Introduction

It is widely acknowledged that invention plays a important function in houses ‘ endurance and growing. Traditionally, invention has been considered one of the major beginnings of growing and competitory advantage, both for states and houses. From the “ originative devastation ” of Schumpeter, when the key participant was represented by the enterpriser who, with his action, could prolong the long-run economic growing ( Schumpeter, 1942 ) , to our recent R & A ; D laboratories, research centres, etc. , invention has ever played an highly relevant function for growing. While at the beginning of economic theory, the function of houses was considered polar, nowadays houses ‘ engagement is a cardinal component for invention. As stated above, the issue is highly relevant non merely for houses, but besides at a macro-level, therefore sing states. In fact, “ the capableness to introduce and to convey invention successfully to market will be a important determiner of the planetary fight of states over the coming decennary. There is turning awareness among policymakers that advanced activity is the chief driver of economic advancement and wellbeing every bit good as a possible factor in run intoing planetary challenges ” ( OECD, 2007, p.3 ) . Furthermore, invention is demonstrated to be critical for endurance and growing of houses, even more in the high competitory planetary environment that houses are now compelled to face. Given the importance of the topic, a whole apprehension of the factors that systematically influence invention is highly of import.

Historically, certain factors have ever been considered to positively impact on houses ‘ invention: for case, size is regarded as a beginning of invention ( e.g. Ebling & A ; Janz, 1999 ; Ganotakis & A ; Love, 2011 ) because greater houses are supposed to have both the fiscal and cognition resources necessary to put in invention. Among the characteristics that might positively act upon invention, exporting has been late brought to attending. While traditionally, exporting has merely been considered an international enlargement scheme, even if the most used one, recently, the positive effects that exports exert on invention have been repeatedly investigated by bookmans. Apart from the relationship with invention, one of the grounds behind the great attending to exports prevarications in the great diffusion of these activities among houses. In fact, as a effect of the altering kineticss of the environment, besides the little and average endeavors ( SMEs ) are recognizing that they can no longer believe international competition will non impact them because they are little or entirely focused on the domestic market. Therefore, SMEs are increasing the exportation activities to last, and hopefully turn, in this ambitious environment. As an illustration, in the information that will be described shortly, more than 60 % of the houses in the Spanish fabrication sector export. All that, has brought to an increased attending to exporting from bookmans.

Furthermore, as anticipated above, presents it is widely shared the position harmonizing to which exports have positive effects on invention. Several researches have demonstrated that exporters have higher advanced productiveness compared to non-exporters. At the same clip, surveies have besides revealed that invention exert positive influence on exporting. The more a house innovates, the higher the export strength ( e.g. Kongomanila & A ; Takahashi, 2009 ; Roper & A ; Love, 2002 ) and the likeliness that house exports ( e.g. Bleaney & A ; Wakelin, 2002 ; Caldera, 2010 ) . Therefore, while before exporting and invention were considered alternate schemes for growing, now they are believed to reenforce each other. Indeed, research have unveiled that the returns in footings of growing of one activity is higher if the house besides engages in the other ( Golovko & A ; Valentini, 2011 ; Ito & A ; Lechevalier, 2010 ) . Invention may drive to an betterment of the bing merchandises, sold in the domestic market, therefore increasing gross revenues, but besides, it may take houses to come in in foreign market with new or modified merchandises. Conversely, runing abroad, exporting houses may come to cognize diverse and improved cognition that, one time incorporated, might be used in the domestic market to bring forth better invention. These constructs are practically embedded in research heads. Yet, despite the great sum of surveies aimed at look intoing this relationship, bookmans ‘ attending has non been focused on the mechanism underlying acquisition by exporting. Although it is widely acknowledged that export exerts positive influence on invention, supposedly driven by information exchange from foreign market, through export mediators or straight from clients ( Salomon & A ; Shaver, 2005a ) , few surveies have tried to understand how houses learn from international trade. Namely, some hypotheses have been advanced but instead unsuccessfully tested. Just to advert the chief one, MacGarvie ( 2006 ) tried to verify whether informal communicating with foreign opposite numbers is the channel through which cognition is obtained, therefore exercising positive influence on invention. Since small grounds was found in favour of the hypothesis, how exporters learn from international trade remains unknown. Clearly, extricating this relationship, therefore understanding how houses, but besides policymakers, can further invention, might be highly relevant for the development both of houses and of states.

In this survey, I hypothesize that exporting houses improve their advanced public presentations through formal engineering transportation understanding. I try to verify whether the improved cognition that houses can obtain through formal understandings fosters the figure of merchandise invention that the house obtains and the figure of patents that the house registries. Therefore, a balanced panel of Spanish fabrication houses, composed by 10,205 firm-years observations, is used to mensurate the impacts of the purchase of licences, engineering and proficient aid abroad on merchandise invention and patent enrollment counts. The consequences confirm that, 1 twelvemonth after the judicial admission of a formal engineering understanding, houses significantly improve their patent enrollment ; moreover, if considered with farther slowdown, the purchase of licence, engineering and proficient aid abroad appears to act upon patent enrollment, although in this instance, the grounds is significantly less strong. On the contrary, merchandise invention does non look to be affected by the variable of involvement, independently of the slowdowns considered.

In the following subdivision, literature linked to the issue is reviewed, followed by the hypotheses of the research. In subdivisions 3, 4, 5 and 6 informations are briefly described and descriptive statistics are provided. In the subsequent subdivisions, the method employed is presented, followed by the consequences of the empirical analysis and extra analyses performed. The concluding subdivision concludes.

Literature reappraisal

The relationship between export, invention and productiveness has been unusually interesting for bookmans. During last decennaries, several researches have been made to extricate the complex relationship among these variables. The purpose of this survey is to travel into more deepness in understanding how learning by exporting takes topographic point, but, before making it, an overview on the old literature ‘s parts is provided.

Self-selection and determiners of export

Until late, most of the literature has chiefly focused on the nexus between productiveness and exportation. First surveies tested the hypothesis that most productive houses self-select into export markets. Self-selection is now embedded into bookmans ‘ heads. Research showed that merely the most productive houses engage in export activities because they are able to confront the high sunk costs of entry and the fiercer competition of international markets ( Bernard & A ; Jensen, 1999 ; Ganotakis & A ; Love, 2011 ) . Hence, a great portion of the positive correlativity between productiveness and exportation is explained by self-selection. Not merely future exporters show higher productiveness and lower variable costs prior to exporting ( Clerides, Lach & A ; Tybout, 1998 ) , but besides houses with higher productiveness have higher returns on exporting in general, and on exporting to advanced states in peculiar ( Trofimenko, 2008 ) . Furthermore, the greater the productiveness works, the greater the fringy benefits of both exporting and R & A ; D activities, therefore increasing the consequence of self-selection ( Aw, Roberts & A ; Xu, 2010 ) .

It is of import to observe that surveies sing determiners of exporting have revealed that several variables, apart from invention and productiveness, influence exporting. Although size is considered to exercise positive influence on exportation, sentiments of the empirical literature are non consentaneous. Ebling and Janz ( 1999 ) , for case, found that steadfast size does non impact exports immediately, but merely indirectly, through invention ; look intoing Gallic houses in biotechnology, Pla-Barber and Alegre ( 2007 ) revealed that size does non impact on export public presentation ; these findings are explained by the characteristics of the industry, where economic systems of graduated table and production efficiency are non peculiarly relevant compared to the competency to develop up-to-date engineerings. As for the presence of foreign equity, alternatively, and as expected by most of the literature, Wignaraja ( 2008 ) , look intoing a sample of Asiatic houses, revealed that the presence of foreign equity has a strong positive consequence on export public presentation. Several factors resulted positively correlated to export such as instruction and good qualified directors ( Braymen, Briggs & A ; Boulware, 2010 ; Higon & A ; Driffield, 2011 ) , antiphonal selling organisation and client focused patterns ( Calantone, Cavusgil, Schmidt & A ; Shin, 2004 ) and labour accomplishments ( Alvarez, 2007 ) . Besides the variables already illustrated, invention is considered to be correlated to export, although the relationship between these two variables is rather more complex.

The function of invention and the relationship with export

Invention is recognized as one of the major beginnings of heterogeneousness among houses, although the relationship between export and invention is still investigated by research workers. An invention is defined as “ the execution of a new or significantly improved merchandise ( good or service ) , or procedure, a new selling method, or a new organisational method in concern patterns, workplace organisation or external dealingss ” ( Eurostat & A ; OECD, 2005, p.46 ) . Traditionally, empirical literature regarded R & A ; D and invention as interchangeable, therefore sing the relationship between R & A ; D and exporting instead than between invention and exportation ( Aw et al. , 2010 ; Barrios, Gorg & A ; Strobl, 2003 ; Braymen et al. , 2010 ) . The importance of invention end product has been late brought into focal point by research workers that began to utilize different variables to mensurate invention, such as merchandise and procedure invention ( e.g. Caldera, 2010 ; Cassiman, Golovko & A ; Martinez-Ros, 2010 ; Damijan, Kostevc & A ; Polanec, 2010 ; Higon & A ; Dreffield, 2011 ; Van Beveren & A ; Vandenbussche, 2009 ) and patent application ( Salomon & A ; Jin, 2008, 2010 ; Salomon & A ; Shaver, 2005a, B ) .

Conventionally, two different theoretical accounts have emerged to depict the relationship between invention and exportation: on one side, product-cycles theoretical accounts claim that invention is one of the drive forces of industrialised states ‘ exports ( Vernon, 1966 ; Krugman, 1979 ) . The nucleus thought is that merchandises follow rhythms and developed states innovate to export to less developed states. Once the development states are able to retroflex the invention by themselves, the initial exporters are forced to introduce once more in order to go on exporting. On the other side, endogenous growing theoretical accounts ( Grossman & A ; Helpman, 1991a ; B ) assert that export can be an impulse for invention for two distinguishable grounds: foremost, higher competition in foreign markets leads exporters to put in R & A ; D to better merchandises and procedures in order to be competitory ; moreover, export helps houses in deriving entree to more developed cognition and engineerings, therefore exciting the productiveness.

Therefore, product-cycles theoretical accounts claim that invention has an impact on exporting and, in the aftermath of these theories, bookmans have tried to through empirical observation turn out the effects of invention on export.

The effects of invention on export

Large portion of the literature agrees on the positive effects of invention on export. It is more likely that a house engages in exporting when it has antecedently undertaken advanced activities ( Ganotakis & A ; Love, 2011 ; Kongomanila & A ; Takahashi, 2009 ; Halpern & A ; MurakO§zy, 2012 ) . Basile ( 2001 ) , analyzing over 4,000 Italian fabrication houses, reaches the same decision, adding interesting characteristics on the impacts of exchange rate fluctuations on export. Treating invention as exogenic, the writer claims merchandise invention schemes have strong positive effects on export merely after a devaluation of the currency. In fact, during a devaluation of the currency, non-innovative houses are able to come in the foreign market. Finally, non merely invention affects export determination, but it besides increases export continuance ( Chen, 2012 ) .

When sing different industries, the consequence of invention on exports can be rather different. A figure of surveies ( e.g. Barrios et al. , 2003 ; Bleaney & A ; Wakelin, 2002 ; Caldera, 2010 ; Lachenmaier & A ; Woessmann, 2004 ; Nguyen, Pham, Nguyen & A ; Nguyen, 2008 ; Roper & A ; Love, 2002 ) have demonstrated that invention exerts a positive influence on exporting in fabrication sectors. Researching this nexus in different states and periods, the writers agreed on the positive consequence of invention on export determination, claiming that the likeliness that houses are exporters is higher if they have had more inventions. Whereas literature place is on the same side in fabrication sector, in other industries surveies ‘ consequences are instead conflicting. Specifically, sing R & A ; D investings as a placeholder for invention of new houses, Braymen et Al. ( 2002 ) revealed that the relationship between R & A ; D and exports is non important for new houses that provide merely services to their clients. On the other manus, Ebling and Janz ( 1999 ) , analyzing more than 1,000 German houses runing in 7 different concern oriented service sectors, found that invention strongly influences export activities. Similarly, look intoing the consequence of engineering and invention on export in a science-based industry, Pla-Barber and Alegre ( 2007 ) unveiled strong positive impacts on export public presentation.

As stated above, the term invention applies both to merchandise and procedure invention. Refering the influence of different types of invention on export public presentation, bookmans ‘ sentiments are instead conflicting. Several research workers agree on the importance of merchandise invention in explicating houses ‘ export determination ( e.g. Cassiman & A ; Golovko, 2007 ; Cassiman et al. , 2010 ; Higon & A ; Driffield, 2011 ) and strength ( Roper & A ; Love, 2002 ) . Another portion of the literature believes that besides process invention is an of import determiner for export determination and public presentation, albeit its part is lower than merchandise invention ‘s ( Becker & A ; Egger, 2007 ; Caldera, 2010 ) . These findings are explained by the greater importance for export public presentation of distinction scheme, to accommodate merchandises to foreign states, compared to treat betterment, that leads to be economy. Therefore, besides merchandise version is presented as a positively correlative factor with export profitableness, even more when the house has a antiphonal selling organisation and client focused patterns ( Calantone et al. , 2004 ) . In a similar vena the part provided by Nguyen et Al. ( 2008 ) which have investigated 2,739 Vietnamese SMEs in the fabrication industry and, apart from merchandise invention and procedure invention, have considered a 3rd step of invention: merchandise alteration, dwelling in the betterment of bing merchandises. The writers found that besides alteration of bing merchandises is of import as a determiner of exporting activities.

Although the positive effects of invention on export have been widely proved, presents besides the position of the impacts of export on invention is rather shared.

The effects of export on invention: acquisition by exporting

As seen above, the empirical literature has historically focused on the positive impact invention has on exporting. Recently, inquiries have been raised on whether export influences invention. To understand this nexus, and sing invention as a step of acquisition, Salomon and Shaver ( 2005a ) have elaborated on an imbalanced panel of Spanish fabrication houses. The writers discovered that exporters increase their merchandise inventions after come ining into foreign markets, and in peculiar, this consequence is more marked two old ages subsequent to exporting. The writers justify these findings claiming that it takes clip to modify merchandises in order to run into clients ‘ demands. Similarly, the writers found a positive consequence of exporting on patent applications, although in this instance the consequence is stronger with further slowdowns since it takes more clip to integrate the technological cognition needed to register patents. Several more surveies confirmed the positive impact of export on procedure betterment, claiming that old experiences as medium-large exporter have positive effects on patent applications ( Salomon & A ; Jin, 2008 ; Salomon & A ; Jin, 2010 ) , on both procedure betterment and merchandise inventions ( Alvarez & A ; Robertson, 2004 ) , and on future productiveness growing because houses improve efficiency by exciting procedure invention ( Damijan et al. , 2010 ) . Furthermore, two elements appeared to be of import in this affair: finish of the market and technological standing of the industry to which the house belongs. With respect to market, Alvarez and Robertson ( 2004 ) showed that houses that export to developed states are more likely to introduce in new merchandises, new tools and reorganisation of production. Exporting to developing states, alternatively, increases the likeliness of prosecuting in merchandise design activities. Similarly, Trofimenko ( 2008 ) , look intoing how exporting to more developed states affects productiveness additions of Colombian fabrication houses, asserts that the quality of the acquisition environment affairs, therefore exporting to more developed states increases future productiveness additions.

As for the 2nd issue, Salomon and Jin ( 2008 ; 2010 ) divided the sectors to which Spanish fabrication houses belonged into technological leaders or dawdlers ; consequences suggested that technological standing of industries significantly influences the sum of cognition that flows from the host environment to the house. In fact, houses belonging to technologically dawdling industries increase their patent activity disproportionately comparative to houses in technologically advanced sectors ( Salomon & A ; Jin, 2008 ) , albeit the benefits gained do non allow them to shut the spread with their technological leaders opposite numbers ( Salomon & A ; Jin, 2010 ) .

In line with acquisition by exporting, an interesting survey made by Aw and Batra ( 1998 ) has sought to analyse the nexus between efficiency and house investings in engineering and exports. To gauge the proficient efficiency of houses by investings in engineering every bit good as by export, the writers used stochastic production frontier techniques on a big sample of Chinese fabrication houses runing in several different industries. The writers stress the importance of technological heterogeneousness among houses to understand the relationship between exports and house productiveness. Basically, for houses that to a great extent and officially put in R & A ; D and engineering, normally big, with skilled labour and foreign capital, efficiency is non related to export orientation, while for little and locally owned houses, which do non officially put in R & A ; D, export is strongly linked to higher future productiveness. Therefore, research workers justify these findings with the prejudice that SMEs have of geting foreign engineering through farm outing agreements with foreign houses, and with the consideration of exporting as a channel through which engineering and cognition are passed from foreign markets and incorporated by exporters.

Part of the literature showed how of import is the acquisition by exporting consequence for houses but, while self-selection hypothesis has been proved by immense part of the literature, the grounds on the acquisition by exporting has been less distinct. A turning organic structure of the empirical work has sought to analyse the impact export has on future productiveness. Although domestic houses start exporting when holding a strength on the domestic market ( Salomon & A ; Shaver, 2005b ) , there is besides a strong grounds in favour of positive effects export has on productiveness additions ( Van Biesebroeck, 2005 ) . Harmonizing to De Loecker ( 2007 ) , houses that enter in foreign markets become on mean 8.8 % more productive than non-exporters. Investigating the consequence of the Free Trade Agreement between Canada and USA, Lileeva and Trefler ( 2010 ) confirmed that non-exporters that began to export to USA, because of the improved entree to foreign market, showed labour productiveness additions. Furthermore, non merely exports conveying a diminution in fringy costs and increase efficiency and capacity use, but besides have positive spill-over effects on the mark-up of the domestic markets ( Bughin, 1996 ) .

Until now, in all the surveies mentioned, the writers entirely examined the relationship between export and invention within the same house. However, the impacts that other houses ‘ invention or export might hold, in footings of spillover effects, should non be underestimated.

The function of R & A ; D and export spillovers

While there is a reasonably big literature look intoing the effects of invention and R & A ; D on exports, fewer bookmans have examined R & A ; D and other houses ‘ export spillovers. The concentration of exporting houses and transnational houses has positive effects on export public presentation, therefore export spillover effects are of import for export success ( Alvarez, 2007 ) . Furthermore, R & A ; D spillovers coming from multinationals of the same sectors have positive impacts on export ratios, both of domestic and of foreign houses, and the consequence is higher if the house export to a developed state ( Barrios et al. , 2003 ) .

Despite the big sum of research aimed at understanding the relationship between invention and exportation, few surveies have hypothesized a positive mutuality between exports and invention.

Complementarity of invention and export

Interactions between invention and export schemes have ever been really interesting ; nevertheless, surveies treated the two schemes as stray from each other, analyzing the effects that each has on productiveness as if it were independent from those of the other. Few surveies shed some new visible radiations on the possible interactions between invention and export. In peculiar, the two schemes are revealed to be complementary and reenforcing each other. The fringy part to SMEs ‘ gross revenues growing of invention is higher if export is besides in topographic point, and frailty versa. While old research demonstrated that the two activities contribute to houses ‘ growing when undertaken individually, Golovko and Valentini ( 2011 ) unveiled the mechanism that may underlie the reenforcing rhythm between export and invention and supply empirical grounds in support for the complementarity. Hence, public presentation betterment do non merely come from the optimisation of one activity, but directors should besides see interactions between them since those contribute the most to houses ‘ public presentation betterment. On the same page Ito and Lechevalier ( 2010 ) , which found that determinations sing invention and export strongly affect public presentation and endurance of houses ; the writers confirmed the complementarity between the schemes and highlighted that the best performing artists in Nipponese fabrication houses were those that exported and innovated.

While a portion of the literature has investigated acquisition by exporting, chiefly reasoning in favour of it, few parts have tried to grok how the phenomenon takes topographic point.

Mechanism of larning by exporting and research hypotheses

Since exportation is progressively being recognized as an of import beginning of acquisition, farther stairss consist in understanding how the acquisition takes topographic point. First ideas occurred to research workers ‘ heads are that, through informal communicating with foreign buyers, a big sum of cognition flows from the opposite number to the house. Therefore, cognition will be incorporated by the house which would utilize it to modify merchandises, better production procedures or use for new patent enrollment ( Salomon & A ; Shaver, 2005a ) . Although this appeared to be rather likely, efforts in demoing it through empirical observation failed. Indeed, MacGarvie ( 2006 ) , utilizing patent commendations as a step of cognition diffusion, tried to find whether exporting increases the likeliness of houses ‘ commendations to foreign patents every bit good as foreign commendations to the houses ‘ patents. Surprisingly, the writer found that, although exporters gain entree to foreign engineering and cognition by pass oning with foreign purchasers, cognition diffusion through these channels is non good measured by patent commendations, and exporting is non statistically important. Conversely, importing houses ‘ patents are significantly more likely to be influenced by engineering in the exporting state than are the patents of the houses that do non import from that state. Therefore the inquiry sing how learning by exporting takes topographic point is still unreciprocated.

Finally, literature agrees that exporters are better compared to non-exporters ; on norm, they produce more than twice the end product of non-exporters, are more productive and pay higher rewards ( Bernard & A ; Jensen, 1999 ) , partially due to self-selection and partially to larning by exporting. But how does the acquisition by exporting work? What are the mechanisms through which exporters incorporate new cognition and new engineerings? Aw and Batra ( 1998 ) stated that export activity is an of import tool through which new or improved engineering is passed from foreign buyer. MacGarvie ( 2006 ) has so unsuccessfully tried to verify whether informal communicating with foreign purchasers allowed the transportation of cognition and engineering from abroad. Therefore, in literature there still is a null sing the procedure through which export exerts the positive influence on invention.

The present survey is aimed at make fulling this bing nothingness. If informal communicating is non a channel through which engineering and cognition are transferred from foreign opposite numbers, it can be hypothesized that formal engineering transportation understandings play the of import function of channel through which new cognition and engineerings are incorporated by the house. In that sense, if a house stipulates a formal engineering understanding in a given twelvemonth, purchasing licence, proficient aid and know-how from abroad, the advanced public presentation of that house should better in the undermentioned periods. Hence, the purpose of this paper is to confirm the undermentioned hypotheses:

Hypothesis 1: A formal engineering transportation understanding in a given period has positive effects on the figure of merchandise inventions in the undermentioned periods.

Hypothesis 2: A formal engineering transportation understanding in a given period has positive effects on the figure patents registered in the undermentioned periods.

As seen above, several variables are demonstrated to act upon advanced productiveness. For this ground, other elements are included in the theoretical accounts, in order to command for the positive effects they might hold on invention. In peculiar, the export and import position, the engagement of foreign stockholders in houses ‘ equity, investings in R & A ; D and size will function as control variables.

Datas

The informations employed in this survey are from a annual study conducted by Fundacion Empresa Publica with the fiscal support of the Spanish Ministry of Industry. The Fundacion surveys a sample of Spanish fabrication houses in order to bring forth an overall image of the state ‘s fabrication sector. The dataset is a balanced panel of 10,205 firm-year observations spread on 13 old ages, from 1990 to 2002. The sample contains 6.3 % observations of houses with no more than 10 employees, 62.4 % with the figure of employees included between 10 and 200 and 31.3 % with at least 200 employees. In all, the dataset is composed by 785 houses, spread in 20 sectors ; in item, Table 1 presents the industry dislocation of the sample used.

Table 2 displays the per centum of firm-year observations in each sector that undertook the interesting activities during the sample period. Stipulation of engineering transportation understandings is non really common ; in fact, on norm, merely 11.4 % of the houses stipulated engineering understandings, although in sectors such as chemicals, motor vehicles and other conveyance equipment, the portion exceeds 25 % . Clearly, sectors with the lowest portion are those where engineering is non a cardinal component ( e.g. meat and wood ) .

Table 3 provides a snapshot of the distribution of firm-year observations across old ages, foregrounding the per centum of houses that transferred engineering and the position as exporter, pioneer and importer. Both export and import activities are undertaken by more than 60 % of the observations and demo a turning tendency across old ages, corroborating the gap of the Spanish economic system to the international markets.

As respects invention, on norm, less than 30 % of the observations obtained merchandise inventions and less than 10 % registered a patent. As stated above, non many houses stipulated engineering transportation understandings and, across the old ages, the per centum of houses has ne’er exceeded 12.2 % .

Dependent variables

As antecedently stated, this survey focuses on understanding the impacts that formal engineering transportation understandings have on houses ‘ invention public presentations. To extricate this relationship, two dependent variables are used: merchandise invention counts and patent enrollment counts. Using two different variables may be helpful in holding a more comprehensive image of the state of affairs as each variable might pick up different facets of invention ( Salomon & A ; Shaver, 2005a ) .

Product Innovation counts

The most direct manner to measure invention public presentation is to straight mensurate it. Data provided by the Fundacion Empresa Publica include non merely information on invention inputs ( i.e. R & A ; D investings ) , but besides on its end products. Indeed, the dataset reports a count variable of the figure of merchandise inventions realized in a given twelvemonth. The maximal value that this variable takes is 950, highly high compared to the mean of 3.15. Albeit this figure may look disproportional compared to the remainder of the dataset, the house that reported this figure has, over clip, really comparable Numberss to this value[ 1 ]. The count variable stand foring merchandise inventions is labeled “ Merchandise Inventions ” .

Patent Registration counts

Another manner to measure invention public presentations is through patent enrollments. A portion of the literature considers patent applications as a step of invention, claiming that if a house decides to confront the dearly-won patent application procedure, it likely believes that it has innovated and it is worthwhile to prosecute patent protection ( Salomon & A ; Jin, 2008, 2010 ; Salomon & A ; Shaver, 2005a, B ) . On the reverse, since the dataset provides information on the figure of patent registered in a period, and the variable surely represents an end product of advanced productiveness, this survey uses a count variable of patent enrollments. This variable is labeled “ Patent Registrations ” .

Independent variables

In order to measure the veracity of the hypotheses, a figure of variables are portion of the theoretical account. Specifically, apart from the variable of involvement stand foring engineering understandings, other variables are besides used to command for elements that affect invention.

Technology understandings

The study collected information on whether a house stipulated formal engineering transportation understandings, or non, during a twelvemonth. Technology transportation is the procedure by which a engineering, expertise know-how or installations developed by one person, endeavor or organisation is transferred to another person, endeavor or organisation ( WIPO, p.1 ) . In peculiar, this survey is focused on the understandings that officially transfer engineering to houses. Depending on the nature of the engineering, different understandings might be used to reassign engineering, such as the sale or the assignment of IP rights, a licence contract, the purchase of proficient aid, etc. The Fundacion Empresa Publica provides informations on the purchase of proficient aid and licences from abroad made by the houses. This is represented by a silent person variable that takes the value 1 if the house bought proficient aid or licence in a given period and, 0 otherwise. This variable is labeled “ Technology Agreements ” .

Since cognition embodied in engineering understandings may take clip to be incorporated by the house and to ensue in an invention or a patent, a 1-year lagged variable of the engineering understanding position is created.

Control variables

Export is demonstrated to positively act upon houses ‘ invention public presentations ( e.g. Alvarez & A ; Robertson, 2004 ; Salomon & A ; Jin, 2008 ; Damijan et al. , 2010 ) ; hence, it is of import to command whether a steadfast exports or does non. For this intent, a silent person variable refering the export position is used and labeled “ Export Status ” ; “ Export Status ” takes the value of 1 if exporting activities are reported in the given period, 0 otherwise. Since Salomon and Shaver ( 2005a ) demonstrated that the effects of export on invention are more marked with slowdowns, a lagged variable of export is besides used.

Similarly, keeping an importing relationships with foreign opposite numbers, may increase cognition flows and, therefore, have positive impacts on invention. In this sense, the usage of a silent person variable, sing the import position of houses in the given period, helps in commanding for import influences on invention. This variable is labeled “ Import Status ” .

The presence of foreign portions in houses ‘ equity is positively associated with an addition of the chance of invention ( Alvarez & A ; Robertson, 2004 ; Basile, 2001 ) ; hence, to command for the presence of the foreign equity, a variable expressed as a per centum of the foreign capital is included in the theoretical account. The variable is labeled “ Foreign Capital ” .

Size is considered a factor that positively influences advanced productiveness ( Ebling & A ; Janz, 1999 ) . Since pioneers are by and large larger than non-innovators, the natural logarithm of the figure of employees is used as a control variable. The usage of figure of employees, alternatively of gross revenues, to command size is due to the positive consequence invention might hold on gross revenues: viz. , a house that produces many inventions in a given twelvemonth is besides expected to increase gross revenues. Therefore, figure of employees is used. This variable is labeled “ Size ” .

Finally, big portion of the literature besides considers positive effects of R & A ; D inputs on invention ( e.g. Salomon & A ; Shaver, 2005b ) . To command for these effects, an R & A ; D strength variable is included in the theoretical account. “ R & A ; D Intensity ” is defined as R & A ; D investings normalized on entire gross revenues, expressed as a per centum. Since by and large, R & A ; D outgos are extremely correlated with size, the standardization of R & A ; D on gross revenues allows commanding for effects related to size.

Descriptive statistics and correlativities

Table 4 and Table 5 present descriptive statistics and correlativity matrix for the variables used in the theoretical account. First of wholly, it is of import to observe that houses that stipulate formal engineering understandings are more productive in footings of invention. In fact, for those houses, both the Numberss of merchandise inventions and patent enrollments, on norm, are higher compared to other houses ‘ . Furthermore, analyzing Table 4, it is clear how, on norm, those houses are more orientated to export and import activities, have a larger portion of foreign capital, put more in R & A ; D activities and hold more employees.

In figure 1 and figure 2, observations are divided based on whether they stipulated engineering understandings or non, in order to compare advanced public presentations of the groups. On Y axes of the graphs, the mean figure of merchandise inventions and patent enrollments for each twelvemonth is shown. Firms that stipulated engineering transportation understandings invariably outperformed the other group, both in footings of merchandise inventions and patent enrollments ; nevertheless, it is necessary to foreground that, albeit interesting, these graphs do non state much about the way of the causality of the relationship.

Table 5 displays the correlativity matrix for the full sample. Although most of the

correlativities are as expected, some deserve peculiar attending. First of wholly, the strength of the correlativity between engineering understandings and patent enrollments is higher than that between engineering understandings and merchandise inventions. Second, and non surprisingly, invention and exporting are positively correlated every bit good as export position with size, R & A ; D expenditures and foreign capital. Furthermore,

the variable stand foring engineering understandings is strongly positively correlated with its 1-year slowdown, therefore connoting that if a house officially transferred engineering from abroad in a period, it is likely that it will besides make it in the undermentioned twelvemonth ; eventually, it is deserving to advert the negative correlativity between foreign capital and merchandise invention,

likely due to inclination of international houses to develop invention in the headquarter alternatively of in subordinates. Although interesting, Table 5 consequences do non state much about the causality of the correlativities ; hence, it is necessary to turn to multivariate analysis to better understand the nature of the relationships and show the hypotheses.

Statistical attack

The attack taken to measure engineering understandings ‘ effects on invention is to regress the steps of invention on the lagged values of engineering understandings. The dependent variables used are both count variables ( figure of merchandise inventions and figure of patent enrollments ) that can merely take nonnegative whole number values. Furthermore, many observations are close or equal to 0. In these instances, when informations are strongly skewed to the right, the OLS arrested developments are non appropriate. Therefore, a Poisson appraisal process is considered ; nevertheless, in Poisson distribution the mean and the discrepancy are the same, while the distributions of the dependent variables display marks of overdispersion, with standard divergences much greater than agencies. In instance of overdispersion, a Negative Binomial theoretical account is more appropriate. Hence, at first, appraisal with random effects are run ; so, to get the better of the job of the uncorrelation of the unseen effects with the regressors, fixed effects appraisals are besides used.

Subsequently, to corroborate the consequences of the Negative Binomial arrested developments, the dependent variables are transformed in their natural logarithms, in order to execute other appraisals. Again, both random and fixed effects theoretical accounts will be used.

Consequences

Table 6 and 7 present the appraisal consequences of the Negative Binomial arrested development theoretical account. In peculiar in ( a ) columns the consequences of random effects specifications are showed while ( B ) columns show fixed effects ‘ . All the specifications include twelvemonth and sector silent persons ; yet, consequences are non presented in order to concentrate on the variables of involvement.

Table 6 nowadayss the consequences from regressing the Product Innovation count variable on Technology Agreements. In columns 1, the theoretical account besides includes 1-year lagged values both of the variable of involvement and of the Export Status. Technology understandings coefficients did non ensue statistically important either in column 1 ( a ) , or in column 1 ( B ) . It is interesting to observe that, in the 2nd specification, besides export coefficients resulted non important, although the strong belief of a positive consequence of export on invention is well-established in the literature. These unexpected consequences might be explained either by the low discrepancy of the variables, which is absorbed by fixed effects, or by the multicollinearity between the export variable and its lagged value[ 2 ]. In columns 2, the variable of involvement and Export Status are merely considered in the instantly old twelvemonth. Again, the coefficient estimations of Technology Agreements in these specifications are positive but statistically non important. On the contrary, and as expected, Export Status ‘ coefficients resulted significantly positive in both random and fixed effects theoretical accounts, at the 1 % and the 5 % significance degrees, severally. In columns 3, the variable of involvement is replaced by its 1-year slowdown, to verify whether, increasing the clip interval from the judicial admission of the understanding to the merchandise invention, the variable of involvement additions statistical relevancy in act uponing the figure of merchandise inventions. Although this seemed sensible, because larger sum of clip might be needed to integrate the engineering and transform it into a merchandise invention, the coefficient of the variable is still statistically non different from nothing. As respects exporting alternatively, the 1-year slowdown of the Export Status variable resulted positively important in both specifications. To sum up, appraisals in Table 6 suggest that the purchase of foreign licence and proficient aid from abroad does non act upon the figure of merchandise inventions that houses obtain, neither in the undermentioned twelvemonth, nor after two old ages. In line with anterior literature, the control variables significantly positively influence invention public presentation, except for the foreign capital engagement whose coefficient is non important in any of the specifications. In peculiar, R & A ; D investings coefficients are positive and statistically important at the 1 % significance degree corroborating that greater investings in R & A ; D boost advanced productiveness. Similarly, size of houses has positive impacts on merchandise inventions.

In Table 7 a different step of invention is used: Patents Registrations. Table 7 displays the consequences of the arrested development of Patent Registrations on Technology Agreements. As for the old arrested development, the coefficients of the variable of involvement in columns 1 are non statistically important in any of the twelvemonth considered. Therefore, when sing both the old ages before the enrollment of the patents, formal engineering transportation understandings do non act upon patent enrollments. On the contrary, being an exporter significantly positively influences the figure of patents registered, systematically with literature ‘s position. In columns 2, the judicial admission of engineering understandings is considered entirely in the twelvemonth instantly predating patent enrollment. In column 2 ( a ) , a random effects arrested development is performed, gauging a positive and statistically important coefficient for Technology Agreements at the 10 % significance degree. To corroborate this interesting determination, a fixed effects appraisal is run. The coefficient estimation remains positive and important, therefore connoting that the formal purchase of engineering from abroad straight positively influences the enrollment of patents in the undermentioned period. Traveling to columns 3, the variable of involvement is substituted by the its 1-year slowdown, in order to understand whether, sing a longer clip interval, engineering understandings are still important in act uponing the figure of patents registered. In this instance, the coefficient of Technology Agreements did non ensue important either in the random effects, or in the fixed effects theoretical accounts.

Findingss sing exports are consistent with those of anterior surveies. Indeed, Export Status in the old twelvemonth resulted positive and important in every specification, corroborating the positive influence of exports on patent enrollments ( Salomon & A ; Jin, 2008 ) . Furthermore, the coefficients of random effects specifications resulted larger both in magnitude and significance. Columns 3, entirely includes a 1-year slowdown value of Export Status: while the variable resulted significantly positive in the random effects specification, in the fixed effects ‘ the coefficient is non statistically important. Similarly, Import Status coefficients appeared positive and important merely in the random effects columns, while in fixed effects, albeit positive, they are non important. These findings might be explained by the low discrepancy of the variables within the groups, since it can be assumed that exporters and importers tend to keep the position across old ages. With regard to the control variables, Size and R & A ; D investings positively influence the enrollment of patents in every specification at 1 % significance degree, systematically with big portion of the literature. Finally, a brief focal point on foreign capital coefficients is needed: harmonizing to the estimations, the engagement of foreign spouses negatively influences the figure of patent registered, although the magnitude of the coefficient is close to nothing. As awaited above, this may ensue from the international organisations ‘ disposition to develop invention in the central offices and non in the subordinates, therefore registering the patents in the state where the headquarter is located.

Wholly, the consequences presented in Table 6 and Table 7 provide grounds that the formal purchase of engineering from abroad in a given twelvemonth does non hold any positive influence on the merchandise inventions of the undermentioned twelvemonth ; hence, there is non statistical grounds in support of the first hypothesis. Conversely, the enrollment of patents is positively influenced by formal purchase of engineering, therefore corroborating the 2nd hypothesis.

To confirm the findings of the Negative Binomial appraisal theoretical account, the dependent variables are transformed into their natural logarithms, in order to execute arrested developments to verify the same hypotheses. Table 8 and 9 show the consequences of the appraisals. Again, twelvemonth and sector effects are included in the random effects specifications while, in the fixed effects theoretical account, sector effects are excluded because their values do non alter during the sample period, therefore bring forthing multicollinearity and being omitted by the fixed effects.

The consequences of the arrested development of the natural logarithm of merchandise invention counts on Technology Agreements appear in Table 8. The estimations confirm old consequences: the coefficients of the variable of chief involvement are non statistically important in any of the specifications, therefore confirming that engineering understandings do non act upon merchandise invention. As before, after analysing impacts of Technology Agreements on Product Innovations, the focal point displacements on the other dependent variable: Patent Registrations.

Table 9 shows the consequences of the appraisal of Technology Agreements coefficient sing the natural logarithm of the Patent Registration as dependent variable. Columns 1 allow for engineering understandings stipulated one and two old ages before the enrollment of the patent. While the coefficients of the lagged variable are non important in columns 1 ( a ) and 1 ( B ) , those of the Technology Agreements are positively important at 1 % and 5 % significance degree, severally. In peculiar, if a house stipulates a formal engineering transportation understanding in a given twelvemonth, the figure of

registered patents in the undermentioned twelvemonth can increase up to 6.8 % . This is besides confirmed by the undermentioned columns, where the coefficients of the dependant variable remain important and positive besides after dropping the lagged variable. In columns 3, a 1-year lagged variable replaces the engineering understandings: while in fixed effects model the coefficient of the variable of involvement is non important, in column 3 ( a ) a positive and important relationship between this variable and patent enrollments is found at the 5 % significance degree. Therefore, patent enrollment is positively affected by engineering understandings non merely in the old twelvemonth, but besides if the house stipulates the understanding two old ages before the enrollment of the patent.

To sum up, a formal engineering transportation understanding with a foreign opposite number does non better advanced productiveness of houses in footings of merchandise inventions ; this is confirmed by every analyses made, therefore connoting that the first hypothesis of the survey is non verified. On the contrary, if the step of invention is changed into the figure of patent enrollments, the purchase of proficient aid or licence from abroad positively and significantly impacts on advanced productiveness, therefore formalizing the 2nd hypothesis of this survey.

Extra analyses

Several extra analyses are conducted to measure the veracity of the findings. So far, a relationship between engineering understandings and the enrollment of patents has been confirmed ; nevertheless, the way of the relationship is still non wholly clear. Technology understandings might positively act upon the enrollment of patents but besides, old patents may hold an impact on the judicial admission of formal engineering understandings. Therefore, in order to guarantee that the presumed way of the relationship is the right one, a dynamic Arellano Bond calculator is used. In this appraisal, lagged values of the dependent variables are included as regressors. Including dependent variables allows accounting for the possible consecutive correlativity of mistakes, every bit good as for the dynamic constituent in the house features ( Golovko & A ; Valentini, 2011 ) .

In peculiar, two appraisals are made. In the first instance, the dependant variable is represented by “ Technology Agreements ” in a given period, while the independent variable, besides the control variables, is the count variable of “ Patent Registrations ” in the old period. As the appraisal allows commanding for dynamic constituent of house features, the lagged value of engineering understandings is besides included in the specification.

The 2nd appraisal uses “ Patent Registrations ” figure as a dependant variable, including its lagged value as covariate, and above all, “ Technology Agreements ” in the old period as regressor. As stated above, the purpose of these appraisals is to clear up the way of the relationship between engineering understandings and enrollments of patents. The way of the relationship will be from “ Technology Agreements ” to “ Patent Registrations ” , if the first variable is a statistically important regressor when the second is the dependent variable, but non vice-versa. Table 10 nowadayss the consequences of the appraisals. In the first column “ Technology Agreements ” is the dependent variable. While the lagged value of the dependant variable resulted positively important, corroborating the univariate analysis consequences, the estimation of the enrollment of patents resulted non important. On the contrary, the consequences displayed in the 2nd column validate the 2nd hypothesis of the survey: engineering understandings positively influence the enrollment of patents in the undermentioned twelvemonth. Among the estimations of the control variable, the most interesting are those sing “ Export Status ” and “ Import Status ” . Indeed, exporting to foreign states does non act upon the figure of patent registered, while, prosecuting in importing activities positively and significantly impacts on the figure of patents registered.

To measure the relevancy of the theoretical accounts, Arellano Bond trials for first and 2nd order consecutive autocorrelation of first-differenced mistakes are implemented for both appraisals. These trials verify whether mistakes are independently and identically distributed. If that is the instance, the first differenced mistakes are first-order serially correlated ; hence the hypothesis of zero autocorrelation in first-differenced mistakes in the trial can be rejected. On the contrary, consecutive correlativity in the first-differenced mistakes at an order higher than 1 implies that the minute conditions used by the theoretical account are non valid. As expected, the Arellano and Bond trial for first-order consecutive autocorrelation for the first and 2nd appraisals are found to be negative and important at the 1 % and 5 % degree severally, while the trials for second-order autocorrelation are rejected. There is no second-order correlativity, which confirms the consistence of the calculators. Therefore, it can be concluded that the conditions are valid.

In about all old appraisals, “ Patent Registrations ” were non affected by “ Technology Agreements ” stipulated two old ages before the enrollment of the patent. At the same clip, the variable “ Technology Agreements ” is positively influenced by its 1-year lagged value. Since it can be assumed that this relationship is valid besides for old periods, 2, 3 and 4-years lagged values of “ Technology Agreements ” can be considered instrumental variables as they influence the independent variable but non the mistakes, therefore the dependant variable. To command for this dynamic relationship, a 3rd Arellano Bond appraisal, whose consequences are shown in Table 11, is made ; as evident from Table 11, besides commanding for the dynamic relationship, the consequences of the 2nd column of Table 10 are confirmed. Indeed, “ Technology Agreements ” coefficient is still positive and important at the 1 % significance degree, albeit lower in magnitude than old specification. Again, to measure the relevancy of the specification, Arellano Bond trials for first and 2nd order consecutive autocorrelation of first-differenced mistakes are implemented, corroborating the cogency of the theoretical account.

To corroborate the consequences of the old appraisals, viz. that engineering understandings merely influence enrollment of patents, a logit theoretical account is used. Indeed, the purpose is to show that the judicial admission of engineering understandings in a given twelvemonth increases the odds of registering patents in the undermentioned twelvemonth, but it does non impact merchandise inventions. For this intent, a silent person variable refering the accomplishment of at least a merchandise invention is used and labeled “ Product Innovator ” ; this variable takes the value of 1 if merchandise inventions are obtained in the given period and, 0 otherwise. Similarly, a silent person variable sing the enrollment of at least one patent is used and labeled “ Patent Registration ” . The theoretical account includes “ Technology Agreements ” as independent variable together with the control variables antecedently used. The consequences of the appraisals are presented in Table 11. Merely random effects appraisals are run including twelvemonth and sector effects in the specifications. The first column presents the appraisal where “ Product Innovator ” serves as dependent variable. In line with old consequences, the coefficient of “ Technology Agreements ” did non ensue statistically important. To the contrary, the independent variable ‘s coefficient is positive and important when the dependant variable is “ Patent Registration ” . In peculiar, if the house has stipulated a engineering transportation understanding in the old period, the log-odds of patent enrollment additions by 0.372. In other words, the odds of registering a patent if the house has stipulated engineering understandings are 1.45[ 3 ]times higher than if the house has non. Furthermore, “ Export Status ” , “ Import Status ” , “ R & A ; D Intensity ” and “ Size ” significantly positively affect the likeliness of registering a patent, while the effects on patent enrollment of the engagement of foreign stockholders to houses ‘ equity are negative, albeit low in magnitude.

Discussion and decision

The purpose of this survey has been to dig deeper into the affair of the relationship between invention and export. Even though a great a portion of research workers agrees on the positive influence invention has on export, and vice-versa, there has been small grounds on the mechanism underpinning acquisition by exporting. Albeit bookmans have investigated the phenomenon, claiming that houses gain knowledge through informal relationship with foreign opposite numbers, the grounds in support of this hypothesis is instead scarce ( MacGarvie, 2006 ) . In this research it has been argued that formal understandings sing the purchase of engineering, licence and proficient aid positively influence advanced productiveness of houses. Therefore, the judicial admission of a formal understanding serves as channel through which houses gain cognition. Using merchandise inventions and patent enrollments as steps of advanced productiveness, the impacts that engineering understandings have on invention have been explored. The empirical attack is to regress the figure of merchandise inventions and patent enrollments on engineering understandings utilizing a unique balanced panel of Spanish fabrication houses provided by the Fundacion Empresa Publica for the period 1990-2002.

The through empirical observation analysis provided grounds partly consistent with the hypotheses. In fact, it has been found that houses do non increase their merchandise inventions if they have officially transferred engineering neither in the old twelvemonth, nor with farther slowdown. Explanations. On the contrary, patent enrollments appeared to be positively influenced by engineering understandings in the old twelvemonth. In add-on, the judicial admission of a engineering understanding besides has positive impacts if considered with farther slowdown, although in this instance grounds is non every bit strong as for 1-year slowdown. Explanations.

Furthermore, this survey gives besides grounds in support of acquisition by exporting. Unlike Higon and Driffield ( 2011 ) , which found that exporting does non act upon merchandise invention, this research ‘s consequences confirm positive impacts of exports on invention. Indeed, being an exporter increases both the figure of merchandise inventions and patent enrollments, besides with biennial slowdowns. Clearly, come ining in a foreign market increase the likeliness of being in contact with improved cognition, methods, engineerings, therefore positively act uponing invention. Furthermore, in line with Alvarez ( 2007 ) , import resulted greatly of import for invention. In fact, both the steps of invention considered turned out to be positively affected if the house imported. MacGarvie ( 2006 ) argues that, after get downing to import, houses register an addition in the cognition available, and, presuming that this will ensue in an addition of invention, these findings are consistent with this survey ‘s. Finally, it is interesting to underscore the function played by the presence of foreign stockholders in houses ‘ equity. Harmonizing to this research appraisal, if a house is a subordinate of an international company, the likeliness that it will register a patent decreases. These findings might be explained by the fact that international companies tend to developed cognition, and therefore registry patents, in the state where the headquarter is located. Therefore, subordinates are inclined to play a secondary function in the affair of invention.

Although extra analyses are performed to corroborate these findings, they should be carefully interpreted because of some cautions that may justify attending. First of wholly, the empirical analysis relies on informations sing entirely a state, Spain. Therefore, it does non take into consideration external factors, such as institutional, economic and fiscal, contexts which can badly impact on houses ‘ public presentations, particularly in footings of invention. Future developments would dwell in spread outing the research in other contexts in order to measure whether the consequences might be applied to other states.

Second, informations regard merely manufacturing houses ; as seen, sentiments sing fabrication sectors are much more consentaneous than for other industries ( e.g. Barrios et al. , 2003 ; Lachenmaier & A ; Woessmann, 2004 ; Nguyen et al. , 2008 ) . Therefore, it might be interesting to estimate whether this research ‘s consequences besides apply to other industries, such as service.

Third, this survey uses entirely dummy variables to depict both engineering understandings and export. On the contrary, the usage of steps of export strength ( such as export normalized on gross revenues ) and of figure of and/or outgos for engineering understandings would give further inside informations on the relationship. Indeed, they would assist in holding a more comprehensive image on how much engineering understandings affect invention.

Fourth, future research would be good served to research the nature of the understanding stipulated. As stated above, the variable of engineering understandings respects both the purchase of a licence and of proficient aid. It can be presumed that the positive effects these elements might hold on invention are rather different. Therefore, there is room for farther probe on differences in magnitude, in footings of positive influences, that understandings might hold.

Finally, the nature of the “ transferor ” is non taken into consideration. With the term nature, I largely refer to two facets: foremost of all, the transferor might be a transnational house, a little medium endeavor, a research centre, etc. Investigating the nature of the transferor might give new penetrations on strategic determinations of houses: as an illustration, houses might make up one’s mind to buy engineering from a little house focused on R & A ; D, alternatively of developing R & A ; D inside