DEM_2015_89to109 © 2015 Nicolaus Copernicus University. All rights reserved. http://www.dem.umk.pl/dem D Y N A M I C E C O N O M E T R I C M O D E L S DOI: http://dx.doi.org/10.12775/DEM.2015.005 Vol. 15 (2015) 89−109 Submitted September 29, 2015 ISSN (online) 2450-7067 Accepted December 15, 2015 ISSN (print) 1234-3862 Katarzyna Andrzejczak, Agata Kliber* The Model of French Development Assistance – Who Gets the Help?∗∗ A b s t r a c t. Development cooperation is an important element of international relations because it influences the power balance between major players on the world markets and in the political debate. The aim of the article was to analyze the French development assistance model based upon the amount of help sent to Africa over the period 2001–2012. The motiva- tion of donor country is a crucial factor of development assistance, which influence not only the relations between donors and recipients, but also the effectiveness of aid. We estimated a series of dynamic panel models to assess whether the poverty-related factors play a domi- nant role in the distribution of help. On the contrary, we found that the most important varia- bles appeared to be the political and economic dependencies, among others: colonial history and oil/gas reserves. K e y w o r d s: development assistance; bilateral aid; development cooperation; developing country; France, dynamic panel model; J E L Classification: F53, F54, I32, O15. Introduction The role and effectiveness of development aid have been repeatedly questioned ever since the system was established. The growing income dis- * Correspondence to: Katarzyna Andrzejczak, Poznan University of Economics e-mail: katarzyna.andrzejczak@ue.poznan.pl, Agata Kliber, Poznan University of Economics, e-mail: agata.kliber@ue.poznan.pl. ∗∗ This work was supported by the Polish Ministry of Science and Higher Education through the Project SONATA: 2013//09/D/HS4/01849 (Mechanizmy wykorzystania technologii w Afryce Subsaharyjskiej). Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 90 parities between the North and the South, energy crisis in 1970s, debt crisis and structural adjustment in the late 1980s, numerous political and military conflicts in the 1990s, the number of people living below the poverty line, as well as interference in political situation in beneficiary countries are only a few examples of system dysfunction, repeatedly analyzed in the develop- ment literature (e.g. Chenery, Carter, 1973; Bonne, 1996; Kosack, 2003; Amprou et al., 2007; Easterly, Pfutze 2007; Doucouliagos, Paldam, 2008; Roodman, 2008). At the same time, the economic and social context, as well as the main character of development cooperation changed over the years. The critics of aid and donors as well as the appearance of new donors on the scene lead to official redefinition of development cooperation in year 2000 (United Nations Millennium Development Goals Declaration). The malfunc- tioning of the development cooperation system has been officially recog- nized and the aid has been declared to serve development issues, not the donors’ interests. In the meantime, the emergence and spreading of global communication means and channels allowed more open debate on migra- tions, detriment of environment, health and natural disasters as well as ter- rorism threats issues. This increased public opinion’s role in evaluation of the Northern countries approach towards less developed regions. As a con- sequence, the role of grass root level of cooperation has been noticed and its importance raised in the development agenda. A philosophy to direct the support towards local communities in need with the omission of state admin- istration often affected by corruption and ineffectiveness (Tanburn, 2008) has been widely recognized The landscape of development cooperation has undergone major changes in the last decade especially due to the changes in the “donors’ club”. This means that the donors represent very different (not only political) contradic- tory interests and approaches, and that they start to “compete” as donors. The appearance of the “generous” Nordic states and multilat-eral organiza- tions, such as European Union, then the so called South countries, compelled the “old donors” to revise their international policy in development coopera- tion (McEwan, Mawdsley, 2012). The appearance of new actors in the system was a main driving force of its evolution. The architecture of aid relations known in 20th century and based on Organization for Economic Cooperation and Development Assistance Committee (hereafter DAC), World Bank and International Monetary Fund is therefore being replaced by a more complex system of actors and approaches (Gore, 2013). The appear- ance of emerging economies and private actors and organizations has changed the international scene and requires rethinking of the traditional development assistance schemes (Eyben, 2012). Main aid donors are no The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 91 longer only the Triada countries. Old donors, which placed the themes of democracy and the market economy as pivotal for aid policy agenda are now only a group in a more complex donor group (Kim, Lightfoot, 2011). The DAC-ability concept, which embraced the philosophy of aid giving and the very definition of what aid is (Official Development Aid term), is now chal- lenged by the non-DAC donors (NDD). There is an increasing volume of development aid provided by the emerging economies. This aid influences international relations between states and redesigns the cooperation scene, influences and interests. Among the symbolic signs of change, the Fourth High Level Forum on Aid Effectiveness was organized at Busan in South Korea. The changes in aid system are undeniable, however, apart from the opti- mistic declarations, it is debatable to say that the system has evolved for the benefit of aid recipients. It is therefore doubtful that the donors stopped fo- cusing on realization of their external policy goals in the scope of develop- ment cooperation. Because of this, the dilemmas that undermine the very concept of development cooperation need to be addressed. The critics of aid donors persist. The ineffectiveness problem remains unresolved, as the Paris Declaration agenda of 2005 does not seem to be fulfilled. The relations be- tween development agents tend to rely on one-sided domination, not partner- ship. Rather than development aid creating growth, Doucouliagos and Paldam (2013), suggest it is possible that economic growth itself may influ- ence donor aid allocation decisions. Since donor dysfunctions burden aid efficiency (Simplice, 2014), in the analysis of aid effects in recipient coun- try, the issue of donors motivation in decision making process is crucial. Development assistance system is given, so the research should concentrate on finding the means to make it more effective and enhance countries’ ca- pacities to develop. For these effects to appear, we assumed that they are more probable, once the aid is provided with such an aim. The goal of this paper is, therefore, to analyze the bilateral development cooperation in the context of the donor motivation to provide assistance. We chose France as a model country representing traditional donors, since it has the characteristics of traditional donors. However, we are aware that the other traditional countries need to be tested for compliance with the model in further research. We concentrate on the aid for African countries. First, we discuss current literature on development assistance cooperation and next we introduce the traditional model of aid, based on qualitative analysis of coop- eration dimensions. Methods of comparative and system analysis were ap- plied in order to elaborate theoretical foundations for the models of coopera- tion. We used the system-GMM approach for dynamic panel models to veri- Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 92 fy the motivation of the donor in the aid distribution over the period 2001– –2012. It appeared that the poverty-related factors do not play any signifi- cant role in the aid distribution, but rather the political and economic ones. The robustness of the results was confirmed by the non-linear correlation test (Kendall’s tau). In the last section we discuss the implication of the results. 1. Traditional Development Aid Donors Development cooperation system is a post-Second World War phenom- enon, which was initiated by the USA as a response to the economic situa- tion and Cold War in Europe (overseas development assistance). It’s further evolution was a consequence of the decolonization process and the inde- pendence of African states (Williams, 2014). In the OECD terminology, the aid is associated with Official Development Assistance (ODA) public flows to beneficiaries placed on the DAC (Development Assistance Committee) List of ODA Recipients and to multilateral development institutions. These transfers are concessional in character – they convey a grant element of at least 25 per cent (calculated at a rate of discount of 10 percent per annum) and are administered with the promotion of the economic development and the welfare of developing countries as the main objective (OECD, 2014). That leaves out a great deal of development initiatives by private entities (enterprises, NGOs) but also public actors (non-ODA initiatives, local com- munities cooperation). Moreover, currently some major donors, such as China, do not report their flows to OECD at all. The character of these flows is often unknown, so there is no possibility to conclude if they meet the requirements of ODA or not. Because of that, the term development cooperation should not be limited purely to ODA transfers recognized by the DAC anymore. Develop- ment aid gains a wider meaning, which include ODA and other development purposes transfers, not reported to DAC. In a broad sense, aid is neither lim- ited to the type of transfer nor to the type of agents in the relation. However, it still does not include some “foreign aid” elements, such as military sup- port. In this article we concentrate on development aid, treated as an instru- ment of a state foreign policy towards less developed countries. The broadening range of donor countries is a consequence of major changes in World economy. While the group of aid recipients is shrinking, the group of donors grows (the good news about the aid system is just about it). The enlarged donor club is currently an amalgam of different countries. Verifying whether their aid is instrumentally subordinated to the long and short term strategic goals of foreign policy is an important aim. Analysis of The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 93 donors motivation requires more than dividing them into DAC and NDD groups, because there are differences between the donors in both of the groups. The division logic may be based on the geography, their public will- ingness to adhere to DAC standards, level of economic development, politi- cal regime, and others. There are OECD countries that are not members of DAC–Mexico, Turkey and several European countries; new EU Members; Middle East and Organization of the Petroleum Exporting Countries, and non-OECD donors from neither of these groups: Brazil, China, India and Russia (Kim, Lightfoot, 2011). In the DAC there are traditional donors, Scandinavian states (generous), new-DAC members, including new EU Members. Such classifications are not exhaustive, but sufficient to under- stand the complexity of interests, which may be related with development cooperation. All these groups are specific in character and realize different development aid policy. In this paper we concentrate on the practice of traditional DAC donors, based on French example. We are interested to see, weather the emergence of new donors affected their actions or only the declarations. We suspect, that as Cumming and Chaffer wrote: “European member states are driven more by their desire to enhance their own relative power within the interna- tional system than by any overriding need to help African countries” (2011: 212). Traditional DAC donors based their foreign aid policy mainly on bilat- eral relations. Major contributors such as USA, France and UK were there- fore criticized for supporting dictators in favor of their geopolitical goals in the bipolarized reality of the second half of XXth century (Williams, 2014). Today, their image of Cold War players is still in force, but the war against terrorism and the peace maintenance in strategic regions of the World are of increased importance (Chou, 2012). The altruistic motivation of traditional donors involvement in any international security issue is questioned, as well as their true willingness to decrease poverty. Traditional donors approach has been first contested by the aid agenda of Scandinavian countries and multilateral organizations. Generally, Sweden and other Nordic states are considered not only most generous donors (Barczak, 2008) but also donors more focused on the idea of effective use of help, on building a civil society in developing countries, and on directing their assistance towards realization of Millennium Development Goals (Thiele et al., 2007). However, even the Nordic states are claimed to enhance their relative power and punch above their weight in international donor circles (Cumming, Chaffer, 2011). Multiparty organizations, on the other hand, are theoretically devoid of egoistic national interests and oriented to- Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 94 wards democracy promotion and human rights protection in recipient states (Der-Chin, 2003). Their policy stance was supposed to reflect the coherent interests of developed and developing countries. Because of this, the in- creased share of multilateral disbursements in the total donors’ aid volume was considered as higher aid quality. In reality, multilateral aid is biased for the Western World with the omission of national labels (see: Charbonneau, 2008). Ohler and Nunnenkamp (2014) found that the multilateral institutions do not take regional needs into account while directing aid, instead favorit- ism plays an important role for location choices. Traditional donors are in- fluential in the international organizations and may profit from their activi- ties indirectly. Therefore, the emergence of South-South development cooperation has influenced the position of traditional donors in a much serious way, than the recommendation to increase the multiparty aid element. Some research val- ues South-originated aid as multidimensional and encompassing the intrinsic and non‐economic roles of development and hence very practical for African economies (Babaci‐Wilhite et al., 2013). Realization of sustainable devel- opment goals investments in joint investment promotion mechanisms, joint programs for absorptive capacity and joint public-private partnership models are said to be more likely to happen in the scope of regional and South-South cooperation (UN WIR, 2014). However, the Chinese development coopera- tion practice is in many ways similar to French or American model (Chaponniere et al., 2009). The major accusations towards traditional donors are their limited involvement in cooperation and self-oriented approach – despite helping to close savings gap, based on the neo-classical theory, aid serves donors’ interests (Page, te Velde, 2004; Cumming, Chaffer, 2011). However, for China and India energy, land and raw materials imports from Africa are equally important (Bearce et al., 2009; Walker, 2008, p. 21). What is especially interesting, is that the appearance of new donors, which negoti- ate with African countries without the colonial history burden, and their increased presence in the region, influenced the need of a change in the tra- ditional donors’ policies. Both individually and as a member of organizations, France has always played leading role in development cooperation system. In the past France was realizing its neo-colonial Françafrique strategy, partly created by Jaques Foccart and supported by the “cellule Africaine” in Presidential Palace, who- ever was its resident since Charles de Gaulle. Historically France kept close relations with the countries of the so called champ and even closer with the group of pre carré, especially in Africa. The strategy embraced economic, political and diplomatic goals (Fuchs, 1993). For Africa – or more precisely The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 95 – for African leaders these relations have been a source of material, political and diplomatic power, while for France – more literally of resources and comparative advantage in international relations (Lancaster, 1999). France was more direct in their policy goals realization than the United Kingdom, for which African policy was a source of relative power – soft influence. French were less liberal than the British and took a more interventionist stance in African relations (Cumming, Chaffer, 2011). The element shared by France and UK was the establishment of language communities, which were enforcing the relations and creating another dimension for a closer link (Hugon, 2008). Today, according to the French development policy, main components of development agenda: economy, society and environment are subordinated to fighting with poverty and sustainable development promotion. Francphone terrirtories are the main area of development aid focus. French cooperation policy aims to address four mutually-supportive issues, promoting peace, stability, human rights and gender equality (1); Equity, social justice and human development (2); sustainable, job-rich economic development (3); protecting the environment and global public goods (4). Strategic partners of French development cooperation are: Benin, Burkina Faso, Burundi, Djibou- ti, Comoros, Ghana, Guinea, Madagascar, Mali, Mauretania, Niger, Centralafrican Republic, Democratic Republic of Congo, Chad, Togo, Sene- gal (MAE, 2015). Among them only two not being a former French colony, both rich in natural resources. Such a selection of partners indicates that the realization of national interests can actually remain a major issue for the development policy. In the next sections, the analysis of French practice will be studied in order to evaluate the motivation of France as a traditional aid donor. 2. Methodology Panel data modeling is a statistical methodology allowing to use infor- mation included both in time and cross-sectional dimension. The regression is therefore run on the two (or more) dimensions. Depending on the assump- tions made about the error term, one can talk about fixed- or random-effect models (see e.g. Hsiao, 2014; Greene, 2011). Let us consider a panel model of the following form: ,,,, tititi uy += βX (1) Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 96 where ity is the dependent variable observed at time t for the i-th cross- sectional unit, while ti ,X – the matrix of explanatory variables. Depending on the error decomposition we receive different model specification. For instance: ,itiitit uy ++= ηβX (2) is a “fixed-effect” panel model, that can be estimated e.g. through least- square dummy variable (LSDV) estimator. Such specification is justified when we assume that apart from the time-varying explanatory variables there are also time-invariant individual effects that influence the dependent variable. Another possible specification is the random-effect one: .itiitit uy +++= νηβX (3) In this formulation iν is a group-specific random element similar to itu , except that for each group there is but a single draw that enters the regres- sion identically in each period (Greene, 2011). If a lagged dependent varia- ble is included in the model, we talk about the dynamic panel data models. In order to utilize all available information: both changes of variables in time, as well as across countries, we composed a panel of the data, including 12 time periods (2001–2012) and 53 cross-section units (53 countries). Since the number of time periods in our model was relatively small (T=12) com- pared to the number of cross-sectional units (N=53), we used the system- GMM approach for dynamic panel models. Let us consider a dynamic panel model of the form: tiitititi uηyy ,,1,, +++= − βXα , (4) where tiy , denotes amount of financial help sent from donor (France) to the i- th African country at time t. In our case t = 1,…,12, while i= 1,…,53. β is a 1×k vector of the coefficients, α is the vector of autoregressive coeffi- cients, ti ,X is a k ×1 vector of explanatory variables observed for country i at time t, iη are so called individual effects (time-invariant), while it u – the disturbance specific for country i at time t. The model can be estimated using the so called “difference estimator”. The main idea of the approach is to first difference the data. Then, one obtains: titititititi ,,,,1,, uγWuXβyαy ∆+=∆+∆+∆=∆ − . (5) The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 97 The error term of the equation (2) is by definition autocorrelated and also correlated with the lagged dependent variable. All values of ktiy −, with k>1 can be used as instruments for 1, −∆ tiy . One-step estimator of equation (5) amounts to computing (Cottrel, Lucchetti, 2014):             ∆                        = ∑∑∑∑ == − == N i iiN N i ii N i iiN N i ii 1 ' 1 ' 1 1 ' 1 ' ˆ yZAZWWZAZWγ , (6) where: [ ] . , 0...000 .................. 0...0 0...00 , ... ... ,,..., 1 1 ' , 4,2,1, 3,1, ,3, 1,2, ,3, − = −       =               ∆ ∆ ∆ =       ∆∆ ∆∆ = ∆∆=∆ ∑ N i iiN Ti iii ii i Tii Tii i Tiii x xyy xy xx yy yyy HZZA Z W Once the 1-step estimator is computed, the 2-step estimated are obtained through replacing the matrix H with the sample covariance matrix of the estimated residuals. The 2-step estimator is consistent and asymptotically efficient. In our paper we used the so-called “system” estimator that complements the differenced data with data in levels, so the lagged differences are used as instruments (see: Blundell and Bond, 1998). The key equation of the system estimator is as follows (Cottrel, Lucchetti, 2014): ,~ ~~~~~~~~ 1 ' 1 ' 1 1 ' 1 '                 ∆                               = ∑∑∑∑ == − == N i iiN N i ii N i iiN N i ii yZAZWWZAZWγ (7) Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 98 where: [ ] .~~ , ...00...000 ........................... 0...00000 ........................... 0...0...000 ........................... 0000...0 0...00...00 ~ , ...... ......~ ,......~ 1 1 *' ,1, 3,2, ,2, 4,2,1, 3,1, ,3,,3, 1,2,1,2, ,3,,3, − = − − −−       =                           ∆ ∆ ∆ ∆ ∆ =       ∆ ∆ = ∆∆=∆ ∑ N i iiN TiTi ii TiTi iii ii i TiiTii TiiTii i Tiitiii xy xy xy xyy xy xxxx yyyy yyyy ZHZA Z W y The choice of matrix *H is not trivial. The details are presented for instance in Roodman (2009) or Hsiao (2014). See also: Dańska-Borsiak (2009). For more detailed description of panel data concept and modelling we refer the Readers to e.g. Hsiao (2014), Longhi and Nandi (2014) or Gruszczyński et al. (2012). 3. French ODA – the Data The amount of French ODA sent to Africa is very heterogeneous and it seems to depend both on the period and on the receiver country. We observe an enormous growth over the period 2004–2006 of the amount of help re- ceived by Nigeria, as well as high transfers to the Democratic Republic of Congo in 2003, to Congo in 2005 and 2010 and periodical higher transfers to Ivory Coast. In order to select a relatively homogeneous group, we excluded from the analysis those countries, where the high jumps in data were present (more precisely: we deleted from the sample those countries, where the standard deviation value was higher than the mean value). We filtered out: Congo, the Democratic Republic of Congo, Cote d’Ivore, Liberia, Mozambique, Nige- ria, Seychelles, Sierra Leone, the United Republic of Tanzania and Zambia. The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 99 In Table 1 We present the descriptive statistics of ODA sent from France to Africa over the period 2001–2012 in the filtered group of countries. The mean value of the help in US 2012 mln dollars amounted to 54.168, but the median only to 20.159. We divided the full sample into two subsamples: COLONY and NON-COLONY countries as well as OIL/GAS vs. NON- OIL/GAS ones. The differences in the amount of help received in each group is striking. In the case of the previous colonies the mean value of help amounted to 104.380 mln USD, while in the case of the non-colonies – only to 9.868. In the case of the countries possessing oil/gas reserves the amount of ODA received equaled 73.246, while in the case of the remaining ones – 36.247. The Cochran-Cox test for equality of mean values rejected the null hypothesis in the case of both pairs. Therefore, we can say that the amount of financial help sent to African countries is larger in the case of the ones that possess natural resources. The analysis of the “within” and “between” standard deviations reveals that the non-colony as well as the NON- OIL/GAS group are more homogeneous than the remaining ones. In all sub- samples the data is right-skewed. Table 1. Descriptive statistics of financial help from France to the African countries FULL SAMPLE COLONY NON-COLONY OIL/GAS NON-OIL/GAS Mean 54.168 104.380 9.868 73.246 36.247 Median 20.159 74.089 5.779 44.145 13.754 Minimum 0.025 1.350 0.025 0.580 0.025 Maximum 811.650 811.650 121.390 811.650 310.700 std.dev. 84.443 101.430 13.032 101.650 59.037 vol.factor 1.559 0.972 1.321 1.388 1.629 skewness 3.780 3.167 4.451 3.566 2.624 Kurtosis 22.150 15.013 29.524 18.033 7.094 5% percentile 0.670 14.011 0.469 2.195 0.445 95% percentile 210.170 268.570 27.311 232.060 189.310 Q3–Q1 66.111 88.910 12.570 93.550 38.716 missing obs. 4 0 4 4 0 obs. No. 512 240 272 248 264 within s.d. 49.744 71.795 10.320 68.146 20.964 between s.d. 70.302 76.229 8.596 79.419 56.705 3.1. Explanatory Variables In order to model the amount of financial help received from France, we chose the following set of explanatory variables: − MIGRATION – number of immigrants in a given year; − FDI – Foreign Direct Investment of the donor country in the receiver country; Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 100 − IMPORT_FUEL – value of import of fuels, lubricates and related prod- ucts (import from Africa); − EXPORT_FUEL – value of export of fuels, lubricates and related prod- ucts (export from France/Great Britain); − IMPORT_CRUDE – value of import of crude materials, inedible except fuel; − EXPORT_CRUDE – value of export of crude materials, inedible except fuel; − IMPORT – total import from the receiver country; − EXPORT – value of the total export to the receiver country; − POLITICAL_STABILITY – index of political stability in the receiver country; − EXTERNAL_DEBT – value of the external debt of the receiver country (as % of GDP); − CORUPT_CONTROL – value of the index of corruption control; − GIRLS_OUT_OF_PS – number of girls out of primary school; − MORTALITY_RATE – infant mortality rate (deaths per 1000 live births); − LIFE_EXPECTANCY – expected length of life in the receiver country; − GDP_PER_CAPITA – value of the GDP per capita in the receiver coun- try; − OIL – binary variable, taking one for the 12 African countries that have documented oil reserves: Algeria, Angola, Chad, Egypt, Equatorial Guinea, Gabon, Libya, ,Sudan1; − OIL_GAS – binary variable taking value 1 for the African countries that have documented gas and/or oil reserves, the gas-producers are: Angola, Benin, Cameroon, Chad, Equatorial Guinea, Ethiopia, Gabon, Ghana, Kenya, Liberia, Madagascar, Malawi, Mauritania, Namibia, Niger, Sen- egal, Sudan, Uganda. The source of the data were the following databases: OECD, and AFMI (Af- rican Financial Markets Initiative). The descriptive statistics of the explana- tory variables are given in Table 8 in the Appendix. The values of import, export and GDP are given in 2012 US dollars. The dependent variable was ODA – the amount of help received from the donor. All the data were col- lected for the time period from 2001 to 2012. The computations were per- formed using GRETL (Cottrel, Lucchetti, 2014, Kufel, 2011). 1 We excluded South Sudan from the whole analysis due to the lack of data for most of the indicators. The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 101 4. Results In Table 2 we present the results of the estimation of the model for the full sample. We observe that the amount of help sent from France to Africa depends on political factors and trade links. First of all, the higher the migra- tion rate from a given country, the higher the amount of help. The amount of help sent seems to depend positively on the amount of export sent to the given country. This suggests that the trade partners receive higher support. Eventually, the historical colonies receive significantly more than the other countries. Table 2. Results of the dynamic panel model estimation for France – the full sample Variable Coefficient Std.error t statistics p-value ODA(–1) 0.318 0.070 4.572 0.000 constant −62.219 24.406 −2.549 0.011 migration (–1) 2.972 0.920 3.231 0.001 colony 38.331 9.309 4.118 0.000 log(eksport)(–1) 3.970 1.448 2.742 0.006 Note: Sargant test statistics of overidentification amounted to 28.4 (p-value: 1), z-statistics for AR(1) test: –1.77 (p-value=0.07), for AR(2): –0.17 (p-value:0.87). Joint Wald test statistics: 168 (p-value:0). Standard error of residuals: 69.5. Table 3. Results of the dynamic panel model for France – the oil&gas countries Variable name Estimate std.error t-statistics p-value ODA(–1) 0.294 0.064 4.575 0.000 Const −93.3005 47.477 −1.965 0.049 Log(IMPORT) 5.424 2.583 2.100 0.036 COLONY 78.742 18.543 4.247 0.000 Note: Sargant test statistics of overidentification amounted to 16.24 (p-value: 1.00), z-statistics for AR(1) test: –1.73 (p-value=0.08), for AR(2): –0.28 (p-value:0.77). Joint Wald test statistics: 84.27 (p-value:0). Standard error of residuals: 79.19. In Table 3 we present the results obtained for the group of countries that have proven oil and gas reserves. The group is not homogeneous, as the de- scriptive statistics in Table 1 show. The heterogeneity of the group can be explained by the political instability and internal conflicts, large debt for- giveness events, and the delayed wealth effect. Moreover, the countries that exploit natural resources since many years are more wealthy than the coun- tries in which the oil/gas reserves have been found only recently. Thus, the explanatory power of the model is very weak. However, it shows the main drivers for the help distribution. It seems that the factors that influence gen- erosity is again the amount of import from France. Supporting outlets for French products is an important goal, which at the same time serves internal Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 102 policy, by the promotion of French industry and companies. Moreover, the countries that used to be French colonies receive on average 70 mln USD more than the other countries. Being a former colony increases the benefit of possession of natural resources. In Table 4 we present the results obtained for the more homogenous group – the countries that do not possess natural resources. From Table 1 we know that the amount of help received by them is significantly lower than those of the countries that have oil and/or gas. We observe again that politi- cal factors do play a role in generosity of the donor – the amount of ODA received grows together with the migration rate. The residuals from the model have the smallest standard deviation from all estimated ones – 29.51. Providing help to the countries of origin of diasporas living in France allows to realize domestic policy goals at the same time indeed. Table 4. Results of the dynamic panel model for France – the non-oil&gas countries Variable name Estimate std.error t-statistics p-value ODA(–1) 0.228 0.103 2.207 0.027 Const 18.660 5.422 3.441 0.001 Migration2 0.333 0.047 7.125 0.000 Note: Sargant test statistics of overidentification amounted to 21.41 (p-value: 1.00), z-statistics for AR(1) test: –1.82 (p-value=0.07), for AR(2): 1.34 (p-value:0.18). Joint Wald test statistics: 1749.29 (p-value:<0.001). Standard error of residuals: 29.51. Next, we estimated the models for the groups: historical colonies and others. In Table 5 we present the results of the model for colonies. We again ob- serve that the amount of help depends on political dependencies (migration) and trade (the more France imports from the donor, the more help is sent in return there). As the reason for the colonies was to ensure the access to cer- tain resources, the same logic explains the development agenda. Table 5. Results of the dynamic panel model for France – the historical colony group Variable name Estimate std.error t-statistics p-value ODA(–1) 0.311 0.063 4.969 0.000 Const −113.774 60.898 −1.868 0.062 Migration(–1) 2.326 1.386 1.679 0.093 Log(IMPORT)(–1) 9.452 3.772 2.506 0.012 Note: Sargant test statistics of overidentification amounted to 14.7 (p-value: 1.00), z-statistics for AR(1) test: –1.91(p-value=0.05), for AR(2): –0.12 (p-value:0.90). Joint Wald test statistics: 67.7 (p-value:0). Standard error of residuals: 82.36 Eventually, in the case of the group of non-colonies we estimated the static fixed-effect model, since the data did not exhibit any autocorrelation. This The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 103 can suggest that French policy is not consequent in this group of countries. The LSDV R2 amounted to 0.47 (the “within” one – to 0.07). In Table 6 we present the fixed-effect model with time-effect. The time effect for the peri- od 2009–2012 was insignificant, so we conclude that the average value of ODA received by the group was the same as in 2001 (in real value). Moreo- ver – the amount of help received by the countries from the group depended on the recipient and the year (this model was the best one from all other es- timated models, including those with additional explanatory variables). Each recipient received different and specific amount of help which varied in dif- ferent years but did not depend on any other factors – neither political nor poverty-related ones. We observe that in 2004 and 2006 the average amount of ODA was higher, while the lowest – in 2002. All the values presented in the table should be interpreted as the surplus in comparison with 2001. Table 6. Results of the static fixed-effect panel model with time effect for France – the non-colony group. Variable name Estimate std.error t-statistics p-value const 6.457 1.179 5.476 0.000 dt_ 2002 1.444 0.462 3.124 0.002 dt_2003 2.286 0.845 2.706 0.007 dt_2004 7.234 2.845 2.543 0.012 dt_2005 6.921 4.040 1.713 0.088 dt_2006 5.308 2.769 1.917 0.057 dt_2007 5.673 1.733 3.273 0.001 dt_2008 4.311 1.423 3.029 0.003 dt_2009 6.400 4.787 1.337 0.183 dt_2010 0.672 0.947 0.710 0.478 dt_2011 0.884 0.919 0.962 0.337 dt_2012 −0.320 0.909 −0.352 0.725 Note: LSDV R2 amounted to 0.47, while the within R2 to 0.08. LSDV F(33,238)=6.43(p-value= 0), F-test statistics for named regressors: F(11,238) =1.8 (p-value=0.05).Durbin-Watson statistics to 1.77. The null hypothesis for common constant in groups was rejected at p-value 4.22e-20. The p-value of the Wald test for common significancy of dummy time effects was equal to 1.75e-13. Standard error of residuals amounted to 10.14. 4.1 Robustness Check – Kendall tau In Table 7 we present the Kendall τ computed for the value of financial help from France and the remaining variables in the set. The τ is the rank- based correlation, indicating the non-linear dependencies in the data. The data in table is sorted according to the decreasing value of τ. According to this simple analysis, the political factors influence the amount of help sent the most. France seems to support mainly its own historical colonies and trade partners or – more precisely – the countries of the highest value of Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 104 import from France. A very important factor seems to be also the migration number, as well as the fact of being a historical colony of France. The fact of possessing oil reserves is also significant, but of lesser importance. France is using the nuclear power to quite a large extend, so this may be one of the explanations. From the poverty-related factors only the illiteracy is signifi- cantly related to the amount of ODA received. Political stability factor is also significantly related to the amount of ODA, but the relationship is in- verse (the less stable the country, the higher amount of help should it re- ceive). This is neither coherent with the Burnside and Dolar paradigm of aid effectiveness in sound policy environment nor with the general conditionali- ty of help based on democracy promoted by the DAC. The conclusions from the panel models seem to be robust. Table 7. Kendall-tau for financial help and other considered variables Correlation between ODA and: Tau p-value colony 0.576 <0.001 Migration 0.551 0 export_crude 0.531 0 export 0.529 0 export_fuel 0.362 0 import_crude 0.361 0 import 0.323 0 FDI 0.328 0 oil/gas 0.238 0 girls out of primary school 0.272 0 life expectancy 0.073 0.006 GDP_per_capita 0.046 0.086 mortality rate 0.017 0.517 corruption control –0.036 0.172 import_fuel –0.05 0.246 external debt –0.03 0.206 political stability –0.116 0 Note: Insignificant variables (p-value higher than 0.05) are put in italics. 5. Conclusions Most of development aid research concentrate on the aid effectiveness in the context of recipient performance. However, the motives of donors are also important. The non-profit and development-driven character of foreign aid flows are questionable. From the early stages of development coopera- tion former colonial powers enjoyed the access to the commodities from newly created states. The economic ties established during colonialism were not sealed off. Despite the reluctance towards former powers, the reconstruc- tion of the economy towards independence was a more complicated process, The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 105 than assumed. Moreover, the increasing political instability in many coun- tries and numerous military conflicts, made many African states fragile and explained traditional aid donors presence in the politics. Establishment of aid structures in African states, allowed donors’ gov- ernments to insert national business (especially of the key branches such as telecommunication, energy, minig etc.) on the key local markets. Companies without government support or local experience wouldn’t easily decide to invest due to high risks. The countries which were politically involved with African leaders, either for the past reasons (France) or the new ones (USA), were automatically predestinated to gain access to the resources of the re- gion. The traditional development cooperation model, represented here by France, is characterized by longtime relationship at relatively high intensity. The evolution of development assistance is the evolution of traditional mod- el policy declaration – from neocolonial relations to development agenda and sustainable development. This policy means both: keep using aid as an instrument of their foreign policy (votes in UN, peacekeeping, access to natural resources), and engage in the initiatives of Millennium Development Goals and the democracy building. They claim to be driven by moral obliga- tion for the past wrongdoing, but at the same time are perceived to use their superior position in negotiations. Our study suggest, that the traditional approach to development coopera- tion is based on donors’ interests in 21st century, just as it was in the prece- dent one. The analysis of recipient structure shows that the aid volume is correlated positively with oil reserves, both export and import and with mi- gration. To a lesser extent literacy rate and mortality rate were depicted as important. The motivation of traditional donor is therefore self-oriented. Development cooperation remains instrumental to realization of foreign and internal policy goals of donors. Governments of donor countries, responding to their voters, tend to subordinate development cooperation to increase real- ization of national and regional interests. Securing the access to markets and low cost imports supports both donor country’s entrepreneurs going abroad as the consumers inside the country. The orientation towards “migration countries” may imply the will to keep stable relations, possibly i.a. for the security reasons. Also, the importance of cultural relations for the develop- ment cooperation, especially because of the use of common language is rec- ognized. We conclude, that despite the official redefinition of the develop- ment cooperation goals, the system tends to serve the traditional donor’s agenda in the first place, and next the recipient’s needs are considered. Prac- tically, for the African states, it means that a stronger negotiation approach with the donors is needed. At the same time the improvement of internal Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 106 transparency to ensure local politicians accountability for the development cooperation agreements could positively influence the system functioning in the future. References Babaci‐Wilhite, Z., Macleans Geo‐JaJa, A., Shizhou, L. (2013), China's aid to Africa: com- petitor or alternative to the OECD aid architecture?, International Journal of Social Economics, 40(8),729–743, DOI: http://dx.doi.org/10.1108/IJSE-09-2012-0172. Barczak, I. (2008), Pomoc Szwecji w rozwoju społeczeństwa obywatelskiego w krajach rozwi- jających się (Supporting Civil Society Development with Swedish Development Aid), in Przybylska-Kapuścińska W. (ed), Gospodarka, finanse i społeczeństwo (Economy, Fi- nances and Society), Poznań, Wyd. AE Poznań, 241–245. Bearce, D. H., Floros, K. M., McKibben, H. (2009), The Shadow of the Future and Interna- tional Bargaining: The Occurrence of Bargaining in a Three-Phase Cooperation Framework, Journal Of Politics, 71(2), 719–732, DOI: http://dx.doi.org/10.1017/S0022381609090562. Blundel, R., Bond, S. (1998), Initial Conditions and Moment Restrictions in Dynamic Panel Data Models, Journal of Econometrics, 87(1), 115–143. Bonne, P. (1994), The Impact of Foreign Aid on Savings and Growth, London School of Economics, London. Brautigam, D. (2011), Aid with Chinese Characteristics: Chinese Foreign Aid and Develop- ment Finance Meet the OECD-DAC Aid Regime, Journal Of International Develop- ment, 23(5), 752–764, DOI: http://dx.doi.org/10.1002/jid.1798. Chaponniere, J., Comolet, E., Jacquet, P. (2009), Les pays emergents et l'aide au developpement (Emerging Countries and Foreign Aid Policy. With English summary), Revue D'economie Financiere, 95, 173–188. Chou, T. (2012), Does Development Assistance Reduce Violence? Evidence from Afghani- stan, Economics Of Peace And Security Journal, 7(2), 5–13. Corkin, L. (2011), Uneasy Allies: China's Evolving Relations with Angola, Journal Of Con- temporary African Studies, 29(2), 16 –180. Cottrel, A., Lucchetti, R. J. (2014), Gretl User’s Guide, GNU Regression, Econometrics and Time-series Library Web-DOC, http://gretl.sourceforge.net/gretl-help/gretl-guide.pdf (25.09.2015). Crescenzi, M. C., Enterline, A. J., Long, S. B. (2008), Bringing Cooperation Back In: A Dy- namic Model of Interstate Interaction, Conflict Management And Peace Science, 25(3), 264–280. Chafer, T., Cumming, G. (2011), From Rivalry to Partnership? New Approaches to the Chal- lenges of Africa, Ashgate Publishing Company, Burlington. DeHart, M. (2012), Remodelling the Global Development Landscape: the China Model and South–South cooperation in Latin America, Third World Quarterly, 33(7), 1359–1375. Dańska-Borsiak, B. (2009), Zastosowania panelowych modeli dynamicznych w badaniach mikroekonomicznych i makroekonomicznych (Dynamic Panel Data Models in Micro- economic and Macroeconomic Research), Statistical Review, LVI(2), 25–41. Charbonneau, B. (2008), Dreams of Empire: France, Europe, and the New Interventionism in Africa, Modern & Contemporary France, 16(3), 279–295, DOI: http://dx.doi.org/10.1080/09639480802201560. The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 107 Der-Chin Horng, A. (2003), The Human Rights Clause in the European Union's External Trade and Development Agreements, European Law Journal, 9(5), 677–701. Deszczyński, P. (2001), Kraje rozwijające się w koncepcjach ekonomicznych SPD. Doktryna i praktyka, (SPD Concept of Developing Countries. Doctrine and Practice), Wyd. AE Poznań, Poznań. Doucouliagos, H., Paldam, M. (2008), Aid Effectiveness on Growth: A Meta Study, Europe- an Journal of Political Economy, 24(1), 1–24, DOI: http://dx.doi.org/10.1016/j.ejpoleco.2007.06.002. Easterly, W. (2006), Les pays pauvres sont-ils condamnés à le rester, Mondes en développement, 3(135), 139–140, DOI: http://dx.doi.org/10.3917/med.135.0139. Easterly, W., Pfutze, T. (2008), Where Does the Money Go? Best and Worst Practices in Foreign Aid, Journal of Economic Perspectives, 22(2), 29–52, DOI: http://dx.doi.org/10.2139/ssrn.1156890. Eyben, R. (2012), Struggles in Paris: the DAC and the Purposes of Development Aid. European Journal of Development Research 25(1), 78–91, DOI: http://dx.doi.org/10.1057/ejdr.2012.49. Fuchs, J. P. (1993), Pour une politique de développement efficace, mâitrisée et transparent: rapport au Premier ministre, Web-DOC. http://lesrapports.ladocumentationfrancaise.fr/BRP/954089600/0000.pdf (25.09.2015) Gore, C. (2013), The New Development Cooperation Landscape: Actors, Approaches, Archi- tecture, Journal of International Development, 25, 769–786, DOI: http://dx.doi.org/10.1002/jid.2940. Greene, W. H. (2011), Econometric Analysis. 7-th Edition, Prentice Hall. Gruszczyński, M., Bazyl, M., Książek, M., Owczarczuk, M., Szulc, A., Wiśniowski, A., Witkowski, B. (2012), Mikroekonometria (Microeconometrics), Wolters Kluwer, War- szawa. Hsiao, Ch. (2014), Analysis of Panel Data. Third Etition, Cambridge University Press, Cam- bridge, DOI: http://dx.doi.org/10.1017/CBO9781139839327. Hughes, A., Wheeler, J., Eyben, R. (2005), Rights and Power: The Challenge for International Development Agencies, IDS Bulletin, 36, 63–72, DOI: http://dx.doi.org/10.1111/j.1759-5436.2005.tb00179.x. Hugon, P. (2008), L’economie du développement et la pensée francophone, Editions des Archives Contemporaines, Paris. Kim, S., Lightfoot, S. (2011), Does ‘DAC-Ability’ Really Matter? The Emergence of Non- DAC Donors: Introduction to Policy Arena, Journal of International Development, 23, 711–721, DOI: http://dx.doi.org/10.1002/jid.1795. Kmita, J. (1991), Essays on the Theory of Scientific Cognition, PWN – Polish Scientific Pub- lishers, Kluwer Academic Publishers, Warszawa. Kragelund, P. (2008), The Return of the Non-DAC Donors to Africa: New Prospects for African Development, Development Policy Review, 26(5), 555–84, DOI: http://dx.doi.org/10.1111/j.14677679.2008.00423.x. Kufel, T. (2011), Ekonometria. Rozwiązywanie problemów z wykorzystaniem program GRETL (Econometrics. Problems solving using GRETL), PWN, Warszawa. Lancaster, C. (1999), Aid to Africa, University of Chicago Press, Chicago. Longhi, S., Nandi, A. (2014), A Practical Guide to Using Panel Data, Sage, London. Mawdsley, E., Savage, L., Kim, S.-.M. (2013), A 'Post-Aid World'? Paradigm Shift in For- eign Aid and Development Cooperation at the 2011 Busan High Level Forum. Geo- graphical Journal, 180, 27–38, DOI: http://dx.doi.org/10.1111/j.1475-4959.2012.00490.x. Katarzyna Andrzejczak, Agata Kliber DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 108 McEwan, C., Mawdsley, E. (2012), Trilateral Development Cooperation: Power and Politics in Emerging Aid Relationships, Development And Change, 43(6), 1185–1209. OECD (2012b), Strategy for Development, Web-DOC, http://www.oecd.org/development/50452316.pdf (23.06.2013). Ohler, H., Nunnenkamp, P. (2014), Needs-Based Targeting or Favoritism? The Regional Allocation of Multilateral Aid within Recipient Countries, Kyklos, 67(3), 420–446. Osei B., How (2005). Aid Tying Can Impose Additional Cost, African Development Review, 17(3), 348–365, DOI: http://dx.doi.org/10.1111/j.1017-6772.2006.00119.x. Ouattara B. (2007). Foreign Aid, Public Savings Displacement and Aid Dependency In Cote d’Ivoire: An Aid Disagregation Approach, Oxford Development Studies, 35(1), 33–46, DOI: http://dx.doi.org/10.1080/13600810601167579. Page, S., Willem te Velde, D. (2004), Foreign Direct Investment by African Countries, Web- DOC, http://www.odi.org/sites/odi.org.uk/files/odi-assets/publications-opinion- files/5739.pdf (25.09.2015). Papanek, G. F.(1972), The effect of aid and other resource transfer on savings and growth in less developed economies, The Economic Journal, 82, 934–950. Przeworski, A. (2004), Democracy and Economic Development, in: Mansfield E.D. and Sis- son R.(eds.), The Evolution of Political Knowledge, Columbus, Ohio State University Press. Quadir, F. (2013), Rising Donors and the New Narrative of ‘South–South’ Cooperation: What Prospects for Changing the Landscape of Development Assistance Programmes?, Third World Quarterly, 34(2), 321–338, DOI: http://dx.doi.org/10.1080/01436597.2013.775788. Radelet, S., Clemens, M., Bhavnani, R. (2004), Aid and Growth: The Current Debate and Some New Evidence, Centre for Global Development Working Paper, https://www.imf.org/external/np/seminars/eng/2005/famm/pdf/radele.pdf (25.09.2015) Roodman, D. (2009), How to do xtabond2: An Introduction to Difference and System GMM in Stata, The Stata Journal, 9(1), 86–136, DOI: http://dx.doi.org/10.2139/ssrn.982943. Rosen, H. (1977), Technology Transfer to developing Nations, Journal of Technology Trans- fer, 1(2), 93–104. Sanfilippo, M. (2010), Chinese FDI to Africa: What Is the Nexus with Foreign Economic Cooperation?, African Development Review/Revue Africaine De Developpement, 22599–614, DOI: http://dx.doi.org/10.1111/j.1467-8268.2010.00261.x. Shirazi, N. S., Mannap, T. A., Ali, M. (2009), Effectiveness of Foreign Aid and Human De- velopment, Pakistan Development Review, 48(4), 853–862. Simplice, A. (2014), Development Thresholds of Foreign Aid Effectiveness in Africa, Inter- national Journal of Social Economics, 41(11), 1113–1155, DOI: http://dx.doi.org/10.1108/IJSE-01-2013-0014. Tajoli, L. (1999), The impact of tied aid on trade flows between donor and recipient countries, Journal of International Trade and Economic Development, 8 (4),373–388. Tanburn, J. (2008), The 2008 Reader on Private Sector Development, Web-DOC, http://www.ilo.org/wcmsp5/groups/public/---ed_emp/---emp_ent/--- ifp_seed/documents/publication/wcms_143158.pdf (25.09.2015). Thiele, R., Nunnenkamp, P., Dreher, A. (2007), Do Donors Target Aid in Line with the Millennium Development Goals? A Sector Perspective of Aid Allocation. Review Of World Economics, Weltwirtschaftliches Archiv, 143(4), 596–630, DOI: http://dx.doi.org/10.1007/s10290-007-0124-x. UN (2014), World Investment Report 2014: Investing in the SDGs: An Action Plan, Web- DOC, http://unctad.org/en/PublicationChapters/wir2014ch4_en.pdf (25.09.2015). The Model of French Development Assistance – Who Gets the Help? DYNAMIC ECONOMETRIC MODELS 15 (2015) 89–109 109 Walker, M. (2008), Indian Ocean Nexus, Wilson Quarterly, 32(2), 21–28. Williams, J. H. (2014), US Foreign Aid, Asian Education and Development Studies, 3(1), 11–30. Williamson, C. R. (2008), Foreign Aid and Human Development: The Impact of Foreign Aid to the Health Sector, Southern Economic Journal, 75(1), 188–207. Appendix Table 8. Descriptive statistics of the explanatory variables Mean Median Min Max Std.dev. Within s.d. Between s.d. Migration 1.726 0.207 0 31.113 4.584 0.719 4.571 FDI 91.711 6.276 –7532.6 12918 789.420 827.64 144.22 Import 5.12e+08 3.55e+07 1841.5 7.36e+09 1.17e+09 5.15e+08 1.06e+09 Import fuel 7.01e+08 9.60e+07 19 7.05e+09 1.33e+09 7.70e+08 9.90e+08 Import crude 2.06e+07 4.03e+06 3 2.68e+08 3.75e+07 1.59e+07 3.31e+07 Export 5.51e+08 1.18e+08 37316 8.42e+09 1.21e+09 3.78e+08 1.16e+09 Export fuel 3.29e+07 1.75e+06 185 8.61E+08 9.72E+07 7.4e+07 6.43e+07 Export crude 6.16e+06 8.40e+05 1157.9 2.54E+08 2.15E+07 1.05e+07 1.83e+07 Political stability –0.539 –0.36 –3.3 1.190 0.93 0.355 0.873 Corruption control –0.607 –0.67 –1.92 1.260 0.58 0.193 0.555 Girls out of PS 3.47e+05 1.57e+05 52 5.07e+06 7.32e+05 1.34e+05 7.63e+05 Life expectancy 56.282 55.708 12.3 114.4 13.203 11.612 7.170 Mortality rate 64.701 63.95 12.2 138.5 27.265 9 26.092 GDP per capita 2024.2 729.88 110.5 24355 3153.8 1434.8 2863.8 External debt 65.195 46.575 0.65 881.95 85.792 58.625 65.423 Model francuskiej pomocy rozwojowej – kto dostaje pieniądze? Z a r y s t r e ś c i. Artykuł przedstawia analizę dystrybucji francuskiej pomocy rozwojowej wśród krajów afrykańskich, w latach 2001–2012. Na podstawie wyników uzyskanych przy pomocy dynamicznych modeli panelowych autorki stwierdzają, że pomoc nie była kierowana do krajów najbardziej potrzebujących, ale do tych, z którymi łączą Francję więzi polityczne i gospodarcze (m.in. byłe kolonie oraz kraje zasobne w ropę i gaz). S ł o w a k l u c z o w e: pomoc rozwojowa, Deklaracja Milenijna, ubóstwo, Francja, dyna- miczny model panelowy Pusta strona