No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Manifesto for a Growth Econometrics
Abstract
While the last three decades have seen remarkable advances in the econometric analysis of many areas of microeconomics and macroeconomics, growth economics has not experienced anything close to such progress. This is so despite the veritable explosion of theoretical and empirical studies of economic growth that has occurred in the last 15 years. A sign of the need for a growth econometrics is made clear when one re#ects on how little the empirical and theoretical literatures on economic growth make contact. For example, while spillover e!ects and attendant nonlinearities continue to be hallmarks of the new endogenous growth theory, empirical practice still largely focuses on linear models whose speci"cation is suggested by the neoclassical exogenous growth model of Solow and whose empirical implementation is often asserted to provide evidence in favor of that model (cf. Barro, 1991). It is no exaggeration to say that the theoretical and empirical growth literatures are evolving with little interaction. It is only through an econometrics that can link theory to data analysis and hypothesis testing that a synergy can be achieved. Empirical analyses of growth generally follow a common strategy. A crosssection or panel of countries is employed in which per capita output growth rates are assumed to depend over a common time horizon on two sets of variables. One set of variables consists of initial per capita output, savings and population growth rates, variables that are suggested by the Solow growth model. The second set of variables consists of control variables that correspond to whatever additional determinants of growth a researcher wishes to examine. In a survey published last year, Danny Quah and I (Durlauf and Quah, 1999) found that over 90 di!erent additional variables had appeared in the literature as of 1998. The current number is certainly much larger, despite the fact that only about 120 countries are available for analysis in the standard data sets. A result of this paucity of data relative to theories is the vast number of nonoverlapping studies that have appeared. A number of the problems that plague empirical growth analysis represent variants of traditional problems with the interpretation of econometric models.
... In essence, in the presence of a heterogeneous relation between FDI and economic growth, the "average effect" that is captured by a homogeneous coefficient in conventional modelling will not be robust to the different conditioning sets, sample sizes or functional forms that are evident across existing empirical studies. Indeed, this notion of relaxing the common assumption of parameter homogeneity in growth regressions is not new to the empirical growth literature (see, e.g., Brock and Durlauf (2001), Durlauf (2001), Masanjala andPapageorgiou (2004), Minier (2007), and the references cited therein). The thrust in this strand of the empirical growth literature is that assuming homogeneous coefficients is tantamount to assuming all countries produce goods and services using identical technologies. ...
... The thrust in this strand of the empirical growth literature is that assuming homogeneous coefficients is tantamount to assuming all countries produce goods and services using identical technologies. Furthermore, Durlauf (2001) argues that coefficients in a growth equation should be functions of "state of development". Our modelling approach incorporates this view. ...
... It is quite plausible and possible that parameter heterogeneities associated with traditional growth determinants are more prevalent and substantial in samples that pool developed and developing, or OECD and non-OECD countries. This is because of the significant differential in growth mechanism between developed and developing countries as argued by Durlauf (2001) and others. Thus, pooling developed and developing countries can substantially bias estimation results in favor of finding parameter heterogeneity. ...
... This essay questions these two assumptions. These questions are motivated by recent papers by Brock and Durlauf (2000) and Durlauf (2001) who argue that the basic Solow (1956) model and conventional cross-country linear regression models, a 0 la Mankiw, Romer and Weil (1992) impose strong homogeneity assumptions on the growth process. This is especially crucial because there is nothing in the theoretical and empirical literature to suggest that all countries obey a common linear production function. ...
... Recent papers by Brock and Durlauf (2000) and Durlauf (2001) argue that the conventional Mankiw, Romer and Weil (1992) (MRW hereafter) cross-country linear regression model based on Solow (1956) imposes strong homogeneity assumptions on the growth process. Assuming parameter homogeneity in growth regressions is equivalent to assuming that all countries have an identical Cobb-Douglas (CD) aggregate production function. ...
... However, in cross-country regressions treatment of endogeneity problems is less than satisfactory because of lack of viable exogenous instruments. Brock and Durlauf (2000) and Durlauf (2001), among others, observe that studies using instrumental variables (IV) to address endogeneity are not convincing as their choice of instruments do not meet the necessary exogeneity requirements. 15 In addition, Romer (2001) shows that IV estimation potentially introduces an upward bias in the parameter estimates due to the fact that most measures of physical and human capital used in the literature vary with levels of per capita output. ...
... As suggested by Baumol (1986) economies might be divided into multiple convergence clubs, and economies within each club converge, but convergence does not occur across all clubs. In addition, Durlauf (2001) emphasizes that the constant coefficient linear model assumptions used in standard growth analyses are not supported by the data, suggesting that alternative models of parameter heterogeneity should be employed in economic convergence analysis. ...
... As highlighted by Durlauf (2001), a key limitation that plagues empirical growth analysis is that the constant coefficient model assumption is not supported by the data. Although rich conceptual constructs such as multiple regimes (Durlauf and Johnson 1995), convergence clusters (Quah 1996) and convergence clubs (Chatterji and Dewhurst 1996) have been provided to close the gap between the theoretical and empirical growth literature, the challenges that spatial effects in the regional growth process pose for our understanding of regional income convergence are equally deserving of attention. ...
This study empirically applies the spatial switching regression method to an analysis of regional income club convergence across the 177 economic areas in the contiguous US states over the period from 1969 to 2008. As functionally defined, these economic areas represent the relevant regional markets for labor, products and information. The result of spatial switching regression reveals that the initial gaps between economic areas relative to average global initial per capita income appear to have declined, but the two spatial clubs exhibit a significant difference in their income convergence processes over the period. The estimated coefficient of the convergence parameter for the peripheral spatial regime is negative and highly significant, indicating that a convergence process exists in this spatial regime. However, there is no statistically significant evidence of convergence in the core spatial regime, implying the possibility of different patterns in the growth dynamics of the core spatial regime.
... Örneğin Durlauf ve Johsen (1995) çalışmalarında Solow büyüme modelinin açıklayıcı gücünün farklı gelir ve eğitim düzeyine sahip ülkeler arasında değiştiğine yönelik güçlü bulgular elde etmiştir (Durlauf ve Johnson, 1995: 365, 370). Bunun yanı sıra Durlauf (2001) literatürde yer alan uygulamalı araştırmalarda ülkelerin büyüme deneyimlerini, ülkelerin kendi özellikleriyle açıklamasını daha makul bulduğunu belirtmiştir (Durlauf ve Johnson, 2001: 68). CCE (veya CCEMG) tahmincisinin hem devamı hem de alternatifi olması amacıyla geliştirilen AMG tahmincisi, Eberhardt ve Teol (2010)'un bir üretim fonksiyonu tahmininde kullanmak üzere geliştirilmiştir. ...
Sağlık harcamalarının ekonomik büyüme üzerine etkisini inceleyen ilk çalışma Mushkin (1962) tarafından ele alınmıştır. Bu çalışmada sağlık harcamalarının ekonomik büyümeyi olumlu etkilediği belirtilmiştir. Devam eden dönemlerde sağlık ve ekonomik büyüme ilişkisi, sağlığa dayalı büyüme hipotezi olarak literatürde yer almıştır. Bu hipotez özellikle yeni sağlık politikası uygulamalarına geçildiği 80’li yıllardan sonra birçok gelişmekte olan ülkelerde test edilmiştir. Bu çalışmada, Türkiye’de 2003-2019 yılları arasında Düzey 2 kapsamında yer alan 26 alt bölgede kamu sağlık harcamalarının ekonomik büyüme ve toplam faktör verimliliği üzerine etkisi araştırılmaktadır. Çalışmada ekonomik büyüme, kamu sağlık harcamaları ve kamu eğitim harcamaları verilerinin yanı sıra kontrol değişken olması amacı ile sabit sermaye yatırımları, ihracat, işgücü stoku ve ortalama yıllık hava sıcaklığı verileri de kullanılmıştır. Verilerin zaman boyutunun kısa olması nedeniyle Panel EKK yöntemi ve bu yöntemden türetilen en uygun tahminciler ile sağlığa dayalı büyüme hipotezinin geçerliliği araştırılmıştır. Elde edilen bulgulara göre sağlığa dayalı büyüme hipotezinin Düzey 2 kapsamındaki TR32, TR41 ve TR83 bölgelerinde kabul edildiği belirlenmiştir.
... Another topic of discussion in the economic growth literature is about the empirical studies that pay little attention to the econometric problems in modeling and testing growth theories (Durlauf 2001). For example, the ordinary least squares (OLS) estimator, frequently used in empirical studies, is inconsistent due to the endogeneity of explanatory variables and as lagged dependent variables and other regressors are correlated with unobserved individual effects. ...
Economic complexity has significant implications for regional imbalances in Turkey, given the pronounced economic differences between its eastern and western provinces. This study aims to provide evidence to support the hypothesis that economic complexity has a significant positive effect on regional growth and long-standing regional imbalances in Turkey prevent the eastern region from benefiting equally from the economic gains of rising complexity. Using annual export data from 2007 to 2020 we calculated economic fitness and economic complexity indexes for the provinces of Turkey. We demonstrate that both economic fitness and economic complexity indexes are important determinants of regional growth in Turkey.
... 5 Baltagi (2021) states that the random effects estimator is an appropriate specification for samples drawn randomly from a large population, while the fixed effects estimator is more appropriate when focusing on a cluster of firms with certain common characteristics, or a specific set of countries or regions, such as OECD countries or states of the United States. In addition, Durlauf (2001) highlighted that there is a contentious issue in the growth literature where empirical studies on growth have tended to overlook or disregard econometric problems when testing and modelling growth theories. ...
This study investigates the relationship between related and unrelated variety and regional development traps in 26 NUTS-2 regions across Turkey from 2014-2021. We propose two unique measures, DT1 and DT2, for these traps based on deviations from EU and emerging market nations. Our findings suggest a notable negative correlation between related variety and DT1, implying that increasing related variety could potentially mitigate the risk of falling into development traps. Furthermore, DT1 appeared to be more reliable than DT2 in accurately representing the risk of regional development traps, particularly in countries experiencing high growth deviations, like Turkey.
... The parametric specification in equation (7) places a fairly strong restriction on the possible link between spatial inequality and economic development, obscuring the true underlying relationship. Studies of growth and convergence must consider parameter heterogeneity; hence, semiparametric methods are the more appropriate approach (Durlauf 2001). A more flexible way of modeling the trajectory of spatial inequality over the course of economic development is thus captured by the following panel semiparametric fixed-effects model: ...
This paper examines spatial income inequality and convergence in the Philippines—a lower middle-income country with historically high inequality—and its course over the process of economic development. Combining higher quality nighttime lights (NTLs) with gridded population data, I construct subnational measures of spatial inequality for the 17 administrative regions in the country in the period 2000–2020. Using this unique dataset, I first document the tremendous improvement in income disparities over the last two decades. Income per capita across provinces has converged rapidly, and income dispersion within administrative regions has narrowed markedly. Then, I uncover a U-shaped relationship between spatial inequality and economic development, which is robust; across alternative measures of inequality; to the outlier effects of highly urbanized cities; across parametric and semiparametric specifications; to business-cycle effects; and to persistence of spatial inequality. Finally, I confirm that structural transformation acts as a transmission channel of this U-shaped link.
... Such differences are explained by the imperfections in the market for knowledge because of monopolistic licensing fees, the requirements for local technology absorption, and protectionist policies (Pack 1994). This is in line with the diversity of growth experiences among countries (Mankiw et al. 1992;Pack 1994;Durlauf 2001;Durlauf et al. 2005). Second, it accounts for cross-sectional dependence across countries, notably in the TFP series. ...
We employ a recent empirical strategy to estimate country-specific and time-varying total factor productivity (TFP) levels for the manufacturing sector of 63 countries over 40 years. The methodology is based on estimated country-specific production functions while accounting for cross-section dependence and nonstationary series. We then analyze the derived TFP series across the entire sample and several regional groupings (Asia, Europe and Central Asia (ECA), Middle East and North Africa (MENA), Latin America and the Caribbean (LAC), and the USA). Our analysis reveals the following. Firstly, the TFP that is common across countries has an upward trend with a significant slump in 2008. Secondly, the leading positions in terms of productivity in the manufacturing sector remained the prerogative of major developed countries. Thirdly, several countries succeeded in climbing the ladder through outstanding productivity growth. Fourthly, despite a clear hierarchy in terms of manufacturing productivity across regional blocs, all regions witnessed an increase in productivity over the period. Fifthly, there is evidence of convergence in the TFP across countries and within Asia and ECA before a potential break in 2008.
... According to Caselli et al. (1996) and Durlauf (2001), the existing empirical literature on economic growth is heavily based on inconsistent estimation procedures due to endogenous explanatory variables. To overcome the endogeneity problem, Caselli et al. (1996) rewrite the equation as a dynamic growth model where the initial value of the dependent variable is added as an additional explanatory variable, the logarithmic difference of this variable is on the left-hand side of the equation, and the given equations are estimated using the generalized method of moments (GMM) framework. ...
Economic fitness is a nonlinear method for measuring the economic complexity of countries and regions. This study aims to measure the economic complexity of regions and its impacts on sectoral labor productivity in Turkey. By developing and employing a regional economic fitness index, the study addresses agriculture, industry, and service sectors at the NUTS‐2 level to have more effective analysis of regions compared to overall labor productivity. Our findings show that economic fitness positively affects the overall labor productivity of the regions and this positive effect mostly stems from the industrial sector. Nevertheless, our findings also highlight that there is no relationship between the regional fitness index and the agriculture sector.
... Second, the empirical models used are linear models, but some authors speculate about possible asymmetries in the response of marriage to change in house prices, depending on the initial level of the price [29]. To overcome the limitations of traditional linear approaches, the use of semiparametric methods is recommended [49]. This alternative approach would allow us to tackle the possible nonlinear effect of initial house prices on marriage in a flexible way because the model is a mixture of both parametric and nonparametric regressions. ...
The aim of this study is to examine the link between house prices and marriage in Spain. We consider data from 50 Spanish provinces (NUTS III regions) and from local civil registries in 282 cities with populations greater than 25,000 inhabitants. The regional data cover the 1995–2018 period, whereas the local sample includes information from 2005 to 2018. The marriage rate is defined as the annual absolute number of marriages per thousand inhabitants in each region or city. We used data on Spain because the Spanish housing market experienced a strong rise in house prices until 2006, when the housing bubble ended and prices dramatically decreased. By using different econometric techniques (panel data models with fixed effects and dynamic panel data models), our results reveal that there is a significant negative relationship between house prices and the marriage rate at both the regional and local levels. Overall, this study highlights the important consequences of rising house prices on family formations. Therefore, public authorities should try to reduce fluctuations in house prices and to facilitate access to home ownership for young couples.
... Such expenditures as spending on efficient legal systems to protect investors and creditors rights, the needed road and transportation infrastructure that spurs private sector investment, and spending on security to allow for a peaceful investment climate will all enhance the role of finance in promoting growth. Going by these results, typical linear cross country regression results may be misleading given the presence of non-linearity (Durlauf, 2001). ...
We ask how government size impacts the relationship between financial development and economic growth for Sub-Saharan African countries during 1980-2014 period. The empirical strategy relies on panel data techniques of panel smooth transition regression (PSTR) and dynamic GMM approaches to test for non-linearities in finance-growth nexus and endogenously model the non-linearity to depend on government size. Preliminary results from the dynamic panel model show that finance-growth nexus is non-linear. Upon delving further into the non-linearities, our results show that financial development impacts economic growth positively only beyond a necessary threshold level of government expenditure share of 12 percent of GDP. Given the large informal sector and the pervasive market failures in Sub-Saharan African economies, governments need a minimum threshold level of expenditure to "correct" the credit markets and "formalize" the large informal financial sector. Therefore, Sub-Saharan African countries should not shy away from government expenditures that are meant to "correct" market failures in the financial sector or "formali-ze" the large informal financial sector to make finance work for economic growth.
... To overcome these limitations, Durlauf (2001) suggested the use of semiparametric methods. This approach allows us to tackle the possible non-linear effect of the different urban variables on COVID-19 incidence in a more flexible way. ...
This paper investigates the determinants of the diffusion and intensity of the COVID‐19 at the country level, focusing on the role played by urban agglomeration, measured using three urban variables: percentage of the urban population, population density, and primacy. We estimate the influence of urban agglomeration on two outcome variables: cumulative number of cases and deaths per 100,000 inhabitants up to 31 December 2020, using both parametric and semiparametric models. We also explore possible spatial effects. The non‐linear effects of the urban variables on the intensity of the disease reveal non‐monotonous relationships, suggesting that it is the size of the urban system that is linked to a stronger incidence.
... A major criticism of these regressions is that it is subject to endogeneity problems. Economic growth variables are endogenous, and hence it is a hard task to estimate the causal effect of an explanatory variable on economic growth [33]. A common strategy to mitigate endogeneity is to use instrumental variables. ...
Understanding how growth factors contribute to explaining the large differences in growth rates across countries remains an important research agenda. The common approach to exploring this issue is based on the use of multiple linear regression analyses. This work contributes to growth literature by applying a new perspective based on the use of variance decomposition procedures: Shapley–Owen–Shorrocks and Oaxaca–Blinder. These methodologies have four main advantages with respect to traditional methodologies: they make possible the quantification of the relative contribution of each factor to economic growth, they allow us to estimate the efficiency in the use of the endowments of each factor, they can be used with any functional form and they can be used with estimation methods that are robust regarding endogeneity issues. We illustrate these advantages by analyzing the causes of the economic growth gap between Latin America and East Asia over the period 1980–2014. We find that the economic growth divergence between the two regions can be primarily explained by the differences in institutions and physical capital. In addition, the results indicate that the higher East Asian performance is not only due to its higher levels of endowments in these factors, but also to the higher efficiency in its use. We connect our results with the 2030 Agenda for Sustainable Development.
... However, such an assumption of homogeneity is vulnerable in practice given that individual heterogeneity has been widely documented in empirical studies using panel data. See, e.g., Durlauf (2001) and Su and Chen (2013) for cross-country evidence and Browning and Carro (2007) for ample microeconomic evidences. In panel threshold regressions, heterogeneity can exist in not only the slopes but also the threshold coefficients. ...
In this paper, we consider the least squares estimation of a panel structure threshold regression (PSTR) model where both the slope coefficients and threshold parameters may exhibit latent group structures. We study the asymptotic properties of the estimators of the latent group structure and the slope and threshold coefficients. We show that we can estimate the latent group structure correctly with probability approaching 1 and the estimators of the slope and threshold coefficients are asymptotically equivalent to the infeasible estimators that are obtained as if the true group structures were known. We study likelihood-ratio-based inferences on the group-specific threshold parameters under the shrinking-threshold-effect framework. We also propose two specification tests: one tests whether the threshold parameters are homogeneous across groups, and the other tests whether the threshold effects are present. When the number of latent groups is unknown, we propose a BIC-type information criterion to determine the number of groups in the data. Simulations demonstrate that our estimators and tests perform reasonably well in finite samples. We apply our model to revisit the relationship between capital market imperfection and the investment behavior of firms and to examine the impact of bank deregulation on income inequality. We document a large degree of heterogeneous effects in both applications that cannot be captured by conventional panel threshold regressions.
... Durlauf and Johnson (1995) reject the cross-country linear model specification in favor of a multiple regime in which different countries obey different linear models when grouped. According to Durlauf (2001) econometric models should consist in large part of the identification of patterns present in observations, rather than in complex parametric model estimations. ...
The purpose of this research is to provide empirical evidence about what institutions are most likely to favor development in its different stages. Firstly, we identify the three development stages that prevailed in the world between 1996 and 2011 according to the income classification of the World Bank corroborated with data from the UNDP Human Development Index. Secondly, we consider that a country had a “successful” behavior if it improved its development stage in that period. Grouping countries based on “success”, instead of according to the income level, allows us to introduce the dynamics of development in the analysis. Thirdly, we formulate a panel data and a probit model to determine the institutions that are behind the success cases. The results identified economic freedom as the most important institution in all development stages; governance was also found essential, but only in the countries in the intermediate stage of development.
... 17 Advocates of this approach argue that the Barro-type growth regressions create biases in the estimated coefficients by pooling data whenever there is heterogeneity in the parameters. Moreover, cross-sectional regressions lead to a waste of information, since they ignore unit-specific time variations in growth rates and prevent the estimation of a steady state for each region or country separately ( e.g. Lee et al., 1997;Temple, 1999;Pritchett, 2000;Durlauf, 2001;Brock and Durlauf, 2001;Masanjala and Papageorgiou, 2004). 18 Canova and Marcet (1995) propose a way to model heterogeneity and calculate steady states for each unit without proxying for the steady state of income with additional variables. ...
... To overcome these limitations, Durlauf (2001) suggests the use of semiparametric methods. This approach allows us to tackle the possible non-linear effect of economic development on regional inequality in a flexible way. ...
... Our model is semiparametric -that is, we represent the coefficients on all regressors in all equations as unknown smooth functions of a measure of institutional quality, and unobserved country-and yearspecific factors (fixed effects) -for a few reasons. One, we adopt the view that substantial linear and nonlinear forms of parameter heterogeneity that stem from, among other sources, cross-country differences in institutional quality, exist in empirical growth models (see, for e.g., Durlauf 2001, Minier 2007, Durlauf, Kourtellos & Tan 2008, Huynh & Jacho-Chávez 2009a, Huynh & Jacho-Chávez 2009b, and in particular in the effect of FDI on economic growth (see, for e.g., Borensztein et al. 1998, Alfaro et al. 2004, Durham 2004, Papaioannou 2009, Kottaridi & Stengos 2010, Delgado et al. 2014, McCloud & Kumbhakar 2012. Thus, we do not assume a priori that all countries use identical technologies to produce goods and services. ...
We characterize the types of interactions between foreign direct investment (FDI) and economic growth, and analyze the effect of institutional quality on such interactions. To do this analysis, we develop a class of instrument-based semiparametric system of simultaneous equations estimators for panel data and prove that our estimators are consistent and asymptotically normal. Our new methodological tool suggests that across developed and developing economies, causal, heterogeneous symbiosis and commensalism are the most dominant types of interactions between FDI and economic growth. Higher institutional quality facilitates, impedes or has no effect on the interactions between FDI and economic growth.
... Una limitaci6n importante que presenta la ma yorfa de las analisis empfricos sabre crecimiento re gional basados en datos de carte transversal ha sido el supuesto de que solo hay un estado estacionario para todas las economfas regionales de la muestra (Durlauf, 2001 ). Si, par ejemplo, las economfas re gional es difieren unas de otras en algunos de sus parametros basicos de crecimiento (coma la capaci dad de innovacion tecnologica, el desarrollo del ca pital humano, etc. ), o bien el desbordamiento o di fusi6n indirecta de conocimiento entre ellas es debil, es posible que no converjan hacia una renta per ca pita comun, sino hacia diferentes niveles de equilibria de renta per capita especfficos de cada economfa. ...
Este trabajo generaliza la metodologla clasica de investigaci6n para el analisis de la convergencia de corte transversal, de modo que sea posible que el proceso de convergencia pueda variar entre los distintos clubes de regiones y se expliquen las posibles interacciones y la code pendencia del crecimiento en Europa en el periodo 1995-2000. Encon tramos claros indicios de clubes de convergencia en Europa durante el periodo de observaci6n. La muestra de economias regionales que per tenecen al club A converge en un sentido incondicional a un ritmo anual del 1,5 par 100, y las que pertenecen al club B (economias regionales de la Europa central y oriental y Europa meridional) a un ritmo del 2.4 par 100. Es importante subrayar que ese 1, 5 6 2.4 por 100 anual, aun que coincide con las resultados de estudios anteriores sabre conver gencia, es una tasa muy baja. En el estudio tambien se expone que la metodologia clasica de ana lisis de convergencia, que ha sido el caballo de batalla de la mayoria de las estudios de convergencia anteriores en la economfa convencional, esta mal disenada para el analisis de la convergencia regional. En particular, pasar par alto la presencia de la autocorrelaci6n espacial en las termi nos de error en el analisis de convergencia llevado a cabo con datos de corte transversal puede conducir a conclusiones err6neas, por ejemplo, con respecto a la valoraci6n de la velocidad de convergencia. Por lo tan to. contrastar la presencia de autocorrelaci6n (heterogeneidad) espacial par medio de los tests adecuados e introducir especificaciones alterna tivas de la ecuaci6n de convergencia cuando sea necesario son asuntos cruciales en el analisis de la convergencia en renta. Palabras clave: regiones europeas, convergencia en renta, clubes de convergencia, econometrfa espacial. I. INTRODUCCION A principios de siglo, todo lo referente a la con vergencia se ha convertido en una cuesti6n de creciente interes para las poHticos de Euro pa, y no solo dentro de la UE-15, sino tambien en re laci6n con los nuevos paises miembros y paises as pirantes del centro y Este de Europa. Evaluar el alcance de la convergencia es alga que dista mu cho de ser sencillo a causa de diversas razones. En primer lugar, hay problemas de medida. En particu lar, las datos sabre el producto regional bruto (PRB) de los paises de la Europa central y oriental son muy poco fiables. Quiza se deba en parte a que han cam biado las convenciones contables que se utilizan ahora en las economfas de estos palses. Como con secuencia de ello, es diffcil comparar las datos del PRB 30 Abstract This study generalises the classic research methodology for cross-sec tion {3 convergence analysis, so that it may be possible for the conver gence process to be able to vary between the different regional clubs and for the possible interactions and co-dependence of growth in Europe to be explained in the period 1995-2000. We find clear signs of convergence clubs in Europe during the period of observation. The sample of regio nal economies that belong to club A converge in a unconditional sen se at an annual rate of 1.5 percent, and those belonging to club B (Cen tral and Eastern European and Southern European regional economies) at a rate of 2 .4 percent. It is important to underline that, although in line with the results of earlier convergence studies, that annual 1.5 or 2.4 per cent is a far lower rate. In the study we also explain that the classic methodology of con vergence analysis, which has been a recurring theme of most of the ear lier convergence studies in conventional economics, is not well designed for analysing regional convergence In particular, overlooking the presence of spatial autocorrelation in terms of error in the convergence analysis carried out with cross-section data may lead to mistaken conclusions, for instance, with regard to assessment of the convergence rate. Therefo re, verifying the presence of spatial autocorrelation (heterogeneity) by means of suitable tests and introducing alternative convergence equa tion specifications when necessary are crucial matters in the analysis of income convergence de las regiones de estos paises con las de la UE has ta mediados de la decada de 1990 (European Com mission, 1999). En segundo lugar, no existe un marco de con senso metodol6gico par el que se guie el trabajo em pirico de analisis de la convergencia regional. En su lugar, hay varios tipos de convergencia, y cada uno de ellos necesita una ecuaci6n distinta. En terminos generales, los analisis em pf ricos se dividen en tres categorias. El primer tipo, y el mas dominante de to dos, analiza la correlaci6n de coste transversal entre los niveles de producci6n per capita y las tasas de crecimiento subsiguientes de los pafses o regiones. Si la correlaci6n es negativa, se considera demostrado que la convergencia como tal implica que (par ter mino media) las economfas con una baja renta per
... While it seems that there is a broad consensus in the profession that a reasonable estimate for the speed of convergence lies around 2%, the results of different econometric studies vary wildly: Abreu et al. (2005) analyze 48 articles with 619 estimated values for the speed of convergence and show that the estimates range from negative values to the maximum of 65.59%. This huge dispersion can be attributed partly to the use of different specifications, different control variables, and different sample sizes, the presence of measurement errors, and to endogeneity issues (see, for example, Durlauf, 2001;Durlauf et al., 2005). However, purely methodological aspects also seem to play an important role: Abreu et al. (2005, p. 410) note that generalized method of moments (GMM) estimators and the corrected least squares dummy variable (LSDVC) technique yield substantially higher estimates than other approaches and Hsiao et al. (2002) show in Monte-Carlo studies that the biases of GMM-based estimators can be large. ...
... 5 Third, endogeneity has been labeled as an important source of bias driving the results of many early empirical macroeconomic studies. Durlauf (2001), for example, argues that most aggregate macroeconomic variables are potentially endogenous in an economic growth specification; many empirical investigations focusing on growth effects of FDI have explored instrumental variables techniques to mitigate 5 Now classic research documenting these findings include Durlauf and Johnson (1995), Liu and Stengos (1999), and Durlauf et al (2001). More recent contributions documenting the importance of parameter heterogeneity and model flexibility include Maasoumi et al (2007), Minier (2007), Henderson et al (2012b), McCloud and Kumbhakar (2012), and Delgado et al (2014). ...
We investigate heterogeneity between foreign direct investment (FDI) and domestic investment induced by corruption and human capital. Controlling for corruption and human capital, inbound FDI has significant, heterogeneous complementarity effects on domestic investment; the effect of outbound FDI on domestic investment is fluid: substitution and complementarity exist, and change direction over time. The fluid effects of outbound FDI oppose the popular dollar-for-dollar hypothesis. Although lower corruption and higher human capital strengthen, weaken, or do not change the degree of these FDI effects, the data are inconsistent with the hypothesis of a global optimum for corruption or human capital. Corruption and human capital do not appear to be binding constraints in all countries. The role of institutional quality appears consistent with the prediction of the General Theory of Second Best.
... Our empirical specification is similar to that used in the convergence literature. Implicit in the empirical specification is the idea that each economy has a steady-state growth path that follows a time trend.Quah (1993) provides cross-country evidence on income growth that refutes this assumption.Durlauf (2001) raises other potential problems, such as nonlinearities, a disconnect between growth theory and empirical modeling (i.e., which variables should be included in growth models and the potential problem of simultaneity), and, finally, heterogeneous parameters. We argue that differences across states in terms of heterogeneous parameters and ...
The authors extend earlier models of economic growth and development by exploring the effect of economic freedom on U.S. state employment growth. They find that states with greater economic freedom-defined as the protection of private property and private markets operating with minimal government interference-experienced greater rates of employment growth. In addition, they find that less-restrictive state and national government labor market policies have the greatest impact on employment growth in U.S. states. Beyond labor market policies, state employment growth is influenced by state and local government policies, but not the policies of all levels of government, including the national government. Their results suggest that policymakers concerned with employment should seriously consider the degree to which their own labor market policies and those of the national government may be limiting economic growth and development in their respective states.
... We also employ the finite mixture models estimation approach to deal with possible sources of heterogeneity in our specification. Parameters may in fact vary in cross-country and panel growth regressions (Durlauf, 2001). Moreover, the standard parametric models do not take into account the unobserved heterogeneity due, for example, to institutional, cultural and geographical factors, initial conditions of the countries (Bloom, Canning, & Sevilla, 2003;Desdoigts, 1999), or omitted variables in the specification (Aitkin, 1999). ...
We investigate the implications of the OECD support for agriculture on the headcount
poverty rates of developing countries. Following the strategy proposed by McMillan,
Peterson Zwane, and Ashraf (2005, 2007), we estimate the cross-country empirical
framework building a new OECD policy index and applying different panel data techniques.
In addition to the standard parametric model estimators, we also use the finite mixture
models to detect heterogeneous effects of the OECD policy index within our sample. We find
statistically significant evidence that OECD agricultural policies worsened the poverty rates
in some developing countries. Most of the main food exporters appear in fact to be
negatively affected by the OECD support for agriculture while the impact detected on food
importers is not significant or near to zero.
... However, if the agents are allowed to be heterogeneous the dynamic system of the neoclassical growth model could lead to multiple steady-state equilibrium in spite of diminishing returns to capital. Durlauf (2001) points out that a key limitation of the majority of empirical analyses of cross-sectional regional growth has been that the assumption of a single steady-state has to hold for all the regional economies in the sample, which is the case for absolute and conditional convergence hypotheses. The club convergence hypothesis, on the other hand, allows multiple and local stable steady-state equilibriums only. ...
Over the past two decades, the issue of regional convergence in the European Union has been the subject of a wide range of empirical research. This paper aims to provide more information on the differences in regional growth patterns of new member states (NMS), as well as Croatia, in addition to the factors influencing regional disparities within each country. This research provides an analysis of regional convergence in the period 2001-2008 at the NUTS II and NUTS III level.
The most widely used model for testing convergence hypotheses is beta-convergence analysis. Other factors commonly included in the econometric modelling of convergence are demographic variables, labour market conditions, industrial structure, institutional factors and overall government policy. The main hypothesis is that the process of regional convergence in NMS and Croatia is not strong enough to dominate over other factors, influencing regional potential growth (mainly industry structure and quality of human capital). Absolute β-convergence can be found at the national level for EU countries. Convergence also can be found for NMS regions, but the pace of convergence on the regional level is lower in comparison to the national level and the estimated β-convergence parameter is less significant.
... Since our generalized model does not require a priori specification of the functional form of the regression coefficient functions, we are able to avoid any potential pitfalls associated with misspecification of the functional form of the coefficients. The advantage of such a generalization is to analyze heterogeneity within the corruption-FDI-growth nexus and address many of the concerns discussed by Durlauf (2001), in a unified framework. ...
Who really wins from foreign direct investment (FDI) and by how much? Should winners care about corruption? Building on evidence of heterogeneity in the FDI-growth relationship, we propose a semiparametric model that allows corruption to influence the relationship between the conditioning variables and GDP growth, parameter heterogeneity of unknown form, and the use of instrumental variables. We find evidence that corruption has a sizeable nonlinear role in the FDI-growth relation, weakening the effectiveness of FDI at improving growth rates in many developing countries. Developing countries with insignificant or low returns to FDI may benefit substantially from reducing corruption.
We analyze the contribution of human capital and the labor augmenting component to per capita GDP using a finite mixture model. We find strong evidence of heterogeneity in sub-models explaining levels of real per capita GDP and extremely weak evidence of heterogeneity in those determining rates of growth, with lack of convergence across groups. Furthermore, by conditioning the country-specific cluster prior probability to the initial level of human capital, we account for the feedback effect between human capital and initial conditions. In this way, we provide an interpretation of the human capital puzzle and document that initial schooling rate levels strongly affect the dynamic of country-specific growth. We also show that the heterogeneous impacts of human capital and the labor augmenting component on GDP are crucially affected by differences in latent variables and, most significantly, by quality of education, legal structure, and property rights’ regulation.
Global value chains (GVCs) promote the diffusion of knowledge and technology. This paper develops an empirical production model that combines spillovers and productivity growth heterogeneity at the industry‐level. We exploit the GVCs linkages from inter‐country input–output tables to describe the interdependencies of technology within the input–output space and combine that with the Asian‐Pacific and US KLEMS database to estimate productivity growth. The spillover effects from capital deepening, intermediate deepening, and technical change are identified and decomposed into domestic and international effects. Our empirical results find that ignoring the industrial interactions leads to an overestimation of China's productivity growth, and underestimation for the US and Japan. The spillovers of capital and intermediate are found to be significantly positive. The spillovers of technical change received account for 32 to 40 percent of total productivity growth. These findings provide a better understanding of how technical changes are distributed and diffused within GVCs.
This paper investigates the evolution of regional income inequality in Turkey from the standpoint of income convergence hypotheses and tested whether regional income convergence is accompanied by poverty convergence during the 2004-2019 period. This study is novel in suggesting the clustered growth patterns of Turkish regions for the absence of regional poverty convergence in a context of strong economic growth accompanied by significant reductions in regional income discrepancies.
This paper discusses parametric, nonparametric, and semiparametric models of urban growth. To illustrate differences across approaches, we test Gibrat's law in the long run, using the three methods and three different datasets: Spanish capital cities and regions (1900–2011, annual data) and US MSAs (1900–2000, decennial data). Our results reveal that the estimation of the relationship between growth and initial size can significantly vary across methods. We suggest and encourage the use of semiparametric methods in future research of urban growth.
Now in its fourth edition, this comprehensive introduction of fundamental panel data methodologies provides insights on what is most essential in panel literature. A capstone to the forty-year career of a pioneer of panel data analysis, this new edition's primary contribution will be the coverage of advancements in panel data analysis, a statistical method widely used to analyze two or higher-dimensional panel data. The topics discussed in early editions have been reorganized and streamlined to comprehensively introduce panel econometric methodologies useful for identifying causal relationships among variables, supported by interdisciplinary examples and case studies. This book, to be featured in Cambridge's Econometric Society Monographs series, has been the leader in the field since the first edition. It is essential reading for researchers, practitioners and graduate students interested in the analysis of microeconomic behavior.
Potential growth and overall economic development across nations since 1990s till 2018 is functioning as a 'wake-up' call for researchers and policy makers to process for investigation the phenomenon of HDI convergence globally via two framework: σ- and unconditional β- convergence. A proper investigation of such knowledge is of utmost important for regulators of economies, suggesting whether the difference in HDI level across nations has narrowed, including the period of global recession (2007/09). The empirical evidence is analysed from different methods that act in a complementary way: i) bootstrap quantile regression approach in static panel framework, which allows the assumption of nonlinearities, especially for those groups that have shared a common path in terms of economic development while estimating the convergence results. ii) pooled OLS model, which seek to examine the soundness of the evaluated convergence results and confirmed the validity of our core applied model. All methods revealed strong evidences of unconditional β- convergence for all HDI groups, reflecting all classified HDI groups had potential to increase their tendency toward convergence. This is mainly driven by both the slower forward mobility with the less developed nations and faster backward mobility with the more developed nations. The σ- convergence estimates provides further insights and suggests that the formation and comprising of multiple convergence HDI clusters is a salient feature of reduction in inequality and dispersion across medium and lower nations.
In studies of firm's innovation performance, regression analysis can involve a significant level of model uncertainty because the ‘true’ model, and therefore the appropriate set of explanatory variables are unknown. Drawing on innovation survey data for France, Germany, and the United Kingdom, we assess the robustness of the literature on inbound open innovation to variable selection choices, using Bayesian model averaging (BMA). We investigate a wide range of innovation determinants proposed in the literature and establish a robust set of findings for the variables related to the introduction of new-to-the-firm and new-to-the-world innovation with the aim of gauging the overall healthiness of the literature. Overall, we find greater robustness for explanations for new-to-the-firm rather than new-to-the-world innovation. We explore how this approach might help to improve our understanding of innovation.
This article investigates the rise of mathematical and quantitative methods in three leading economics journals - American Economic Review, Journal of Political Economy, and Quarterly Journal of Economics - from 1940 to 2010. We show that the trajectories describing the evolution of mathematical and quantitative methods in these outlets are strikingly similar. Analyzing theoretical and applied research separately, the former follows an inverted-U path while the latter has been growing in importance throughout the period, especially after 1990. Moreover, cointegration methods are used to investigate the existence of a long-term relationship between the time series. Our results suggest that whereas the three journals mutually influenced one another, different dynamics emerge when theoretical and applied research are considered separately.
This paper analyses the long‐term relationship between regional inequality and economic development. Our data set includes information on national and regional per‐capita GDP for four countries: France, Italy, Portugal and Spain, compiled on a decadal basis for the period 1860‐2010. Using parametric and semiparametric regressions, our results confirm the rise and fall of regional inequalities over time although in recent decades they are on the rise again. Finally, we identify structural change as being a significant transmission mechanism of the inverted‐U relationship. The arrival of technological shocks, beginning during the onset of industrialization, and the transition from agrarian to industrial economies, would explain this result.
The empirical growth literature is dominated by accounting and regression methods that assume common production technology across countries. Our empirical model relaxes this assumption and further allows unobservable determinants of output (total factor productivity, TFP) to differ across countries and time, while accounting for endogeneity and cross‐section correlation arising from global shocks. Using manufacturing sector data for 48 economies, we show that the assumption of common technology is rejected in our regressions. We illustrate that the erroneous choice of homogeneous technology has a substantial impact on the patterns and magnitudes of the resulting TFP estimates.
This paper uses a general Lotka–Volterra model to estimate convergence for 93 countries over the period 1960–2007. It employs an equation with a spatial time lag and common factors. The spatial lag controls for spatial dependence, while the common factors control for strong cross-sectional dependence. As spatial weights matrices, the shares of high-skilled migrants, trade shares and foreign direct investments are used. A simultaneous least squares estimator and a dynamic common correlated effects (DCCE) estimator are employed. The DCCE estimator finds conditional convergence. The paper highlights the importance of controlling for both types of cross-sectional dependence.
New macro empirical evidence is provided to assess the relative importance of object and idea gaps in explaining the world income distribution dynamics. Formal statistical hypothesis tests allow us to discriminate between two competing growthmodels: (i) the standard neoclassical growth model similar to that employed by Mankiw, Romer, and Weil (1992), (ii) an extension of the Nelson and Phelps’ approach (1966) that emphasizes the importance of technology transfer in addition to factors accumulation. First, the latter model better characterizes international data at an aggregate level. It cannot be rejected as a null hypothesis and is significantly preferred to a standard neoclassical model. Second, robust to sample selection evidence suggests that the high social returns to investment in equipment (as opposed to structure) reflect technology transfer mediated through capital goods. Finally, technological catch-up mostly benefits "socially" advanced economies and largely contributes to the polarization of the world income distribution
The East Asian experience shows that fast economic growth could be achieved along with narrowing income inequality, especially for the case of Japan, Korea and Chinese Taiwan. This chapter tries to explore the growth–inequality nexus in China by argues that the conventional approach of data averaging is problematic. It introduces the polynomial inverse lag (PIL) framework so that the impacts of inequality on investment, education, and ultimately on growth can be measured at precisely defined time lags. Combining PIL with simultaneous systems of equations, we analyze the growth–inequality relationship in postreform China, finding that this relationship is nonlinear and is negative irrespective of time horizons.
In this paper we argue that informal institutions, or more generally, what is called social capital, can act as one of the mediating factors determining the size and the direction of foreign direct investment-induced spillovers on growth and productivity at regional level. The idea is that when a foreign firm sets up a new production plant in a location, the nature and the quality of its relationships with local workers and firms are affected by the endowment of social capital of that location. A ‘wrong’ social capital may make these relationships difficult, thus limiting the capacity of the host economy to convert foreign direct investment-induced spillovers into local competencies conducive to growth. We operationalized informal institutions in terms of generalized trust, associational activity, and cultural closeness and we found that spillover effects do not arise: (1) when generalized trust is too ‘self-referential’, (2) the level of associational activities is low, and (3) cultural closeness towards foreigners and external culture dominate the society. We also found that foreign presence is not universally beneficial, since positive spillovers are associated with EU-originating foreign firms and foreign direct investment in services only. These results have policy implications for the EU regions. In order to maximize the returns from foreign direct investment, the issue of the origin of foreign investors as well as the sectoral composition of foreign direct investment inflows should be carefully considered. Furthermore, investments in education may help regions to benefit more from the foreign presence because human capital and social capital are likely to be complementary.
Purpose
This paper aims to test the political-economy hypothesis that country sizes are related with constraints associated with Chinese trade.
Design/methodology/approach
This study applies a generalized linear mixed approach on panel data of Southern African (henceforth SADC) economies from 2001 to 2014 to observe common Chinese trade patterns among SADC countries.
Findings
Empirical results support the hypothesis that structural differences exist and smaller SADC countries are disadvantaged in their trade relations with China.
Research limitations/implications
This paper is exploratory by nature. Its scope and the depth of analysis is constrained by data availability.
Originality/value
The manuscript has been approved by the author and has never been published, or has been considered for publication elsewhere.
This paper analyzes the causal relationships in regional technological systems within a structural vector autoregression framework. Applying a data-driven independent component analysis, it shows how the regional dynamics of economic, research, innovation and educational activities affect each other instantaneously and over time. By matching differently classified data on employees, patents and graduates, the analysis is based on a unique database which embraces multi-dimensional aspects for five clearly separated industries. The findings on how industry specific growth processes unfold are explained by referring to the type of industry and its knowledge base. For instance, it is found that more engineering-oriented industries like machine tools show a success-driven pattern, whereas science-based industries show a stronger dependence on universities. Such knowledge on the causal relations is of utmost relevance to the design and implementation of policy instruments.
This paper applies Zipf rank approach to measure how long it will take for the individual economy to reach the final state of equilibrium by using local-level data of China's urban areas. The indicators, the gross domestic product (GDP) per capita and the market capitalization (MCAP) per capita of 150 major cities in China are used for analyzing their convergence. Besides, the power law relationship is examined for GDP and MCAP. Our findings show that, compared to the classical approaches: β-convergence and σ-convergence, the Zipf ranking predicts that, in approximately 16 years, all the major cities in China will reach comparable values of GDP per capita. However, the MCAP per capita tends to follow the periodic fluctuation of the economic cycle, while the mean-log derivation (MLD) confirms the results of our study. Moreover, GDP per capita and MCAP per capita follow a power law with an average value of α=0.41 which is higher than α=0.38 obtained based on a large number of countries around the world.
I review the empirical comparative law literature with an emphasis on quantitative work. After situating the field and surveying its main applications to date, I turn to methodological issues. I discuss at length the obstacles to causal inference from comparative data, and caution against inappropriate use of instrumental variables and other techniques. Even if comparative data cannot identify any single causal theory, however, they are extremely important in narrowing down the set of plausible theories. I report progress in measurement design and suggest improvements in data analysis and interpretation using techniques from other fields, particularly growth econometrics.
Recent years have seen a wealth of articles on spatial effects in empirical growth specifications. Endogenous growth models, together with the arguments of the new economic geography, have led to spatial dependence being identified with the existence of externalities which cross regional borders. This article continues in this line of research and offers new empirical evidence on the contribution of human capital and agglomeration economies to the differences in productivity in European Union regions. The article uses the spatial Durbin model with different weight matrices to explain the relation between human capital variables and labour productivity, with the advantage that it allows the effects of spatial externalities associated to human capital and agglomeration economies to be quantified.
We re-examine the concept of beta-convergence in living standards across countries during the period 1980–2012. In this study, well-being is assessed using the Human Development Index (HDI) which considers income aspects as well as social indicators, thus reflecting the multidimensional nature of this process. The existence of sigma convergence is evidenced in this study and hence beta convergence, as a necessary condition, is also pointed out. However, the linearity of this process has been questioned. Therefore, we apply a semiparametric specification of this process to the HDI and each of its intermediate indices. These models allows for nonlinearities in the estimation of the convergence speed. Our results reveal that absolute convergence in human well-being is satisfactorily represented by the conventional linear specification. However the income and education indices show nonlinear patterns. We also include structural variables to capture differences in the steady-state (conditional convergence). Under this model, convergence speed of all indicators is higher and the convergence process seems to be linear only for the health index.
This paper uses a model of growth and imperfect capital mobility across multiple economies to characterize the dynamics of (cross-country) income distributions. This allows convenient study of the convergence hypothesis, and reveals, where appropriate, polarization and clumping within subgroups. The data show little cross-country convergence; instead, the important features are persistence, immobility, and polarization, exemplified by "convergence club" or "twin peaks" dynamics. Copyright 1996 by Kluwer Academic Publishers
This paper studies cross-country patterns of economic growth from the viewpoint of income distribution dynamics. Such a perspective raises new empirical and theoretical issues in growth analysis: the profound empirical regularity is an "emerging twin peaks" in the cross-sectional distribution, not simple patterns of convergence or divergence. The theoretical problems raised concern interaction patterns among sub-groups of economies, not only problems of a single economy's accumulating factor inputs and technology for growth.
For experiments where the strength of association between a response variable Y and a covariate X is different over different regions of values for the covariate X, we propose local nonparametric dependence functions which measure the strength of association between Y and X as a function of . Our dependence functions are extensions of Galton's idea of strength of co-relation from the bivariate normal case to the nonparametric case. In particular, a dependence function is obtained by expressing the usual Galton-Pearson correlation coefficient in terms of the regression line slope and the residual variance and then replacing and by a nonparametric regression slope and a nonparametric residual variance , respectively. Our local dependence functions are standardized nonparametric regression curves which provide universal scale-free measures of the strength of the relationship between variables in nonlinear models. They share most of the properties of the correlation coefficient and they reduce to the usual correlation coefficient in the bivariate normal case. For this reason we call them correlation curves. We show that, in a certain sense, they quantify Lehmann's notion of regression dependence. Finally, the correlation curve concept is illustrated using data from a study of the relationship between cholesterol levels x and triglyceride concentrations y of heart patients.
Countries' geographic characteristics have important effects on their trade, and are plausibly uncorrelated with other determinants of their incomes. This paper therefore constructs measures of the geographic component of countries' trade and uses those measures to obtain instrumental variables estimates of the effect of trade on income. The results suggest that ordinary least squares estimates understate the effects of trade, and that trade has a quantitatively large, significant, and robust positive effect on income.
This paper studies cross-country patterns of economic growth from the viewpoint of income distribution dynamics. Such a perspective raises new empirical and theoretical issues in growth analysis: the profound empirical regularity is an "emerging twin peaks" in the cross-sectional distribution, not simple patterns of convergence or divergence. The theoretical problems raised concern interaction patterns among subgroups of economies, not only problems of a single economy's accumulating factor inputs and technology for growth. Copyright 1997 by Kluwer Academic Publishers
A vast literature uses cross-country regressions to search for empirical linkages between long-run growth rates and a variety of economic policy, political, and institutional indicators. This paper examines whether the conclusions from existing studies are robust or fragile to small changes in the conditioning information set. The authors find that almost all results are fragile. They do, however, identify a positive, robust correlation between growth and the share of investment in GDP and between the investment share and the ratio of international trade to GDP. The authors clarify the conditions under which there is evidence of per capita output convergence. Copyright 1992 by American Economic Association.
[eng] Transportation costs and monopoly location in presence of regional disparities. . This article aims at analysing the impact of the level of transportation costs on the location choice of a monopolist. We consider two asymmetric regions. The heterogeneity of space lies in both regional incomes and population sizes: the first region is endowed with wide income spreads allocated among few consumers whereas the second one is highly populated however not as wealthy. Among the results, we show that a low transportation costs induces the firm to exploit size effects through locating in the most populated region. Moreover, a small transport cost decrease may induce a net welfare loss, thus allowing for regional development policies which do not rely on inter-regional transportation infrastructures. cost decrease may induce a net welfare loss, thus allowing for regional development policies which do not rely on inter-regional transportation infrastructures. [fre] Cet article d�veloppe une statique comparative de l'impact de diff�rents sc�narios d'investissement (projet d'infrastructure conduisant � une baisse mod�r�e ou � une forte baisse du co�t de transport inter-r�gional) sur le choix de localisation d'une entreprise en situation de monopole, au sein d'un espace int�gr� compos� de deux r�gions aux populations et revenus h�t�rog�nes. La premi�re r�gion, faiblement peupl�e, pr�sente de fortes disparit�s de revenus, tandis que la seconde, plus homog�ne en termes de revenu, repr�sente un march� potentiel plus �tendu. On montre que l'h�t�rog�n�it� des revenus constitue la force dominante du mod�le lorsque le sc�nario d'investissement privil�gi� par les politiques publiques conduit � des gains substantiels du point de vue du co�t de transport entre les deux r�gions. L'effet de richesse, lorsqu'il est associ� � une forte disparit� des revenus, n'incite pas l'entreprise � exploiter son pouvoir de march� au d�triment de la r�gion l
Varying-coefficient models are a useful extension of the classical linear models. The appeal of these models is that the coefficient functions can easily be estimated via a simple local regression. This yields a simple one-step estimation procedure. Weshow that such a one-step method can not be optimal when different coefficient functions admit different degrees of smoothness. This drawback can be repaired by using our proposed two-step estimation procedure. The asymptotic mean-squared errors for the two-step procedure is obtained and is shown to achieve the optimal rate of convergence. A few simulation studies show that the gain by the two-step procedure can be quite substantial. The methodology is illustrated by an application to an environmental dataset.
We explore a class of regression and generalized regression models in which the coefficients are allowed to vary as smooth functions of other variables. General algorithms are presented for estimating the models flexibly and some examples are given. This class of models ties together generalized additive models and dynamic generalized linear models into one common framework. When applied to the proportional hazards model for survival data, this approach provides a new way of modelling departures from the proportional hazards assumption.
We provide an overview of recent empirical research on patterns of cross-country growth. The new empirical regularities considered differ from earlier ones, e.g., the well-known Kaldor stylized facts. The new research no longer makes production function accounting a central part of the analysis. Instead, attention shifts more directly to questions like, Why do some countries grow faster than others? It is this changed focus that, in our view, has motivated going beyond the neoclassical growth model.
For 98 countries in the period 1960–1985, the growth rate of real per capita GDP is positively related to initial human capital
(proxied by 1960 school-enrollment rates) and negatively related to the initial (1960) level of real per capita GDP. Countries
with higher human capital also have lower fertility rates and higher ratios of physical investment to GDP. Growth is inversely
related to the share of government consumption in GDP, but insignificantly related to the share of public investment. Growth
rates are positively related to measures of political stability and inversely related to a proxy for market distortions.
We investigate the issue of model uncertainty in cross-country growth regressions using Bayesian Model Averaging (BMA). We find that the posterior probability is spread widely among many models, suggesting the superiority of BMA over choosing any single model. Out-of-sample predictive results support this claim. In contrast to Levine and Renelt (1992), our results broadly support the more 'optimistic' conclusion of Sala-i-Martin (1997b), namely that some variables are important regressors for explaining cross-country growth patterns. However, care should be taken in the methodology employed. The approach proposed here is firmly grounded in statistical theory and immediately leads to posterior and predictive inference. Copyright © 2001 John Wiley & Sons, Ltd.
This paper provides some new evidence on the behavior of cross-country growth rates. We reject the linear model commonly used to study cross-country growth behavior in favor of a multiple regime alternative in which different economies obey different linear models when grouped according to initial conditions. Further, the marginal product of capital is shown to vary with the level of economic development. These results are consistent with growth models which exhibit multiple steady states. Our results call into question inferences that have been made in favor of the convergence hypothesis and further suggest that the explanatory power of the Solow growth model may be enhanced with a theory of aggregate production function differences. Copyright 1995 by John Wiley & Sons, Ltd.
this paper, we produced the curves fi 0 (E) and
The article proposes a technique, based on the predictive density of the data, conditional on the parameters of the model, to jointly tests for groups of unknown size in a panel and to estimate the parameters of each group. The procedure is applied to the problem of identifying convergence clubs in scaled income per capita data. The steady-state distribution of European regional data clusters around four poles of attraction with different economic features. The distribution of income per capita of OECD countries has two poles of attraction and each group clearly identifiable economic characteristics.
We share the uncommonness of being different.
J. P. Roche
Determinants of long-term economic growth: robustness tests and model averaging. Columbia University
- G Doppelhofer
Doppelhofer, G., 1999. Determinants of long-term economic growth: robustness tests and model averaging. Columbia University, Mimeo.
Sensitivity analysis of cross-country growth regressions Empirical limits for time series econometric models Convergence empirics across economies with (some) capital mobility Empirics for growth and distribution: polarization, strati"cation, and convergence clubs
- R Levine
- D Renelt
- P Phillips
- W Ploberger
Levine, R., Renelt, D., 1992. Sensitivity analysis of cross-country growth regressions. American Economic Review 83, 942}963. Phillips, P., Ploberger, W., 1999. Empirical limits for time series econometric models. Yale Univer-sity, Mimeo. Quah, D., 1996. Convergence empirics across economies with (some) capital mobility. Journal of Economic Growth 1, 95}124. Quah, D., 1997. Empirics for growth and distribution: polarization, strati"cation, and convergence clubs. Journal of Economic Growth 2, 27}59. S.N. Durlauf / Journal of Econometrics 100 (2001) 65}69