Printer Friendly

Are subsidies really needed? The case of EU regional policy in the Czech and Slovak Republics.


The idea of public support programmes to strengthen local economic performance is widely recognized and used by many national or local governments. The basic rationale for support programmes could be found in several economic theories. The support arguments for subsidies come from existence of externalities and market failures see for example [31], for regional conditions in the Slovak Republic see for example [24], [25]. The private companies do not want to invest into activities with positive externalities (since they are not able to absorb all the benefits), but these investments are important for society itself, so the governments must subsidy such an investment to be interesting for private enterprises [23]. Another barrier arises due to risk adverse views of financial institutions. Some firms can deny access to credit despite the fact that they have reasonable projects, but without proper financial coverage.

However, subsidies must also fulfil certain criteria to be valuable for economic growth and prosperity. It is very difficult to correctly set up and evaluate public support interventions. The evaluation of public support should identify whether, and to what extent, the state or regional support aid schemes have been able to reduce the targeted market failure. This approach involves quantifying both the market failure and the influence of the public support scheme on the market failure. However, it is quite impossible to quantify market failure, let alone the impact of a policy measure on the extent of that market failure [22]. We could find that many jobs have been created in supported companies or how high growth they achieved but key questions are different--how many jobs would be created without this support or with other use of this support [20]. This leads us to the question of effectiveness and real need of public support policies.

Support subsidies could have various forms [3], but we will concentrate on public support programmes. These programmes are the main form of regional policy instruments in the recent years. Some main features of these programmes are competitiveness in project selection and evaluation, selection of several different projects and usually longer implementation period with several calls. They are also not negotiated subsidies (e.g. subsidies for foreign investors).

Another important aspect of this kind of support is related especially to regional policy. The main goal of such programmes is to support development of lagging part of the country of group of countries. This is also the case of regional policy of the European Union. This policy should be no longer viewed as redistributive policy, but rather the development policy [9] with ability to define investment priorities, which are important for region, but companies do not want to invest into them alone. Otherwise, the whole regional policy could be only in the form of unconditional cash transfers [4]. As summarized, the redistributive function is not sufficient for achievement of efficiency, effectiveness and social inclusion in the regional policy of the European Union. To achieve effectiveness of subsidies we must reassure that public sources are spent on activities otherwise not realized by private sector. This ineffectiveness is called "deadweight" effect and we will analyze this effect in the case of the Czech and Slovak Republics. In these countries, no detailed analysis on the deadweight effect of subsidies of private companies has so far been conducted.

The main purpose of this article is to measure the deadweight effect and analyze the internal and external conditions under which subsidised projects would be abandoned without public assistance in case of the Czech and Slovak Republics. The article is organized as follows. In the first part we look more closely at the definition of deadweight effect and previous studies analyzing it. In the methodology part we discuss different approaches how to measure this effect and we introduce methodology for our research. We also discuss potential problems with interpretation of the results. In the final part we present the results of our research and some policy implications.

1. Deadweight Effect

Deadweight effect is one of the forms of ineffectiveness in additionality principle. The additionality is one of the key principles of regional policy of the European Union. Applying this principle the public support, as market failure correction should be used only if really needed and adds value to regional development in supported region. Tab. 1 summarizes all types of potential ineffectiveness within this principle.

Deadweight could be defined as outcomes of the projects which would have occurred without public intervention. In our research deadweight is understood as proportion of projects planned to be realized also without obtaining support from the EU funds. One of the main goals of intervention should minimize this effect by focusing the finance on projects that would not be implemented without public support. This effect is usually the biggest one from the above mentioned ineffectiveness and support policy designers must pay adequate attention to diminish it.

In reality, it is impossible to fully eliminate deadweight effect. One of main issues here is asymmetry of information [27]. The government or other public agencies providing grants will never have enough information to be sure the project will not be realized without their support. The enterprises will always have a tendency to use public support for their own benefits. They will try to prepare their projects in the way to persuade evaluators on necessity to support them.

The opposite problem is "good choice paradox" [33]. The government tries to select the best projects for support not the projects that really need a support. There are two basic reasons for this. Firstly, executive agencies try to show the best direct results from their support programmes. The more direct jobs from programmes are created or more exports are achieved, the better for politicians responsible directly for these programmes and for sustainability of responsible agencies. The more detailed consequences of programmes are evaluated very rarely, especially in the Central and Eastern European countries. But it is very probable that these projects are projects with the highest deadweight effect [35]. This should be eliminated by adjusting evaluation criteria to be more sensitive to real needs of the region and it selects to support enterprises to bring new development into the region. Most of aid schemes really contain only very limited criteria for the access to real need of the support for the region.

Empirical studies monitoring this effect are rather limited. Most of the studies were done in Western European countries; the studies from the rest of the world are very scarce. Most of the recent studies estimate the effect between 30-40%. The summary of most relevant studies is in the table 2 below. The lowest deadweight effect of 20% was identired by [41] in Great Britain. On the other hand the effect could be found in the study of [16] with more than 70% of deadweight. The total level of this effect is very high, indicating more than one third of public resources spent on activities that also happened without this support.

Differences could be found not only in total deadweight estimations in the studies, but also among different kinds of the support of enterprises. BIS [3] identired 45% deadweight in direct support programmes to enterprises compared to 33.9% in investment into physical infrastructure. Internatiolization of enterprises was the most effective (only 26% deadweight) among direct support of enterprises programmes. The similar results could be found in study [7]. The deadweight in direct support programmes was 36%, compared to 15% in education programmes. Despite the deadweight effect, the type of supported activities could primarily depend on goal of particular regional policy. Some authors [26] showed that the investment subsidies tend to decrease employment. On the other hand employment subsidies stimulate employment. If the main goal for regional policy is oriented on unemployment, there should be support of the employment grants in the first place.

Studies also try to identify reasons for deadweight. Found deadweight effect to be statistically significantly dependent on region development position and the length of firm existence [35]. The EU evaluation also suggests that the deadweight tends to be higher in richer regions.

Another important variable is the size of enterprises. In Hungary there was 90% deadweight in the group of large enterprises compared to 50% in SME [11]. Large enterprises usually do not need a public support and this support only represents additional benefits, but it is not decisive for the project [18]. But for SME the support in research activities is very important for their final decision to implement the project. The size of enterprise as well as grant type and number of earlier grants have a significant impact on the likelihood of a firm to report deadweight [17]. This is mainly explained by the fact that firm's access to finance is likely to increase with business size [40].

The authors [16] analyzed the ownership of company on deadweight. They found foreign companies achieved lower level of deadweight (7% less), but statistically not significant. However, we must mention that studies did not consist of the variables that could cause deadweight effect. Some studies confirm the importance of the age of firm [35] received but some do not [17] and we could find several other examples in the mentioned studies from the Tab. 2.

As previous studies show the deadweight effect is quite substantial and creates pressure on the public sector to be more effective in the management of their support programmes. Especially this is the case in present economic crisis with strong negative influence on public revenues and the need to cut public debts.

2. Methodology

There are several methods how to measure deadweight effects. The main problem with all of them is counterfactual situation. It means we try to find out what happens in the case of non--intervention and its always hypothetical question [1]. We are not able directly to compare these two situations to analyze "the real difference" of applied public support.

Most of the studies were based on interviews or surveys with enterprises obtaining public support (e.g. [16], [35], [36]). The degree of the deadweight effect of the project is estimated by posting a hypothetical question of what would happen if the project were not subsidised. For example [35] used five answers to this question--the project will be abandoned, the project will be implemented on a reduced scale, the project will be implemented on a reduced qualitative level, the project will be implemented at a later date and the project will be implemented unchanged. Then they add to each question a deadweight from 0 (zero deadweight) to 1 (full deadweight). We changed this methodology and mainly due to time aspect. During preliminary research we found out that there were very long time delays in the evaluation process for subsidies, so most of enterprises reported no time savings, but time delays in the project implementation due to public support. So we asked only the question regarding the financial aspect of the project support: "If you did not get public support, how big part of the project budget do you estimate to spend on planned activities?" We asked this at the end of their approved project implementation process. We complement this survey with several interviews with supported companies to be able to understand better the reasons why they try to use public support programmes. The summary of statistical evaluation of this question is in the table 3, later in the text.

Another possibility how to evaluate deadweight is to ask unsupported projects which fulfil all the criteria how to obtain support, but there were not enough funds in that particular call to support them. The advantage of this method will be the ability to see the reality not hypothetical question about what they implemented from the project. Unfortunately, information about these projects is no longer available from the government agencies granting the subsidies. One pilot study with these companies shows lower deadweight effect than in the case surveys of approved projects [32]. This could be expected result taking into account the mentioned "good choice" paradox.

In some cases control group of unassisted enterprises is created [37] trying to use difference in different techniques [21] to compare their development before and after support within both groups. It is extremely difficult to properly construct this group, mainly because mutual influence between these groups. For example if you select to support bakery and you choose another bakery to control groups you are not able to answer the question, if the support itself (giving an advantage to one of them) or other external characteristics changed the development path of these two enterprises. Another problem with control groups could be the impossibility to construct it, because support enterprises could have unique structure or be somehow else different from the rest of enterprises (selection bias is in the selection process itself). This is specially the case of innovation and research support schemes, where very often best companies with previous research activities are supported and it is not possible to find adequate control group. The methodology of control groups was used e.g. by [28] or [39].

We conducted our survey on several calls for projects (KaHR-111DM-0901, KaHR 111DM-0801, SIA 2009 121 01, Rozvoj I and EDUCA) from 5 different operational programmes. We made a survey and obtained results from 414 enterprises (291 from the Czech Republic, 123 from the Slovak Republic). This sample was representative from the point of selected programmes and countries. The survey was conducted during the year 2012. Only privately owned companies were included in the survey. The public companies as e.g. hospitals were excluded, because even if there is deadweight in these companies, the outputs will serve the public interests, so we cannot consider this directly as ineffectiveness of public spending. Answers to the question: "If you did not get public support, how big part of the project budget do you estimate to spend on planned activities?", could be on the scale from 0% (it means project would not be realized at all without public support) to 100% (it means project would be fully realized even without public support). The summary of the answers for research question is in the table 3.

We statistically evaluated the surveys and estimated the deadweight effect in different categories. After this, we used logit regression model to identify key characteristics that influence the deadweight effect. Before constructing the model we used stepwise selection to identify only statistically significant characteristics. These methods are appropriate to such evaluation [34], [29] and were used also in several previous studies [16]. Sometimes there were used probit instead of logit models, but these models show very similar results [17].

The selection of factors possibly influencing the deadweight was also inspired by several previous studies. We must leave away two commonly used characteristics--number of employees and turnover of enterprises, because some evaluated programmes were open only for SME and some also for bigger companies, so the results for these two characteristics could be biased by this. We divided model variables into three main groups, which could influence the deadweight characteristics of enterprise, project and region (external environment). The Tab. 4 summarizes all used variables in the model. This was the first study conducted in two different countries and tries to identify if this influences the results of a model compared to previous studies.

3. Results

The results of our model could be found in the following tables. The Tab. 5 shows the results of stepwise selection. We found out only two variables that were statistically important--the amount of support granted and type of projects (investments). All other variables were not statistically significant on more than 5% level. Tab. 5 shows results of statistical significance of the constructed model.

This shows one very interesting fact. The variables related to project itself have much higher influence on deadweight level than other variables related to a region or an enterprise. The type of grant and amount of support received shows as the most important to determine deadweight effects. The amount of support received is not a surprised result. The higher the grant support was, the less likely the enterprise would realize the activities without this support. The following interviews clearly show that there were project proposals of an enterprise behind its present scope of activities. They clearly try to get the support to help them to reach "higher" level of their entrepreneurship, e.g. to buy a new machine equipment, otherwise too expensive for them, that allows them to enter new market segment.

The type of investment shows different results than in previous studies in the Western European countries. Based on interviews with some applicants we think the reason undervalues the investments in educational activities. Basically typical applicants in the investment support schemes are companies which try to replace presently used equipment with new ones, mostly as a part of their "normal" renewal of technological equipment. These companies will usually realize their replacement without public support. On the other hand typical applicants for educational grants are companies that carry out the education of workforce, understand how it is important, but have limited budget for it. They try to use an opportunity to cover their costs for the kind of education they have never had their own funds for. So without the support they usually dramatically reduce the amount spent on educational activities (or come back to previous level of their expenditures on education of workforce). Then the amount of funds required for grant by an applicant in investments subsidies is usually not higher than their two or three years annual budget for such activities. Compared to education grants where the amount required is in many cases higher than five years' company budget on this priority.

The Tab. 7 shows the odds ratio estimates for statistically significant variables. As you can see there is more than four times higher probability of deadweight in case of investment type of grant. The probability of full deadweight in the project with the budget up to EUR 116,000 is eight times lower than the probability of project with the budget higher than EUR 240,000.

Except the model, we statistically evaluate the achieved level of deadweight in the different categories. Some of the results are summarized in the table 8. The total deadweight effect was 36.8%, so more than one third of public support was used on projects that were realized also without this support. This is comparable number to other studies from Western Europe and there is no significant difference.

In spite of the characteristics of regions that were not statistically significant, they are related to the type of projects. It shows that more educational projects are from more developed regions and bigger cities. The effect of external environment could also be diminished because two countries with different development conditions were evaluated.

Very interesting was the deadweight effect from newly established enterprises. It was slightly higher than average deadweight. Compared to established enterprises, the levels of deadweight effects were very similar in both investments and educational types of project. It seems these companies need to invest into labour force at the beginning of their existence with or without public support, so the deadweight effect is higher in these projects than in established enterprises. On the other hand, the bigger investments into technology, the less affordable are the established enterprises. Therefore without public support they are less likely to realize the proposed projects. We also found out lower (27.85%), but statistically not significant deadweight effect in foreign enterprises (see Tab. 8).


We found out that more than a third of public support in the EU regional policy programmes was spent on the projects which will be realized even without this support. The main variables influencing the deadweight effect were the amounts of grant received and the type of grant. Investment grants have much higher deadweight than educational or employment grants. The more financial funds enterprises obtained through the grant, the more deadweight effect occurred. All other variables were not statistically significant in our model. Applying this to regional policy, the higher support for educational projects leads to the lower deadweight effect. The support of technology transfers looks more appropriate for newly established companies.

The deadweight effect could substantially limit the effectiveness and efficiency of programmes. However, the deadweight is only one possible ineffective use of public support. In order to correctly evaluate impact and effectiveness of programmes, we need to make a complex evaluation based on several micro studies of different effects. There also could be other negative effects. Even if there is 'zero deadweight', and all the firms indicate that in the absence of grant assistance they would not have been able to realize projects, there is the threat that the assistance given to one firm could displace jobs elsewhere in the region [17] and thus the whole region effect of support will be negative. On the other hand, even in the existence of deadweight spending, public subsidies may have a variety of other positive impacts on regional development, then also other implications that investment subsidies might have for economic activity, employment, growth, cooperation or networks must be evaluated.

In addition, comparison of ex ante analysis with ex post information could provide valuable information on appraising the deadweight effect of a project [17]. It could help the identification of inefficiencies so that the support scheme could be improved at least in some aspects.

The change of criteria in evaluation process to avoid the deadweight could be recommended for policy actions. The present criteria are much more oriented to select "the best performance" projects. Such projects usually are the best candidates for the deadweight effect. More relevant indicators and criteria should be applied to analyse the need of the project for enterprise and especially for region where support is provided. Another possible policy implication could be more oriented on newly established enterprises in the transfer technology projects. Stronger orientation also can be recommended on "soft" educational projects. The total level of deadweight (36.8%) also raises the question of effectiveness of direct support of a private enterprise as such. One third of the funds is spent ineffectively. If we add problems with possible distortion of the competitive environment in this support, more orientation of public support to more general activities supported business environment could be recommended.

The article was supported from the VEGA project 1/0093/12 Effectiveness of EU regional policy in Slovak Republic".


[1] BASLE, M. Strengths and Weaknesses of European Union Policy Evaluation Methods: Ex-Post Evaluation of Objective 2: 1994-99. Regional Studies. 2006, Vol. 40, Iss. 2, pp. 225-236. ISSN 0034-3404.

[2] BIS. Research to improve the assessment of additionality [online]. BIS OCCASIONAL

PAPER NO. 1. 2009. [cit. 2012-12-11]. Available from: < economics-and-statistics/docs/09-1302bis-occasional-paper-01>. [3] BUIGUES, P.A., SEKKAT, K. Public Subsidies to Business: An International Comparison. Journal of Industries, Competition and Trade. 2011, Vol. 11, Iss. 1, pp. 1-24. ISSN 1566-1679.

[4] CSIL. Impact of Additionality on the Real Economy of the EU MemberStates [online]. Milan: Csil--Centre for Industrial Studies, 2010. [cit. 2012-12-11]. 72 p. (PDF). Available from: docgener/studies/pdf/2010_additionality.pdf>.

[5] DAVENPORT, S., GRIMES, C., DAVIES, J. Research collaboration and behavioural additionality: A New Zealand case study. Technology Analysis and Strategic Management. 1998, Vol. 10, Iss. 1, pp. 55-68. ISSN 0953-7325.

[6] DE KONING, J. Measuring the placement effect of two wage-subsidy schemes for the long-term unemployed. Empirical Economics. 1993, Vol. 18, Iss. 3, pp. 447-468. ISSN 0377-7332.

[7] DETR STUDY. Final Evaluation of City Challenge. 2000. London.

[8] EDERVEEN, S., GROOT, H., NAHUIS, R. Fertile Soil for Structural Funds? A Panel Data Analysis of the Conditional Effectiveness of European Cohesion Policy. Kyklos. 2002, Vol. 59, No. 1, pp. 17-42. ISSN 1467-6435. [9] ERPC. Comparative study on the visions and options for cohesionpolicyafter 2013. 2011. Brussels.

[10] ESPOSTI, R., BUSSOLETTI, S. Impact of Objective 1 Funds on Regional Growth Convergence in the European Union: A Panel-data Approach. Regional Studies. 2008, Vol. 42, Iss. 2, pp. 159-173. ISSN 0034-3404.

[11] EUROPEAN COMMISSION. Ex post evaluation of Cohesion Policy Programme 2000-06 cofinanced by the ERDF. Publications Office of the European Union. Luxembourg, 2010.

[12] EUROPEAN COMMISSION. Annual evaluation review 2003: overview of the Commission's evaluation activities and main evaluation findings. Brussels, 2004. 314 s. SEC (2004) 662. Directorate-General for the Budget.

[13] FOLEY, P. Local economic policy and job creation: a review of evaluation studies. Urban Studies. 1992, Vol. 29, No. 3/4, pp. 557-598. ISSN 0042-0980.

[14] GEFRA. Ex post evaluation of Cohesion Policy programmes 2000-2006 financed by the European Regional Development Fund. Munster, 2010.

[15] IEU. Evaluation of Micro Enterprise Supports Across National and Local Development Agencies, Industrial Evaluation Unit. Dublin, 1999.

[16] LENIHAN, H., HART, M. The use of counterfactual scenarios as a means to assess policy deadweight: an Irish case study. Environment and Planning C: Government and Policy. 2006, Vol. 22, Iss. 6, pp. 817-839. ISSN 0263-774X.

[17] LENIHAN, H. Evaluating Irish industrial policy in terms of deadweight and displacement: a quantitative methodological approach. Applied Economics. 2004, Vol. 36, Iss. 3, pp. 229-252. ISSN 0003-6846.

[18] LUUKKONEN, T. Additionality of EU framework programs. Research Policy. 2000, Vol. 29, Iss. 6, pp. 711-724. ISSN 0048-7333.

[19] MARTIN, R., TYLER, P. Evaluating the Impact of the Structural Funds on Objective 1 Regions: An Exploratory Discussion. Regional Studies. 2006, Vol. 40, Iss. 2, pp. 201-210. ISSN 0034-3404.

[20] MOLLE, W. European Cohesion Policy. Oxon: Routledge, 2007. ISBN 0-203-94527-1.

[21] MORTON, M. Applicability of Impact Evaluation to Cohesion Policy, working paper report for European Union [online]. University of Oxford, 2009 [cit. 2012-12-21]. 30 p. (PDF). Available from: < licy/future/pdf/4_ morton_final-formatted.pdf>.

[22] MOSSELNAM, M., PRINCE, Y. Review of methods to measure the effectiveness of state aid to SMEs: Final Report to the European Commission [online]. Zoetermeer: EIM, 2004 [cit. 2012-12-11]. 116 p. (PDF). Available from: <>.

[23] NELSON, R.R. The simple economics of basic scientific research. Journal of Political Economy. 1959, Vol. 67, No. 3, pp. 148-163. ISSN 0022-3808.

[24] NEMEC, J., MERICKOVA, B., STRANGFELDOVA, J. The Ownership Form of Hospitals from the Viewpoints of Economic Theory and Slovak Practice. E+M Ekonomie a Management. 2010, Vol. 13, Iss. 2, pp. 19-31. ISSN 1212-3609.

[25] NEMEC, J., OCHRANA, F., SUMPIKOVA, M. Czech and Slovak lessons for public administration performance evaluation, management and finance. Ekonomicky casopis. 2008, Vol. 56, No. 4, pp. 353-369. ISSN 0013-3035.

[26] PETRUCCI, A., PHELPS, E. Capital Subsidies versus Labor Subsidies: A Trade-off between Capital and Employment? Journal of Money, Credit, and Banking. 2005, Vol. 37, No. 5, pp. 907-922. ISSN 1538-4616.

[27] PICARD, P.M. Job additionality and deadweight spending in perfectly competitive industries: the case for optimal employment subsidies. Journal of Public Economics. 2001, Vol. 79, Iss. 3, pp. 521-541. ISSN 0047-2727.

[28] ROPER, S., HEWITT-DUNAS, N. Grant assistance and small firm development in Northern Ireland and the Republic of Ireland. Scottish Journal of Political Economy. 2001, Vol. 48, Iss. 1, pp. 99-117. ISSN 1467-9485.

[29] RUBLIKOVA, E., LABUDOVA, V., SANDT NEROVA, S. Analyza kategorialnych udajov. Bratislava: Vydavatelstvo EKONOM, 2009. 172 s. ISBN 978-80-225-2710-1.

[30] SHEEHAN, M. Government financial assistance and manufacturing investment in Northern Ireland. Regional Studies. 1993, Vol. 27, Iss. 6, pp. 527-540. ISSN 0034-3404.

[31] STIGLITZ, J.E. Economics of the public sector. 3rd ed. London: Norton, 2000. 848 p. ISBN 978-0393966510.

[32] SIPIKAL, M. Deadweight effect. Region Direct. 2010, Vol. 3, Iss. 2, pp. 71-79. ISSN 1337-8473.

[33] SIPIKAL, M. Strategia projektovo orientovanej podpory regionalneho rozvoja Slovenskej republiky (kriticka analyza). Ekonomicky casopis. 2011, Vol. 59, No. 9, pp. 954-968. ISSN 0013-3035.

[34] TEREK, M., HORNIKOVA, A., LABU DOVA, V. HIbkova analyza udajov. Bratislava: Iura Edition, 2010. ISBN 978-80-8078-336-5.

[35] TOKILA, A., HAAPANEN, M., RITSILA, J. Evaluation of investment subsidies--When is deadweight effect zero? International Review Ekonomie of Applied Economics. 2008, Vol. 22, No. 5, pp. 585-600. ISSN 0269-2171.

[36] TOKILA, A., HAAPANEN, M. Evaluation of Deadweight Spending in Regional Enterprise Financing. Regional Studies. 2010, Vol. 40, Iss. 1, pp.1-17. ISSN 0034-3404.

[37] UNTIEDT, G. Impact analysis and counterfactuals in practice: The case of Structural Funds support for enterprise [online]. Munster: Gesellschaft fur Finanz--und Regionalanalysen, 2009 [cit. 2012-12-11]. 17 p. (PDF). Available from: < conferences/evaluation2009/abstracts/ untiedt.pdf>.

[38] URAMOVA, M., KOZIAK, R. Regionalne disparity na Slovensku z aspektu priemernej nominalnej mzdy. E+M Ekonomie a Managemenf. 2008, Vol. 11, Iss. 2, pp. 6-17. ISSN 1212-3609.

[39] WREN, C., STOREY, D. Evaluating the effects of soft business support upon small firm performance. Oxford Economic Articles. 2002, Vol. 54, Iss. 2, pp. 334-365. ISSN 1464-3812.

[40] WREN, C. Subsidies for job creation: is small best? Small Business Economics. 1998, Vol. 10, No. 3, pp. 273-281. ISSN 0921-898X.

[41] WREN, C. Regional Grants: Are they Worth it? Fiscal Studies. 2005, Vol. 26, No. 2, pp. 245-275. ISSN 0143-5671.

Mgr. Miroslav Sipikal, PhD.

University of Economics in Bratislava Faculty of National Economy Department of Public Administration and Regional Development

doc. Ing. Peter Pisar, PhD.

Matej Bel University in Banska Bystrica Faculty of Economics Department of Finance and Accounting

RNDr. Viera Labudova, PhD.

University of Economics in Bratislava Faculty of Economic Informatics Department of Statistics

Doruceno redakci: 4. 4. 2013

Recenzovano: 2. 5. 2013, 21. 6. 2013

Schvaleno k publikovani: 27. 9. 2013
Tab. 1: Type of Ineffectiveness in Applying Additionality Principle

Type of effect   Description

Deadweight       The proportion of total outputs/outcomes that would
                 have been secured anyway without the public support

Displacement     The proportion of outputs/outcomes that are reduced
                 elsewhere in the target area

Substitution     This effect arises from firm substitutes of their own
                 expenses by public support e.g. a firm substitutes a
                 jobless person to replace an existing worker to take
                 advantage of the public sector assistance

Leakage          The proportion of outputs/outcomes that benefits
                 those outside the target area of the intervention

Source: BIS (2009), adjusted by authors

Tab. 2: Studies Measuring the Deadweight Effect

Study                                   Country

Lenihan and Hart (2006)                 Ireland
Sheehan (1993)                          Northern Ireland
De Koning (1993)                        The Netherlands
EC (2010)                               Italy
IEU (1999) Micro Enterprise Supports    Ireland
Lenihan and Hart (2004)                 Ireland
Davenport et al. (1998)                 New Zealand
Stierwald and Wiemers (in GEFRA 2010)   Germany
Wren (2005)                             Great Britain
BIS (2009)                              Great Britain
Tokila (2010)                           Finland

Study                                   Deadweight effect

Lenihan and Hart (2006)                      73.2%
Sheehan (1993)                                59%
De Koning (1993)                              40%
EC (2010)                                     50%
IEU (1999) Micro Enterprise Supports          45%
Lenihan and Hart (2004)                   42.6-55.8%.
Davenport et al. (1998)                      37.5%
Stierwald and Wiemers (in GEFRA 2010)        28-35%
Wren (2005)                                   20%
BIS (2009)                                    43%
Tokila (2010)                                35.9%

Source: mentioned studies, summarized by authors

Tab. 3: Statistical Distribution of Answers to the Research Question

Deadweight level           Number of answers   Number of answers
                           (Czech Republic)       (Slovakia)

0%                               94                  48
1-10%                            18                   6
11-20%                           25                   5
21-30%                           25                  14
31-40%                            4                   6
41-50%                           39                  18
51-60%                            8                   1
61-70%                           13                   3
71-80%                            9                   1
81-90%                            1                   3
91-99%                            0                   0
100%                             55                  18
Total average deadweight        36.8%

Source: own research

Tab. 4: Independent Variables Used in the Logistic Regression

Characteristics of enterprise

New firm                          1 if company was less than 4 years
                                  old, 0 otherwise

Foreign Ownership                 1 if owned by a foreign owner, 0

Personal Ownership                1 owned by one person, 0 otherwise

Characteristics of project

Amount granted                    Total amount of grants, divided into
                                  5 categories (code--BUDGET_CAT in
                                  the following tables of results)

Investment project                1 if investment project (or "hard"
                                  project--project aimed at buying new
                                  technology), 0 otherwise (or "soft"
                                  project--aimed at the education of
                                  workforce), (code--TYP_INV in
                                  following tables of results)

Characteristics of region

% of unemployed in the district   Level of unemployment in the
                                  district of residence of applicants

Region                            NUTS III regions according to the
                                  residence of applicants

City size                         City size of subsidies company
                                  headquaters (4 categories)

Country                           1 if from the Czech Republic, 0
                                  otherwise (Slovak Republic)

Source: Authors

Tab. 5: Stepwise Selection Results

Summary of Stepwise Selection

Step            Effect

         Entered     Removed   DF   Number     Score
                                      In     Chi-Square

1        TYP_INV               1      1       59.6942
2      BUDGET_ CAT             3      2       18.8755

Step      Wald       Pr>ChiSq    Variable
       Chi- Square                 Level

1                     <.0001      TYP_INV
2                     0.0003    BUDGET_ CAT

Source: output of SAS Enterprise Guide 4, own research

Tab. 6: Parameters and Statistical Significance of the Model

Analysis of Maximum Likelihood Estimates

Parameter                              DF   Estimate

Intercept                              1    -3.0099
TYP_INV                0.0             1    -0.6952
BUDGET_CAT      01:low -116813.2       1     2.1187
BUDGET_CAT   02:116813.2-240230.135    1     2.3658
BUDGET_CAT   03:240230.135-764149.14   1     2.0923

Parameter    Standard      Wald      Pr>ChiSq
              Error     Chi-Square

Intercept     0.5940     25.6789      <.0001
TYP_INV       0.1378     25.4417      <.0001
BUDGET_CAT    0.6395     10.9782      0.0009
BUDGET_CAT    0.6391     13.7044      0.0002
BUDGET_CAT    0.6474     10.4456      0.0012

Testing Global Null Hypothesis: BETA=0

Test             Chi-Square  DF   Pr>ChiSq

Probable Ratio    83.8691    4     <.0001
Score             71.4221    4     <.0001
Wald              49.5945    4     <.0001

Source: output of SAS Enterprise Guide 4, own research

Tab. 7: Odds Ratio Estimates

Odds Ratio Estimates

Effect                        Point Estimate        95% Wald
                                               Confidence Limits

TYP_INV 0.0 vs 1.0                0.249         0.145   0.427

BUDGET_CAT 01:low -116813.2       8.321         2.376   29.138
vs 04:764149.14-high

BUDGET_CAT                        10.653        3.044   37.279
02:116813.2-240230.135 vs

BUDGET_CAT                        8.104         2.278   28.822
03:240230.135-764149.14 vs

Source: output of SAS Enterprise Guide 4, own research

Tab. 8: Deadweight Effect for Certain Specific Groups of Enterprises

                                                  Deadweight effect %

All enterprises                                          36.8%

All enterprises with less than 4 year existence          38.8%

Enterprises with less than 4 year existence              39.6%
(investments projects)

Enterprises with less than 4 year existence              35.4%
(education projects)

Foreign enterprises                                     27.85%

Enterprises from the Slovak Republic                    33.15%

Enterprises from the Czech Republic                     38.46%

Educational projects in the cities with under           22.05%
5,000 inhabitants

Investments projects in the cities with under           62.19%
5,000 inhabitants

Investments projects in the cities with over            62.17%
50,000 inhabitants

Educational projects in the cities with over            19.27%
50,000 inhabitants

In regions with unemployment over 17%                    33%

In regions with unemployment under 7%                  39.78%

Enterprises with investment projects                     53.4%

Enterprises with educational projects                    22.4%
Source: own research
COPYRIGHT 2013 Technical University of Liberec
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Economics/Ekonomie
Author:Sipikal, Miroslav; Pisar, Peter; Labudova, Viera
Publication:E+M Ekonomie a Management
Geographic Code:4EXSV
Date:Oct 1, 2013
Previous Article:Creative industries in the Czech Republic: a spatial perspective.
Next Article:Modeling of wage distribution in recent years in the Czech Republic using L-moments and the prediction of wage distribution by industry/Modelovani...

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters