Tải bản đầy đủ - 0 (trang)
Chapter 14. Methodological and Practical Issues for the Evaluation of Territorial Pacts The Experience of Italy

Chapter 14. Methodological and Practical Issues for the Evaluation of Territorial Pacts The Experience of Italy

Tải bản đầy đủ - 0trang

14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Italian territorial pacts

A territorial pact is a specific policy instrument aimed at promoting local

development through financial incentives to a group of locally based and

integrated projects designed by a coalition of local actors (private and public).

In some cases financial resources for technical assistance are also made

available to the coalition.1

In Italy, as of September 2002, the number of territorial pacts approved

and considered eligible for financial support is quite large. There are 230 pacts,

of which 220 (including 91 pacts specialised in agriculture and fisheries) were

selected through a national procedure, and 10 selected during the update of

the programming of the Community Support Framework for OB.1 1994-1999

on the basis of a procedure agreed with the European Commission. The two

procedures differ. In the national procedure the selection of pacts projects

(initiated by a public national bid) simultaneously ends in the approval of the

general pact project and of all the single initiatives included. The national

procedure has been repeated and modified over time, so we now have

different cohorts of national territorial pacts approved. The European

procedure was only implemented once. It was carried out in two stages. First,

10 general projects and territories were chosen and later – through territorial

bids – single initiatives were selected.2

Although single territorial pacts do not cover very large areas, the

instrument is so widespread that a very large portion of the national territory

is affected (see Figure 14.1).

Territorial pacts in Italy have a very mixed reputation. Among politicians,

general observers and territorial experts they have fierce enemies and

determined defenders. Most of the debate (and most of the arguments in favour

or against) have not however been based on scientific evidence, rigorous

monitoring or evaluation research. They have rested mainly on conjectural

arguments, direct experience and, often, prejudices. In what follows, however,

we do not consider this general – and mainly media-driven – debate in detail.

Rather, we concentrate on what and why we might want to learn from

evaluating territorial pacts and report on what has been done so far in this

respect.



356



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Figure 14.1. Italy: areas covered by territorial pacts* (September 2002)



1. *The 91 pacts specialised in agriculture are not included (as they mainly overlap with other pact

areas).



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



357



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Objectives of territorial pacts and evaluation questions

The territorial pact is built around the idea that financing a coalition of

actors with a project could serve the purpose of bringing individual agents

together, giving rise to economies of agglomeration and capable partnerships.

In theory, if the desired outcome is to promote economies of agglomeration

and collective capacity, this line of action should be superior to financing the

single worthy but uncoordinated projects of individual private actors or public

entities (as it is the case for common incentives for private investments, or the

financing of single public initiatives such as infrastructure, training, or

communal and social services).

The idea is not new in local development promotion. It is based on both

theoretical and empirical findings showing that areas where development has

been spurred often possess dense social and economic interrelations. These

interrelations might appear as formal, informal, market-driven or

institutional. Linkages in production activities and related services also match

these relations among the relevant local actors. Local economies based on

coherent agglomerations of a variety of activities have often proved to be

associated with local prosperity, comparable to – or even more long-lasting

than – forms of territorial development coming from the presence of single

large firms and plants. This is, for instance, the case of the so-called Italian

industrial districts. Economists and social scientists have studied the latter

extensively. However, these studies concentrate on natural evolution and

equilibria, hence their findings do not necessarily support the idea that it is

possible for a policy maker to promote or accelerate local development by

devising incentives for a coalition to form or progress more speedily. The need

for evaluation comes in part from this last consideration. 3 The general

evaluation question in the background of this line of reasoning is whether, to

what extent and in which circumstances the policy maker can induce or

accelerate local development dynamics.

In order to identify the specific relevant evaluation questions concerning

territorial pacts, it is useful to consider the explicit objectives of the policy. The

policy has two specific ambitions seen as crucial mechanisms for inducing or

enhancing local development. The first is to support the start-up of a sound,

locally-rooted, integrated project made up of a set of different initiatives involving

responsibilities on the part of many public and private actors. If the project is

successful, it is expected to generate positive spillovers for the economy of the

area. The second ambition is to promote the formation of a robust local coalition,

a group of actors that might – through the setting and implementation of the

original project – learn how to interact with each other and promote further

development. This second objective is a key one, as permanent changes are

associated with the creation of a long-lasting local coalition.



358



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



These two objectives of territorial pacts call for two sets of evaluation

questions. The first set relates to the nature of the integrated project financed

(for example, under which circumstances has the policy mechanism proved

successful in inducing – and selecting for financial support – a good project? Is

the project well defined and rooted in a real knowledge of the potential of a

territory? Is it capable of triggering a local process of development? And how?

etc.). A somewhat different set of questions is related to the nature of the local

coalition that the pact is to promote (is it a good coalition? Is the process helping

in inducing or enhancing fiduciary relations among actors? Has their collective

capacity in problem solving been augmented? In which way? etc.).

The questions sketched above are related to the specific mechanisms of

development that the pact should directly activate (good projects and

institutional capacity). However most policy makers are mainly interested in

the final economic results for the territories where pacts are implemented. In

other words, there is a third and more explicit set of questions of the following

type: have the pacts promoted local development, firms’ growth, employment

opportunities? These latter questions are indeed important, but they can also

be misleading if not addressed properly. Two issues must in fact be

considered: the time span between implementation of the original project and

the desired spill-over effects; the circumstance that questions of this type are

indeed very general. In discussing evaluation of territorial pacts in Italy this

latter point is particularly important as pacts are implemented in territories

(especially the Italian Mezzogiorno) in which other policies are at work at the

same time and also rely on different mechanisms. The need to disentangle

effects coming from -or cumulating by – different kinds of interventions is of

primary importance in discussing evaluation methods. Hence we need to keep

this in mind when we look at territorial pacts in order to make clear what we

expect from evaluation exercises.



The cultural and institutional environment for carrying out

evaluation exercises: theory and practice

As in other contexts, we can evaluate both for learning (how the

instrument actually works; where it works well, where it doesn’t and why) and

accountability4 (to the fund givers). In the Italian case, however, most national

policies are not evaluated ex post (or an on-going basis) and systematic policy

monitoring has only recently begun. To some extent this is a good opportunity

for territorial pacts. In fact, it opens a window of opportunity for promoting a

learning approach, which is both less threatening for the policy maker and

particularly advisable when the policy in question does not have an evaluation

history.



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



359



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



In evaluating pact performance a crucial issue is timing. In other words,

given the characteristics of the instrument and its mechanisms, it is

important to be both patient and forward looking. First, for assessing final

results (both in terms of local development and institution building) it is

necessary to wait for the pacts to accumulate enough implementation history.

However, in order to be able to assess results it is necessary to collect a certain

amount of information while the pacts are under implementation and to

prepare the necessary statistical information at the desired level of territorial

detail. These arguments were perceived as very abstract only a couple of years

ago. However, both these concepts have now been fairly well understood by

many policy makers.

Despite the presence of a certain residual degree of impatience,5 the

administration has decided in April 2002 to launch a study involving fieldwork

on a group of pacts that have been in place for several years (signalling

awareness that only in this case can a study come up with some answers

about performance). Moreover, a new monitoring system has been set up

which should be able to provide precise information on the administrative

history of financial contributions to the pacts and offer other information

which in principle could be used in conjunction with other territorial data. To

implement a real evaluation, however, it is necessary that the policy maker

express an entire and detailed set of evaluation questions in which s/he is

genuinely interested. If there is an interest in learning from evaluation, these

questions should be expressed not only in terms of socio-economic results

occurring in the territories, but also in terms of the functioning of the

instrument itself (does the set-up procedure to gather and select the projects

work? what are the flaws? what are the good and the bad incentives?).6 In this

respect, more progress is needed. Even in the study recently launched, despite

an indisputable genuine interest about what happened in the different

contexts of the pacts, not very much time has been spent in detailing

questions challenging the role of the administration in designing and

implementing the instrument.

Another key issue is related to evaluation research methods.

Methodology is crucial for at least two reasons. First, as a territorial pact is a

package of different things (it is not a single well defined policy intervention,

but an entire set of different instruments pooled together), in order to learn

from evaluation we need to know about key ingredients (which aspects of the

policy are the most important/effective) and the methods used have to be

appropriate to this scope. Second, even more than in other policy

interventions, context matters. The evidence coming from administrative

monitoring of financial contributions to single initiatives included in the pacts

makes clear that they proceed very differently. There appear to be different

mechanisms at work. We want to know why some pacts seem to work better



360



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



that others, to what extent this is true and why this happens. The

methodology used has to deliver answers that help in uncovering reasons for

differentiated outcomes. In other words, even more than in other

circumstances, we need a method of investigation which is able to give

explanations. All these considerations call for a methodological approach

which considers primarily comparative analysis between pacts (coming from

accurate fieldwork to be read in conjunction with more macro statistical

evidence). In this respect, even if the study recently commissioned is not

strictly speaking an evaluation study, the methodology chosen (fieldwork and

direct interaction with relevant stakeholders in territories where the pacts are

active) appears to be adequate enough to give some revealing answers.

Another piece of good news for evaluation is the increasing institutional

attention devoted to building territorial statistics at a very detailed

administrative and economic level (“local systems” and council level). This

kind of information has important implications for evaluation of territorial

pacts – or other similar forms of local development promotion – as it allows

the assessment of changes in target areas using statistical concepts similar

and thus comparable to those available for larger territories (at regional, or

province level). Of course, as suggested before, the availability of more detailed

territorial statistics does not imply that evaluation of the effects of territorial

pacts can be limited to comparing areas where the instrument is active and

areas where the instrument is not active.7 However, standardized detailed

information allows for several kinds of informative analyses that can be used

in conjunction with other methods of investigation and that also offer a

possible guide for picking areas of interest (for instance, areas where target

case studies might be carried out on specific instruments (like territorial

pacts), or also other very informative kinds of evaluations, as for instance

area–based evaluations).8



What we have learned so far: academic studies, preliminary

evaluations, direct experience and learning by monitoring

Most of the studies carried out in the last few years on territorial pacts

have involved independent academic research, mostly fieldwork by

sociologists who focussed on the issue of social capital accumulation.9 The

nature of the coalitions, and the development of trust relations has been the

object of many of these studies. As these are mainly case studies undertaken

with different methods, they were not aiming to reach general conclusions.

However, they all seem to suggest that the process associated with the pact, in

the various cases examined, did trigger some social capital dynamic, even if in

a differentiated way. In the academic debate on these issues these are

important findings that speak loudly in favour of the possibility that pacts

may affect local development. Nevertheless these studies also seem to suggest



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



361



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



that if the pact offered an opportunity, it took some pre-existent local social

leader to glue the coalition together, an issue that should be studied more

extensively.

Sociologists also found different models at work in the coalitions’

formation, some of which are in fact quite perverse (they observe in some

instances the formation of opportunistic coalitions). As we do not have these

kinds of studies available for a sufficiently large number of pacts, we do not

know whether good coalitions outnumber opportunistic ones. However it is

clear that this variety of outcomes shows that the procedure used to select the

pacts had some limitations in discouraging opportunistic coalitions. It would

be useful to investigate whether the magnitude of the financial incentives and

the mechanism for selecting the projects played any role.

In the first part of 2000 a survey was carried out on all the pacts in

operation (46). This survey was based on a structured questionnaire for

entrepreneurs and key informants.10 The survey was part of an autonomous

study aimed at uncovering the motivation of actors who joined the pact

coalition, learning directly from involved entrepreneurs about the

opportunities and needs of the territories. The main result of the study was

that the policy was particularly well received in most of the cases

(entrepreneurs were on average quite happy), despite some complaints about

lengthy procedures.

What is most important, however, is that the study produced a database

with coded information on the agents’ perceptions and point of views on a set

of issues.11 In the same year, within the Department for development (the

administrative authority in charge at the time for implementing the pacts)

another project was carried out to build economic statistics defined at the pact

level, in other words referring to the territory included in the pact. That

exercise showed that pacts were starting in territories with different economic

conditions, though the very first cohort of pacts appeared in areas that were

relatively less disadvantaged.

The most thought provoking finding came, almost casually, from

monitoring evidence, as early as the end of 2000.12 Up to that point the general

debate on territorial pacts had considered pacts as homogeneous (pacts were

thought and portrayed as being everywhere the same, suffering from similar

weaknesses that were mostly due to bureaucratic inefficiencies). Monitoring

showed that pacts were behaving differently with respect to the progress in

the pact project (see Figure 14.2).

Some pacts, after controlling for their starting time, showed a

significantly higher rate of active initiatives and a significantly higher rate of

expenditure. The issue seemed worth further investigation.



362



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Figure 14.2. Evidence from monitoring the implementation of the projects

in the first 61 approved territorial pacts

(Centre North and Mezzogiorno; percentage of financial contribution used by each pact,

standardized data referring to April 2001)

80.0

70.0

60.0

50.0

40.0

30.0

20.0

10.0

0

Pacts in the Centre-North



Pacts in the Mezzogiorno



We then tried to investigate whether and to what extent the initial

conditions of the territories, as measured by a simple set of variables, had

affected the formation of the local partnership, its objectives and its

functioning.13 In particular we were interested in uncovering whether those

partnerships which seemed to perform better (as they appeared to progress

more quickly in the implementation of the pact’s project) could be predicted

by more favourable initial economic conditions in the territories or, instead,

whether they had acquired their skills in the process of building and

implementing the pact. The main idea was to look at differences in ongoing

performance in order to uncover general characteristics of the instrument at

work, asking explicit evaluation questions (where a pact works better, why? Is

this because of things that the policy cannot affect – like more favourable

initial conditions? Is this because of something that can be incorporated in a

better policy design?)

The study used statistical techniques exploiting a quite rich data set built

on the basis of administrative monitoring information about the pacts’

relative realization speed (assumed as a proxy for intermediate performance),

characteristics of the pacts’ areas and data coming from the 2000 survey.

The exercise was carried out over the first 61 pacts, with a specific focus

on those implemented through the national procedure, for which the

performance variable was measured more accurately. The results of the

analysis can be summarized as follows. Relative initial economic conditions

did not seem to affect the implementation of the pact (see Table 14.1).14



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



363



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Table 14.1. Pacts’ implementation and initial conditions of the territories

PERF



A



A1



Variables



Coefficient



(t)



Coefficient



(t)



Constant



3.301



(0.752)



2.012



(0.494)



DC1



0.547



(1.725)



0.312



(1.036)



DC4



0.455



(1.327)



0.289



(0.900)



–1.242



(–2.286)**



–0.652



(–1.221)



0.265



(0.745)



–0.241



(–0.662)



0.99



(0.517)



0.092



(0.523)



1.186



(3.242)**



LVALAG

LUNR

LSIZEFC

DS



F (5.55) = 6.04

Prob > F = 0.0002

Adj R2 = 0.2957

Num Obs 64



F (6.54) = 7.65

Prob > F = 0.0000

Adj R2 = 0.3995

Num Obs 61



PERFi = CONSTANT + b1 (DC1) + b2 (DC4) + b3 (LVALAGi) +b4 (LUNRi) + b5 (LSIZEFCi). PERF is a variable

proxying for the ongoing performance of the pact at April 2001 (see explanation in the text); DC1 is a

dummy for the first cohort of National territorial pacts which were approved much earlier; DC4 is a dummy

identifying territorial pacts approved with the European procedure; LVALAG is the log of per-capita value

added in the area of each Pact in 1996; LUNR is the log of the unemployment rate in the area of each pact

in year 1996 (chosen as the initial conditions date); LSIZECF is a control for the (log) dimension of the total

public contribution available for each pact. The same regression is then run adding a dummy DS for

location in the South. OLS estimates, bounded variable are treated with a logistic transformation.

Source: Casavola, P and Utili, F.(2002) Promozione di partnership locali per incoraggiare lo sviluppo

locale: una valutazione preliminare dei patti territoriali, Sviluppo Locale, Vol. IX, n. 20.



This turned out to be a quite robust result holding among the various

subgroups of pacts examined. If initial economic conditions do not count very

much in explaining differentiated performance, we might conclude that it is

possible to foster local development through incentives to form coalitions

even in areas that are very disadvantaged.

Instead, the performance of the pacts [as supported by an exercise

carried out on a smaller number of pacts for which additional information was

available (see Table 14.2)] seemed to be particularly related to the quality of

the process of setting up the pacts at the local level.15 Pacts in which relevant

actors had been involved in discussing the project from the very beginning

appeared to proceed faster in their implementation, a result confirmed by

other fieldwork.16 The result is interesting, as the way the local process is

organized is a variable that can be directly affected by the mode in which the

policy is delivered in a territory. It signals the importance of giving enough

time and resources to the initial phase when designing a policy for promoting

the formation of a local coalition. It also suggests that accurately screening the

process of coalition formation (in order to prevent weak projects) could be

crucial for their future success. This result is also particularly informative as it

can be generalized to a family of policy instruments based on a bottom-up

approach to development.



364



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



Table 14.2. Pacts’ implementation and the quality of the local setting-up

process

PERF



B1



C



Variables



Coefficient



(t)



Coefficient



Constant



(t)



–5.3035



(–1.37)



–5.058



DC1



0.228



(0.815)



0.206



(0.737)



LVALAG



0.479



–0.054



(0.896)



LSIZEFC



0.097



(0.483)



0.054



(0.272)



DS



1.236



(3.417)***



1.186



(3.294)***



(0.77)**



(–1.326)



LPARTIC



0.279



TP_TRUST



–0.99



(–1.014)



–0.076



(–0.674)



TP_INFR

F (4.31) = 6.02

Prob > F = 0.0011

Adj R2 = 0.364

Num Obs 36



(2.195)**



F (7.28) = 4.35

Prob > F = 0.0023

Adj R2 = 0.401

Num Obs 36



PERF= CONSTANT + b1 (DC1 ) + b2 (LVALAG) + b3 (LSIZEFC) + b5 (DS) + b6 (LPARTIC) + b7(TP_TRUST) + b8

(TP_INFR). In the equation the new variables are all derived from a survey carried out on a smaller number

of pacts, asking questions to entrepreneurs taking part in the pacts: LPARTIC is a variable derived from

positive answers to a question in which entrepreneurs were asked whether in the initial process of setting

up the pact all the key local actors had been involved; TP_TRUST is a variable derived from answers to a

question in which entrepreneurs were asked about the role of the pact in inducing, enhancing or

hampering trust relations among local actors. It represents the share of entrepreneurs that perceived a

positive role of the pact in this respect; TP_INFR is derived from another set of answers to questions related

to the most urgent necessities of the territory and it represents the share of entrepreneurs who signaled

the need for better infrastructures.



To learn more however it appears necessary to carry on more targeted

fieldwork. This is the task of the study implemented by a group of researchers

working for the department of Development Policies. In particular, the

researchers have been asked to come up with a more appropriate definition of

success and investigate more deeply what makes the instrument work

better.17



Notes

1. Resources for technical assistance are provided for preparation of the projects and

during implementation to support the coordination process.

2. There are other relevant differences among the two groups of pacts, the main ones

being the greater role in European pacts played by the local agency responsible for

coordinating implementation (a feature that in time should be extended to the

national pacts as well) and the presence of a much larger variety of initiatives

(whereas in the national pacts the public contribution is given mainly to finance

entrepreneurial activities and to a much lesser degree public infrastructure).

3. The industrial districts literature has stressed the importance of context and local

relations as a basis for different paths to development and this argument

apparently speaks in favor of policy instruments like territorial pacts. However,



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



365



14. METHODOLOGICAL AND PRACTICAL ISSUES FOR THE EVALUATON OF TERRITORIAL PACTS



the literature on social capital inspired by other work (following Putnam’s famous

contributions on the origin of Italian regional development) has often reached the

conclusion that it takes a long historical path to build social capital. Some

researchers have hence inferred that there is no room for the policy maker in

trying to speed up this process.

4. I am using the term accountability in a somewhat improper and restrictive sense

for the sake of the argument and just to keep in mind that some of the questions

usually raised in relation to territorial pacts are of the following sort: are we

spending all the money well? Overall, is the instrument providing the promised

increase in development?

5. In the recent political debate some commentators, understandably concerned

with the large number of existing pacts, have asked for a quick decision on which

(of the many in place) are the good pacts – to keep supporting – and which are the

bad pacts – to possibly close down.

6. In other words an important set of questions relates to the relation between the

policy design and the policy outcomes. In fact the general characteristics of the

policy (financing a coalition with a project) do not suffice to completely describe

the nature of the intervention. A series of other design features should be

considered: what kind of projects and with which characteristics; which actors are

eligible to present the project; how large is the public contribution offered; how

and when is the project submitted; which procedures should be followed to select

projects to finance. All these elements might play a role in determining at least in

part the future success of the coalition selected.

7. There are numerous theoretical reasons to discourage a mechanical application of

the so called standard evaluation paradigm (which compares a treated group to an

untreated group) to instruments of local development like pacts.

8. Area-based evaluations could be a promising method for evaluating local

development policies as investigators are required to start their work looking at a

limited portion of the territory and from that angle reconstruct the possible causal

chains linking observed facts and behaviors to policy interventions in that area.

9. See for instance the 2001/3 issue of the Italian journal Stato e Mercato, almost

entirely dedicated to territorial pacts.

10. Sviluppo Italia – Iter (2000) “Caratteristiche e potenzialità dei Patti territoriali”,

Roma.

11. Most of the sociological studies quoted above had not produced coded or

standardized information (to be used in other studies).

12. Ministero del Tesoro, Bilancio e Programmazione Economica, “Terzo Rapporto

sullo sviluppo Territoriale”, November 2000.

13. The results reported briefly here come from Casavola, P – Utili, F. Promozione di

partnership locali per incoraggiare lo sviluppo locale: una valutazione preliminare dei patti

territoriali, Sviluppo Locale, Vol. IX, No. 20, 2002.

14. We looked at the relation between a dependent variable describing the ongoing

performance of the pact at some point in time (the amount of public financial

contribution actually used at April 2001 over the total amount available at the

same date, on the hypothesis that all the initiatives were progressing at the fastest

rate) and indicators proxying the initial economic condition of the area of the pact

(per capita value added and unemployment rate computed for a period before the

activation of the pact ). A few other controls were also added.



366



EVALUATING LOCAL ECONOMIC AND EMPLOYMENT DEVELOPMENT – ISBN 92-64-01708-9 – © OECD 2004



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Chapter 14. Methodological and Practical Issues for the Evaluation of Territorial Pacts The Experience of Italy

Tải bản đầy đủ ngay(0 tr)

×