee, C. 1990. Growth and changes in the structure of the US agricultural economy, 1972-82: an input-output perspective. Econ Systems Res 2:303-311. Leontief, W. 1953. Studies in the structure of the American economy. Oxford University Press, London. Mattas, K., A. Pagoulatos, and D.L. Debertin. 1984. Building input-output models using non-survey techniques. Southern Rural Development Center, Mississippi Series 721. Midmore, P. 1993. Input-output forecasting of regional agricultural policy impacts. J Agr Econ 44: 284-300. Miernyk, W., E.R. Bonner, J.H.Jr. Chapman, and K. Shellhanner. 1967. Impact of the space program on a local economy: an input-output analysis. West Virginia University Press, Morgantown. Mylonas, N.A., 1989. Analytical input-output tables of the Greek economy – year 1980. Center of Planning and Economic Research, Athens. National Statistical Service of Greece 1998. National accounts of Greece 1988-1997. National Statistical Service of Greece, Athens. Psaltopoulos, D. 1995. Input-output analysis of Scottish forestry strategies. Department of Eco- nomics, University of Aberdeen, unpublished PhD thesis. Psaltopoulos D., and K.J. Thomson. 1993. Input-output evaluation of rural development: a forestry- centered application. J Rural Stud 9:351-358. Psaltopoulos, D., K.J. Thomson, S., Efstratoglou, J. Kola, and A. Daouli. 2004. Regional social accounting matrices for policy analysis in lagging EU rural regions, Eur Rev of Agric Econ, 31:149-178 Rasmussen, P.N. 1956. Studies in intersectoral relations. North-Holland, Amsterdam Regional innovation and technology transfer strategies (2001) Region of East Macedonia and Thra- ce. http://liaison.xan.duth.gr/ritts_emth/eng/index.htm Rose, A., and W. Miernyk. 1989. Input-output analysis: the first fifty years. Econ Systems Res 1:229-271 Round, J.I. 1983. Nonsurvey techniques: a critical review of the theory and the evidence. Int Reg Sci Rev 8:189-212 Sawer C.H., and R.E. Miller. 1983. Experiments in regionalization of a national input-output table. Environ Planning A 15:1501-1520 Schaffer, W.A. 1976. On the use of input-output models for regional planning studies in applied regional science. Martinus Nijhoff, Leiden Simpson, D., and J. Tsukui. 1965. The fundamental structure of input-output tables: an international comparison. Rev Econ Statist 47:434-446 Skolka, J. 1989. Input-output structural decomposition analysis for Austria. J Pol Modeling 11:45- 66 Skountzos, T. 1980. Structural changes in the economy: intertemporal analysis in input-output framework. Center of Planning and Economic Research, Athens k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72 71

Skountzos, T., N. Stromplos, A. Vozikis, and F. Theofanides. 2007. Intersectoral relations of the Greek economy at the national and regional levels. Study no. 7, Academy of Athens, Athens. Sonis, M., G.J.D. Hewings, and J. Guo. 1996. Sources of structural change in input-output systems: a field of influence approach.Econ Systems Res 8:15-32 Sonis, M., and G.J.D. Hewings. 1998. Economic complexity as network complication: multire- gional input-output structural path analysis. Ann Reg Sci 32:407-436 Soofi, A. 1992. Industry linkages, indices of variation and structure of production: an international comparison. Econ Systems Res 4:349-375 Staglin, R., and H. Wessels. 1972. Intertemporal analysis of structural change in the German economy. In: Carter AP, Brody A (eds.) Input-output techniques, North-Holland, Amsterdam Tohmo, T. 2004. New developments in the use of location quotients to estimate regional input- output coefficients and multipliers.Reg Studies 38: 43-54. Tzouvelekas, V.M., and K. Mattas. 1995. Revealing a region’s growth potential through the internal structure of the economy. Int Advances Econ Res 1:304-313 Urata, S. 1988. Economic growth and structural change in the Soviet economy, 1959-72. In: Cias- chini M (eds.) Input-output analysis: current development, Chapman and Hall, London Yan, C., and E. Ames. 1965. Economic interrelatedness. Rev Econ Stud 32:290-310 72 k. MAttas et al., South-Eastern Europe Journal of Economics 1 (2010) 55-72

Appendix 1

Industry Description (sector no.) 1 Cereals 2 Vegetables 3 Fruits 4 Livestock 5 Forestry 6 Fishing 7 Extraction of crude oil and natural gas 8 Mining and quarrying 9 Food and beverages 10 Tobacco products 11 Textiles 12 Clothing 13 Leather industry 14 Wood industry 15 Paper and publishing 16 Chemicals 17 Rubber and plastic products 18 Non-metal products 19 Metal products 20 Machinery and equipment 21 Transport equipment 22 Other industries 23 Electricity and water 24 Construction 25 Trade 26 Hotels and restaurants 27 Transport and communication 28 Financial intermediation 29 Real estate, renting and business activities 30 Public administration and defense 31 Education 32 Health 33 Other services 34 Domestic services South-Eastern Europe Journal of Economics 1 (2010) 73-86

ECONOMICS EDUCATION: CAN LIBERAL ARTS MAKE ECONOMICS MORE APPEALING?

SERDAR ONGAN* Istanbul University, Turkey

Abstract Far better and more efficient teaching and understanding of economics, which predominantly encompasses human behaviors, is of the utmost importance. Both the correctness of personal financial decisions that individuals will make and the success of economic policies implemented and to be implemented at the macro level, require a through knowledge of economics and its fundamental rules. This study is an attempt to attract attention to the importance of utilizing literature, theatre, cinema, and humor, which are matters as everyday as economics, rather than employing classical presentation methods in economics education, in order to render economics more enjoyable, understandable, and appealing to college students.

JEL Classification: A20, A22, A23, A29. Key words. Economics Education, Literature-Theatre, Cinema, Humor

* Corresponding address: Istanbul University, Department of Economics, Beyazit-Istanbul, Turkey. e-mail: [email protected] This work was supported by the Research Fund of the Istanbul University. Project number: 3282 74 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86

Introduction One of the primary responsibilities and therefore a significant expenditure item of a social state, education is a long-term and expensive investment. In this context, the economic dimension of education or, in other words, education economics is extensively covered in various studies (Sarı and Uğur, 2006; Parem and Russell, 2006; Vibeke, 2005). In fact, there are 19 million students and 900,000 edu- cators in total from primary education schools up to higher education and the annual education budget accounts for 13.6% of the consolidated/centralized state budget in Turkey (MEB, 2008). Therefore, the economic aspects of education are frequently discussed and have an importance in Turkey and other developing countries. In spite of this significance, economics education is rarely given the attention its importance merits and is not covered extensively enough. However, positive correlations have been pointed out even between economics education and economic growth (Grimse Ana Lee, 2000). It can be argued that those individuals who get economics education can behave more rationally in making choices in their own lives.

2. The Importance of Economics Education Economics education can help individuals to act in line with cost-benefit analysis in the decision-making processes of their private daily lives, including work, finance and political issues (Stiller, 1983). According to Alan Greenspan, a basic economics education offered during primary and secondary school years will largely prevent young people from making wrong financial decisions that can adversely affect their later lives (Nysed, 2002). It can be argued that such an evaluation at an individual level is likely to have positive economic results for the society. A comparative study based on regression analysis has been carried out among the colleges in the states of the USA with and without economics education for at least one semester. In this analysis covering the period between 1982-1997, a statistically positive and more significant relationship has been found between the growth rate in gross state product (GSP) and economics education in the states offering economics education in their colleges compared to those states without such an education program (Grimse Ana Lee, 2000). It is pointed out that in countries like Japan and Korea, which struggled with recession during the 1990s, economics education based on the provision of basic economic knowledge played an important role in helping these countries to escape recession (Chin-Young, 2000). It is also argued that the economics education offered in the socialist countries equips individuals with an economic perspective and thinking, which played a significant role in the economic reform process of these countries (Shen and Shen, 1993; Shiller et al., 1991). When China entered into diplomatic contacts with the USA in 1979, the teaching of economics was totally Marxian and socialist and it contained neither contemporary S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 75 economic theories nor empirical analyses. Economics was treated as a philosophy rather than a science. There were no written materials in university libraries on any characteristic of the Chinese economy, and economics professors conducted no empirical research on China (Chow, 2000). Moreover, even though lawyers represent a completely different discipline and professional group, those who have boosted their human capital with an economics education are shown to earn more compared to their peers who have not received such an education (Craft and Baker, 2003).

3. Teaching and Understanding Economics In spite of this importance of economics, it is known that Thomas Carlyle1 defines economics as a dismal science (Barber, 1967; Galbraith, 1977; Heilbroner, 1986 and Oser and Brue, 1988), voicing his reaction to Malthus’ work “An Essay on the Prin- ciple of Population” (Malthus, 1798). On the other hand, students sometimes regard economics, which encompasses unforeseen human behavior, as boring, uncertain, and difficult to understand compared to traditional positive sciences like physics, mathematics and biology, which are based on foreseeable natural laws. At this point it would not be incorrect to assert that academicians in the field of economics frequently experience differences in opinion regarding economic theories and policies due to the nature of economics (Frey, Schneider, Pommerehne and Gil- bert, 1984; Alston, Kearl and Vaughan, 1992a, 1992b) or that even when they come to an agreement on a certain issue, they fail to express this openly and strongly (Fuller and Geide-Stevenson, 2003), which reinforces the views above.

“If you put two economists in the same room, you get two opinions, unless one of them is Lord Keynes, in which case you get three opinions.” Winston Churchill

While it is a fact that only a few of the many students who get an economics education will eventually become professional economists, it should not be forgotten that all of these students will be using their economic knowledge throughout their lives (Klein, 1999). Therefore, it is of the utmost importance to choose teaching methods in economics education that will make the topics more interesting, understandable, and attrac- tive for students and that will ensure their active participation in the classes. Empiri- cal studies on pedagogical techniques show that most academicians in the field of

1. Thomas Carlyle (1795-1881) Scottish satirical writer, essayist, and historian. 76 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 economics still mainly rely on classical presentation techniques of chalk and talk (Becker and Watts, 2001). It is emphasized that certain new pedagogical techniques, which involve the active participation of students in experimental games, should replace these classical teaching techniques (Becker, 2000; Hazlett, 1999; Gremmen and Potters, 1997). A very interesting, even unique experimental device can be considered for use in economics education at this stage.

3.1. The Phillips Machine (MONIAC) Phillips had a deep interest in the Keynesian economy and he discovered obvious parallels between the configuration of the IS-LM model and the macro-stabilization matters focused on by its users and the continuous time dynamic systems and the control problems analyzed by electrical engineers. Phillips was an electrical engineer at the same time, which must have been a key factor in this insight (Laidler, 2001). Building upon his knowledge of hydrodynamics as well, Phillips built a pedagogic machine that enabled students to quickly and visually understand the complex relations, which may be difficult to grasp from time to time, between such several macroeconomic variables as exchange rates, interest rates, savings, investments, government expenditures, taxes, imports and exports. Phillips published an article in 1950, in which he outlined his idea to design a mechanism called the “Phillips Machine” or “MONIAC”2 (Monetary National In- come Automatic Computer) as a mechanical tool and an economic model (Phillips, 1950). The Phillips Machine was approximately 2 meters high and 1.2 meters wide and the circulation of money in an economy is symbolized on the machine via the col- ored water that passes through the translucent plastic pipes while the accumulation of water in a series of translucent tanks stand for money stocks in the same economy (Barr, 2000). The economic system is presented by Phillips as an engineering problem (Leeson 2000), which was much later called “Hydraulic Keynesianism” (Cod- dington, 1976). Consistent with the variables above, the machine visually illustrates the following Keynesian equilibrium condition that economy students know very well: Y = AE = C + I + G + (X − M)

2. The name of the machine called the MONIAC in the USA is derived from the words MONEY, ENIAC (Electronic Numerical Integrator and Computer), and MANIAC which symbolizes that it is an extraordinary, crazy thing (Lerner, 1952). The name of the Phillips Machine was inspired from (Maniac → Moniac). S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 77

The figure below is a simplified drawing that portrays the mechanical functioning of the Phillips machine, which will be considered in this respect in this study.

Chart:1: Simplified Version of the Phillips Machine

Moreover, academicians in the field of economics suggest the use of literature (Watts, 2002; Hartley, 2001; Watts and Smith, 1989), cinema (Leet and Houser; Mateer, 2004), music (Tinari and Khandke, 2000) and humor (Clotfelter, 1996) in economics courses to make the students more interested in these classes. Academicians teaching economics can be more effective at explaining economic concepts and theories if they make use of literature and theatre works that reflect all aspects of our daily lives, just as economics, rather than filling the sessions with statistical and analytical tools that contain complex mathematical models and econometric studies (Kish-Goodling, 1998). 78 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86

4. Economics and Philosophy Several philosophers have made analyses of the development and evolution of a variety of economic theories and thus contributed greatly to the science of economics and they can be considered as the under-laborers of economists (Little, 1995). Aristotle has a specific importance among the first thinkers of the early ages, as he dealt exclusively with economic events. According to Aristotle, having personal belongings is pleasant and legitimate. Having your friends benefit from your posses- sions is a particular pleasure. The way to achieve this pleasure intersects with having private property. The famous philosopher used the concept of economics (oeconom- ics), which he limited to household management and obtaining the usage values required to make a living, for the first time between 384 - 322 BC (Finley, 1970). An- other famous philosopher, Thomas Aquinas, outlined many basic concepts of today’s modern economics a long time ago through his considerations on the requirement to bring about reasonable arrangements in terms of equality, fair prices, and fair wages as well as on the view that the changes in the value of money should be decided upon by the society, not the ruler (Blaug, 1991).

5. Economics and Literature & Theatre The Wonderful Wizard of Oz by Frank Baum is often used as a pedagogical tool to teach economics. The characters and stories in the book cover the populist monetary expansionary policies at the end of the 19th century, which also coincides with heated debates on bimetallism. The book is able to present the topics in money, banking and economic history classes in a more understandable and enjoyable way (Rockoff, 1990). It is also possible to evaluate the borrowing and lending relationships in the historical and philosophical perspective of the Middle Ages by studying the characters in The Merchant of Venice by Shakespeare, highlighting the element of “interest” in the modern capitalist economies of today (Kish-Goodling, 1998). In the famous Robinson Crusoe by Defoe, we, the economists, come across the existence and applications of very significant economic concepts and theories along with various socio-psychological behavior and views of the characters in terms of the location of the setting, its characteristics, heroes and developments as the story unfolds. Crusoe’s optimal selection as shown in the chart below, which is also called the “Robinson Crusoe Economy,” will help students beginning economics to understand the basic philosophy of economics easily and effectively. In this novel with a simple economic setting, Crusoe is a single producer and at the same time a consumer of only two products (coconuts– collected on the beach and leisure– time not spent col- lecting coconuts) and makes the optimal choice (Wooton, 2003). S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 79

Chart 2: Robinson Crusoe Economy

The whole life of Robinson Crusoe on the island as the main character of the novel is a kind of application of the Life Cycle-Permanent Income Hypothesis of Irving Fisher, while at the same time fulfilling the Keynesian open economy (other islands in this case) equilibrium Υ = AE = C + I + G + X with all of its parameters, which is well-known to students of economics (Hartley, 2001). Apart from being a story like the ones mentioned above, Guns, Germs, and Steel by the biologist-physiologist Jared Diamond (Diamond, 2004) is a very interesting novel that ties the differences in the level of development between primitive societies/today’s countries to the biological, geological and geographical characteristics of the natural environment inhabited by these societies. This book gives answers to questions involving the ultimate aim of economics and the fundamental issue and concept of growth-development/non-development with simple but equally impres- sive explanations and is likely to be very useful for students of economics. It is also argued that good economic analyses are based on the assumption that human beings will act rationally, just as in classical detective novels. Based on this analogy, the leading characters created by two economists using the nickname Mar- shall Jevons in a series of three novels make use of economics and game theories to solve crimes (Elzinga and Breit, 2002). In addition to being a very important and famous masterpiece, the play Hamlet features many everyday theories and practices of psychology, politics, and economics (MacCary, 1998). The vital significance of the fundamental variables of various theories and practices of economics is demonstrated by the imagined absence of main characters from the aforementioned play. 80 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86

Laidler, one of the pioneers of monetarist economics, argues that a monetary policy in which money (M) does not exist within the model would be like a production of Hamlet that lacks one of the main characters of the play — the Ghost (Laidler, 2003). Similarly, Minsky makes Keynes (without uncertainty), who made a considerable contribution to the concept of uncertainty that is influential on the behaviors of decision makers in economics, resemble the play Hamlet without the Prince, another main character, when this very concept is lacking (Minsky, 1975). The example of a Hamlet without the Prince is used by Lachmann, one of the pioneers of the Austrian School, in an investment theory without a capital theory (Lachmann, 1956), and by the Post-Keynesian Kregel as a Cambridge macroeconomics without money (Kregel, 1985).

6. Economics and Cinema The movie, A Beautiful Mind, is probably the best way to explain the Nash equilibrium, which forms the fundamental concept of game theory and was developed by John Nash, a good mathematician who also had mental disorders (Dixit, 2005). Modern Times, a classic movie by Charlie Chaplin produced in 1936, transforms the economic and psychological effects of industrial technology such as efficiency and scale economies from a tragedy to a comedy and provides important clues to the economic development of the world and especially of the USA (Cohen, 2004). Similarly, the 1946 movie called It’s A Wonderful Life shows how the financial system can collapse if all the depositors rush to withdraw their money from the banks (Leet and Houser, 2003). The Turkish comedy movie, Banker Bilo, which was produced in 1980, covers the financial problems caused by the banker rush on the bank experienced in the Turkish economy. Both of these films help us to watch-observe and comprehend the applications and results of money, banking, and monetary policies very easily (Orta, 2007). The 1987 production Wall Street is also considered very useful for finance students to understand the concepts of the mixed economy, financial markets, public regulations, and asymmetrical information systems (Dyl, 1991 and Belden, 1992). Similarly, On the Waterfront, the 1954 movie directed by Elia Kazan, covers workers’ unions and corruption, while analyzing such concepts as capitalism and the working class. The 2005 documentary Wal-Mart: The High Cost of Low Price, directed by Robert Greenwald, strikingly portrays how Wal-Mart rejects unionization and, consequently, obtains high profits, illustrating the personal and social costs of these earnings (Darlington, 2009). In another movie directed by Robert Zemeckis in 1994, Forrest Gump (Tom Hanks) enters the shrimping industry. But he has trouble finding shrimp and also the industry is a highly competitive industry. However, a hurricane destroys all the other shrimp boats, except Forrest’s. Since Forrest has the only boat left, he gains a S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 81 temporary monopoly and catches all the shrimp (Sources of Monopoly, Competition) (Pfeiffer, 1997).

7. Economics and Humor The educational use of humor goes all the way back to Babylonian times (Amoo et al. 2002). The educational use of humor - which strengthens the communication between the teacher and the student, makes the classes more interesting and fun and as- sists in the better understanding of topics and concepts - has been studied extensively (Aylor and Oppliger, 2003; White, 2001; Berk, 1996; Bryant et al., 1980; Kaplan et al., 1977). It can also be very effective to make use of humor in economics education. The use of humor to explain certain concepts and theories in different fields of economics can be seen in various studies (Blinder, 1974; Christopher, 1980; Edi and Barbara, 1980; Dennis, 1982; Clotfelter, 1996). There seems to be no better way than the anecdote below to illustrate the primary effect of money supply on inflation and economy, which has been frequently pointed out by Milton Freidman, who outlined the fundamental principles of monetarism in Studies in the Quantity Theory of Money he edited (Freidman, 1956).

“A student was taking the economics course given by Milton Friedman at the Chicago University and was so tired due to studying late on the previous night that he fell asleep during the class. Seeing this sleeping student, Friedman wanted to make a joke and knocked on his desk ask- ing him to answer the question he had just asked the class. The student was shocked but became wide-awake and replied “I am sorry, Sir. I missed the question, but the answer is ‘to increase the money supply’”. Anonymous

Another anecdote,

With simultaneous low inflation and low unemployment in the US per- sisting for some time, certain policy-makers at the Fed are beginning to think NAIRU stands for Nothing About Inflation (is) Related (to) Un- employment. Chris Varvares 2002

8. Economics and Religion In addition to economics, religion is another great constituting agent of human history (Marshall, 1920). It is certainly possible to accept sinners’ performance of pen- ance by proxy - paying someone to perform sacrifices on their behalf in the early 82 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86

Christian Church (Mahoney, 1987) — as well as similar practices in Islam such as the zakat and sadaqa, which have been institutionalized over time, as the foundations of such concepts of today’s modern economies as tax justice and wealth distribution (Eskicioğlu, 1989). According to Weber, who arrives at the economics–religion relationship from a different perspective, the spirit of capitalism overlaps with that of Protestantism (To- run, 2002). In other words, the capitalist mentality is intrinsic in Protestant ethics.3 The most important virtue delivered to the economic life by the Protestant ethics, which considers successful economic activities and revenues a religious activity (Bo- dur,1991), is advocating the awareness that regular and methodical work in one’s profession is a sort of religious duty (Turkdogan, 1981).

9. Conclusion The teaching and comprehension of economics, which is the leading branch of science that directs and shapes human life — in other words, economics education — is of the utmost importance. The correctness of the financial decisions taken and to be taken by individuals and the success of macroeconomic policies depend largely on this. Therefore, apart from classical presentation techniques, it is important to adopt new methods that will make economics courses more enjoyable and understandable for students and encourage their active participation. In addition to several pedagogical techniques, introducing pieces of literature, theatre, cinema, humor and other liberal arts that are actually part of daily life, just as economics is, will prove very useful in economics education. Un- derstanding and teaching economics through the use of such popular works, that have endured the passage of time and that are widely remembered, will be much easier, more effective, and enjoyable.

References Alston, R.M., Kearl, J.R. and Vaughan, M.B.(1992a). “Is There a Consensus Among Economists in The 1990’s”, The American Economic Review, 82(2), 203-209. Alston, R.M., Kearl, J.R. and Vaughan, M.B.(1992b). Consensus Among Economists: The Role of Employment Affiliation. Unpublished Paper. Presented at the Empirical Studies on Economic Consensus Session, Western Economic Association, San Francisco, Calif., July 1992. Amoo, T., Friedman, L.W., and Friedman, H.H.2002. “Using Humor in the Introductory Statistics Course”, Journal of Statistics Education, 10(3).

3. Weber, M. 1904. The Protestant Ethic and the Spirit of Capitalism, Translated by Peter Baehr and Gordon C. Wells in 2002, Penguin Books, USA. S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 83

Aylor, B., and Oppliger, P.2003. “Out-of-class Communication and Student Perceptions of Instruc- tor Humor Orientation and Socio-communicative Style”, Communication Education, 52(2), 122-/134. Barber, W. 1967. A History of Economic Thought, Harmondsworth: Penguin Books Barr, N.2000. “Developments Between Phillips and the London School of Economics,” in Robert Leeson, eds., A. W. H. Phillips: Collected Works in Contemporary Perspective. Chapter 10, Cambridge: Cambridge University Pres. Becker, W.E. 2000. “Teaching Economics in the 21st Century, Journal of Economic Perspectives, 14(1), 109-119. Becker, W.E., and Michael W. 2001. “Teaching Methods in U.S. Undergraduate Economics Cours- es”, Journal of Economic Education 32(3), 269-279. Belden, S. 1992. “A Comment on Wall Street: A Case in Ethics”, Financial Practice & Education 2(1), 53–54. Berk, R.A. 1996. “Student Ratings of 10 Strategies For Using Humor in College Teaching”, Jour- nal of Excellence in College Teaching, 7(3), 71-92. Blaug, M. 1991. St.Thomas Aquinas, 1225-74 (Pioneers in Economics), Edward Elgar Publishing Ltd. Blinder, A.S. 1974.“The Economics of Brushing Teeth”, Journal of Political Economy, 82(4), 887- 891. Bodur, H.E. 1991. Modern Kapitalizm’in Doğmasında Dinin Rolü”, Atatürk University Theology Faculty Journal, Number:10, 80-108, Erzurum. Bryant, J., Comisky, P.W., Crane, J. S., and Zillmann, D. 1980. “Relationship Between College Teachers’ Use of Humor in the Classroom and Students’ Evaluations of Their Teachers”, Jour- nal of Educational Psychology, 72(4), 511-519. Chin-Young, K. 2000. “Comparative Study on Economic Education: Korean and Japanese Schools”, NIRA (National Institute for Research Advancement) Review, Spring 7(2), 41-43. Clotfelter, C. P.1996. On The Third Hand: Humor in The Dismal Science, An Anthology, Ann Ar- bor, Michigan: University of Michigan Press. Coddington, A.1976. “Keynesian Economics: the Search for First Principles.” Journal of Economic Literature,14(4), pp. 1258-1273. Cohen, D.2004. Our Modern Times: The New Nature of Capitalism in The Information Age, Cam- bridge, Mass. ; London: MIT. Chow, G.2000. The Teaching of Modern Economics in China, Comparative Economic Studies (2000) 42, 51-60. Craft, R.K. and Baker, J.G. 2003. “Do Economists Make Better Lawyers? Undergraduate Degree Fields and Lawyer Earnings”, Journal of Economic Education, 34(3), 263-281. Christopher, U.L. 1980. “The Birth of Stagflation: A Fable”, Journal of Political Economy, 88(6), 1255-1259. Darlington,R. 2009.Roger darlington’s World. http://www.rogerdarlington.co.uk/index.shtml Dennis, S. 1982. “Macroeconomic Policy and the Optimal Destruction of Vampires”, Journal of Political Economy, 90(3), 647-655. Diamond, J. 2004. Tüfek, Mikrop ve Çelik, Tübitak Yayınları. Dixit, A. 2005. “Restoring Fun to Game Theory”, Journal of Economic Education, 36(3), 205-219. Dyl, E.A. 1991. “Wall Street: A Case in Ethics”, Financial Practice & Education, 1(1), 49-51. Edi, K. ve Barbara, S.1980. “Tales of Horror From Ivory Towers”, Journal of Political Economy, 88(1), 210-212. Elzinga, K.G and Breit,W.2002. “Economics As Detective Fiction”, Journal of Economic Educa- tion, 33(4), 367-376. Eskicioğlu, O.1989. “Modern Vergi Anlayışı ve Zekat.” D.E.Ü. İlahiyat Fakültesi Dergisi, Sayı: V, İzmir, Turkey Friedman, M.1976. Studies in The Quantity Theory of Money, Chicago: University of Chicago Press. Galbraith, J. 1977. The Age of Uncertainty, Boston: Houghton Mifflin. Gremmen, H.J.F.M., ve Potters, J.J.M. 1997. “Assessing The Efficacy of Gaming in Economics Education”, Journal of Economic Education, 28(4), 291-303. Finley,M.1970. Aristotle and Economic Analysis, The Past and Present Society. Frey, B.S., Pommerehne, W.W., Schneider; F. and Gilbert, G. 1984 “Consensus and Dissension Among Economists: An Empirical Inquiry”, The American Economic Review, 74 (5): 986-994. Fuller, D. and Stevenson, D.G. 2003. “Consensus Among Economists: Revisited”, Journal of Eco- nomic Education, 34(4), 369-387. Grimes, W.P. and Lee, O.D. 2000. “Economic Education and Economic Growth”, Atlantic Eco- nomic Journal, 28(4), 490. Hartley, J.E. 2001. “The Great Books and Economics”, Journal of Economic Education, Volume 31(2), 147-159. Hazlett, D. 1999. Economic Experiments in the Classroom, Reading, MA: Addison Wesley Long- man. Heilbroner, R. 1986. The Worldly Philosophers, Sixth edition. London: Penguin Books. Kaplan, R.M., and Pascoe, G.C. 1977. “Humorous Lectures and Humorous Examples: Some Ef- fects Upon Comprehension and Retention”, Journal of Educational Psychology, 69(1), 61-65. Kish-Goodling, D.M. 1998. “Using The Merchant of Venice in Teaching Monetary Economics”, Journal of Economic Education 29 (4), 330–39. Klein, D. 1999. What Do Economists Contribute?, New York: New York University. Press. Kregel, J. 1985. “Hamlet Without the Prince: Cambridge Macroeconomics Without Money”, Amer- ican Economic Review, 75(2), 133-139. Laidler, D. 2001. “Phillips in Retrospect.” UWO Department of Economics Working Papers: Num- ber: 20013. Laidler, D. 2003. Monetary Policy without Money: Hamlet without the Ghost, UWO Department of Economics Working Papers Lachmann, L.M. (1956), Capital and Its Structure, Bell & Sons, London Leeson, R.2000. “A.W.H. Phillips M.B.E. (Military Division).” in Robert Leeson, eds., A.W.H. Phillips: Collected Works in Contemporary Perspective, Chapter 1, Cambridge: Cambridge University Pres. S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86 85

Leet, D. and Scott H. 2003. “Economics Goes to Hollywood: Using Classic Films and Documen- taries to Create and Undergraduate Economics Course”, Journal of Economic Education 34(4), 326-332. Lerner, P.A. 1952. “The Moniac: Economics in Thirty Fascinating Minutes”. Fortune, March, pp. 101. Little, C.D. 1995., ed. On The Reliability of Economic Models: Essays in the Philosophy of Eco- nomics. Boston: Kluwer Academic Publishers. MacCary, T. 1998. Hamlet: A Guide to the Play, Greenwood Press. Mateer, G.D. 2004. Economics in The Movies. Mason: Thomson/South-Western. Malthus, T.R. 1798. An Essay on The Principle of Population. First edition. Volume 1 of The Works of Thomas Robert Malthus. Edited by E.A. Wrigley and D. Souden. London: W. Pickering. Marshall, A. 1997., ed. Principle of Economics. 1920, Prometheus Books, Originally Published: 8th ed., London: Macmillan. Mahoney, J. 1987. ed. The Making of Moral Theology. Oxford: Clarendon Pres. Minsky, H.M., 1975, John Maynard Keynes, Columbia University Press, New York. Nysed. 2002. Economics, The Enterprise System, and Finance, The State Education Department, The University of the State of New York, February 2002, http://www.eric.ed.gov/ERICDocs/ data/ericdocs2sql/content_storage_01/0000019b/80/1b/02/ac.pdf MEB.2008. (Ministry of Education). The Statistics of Ministry of Education, 2007-2008, Ankara. Orta, N. 2007. “The Social Fact-Events Faced in Turkey and Its impacts to Turkish Cinema (1980- 2004)”, Edebiyat Dergisi, Fen-Edebiyat Fakültesi, 18, 125-143 (In Turkish). Oser, J., and S. Brue. 1988. The Evolution of Economic Thought, Fourth edition. San Diego: Har- court Brace Jovanovich. Pares, K.N. and Russell, S.2006. “Higher Education, Real Income and Real Investment in China: Evidence From Granger Causality Tests”, Education Economics, Volume 14, Number 1, Num- ber 1/March 2006, 107-125(19). Pfeiffer, G. 1997. Forrest Gump-Accountant: A Study of Accounting in the Motion Picture Industry, Journal of Accounting Education, 15(3), 319-344. Phillips, A.W. 1950. “Mechanical Models in Economic Dynamics.” Economica, 17(67), pp. 283- 305. Rockoff, H. 1990. “The Wizard of Oz As A Monetary Allegory”, Journal of Political Economy, 98 (4), 739-60. Sarı, R. And Soytas, U. 2006. “Income and Education in Turkey: A Multivariate Analysis”, Educa- tion Economics, Volume 14, Number 2, June 2006, 181-196. Shen, R. and Shen, T.Y. 1993. “Economic Thinking in China: Economic Knowledge and Attitudes of High School Students”, Journal of Economic Education, 24(1), 70-84. Shiller, R.M., Boycko,M. and Korobov,V.1991. “Popular Attitudes Toward Free Markets: The So- viet Union and The United States Compared”, American Economic Review, 81(3), 385-400. Stigler, G.J. 1983. “The Case, If Any, For Economic Literacy”, Journal of Economic Education, Summer 1983, 14(3), 60-66. Tinari, F.D., and Khandke, K. 2000. “From Rhythm and Blues to Broadway: Using Music to Teach Economics”, Journal of Economic Education 31(3), 253-70. 86 S. ONGAN, South-Eastern Europe Journal of Economics 1 (2010) 73-86

Torun, I.2002. Kapitalizmin Zorunlu Şartı “Protestan Ahlak”, Cumhuriyet University. İktisadi ve İdari Bilimler Dergisi, 3(2), 89-98, Sivas. Türkdogan, O. 1981. Endüstri Sosyolojisi, Türkiye’nin Endüstrileşmesi: Dün-Bugün-Yarın, Töre Devlet Publishing House, Ankara. Vibeke, O. 2005. “The Economic Burden of Student Loan Repayment in Norway”, Education Eco- nomics, Volume 13, Number 4, December 2005, 427-447. Watts, M. 2002. “How Economists Use Literature and Drama”, Journal of Economic Education 33(4), 377-386. Watts, M., and R.F. Smith. 1989. “Economics in Literature and Drama”, Journal of Economic Education, 20(3), 291-307. Weber, M. 1904. The Protestant Ethic and the Spirit of Capitalism, Translated by Peter Baehr and Gordon C. Wells in 2002, Penguin Books, USA. White, G.W. 2001. “Teachers’ Report of How They Used Humor With Students’ Perceived Use of Such Humor”, Education, 122(2), 337-/347. Wooton, I. 2003. Production, Lecture 12, University of Strathclyde, Glasgow, http://www.economics.strath.ac.uk/ian/Teaching/PricesandMarkets/Lecture%20Notes/Lecture%2012%20Notes. pdf South-Eastern Europe Journal of Economics 1 (2010) 87-110

Multidimensional Poverty in Greece: A deep, persistent grey?

ANTIGONE LYBERAKΙa*, PLATON TINIOSb and THOMAS GEORGIADISa aPanteion University, Greece bUniversity of Piraeus, Greece

Abstract This study fleshes out the picture of poverty and the poor in Greece, presenting findings with important implications for the Greek social inclusion strategy. Assessing poverty using both monetary and non-monetary dimensions of well- being it becomes evident that, for certain population groups, poverty risk is associated with deprivation risk (multidimensional nature of poverty). Focusing on the age dimension, the stochastic dominance analysis indicates that old age income poverty in Greece appears to be remarkably robust and is not simply due to the choice of poverty lines. Moreover, what is at work statistically is that old age income has an effective ‘floor’ which is constraining inequality among the poor. Such a floor is not evident in the case of the younger group whose distribution of income below the poverty line is much more dispersed. These findings support the argument that much of the effect of the old age poverty alleviation policies over the ten last years in Greece is concentrated on the formation of ‘an effective floor’ for the elderly population rather than on decreasing poverty rates.

JEL Classification: I31; I32. Key words: poverty, composite deprivation index, stochastic dominance analysis

* Corresponding author: Panteion University of Athens, Department of Economics and Regional Development, 136, Syngrou Ave., 17671, Athens, Greece e-mail address: [email protected] The authors would like to acknowledge valuable comments on earlier drafts received from: Sergio Perelman, Clive Richardson and Panos Tsakloglou. Any remaining errors are the authors’ own responsibility. 88 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

1. Introduction Is poverty in Greece ‘grey in colour’? This paper investigates poverty in Greece focusing on the age dimension. Beyond monetary poverty the present empirical analysis deals with the concept of social and material deprivation, incorporating aspects such as financial stress, capacity to afford leisure and social activities, ownership of durable goods and housing conditions. Given that information on income poverty is available, what extra information does additional evidence on material deprivation add? First and foremost, there is a widespread agreement that monetary measures are certainly instructive in assessing income poverty, but they might not capture fully non-monetary dimensions of well- being (Boarini and Mira d’Ercole, 2006; 2008). This is because the concept of poverty goes beyond income or consumption; it is further enriched with non-economic criteria such as vulnerability as well as aspects reflecting participation in social life. There are still many ‘pragmatic’ reasons why looking at material deprivation is not likely to be void of information. First, there are many markets where allocation of resources bypasses the price mechanism: health and education are the most prominent examples; many aspects of housing are subject to criteria wider than market ones. Second, measured income may be an imperfect measure of permanent income, but may also be subject to measurement and reporting errors. Moreover, these errors are not random but are likely to depend on social and economic characteristics; poverty profiles derived from income alone are likely to be biased. Third, we know that the distribution of resources within family units always bypasses the market. Given the very extensive transfer of resources between generations, especially prevalent in Greece and in Southern Europe, information on material deprivation of the aged can be valuable for social policy. This analysis employs the data of EU-SILC, 2004, (Survey of Income and Living Conditions) for Greece. The EU-SILC data are the first release of a new database, designed to replace and improve on the ECHP (European Community Household Panel) and to form the basis on which European Union policy is formulated. This new dataset will be brought to bear on some of the key research questions, which have hitherto been investigated using the ECHP: ●● How does the poverty rate vary across sub-groups of the society? What are the population groups facing significantly higher poverty risk? ●● How much does income status affect material deprivation? How does deprivation vary by socio-demographic characteristics? Are there any population groups at high risk of both poverty and deprivation? ●● Is poverty grey in colour in Greece, irrespective of the selection of a poverty line? A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 89

●● Given that the elderly in Greece are the only group who enjoy effective income guarantees, how effective are these guarantees in securing well-being? Are there signs that this policy is bearing fruit? The rest of this paper proceeds as follows: section 2 provides a brief background of poverty in Greece. Section 3 discusses conceptual and methodological issues related to the poverty analysis, while section 4 presents and comments on the empirical findings. A final section draws some conclusions for public policy.

2. Brief background of poverty and deprivation in Greece This section aims to illustrate certain aspects and characteristics of poverty and deprivation in Greece. To serve this purpose, a number of recent poverty studies are reviewed, portraying some of the ‘stylised facts’ of poverty and material deprivation in Greece. Starting from the poverty risk, using the threshold of 60% of equivalent median income (the central ‘Laeken indicator’ used by the EU in the context of the Open Method of Coordination –see Commission of the European Communities [CEC], 2002), estimates provided by Eurostat indicate that, since 1996, the poverty risk rate in Greece has never been far from 20%. Indisputably, this figure indicates that Greece’s relative position with respect to poverty rate lags significantly and persis- tently behind the better-performing countries in EU (CEC, 2006a; Guio, 2005a; Den- nis and Guio, 2003). As regards trends over time, Tsakloglou (1990) documents a very clear declining trend in both absolute and relative poverty in Greece over the period 1974-1982. The period after 1982, is characterised in political terms by a reforming socialist government pledging ‘Change’, and a considerable increase in social expenditure focused on old age protection (ESSPROS data show total social protection expenditure rose from 10% of GDP in 1980 to 16% in 1985 and 22.9% in 1990). However, contrary to expectations, the poverty picture is characterised by constancy: Tsakloglou (1999) as well as Sarris and Zografakis (1997) using data of Household Budget Surveys until the mid-1990s show that relative poverty actually increased between 1982 and 1988, followed by a further decline during the period 1988-1994. The level of relative poverty then remained essentially unchanged over the period 1995-2003, ranging between 20% and 22%. Regarding the poverty profile, the National Action Plans for Social Inclusion for 2003-2005 and for Pension Sustainability report that poverty is far more prominent among the elderly and is more prevalent in rural than in urban areas in Greece (Min- istry of Labour and Social Security, 2003). At the same time, education is highlighted as the most fundamental factor associated with poverty. Interestingly enough, Tsak- loglou and Panopoulou (1998) demonstrate that persons in old age, persons with 90 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 low educational qualifications and households residing in rural areas are consistently classified as high poverty-risk groups. Focusing more on the old age dimension of poverty in Greece, Lyberaki and Ti- nios (2005; 2006; 2008) based on data obtained from the longitudinal SHARE survey (Survey on Health, Ageing and Retirement in Europe) report that the group of persons aged over 65 appears to be at substantially greater poverty risk compared to the group still of working age (50-65), documenting that advanced age remains an important poverty risk factor in Greece. It is noteworthy that most of the difference between Greece and the EU-15 in poverty risk is due to a worse performance for the over-65s. Interestingly, only Cyprus (and to a lesser extent Spain and Portugal) matches the Greek experience in this area (CEC, 2007). All poverty studies in Greece underline the importance of housing tenure. Own- er occupiers enjoy a stream of housing services which are an important input into household finances. Suffice it to say that low income tenants spend around 20% (and some around 30%) of their income on rents; owner occupiers (netting out mortgage payments) are correspondingly better off. EU-SILC income data do not (to date) include imputations for this type of housing income. Given the wide spread of owner occupancy in Greece, across all income and age classes, allowing for housing tenure effects (as is possible in Household Budget studies) reduces poverty rates by around three percentage points, from 20% to 17% (Lyberaki and Tinios, 2002). Moreover, more recent studies (Koutsambelas and Tsakloglou, 2008) suggest that poverty is reduced according to all versions of the FGT index after the inclusion of imputed rent in the concept of resources, while the reduction is slightly larger for the older groups than the non-elderly. A fuller housing analysis would also need to take other factors on board: quality differences, housing market imperfections affecting the welfare interpretation of the imputation procedure, second homes, as well as sample selection effects arising both from cohabitation with younger relatives and the existence of old age homes. Finally, focusing on studies that have made use not only of household income but also of non-monetary dimensions, Tsakloglou and Papadopoulos (2001a) estimated that 47% and 41% of those classified as deprived in the fields of living conditions and necessities of life, respectively, in Greece were also falling below the poverty line. This is consistent with the findings of other poverty studies (Guio, 2005b; Förster, 2005; Tsakloglou, 1996; Tsakloglou and Papadopoulos, 2002a; 2002b) which indicate that although the overlap between income and material deprivation in Greece seems to be less than complete there is a remarkable share of income poor in Greece that also suffer from material deprivation. A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 91

3. Methodology Next we discuss some methodological issues relevant to the empirical analysis. In particular, section 3.1 focuses on issues related to poverty measurement, while section 3.2 deals with the construction of a deprivation index. Finally, section 3.3 em- phasises the issue of making poverty comparisons, describing the stochastic dominance technique.

3.1 Poverty: identification and aggregation According to Sen (1981) the measurement of poverty can be split into two distinct operations: the identification of the poor and the aggregation of their poverty characteristics into a useable and meaningful measure of poverty. The first step in the identification of poverty is to choose an indicator of welfare (Atkinson, 1989) such as consumption, expenditure or income. Lipton and Raval- lion (1995) argue that consumption is often preferred over current income because it is believed to be a better indicator of long-term average well-being, reflecting the ability of saving to smooth out income fluctuations (it is thus a better measure of ‘permanent income’). If the focus is on age comparisons, we might expect (following life cycle models) the relationship between the two measures to vary with age, while factors such as liquidity constraints will also play an important role. In addition to theoretical considerations, the provenance of the data is likely to affect data quality. In budget surveys total expenditure is thought more reliable than income; in income surveys (such as EU-SILC) the opposite is likely to hold. The present poverty analysis thus selects income as a monetary indicator of poverty, primarily because it is a well-developed module in the EU-SILC (2004) survey (Eurostat, 2005). Choosing income as the measure of access to resources thus follows well-established practice and allows comparisons with other studies. Moreover, when computing poverty measures it has to be taken into account that household size and demographic composition vary across households. A widely-used approach that deals with both size and composition effects is the use of equivalence scales. Though equivalence scales and the interpersonal comparisons they encompass may be rigorously based on social welfare functions (e.g. Deaton and Muellbauer, 1980, ch. 9), we follow most researchers (and Eurostat) who on pragmatic grounds, choose the ‘OECD modified scale’, (1.0 for the head of the household, 0.5 for other adults and children over thirteen years and 0.3 for younger children). Nevertheless, as the choice of equivalence scales might have implications regarding the structure of poverty (de Vos and Zaidi, 1997; Förster, 1994; Tsakloglou and Panopoulou, 1998), the robustness of the findings regarding the age structure of poverty is tested in the present analysis by alternative equivalence scales (Table A1 in the Appendix). The detailed discussion in the next section concludes that use of any equivalence scales 92 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

(as opposed to simple per capita magnitudes) leads to qualitatively similar poverty results. Having chosen the equivalent income as the measure of well-being at individual level, the next step is to define a poverty line in order to identify the poor. The relative poverty line used in the present poverty analysis follows European practice, as embodied in structural indicators: the poverty risk line is defined in relation to the overall distribution of equivalent income, and is set at 60% of the median equivalent income in Greece. The effects of selecting other poverty lines will, nonetheless, be investigated. The next problem is aggregation; that is, to construct summary measures of the extent of poverty. The present poverty analysis focuses on the Foster-Greer-Thor- becke (1984) class of poverty measures, which is defined as follows:

(1)

where yi is the equivalent income of individual i; N is the total population; z is the poverty line; k is the number of the poor; and α is the parameter that reflects the degree of aversion to inequality among the poor. For instance, setting α=0, derives the head-count index that corresponds to the fraction of individuals falling below the poverty line. For α=1 the poverty gap index is derived, which presents the mean aggregate shortfall of the income of the poor from the poverty line. For α=2 the squared poverty gap index is obtained, which takes into account the inequality among the poor, capturing the severity of poverty.

3.2 Deprivation index How should one measure material deprivation in a way that it can be brought to bear on what we know about income poverty? In measuring material deprivation, there is a variety of typologies based on different measurement approaches (recent studies include Halleröd et al., 2006, for Britain, Finland and Sweden; Mckay and Collard, 2003, for Britain; Perez-Mayo, 2005, for Spain; Perry, 2002, for New Zealand; Tsak- loglou and Papadopoulos, 2001b, for Greece; Whelan, and Maître, 2007, for Ireland). Starting from a broader context, building a multi-dimensional measure of material deprivation requires, first, a selection of a subset of events (items) of deprivation at individual level which are then summed over individuals to form an aggregate index (Atkinson, 2003). With respect to the former, the selection of the events depends, inter alia, on the culture of a community, corresponding to socially perceived necessities. Having identified an appropriate set of events, their aggregation into a composite index is usually based either on a simple count or on a weighted approach. The simple count approach is based on binary deprivation scores, namely one or zero, capturing whether a person lacks each of the selected events or not. Next, a A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 93 simple count index is constructed based on the number of events lacked. Townsend (1979) originally proposed this approach as a measure of deprivation assuming ho- mogeneity in tastes for the members of a community – an assumption open to criticism when the focus of interest is on different age groups (or cohorts). The major shortcoming of the simple count approach is that a not-widely owned event is ranked equally with an event that is perceived as much more important to the society. As a result a single event may exert a disproportionate effect on the overall deprivation measure (Willits, 2006). One way of weighting events unequally is the prevalence weighting approach, according to which each of the selected events is weighted by the proportion of the individuals possessing the particular event. The rationale underlying this approach is that assigning higher weights to the events that most people experience, makes the level of deprivation of those who are lacking such events more severe (Tsakloglou and Papadopoulos, 2001b). Going one step further, to control for the influence of tastes in consumption behaviour Desai and Shah (1988) introduced a two-stage econometric methodology for the construction of a deprivation index. This two-stage approach has been employed in other studies of relative deprivation (e.g., Delhausse, Luttgens and Perelman, 1993). In line with Desai and Shah (1988) and Delhausse et al. (1993) the present empirical analysis employs an econometric methodology for the construction of a deprivation index, in order to control for the influence of tastes in consumption. In particular, a probit model is estimated for every selected event in which the dependent variable equals one if the individual experiences the event or zero if the individual lacks the event. The estimation of these probabilities constitutes a crucial aspect in the construction of the index, since it controls for socio-demographic characteristics and taste elements. The outcome of these estimations is presented as adjusted probabilities for each event and for each individual as a member of a particular group in the society. The distances between individual and community estimated probabilities are then computed and the index of relative deprivation is obtained as a weighted sum:

(2)

where i denotes the consumption event, δij the estimated distance between the indi- th vidual and the community experience for the i event and λi the weight given to the ith event. To be more specific, following the approach employed by Delhausse et al.

(1993) the distance δij can be defined as follows: (3) 94 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 where is the estimated mean probability for the ith event and is the estimated probability for individual j to experience the ith event controlling for socio-economic and demographic characteristics. In essence, this approach assumes that being deprived of one good, matters more if that good is more widespread (and hence is more likely to be thought to be part of the ‘social norm’) than if it is not. Being able to assign a ‘deprivation value’ to each event depending on its distance from the social norm, the way is open to aggregate over all deprivation values to derive an (unobserved) ‘overall pain of being deprived’ for every household in the sample. Whereas on the individual level deprivation is always discrete (one is either deprived or not deprived), this procedure essentially gives a smoother overall picture, as many individual characteristics may be correlated with the probability of deprivation. If, however, the unobserved ‘pain of deprivation’ variable is thought itself to be continuous, the force of this objection is reduced.

3.3 Making comparisons When comparing poverty measures over time or between groups, it is crucial to test the robustness of the observed changes in poverty indexes (Coudouel et al., 2002). This is because the robustness of poverty comparisons, as Ravallion (1992) argues, can be compromised by errors in survey data and arbitrariness about both the poverty line and the precise poverty measure. In order to deal with the sensitivity of the ranking of poverty levels (between individuals aged less than 65 years and individuals aged 65 years or more) to the use of different poverty lines, the poverty analysis employed in this paper is based on the stochastic dominance technique. As Deaton (1997) states, stochastic dominance is about ranking distributions. For instance, consider two income distributions y1 and y2 (for two groups A and B respectively) with cumulative distribution functions F(y1) and F(y2). These two cumulative distribution functions may also be thought of as the poverty incidence curve for each group, since each point of the curve gives the proportion of the population with income less than the amount on the horizontal axis. If the poverty incidence curve of group A is some- where below and nowhere above the poverty incidence curve of group B, then poverty is lower for the first group than the second group, independently of the selection of a poverty line. This is called first order stochastic dominance. If poverty incidence curves cross each other, then some poverty lines are likely to rank the situation differ- ently. In this case, the analysis can be restricted by applying the second order and the third order dominance tests. Second order stochastic dominance refers to the poverty gap only, while third order dominance refers to the squared poverty gap. Quisumb- ing et al. (2001) provide an application of both first and second-order dominance criteria in their study on the association between poverty and gender, while Justino and Litchfield (2003) used the stochastic dominance technique to examine poverty dynamics over the 1990s in Vietnam. However, poverty comparisons based on sto- A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 95 chastic dominance are rarely employed in poverty studies in Europe -to some extent this offers an opportunity for added value to the present study.

4. Empirical results This part of the paper presents evidence from the empirical analysis. In particular, section 4.1 provides a detailed poverty profile (the incidence of poverty and the distribution of the poor along socio-economic and demographic characteristics of the population). Next, section 4.2 offers a broad assessment of material deprivation and its determinants in Greece. Finally, section 4.3 compares poverty measures for individuals aged less than 65 years and individuals aged 65 years or more using the stochastic dominance technique as well.

4.1 Poverty Profile Table 1 presents summary statistics on income measures based on the data of EU- SILC, 2004, survey for Greece. As reported in the fourth column of Table 1, the poverty line set at 60% of the median equivalent income classifies one in five individuals (20.1%) as being poor. In order to give a more complete picture, Table 1 also decomposes the group of the poor and reports poverty odds ratios for each individual characteristic. Focusing on the demographic characteristics, age emerges as an important and overwhelmingly significant dimension of poverty in Greece. In particular, while poverty incidence is estimated at about 14% for individuals aged between 26 and 40 years, it rises steadily (reaching 23.2%) for persons aged between 65 and 74 years. Poverty appears to be most acute for the oldest members of the population, with the estimated poverty rate (34.4%) suggesting that one in three persons aged 75 years or more faces the risk of being poor. Given that household income is set, by construction, to be equivalently distributed to all members of a household, the gender dimension of poverty seldom produces meaningful results. Despite this, in Greece, women face significantly higher poverty risk — chiefly due to the poverty faced by widows who comprise the majority of female-only households. One of the distinguishing features of poverty in Greece is the high percentage of owner-occupation amongst the poor (CEC, 2006b). Indeed, home owners are more likely to be poor than tenants. Taking a deeper look by focusing on the interaction of age with tenure status, this prima facie ‘perverse’ result is due to home owners aged over 65 years of age being at greater risk of poverty than tenants aged over 65 years. This is a noteworthy finding for the elderly population in Greece, given that in the majority of countries in EU the emerging picture (e.g., Zaidi et al., 2006) suggests that elderly tenants have much higher poverty rates than those observed for elderly home owners. The reverse, more expected relationship holds for the younger population group, since home ownership reduces the poverty risk for those aged less than 65 years. Another feature of the Greek poverty picture is that the presence of children is 96 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

Table 1: Summary statistics on income measures, EU-SILC, 2004

Average income Poverty risk: Poor Population Odds as % of sample (%) below share share (%) Ratio* average poverty line (%)

Age and Gender <16 years 16.3 102.8 19.1 14.6 89.6 16 – 25 years 12.5 90.4 22.6 14.1 112.8 26 – 40 years 21.1 109.4 14.4 16.9 80.1 41 – 54 years 19.2 107.3 18.0 17.4 90.6 55 – 64 years 11.1 103.8 19.6 10.9 98.2 65 – 74 years 11.6 89.1 23.2 12.9 111.2 75 + years 8.2 76.4 34.4 13.2 161.0 < 65 years 80.2 104.0 17.9 74.0 92.3 65+ years 19.8 84.0 27.8 26.0 131.3 Male 48.0 102.4 18.5 44.9 93.5 Female 52.0 97.8 21.5 55.1 106.0 Marital Status and Nationality Single 25.2 102.1 17.7 22.9 90.9 Married 62.9 101.3 19.9 62.1 98.7 Widowed 9.1 85.2 28.1 12.3 135.2 Divorced / separated 2.8 98.3 19.8 2.8 100.0 Foreign-born 6.9 86.5 23.8 9.2 133.3 Greek-born 93.1 101.1 19.7 90.8 97.5 Education Compulsory 9-year or less 51.3 78.4 28.4 72.7 141.7 High School 32.9 106.5 14.2 23.3 70.8 Tertiary 15.8 156.9 5.0 4.0 25.3 Region and Health Status Urban 82.6 106.0 15.7 64.8 78.5 Rural 17.4 71.4 40.7 35.2 202.3 No Chronic Disease 79.9 104.4 18.2 73.5 92.0 Health Disability 20.1 81.2 27.9 26.5 131.8 Employment Status In Employment 48.2 114.7 13.5 32.6 67.6 Unemployed 5.3 79.6 29.3 7.7 145.3 Retired 21.2 89.4 26.1 27.7 130.7 Domestic tasks 15.5 83.2 24.6 19.1 123.2 Students / other inactive 9.8 88.4 26.3 12.9 131.6 # of children aged <16 0 71.4 99.4 20.2 69.0 96.6 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 97

1-2 25.7 102.2 19.2 26.8 104.3 3 + 2.9 77.1 32.6 4.2 144.8 Tenure status and age Home owner/rent free 77.2 101.1 20.3 77.1 99.9 Tenant 22.8 94.6 18.7 22.9 100.4 <65 & Home owner 57.4 106.9 17.6 49.2 85.7 <65 & Tenant 20.4 95.2 18.5 20.0 98.0 65+ & Home owner 19.8 83.9 28.2 27.9 140.9 65+ & Tenant 2.4 86.5 21.1 2.9 120.8 Total 100.0 100.0 20.1 100.0 100.0

Note: * Odds ratio defined as [ (Poor share % / Population share %)* 100]. Values over 100 indicate a greater representation of a group among the poor than in the total population. Source: EU-SILC, 2004, authors’ calculations. a far less important poverty determinant than in other countries (CEC, 2006b). Disag- gregating by the number of children, this result holds for households with two children or one child, where the poverty risk is indeed slightly lower than for households with no children. On the contrary, households with three or more dependent children face a poverty risk exceeding 32%. Thus, while overall the presence of children does not increase poverty risk, families with three children and more are at a definite disadvantage. It should be noted that in public discussion a live issue is the extent to which benefits intended for large families should be extended to the group with three children. As expected, given that education is the chief correlate of permanent income, the results offer evidence of a strong inverse relationship between poverty and education level. Indeed, the emerging picture is quite straightforward: the poverty rate is estimated 28.4% for individuals with nine years of compulsory education or less, 14.2% for individuals possessing high school education and only 5% for those with tertiary education. Therefore, these results highlight the positive contribution of education in raising income and living standards. Concerning the geographical variation in poverty incidence, it becomes evident that four out of ten persons residing in rural areas fall below the poverty line, indicating that poverty is more acute in rural than in urban areas in Greece. Residence in rural areas (defined as townships with fewer than 1000 inhabitants) appears as one of the most potent predictors of poverty. Finally, as in other countries, participation in the labour market reduces poverty risks considerably. In particular, the estimated poverty rates suggest that poverty incidence is higher than average for unemployed persons (29.3%), for retired (26.1%), for persons engaged in domestic tasks (24.6%), as well as for other inactive population groups (26.3%) –chiefly students. On the other hand, persons participating in the labour market face a considerably lower poverty risk (13.5%). 98 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

4.2 Income status and material deprivation This section aims to explore certain aspects of deprivation, to identify population groups that are at deprivation risk and to investigate the relationship between income and deprivation in Greece. Twelve events that serve as indicators of material and social deprivation were selected. Starting from the subjective dimensions of deprivation, the event that captures the financial stress of the households is described as follows: (i) the household has not been in arrears on utility bills -electricity, water, gas- in the last 12 months. Turning to the objective dimension, these include eight events relating to the household’s ability to afford basic leisure and the availability of consumer durables. In particular, the three events that capture the capacity to afford leisure and social activities are: (ii) ability to pay for one week’s annual holiday away from home, (iii) ability to afford one meal containing meat or fish every two days, (iv) ability to face unexpected financial expenses. The selected durable goods are: (v) television, (vi) personal computer, (vii) washing machine, (viii) car and (ix) home-ownership. It is worth mentioning that in the case of durable goods the deprivation analysis is based on the ability to afford –not availability- of the four items (television; personal computer; washing machine and car). Finally, to capture the availability of basic facilities in housing, the three events that have been selected are: (x) heating, (xi) water closet and (xii) bathroom. The second column of Table 2 presents the mean value of the deprivation index for population groups defined by age, gender, employment status, education, health, marital status, region and nationality. Generally, positive mean values of the deprivation index indicate high deprivation risk, while a value close to zero corresponds to the norm in the community. Consequently, if a population group exhibits a negative mean value of the deprivation index, this is a sign of low deprivation level among the group. Starting from the age dimension, the emerging picture suggests that deprivation appears to be more prevalent for the elderly (the value of the index reaches 0.0211 for those aged 65+) than for the youngest (persons aged less than 65). Turning to the gender dimension, women exhibit higher mean value of the deprivation index (0.0038) compared to men (-0.0040). Married persons are, on average, less deprived than the unmarried, while persons of other marital status (such as widows, divorced etc) appear to be especially vulnerable to deprivation. Focusing on ethnicity, it becomes evident that material hardship is on average higher among foreign-born, compared to Greek-born individuals. Turning to the education dimension, the emerging picture indicates that material deprivation is inversely related to the level of education. Thus, persons with low education are more affected, on average, by deprivation compared to individuals with higher level of education. Moreover, deprivation substantially diminishes among those with tertiary education. A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 99

Table 2: Levels of the deprivation index by population groups in Greece

Classes Mean index value 95% Confidence interval Age Less than 65 years -0.0060 -0.0072 -0.0048 65+ years 0.0211 0.0188 0.0234 Gender Male -0.0040 -0.0055 -0.0025 Female 0.0038 0.0023 0.0054 Marital Status & Nationality Single 0.0087 0.0065 0.0110 Married -0.0117 -0.0129 -0.0104 Widowed 0.0474 0.0435 0.0513 Divorced or separated 0.0340 0.0264 0.0416 Foreign-born 0.0863 0.0819 0.0907 Greek-born -0.0072 -0.0082 -0.0062 Education Compulsory 9-year or less 0.0297 0.0284 0.0310 High School Education -0.0178 -0.0195 -0.0162 University Education -0.0595 -0.0615 -0.0575 Region – Type of household Urban -0.0064 -0.0076 -0.0052 Rural 0.0308 0.0286 0.0330 Health Status No Chronic Disease -0.0090 -0.0101 -0.0078 Health Disability 0.0384 0.0360 0.0408 Employment Status In Employment -0.0146 -0.0161 -0.0132 Unemployed 0.0556 0.0508 0.0604 Retired 0.0142 0.0118 0.0166 Domestic tasks 0.0017 -0.0005 0.0041 Students or other inactive 0.0087 0.0054 0.0120 # of children aged <16 0 0.0040 0.0027 0.0052 1-2 -0.0111 -0.0133 -0.0089 3 + -0.0058 -0.0108 -0.0009 Equivalent Income 1st quintile 0.0741 0.0724 0.0758 2nd quintile 0.0399 0.0382 0.0417 3rd quintile 0.0053 0.0037 0.0069 100 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

4th quintile -0.0303 -0.0317 -0.0289 5th quintile -0.0730 -0.0737 -0.0722

Source: EU-SILC, 2004, authors’ calculations

As expected, income status is estimated to be inversely and monotonically related to the level of deprivation. In particular, people who are classified in the first (bottom) income quintile record, on average, higher level of deprivation compared to persons in higher income quintiles. In addition, moving towards higher income quintiles (that stand for higher incomes) the level of deprivation diminishes significantly, becom- ing remarkably low for those who are classified into the fifth (top) income quintile. Finally, in contradiction to the poverty profile, the presence of a third child is not associated with a greater deprivation risk. Furthermore, material deprivation appears to be strongly associated with health status, though the direction of causation is, of course, indeterminate: the estimated results indicate that people who face sickness and disability problems record, on average, higher level of deprivation compared to the rest of the population. With ref- erence to the employment status, being unemployed is associated, on average, with higher level of deprivation compared to individuals in any other labour market status, indicating that participation in the labour market is an important factor preventing deprivation. Making the link to the findings of the poverty profile, it becomes evident that many groups that are found to be of high poverty risk, such as elderly persons, those with low education, persons with health disabilities, unemployed persons and persons residing in rural areas appear to be at high deprivation risk as well. This implies that for many vulnerable population groups, (income) poverty risk is associated with (material) deprivation risk (though the fit is not perfect).

4.3 Comparing poverty measures: the age dimension This section compares poverty measures for individuals aged less than 65 years with those aged 65 years or more. In order to deal with the sensitivity of the ranking of poverty levels between these two age-groups to the use of different poverty lines, the poverty analysis is based on the stochastic dominance technique. Table 3 shows poverty estimates of the Foster-Greer-Thorbecke (1984) group of indicators (headcount, poverty gap and squared poverty gap index) for those aged less than 65 and for those aged 65 years or more using a wide range of relative poverty lines. Estimates for the headcount index, which show the proportion of people counted as poor, indicate that the poverty rate is higher for those aged over 65 years than those aged less than 65 for all the selected poverty lines. For the baseline poverty line set at 60% of the median equivalent income, the poverty rate of those aged more A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 101 than 65 years (27.8%) exceeds almost by ten percentage points the corresponding figure of those aged less than 64 years (17.9%). Interestingly the percentage by which the older group is more poverty prone than the younger group is remarkably constant: around 50% of the value of the younger group.

Table 3: FGT class of measures across different poverty lines, by age group

Poverty lines 40% of median 50% of median 60% of median 70% of median FGT index <64 65+ <64 65+ <64 65+ <64 65+ Headcount ratio 7.1 10.4 11.4 18.3 17.9 27.8 24.8 36.8 (a=0) Stand. error 0.27 0.57 0.34 0.73 0.42 0.84 0.47 0.9 Poverty Gap (a=1) 4.9 3.0 5.7 5.3 7.2 8.2 9.2 11.7 Stand. error 0.52 0.28 0.44 0.31 0.39 0.35 0.36 0.39 Squared Poverty 26.0 2.2 18.5 2.9 14.8 4.1 13.1 5.7 Gap (a=2) Stand. error 0.52 0.71 5.66 0.53 4.04 0.44 3.05 0.41

Source: EU-SILC 2004, authors’ calculations

Turning to the poverty gap index, which presents the mean aggregate shortfall of the income of the poor from the poverty line, the emerging picture suggests that for relatively low poverty lines (set at 40% or 50% of the median equivalent income), poverty gaps are lower for those aged over 65 years compared to the younger group, reversing the familiar relationship. On the other hand, the reverse becomes evident according to the baseline poverty line (60% of the median) or the relatively high poverty line set at the 70% of the median equivalent income. Concerning the severity of poverty, the squared poverty gap index indicates lower income inequality among the elderly poor than the non-elderly, despite the fact that the latter group’s poverty rate is lower. The above analysis based on summary statistics is further confirmed by the stochastic dominance analysis. By plotting the poverty incidence curves, it is possible to check graphically which of the two age-groups shows a higher level of poverty. As already mentioned above, each point of the poverty incidence curves corresponds to the proportion of the population with income less than the amount given on the horizontal axis. Figure 1a confirms the results obtained by computing the poverty indices, indicating that poverty incidence is unambiguously lower for those aged less than 65 compared to those aged 65 years or more, over the relevant range of poverty lines. In the same way, Figure 1b focuses on the lower part of the distribution providing a 102 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 clearer exposition of the poverty incidence curves by age group for annual equivalent incomes up to 8000 euros.

Figure 1a: Stochastic Dominance Analysis: poverty incidence by age group in Greece

Source: EU-SILC, 2004, authors’ estimates

Old age income poverty in Greece appears to be remarkably robust: it is not simply due to the choice of poverty lines. Even using a poverty line of 40% median the poverty risk index remains around 50% greater for the over 65s than the under 65s. Furthermore, the robustness of the findings regarding the age structure of poverty in Greece doesn’t appear to be affected by the choice of alternative sets of equivalent scales (Table A1 in the Appendix): all three scales lead to qualitatively similar results; the results only differ if per capita quantities are used instead. The other results, together with the stochastic dominance analysis, indicate that what is at work statistically is that old age income has an effective ‘floor’ which is constraining inequality among the poor. Such a floor is not evident in the case of the younger group whose distribution of income below the poverty line is much more dispersed. Furthermore, it is worth making a comparison of the estimated poverty rates with the corresponding picture of the mid-1990s, namely since the period of introduc- A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 103

Figure 1b: Stochastic Dominance Analysis by age group: focusing on low incomes

Source: EU-SILC, 2004, authors’ estimates tion of a supplementary pension benefit (EKAS) targeted at low income pensioners (described in more detail below). The emerging picture suggests that the poverty incidence of persons aged between 16 and 64 remained rather unchanged since the mid-1990s (18% in 1996). At the same time, the poverty rate of persons aged 65+ decreased by almost five percentage points over the period 1996-2004 (from 33% in 1996 to 27.8% in 2004). Hence, it can be unambiguously argued that the continuous implementation of such measures almost for ten years has not dented the effect of old age as a poverty risk. This is certainly an important remark; however, it neglects other equally important aspects of the effectiveness of social policy in Greece. For instance, focusing on the relative poverty gap (defined as the difference between the median equivalent income of persons below the at-risk-of poverty threshold and the threshold itself, expressed as a percentage of the at-risk-of poverty threshold) it becomes evident that the relative poverty gap for the poor persons aged over 65 years — chiefly pensioners entitled to this benefit, has been significantly reduced from 40% in 1995 to 26.6% in 2004 — being less than 30% from 2000 onwards. Therefore, much of the effect of the 104 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 old age poverty alleviation policies over the ten last years in Greece is concentrated on the formation of ‘an effective floor’ for the elderly population rather on decreasing poverty rates. Greece also has in-kind income not accounted for (housing income accruing to owner occupiers). However, that is fairly evenly spread amongst the age classes and does not account for the differential poverty results. Explanations may be sought in the structure of income of the aged in Greece, as well as in the structure and practices of what has been called the ‘informal system of social protection’: ●● Around 50% of that is in the form of pensions. Though the pension system is extremely fragmented, a number of ‘floors’ in the form of guaranteed minima exist (Börsch-Supan and Tinios, 2001): The uninsured are entitled to a basic farmer’s pension. The main private sector social insurance fund (IKA) has a minimum pension which 70% of IKA pensioners collect. To this informal em- ployee floor is added a means tested pension top up called EKAS, which has been in force since 1996 and was explicitly introduced as a targeted measure to cope with old age poverty. The self-employed are also entitled to EKAS, which hence (added to the minimum pension which is differentiated by pension fund) should be the effective floor of urban sector income. The relationship of these floors to the poverty line, their relative size, together with household composition are thus the determinants of poverty incidence for pensioners. Indicative calculations bear this out: an elderly couple living on one IKA minimum pension would be well below the poverty line, their equivalent income in 2004 being 69% of the poverty line. Should they collect the low pension supplement EKAS, their income will rise to 89% of the poverty line, but they will still be poor. Thus the implementation of EKAS may be expected in this case to reduce the poverty gap but to leave the headcount ratio unaffected (the poverty line is crossed only in the case of self-employed minimum pension in a single member household). The data, unfortunately, do not distinguish EKAS from other pensions and preclude an analysis of its effect. ●● Not all over 65s are pensioners. Based on EU-SILC 2004, 23% of those aged over 65 years don’t receive income transfers from an old-age pension. Simi- larly, 13.6% of the persons aged over 65 years live within a household where no one receives an old-age income transfer. ●● Family solidarity. Only 53% of pensioners’ income is derived from old age pension. ●● Cohort effects. Cohorts born before 1930 had to live through 10 years of war and civil war in the 40s, while they lived their productive years at times when social insurance was not universally available. Hence, the older old must be A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 105

expected to be worse off and old age poverty must be decreasing over time reflecting the maturation of social insurance. A direct comparison with older results is not possible, given that SILC is different in important respects from ECHP. Yet, the existing evidence shows that the potency of old age as a poverty inducing factor remains. Its persistence surely implies that old age poverty is reproduced by features of the pension system and is systemic in nature.

5. Conclusions Using data obtained from the EU-SILC, 2004, survey for Greece, the main objective of this paper has been the investigation of poverty in Greece, focusing particularly on the age dimension. Overall, the poverty profile revealed that poverty in Greece has distinct and complex demographic and socio-economic dimensions, a finding that is in line with the conclusions of other poverty studies in Greece using ECHP. Thus, on the one hand, poverty rates are quite low among the well educated and among those who participate in the labour market. On the other hand, poverty increases with age and becomes quite acute for the oldest-old. Moreover, poverty risk is also high for those with low education, for individuals residing in rural areas and for individuals who face health disabilities. Finally, families with three or more children (though not those with one or two) are at greater poverty risk. Housing tenure plays an interesting and complex role. While overall owner-occupation is associated with greater poverty risk, this only holds for the older group (aged over 65); for those aged less than 65 owner occupation is associated with lower poverty. With respect to the identification of the population groups that are at deprivation risk, the empirical findings reveal that individuals aged over 75, the divorced or widowed, those with low education, individuals residing in rural areas, persons with health disabilities and unemployed persons exhibit, on average, high deprivation levels. In other words, many of the vulnerable population groups face both income poverty and material deprivation risk. Concerning the poverty comparison between persons aged under and over 65, estimates for the headcount index indicate that poverty incidence is higher for those aged over 65 years over a wide range of poverty lines (ranging from 40% to 70% of the median equivalent income). These summary statistics are further supported by the stochastic dominance analysis. These findings have an important implication for the Greek social inclusion strategy. That strategy forswears a generalised guaranteed income, in favour of a more pragmatic approach, focused on specific groups offering group-specific income guarantees (Ministry of Labour and Social Security, 2003). The idea is that by giving 106 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 priority to groups that can be clearly identified and where mechanisms exist for as- certaining and serving that need, targeted actions will be more effective in combating real need. Such groups are precisely those that are characterised by greater incidence of poverty risk. In this context two observations are in order. First, most of the dimensions of poverty and deprivation have been identified for a long time and indeed have been subject to poverty-alleviating policies implemented in Greece for at least a decade. That the risk factors identified in our study, using 2004 data are not qualitatively different from those for the mid-1990s identified by Mitrakos and Tsakloglou (2006), who used HBS data, speaks volumes about the lack of effectiveness of such policies. Secondly, persons aged over 65 (and indeed the very old — over 75) can be taken to be the group that fulfills the preconditions of the targeted approach. They can be easily identified, and given that their most significant income source is state pensions, can be easily reached. Indeed, low income pensioners have been the explicit object of targeted measures at least since 1996. Almost 10 years of continuous implementation of such measures have not dented the effect of old age as a poverty risk, though they might have increased the floor a little. Further progress must necessarily come to grips with those characteristics of the pension system that reproduce and seemingly accentuate poverty. Most of the preceding analysis dealt with income poverty. However, it is unde- niable that changes in the well-being of the elderly also depend on other factors: benefits in kind are a case in point, such as the provision of services at home by the promising ‘Help at Home’ programme (Ministry of Labour and Social Security, 2003). However, the small number of beneficiaries and lack of clarity in eligibility rules meant that in 2004 its overall effect was limited. Of greater quantitative importance are changes on the supply side of personal care services which resulted from the availability of (female) immigrant labour from the mid-1990s on. These ‘Deae ex machina’ (Lyberaki, 2008) contributed to the availability and affordability of personal care services to the old, while they permitted the increase in female labour participation to take place, thus adding both directly and indirectly to real incomes of the elderly. The poverty profile emerging from our work, together with the observation that it is no different from that holding a decade ago, means that the efficacy of the Greek social inclusion strategy is placed in serious doubt. If Greece is to take poverty alleviation seriously it should be more closely attuned both to the characteristics of poverty and to the mechanisms that give rise to them. A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 107

References Atkinson, A.B., 1989, Poverty and Social Security, Hempstead: Harvester Wheatsheaf. Atkinson, A.B., 2003, “Multidimensional deprivation: contrasting social welfare and counting approaches”, Journal of Economic Inequality 1, 61-65 Boarini, R. and Mira d’Ercole, M., 2006, “Measures of Material Deprivation in OECD Countries”, OECD Social Employment and Migration Working Papers No.37, OECD, Paris. Boarini, R. and Mira d’Ercole, M., 2008, Non-income poverty: What can we learn from indicators of material deprivation?, in OECD (ed), Growing Unequal? Income Distribution and Poverty in OECD Countries. OECD, Paris. Börsch-Supan, A. and Tinios, Pl., 2001, The Greek Pensions System: Strategic Framework for Reform, in R. Bryant, N. Garganas and G. Tavlas (eds), Greece’s Economic Performance and Prospects. Bank of Greece and Brookings Institution. Commission of the European Communities, 2002, Joint Report by the Commission and Council on Social Inclusion. Directorate of Employment and Social Affairs. Commission of the European Communities, 2006a, Adequate and Sustainable Pensions: Synthesis Report. Directorate-General for Employment, Social Affairs and Equal Opportunities. Commission of the European Communities, 2006b, Joint Report by the Commission and Council on Social Protection and Social Inclusion. Directorate of Employment, Social Policy, Health and Consumer Affairs. Commission of the European Communities, 2007, Joint Report by the Commission and Council on Social Protection and Social Inclusion. Directorate of Employment, Social Policy, Health and Consumer Affairs. Coudouel, A., Hentschel, J. and Wodon, Q., 2002, Poverty Measurement and Analysis, in J. Klug- man (ed), A Sourcebook for Poverty Reduction Strategies. Washington, DC: World Bank. Deaton, A. and Muellbauer, J., 1980, Economics and Consumer Behaviour, Cambridge, CUP. Deaton, A., 1997, The Analysis of Households Surveys: A Microeconometric Approach to Develop- ment Policy. Baltimore and London: John Hopkins Press and World Bank. Delhausse, B., Luttgens, A. and Perelman, S., 1993, “Comparing measures of poverty and relative deprivation: An example from Belgium”, Journal of Population Economics, 6, 83-102. Dennis, I. and Guio, A., 2003, Poverty and social exclusion in the EU after Leaken –Part 1. Euro- stat, Statistics in Focus. Theme 3 8/2003 Desai, M. and Shah, A., 1988, “An econometric approach to the measurement of poverty”, Oxford Economic Papers, 40, 505-522 De Vos, K. and Zaidi, A.M., 1997, “Equivalence Scale Sensitivity of Poverty Statistics for the Members States of the European Community”, Review of Income and Wealth, 43, 319-333. Eurostat, 2005, The Continuity of indicators during the transition between ECHP and EU-SILC. Office for Official Publications of the European Communities, Luxembourg. Foster, J., Greer, J. and Thorbecke, E., 1984, “A class of decomposable poverty measures”, Econo- metrica, 52, 761-766. Förster, M., 1994, Measurement of Low Incomes and Poverty in a Perspective of International Comparisons, OECD Labour Market and Social Policy Occasional Paper, No.14, Paris. 108 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

Förster, M., 2005, “The European Social Space Revised: Comparing Poverty in the Enlarged Euro- pean Union”, Journal of Comparative Policy Analysis, 7, 29-48. Guio, A., 2005a, Income Poverty and Social Exclusion in the EU25. Eurostat, Statistics in Focus 13/2005, Luxembourg. Guio, A., 2005b, Material Deprivation in the EU. Eurostat, Statistics in Focus 21/2005. Halleröd, B., Larsson, D., Gordon, D. and Ritakallio, V., 2006, “Relative deprivation: a comparative analysis of Britain, Finland and Sweden”, Journal of European Social Policy, 16, 328-345 Justino, P. and Litchfield, J., 2003, Welfare in Vietnam During the 1990s: Poverty, Inequality and Poverty Dymanics, PRUS Working Paper No. 8, Poverty Research Unit at Sussex, University of Sussex. Koutsambelas, Ch. and Tsakloglou, P., (2008), Estimates of imputed rents and their distributional impact in Greece. Paper Prepared for the 30th General Conference of The International Associa- tion for Research in Income and Wealth, Portoroz, Slovenia. Lipton, M. and Ravallion, M., 1995, Poverty and Policy, in J, Behrman and T.N. Srinivasen (eds) Handbook of Development Economics vol. 3B (Amsterdam: Elsevier). Lyberaki, A., 2008, “Deae ex Machina”: migrant women, care work and women’s employment in Greece. Hellenic Observatory Papers on Greece and Southeast Europe, GreeSE Paper No 20, London School of Economics. Lyberaki, A. and Tinios, Pl., 2002, Work and Cohesion: the Greek National Plans for Employment and Social Inclusion. Athens: Papazisis (in greek). Lyberaki, A. and Tinios, Pl., 2005, Poverty and Social Exclusion: a new approach to an old issue, in A. Börsch-Supan, A. Brugiavini, H. Jurges, J. Mackenbach, J. Siegrist and G. Weber (eds) Health Ageing and Retirement in Europe: First Results from the Survey of Health, Ageing and Retirement in Europe. Mannheim. Lyberaki, A. and Tinios, Pl., 2006, Poverty and Social Exclusion. Paper presented in AMANDA Final Conference: Lessons from SHARE Wave 1, Como, Italy. Lyberaki, A. and Tinios, Pl., 2008, Poverty and Persistent Poverty: Adding Dynamics to Familiar Findings, in A. Börsch-Supan, A. Brugiavini, H. Jurges, A. Kapteyn, J. Mackenbach, J. Siegrist and G. Weber (eds) Health Ageing and Retirement in Europe (2004-2007): Starting the Longi- tudinal Dimension. Mannheim. Mckay, S. and Collard, S., 2003, Developing Deprivation Questions for the Family Resources Sur- vey, Department for Work and Pensions Working paper Number 13, Personal Finance Research Centre, University of Bristol Ministry of Labour and Social Security, 2003, National Action Plan for Social Inclusion 2003- 2005, Greece. Mitrakos, Th. and Tsakloglou, P., 2006, Inequality and Poverty in the Last Quarter of the 20th Centu- ry, in M. Petmetsidou and El. Mossialos (eds) Social Policy Developments in Greece. Ashgate. Perez-Mayo, J., 2005, “Identifying Deprivation Profiles in Spain: A New Approach”,Applied Eco- nomics, 37, 943-955. Perry, B., 2002, “The Mismatch Between Income Measures and Direct Outcome Measures of Pov- erty”, Social Policy Journal of New Zealand, 19, 101-127. A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110 109

Quisumbing, A., Haddad, L. and Pena, C., 2001, “Are women overrepresented among the poor? An analysis of poverty in 10 developing countries”, Journal of Development Economics, 66, 225-269. Ravallion, M., 1992, Poverty comparisons: A Guide to Concepts and Methods, LSMS Working Paper No. 88, World Bank, Washington, DC. Sarris, A. and Zografakis, St., 1997, Poverty and income inequality in Greece after 1974.University of Athens, Greece (in Greek). Sen, A., 1981, Poverty and Famines: An Essay on Entitlement and Deprivation, Oxford University Press. Townsend, P., 1979, Poverty in the United Kingdom, Harmondsworth, Penguin. Tsakloglou, P., 1990, “Aspects of Poverty in Greece”, Review of Income and Wealth, 36, 381-402. Tsakloglou, P., 1996, “Elderly and Non-Elderly in the European Union: A Comparison of the Living Standards”, Review of Income and Wealth, 3, 271-291 Tsakloglou, P., 1999, Poverty and anti-poverty policies in Greece and a comparison with other Mediterranean EU Member-States, Athens University of Economics and Business. Tsakloglou, P. and Panopoulou, G., 1998, “Who Are the Poor in Greece? Analysing Poverty under Alternative Concepts of Resources and Equivalence Scales”, Journal of European Social Policy, 8, 213-236. Tsakloglou, P. and Papadopoulos, F., 2001a, Identifying Population Groups at High Risk of Social Exclusion: Evidence from the ECHP, IZA Discussion Paper Series, No. 392. Tsakloglou, P. and Papadopoulos, F., 2001b, Indicators of Social Exclusion in EUROMOD, EURO- MOD, Working Paper No. EM8/01. Tsakloglou, P. and Papadopoulos, F., 2002a, Poverty, material deprivation and multidimensional disadvantage during four life stages: Evidence from the ECHP, in C. Heady, M. Barnes, J. Millar, S. Middleton, P. Tsakloglou and F. Papadopoulos (eds.) Poverty and social exclusion in Europe, Edward Elgar, Cheltenham. Tsakloglou, P. and Papadopoulos, F., 2002b, “Aggregate level and determining factors of social exclusion in twelve European countries”, Journal of European Social Policy, 12, 211-225. Whelan, C.T. and Maître, B., 2007, “Measuring Material Deprivation With EU-SILC: Lessons From the Irish Survey”, European Societies, 9, 147-173. Willits, M., 2006, Measuring child poverty using material deprivation: possible approaches, De- partment for Work and Pensions, Working Paper No. 28, Corporate Document Services, Leeds. Zaidi, A., M. Makovec, M. Fuchs, B. Lipszyc, O. Lelkes, M. Rummel, B. Marin and de Voos, K., 2006, Poverty of Elderly People in EU25, European Centre for Social Welfare Policy and Re- search, 10 July 2006 (revised version), Vienna: Austria 110 A. LYBERAKI et al., South-Eastern Europe Journal of Economics 1 (2010) 87-110

Appendix Table A1: FGT class of measures across different equivalence scales, by age group

Equivalence scale Poverty line: 60% median income OECD- Old-OECD Square root Per capita modified scale * (Oxford scale)** scale *** income Age <64 65+ <64 65+ <64 65+ <64 65+ Headcount ratio 17.9 27.8 18.6 24.4 17.8 30.6 20.8 21.5 Stand. error 0.42 0.84 0.42 0.81 0.41 0.86 0.44 0.7 Poverty Gap 7.2 8.2 7.4 7.0 7.1 9.3 8.1 5.9 Stand. error 0.39 0.35 0.38 0.33 0.39 0.36 0.38 0.32 Squared Poverty 14.8 4.1 14.4 3.5 15.3 4.6 13.9 3.0 Gap Stand. error 4.04 0.44 3.83 0.43 4.24 0.46 3.42 0.43

Notes: * 1.0 for the head of the household, 0.5 for other adults and children over thirteen years and 0.3 for younger children. ** 1.0 for the head, 0.7 for adults and children over thirteen and 0.5 for younger children *** Estimated as the square root of household size. Source: EU-SILC 2004, authors’ calculations South-Eastern Europe Journal of Economics 1 (2010) 111-128

IDENTIFYING SPATIAL LABOR MARKETS IN GREECE FROM THE 2001 TRAVEL-TO-WORK PATTERNS

Prodromos-Ioannis K. Prodromidis* Centre for Planning and Economic Research [KEPE]

Abstract The article examines inter-municipal commuting flows in Greece, collected via the 2001 Census, and delineates the country’s labor market areas (LMAs). It finds that the LMAs of Athens (3.9 million inhabitants) and Thessaloniki (1.1 million) exceed the homonymous urban-planning complexes by 8 and 15 times, respectively. These LMAs, along with the LMAs of Patras (245 thousand) and Iraklion (233 thousand), host about half the country’s population. Another thirty-eight clusters of municipalities and eight self-contained municipalities of 20-184 thousand inhabitants jointly host a quarter of the country’s population. The picture is complemented by the presence of ten clusters of municipalities and 607 self- contained municipalities with smaller populations. Overall, the article advances our understanding of how the country functions at the sub-national level.

JEL Classification: J49, R12. Key words: Labor market areas. Functional economic areas. Commuting flows. Localities. Urban and micro-regional policy areas

* Corresponding address: KEPE, Amerikis 11, Athens 10672, Greece. e-mail: [email protected] Earlier versions of the article were presented at the 26th Conference of the Standing Committee on Regional and Urban Statistics (Europe Group), and in seminars at the University of Thessaly and KEPE. Thanks are due to the National Statistical Service of Greece for supplying the data and promptly answering the queries regarding their collection, the participants in the above conference and seminars for constructive comments, and an anonymous referee for useful suggestions. I am indebted to Prof. Martin Schuler of the École Polytechnique Fédérale de Lausanne for bringing me up to date on relevant issues prior to my involvement with the OECD’s Working Party on Territorial Indicators in 2004. The usual disclaimer applies. 112 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128

1. Introduction The purpose of this article is to delineate, for the first time, the Labor Market Ar- eas (LMAs) of Greece by utilizing disaggregated travel-to-work data solicited in the 2001 Population Census. Thus, it enhances our understanding of how the economy operates at the sub-national level in a manner that may not be possible through the conventional territorial partitions (administrative regions or provinces; counties, districts or prefectures; etc.). This is especially true in cases where the aforementioned partitions represent a mixture of geographical factors, historical memories, relics of commercial life from previous centuries, administrative contingencies, political/electoral considerations, geometry or chance. The stimulus was provided in conferences organized by the OECD and the EU, dealing with the issue. Indeed, a number of studies have already delineated and re- delineated the self-contained LMAs or basins (zones) of employment of Great Britain (Smart, 1974; Ball, 1980; Coombes and Openshaw, 1982; Coombes et al., 1986; Of- fice for National Statistics and Coombes, 1998; Coombes et al., 2005), Italy (Istituto Nazionale di Statistica, 1997), Denmark (Kristensen, 1998), the Netherlands (van der Laan, 1991; van der Laan and Schalke, 2001), a number of French and Spanish regions (Schmitt and Henry, 2000; Casado-Díaz, 2000; Poper, 2005; Royuela and Vargas, 2007), New Zealand (Papps and Newell, 2002), and the travel-to-work areas of several national capitals (OECD, 2007). For the most part, these approaches are based on an iterative process (algorithm) of grouping in a consistent manner contiguous or nearby localities, wards or municipalities, according to daily commuting flows from the place of residence to the place of work, and residence- and work-place-based self-containment criteria. Thus, unlike the conventional territorial partitions, the spatial formations recovered via this process display a good deal of the functional linkages between a ‘core’ area and its surrounding territories, and constitute territorial partitions on the basis of the economic interdependencies of localities. Consequently, they have gained considerable acceptance by the scientific community and governments as the appropriate territorial grid for diagnosing and analyzing regional disparities, engaging in policy interventions, relying on spillovers, etc. Indeed, in view of the complexity and continuing restructuring of modern economic reality, the determination and periodic revision of a country’s functional areas are probably very much to be desired. However, the instruments employed in these calculations, such as the definitions of cores and surrounding areas, commuting directions, and the criteria for attaching together or detaching areas, vary considerably from one case to another (Casado-Díaz and Coombes, 2005). A notable exception is provided by Eurostat, the EU’s principal statistical agency, the staff of which studies sub-national development across countries in a uniform manner. Hence, a dataset consisting of the thirty biggest commuting outflows from each municipality (or LAU 1 level district, in terms of NUTS P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 113 typology)1 is employed in order to identify, via an iterative process, the boundaries of the metropolitan and other main urban LMAs by attaching to city-cores the surrounding municipalities that exhibit substantial commuting flows to the city-core or to an iteratively enlarged core. In fact, the agency’s statisticians have tested the use of a narrow commuting threshold of 20%, as well as a wider commuting threshold of 15%, which is more challenging to work with (Carlquist, 2006).2 However, a growing recognition of the possibilities of LMA statistics in offering insights with regard to the performance of regions, prompted Eurostat, in late 2007, to invite the member states to report not only their main urban LMAs but all LMAs. Among the member states, Greece is characterized by a very fragmented landscape. Located at the southern part of the Balkan peninsula (Figure 1) and inhabited by 10.934 million residents, the country covers an area of some 132 thousand square kilometers (2001 Census figures). The terrain is dominated by high mountain-chains (about 42.2% of the country’s surface), small valleys traversed by rivers or inlaid with lakes, narrow coastal strips, a multitude of islands (about 35.1% of the country’s surface), and a very jagged coastline, extending for 15 thousand kilometers of which 6.1 thousand kilometers are on the mainland and 8.9 on the islands. (To give a measure of comparison: Greece’s coastline accounts for 13.6% of the EU-27 total, packed in a rather small area, about 3.1% of the EU landmass.) These natural features greatly fragment the country into a host of tiny districts. Obviously, the splintering impact of the landscape is, to some extent, tempered by the effectiveness of the transportation network linking these districts (coastal strips, plateaus, and islands). Yet, it might not come as a surprise, if a uniform set of criteria was applied across the EU, to see the number of the Greek functional economies per square kilometer exceed the EU average. In Greece, the need to empirically determine the country’s travel-to-work areas and delineate the labor markets areas within which policies may be most effective has been persuasively articulated by Efstratoglou (2006). However, up until early 2008,

1. The Nomenclature des Unités Territoriales Statistiques (NUTS) is the five-tier hierarchical structure used in the EU to standardize territorial units. In Greece, the administrative regions (periferies) correspond to NUTS level 2 sized-districts; prefectures (nomoi) correspond to NUTS level 3 sized- districts; municipalities (demoi or koinotites) to upper level Local Administrative Units (LAU 1, formerly NUTS level 4), and their subdivisions or wards (e.g., demotika diamerismata or koinotika diamerismata) to lower level Local Administrative Units (LAU 2, occasionally still termed NUTS level 5). 2. In essence, if a relatively large share of a municipality’s employed residents (say, 15% or more) commute to the city-core, then the municipality is considered as sufficiently integrated with the core and is treated as a part of its LMA. Moreover, if other municipalities exhibit similar commuting patterns vis-à-vis the city-core or the iteratively enlarged core then these municipalities are treated as parts of the particular LMA as well. 114 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 the detection of Greek LMAs could only be made by means of the thirty biggest commuting outflows from each municipality as solicited in the course of the 2001-Cen- sus, i.e., the dataset used by Eurostat. The recent availability of the complete set of commuting flows, in response to Eurostat’s request for more sophisticated treatment and a thorough report, has enabled the production of a more accurate determination of the country’s LMAs at both the 20% and the 15% incoming and outgoing commuting thresholds. The relevant calculations according to the former threshold bring the number of LMAs to 754 and according to the latter to 667.3 Understandably, the employment of a lower threshold affects the aggregation of localities into broader formations on the basis of fewer commuters traversing the mountainous or other ter- rains. This yields a smaller number of rather larger LMAs, which, on average, include a number of less integrated communities compared to the formations drawn on the basis of a higher threshold. Considering that the 15% threshold employed hereinafter is also one of the lowest used by EU member states in their replies to Eurostat, the recovery in Greece of one of the largest number of LMAs across the EU is probably reflective of the country’s idiosyncratically fragmented terrain. The rest of the article is concerned with the inter-municipal commuting flows recorded in the 2001 Census in order to determine the LMAs of Greece at the 15% commuting threshold. Section 2 explains how the LMAs are determined, Section 3 supplies the results, and Section 4 concludes.

2. The methodology used in the delineation of the LMAs As already mentioned, we use the same building blocks as Eurostat (namely, LAU 1 areas), and employ all incoming and outgoing commuting data. However, we en- hance the method by considering two-way commuting flows, i.e., from the fringes to the core and vice-versa, which, despite increasing the project’s complexity, provides a more complete picture of labor market linkages than one-way commuting. Indeed, two-way commuting is employed in both the UK self-containment algorithm and the North American labor market definitions. Additionally, we take the labor market delineation technique to the next logical step, and apply it throughout the country by examining not only the commuting patterns around the main urban centers, but across all 1,034 Greek municipalities. Thus, we codify the commuting origins and destinations in a non-symmetrical 1,034 by 1,034 interaction matrix, and proceed

3. A comprehensive list of each LMA’s constituent communities and of the communities that meet each threshold is provided in the discussion paper under the title Deriving Labor Market Areas in Greece from Commuting Flows, located at http://www.kepe.gr/pdf/D.P/dp_99.pdf. P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 115 to cluster them without contiguity restrictions.4 In effect, a municipality (or an iteratively enlarged LMA) will be grouped with another municipality or LMA if either (a) 15% or more of its employed residents commute to the other municipality or LMA, (b) 15% or more of all persons employed in the municipality commute from the other municipality or LMA, or both (a) and (b).5 We illustrate this with an example which makes use of the travel-to-work flows displayed by the employed residents in the town of Kozani (pop. 49,812 in West Mace- donia) and its environs (Table 1). Apparently, Kozani attracts 25% of the employed persons residing in the municipality of Dimitrios Ipsilantis (pop. 2,861), 23% of the employed persons residing in the municipality of Elimia (pop. 6,320) and 23% of the employed persons residing in the municipality of Eani (pop. 3,746).6 At the same time, commuters from the town of Kozani make up 23% of all who are employed in the municipalities of Ellispontos (pop.7,481) and Dimitrios Ipsilantis.7 As all five localities meet or exceed either the incoming or the outgoing commuting threshold of 20%, they are taken to comprise a functional area at the particular threshold. Col- lectively, they attract 19% of the employed residents in neighboring Velvendos (pop. 3,549),8 while commuters from them make up 1,9% of all persons employed in Vel- vendos.9 As neither of the two commuting ratios meets the 20% threshold, Velvendos is left outside the functional area formed by the other five localities at the 20% incom-

4. Hypothetically speaking, the imposition of such restrictions, while allowing the consideration of adjacent localities that may not be well linked (as contiguity does not necessarily translate to actual rail/motorway or direct and steady ferry connectivity), may prevent the consideration of detached or somewhat distant continental localities that are well linked and may exhibit substantial travel flows through the transportation network. Additionally, it is at variance with the commitment to exclusively rely on (and map) economic interactions. At any rate, only three non-contiguities are detected, involving localities attached to the Thessaloniki, Patras and Serre LMAs. 5. The prevailing view is that LMAs are not supposed to overlap. In our case, only two municipalities raise any concern regarding the broader travel-to-work area to which they ought to be attached: Avlis (to the LMA of Athens or to the LMA of Halkis) and Dimitrios Ipsilantis (to the LMA of Kozani or to the LMA of Ptolemais). The issue is resolved in favor of their incorporation into the Athens and Kozani zones, respectively, on the basis of both (a) the relative sizes of the flows (for they join the Athens and Kozani LMAs on the basis of the 20% rather than the 15% threshold), and (b) the similarities displayed in terms of their male and female employment, unemployment, and non-participation patterns (i.e., better t-statistics and fits in a number of regressions). 6. I.e., 209 out of the 831 employed persons residing in Dimitrios Ipsilantis, 244 out of the 1,062 employed persons residing in Eani, 482 out of the 2,070 employed persons residing in Elimia. 7. I.e., 615 out of 2,711 in the case of the former, and 225 out of 998 in the case of the latter. 8. In particular, 155 of the municipality’s 1,183 employed residents commute to Kozani, 27 to El- lispontos, 3 to Eani, 1 to Dimitrios Ipsilantis. 9. Involving 16 residents from Kozani, and 1 from each Elimia and Ellispontos out of the 942 people employed in Velvendos. 116 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 ing or outgoing threshold, but it is included in the enlarged LMA formed at the lower threshold of 15%. It follows that LMA sizes may change as commuting thresholds are modified. Having shown how an LMA is formed, we turn to the results obtained across the country, at the lower of the two thresholds.

3. The LMAs We find that the largestL MAs are located around Athens, Thessaloniki, and the urban centers of Patras, Iraklion, Larisa, Volos, and Ioannina (Table 2). All encompass a number of urban and rural municipalities, which are denoted in Figure 2 with black and dark gray, respectively. In particular: ●● The Athens LMA seems to consist of 120 municipalities, the majority of which are situated in the region of Attiki and five in the prefecture ofB oeotia (in Central Greece). The LMA occupies a surface of 3,609 square kilometers (2.7% of the country’s land-surface)10 and, at the time of the Census hosted a population of 3.887 million inhabitants (35.6% of the country’s population); i.e., 600 thousand more residents and about eight times more than the area of the homonymous urban-planning complex. This corresponds to 79.7% and 3.7% of the overall surface of the administrative regions of Attiki and Central Greece-Euboea, respectively, and 99.1% and 5.2% of the corresponding regional populations.11 ●● The Thessaloniki LMA spans 45 municipalities situated in the homonymous prefecture and the neighboring prefectures of Halkidiki and Kilkis. It occupies a total surface of 3,718 square kilometers (2.8% of the country’s surface)12 and, at the time of the Census, hosted a population of 1.090 million inhabitants (10.0% of the country’s population); i.e., 312 thousand more residents and 15-times more than the area of the homonymous urban-planning complex. This corresponds to 19.8% of Central Macedonia’s surface and 58.1% of the region’s population. ●● The LMA of Patras (12 municipalities) hosted 245 thousand people, and those of Iraklion (17 municipalities) 233 thousand, Larisa (11 municipalities) 184 thousand, Volos (11 municipalities) 154 thousand, and Ioannina (17 municipalities) 131 thousand. Taken together, these areas accounted for 8.7% of the country’s population and occupied 5.4% of the land.

10. The minor divergence from the figure provided in Table 2 is accounted by inland water-covered surfaces, such as lake Marathon. 11. The figures relate to the continental part of Attiki and the island of Salamis, thus leaving out a number of islands and a small slice of the east Peloponnesian coast that form part of the administration of Attiki. 12. This corresponds to 3,649 square kilometers if inland water-covered areas are excluded (see Table 2). P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 117

Thirty-five LMAs and eight self-contained municipalities host populations ranging from 21 to 115 thousand inhabitants. They are indicated in Figure 2 with black and standard gray, which denote the urban and rural municipalities, respectively. (Their population, land-surface (water-covered areas excluded) and altitude figures are provided in Table 3). These are: ●● The LMAs of Hania (115 thousand people), Rhodes (109 thousand), Halkis (98 thousand), Trikala, Kalamata, Serre (80-82 thousand), Kavala, Kerkira, Kozani, Lamia, Corinth, Agrinion (70-74 thousand), Komotini, Xanthi, Drama, Kateri- ni, Veria (61-69 thousand), Alexandroupolis, Karditsa, Tripolis, Kastoria (50-56 thousand), Rethimnon, Arta, Ptolemais, Hios, Mitilini (41-48 thousand), the island of Zakinthos (39 thousand), and the LMAs of Egion (35 thousand), Kilkis, Thebes, Sparta, Levadia, Kos, Florina, Nafpaktos (21-29 thousand people). ●● The individual municipalities of Pirgos (36 thousands), Giannitsa, Amalias, Argos (30-32 thousand), Edessa, Ierapetra, Naousa, Orestias (22-26 thousand). These exhibit rather low commuting ratios to/from the surrounding communities, i.e., they appear to be self-contained. The majority of local authority units (indicated in Figure 2 with light gray) are quite small in terms of population size. (a) A small number of these units form ten inter-municipal LMAs, namely the island of Siros and the travel-to-work areas of Aliverion, Grevena, Argostolion, Hri- soupolis, Igoumenitsa, Amfissa, Lefkas (14-20 thousand), Distomon, Tinos (8-9 thousand). (b) The rest exhibit low inward and outward commuting ratios. This suggests that the majority of these (mainly rural) municipalities are rather self-contained. In particular: ●● 190 municipalities host populations ranging from 5 to slightly less than 20 thousand inhabitants. They collectively account for 14.9% of the country’s population and 30.3% of the land. ●● 417 municipalities have fewer inhabitants. They collectively account for 9.0% of the country’s population and 38.3% of the land. A good number of them are insular communities or communities that lie along the Pindos mountain-range and its Peloponnesian extensions that splits the mainland into east and west. A brief summary of the population and land-surface features of the above clusters of municipalities and self-contained municipalities under (a) and (b), is provided in Tables 4 and 5. Considering that, on average, the smallest of these in terms of: ●● acreage (less then 115 km²) are situated on the insular Ionian, South and North Aegean regions and the insular sub-region of Attiki; and 118 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128

●● population size (about 3,050 inhabitants) are situated on the mountainous regions of Epiros and West Macedonia (at average altitudes of 566 and 820 meters, respectively: the highest in the country); it appears that the feature of economic fragmentation, relative seclusion and low labor market integration/interdependency may be associated with the country’s fragmented terrain (i.e., with the existence of many islands or mountains). Table 6 probes the issue by considering the correlations (i) between population size and both land surface and population density across all LMAs, and (ii) among land-surface, population density, and altitude across the three types of LMAs.13 According to the estimates with absolute values in excess of 65%, the most populous LMAs are generally more extensive in terms of surface and also more crowded (column 1). Additionally, of the seven largest LMAs, those that are more extensive in terms of surface are generally more densely populated as well. By contrast, those of the small LMAs that are more extensive are by-and-large thinly populated (column 2), and those that are thinly populated are generally situated at higher altitudes (column 4). Indeed, as the cost of energy in traversing mountainous formations is higher, it is reasonable to assume that highland communities may be seen by many as less attrac- tive domiciles and workplaces to commute to/from. Accordingly, labor market integration is more likely to occur along transport corridors, such as the small valleys and narrow coastal strips, where the cost of moving or hauling is lower. In an interesting departure from J.H. von Thünen’s theory of concentric rings of settlements formed on flat homogeneous land around a central city, or as a paradigm ofL MA adjustment to a distinctively idiosyncratic terrain,14 a good number of large and modestly-sized Greek LMAs appear to follow the mountainous contours and to have assumed stripe- form shapes that have little or no interaction with the surrounding areas lying further away.15 Likewise, coastal localities separated by long inlets of water, and islands constitute separate economic spaces. Obviously, spatial seclusion may have important implications regarding the degree of attractiveness of these localities to a number of industries, as well as the effectiveness and spillover-potential of regional development policies. Hence, it draws attention to the importance of the transportation network in facilitating traveling be-

13. I.e., (a) the seven largest LMAs, (b) the forty-three medium-size LMAs with populations ranging between 21 and 115 thousand, and (c) the thirteen regional types of smaller LMAs with less than 20 thousand inhabitants. 14. It is this very environment that gave rise to the diverse patchwork of city-states and tribes in classical antiquity. 15. I am indebted to Prof. Helmut Maier of the Berlin School of Economics for bringing it to my attention. P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 119 tween localities and effecting the economic unification of neighboring communities, and – by extension – the country. In closing, we note that there is not much evidence of labor market integration between major cities other than Athens and Piraeus (the port of Athens) or of transnational commuting. The lack of transnational commuting is not surprising given that the statistics date from a period when Greece did not share common borders with other EU member-states. In short, the country’s national borders were also EU borders, with all the limitations in trans-border commuting flows this entails.

4. Conclusions The daily commuting information obtained from employed persons permits the determination of the labor markets of Greece across the country’s 1,034 municipalities, in a manner that was not previously possible. We analyze all two-way inter-municipal commuting flows at the 15% threshold, which is used in Eurostat’s pilot study regarding national capitals and major urban centers. The iterative computation process used suggests that Greece contains 667 distinct LMAs. The largest ones are situated around the cities of Athens, Thessaloniki, Patras and Iraklion, which taken together host 49.9% of the country’s total population. There also exist thirty-eight smaller clusters of municipalities and eight self-contained municipalities hosting populations ranging from 21 to 185 thousand inhabitants, which jointly host 25.0% of the country’s population. The remaining LMAs and self-contained municipalities are smaller. The exercise has allowed us to take a fresh look at the economy as it truly is (i.e., a collection of clusters and communities) without preconceptions that localities or economic sub-spaces must fit into the inherited regional administrative framework. Indeed, the visual representation of these micro-regional formations on a map challenges the conventional perception of how modern-day Greece is organized and functions. It recalls the geographic relief of the country (e.g., a good number of small self-contained localities is situated along the Pindos mountain-range), but bears little resemblance to the patchwork of the NUTS levels 2 and 3 administrative divisions (i.e., the 13 regions and 54 prefectures) employed by the national and the EU authorities to design the country’s regional development. This suggests that economic life may not (a) correspond to the presumed 13 or 54 labor markets or territorial units of policy intervention of Greece or (b) run along the urban/rural divide. As a result, the survey areas for soliciting a good number of economic and social statistics may have to be re-drawn; and the spatial dimension of employment, unemployment or other policy-initiatives (pertaining to social cohesion, transportation, the environment, the spread of epidemics, the attraction of foreign investment etc.) to be accordingly re- 120 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 focused.16 In this context, it might make sense if a number of such issues were assigned to authorities operating at the LMA level, even if the constituent localities be- long to different administrative districts for historical, electoral or other reasons. For instance, the Thessaloniki LMA stretches beyond the homonymous urban-planning complex, over the western part of the Thessaloniki prefecture, as well as parts of the neighbouring Halkidiki and Kilkis prefectures. It seems that the localities forming an LMA ought to enjoy/engage in a uniform treatment of, say, labor-related issues on which they are highly interdependent. Understandably, a prescription that meets their needs may not be as suitable for the localities in another LMA, the authorities of which may have to work out what is appropriate in their case. In contrast, in the case of an LMA that cuts across administrative lines, reliance on the coordination of two authorities set up in different administrative districts (with each of these authorities adopting different policies in the different territories (other LMAs) under its purview) appears to be more cumbersome. Last but not least, this first snapshot of the country’s LMAs is quite relevant to discussions taking place in Greece regarding mergers of municipalities. Rezoning municipalities is expected to (a) generate economies of scale in terms of staffing and policy coordination, as well as (b) align the average size of local authorities with the average size encountered in a number of EU countries. However, our findings suggest that it might be unwise to reduce the number of micro-regional areas in which economic initiatives can be planned/combined (i) below the figure of 667, (ii) before fresh results are obtained via the forthcoming Census. The next Census is scheduled to take place in 2011 and supply commuting data at an even more disaggregated level (LAU 2). This will facilitate the production of a more thorough map of LMAs. On the other hand, if the existing municipalities are merged into a considerably smaller number of larger formations (say 300 to 400), presumably on the basis of a lower commuting threshold or other criteria (historical, political etc.), then there is a good chance that many of the new, enlarged authorities will constitute little more than amalgamations of poorly integrated areas. This may seriously compromise the assumption of suitable place-based policies across large tracts of the country.

16. For example, in building a case for attracting private investment to a place, it is probably wiser to factor in the age and skill profile of the resident workforce or other assets at the LMA level, as opposed to the age and skill profile of inhabitants or the other assets to be found in the individual community where the plant is to be located. P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 121

Table 1: In- and out-commuting flows observed at the town of Kozani and its environs (in West Macedonia)

Residents Employed Sum of (a) in the (i) in the municipalities of (ii) employed municipality of Kozani D. Ipsilantis Eani Elimia Ellipsontos Velvendos elsewhere residents

● Kozani 14,329 225 40 69 615 16 1,760 17,054 ● D. Ipsilantis 209 482 1 12 127 831 ● Eani 244 11 609 5 44 149 1,062 ● Elimia 482 10 4 1,322 70 1 181 2,070 ● Ellispontos 302 71 2 1,704 1 180 2,260 ● Velvendos 155 1 3 27 892 105 1,183

(b) in other places 1,150 198 24 18 239 32

Sum of residents and commuters 16,871 998 680 1,417 2,711 942 employed in a place

Source: Own calculations based on the National Statistical Service of Greece (NSSG), the 2001 Population Census figures.

Table 2: The seven largest LMAs of Greece on the basis of the 15% in- and out-commuting ratios

Land surface Average alti- Population (excl. water) tude (meters)

Of the Number of municipalities In In urban In urban Km² Total urban thousands wards wards wards

1 Athens-Piraeus 3,887 97% 3,607 38% 175 110 115 in Attiki, 5 in Central Greece 2 Thessaloniki 1,090 86% 3,649 13% 259 93 45 in Central Macedonia 3 Patras 245 86% 1,251 11% 306 27 12 in Western Greece 4 Iraklion 233 75% 1,202 13% 316 98 17 in Crete 5 Larisa 184 78% 1,660 9% 173 70 11 in Thessaly 6 Volos 154 84% 825 8% 187 38 11 in Thessaly 7 Ioannina 131 70% 1,882 5% 671 499 17 in Epiros

Source: See Table 1. 122 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128

Table 3: The forty-three medium-size LMAs of Greece on the basis of the 15% in- and out-commuting ratios

Land surface Average alti- Population (excl. water) tude (meters) Number of Of the In In urban In urban municipalities Km² Total urban thousands wards wards wards

1 Hania 115 82% 794 17% 259 56 12 in Crete 2 Rhodes 109 73% 787 12% 146 27 8 in South Aegean Isl. 3 Halkis 98 82% 749 14% 204 17 6 in C. Greece-Euboea 4 Trikala 82 71% 561 12% 171 115 8 in Thessaly 5 Kalamata 81 81% 524 16% 314 32 5 in Peloponnesos 6 Serre 80 70% 669 14% 166 53 5 in Central Macedonia 7 Kavala 74 82% 351 11% 256 53 2 in East Macedonia 8 Kerkira 74 80% 283 39% 139 70 7 in Ionian Islands 9 Kozani 74 58% 1,180 5% 713 669 6 in West Macedonia 10 Lamia 73 72% 772 9% 290 98 3 in Central Greece 11 Corinth 71 59% 557 10% 127 22 5 in Peloponnesos 12 Agrinion 70 80% 274 37% 105 85 3 in West Greece 13 Komotini 69 69% 927 7% 84 44 4 in Thrace 14 Xanthi 67 69% 445 7% 51 80 3 in Thrace 15 Drama 67 67% 1,930 3% 498 115 5 in East Macedonia 16 Katerini 63 98% 118 82% 41 31 2 in Central Macedonia 17 Veria 61 81% 526 14% 534 99 3 in Central Macedonia 18 Alexandroupolis 56 89% 804 19% 183 10 2 in Thrace 19 Karditsa 54 66% 491 7% 379 105 4 in Thessaly 20 Tripolis 54 67% 1,583 3% 732 582 8 in Peloponnesos 21 Kastoria 50 58% 1,262 7% 895 679 11 in West Macedonia 22 Rethimnon 48 73% 497 10% 356 45 4 in Crete 23 Arta 48 55% 426 9% 234 26 6 in Epiros 24 Ptolemais 47 64% 709 8% 750 600 5 in West Macedonia 25 Hios 46 62% 472 7% 216 28 6 in North Aegean Isl. 26 Mitilini 41 77% 187 10% 83 19 2 in North Aegean Isl. 27 Zakinthos Isl. 39 58% 405 15% 230 53 6 in Ionian Islands 28 Pirgos 36 73% 169 31% 47 16 1 in Peloponnesos 29 Egion 35 84% 237 19% 372 44 2 in Western Greece 30 Giannitsa 32 93% 208 90% 51 45 1 in Central Macedonia 31 Amalias 32 71% 247 31% 105 42 1 in Peloponnesos 32 Argos 30 87% 138 38% 118 40 1 in Peloponnesos 33 Kilkis 29 66% 478 23% 197 274 2 in Central Macedonia 34 Thebes 29 77% 482 29% 246 180 2 in Central Greece 35 Edessa 26 74% 321 12% 444 320 1 in Central Macedonia 36 Sparta 26 72% 518 5% 589 217 3 in Peloponnesos 37 Levadia 24 89% 278 50% 180 204 2 in Central Greece P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 123

38 Kos 24 75% 130 52% 33 15 2 in South Aegean Isl. 39 Ierapetra 24 66% 395 18% 232 18 1 in Crete 40 Florina 23 67% 362 7% 851 663 2 in West Macedonia 41 Naousa 22 0% 301 383 1 in Central Macedonia 42 Orestias 22 79% 255 41% 73 31 1 in Thrace 43 Nafpaktos 21 78% 211 13% 249 20 2 in Western Greece

Source: See Table 1.

Table 4: The regional distribution of the small LMAs of Greece (with less than 20 thousand inhabitants) on the basis of the 15% in- and out-commuting ratios

Land surface Average alti- Population (excl. water) tude (meters) Number Of the of LMAs In In urban In urban Km² Total urban involved thousands wards wards wards

1 Central Macedonia 473 12% 12,616 2% 238 27 75 2 Peloponnesos 337 19% 12,154 1% 425 29 85 3 Central Greece-Euboea 305 6% 12,571 1% 435 83 72 4 Western Greece 284 13% 8,678 2% 399 43 53 5 Thessaly 266 13% 10,467 2% 429 154 70 6 East Macedonia-Thrace 251 8% 9,351 1% 240 23 37 7 Crete 175 24% 5,448 2% 333 108 36 8 South Aegean Isl. 165 23% 4,369 2% 165 76 45 9 Epiros 158 25% 6,776 2% 566 18 52 10 North Aegean Isl. 118 20% 3,164 2% 165 32 28 11 West Macedonia 101 17% 5,746 2% 820 541 33 12 Ionian Isl. 97 42% 1,611 9% 210 49 22 13 Attiki 36 29% 782 7% 139 31 9

Source: See Table 1. 124 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128

Table 5: Average population and land-surface figures of the small LMAs of Greece (with less than 20 thousand inhabitants), by region

Population (in thousands) Land surface excluding water (in km²)

1 Epiros 3,033 1 Ionian Isl. 73 2 W. Macedonia 3,065 2 Attiki 87 3 S. Aegean Isl. 3,668 3 S. Aegean Isl. 97 4 Thessaly 3,802 4 N. Aegean Isl. 113 5 Peloponnesos 3,966 5 Epiros 130 6 Attiki 4,013 6 Peloponnesos 143 7 N. Aegean Isl. 4,213 7 Thessaly 150 8 C. Greece-Euboea 4,240 8 Crete 151 9 Ionian Isl. 4,404 9 W. Greece 164 10 Crete 4,851 10 C. Macedonia 168 11 W. Greece 5,358 11 W. Macedonia 174 12 C. Macedonia 6,309 12 C. Greece-Euboea 175 13 E. Macedonia-Thrace 6,789 13 E. Macedonia-Thrace 253

Source: See Table 1.

Table 6: Correlations across the features of the seven large LMAs, the forty- three medium-size LMAs and the thirteen regional types of small LMAs

Land surface and Land surface Altitude and All LMAs LMAs pop. density and altitude pop. density (1) (2) (3) (4) Population Large 0.6765 -0.1264 -0.4043 Land surface 0.7089 Medium-size -0.4867 0.5383 -0.5145 Pop. density 0.8472 Small -0.6731 0.3375 -0.7394

Notes: In columns (2)-(4) the correlation coefficients are calculated for each group ofL MAs separately. The population size of the large, the medium-size and the regional representatives of small- size LMAs is above 131 thousand, between 21-115 thousand, and less than 20 thousand inhabitants, respectively. Source: See Table 1. P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 125

Figure 1: The terrain of Greece and neighboring regions

Rodopi Mts.

Limnos Mt. Olympus isl. P I N D O S M T.

Lesvos Corfu isl. isl. Euboea Kefallinia isl. Hios isl. isl.

Samos isl.

Peloponnese peninsula Rhodes isl. Crete isl.

Source: NASA/GSFC (2003) 126 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128   Rural munic.    S. Aegean Isl. S. N. Aegean Isl. N. (>150 people/km²) Urban municipalities E. Macedonia-Thrace Attiki Crete with 20-115 thousand inhabitants with 20-115 with fewer inhabitants C. Greece-Euboea L Ms with over 130 thousand inhabitants L MAs and self-contained municipalities Key to color classifications Key to color Thessaly C. Macedonia R W. Greece W. Epiros Peloponnesos The administrative regions of Greece The administrative regions Macedonia W. Ionian Isl. Ir Ha S A H T V L K P Tr I Figure 2: The country’s LMAs on the basis of 15% in- and out-commuting ratios those employed in 2001 The country’s 2: Figure Athens Ioannina Iraklion Patras Rhodes Serre Thessaloniki Trikala Volos

A h h alkis H a h ania I Ir k k alamata l l arisa P R S T Tr V P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 127

References Ball, R.M., 1980, “The Use and Definition of Travel-to-Work Areas in Great Britain: Some Prob- lems”, Regional Studies, 14, 125-39. Carlquist, T., 2006, “The Larger Urban Zones in the Urban Audit data collection”, 25th Conference of the Standing Committee on Regional and Urban Statistics (Europe Group). Wroclaw. Casado-Díaz. J.M., 2000, “Local Labour Market Areas in Spain: A Case Study”, Regional Studies, 34, 843-56. Casado-Díaz, J.M. and M. Coombes, 2005, “The Delineation of 21st Century Local Labour Market Areas (LLMAs)”, Proceedings of the 8th Network of European Communications and Transport Activity Research Conference, Las Palmas G.C. Coombes, M., S. Raybould S., and C. Wymer, 2005, Travel to Work Areas and the 2001 Census: initial research. London: Office for National Statistics. Coombes, M.G., A.E. Green and S. Openshaw, 1986, “An Efficient Algorithm to Generate Official Statistical Reporting Areas: The Case of the 1984 Travel-to-Work Areas Revision in Britain”, Journal of the Operational Research Society, 37, 943-53. Coombes, M.G., and S. Openshaw, 1982, “The Use and Definition of Travel-to-Work Areas in Great Britain: Some Comments”, Regional Studies, 16, 141-49. Efstratoglou, A, 2006, Local Labor Markets in Greece, [In Greek.] Athens: Labor Institute of the Greek General Confederation of Labor, and Confederation of Civil Servants. Istituto Nazionale di Statistica, 1997, I sistemi locali del lavoro 1991, edited by F. Sforzi, Rome: ISTAT. Kristensen, K, 1998, Functional Economic areas of Denmark: Applying Input-Output Techniques to Commuting, Nexo: Center for Regional and Tourism Research / Research Centre of Bornholm. OECD, 2007, Defining and Measuring Metropolitan regions – Follow up of the International Workshop 2006, edited by J.E.Garcilazo. Paris: OECD Working Party on Territorial Indicators. [GOV/TDPC/TI(2007)2] Office for National Statistics and M.G. Coombes, 1998,1991-based Travel to Work Areas, London: Office for National Statistics. NASA/GSFC, 2003, Image of Balkans of March 26th from the Terra satellite, http://visibleearth. nasa.gov/images /5232/Greece A2003085.0920.500m.jpg. Papps, K.L, and J.O. Newell, 2002, Identifying Functional Labour Market Areas in New Zealand: A Reconnaissance Study Using Travel-to-Work Data, Discussion Paper 443, Bonn: Institute for the Study of Labor. Poper, S., 2005, “The Interaction Value: Its Scope and Limits as an Instrument for Delimiting Urban Systems”, Regional Studies, 39, 257-373. Royuela, V., and M. Vargas, 2007, Defining housing market areas using commuting and migration algorithms. Catalonia (Spain) as an applied case study, Barcelona University IREA Working Paper Series #200707. Schmitt, B., and M.S. Henry, 2000, “Size and growth of urban centers in French labor market areas: consequences for rural population and employment”, Regional Science and Urban Economics, 30, 1-21. Smart, M.W., 1974, “Labor market areas: Uses and definition”,Progress in Planning, 2, 239-351. 128 P.-I. K. PRODROMIDIS, South-Eastern Europe Journal of Economics 1 (2010) 111-128 van der Laan, L., 1991, Spatial Labour Markets in The Netherlands, Delft: Eburon. van der Laan, L., and R. Schalke, 2001, “Reality versus Policy: The Delineation and Testing of Lo- cal Labour Market and Spatial Policy Areas”, European Planning Studies, 9, 201-221 Book Review

Wealth, Welfare and the Global Free Market A Social Audit of Capitalist Economics by Ibrahim Ozer Ertuna Gower Publishing Ltd, England 2009. Pp. 233 reviewed by Theodore P. Lianos*

The subtitle of this book, “A Social Audit of Capitalist Economics”, gives a good idea of what it contains. The last paragraph of the text on p. 223, is also a good but different summary “The world will have to create an alternative or alternatives to capitalism. Today, as yet, there is no ready recipe we can use. The solution will emerge through the wide and increased participation of people enabled by the advances in technology. Creating solutions in line with the aspiration of humankind requires increased awareness. This is what we have tried to accomplish in this book”. The first ten chapters of the book present textbook material of economic theory from a critical point of view. However, it is not always clear whether the criticism is against economic theory or against capitalism as an economic system. For example, on p. 41, he says “Capitalism does not take into consideration these ethical and moral values and rules in making its assumptions and reaching its conclusions”. Chapter 11 examines what the author calls the new economic design which is the “neoliberal economy” or “gelatinized economy” or the “free market democracy”. The last chapter is a critique of the new economic design and the undesirable results that it has brought to the world. The book is well written and in some cases convincing. I also share the feelings of the author regarding his dissatisfaction with the ugly faces of capitalism. Everyone with common sense would recognize that capitalism is far from perfect, and would agree with the author of this book that we should always be searching for a better system. However, the book has weaknesses, some methodological and some substantive. In reading this book, the reader has the impression that the author makes the error of personifying capitalism, in other words, that capitalism is a humanoid entity that thinks, make decisions and acts accordingly. This conception of capitalism leads the

* Athens University of Economics and Business, Greece 130 BOOK REVIEW, South-Eastern Europe Journal of Economics 1 (2010) 129-132 author to statements that could easily be put in the mouths of activists. For example, “The new economic design was established under the name of globalization, in the framework of the capitalist market economy and in two phases. In the first phase, through an efficient propaganda machine, both developed and developing counties were persuaded of the virtues of capitalist market economies. In this stage, the capitalist market economy was converted into a creed. In the second phase, the capitalist market economy was reshaped to serve the interests of the developed rich countries”. The truth of these statements is not obvious. Theoretical analysis and empirical evidence are required before one can accept them as valid. Also, they are incomplete. For example, who is operating the propaganda machine and for what purpose? Who is persuaded and why? If globalization has reshaped the capitalist market economy to serve the interests of the developed rich countries, has this happened without benefits for the undeveloped poor countries? Has China, for example, become poorer because of globalization? And if so, why have poor countries agreed to become part of the new economic design? Or were they forced to participate? These and other questions need to be answered before the author’s critique can be accepted on a scientific level. On a more important level, the author of this book reverses the causality between social arrangements and ideas. On p. 221 he writes “Social systems are the products of people’s faith, beliefs and values” I believe the opposite to be true. It is the real conditions under which people live, namely, the existing economic reality that shapes beliefs and values, ideology in general. Finally, the author seems to feel free to make general statements of a questionable nature. For instance, the first paragraph of this book reads “…economics, which is one of the fields of science, is fast losing its scientific nature”. Despite, my disagreements with many statements in this book, I believe it is well written and readers will benefit from reading it. Book Review

Human Development in the Era of GLobalization Essays in Honor of Keith B. Griffin edited by J.K. Boyce, St. Cullenberg, Pr. K. Pattanaik and R. Pollin Edward Elgar Publishing Ltd. (2009), Cheltenham, UK, p.p. 392 reviewed by Theodore P. Lianos*

This volume is in honor of Professor Keith B. Griffin, a well known economist with important contributions to economics for a time period of almost half a century. It includes sixteen papers and in addition a substantive and lengthy introduction by the Editors and an autobiographical paper by Griffin, where he gives an account of how he saw and judged the economic affairs of the world after the Second World War. The presentation of the papers is organized in five parts. The first part, Perspec- tives on Chinese Development, includes a paper by C. Riskin on Chinese poverty, a paper by J. Knight, L. Shi and L. Song on the rural-urban divide in China, and a paper by M.D. Brenner on land tenure and growth in rural China. The second part, Agricul- ture and Rural Poverty, includes a paper by J.K. Boyce on biodiversity and sustainable agriculture, a paper by St.M. Helfand and E.D. Levine on policy reforms on rural poverty in Brazil and a paper by A. Berry on agrarian reform in Colombia. The third part, Dimensions of Human Development, includes a paper by Bob Sutcliffe on death and Development, a paper by Pr. Pattanaik on Functioning Bundles and Capability Sets and a paper by Ch. Perrings on environment, poverty and development. The fourth part, Globalization and Inequality, includes a paper by R. Pollin on egalitar- ian development, a paper by D. Elson on gender equality, a paper by A. Chakrabarti, St. Cullenberg and M. wa Githinji on class transition and a paper by V.D. Lippit on the welfare state. The last fifth part includes a paper by A. Rahmanand and B. Sen on personal income and poverty, a paper by R. Sobhan on poverty as injustice and a paper by T. McKinley on poverty reduction in Asia and the Pacific. As the editors note in their introduction, there are four basic themes that recur in the papers of this volume. First, the need to eradicate poverty and also to reduce inequality in income and wealth within as well as among countries. Second, the relationship between growth and distribution as a two-way relationship. Third, the rec-

* Athens University of Economics and Business, Greece 132 BOOK REVIEW, South-Eastern Europe Journal of Economics 1 (2010) 129-132 ognition that policies are not formulated on the basis of an optimizing procedure for the economy as a whole, but rather on the basis of specific interests of social classes. Fourth, the heterogeneity among various economies and the implication it has for economic development strategies. Some of the papers in this volume are highly technical, some contain econometric analyses, some are based on descriptive statistics, some analyze problems of specific countries etc. But to the extent one can make a general judgment on such a diverse collection, it is this reviewer’s opinion that the papers in this volume are of a very high scientific standard. Those professionals and graduate students who want to enter the area of economic development in a serious way will find this volume extremely useful. reminder

6th INTERNATIONAL CONFERENCE of

organized by Faculty of Economics, University of Montenegro

Economic Development, Tax Systems and Income Distribution in the Countries of Southern and Eastern Europe

May 21-23, 2010 Podgorica

[email protected] refereeacknowledgment The South Eastern Europe Journal of Economics gratefully acknowledges the continued support of the distinguished members of the Editorial Board and the following roster of economists who have contributed their services as referees during the period Spring 2008-Spring 2010. Adriotis Konstantions Nottingham Trent University, United Kingdom Aiginger Karl Austrian Institute of Economic Research, Austria Alagidede Paul University of Stirling, United Kingdom Andrikopoulos Andreas Athens university of Economics and Business, Greece Aristotelous Kyriacos Otterbein College, USA Athanassiou Ersi Centre for Planning and Economic Research, Greece Baharumshah Zubaidi Ahmad University Putra Malaysia, Malaysia Balfousias Stella Centre for Planning and Economic Research, Greece Baltas Nikolaos Athens University of Economics and Business, Greece Bitros Georgios Athens University of Economics and Business, Greece Bonin P.J. Wesleyan University, USA Carow Kenneth Indiana University, USA Castaldo Adriana University of Sussex, United Kingdom Cavounidis Jennifer Athens University of Economics and Business, Greece Daouli Joan University of Patras, Greece Demoussis Michael University of Patras, Greece Denčić-Mihajlov Ksenija University of Niš, Serbia Dinopoulos Elias University of Florida, USA South-Eastern Europe Journal of Economics 1 (2010) 135

Djukic Djordje University of Belgrade, Serbia Droucopoulos Vassilios University of Athens, Greece Economidis Georgios Athens University of Economics and Business, Greece Foelster Stefan Confederation of Swedish Enterprise, Sweden Gibson D Heather Bank of Greece, Greece Glytsos P. Nikolaos Center for Planning and Economic Research, Greece ✝ Hatziprokopiou Michalis University of Macedonia, Greece Karathanassis Georgios Athens University of Economics and Business, Greece Karayanni Despina University of Patras, Greece Köthenbürger Marko University of Munich, Germany Koutsomanoli-Filippaki Anastasia Council of Economic Advisors, Ministry of Finance, Greece Mantakas Georgios Piraeus Bank, Greece Mattas Konstantinos Aristotle University of Thessaloniki, Greece Michael S. Michael University of Cyprus, Cyprus Miliotis Panagiotis Athens University of Economics and Business, Greece Molyneux Philip Bangor Business School, United Kingdom Panigyrakis George Athens University of Economics and Business, Greece Papadogonas Theodore Technological Educational Institute of Chalkis, Greece Papapanagos Harry University of Macedonia, Greece Papavassiliou Nikolaos Athens University of Economics and Business, Greece 136 South-Eastern Europe Journal of Economics 1 (2010)

Pournarakis Efthimios Athens University of Economics and Business, Greece Prodromidis Kyprianos, Athens University of Economics and Business, Greece Prodromidis Prodromos Centre for Planning and Economic Research, Greece Pseiridis Anastasia University of Central Greece, Greece Rapsomanikis George Food and Agriculture Organization of the United Nations (FAO), Italy Sarris Alexander Food and Agriculture Organization of the United Nations (FAO), Italy Seskar Lidija Belgrade Stock Exchange, Serbia Sissouras Aris University of Patras, Greece Skarmeas Dionysis Leeds University Business School, United Kingdom Skountzos Theodore University of Piraeus, Greece Stringham Edward Trinity College, USA Tödtling Franz Vienna University of Economics and Business Administration, Austria Trihopoulou Anna University of Macedonia, Greece Tsakloglou Panos Athens University of Economics and Business, Greece Tsartas Paris University of Aegean, Greece Turley Gerard National University of Ireland, Galway, Ireland Visvikis Ilias ALBA Graduate Business School, Greece Von Hagen Jürgen Institut für Internationale Wirtschaftspolitik, Germany Zervogiannis Athina University of Patras, Greece Zotos George Aristotle University of Thessaloniki, Greece callfor papers

7th INTERNATIONAL CONFERENCE of

organized by Rostov State University of Economics Rostov-on-Don, Russia

Recent Economic Crisis and Future Development Tendencies

May 19-21, 2011 in Rostov-on-Don, Russia

Deadline for receipt of abstracts: January 16, 2011 Deadline for receipt of full papers: April 17, 2011

[email protected] callfor papers

General information

We welcome submissions addressing the ASECU 2011 International Conference theme:

Recent Economic Crisis and Future Development Tendencies

The conference is aimed to encourage development, cooperation and mutual assistance among the countries, industries and companies of South and East Europe and Black Sea Region. The conference will include contributed paper sessions, invited and plenary presentations, round tables and panels. Accepted papers will be published in the Conference Proceedings after their presentation at the Conference. The Editorial Board of South Eastern Europe Journal of Economics (SEEJE), the official journal of ASECU, after having the agreement of the authors, plans to publish selected conference papers following a standard refereeing process. The official language of the Conference will be English.

Conference Topics

•• Recent financial-economic crisis •• Innovation technologies •• Globalization and globalization as development and world economy two main vectors of modern world globalization economy development •• Transformation of economic systems of Russia, EU and South- •• Contemporary economics: Eastern countries as response to vindication of state economy globalization challenges reform •• Modernization of state economic •• Originalities of perception of doctrine of Russia, EU and South- international experience in Eastern countries with a glance to economic and political system post-crisis changes transformation of Russia, EU and •• Financial system role and functions South-Eastern countries in realization of foreground tasks •• Topical problems of financial law of Russia, EU and South-Eastern state regulation of economy in post- countries’ development in post- crisis period crisis period callfor papers

•• Peculiarities of creation of •• Modern information technologies innovation development models in in economic activity as a basis of developed and developing countries innovative development •• New methods of financial-economic •• Innovative anti-crisis strategies system development management in optimization of companies’ •• Development of theory and resource potential methodology of investment activity •• Peculiarities of investment in modern conditions activity management: national and •• Development of human potential international experience as a basis of economy and society •• Competitiveness of human potential sustainability in Russia, EU and South-Eastern •• Business social responsibility in countries developed and developing countries •• Social differentiation of society in •• Future of regional development: Russia and South-Eastern countries: international and national threat of political and economic experience in creation of cluster development disintegration in post- economy crisis conditions •• Corporate sector of economy in •• Companies’ competitiveness post-crisis conditions management in crisis conditions: •• Modern education policy and response of national entities to its influence on country’s socio- economic downturn economic development •• Modern tendencies of integration •• Supranational, national, regional development in Russia, EU and and municipal government: South-Eastern countries international and national •• Efficient technologies of anti-crisis cooperation experience management

Conference Program Committee yy Prof. Dr. Nikolay Kuznetzov Rector of Rostov State University of Economics yy Prof. Dr. Vladimir Zolotarev President of Rostov State University of Economics yy Prof. Dr. Adam Albekov Senior Pro-rector of Rostov State University of Economics yy Prof. Dr. Ludmila Usenko Pro-rector of scientific researches of Rostov State University of Economics callfor papers yy Prof. Dr. Ludmila Nivorozhkina Head of the Mathematical statistics, econometrics and actuarial evaluation department, Rostov State University of Economics yy Prof. Dr. Yannis Tsekouras, President of ASECU, University of Macedonia, Greece yy Prof. Dr. Atanas Damyanov, Vice President of ASECU, D.A. Tsenov Academy of Economics, Bulgaria yy Prof. Dr. Božidar Cerović, General Secretary of ASECU, University of Belgrade, Serbia yy Representative of Ministry of Education and Science of Russian Federation yy Representative of State Duma of Russian Federation yy Representative of the Ministry of Economics, Trade, International and External Economic Relations of Rostov region, Russia yy Representative of Presidential Envoy in Southern Federal District, Russia

Guidelines for Submission (All submissions are refereed): •• Submissions must not have been published, submitted or presented at other conferences. •• The full paper should be a maximum of fifteensingle-spaced pages including references and exhibits. Requirements for the paper formatting and abstract: Page: A4; Font: Times New Roman, Size: 12 pt; Line spacing single; Page Setup: Margins: Left: 4; Right: 4.5; Top: 6; Bottom: 3.5. •• Abstracts should be a maximum of one single-spaced page and up to four Keywords that describe your paper. •• The name(s) of the presenting author(s) and other co-author(s), affiliation(s), complete mailing address(es), telephone/fax numbers and e-mail address(es), title of the papers should be filled in the Application Form. •• Special scheduling requests (dates or times) must be made at time of paper submission. •• Electronic submissions are expected. Submissions should be in Microsoft Word format. In the subject should be written: ASECU conf. (name of the leading author) Submissions should be sent to the Organizing Committee: Rostov State University of Economics e-mail: [email protected] Tel /Fax: +7 863 2370253 callfor papers

Important Dates: January 16, 2011 Deadline for receipt of abstracts and application forms. February 18, 2011 Notification of acceptance. April 17, 2011 Deadline for submission of full papers.

Conference fee and Hotel Accommodation: The conference fee is 50 € / participant for ASECU members and 70 € / participant for non-ASECU members. Each participant should cover travel costs to and from Rostov-on-Don. Each participant should directly book accommodation in Rostov-on-Don. The following hotels are within walking distance from the Rostov State Uni- versity of Economics: •• Rostov Hotel (http://www.rostovhotel.ru) •• Don Plaza Congress-Hotel (http://www.don-plaza.ru) •• Pushkinskaya Hotel (http://hotel-pushkinskaya.ru) •• Radisson SAS Don Hotel (http://www.radissonblu.com/hotel-rostovon- don) The hotel reservation, the exact date and hour of the arrival and the departure of each participant, as well as the way of travelling have to be sent by email to the Organizing Committee of the conference in Rostov-on-Don ([email protected]) Participants who need assistance in booking accommodation in Rostov-on-Don should contact the Organizing Committee ([email protected]) not later than April 17, 2011. guidefor authors

The articles should be written as follows: (1) Papers must be in English. (2) Papers for publication (two copies) should be sent to: Mrs. Melina Petromelidou Editorial Secretary South-Eastern Europe Journal of Economics ASECU, University of Macedonia, 156, Egnatia Str., 540 06 Thessaloniki, Greece The Journal’s phone number is (+30) 2310891793, fax: (+30) 2310891748 e-mail: [email protected] Submission of a paper will be held to imply that it contains original unpublished work and is not being submitted for publication elsewhere. The Editor does not accept responsibility for damage or loss of papers submitted. Upon acceptance of an article, author(s) will be asked to transfer copyright of the article to the publisher. This transfer will ensure the widest possible dissemination of information. (3) Papers will be considered in any form, but authors of papers accepted for publication will be expected to provide a final copy conforming to the general style of the Journal as outlined in notes 4 through 13 below. (4) Manuscripts should be double spaced, with wide margins, and printed on one side of the paper only. All pages should be numbered in sequence. Titles and subtitles should be short. References, tables, and captions for the figures should be printed on separate pages. (5) The first page of the manuscript should contain the following information: (i) the title; (ii) the name(s) and institutional affiliation(s) of the author(s); (iii) an abstract of not more than 100 words. A footnote on the same sheet should give the name, address, and telephone and fax numbers of the corresponding author [as well as an e-mail address]. (6) The first page of the manuscript should also contain at least one classification code according to the Classification System for Journal Articles as used by the Journal of Economic Literature; in addition, up to five key words should be supplied. The classification system used in JEL can be found at: http://www.aeaweb.org/journals/jel_class_system.html. (7) Acknowledgements and information on grants received can be given in a first footnote, which should not be included in the consecutive numbering of footnotes. (8) Footnotes should be kept to a minimum and numbered consecutively throughout the text with superscript Arabic numerals. (9) Displayed formulae should be numbered consecutively throughout the manuscript as (1), (2), etc. against the right-hand margin of the page. In cases where the derivation of formulae has been abbreviated, it is of great help to the referees if the full derivation can be presented on a separate sheet (not to be published). (10) References to publications should be as follows: ‘Smith (1992) reported that...’ or ‘This problem has been studied previously (e.g., Smith et al., 1969)’. The author should make sure that there is a strict one-to-one correspondence between the names and years in the text and those on the list. The list of references should appear at the end of the main text (after any ap- pendices, but before tables and captions for figures). It should be double spaced and listed in alphabetical order by author’s name. References should appear as follows: For monographs Sen, A., 1970, Collective Choice and Social Welfare, San Francisco: Holden Day. For contributions to collective works Kornai, J., 1991, Stabilization and Economic Transition in Hungary: The Next Two Years, in J. de Melo and A. Sapir (eds.), Trade Theory and Economic Reform: North, South and East, Oxford: Basil Blackwell, 307-326. For periodicals Magdalinos, M., 1990, “The Classical Principles of Testing Using Instrumental Variables Es- timates”, Journal of Econometrics, 44, 241-279. Note that journal titles should not be abbreviated. (11) Illustrations will be reproduced photographically from originals supplied by the author; they will not be redrawn by the publisher. Please provide all illustrations in quadruplicate (one high-contrast original and three photocopies). Care should be taken that lettering and symbols are of a comparable size. The illustrations should not be inserted in the text, and should be marked on the back with figure number, title of paper, and author’s name. All graphs and diagrams should be referred to as figures, and should be numbered consecutively in the text in Arabic numerals. Illustration for papers submitted as electronic manuscripts should be in traditional form. (12) Tables should be numbered consecutively in the text in Arabic numerals and printed on separate sheets. (13) Accepted papers should be submitted in electronic form, i.e., on disk with accompanying manuscript. Electronic manuscripts have the advantage that there is no need for re-setting of text, thereby avoiding the possibility of introducing errors and resulting in reliable and fast delivery of proofs. The preferred storage medium is a 3.5 inch disk in MS-DOS system. The preferred format is either WORD or Word Perfect. Make absolutely sure that the file on the disk and the printout are identical. Use a new and correctly formatted disk and label this with your name; also specify the software and hardware used as well as the title of the file to be processed. (14) Page proofs will be sent to the corresponding author. Proofs should be corrected care- fully; the responsibility for detecting errors lies with the author. Corrections should be restricted to instances in which the proof is at variance with the manuscript. There are neither submission fees nor page charges. Twenty reprints of each paper are supplied free of charge to the corresponding author.