Abstract
The empirical results of the study are presented in Chap. 6. The chapter starts with a descriptive analysis of different aspects of knowledge generation and use, whereby it is already seen that more regions than just the larger Moscow area and St. Petersburg are active in the Russian innovation system.The analysis of the knowledge production function shows that the Russian Federation indeed possesses a functioning innovation system however it is less effective than other systems.The first three approaches to knowledge spillovers – knowledge production function, patent citations and inventor mobility – show that spillover do exist take place however few they might be. Trying to explain these flows it is shown that indicators of spatial autocorrelation do not provide a suitable means to do so, however basic economic indicators like the patent stock, amount of researchers and GRP per capita are effective impact factors.Finally, some basic results on the impact of knowledge related aspects on economic growth have been gained however not all of them are consistent.
Economics is a subject that does not greatly respect one’s wishes. Nikita Khrushchev
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In the Russian Federation a number of institutions like academies or business schools exist that are not by name a university, but offer tertiary education programs comparable to those of regular universities. Therefore this study observes the broader concept of institutions of tertiary education, even if in some places the term university or simply institution is used.
- 2.
Shortened in this section to institutions.
- 3.
Note that in the figure the autonomous okrug of Chukothka is not considered as it registers no official institution of tertiary education.
- 4.
While using only researchers as input variables is the standard approach, the AK growth model would suggest using knowledge capital as a single input factor.
- 5.
Theoretically, it is available since 1987 and up to 2008. The Russian Federation as an independent national entity, however, only came into being in 1991. Furthermore, data on the year 2007 and the beginning of 2008 is available, but a closer look reveals that a number of patent applications are still pending. It can therefore be assumed that the data is reliable enough to be used in an econometric analysis only up to 2006.
- 6.
Due to lags of up to 3 years and a general availability and reliability problem for data on a regional basis from before 1994, an earlier beginning of the estimation horizon might be possible from the output side, but not from the input side.
- 7.
In the context of this study the researcher variable does not include technical research staff, assistant staff or miscellaneous research-oriented personnel, but only the researchers themselves.
- 8.
In the context of this study the expenditures on R&D encompass all domestic expenditures on R&D in real prices from 1995. As researchers are generally paid through the expenditures on R&D, on the one hand an implicit collinearity between both variables can be assumed, and effects of research personnel other than the researchers are indirectly covered by the variable on R&D expenditures.
- 9.
Note that for all of the following regressions the statistics program Stata/SE 10.1 has been used.
- 10.
A detailed description of both measurements can be found in Sect. 3.2.1.3.
- 11.
On the one hand this is due to the selection of larger regional entities and the disregard of all but one of the autonomous okrugs – Chukotka. On the other hand the reallocation of space from the Moscow Oblast to the city of Moscow did not take place until 2012 and therefore does not fall into any of the observed time frames.
- 12.
Nevertheless, in the cases where spatial effects are included in the estimation, a pooled data approach is used. A panel version of the spatial error and spatial lag model such as can be found in Arbia and Piras (2005) or Kapoor et al. (2007) has not been implemented to study the robustness of regression results over different types of estimation procedures.
- 13.
The three possibilities are therefore referred to as the 1,200 observation and the 800 observation case or alternatively the long term and the short term time frame.
- 14.
Missing data has been generated using inter- or extrapolation measures. In some cases missing data for Chechnya has been generated by assuming a structure similar to that of Ingushetia.
- 15.
See Levin et al. (2002).
- 16.
- 17.
A detailed summary of all test statistics can be found in Tables C.1–C.21 in the appendix.
- 18.
See Woolridge (2002).
- 19.
An alternative indicator would have been the panel version of the Durbin-Watson statistic as introduced by Bhargava et al. (1982).
- 20.
Note that for the single year cases a Breusch-Pagan Test for heteroskedasticity is performed as suggested by Verbina and Chowdhury (2004).
- 21.
In the literature, such as Kutner et al. (2004), it is stated that in a cross-section analysis, VIF up to a value of ten are unproblematic, while for panel data VIF can even be much higher; however no distinct cut-off value is given. Therefore, in the following study a cut-off value of ten is assumed implicitly.
- 22.
- 23.
The method described in Kapoor et al. (2007) would be able to cope with the panel structure and serial and spatial autocorrelation at once. However, in this study this approach was decided against for studying the influence of the different effects separately.
- 24.
Note that regional dummy variables to account for fixed regional effects are not included in the spatial versions of the model as they would lead to high multi-collinearity and non-convergence of the estimation method used for the spatial models. In the other versions they are not included as they are already represented by the fixed effects.
- 25.
Note that in the literature, either threshold VAR models – refer to Tsay (1998) for an example – or time-varying smooth transition models – refer to Lundbergh et al. (2003) for an example – are the usual approach to account for structural breaks in the data. However, in both cases the point of the structural break is either known in advance of results for both periods – before and after the break – or only reported on indirectly.
- 26.
The results of these tests are not reported in this study, though in some cases they are referred to in the ongoing text. It can be summarized that for all estimations performed the tests showed that significant differences between the coefficients for the transition period and the coefficients for the later period exist.
- 27.
The basic idea behind the link-test is rather simple. The estimated coefficients can be used to calculate an estimate of the dependent variable. In theory this estimate should be highly correlated to the real dependent variable. Furthermore, if the estimated variable is squared, it should no longer be correlated to the dependent variable at all. Therefore, a regression is conducted with the estimated dependent variable and its squared versions as regressors. If the estimated variable is significant, it can be said that the model uses fitting regressors – and the chance for omitted variables is rather small – while if the squared estimated variable is significant it can be seen as an indicator for model miss-specification.
- 28.
Note, that the link-test was not available for the panel estimation and has been conducted for the pooled version of the model only. Furthermore, as mentioned by Arbia and Piras (2005) and others, the panel structure of the model already takes care of some of the model bias generated by omitted variables.
- 29.
A higher statistic signifies a higher chance of omitted variables. The error probability is reported in brackets below the statistic.
- 30.
Both input variables enter the regression in logarithmized form. For each set of results model I includes only the researchers, model II includes only the R&D expenditures and model III includes both.
- 31.
If not reported explicitly, a structural break in levels between the transition and the later years is assumed to exist in all cases.
- 32.
In contrast to patents granted by Rospatent this implies that internationally important patents were granted – and therefore also filed – independent of the observed time period.
- 33.
While the Heckman selection model approach seems a prudent alternative to the implemented procedure, it necessitates at least one unique instrument variable for each of the two steps. As the variables in this approach are already predefined, the Heckman selection model is therefore not applicable.
- 34.
The single year case observes the year 2009 or 2008, depending on the lag-structure of the model. With a 2 year lag, variables from 2009 are regressed on patents from Rospatent for 2011.
- 35.
On the downside the LLC unit root test suggested that the 2 year lag structure might be better suited, at least for the short time horizon.
- 36.
Note that all of the variables implemented in this section enter the regression in logarithmized terms, except for shares or dummy-variables. Population based variables like researchers, government personnel and students are implemented on the one hand in logarithmized terms and on the other hand in per capita terms.
- 37.
Note that integrating the GRP also allows one to control for business cycle effects.
- 38.
As size is time invariant, it can even a priori be assumed to be of only marginal importance as all time-invariant effects should already be covered by the fixed effects structure of the estimation models. However, Kao et al. (1999) argue that size plays a role, as small countries benefit more from international R&D than larger countries.
- 39.
In an economy with an overall share of the shadow economy of between 20 and 50 % quite a large portion of effects is not covered by official GRP. Note that the share of the shadow economy is calculated with 1995 as a base year, as the results of a study by Afontsev (1998) have been used as base values and results are extrapolated backward and forward using the electricity method introduced in Sect. 5.1 and discussed in more detail in Kaufmann and Kaliberda (1996) or Johnson et al. (1997). It needs to be mentioned that no energy consumption statistics were available on a consistent basis for all years of the study, therefore energy production data has been used – under the assumption that energy production patterns in Russian regions roughly match energy consumption patterns. Use of the electricity method is strongly criticized by authors like Eilat and Zinnes (2002) and Alexeev and Pyle (2003).
- 40.
Lichtenberg and van Pottelsberghe de la Potterie (1998) stresses that it is not so much the intensity of imports, but the distribution of the countries of the origin of imports that matters.
- 41.
The stock of knowledge is calculated under the assumption that over the time horizon of 10 years patents lose their value in a linear fashion. The sum of the correspondingly depreciated patent numbers then produces the patent stock. Note that the implemented depreciation scheme partially mirrors an exponential smoothing.
- 42.
- 43.
- 44.
See Netter and Megginson (2001).
- 45.
As not all regions are oil and gas producers and therefore report zero oil and gas output, taking the logarithm is not advised, instead the absolute output is used as a variable, measured in billions of tons.
- 46.
- 47.
Note in this context as well the proclaimed negative relation between resource endowments and economic growth, and therefore the importance of this variable, which in the literature is referred to as the resource curse, in the later Sect. 6.5, See for a discussion of this phenomenon for example Auty (1993).
- 48.
The indicator is calculated as the relation of the sum of exports and imports against the GDP.
- 49.
Note that the inclusion of the openness indicator might not generate additional information as it basically replicates the effects of exports, imports and GDP in a composite form.
- 50.
Note, all monetary variables including especially the GRP, the expenditures on R&D, the exports and imports as well as the FDI enter the regression equation in real terms with the base year 1995.
- 51.
Mowery and Sampat (2006), for example, stress the role of universities in innovation systems as well.
- 52.
The Krugman index is used as well in the study by Breschi and Lissoni (2001). Connolly (2008) advocates the Krugman index to measure changes in the export structure and thereby account for the export pattern. In this study, however, the Krugman index has only been calculated for domestic production at the Russian top-level classification.
- 53.
In Russian called: Indeks gotovnosti regionov Rossii i informazionnomu obshchetvu.
- 54.
Note that only those variables that have not already been discussed in the Sect. 6.2.1 have been in considered here.
- 55.
In Table 6.20 model I is based on logarithmized absolute values for the researchers, students and government personnel, whereas model II is based on logarithmized per capita values.
- 56.
This would be reasonable if a deterioration of the Soviet Science system is assumed. However, it could also imply that relatively new knowledge entered the Russian Federation in the transition years – relative in the sense that it might have been new to the Russian Federation, but not in a global sense.
- 57.
The results of SMS are not remarkable considering that SME are supposed to have a non-linear-bowl-shaped influence on innovativeness.
- 58.
Kutner et al. (2004) argue that ten is a reasonable threshold for a cross-sectional model. The same threshold is assumed here for the panel models as well, even if a panel model is able to cope effectively with higher VIF.
- 59.
Model I shows the results for the whole time horizon; model II shows the results for the whole time horizon when an additional dummy variable for the transition years is introduced. Models III and IV show the results for the transition and the later years, respectively.
- 60.
Model I refers to the case where all years are observed. In model II only the transition years and in model III only the later years are observed.
- 61.
Note that the base year for the calculation of the shadow economy variable is 1995 and therefore the variable can be assumed to be more accurate for the transition years than for the later years.
- 62.
Model I uses a basic OLS estimation, while model II assumes a spatial error and model III a spatial lag structure.
- 63.
It should be noted that this method is lacking in two ways. First, the widths of the circles are not dynamic, but more or less set arbitrarily roughly following some general results on spillover reaches. In Sect. 5.2.1 results on different spillover reaches have been listed and, where possible, an assignment to the different annuli has been performed. However, these results mostly refer to the situation in the US or Europe and are not necessarily stable across different regional set-ups. Second, the point at which a region becomes part of a certain circle needs to be clarified. Here we use the artificial economic center of the region as introduced in Perret (2011). If the economic center of a region i is within a radius r around the economic center of region j then i is considered to be part of the annulus with outside radius r.
- 64.
Pat j,t represents the number of patents granted at time t in region j, Res j,t represents the number of researchers at time t in region j and RD j,t represents the expenditures on R&D at time t in region j.
- 65.
Model I refers to the case where the researchers are the observed input factor, while in model II R&D expenditures are the observed input factor.
- 66.
Note that for the researcher based spillover effects in the transition period the 75 km disc is marginally insignificant.
- 67.
Later referred to as the regional level.
- 68.
Later referred to as the regional-international level.
- 69.
Later referred to as the international level. In this context it needs to be noted that international spillovers do not necessarily need to be centered on the Russian Federation, as all patents with Russian participation – patents where at least one of the inventors has been Russian – are considered. This includes patents issued to applicants not situated in the Russian Federation as well.
- 70.
The first 4 years of the analysis take place with the Russian Federation still being part of the Union of Soviet Socialist Republics, which by itself does not present a problem as only a very limited number of patents were issued in this time and only those patents were considered that can be clearly attributed to the Russian Federation.
- 71.
Only a limited number of patents were available for 2007 and 2008, though.
- 72.
All of the figures in this and the following section are constructed using the statistics program UCINET 6 and the visualization program NetDraw 2.123. Each node in the figures represents a region or a country and the edges of the graphs give the knowledge flows between the nodes. The nodes/regions/countries in this section represent the origin of the applicant, while the inventors are only indirectly represented by the flows. In the following section the nodes represent the origin of the inventors. Here applicant addresses have been used as the applicant is the actual owner of the patent and therefore of the knowledge incorporated in it. Using applicant address therefore means using the place where the owners of the knowledge are situated, at least officially. A detailed list of the abbreviations used to name the regions and countries can be found in the appendix. The line size, as far as it differs, describes the amount of flows along that edge in comparison to the other flows. While two edges are comparable inside the same figure, comparability across the different figures is not assured. On the regional level, the numbers attributed the edges describe the amount of knowledge flows. As knowledge flows occur in both directions, the outgoing knowledge flow is always described by the number nearest to the respective node.
- 73.
In the ongoing study this type of spillover will be referred to as passive citing.
- 74.
For example, in the city of Moscow a significant a number of intra-regional spillovers take place.
- 75.
Note that only patents with Russian research participation were used.
- 76.
Henceforth this type of knowledge spillovers is referred to as active citing.
- 77.
As mentioned in Sect. 5.2.3, data for the years 2007 and 2008 is only available sporadically, as the data access was limited to the beginning of 2008.
- 78.
See for example Andrienko and Guriev (2004) for an overall low labor mobility across Russia.
- 79.
See for example the Nations in Transit or the Doing Business Index.
- 80.
Only lags of up to three periods are considered, but actual knowledge flows might need longer time horizons, even if only due to the patenting process itself. Besides, the results from the preceding section suggest that knowledge spillovers across the regions of the Russian Federation occur mostly over a long time horizon.
- 81.
The significance has been calculated by using the sample variance, based on the resulting Moran’s I statistics for every year and interpreting this as the actual variance. In a second step a basic t-test testing for difference from zero is conducted using this variance. A region is considered significantly positive or negative if this replacement t-value is absolutely larger than 1.6, equaling an approximate error level of 10 %.
- 82.
Two for patent citations and two for inventor mobility.
- 83.
The estimation procedure was introduced by Arellano and Bond (1991). While it has been noted that, especially in the growth context, Bond et al. (2001) argue that the Arellano-Bond/GMM estimator is insufficient and the Arellano-Bover or Blundell-Bond system estimator introduced by Blundell and Bond (1998) should be used instead. However, in the context in which it is implemented in this section, it is more than sufficient as its results do not significantly differ from those of the Blundell-Bond estimator. In the next section and especially in the growth context of Sect. 6.5, even though the perspective is more of an estimation of the Solow residual than one on economic growth itself, both estimators are used. It can be argued that a panel version of the Cochrane-Orcutt transformation might be used to cope with autocorrelation as well, however it has not been implemented in this study.
- 84.
Note, if an F-statistic or a χ2-statistic exceeds 1,000 the last three digits are abbreviated as k. If the statistic exceeds 1,000,000 the last digits are abbreviated as M.
- 85.
Model I like model II refers to the whole time horizon, however model II includes a dummy variable for the transition years. Model III refers to the transition and model IV to the later years.
- 86.
- 87.
Note that this assessment of the Getis-Ord statistic is not generally shared. Aldstadt and Getis (2006) for example use a procedure based on the local Getis-Ord statistic to generate an indicator of cluster detection.
- 88.
Note, tests on the model structure yield more or less the same results as before and lead to the same conclusions concerning the model structure. For completeness sake they are summarized as well in the appendix.
- 89.
See SREDA (2012).
- 90.
Note that the first years 1994 and 1995 include only the last 8 or 9 years, respectively, in their calculation of the stock of patents and not 10 years as is the rule.
- 91.
In the linear versions of the estimated model, researchers enter the estimation, measured in thousands of researchers, and GRP per capita enters the estimation measured in millions of rubles.
- 92.
Note, if an F-statistic or a χ2-statistic exceeds 1,000 the last three digits are abbreviated as k and if the statistic exceeds 1,000,000 the last six digits are abbreviated as M.
- 93.
Note that in the following regressions, where absolute numbers have been used, for clarity’s sake researchers and GRP per capita are given in thousands of researchers or thousands of rubles, respectively.
- 94.
According to Kremer (1993), having more people to draw on for ideas is an advantage for generating new knowledge.
- 95.
- 96.
It is assumed that the central role of Moscow and the Moscow Ooblast in the Russian science system biases the results significantly.
- 97.
Again the city of Moscow and the surrounding oblast might have had a biasing effect on the dynamics.
- 98.
Note that indeed in the later years, nominal as well as real income levels rose more than in the transition years. See Rosstat (2012).
- 99.
Note that a number of studies like Guellec and Van Pottelsberghe de la Potterie (2004) exist that analyze the importance of R&D and the institutional environment on the output of an economy. For the Russian Federation, however, Ahrend (2002) argues that political and institutional features are almost unimportant and can therefore be left out of a growth-related analysis.
- 100.
The theory of growth poles is basically the theoretical foundation for the formation of cluster, where one large enterprise dominates the cluster structure and related enterprises.
- 101.
- 102.
The model underlying the following estimations – unless stated otherwise – is always considered to report a log-linear form. Therefore, all variables, as long as they do not represent a quota or percentage, are logarithmized versions.
- 103.
Additionally, it has been shown in the context of Sect. 6.2.2 that the geographic size is highly correlated with other input variables as well as indirectly accounted for by the fixed effects.
- 104.
While the results in Netter and Megginson (2001) strengthen this argument, for the Russian Federation Berkowitz and DeJong (2003) show that ownership has no impact on firm performance; they instead highlight more the firms’ distance from Moscow, which in this study is implicitly included in the fixed effects.
- 105.
It would be more suitable to include a variable like the Corruption Perception Index, or the ICRG index of corruption advocated by Kim (2010) or the Bribe Payers Index advocated by Ofer (2010); however, they are not available on a regional level for a continuous span of years. Note as well the arguments by Brown and Shackman (2007) who links corruption and the long-term development of the GDP per capita and a continuing deterioration of law and order.
- 106.
As not all regions are oil and gas producers and therefore report zero oil and gas output, taking the logarithm is not advised; instead the absolute output is used as a variable, measured in billion tons.
- 107.
Refer for the resource curse hypothesis to Auty (1993), for example.
- 108.
In the linear versions of the estimated model researchers enter the estimation measured in thousands of researchers.
- 109.
For the Russian Federation Popov (2001) for example stresses the importance of the level of export shares for the regional performance.
- 110.
See Doehrn and von Westernhagen (2003) as one article that stresses the importance of FDI for growth in transition economies.
- 111.
- 112.
For the influence of single variables, model I refers to the whole time horizon, while model II refers to the transition and model III to the later years. Note, if an F-statistic or a χ2-statistic exceeds 1,000 its last three digits are abbreviated as k. If the statistic exceeds 1,000,000 its last six digits are abbreviated as M.
- 113.
It needs to be kept in mind that even in the two sub-periods the two variables are still highly correlated and the estimation of a basic production function would suffer from these severe multi-collinearity problems.
- 114.
Note, however, that the arguments in Ahrend (2002) mostly refer to the 1990s and Russia’s transition period.
- 115.
Note furthermore the results stated in Lundvall and Borrás (1997), amongst others, that the importance of tacit and codified spillovers differs across sectors. Therefore, if structural change has taken place, it is only reasonable to assume that the importance of tacit and codified spillovers has changed as well.
- 116.
Here model I refers to the OLS estimation, while model II refers to the spatial error estimation and model III refers to the spatial lag estimation.
- 117.
- 118.
Bähr et al. (2007) argue that high-demand regions are usually more strongly specialized. The inclusion of an indicator of regional specialization therefore seems prudent.
- 119.
Srholec (2007) sees the ICT infrastructure as an essential part of the technological capability of a region and therefore its absorptive capacity. By linking the absorptive capacity to growth it becomes obvious that the ICT infrastructure should also play an essential part in economic growth. Additional arguments in favor of the ICT infrastructure as a driver of economic growth can be found, for example, in Welfens (2012) or Meijers et al. (2008).
- 120.
Note still that if the influence of each institution is considered separately, each reports a consistently significant and positive impact on economic growth and also is able to explain quite a lot the variance of the GRP.
- 121.
Note that for a better visibility of the results, the openness indicator has been measured in percentages.
- 122.
The number of universities is highly correlated to the number of students.
References
Afontsev S (1998) The shadow economy in the Russian regions/tenevaya ekonomika v rossijskikh regionakh. In: Erchov E (ed) Review of the Russian Economy – Main Development Trends/Obzor Ekonomiki Rossii – Osnovnye Tendentsii Razvitiya, 1st edn. Recep, Moscow, pp 232–247
Ahrend R (2002) Speed of reform, initial conditions, political orientation, or what? Explaining Russian regions’ economic performance. DELTA working paper 2002-10
Aldstadt J, Getis A (2006) Using AMOEBA to create a spaital weights matrix and identify spatial clusters. Geogr Anal 38:327–343
Alexeev M, Pyle W (2003) A note on measuring the unofficial economy in the former Soviet Republics. Econ Transit 11:153–175
Andrienko Y, Guriev S (2004) Determinants of interregional mobility in Russia. Econ Transit 12:1–27
Anselin L (1988) Spatial econometrics: methods and models, 1st edn. Kluwer, Dordrecht/Boston
Arbia G, Piras G (2005) Convergence in per-capita GDP across European regions using panel data models extended to spatial autocorrelation effects. ISAE working paper 51
Arellano M, Bond S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev Econ Stud 58:277–297
Asheim B, Gertler M (2005) The geography of innovation: regional innovation systems. In: Fagerberg J, Mowery D, Nelson R (eds) The Oxford handbook of innovation, 1st edn. Oxford University Press, Oxford, pp 291–317
Auty R (1993) Sustaining development in mineral economies: the resource course thesis, 1st edn. Routledge, London/New York
Ayyagari M, Beck T, Demirgüç A (2003) Small and medium enterprises across the globe: a new database. Worldbank Policy Research working paper 3127
Bähr C, Stierle-von Schütz U, Wrede M (2007) Dezentralisierung in den eu-staaten und räumliche verteilung wirtschaftlicher aktivitäten. Perspektiven der Wirtschaftspolitik 8:110–129
Baltagi B (2003) Econometric analysis of panel data, 2nd edn. Wiley, Chichester
Berkowitz D, DeJong D (2003) Policy reform and growth in post-Soviet Russia. Eur Econ Rev 47:337–352
Bhargava A, Franzini L, Narendranathan W (1982) Serial correlation and the fixed effects model. Rev Econ Stud 49:533–549
Blume L, Fromm O (2000) Regionalökonomische Bedeutung von Hochschulen, 1st edn. Gabler Verlag, Wiesbaden
Blundell R, Bond S (1998) Initial conditions and moment restrictions in dynamic panel data models. J Econom 87:115–143
Bond S, Hoeffler A, Temple J (2001) GMM estimation of empirical growth models. CEPR discussion paper 3048
Bottazzi L, Peri G (2003) Innovation and spillovers in regions: evidence from European patent data. Eur Econ Rev 47:687–710
Breschi S, Lissoni F (2001) Localised knowledge spillovers vs. innovative milieux: knowledge ‘tacitness’ reconsidered. Pap Reg Sci 80:255–273
Brown S, Shackman J (2007) Corruption and related socioeconomic factors: a time series study. KYKLOS 60:319–347
Coe D, Helpman E (1993) International R&D spillovers. Eur Econ Rev 39:859–887
Cohen W, Levin RC, Mowery D (1987) Firm size and R&D intensity: a re-examination. J Ind Econ 35:543–565
Connolly R (2008) The structure of Russian industrial exports in comparative perspective. Eurasian Geogr Econ 49:586–603
Doehrn R, von Westernhagen N (2003) The role of foreign direct investment in transformation. In: Lane T, Oding N, Welfens P (eds) Real and financial economic dynamics in Russia and Eastern Europe, 1st edn. Springer, Heidelberg, pp 251–278
Eggertson T (1990) Economic behavior and institutions, 1st edn. Cambridge University Press, Cambridge/New York
Eilat Y, Zinnes C (2002) The shadow economy in transition countries: friend or foe? A policy perspective. World Dev 30:1233–1254
Emons O (2012) Innovation and specialization dynamics in the European automotive sector: comparative analysis of cooperation and application network, 1st edn. Springer, Heidelberg, pp 185–240
Falvey R, Foster N, Greenaway D (2002) North-South trade, knowledge spillovers and growth. J Econ Integr 17:650–670
Feldkircher M (2006) Regional convergence within the EU-25: a spatial econometric analysis. In: Proceedings of OeNB Workshops, Vienna, vol 9, pp 101–119
Fritsch M (2000) Interregional differences in R&D activities – an empirical investigation. Eur Plan Stud 8:409–427
Fritsch M (2003) Does R&D-cooperation behavior differ between regions. Ind Innov 10:25–39
Fritsch M, Brezinski HE (1999) Innovation and technological change in Eastern Europe: pathways to industrial recovery, 1st edn. Edward Elgar, Cheltenham/Northampton
Fritsch M, Franke G (2004) Innovation, regional knowledge spillovers and R&D cooperation. Res Policy 33:245–255
Fritsch M, Slavtchev V (2006) Universities and innovation in space. Ind Innov 14:201–218
Funk MF (2001) International R&D spillovers and convergence among OECD countries. J Econ Integr 16:48–65
Furubotn E, Richter R (1991) The new institutional economics: an assessment, 1st edn. A&M University Press, College Station
Gambardella A, Malerba F (eds) (1999) The organization of economic innovation in Europe, 1st edn. Cambridge University Press, Cambridge/New York
Getis A, Ord J (1992) The analysis of spatial association by use of distance statistics. Geogr Anal 24:189–206
Greene W (1997) Econometric analysis, 3rd edn. Prentice-Hall, Saddle River
Guellec D, Van Pottelsberghe de la Potterie B (2004) From R&D to productivity growth: do the institutional settings and the source of funds of R&D matter? Oxf Bull Econ Stat 66:353–378
Johnson S, Kaufmann D, Shleifer A (1997) The unofficial economy in transition. Brook Pap Econ Act 2:159–221
Johnson B, Lorenz E, Lundvall BA (2002) Why all this fuss about codified and tacit knowledge? Ind Corp Change 11:245–262
Kao C, Chiang MH, Chen B (1999) International R&D spillovers: an application of estimation and inference in panel cointegration. Oxf Bull Econ Stat Spec Issue 61:691–709
Kapoor M, Kelejian H, Prucha I (2007) Panel data models with spatially correlated error components. J Econom 140:97–130
Kaufmann D, Kaliberda A (1996) Integrating the unofficial economy into the dynamics of post-socialist economies: a framework for analysis and evidence. In: Kaminski B (ed) Economic transition in Russia and the New States of Eurasia, 1st edn. M.E. Sharpe, London, pp 81–120
Keller W (1998) Are international R&D spillovers trade-related? Analysing spillovers among randomly matched trade partners. Eur Econ Rev 42:1469–1481
Khokhlova Y, Shaposhnika S (eds) (2010) E-readyness index of the Russian regions – Indeks Gotovnosti Regionov Rossii k Informatsionnomy Obshchestvu, 1st edn. IRIO, Moscow
Kim MJ (2010) Corruption and economic growth. In: EACGS conference, Cambridge
Klochikhin E (2012) Russia’s innovation policy: stubborn path-dependencies and new approach. Res Policy 41:1620–1630
Kremer M (1993) Population growth and technological change: one million b.c. to 1990. Q J Econ 108:681–716
Kutner M, Nachtsheim C, Neter J (2004) Applied linear regression models, 4th edn. McGraw-Hill, Irwin
Leite C, Weidmann J (1999) Does mother nature corrupt? Natural resources, corruption, and economic growth. IMF Working paper WW/99/85
Levin R, Klevorick A, Nelson R, Winter S, Link A, Bozeman B (1991) Innovative behavior in small-sized firms. Small Bus Econ 3:179–184
Levin A, Lin CF, Chu CS (2002) Unit root tests in panel data: asymptotic and finite sample properties. J Econom 108:1–24
Lichtenberg F, van Pottelsberghe de la Potterie BJMTG (1998) International R&D spillovers: a comment. Eur Econ Rev 42:1483–1491
Lucas RE Jr (1988) On the mechanics of economic development. J Monet Econ 22:3–42
Lucas RE Jr (1993) Making a miracle. Econometrica 61:251–271
Lundbergh S, Teräsvirta T, van Dijk D (2003) Time-varying smooth transition autoregressive models. J Bus Econ Stat 21:104–121
Lundvall BA (ed) (2010) National systems of innovation: toward a theory of innovation and interactive learning, 1st edn. Anthem Press, London
Lundvall BA, Borrás S (1997) The globalising learning economy: implications for innovation policy, 1st edn. DGXII-TSER, The European Commission, Brussels
Mansfield E (1981) Composition of R&D expenditures: relationship to size of firm, concentration and innovative output. Rev Econ Stat 63:610–615
Meijers H, Dachs B, Welfens P (2008) Internationalisation of European ICT activities, 1st edn. Springer, Heidelberg
Michailova S, Hutchings K (2006) National cultural influences on knowledge sharing: a comparison of China and Russia. J Manag Stud 43:383–405
Mowery D, Sampat B (2006) Universities in national innovation systems. In: Fagerberg J (ed) The Oxford handbook of innovation, 1st edn. Oxford University Press, Oxford, pp 209–239
Netter J, Megginson W (2001) From state to market: a survey of empirical studies on privatization. J Econ Lit 39:321–389
Ofer G (2010) Twenty years later and the socialist heritage is still kicking. UNU-WIDER working paper 59
Ord J, Getis A (1995) Local spatial autocorrelation statistics: distributional issues and an application. Geogr Anal 27:286–306
Perret J (2011) A proposal for an alternative spatial weight matrix under consideration of the distribution of economic activity. Schumpeter discussion papers 2011-002
Pisati M (2001) Tools for spatial data analysis. Stata Tech Bull 10:21–37
Podmetina D, Vaatanen J, Smirnova M (2011) Open innovation in Russian firms: an empirical investigation of technology commercialization and acquisition. Int J Bus Innov Res 5:298–317
Popov V (2001) Reform strategies and economic performance of Russia’s regions. World Dev 29:865–886
Richter R, Furubotn E (2010) Neue Institutionenökonomik: Eine Einführung und kritische Würdigung, 4th edn. Mohr, Tübingen
Rodrik D, Subramanian A, Trebbi F (2004) Institutions rule: the primacy of institutions over geography and integration in economic development. J Econ Growth 9:131–165
Rosstat (2012) Regions of Russia 2012/regiony rossii. http://www.gks.ru. Accessed 27 Feb 2013
Rothwell R (1989) Small firms, innovation and industrial change. Small Bus Econ 1:51–64
Roud V (2007) Firm-level research on innovation and productivity: Russian experience. ISSEK working paper
Russian Public Opinion Research Center/VCIOM (2012) About us. http://www.wciom.com/index.php?id=139. Accessed 27 Feb 2013
Sala-I-Martin X (1997) I just ran two million regressions. Am Econ Rev 87:178–183
Scherngell T (2007) Interregionale Wissensspillovers in der Europäischen High-Tech Industrie – Eine Empirische Analyse, 1st edn. Deutscher Universitäts-Verlag, Wiesbaden
Schumpeter J (1911) Theorie der wirtschaftlichenEntwicklung, eine Untersuchung über Unternehmergewinn, Kapital, Kredit, Zins und den Konjunkturzyklus, 2nd edn. Harvard University Press, Cambridge
Silva A, Afonso O, Africano A (2010) Learning-by-exporting: what we know and what we would like to know. Int Trade J 26:255–288
SREDA (2012) Arena – atlas of religions and nationalities of the Russian Federation/arena – atlas religiy i natsionalnostey rossiyskaya federatsiya. http://www.sreda.org/arena. Accessed 27 Feb 2013
Srholec M (2007) High-tech exports from developing countries: a symptom of technology spurts or statistical illusion? Rev World Econ 143:227–255
Tiefelsdorf M, Griffith D, Boots B (1999) A variance-stabilizing coding scheme for spatial link matrices. Env Plan 31:165–180
Tompson W (2006) A frozen Venezuela? The resource curse and Russian politics. In: Ellman M (ed) Russia’s oil and natural gas: bonanza or curse?, 1st edn. Anthem Press, London, pp 189–212
Torkkeli M, Podmetina D, Ylä-Kojola AM, Väätänen J (2009) Knowledge absorption in an emerging economy – the role of foreign investments and trade flows in Russia. Int J Bus Excell 2:269–284
Tsay R (1998) Testing and modeling multivariate threshold models. J Am Stat Assoc 93:1188–1202
Vadi M, Vereshagin M (2006) The deposit of collectivism in organizational culture in Russia: some consequences of human resource management. Balt J Manag 1:188–200
Verbina I, Chowdhury A (2004) What determines public education expenditures in Russia? Econ Transit 12:489–508
Wagner J (2002) Exports and productivity: a survey of the evidence from firm-level data. World Econ 30:60–82
Welfens P (2012) Regional innovation and cluster policies in the new and old economy. In: Welfens P (ed) Clusters in automotive and information & communication technology, 1st edn. Springer, Heidelberg, pp 1–58
Williamson O (1975) Markets and hierarchies: analysis and antitrust implications, 1st edn. Free Press, New York
Wilson T (2002) The nonsense of ‘knowledge management’. Inf Res 8
Woolridge J (2002) Econometric analysis of cross section and panel data, 1st edn. MIT, Cambridge
Yuri Levada Analytical Center (2012) Levada Center. http://www.levada.ru/eng/levada-center. Accessed 27 Feb 2013
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Perret, J.K. (2014). On the Importance of Knowledge in the Russian Federation. In: Knowledge as a Driver of Regional Growth in the Russian Federation. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-40279-1_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-40279-1_6
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-40278-4
Online ISBN: 978-3-642-40279-1
eBook Packages: Business and EconomicsEconomics and Finance (R0)