Research methods unit I article review

International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 51 The impact of research and development (R&D) on eco nomic growth: The case of the MENA countries.

1Lamia Ben Amor and 2Naceur Ben Zina 1. Member and researcher at the unit of reasearch, ec onomic and developpement at the University of Economics and Management of Sfax, Tunisia. 2. Associate Professor of Economics at the Unit of Dy namic Economics and Environmental Research (URDEE) at the University of Economics and Manageme nt of Sfax, Tunisia.

* Corresponding Author E-mail: lamiabenamor.fseqs@g mail.com --------------------------------------------------- ----------------------------------------------------------------------- ABSTRACT The aim of this paper is to study the relationship between research and development (R & D) and economic growth. We will try to enlighten, from a theoretical side, the importance of R&D as a critical factor for recovery. This empirical study is based on different estimati on methods developed in the context of a static pan el on a sample of 15 countries over the period 1980-2009 . The methods used are the GMM (Generalized Method of Moments), the OLS (Ordinary Least Square s) and causality tests, a unit root applied to panel data. The overall findings identify a positive and significant relationship. JEL Codes: C33, D83, F43. Keywords: R & D, Economic Growth, Static Panel Model.

1. INTRODUCTION To cope with the increased globalization, competiti on from emerging economies and ever-changingneeds of consumers, the economies of industrialized coun tries are increasingly relying on the activities of research anddevelopment, which are essential to pro vide the firms with the competitive advantages necessary for their competitiveness. According to the OECD (2004), expenditure on resear ch and development (R & D) conducted International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 52 inindustrialized countries (EU, U.S., Japan) incre ased from 396 billion euros in 1995 to more than680 billion in 2003. In the same way, some emerging eco nomies seem to be determined also to increase their technological capital. Let us takeChina as an examp le: by reference to OECD, its expenditure of R& D expressed as a percentage of the GDP (gross domesti c product) doubled during the period 1995 -2002.

Indeed, they went from 0.6% to 1.2%. The key role of R & D in the competitiveness of fir ms led by industrialized countries to seekterritories and more attractive partners to carry out their inn ovation strategies, including emergingmarkets, whic h should increase their technological capital. R&D o ften implies a set of cumulativeprocesses of inventions and innovations, skills and know-how. Th rough its horizontal and verticaldiversification of goods and services, it is considered as the generat or of technical progress. R&D, innovation, technological developments have been regarded since the second half of the eighteenth century as an indisputable source of productivity growth. However , the non-rival and partially excludable product of R&D recognizes that certain agents or sectors have not fully developed the capabilities of their own R&D, thus benefiting from other’s R&D. The major objective of this paper is to examine the impact of research and development (R&D) on economic growth in MENA countries as host countrie s. A developing country that suffers from under-capacity in R&D efforts can benefit from int ernational R&D. We have choosen a sample of these countries ( we selected 15 countries: Algeria, Bahr ain, Djibouti, Egypt, Iran, Israel, Jordan, Kuwait, Malta, Morocco, Palestine, Saudi Arabia, Syria, Tun isia, UAE) during the period 1980 - 2009 in which information is available on R&D expenditure. (We ch ose this period starting from 1980 due to unavailability of data for some countries before th e year 1980 as Tunisia and Algeria). This paper is structured as follows: The next section describes the literature on research and development (R&D) where we particulary analyze the importance of R&D in economic growth. Indeed, this paper is a general summary of the reason for having a link bet ween R&D and economic growth. As for the the third section, it presents our empir ical validation test based on static panel data. The appropriation of R&D is the subject of section four . International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 53 2. Literature review on research and development (R &D) The recent economic studies whether on theories of endogenous growth or on international trade show that thebenefits of R&D do not stop at national bo rders. In this context, Eaton and Kortum (1996) estimated that the foreign R&D accounted for 87% o f productivity growth in France. While 60% gain inproductivity in Japan is provided byU.S R&D, aga inst 20% of U.Searnings due to Japanese R & D. The importance of technological innovation and R&D expenditure in economic growth has drawnthe attention of economists since the launch of the re search on the origin of the concept of"residue" stated by Solow (1957). It has become much more explicit in the recent years due tothe theoretical developments undertaken in the framework of the th eory of endogenous growth. Moreover, the impact of R&D on productivity growth has been the subject of several studies:Griliches (1979), Mairesse and Cuneo (1985), Dosi (1988). Th e true modelling of R&Dstarted as from the years 1990s. Indeed, the activity of R&D represents the important source ofdevelopment of new knowledge and technological innovation (Crossman and Helpman (1991)). In the same way for Romer (1990), it leads to techn ological innovations. Contributions came then With Aghion and Howitt (1998). In addition, it is a domi nating source of economic growth. In fact, theactivity of R& D took a great importance in the recent mode ls of the endogenous growth whosecreative processes and diffusion of technology constitute t he important engines. Moreover, R&D isconsidered as a product of technology. It is used to improve the living standard, increase theproductivity of the f actors of production and stimulate economic growth. In this context, we present the theoretical and emp irical basis as well as the nature of research anddevelopment (R&D). In other words, we highlight the endogenous theory of growth where R&D plays a fundamental role. Then, we will try to ide ntify the effect of R&D on the economic growthof th e MENA countries starting from a static panel model. Within this context, it makes thetheoretical, empirical, and the nature of research and developme nt (R&D). It focuses on the theory of endogenous growth where R&D occupies a fundamental role. International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 54 Thus, with the increased interdependence of economi es, this process of dissemination of knowledge and expertise is gaining importance internationally. I n addition, international agreements between countr ies require them to confront international competition based on competitiveness and differentiation capabilities. In addition, total factor productivit y (TFP) of a given country depends not only on its own stock of R&D but also on that of its partners. In this context, we present the theoretical and emp irical basis as well as the nature of research and development (R&D). In other words, we highlight th e endogenous theory of growth where theR& D plays a fundamental role. Then, we will try to ide ntify the effect of R&D on the economic growth of t he MENA countries starting from a static model of pane l. Thus, with the increased interdependence of economies, this process of dissemination of knowled ge and expertise is gaining importance internationally. In addition, international agreeme nts between countries require them to confront international competition based on competitiveness and differentiation capabilities. In addition, total factor productivity (TFP) of a given country depen ds not only on its own stock of R&D but also that of its partners. Thanks to modernization and fast technological deve lopment, the research and development (R&D) constitutes an important factor of growth. Moreove r, it is essential to follow the activities of R&Dto work out adequate policies and analyze them well. This idea is dictated by a certain number oftheoretical and factual arguments. Concerning the theoretical arguments, we notice th at the numberof new models of the endogenous economic growth rises by giving a particular impor tance to theactivity and expenditure in the research and development. Indeed, R&D constitutes a primary source for technological advance and economic growth. Besides, the development andenrichment of t he research fields of the international macro-economy, following the intensificationand the diversification of the sources of interdependence of the economies, represent important factors. As for the factual arguments, we notice the attenti on given by certain countries, such asTunisia which has highlighted the activity of the research and devel opment since theimplementation of a program of International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 55 structural adjustment (PSA). This program created, in 1991, a newpublic agency which takes care of the promotion of innovation (the Secretariat of State to the“Scientific research” and technology). Hence, the very fast evolution of telecommunication and inform ation technologies in the world and their implication on the total productivity of the factor sof the companies constitutes an important factor. In fact, several authors are interested in thisphenome non by taking it as a fundamental subject of resear ch.

Frascati (1993), for example, definedR&D as the who le set of work of creation undertaken in a systematic way to increase the stock ofknowledge and its impo rtance for new applications. In his Handbook entitled “Standard Methodsuggested for the investi gations into research and experimental development” , OECD distinguishes three forms of R&D: Fundamental R&D: the expenditure is devoted to the analysis of the properties, the structures andthe physical and natural phenomena, in order to organize the facts in general laws by means ofexplanatory diagrams and interpretative theory. Applied research: it allows the operational working of ideas. The knowledge orinformation drawn from this research is likely to be patented and ca n be kept secret. This typeisundertaken either to distinguish the possible applications of the resul ts of a basic research or to findnew solutions.

Experimental R&D: it consists of a systematic work founded on the knowledge obtained by theapplied research whose objective is to acquire n ew materials, devices, products, processes and services.

The endogenous theory of growth devotes a great imp ortance to the part played by research anddevelopment in the productivity growth making it in the centre of several studies. R&D isregarded as a primary source of technological advance, and esp ecially represents one of the enginesof economic growth for the countries which devote much effort t o research and development inthe form of expenditur e.

Since the activity of R&D is a form of knowledge, i t represents the mainway of developing new knowledge and technological innovation for the deve loped countries (Crossman-Helpman (1991)). These countries devote important financial resources to t his activitywhich acts positively on growth in term s of productivity. Therefore the real expenditure of R&D in these countries increases in a remarkable way. The International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 56 case for the developing countries is completelydif ferent. These countries do not carry out research activities by themselves. However, they benefitmainly from the technical chan ge which takes place in the developed countries thanks to their opening on the outside world throug h the importation of new technologies incorporated in capitalequipment. The theory of growth shows the role of the technolo gical advance in the invention of new machines and new intermediate goods which give way to the possib ility of investment. The human capital andtechnology are two facets of t he same phenomenon, two factorsfor the accumulation ofknowledge as the strategic role of k nowledge explains the increase in the immaterial investment. Lucas (1988) considers that education i s an investment in human capital ; in his model, knowledge is considered under its individual aspect (education, formation) andhe seek to integrate more significant dimensions as well as the objectives o f knowledge. As knowledge is the result of a collective activity and research takes the form oftechnology. The key idea is that research is an investment. Actually, we no tice that the 20 last years have been characterized by a development inthe activities of R&D. Indeed, in al l OECD countries, the firms have strongly raised theirexpenditure on R&D. Investment in R&D is toda y more than even before, a conditionof the growth of nations. In fact, the activity of R&D necessaril y goes through the existence of certainqualificati ons which can control the tools of research. Innovation, technological advance and research and development (R&D) evolve by playing a very important part in the process of economic growth. These last years have been characterized by theappearance of many research orientations, thank s to the differences in terms of the growth rate of thereal income per capita and also to other factors which propel the process of research. In this res pect, we can primarily distinguish three factors in favou r of the evolution of this process. But, what isinteresting is that th ere is always a persistence of differences in terms of income by capita on the international level. These successive factors are: An increase in the volume of work. International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 57 An increase in the stock of capital.

An increase of the total productivity of the factor s of production “PTF”. The increase in productivity is directly connected with the evolution of labourqualifications, on the one hand, and with innovation and technological advanc eon the other hand. It involves theexistence of more powerful capital equipment (built-in technological advance) and results in a greateffectiveness of the two joined factors of production. Investments in R&D directly contribute tothe accumu lation of knowledge by giving place to new products and processes of production thuscontributi ng to the improvement of productivity. The advantages which ensue from the investments in R&D are spread between the companies and the organizations. Indeed, the productivitygrowth of a company depends not only on its activities in R&D, but also on its efforts towardsthe R&D of the other generating sectors of knowledge. Therefore, investment in R&D is regarded as a sourc e of improvement of productivity. The technological advance constitutes adetermining factor of growth r ate through its impact on the productivity of the factors of production. It includes innovations of a technical and organisa tional nature. We distinguish two major types of innovation: • Innovations of the products which are used to int roduce new products on the market (example:cds, car s) or raw materials. Their objective is to offer the c onsumers a product which meets theirneeds (Improvement of the well being). This type of innovation depends directly on the act ivities ofthe R&D. The innovation of a procedure based on the implemen tation of new methods of production. To day, we clearly observe an expansion of the poli cies of innovation mainly in the developed countrieswhich devote a great investment to the R& D. This shows the existence of a suitable climate forinnovation in these countries. It is remarkable that during these last years, companies of hightechnology or state-of-the-art technology (data -processing, pharmaceutical) have significantlyincreased their expenditure in researc h and development. The role of the government International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 58 policiesregarding the R&D should not be neglected. Indeed, the policies of innovation define the specificactions of the State, which must, on the on e hand, encourage the accumulation of a qualified labour and, on the other hand, help the companies p rospect the markets. This justifies the need of thepublic administrations to support the R&D. What are then the reasons for the government aid an d the alternative mechanisms available to thepublic administrations to support the R&D? To answer this question, we will try to analyze thejustification of the government aid in the R&D starting from the theorie s of economic growth.

In the neo-classical theory of growth, the technolo gical advance is supposed to be exogenous (Solow 1956). With the long run balance, the growth of th e population and technological advancedetermine the level of growth rate. This implies, according to basic assumptions, that long-termgrowth rate is stable and given in an exogenous way. Within this framework, the impact o f anaction of the public authorities is practically ignored. The neo-classical theory of growth presumes that th e economy starts from a weak relationship between capital and labour. Similarly, the marginalreturns on capital are decreasing, which reduces the motiva tion for investment in the new capital. Therefore, each new unit of the capital produces a lower income and less large savings. In the longrun , there will be an absence of incentives for investm ent. Consequently, the capital growth rate bycapita is cancelled. The neo-classical theory of growth presumes an effi cient distribution of resources in the economy. The assumptions which underlie this theory exclude the modification from this distribution. Therefore, the intervention of the State in such an economy is not justified at least in the field ofefficiency.

Any policy affecting the resource allocation would decrease the total production anddelay the economic growth. However, the intervention of the state coul d be founded on the criterionof equity. But, the assumptions underlying the neo-classical theory are not realistic. Thetechnological change is not always an exogenous factor outside the market, determined by anun known process. In the 20th century, a good International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 59 number of discoveries and improvements were madein the commercial sector by companies with a lucrative goal and not by public administrations or universities where researches are directed by non commercial forces. Markets are seldom in aperfect competition. Moreove r, the private sector cannot produce all the desired goods and services because some of them are public goods and others produce external effects. The endogenous theory of growth recommends the rela xation of certain neo-classical assumptionsand incorporates the failures of the market. However, t he long-term economic growth is directed bythe accumulation of the factors of production namelyfou nded this time, on knowledge, humancapital, training, R&D and innovation. The neo-classical theory of gro wth identifies only one sourceof growth - the accumulation of the physical capital - which genera tes machines, structures and stocks and differs from the “R&D capital”. The latter includes the highly s pecializedor very specialized labour. The neo-classical theorists do not obviously neglect the other source s, but they explicitly do not integrate them in the models by considering that technologicaladvance is an exogenous variable. On the other hand, the model s of endogenous growth arecharacterized by a great di versity of selected resources such as the investment in physical, human and public capital, in the divis ion of labour, practical training (Learning by doing), research and technological innovation. The endogeno us theory of growth recommends that technologicaladvance ensues from the R&D carried ou t by profit-making companies. Thus, research and development constitutes an important factor in the production process. The models worked out in this theory try to explain the origin of the endogenous technologicaladvance which is characterized by an increase in the variet y of the goods produced or consumed, orby the improvement of quality. We distinguish, in general, four ways of research: The first one results from the work of Romer (1986) and seeks the engine of growth in thephenomenon of practical training which takes pla ce inside the companies .

The second, introduced by Lucas in 1988, favours th e accumulation of the human capitalwithin the educational system. International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 60 The third insists on the driving role of infrastruc ture and public expenditure. This current is initiated by the work of R. Barro (1990-1991). The fourth, based on the work of Romer (1990), Aghi on and Howitt (1998), highlights the role of R&D or innovation. We can distinguish three levels of innovationsystem s:

A system of the innovating company; A system of the immediate environment of the compan y; A system of the overall environment concerning the general elements which support innovation. In this respect, innovation is first of all: A busi ness of the companies, which, to innovate, areinter acting with other companies, government organizations, uni versities, etc. Then, it is a system of innovation which includes these whole factors and their intera ctions. In short, the assumption, according to which the factors determining the long-term growth are endoge nous with the decisionmakingprocess, constitutes one of the main exemptions from the neo-classical t heory of growth andinvolves important effects on th e policy. Indeed, if the long-term growth is directed by the factors of production based on knowledge, which belongs to the normal structure of the costs of the company, then the public administrations can influ ence the long-term growth by changing the cost of thesef actors via the direct subsidies, taxincentives or marketing policies. The endogenous theories ofgrowt h provide a framework of analysis of growth and its determinants which can also be used to study the im pact of the public policies on the economic growth and investment in R&D. 3. METHODOLOGY R&D is a cumulative process of acquisition of compe tences and know-how. The transfer oftacit knowledge is delicate and represents a specific cha racteristic of the activity of R&D.Moreover, some expenses necessary to R&D, such as the capital expe nditures ofinfrastructures, are irrecoverable.

Therefore investment in R&D is an irreversible one and subjected to uncertainty. From the uncertainty of International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 61 the cumulative process (R&D) emerges adynamic behav iour research, which provides information on the intrinsic nature of the good. Themajor objective of this paper is to know the impact of research and development (R&D) oneconomic growth in the MENA cou ntries. (The countries taken in our sample are about 15; these are: Algeria, Saudi Arabia, Bahrain , Djibouti, Egypt, UAE, Iran, Israel, Jordan, Kuwait, Malta, Morocco, Palestine, Syria and Tunisia) durin g the period 1980 - 2009 where information onexpenditure on the R&D is available. We try to follow an approach in which we consider r esearchand development to be a fundamental factor in the economic growth in these countries. This emp irical study and the nature of research and development (R&D). It focuses on the theory of endo genous growth where R&D occupies a fundamental role. Thus, we try to identify the effect of R&D on economic growth in MENA countries from a static panel model. This empirical study is based on different estimati on methods developed recently such as the GMM (GeneralizedMethod of Moments) and the OLS (Ordinar y Least Squares). Hence, we use the causality tests andunit root while adopting the recent litera ture on the econometrics of panel data. Finally, we estimatethis model by the method "Full Modified Ordinary Least Square” (FMOLS), and then interpret the results. 3.1 Model Overview Within the framework of this study, we consider a t ransformed linear log function Cobb – Douglas of th e following production: The examined basic model takes as a starting point the work of Sadraoui and Ben Zina (2007). We limit ourselves to the study of the simple linear models of the panel data which are defined inopposition to the dynamic models using delayed endogenous variables. We assume that the per capita GDP incurred by governments is a function of the R & D, capital, an d labour. The model is presented as a log-linear relationship between variables. Thus, we define: International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 62 Y: the real gross domestic product per capita for cou ntry (i) in year t; K: the value of the real capital for country (i) in y ear t; L: the labor for country (i) in year t; RD: the ratio of the expenditure in research and devel opment to the GDP for country (i) in year t; : the term of error. 3.2 Results and interpretations 3.2.1 Graphical representation of the different va riables Figure 1 : Logarithm of Gross Domestic Product Figure2 : Logarithm of Labor Figure3 : Logarithm of Capital 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 80 82 84 86 88 90 92 94 96 98 00 02 04 06 LPIB 9.88 9.92 9.96 10.00 10.04 10.08 80 82 84 86 88 90 92 94 96 98 00 02 04 06 LL International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 63 Figure4 : Logarithm of research and development (R& D) We note that the gross domestic product (GDP) in th is country is changing significantly. Indeed, theGDP has been trending upward.

Moreover, the GDP trend is different according to c ountries of MENA. Still, on the other variables, we distinguish a particular development whose series show a trend, which makes non-stationary variables. A study of the series in level shows that they are not stationary; the different analysis of each series shows that they are differenced stationary. 3.2.2 Study of stationarity and cointegration of s eries The unit root tests have become a common approach f or the analysis of stationary time series.

However, the implementation of these tests on panel data is recent. If we refer to the article "R&DCooperation and Economic Growth", we find that the tests most frequently used are those of Levinand Lin (LL) and Im, Pesaran and Shin for (IPS ). Indeed, these two tests are used to study the 10.0 10.5 11.0 11.5 12.0 12.5 80 82 84 86 88 90 92 94 96 98 00 02 04 06 LK 0.4 0.6 0.8 1.0 1.2 1.4 80 82 84 86 88 90 92 94 96 98 00 02 04 06 RD International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 64 nonstationary series. To overcome the low power of tests "LL" and check the stationary state of thegroup, we use the IPS method which suggested a u nit root test in the context of a panel data model using the average individual statistics regressions "ADF". The longitudinal cross-sectional datashould ideally meet the assumptions necessary for the appl ication of the alternative statistica t-bar totest the null hypothesis of unit root for all i ( =0) t p 1 N t p !" Where: ti T (pi) : represents ADF tests estimated w ith delayed différences ; N : is number of groups N= 1,2,……………….15. T : The period of time T =1,2,………………..30 IPS suggests the use of fllowing of standardized st udies: Z √ N t % E't ( ) vart Where E ( t ) and Var ( t ) are respectively the arithmetic means and varian ces of the individual ADF statistics, since 0. The IPS study shows that the standardized stati stics converges weakly to the normal standard distribution, allowing to compa re it to the critical values of distribution. The application of the unit root tests show that LL and IPS are all statistics assigned to a unit root. Table 1: Unit root tests Statistique Y L K RD Levin- Lin rho-stat -0.04484 0.86878 -0.16490 0.17205 Levin-Lint-rho-stat 0.79204 1.99911 0.71672 1.20149 Levin-Lin ADF-stat 1.84314 2.15673 0.81625 1.34737 IPS ADF- stat 1.13540 1.89654 -0.83109 0.59321 International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 65 The checking of properties of non stationary state for all the variables of the panel leads us to study the existence of a long-term relationship between these variables. In other words, the study of the existence of a relationship of cointegration, by applying Pedron i’s tests of co-integration, is based on the unit root tests of the estimated residuals. The co-integration tests on panel data are to test the presence of the unit root in the estimated resi duals.

However, the problem of deceitful regressions, well known in econometrics time series, also arises in the case of the panel data. Pedroni‘s tests are of null hypothesis of no co-int egration based on the unit root tests of the estima ted residuals. Pedroni has developed seven homogeneous and heterogeneous co-integration tests on the panel data. These tests take into account the heter ogeneity in the co-integration relationship, that is to say that for every individual, one or more co-integrati on relations are not necessarily identical for each panel individual. The implementation of Pedroni’s tests r equires a first step in estimating the long-term relationship for each individual described by:

y / α δ t β "x"/ ⋯ β 5 x5/ ε / such as : i =1…….N, t =1…….T et m =1………M In Pedroni’s seven tests, four are based on the dim ension "within" (intra) and three on the dimension "between" (inter). These two categories are based o n the null hypothesis of no co integration (non-stationary state of the estimated residuals). The distinction between the two categories is made on the level of the alternative hypothesis:

H " 8ρ ρ : 1∀i: within ρ : ρ∀i: between Pedroni has shown that under appropriate normalizat ions, based on functions of Brownian motion, each of the seven statistics follows a normal rule, cent red, and limited to N and T which are sufficiently important: z % μ√N √v → N0, 1 The results are shown in the following table: Table 2 : Pedroni’s cointegration tests International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 66 Rho-stat V-stat PP-stat Adf-stat Rho_stat 1 Pp _stat 1 Adf_stat 1 (y, L, k, RD) 0 .31934 1.72137 -1.09983 -1.80154 1.76313 -0.94537 -2.62449 Note : 1 : These are tests based on the dimension "BETWEEN" From the results found in this table, we notice the existence of a long-term relationship between the variables, that is to say, a co-integration. To car ry out tests of Co-integration on panel data and to obtain an estimate of the vectors of Co-integration, it is necessary to apply an effective method of estimati on.

Within this framework, we can distinguish several t echniques such as, the FMOLS method “least square completely modified” used by Pedroni, the DOLS meth od (dynamic least square), the GMM method “Method of Moments Generalized” and the ML “Maximum of Probability”. Philips and Moon (1999) showed that within the framework of the panel data, the FMOLS and DOLS techniques lead to asymptotically distributed estimators according to a reduced centred normal law. Similarly, Pedroni (1 996) affirms that MCO estimators “ordinary Least Square” are highly- convergent, whereas their asymptotic distributions are distorted and depend on the para meters of bad effects. According to Pedroni, these problems can be more marked in the presence of hete rogeneity. For this model, the estimation of these co-integral vectors by the FMOLS method and for the whole of the panel, is given by (T - Student between brackets):

G H I. K I. L I. M % % N. NL . N O. P Based on the estimator of Pedroni, the econometric specification of this model combines the use of the instrumental variables and the FMOLS method. The us e of the instrumental variables makes it possible to obtain consistent estimations. Thus, the estimate b y the GMM makes it possible to obtain a diagonal matrix of correlations. Table 3 : Correlation matrix LPIB LL LK RD LPIB 1.000000 0.954639 0.981909 0.898691 International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 67 LL 0.954639 1.000000 0.976842 0.944499 LK 0.981909 0.976842 1.000000 0.911354 RD 0.898691 0.944499 0.911354 1.000000 Yet, we use the Haussman Test (1978) which can be a pplied to many problems of specification in econometrics. However, its most widespread use is that of the tes ts of specification of the individualeffects as a panel. It is thus used to explain the fixed and random effec ts. Its general idea is at thesame time simple and brilliant. Let us actually suppose that we seek to test the possible presence ofa correlation or a defect of specification. Let us admit that we have two types of estimators for theparameters of the studied model:

the first estimator is supposed to be predetermined while the second is not. Consequently, all what we have to do is to compare a distance balanced by a m atrixof variance- covariance between the two estimators to be able to determine whether thespeci fication is correct or not. In fact, the technical use of this principle in this model makes itpossible to ob tain a matrix of variance -covariance having good properties. Table 4 : Variance- covariance Matrix C LL LK RD C 5.812536 -0.615414 0.025530 0.033289 LL -0.615530 0.065199 -0.002745 -0.003432 LK 0.025530 -0.002745 0.000157 438 E-05 RD 0.033289 -0.003432 438 E-05 0.000476 The econometrics of the panel makes it possible to control the heterogeneity of the observations intheir individual dimensions either by taking into accoun t a presumably certainspecific effect, “Fixedeffects”, or by considering a non observable specific effect (Radom effects). In this work, we examine the link between R& D and growth in various countries. On a sample of 15 countries during the period 1980 - 2009 , the results of the obtained estimates using the va rious methods instatic panel, show that the International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 68 impact of R& D on growth varies according to the in dicator of theinternal expenditure of research and development “DIRD” of each country taken in the sam ple. Onthe basis of this last indicator, it arises from the estimates that the increase in apoint of percen tageof this indicator is brought to 0.427 point of additional growth. The application of the tests of unitroot LL and IPS shows that the whole of the statistical series is affected of a unit root. The checking of the properties of non stationary state for all the variables of the panel leads us to Study the ex istence of a relation of long run between these var iables.

In other words, we study theexistence of a relation of co-integration by applying the co-integration tests of Pedroni. Moreover, we use the technique of the t ests of unit root in order to represent the estimated “residues” of the various countries included in our sample model. 4. CONCLUSION At the end of this research, the inventory of the m ain theoretical arguments and empirical results obtained reveal and confirm the importance of the R&D in the stimulation of growth. The drawn conclusions are also of a theoretical and empirica l nature. On the theoretical level, we have showed that the endogenous theory of growth recommends the loosening of certain neo-classicalassumptions to highlight other underly ing ones. Besides, the models worked out in this theory arecharacterized by a diversity of resources ; for instance, investment in physical and human capitalwithin the educational system “Lucas (1988)” , practical training (Learning by doing) “Romer (1990)”, the driving role of infrastructure and pu blic expenditure “R.Barro (1991)”, research and technological innovation “Aghion and Howitt (1998)” , etc. Moreover, this theory provides a well defined frame work of analysis for the research anddevelopment as a fundamental factor in the economic growth. Al l in all, the empirical steps that wehad adopted in this work, made it possible to meet our need to ev aluate to what extent the MENAcountries have profited from the R&D incorporated over the period 1980 -2009. The results areconvincing, indicating that there is a positive relationship between resea rch and development and growth. In fact, this empirical study of the MENA countries has brought i mportant results: The existence of a positive International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 69 relationship between research and development and e conomic growth inthese countries. Indeed, the volume of expenditure of R&D acts on the speed of convergencebetween the countries. So, the research and development constitutes a crucial fac tor forrecovery. Therefore, stocks of R&D are necessary to adapt the advantages of the technolog ical diffusion. Moreover, on the level of the developed capital, a policy of research and development and adynamic specialization constitute the principal determinan ts of recovery. Besides, the MENA countries must profit from their closeness to Europe to increase their exports and attract the FDE (Foreign Direct Investment) inorder to increase their creative cap acities and improve their speed of convergence compared to thedeveloped countries. REFERENCES Aghion, P. Howitt, P. (1998).Endogenous economic gr owth, MIT Press, Cambridge, Mass Barro, R. (1991). Economic growth in a cross sectio n countries’, Quarterly Journal of Ecomics, Vol. 106, pp. 407-443. Chantal, K. Cincera, M. (2001). Determinants of inn ovation activities of firms in has Newly Industrialized Country: Year econometric analysis w ith firm-level dated for the Turkish manufacturing sector', Economy and Forecast, vol. 150-151, pp. 1 39-158. Christophe, H. (2000).The econometrics of the data of panel; Linear models simples', Doctoral School, Methodological Seminar. Dosi, G. (1988). The nature of the innovative proce ss, Technical change and economic theory, London:

Printer Publishers. Eaton, J and Kortun, S. (1996).Engines of Growth: D omestic and Foreign sources of innovation, NBER Working paper, N°5207. Griliches, Z. (1979). Issues in Assessing the Contr ibution of Research and Development to Productivit y Growth’, Bell Journal of Economics, 10: 92-116. Grossman, GM. and Helpman, E. (1991). Trade, Knowle dge Spillovers, and Growth, European Economic Review, Vol. 35, pp. 517-26. Haussman, JA. (1998).Specification Tests in Econome trics, Econometrica, Vol. 46, pp. 1251- 1271. Frascati. (1993).Méthode type proposée pour les enq uêtes sur la recherche et développement expérimental, OCDE, Paris. International Journal of Information, Business and Management, Vol. 7, No.3, 2015 ISSN 2076-9202 70 Lucas, RE. (1988).On the mechanics of economic deve lopment, Journal of MonetaryEconomics, Vol. 22, pp.3-42. Levin, A. Lin, C. ET Chu, CJ. (2002) .Unit roots te sts in panel data: Asymptotic and nite – sample properties ’, Journal of Econometrics, Vol 108, pp . 1-24. OCDE, (2004b). The internationalisation of industri al R & D: policy issues and measurement problems’, document de travail, T. Hatzichronoglou. Pedroni, P. (1999). Critical values for cointègrati on Tests in Heterogeneous panels with Multiple Regressors, Oxford Bulletin of Economics and Stati stics, Vol.61, pp.653 -670. Pedroni, P. (2004).Panel cointègration: Asymptotic and Finite Sample properties of Pooled Time Series Tests with an Application to the PPP Hypothesis, Ec onometric Theory, Vol .3, pp.579-625.

Philips, PCB and Moon, H. (1999).Linear Regression Limit Theory for Non Stationary Panel Data, Econometrica, Vol. 67, pp.1057-1111. Romer, PM. (1986). Increasing returns and long-run growth, Journal of Political Economy, Vol.94, pp.

1002 – 37. Romer, P. (1990). Endogenous technological change, Journal of Political Economy, Vol. 98, pp. 71-120. Sadraoui, T. Zina, NB. (2007).A dynamic Panel data analysis for R&D Cooperation and economic growth, Int .J.Foresight. And innovation policy, MP RA PAPER N°.3415, Posted 7. Solow, RM. (1956).A contribution to the theory of e conomic growth, Quarterly Journal of Economics, Vol.70, pp. 65-94. Solow, RM. (1957).Technical change and the aggregat e production function’, Review of Economic Studies, Vol. 73, pp. 217-35. R epro duce d w ith p erm is sio n o f th e c o pyrig ht o w ner. F urth er r e pro ductio n p ro hib ite d w ith out p erm is sio n.