#### DMCA

## The new empirics of economic growth (1998)

Citations: | 351 - 14 self |

### Citations

3445 | X.: Economic growth - Barro, Sala-i-Martin - 1999 |

2991 | Endogenous Technological Change
- Romer
- 1990
(Show Context)
Citation Context ...h choices. Our immediate concern is: How do the empirical implications then dier? Certain key details dier, but the models of Aghion and Howitt [1], Grossman and Helpman [53], Jones [56], and Romer =-=[99]-=- all associate the evolution of A with a measurable input such as research and development expenditure, the number of scientists and engineers, and so on. By contrast, models such as those in Lucas [7... |

2882 | Increasing Returns and Long-Run Growth
- Romer
- 1986
(Show Context)
Citation Context ...ly on the topic, economists from at least Kaldor [60] on have marshalled cross-country stylized facts as compelling starting points for discussions about economic growth. Indeed, Lucas [74] and Romer =-=[98, 101]-=- use exactly income comparisons across countries to motivate their endogenous growth analyses. { 3 { Section 5 presents empirical models and critically evaluates the empiricalsndings and methodologies... |

2068 |
A contribution to the theory of economic growth.
- Solow
- 1956
(Show Context)
Citation Context ..._~k1=~k1 can occur only at _~kl=~kl = _~k1=~k1 = _~y=~y = 0; independent of any other economic structure beyond the technology specication. (We will see this below when we study the Solow-Swan model =-=[106, 107, 111]-=-, its general equilibrium Cass-Koopmans version [23, 65], and the modication due to Mankiw, Romer, and Weil [78].) The reasoning just given extends naturally to production technologies beyond Cobb-Do... |

1904 | A Model of Growth Through Creative Destruction
- Aghion, Hewitt
- 1992
(Show Context)
Citation Context ... But h rf(~k) i0 d~k dt = ( ~k1@f(~k)=@~k1; ~k2@f(~k)=@~k2; ) 0B@ _k1=k1 − _H=H − _A=A _k2=k2 − _H=H − _A=A ... 1CA ; so that dening sl(~k) = " ~kl @f(~k)=@~kl f(~k) # (necessarily, sl(~k) 2 =-=[0; 1]-=- and P l sl(~k) 1) we have the growth equation _y=y = ( _H=H + _A=A) + X l sl(~k) n _kl=kl − _H=H − _A=A o ; or _~y=~y = X l sl(~k) _~kl=~kl: (4a) (Equation (4a) refers to both the expressions abo... |

1636 |
A sensitivity analysis of cross-country growth regressions.
- Levine, Renelt
- 1992
(Show Context)
Citation Context ...nditionals-convergence.13 In most empirical studies, the choices of additional control variables are ad hoc across datasets and political units. As one example, the data appendix in Levine and Renelt =-=[72]-=- lists over 50 possibilities. Among the range of controls that have appeared in the literature are the growth of domestic credit, its standard 12 Well-known examples include Barro and Sala-i-Martin [9... |

1579 |
A Contribution to the Empirics of Economic Growth,” Quarterly
- Mankiw, Romer, et al.
- 1992
(Show Context)
Citation Context ...gy specication. (We will see this below when we study the Solow-Swan model [106, 107, 111], its general equilibrium Cass-Koopmans version [23, 65], and the modication due to Mankiw, Romer, and Weil =-=[78]-=-.) The reasoning just given extends naturally to production technologies beyond Cobb-Douglas when the counterpart to P l l (or, more generally, P l sl(~k)) is not constant but always remains strictly... |

1480 |
Technical change and the aggregate production function.
- Solow
- 1957
(Show Context)
Citation Context ...ries of growth. The questions studied in the new empirical growth literature dier from those in earlier empirical work embodying Kaldor’s stylized facts [60] or those in a production function (Solow =-=[107]-=--Denison [32]) accounting exercise. The new literature emphasizes understanding cross-country patterns of income, not the stability within a single economy of factor shares or \great ratios" (the rati... |

1325 | Finance and Growth: Schumpeter might be right. The Quarterly - King, Levine - 1993 |

1074 | Corruption and growth. The Quarterly - Mauro - 1995 |

894 |
The Great Crash, the Oil Price Shock, and the Unit Root Hypothesis. Econometrica 57(6): 1361--1401
- Perron
- 1989
(Show Context)
Citation Context ...apita growth rates. This should be compared with that line of research beginning from the unit-root analyses of Nelson and Plosser [82], extending through the { 34 { breaking-trend research of Perron =-=[84]-=- (and numerous others since), arguing that, over dierent timespans, the time-series properties of dierent income measures do show permanent changes. We do not suggest here that the evidence is decis... |

861 |
Unit root tests in panel data: Asymptotic and finitesample properties
- Levin, Lin, et al.
- 2002
(Show Context)
Citation Context ...eterization of the individual eects, and conclude that those eects do, indeed, dier across economies. Evans, using a dierent statistical technique, concludes the same. Evans follows Levin and Lin =-=[71]-=- and Quah [90] in taking an underlying probability model where both time and cross-section dimensions in the panel dataset are large. This contrasts with standard panel-data studies where the time dim... |

825 |
R&D-Based Models of Economic Growth
- Jones
- 1995
(Show Context)
Citation Context ...he result of such choices. Our immediate concern is: How do the empirical implications then dier? Certain key details dier, but the models of Aghion and Howitt [1], Grossman and Helpman [53], Jones =-=[56]-=-, and Romer [99] all associate the evolution of A with a measurable input such as research and development expenditure, the number of scientists and engineers, and so on. By contrast, models such as t... |

789 | Distributive Politics and Economic Growth,” Quarterly - Alesina, Rodrik - 1994 |

746 | Is Inequality Harmful for Growth - Persson, Tabellini - 1994 |

742 | Long-Run Policy Analysis and Long-Run Growth
- Rebelo
(Show Context)
Citation Context ...rs also in endogenous growth models that exploit externalities and increasing returns (Romer [98]) or in models with the production technology \asymptotically linear" (Jones and Manuelli [59], Rebelo =-=[96]-=-). Our denition of balanced-growth equilibrium compares the growth rates _y=y and _kl=kl. This is not sensible for technology (1b) where we see that _A=A appears with _y=y− _H=H but not with _kl=kl− ... |

738 |
Trends and random walks in macroeconomic time series: Some evidence and implications
- Nelson, Plosser
- 1982
(Show Context)
Citation Context ...Table 1, showing the time-series stability properties of US GDP per capita growth rates. This should be compared with that line of research beginning from the unit-root analyses of Nelson and Plosser =-=[82]-=-, extending through the { 34 { breaking-trend research of Perron [84] (and numerous others since), arguing that, over dierent timespans, the time-series properties of dierent income measures do show... |

734 |
Income Distribution and Macroeconomics
- Galor, Zeira
- 1993
(Show Context)
Citation Context ...vel of per capita output, while economies with high initial incomes or capital stocks converge to a dierent steady-state level. Examples of such models include those by Durlauf [37], Galor and Zeira =-=[51]-=-, and Murphy, Shleifer, and Vishny [80]. The model due to Azariadis and Drazen [4] is particularly convenient for illustrating the empirical dierences between this framework and the neoclassical appr... |

665 | Cross-country evidence on the link between volatility and growth - Ramey, Ramey - 1995 |

594 | Reopening the Convergence Debate: A New Look at Cross-Country Growth Empirics
- Caselli, Esquivel, et al.
(Show Context)
Citation Context ...hat have been employed. Panel data analysis To permit unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet [21], Caselli, Esquivel, and Lefort =-=[22]-=-, Evans [43], Islam [55], Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e... |

575 |
Growth Empirics: A Panel Data Approach', Quarterly
- Islam
(Show Context)
Citation Context ...Panel data analysis To permit unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet [21], Caselli, Esquivel, and Lefort [22], Evans [43], Islam =-=[55]-=-, Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e.g., Chamberlain [24]), ... |

542 | Economic Growth in a Cross-section of Countries,” Quarterly - Barro - 1991 |

488 | The allocation of talent: Implications for growth. Quarterly - Murphy, Shleifer, et al. - 1991 |

475 |
Panel Data
- Chamberlain
- 1984
(Show Context)
Citation Context ...am [55], Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e.g., Chamberlain =-=[24]-=-), many such studies seek to eliminate, in the notation of Section 4, unobservable country-level heterogeneity in A(0). Those heterogeneities, denoted individual eects in the language of panel-data e... |

456 |
Optimum growth in an aggregative model of capital accumulation.
- Cass
- 1965
(Show Context)
Citation Context ...dependent of any other economic structure beyond the technology specication. (We will see this below when we study the Solow-Swan model [106, 107, 111], its general equilibrium Cass-Koopmans version =-=[23, 65]-=-, and the modication due to Mankiw, Romer, and Weil [78].) The reasoning just given extends naturally to production technologies beyond Cobb-Douglas when the counterpart to P l l (or, more generally... |

450 |
The Penn world table (Mark 5): An expanded set of international comparisons
- SUMMERS, HESTON
- 1991
(Show Context)
Citation Context ...n willsgure prominently in discussions below. 2 Hereafter, \the world’s economies" refers to the 122 countries with essentially complete income and population data for 1960{1989 in the Summers-Heston =-=[110]-=- V.6 database. These countries are identied in the Data Appendix below. { 4 { Economists have typically been most interested in growth models as a way to understand the behavior of per capita income ... |

449 | Econometrics . - MADDALA - 1977 |

429 | Time Series Tests of Endogenous Growth Models,” Quarterly
- Jones
- 1995
(Show Context)
Citation Context ...ome will be indistinguishable across A and H improvements. The empirical approach suggested by this reasoning focuses on variables that proxy the eects and economic costs of research activity. Jones =-=[57]-=- notes that the US, for one, has seen neither permanent changes in growth rates nor trend path levels of per capita GDP since 1880. Yet, resources devoted to R&D, by almost any measure, have increased... |

406 |
The Origins of Endogenous Growth.”
- Romer
- 1994
(Show Context)
Citation Context ...ly on the topic, economists from at least Kaldor [60] on have marshalled cross-country stylized facts as compelling starting points for discussions about economic growth. Indeed, Lucas [74] and Romer =-=[98, 101]-=- use exactly income comparisons across countries to motivate their endogenous growth analyses. { 3 { Section 5 presents empirical models and critically evaluates the empiricalsndings and methodologies... |

402 | Convergence across states and regions
- BARRO, SALA-I-MARTIN
- 1991
(Show Context)
Citation Context ...s can place further qualitative restrictions on the behavior of particular aggregates. However, it provides no fundamentally new empirical perspective. Indeed, studies such as Barro and Sala-i-Martin =-=[9, 10]-=-, while using general equilibrium formulations to justify their empirical analyses, typically consider regression models observationally equivalent to the Solow-Swan model with exogenous savings rates... |

390 |
Monitoring the World Economy,
- Maddison
- 1995
(Show Context)
Citation Context ...rd and Durlauf [16, 17], restricts analysis to particular subgroups of economies, for instance the OECD. This allows the researcher to use long time series data, such as those constructed by Maddison =-=[77]-=-. Multivariate unit root and cointegration tests reject the null hypothesis that there is a single unit root process driving output across the OECD economies|thus, across all the economies in the OECD... |

348 |
Intergenerational transfers and the distribution of earnings
- Loury
- 1981
(Show Context)
Citation Context ...7 Economic applications of these tools have also appeared in stochastic growth models (e.g., the examples in Stokey and Lucas (with Prescott) [109, Ch. 16]), income distribution dynamics (e.g., Loury =-=[73]-=-), and elsewhere. Using these ideas for studying distribution dynamics rather than analyzing a time-series stochastic process, say, exploits a duality in the mathematics. This is made explicit in Quah... |

342 |
Multiple regimes and cross-country growth behaviour,
- Durlauf, Johnson
- 1995
(Show Context)
Citation Context ...owever, the results of such a procedure can be dicult to interpret within the Solow-Swan (or Cass-Koopmans) growth model.5 Empirical studies such as Bernard and Durlauf [16, 17], Durlauf and Johnson =-=[39]-=-, and Quah [94] seek to circumvent some of the criticisms we have just described. One strand of this work estimates models that explicitly nest the traditional neoclassical setup. Another strand seeks... |

329 |
Threshold Externalities in Economic Development”,
- Azariadis, Drazen
- 1990
(Show Context)
Citation Context ...ks converge to a dierent steady-state level. Examples of such models include those by Durlauf [37], Galor and Zeira [51], and Murphy, Shleifer, and Vishny [80]. The model due to Azariadis and Drazen =-=[4]-=- is particularly convenient for illustrating the empirical dierences between this framework and the neoclassical approach. The Azariadis-Drazen model works o 9 Increasing returns to scale, of the ki... |

327 |
Empirical cross-section dynamics in economic growth
- Quah
- 1993
(Show Context)
Citation Context ...e accounting explicitly for the multivariate nature of aggregate socioeconomic characteristics, warrants further study. Distribution dynamics Bianchi [18], Desdoigts [33], Jones [58], Lamo [67], Quah =-=[88, 89, 92, 94]-=- have studied the predictions of the theoretical growth models in terms of the behavior of the entire cross-section distribution. While this work is often quite technical, it can be viewed as just a w... |

317 | Macroeconomic determinants of growth: Crosscountry evidence - Kormendi, Meguire - 1985 |

302 |
I just ran two million regressions
- SALA-I-MARTIN
- 1997
(Show Context)
Citation Context ...ions are not typically normal. { 48 { about the fragility of regression coecients. Hence, while wesnd the Levine and Renelt analysis suggestive, the import of the challenge is unclear. Sala-i-Martin =-=[104]-=- has attempted to deal with this limitation by calling \robust" only those variables found statistically signicant in 95% of a group of regressions in a wide range of possible combinations of control... |

289 | Galton’s fallacy and tests of convergence hypothesis
- Quah
- 1993
(Show Context)
Citation Context ...nce statistics are found in all three cases, even though implications on the poor catching up with the rich dier across them. We can make this argument explicit by drawing on reasoning given in Quah =-=[89]-=-. Remove from each observed y its upward-sloping steady state growth path in Fig. 10a{c, so that all y’s have mean zero. Suppose, moreover, that in the long run these transformed y’s satisfy two condi... |

284 | Openness and growth: A time-series, cross-country analysis for developing countries - HARRISON - 1996 |

269 | Good policy or good luck? Country growth performance and temporary shocks - EASTERLY, KREMER, et al. - 1993 |

267 |
Markov chains with stationary transition probabilities
- Chung
- 1960
(Show Context)
Citation Context ...ce nor technology parameters (apart from ) influences the long-run growth rate. Distribution dynamics Rigorous expositions of the mathematics underlying a formulation like (35) can be found in Chung =-=[26]-=-, Doob [34], Futia [49], and Stokey and Lucas (with Prescott) [109].27 Since we are concerned here with real-valued incomes, the underlying state space is the pair (R;R), i.e., the real line R togethe... |

267 | Industrialization and Big Push
- Murphy, Shleifer, et al.
- 1989
(Show Context)
Citation Context ...es with high initial incomes or capital stocks converge to a dierent steady-state level. Examples of such models include those by Durlauf [37], Galor and Zeira [51], and Murphy, Shleifer, and Vishny =-=[80]-=-. The model due to Azariadis and Drazen [4] is particularly convenient for illustrating the empirical dierences between this framework and the neoclassical approach. The Azariadis-Drazen model works ... |

264 |
A theory of persistent income inequality
- Durlauf
- 1996
(Show Context)
Citation Context ...riety of empirical approaches that have been used in growth analysis. 11 Models displaying persistent inequality between families due to neighborhood spillover eects (e.g., Benabou [14] and Durlauf =-=[38]-=-) are also driven by endogenous formation of interaction networks. { 38 { Cross section regression:s-convergence The most common approach to growth and convergence applies cross-section regression ana... |

261 | Divergence, Big Time
- Pritchett
- 1995
(Show Context)
Citation Context ...the current state of macroeconomists’ knowledge on cross-country growth. Since a number of excellent summaries on this subject already exist (e.g., Barro and Sala-i-Martin [11], Jones [58], Pritchett =-=[86]-=-, D. Romer [97]), it is useful to clarify how our presentation diers. First, our emphasis is empirical: We develop dierent growth models focusing on their observable implications for cross-country i... |

233 | The history of statistics – the measurement of uncertainty before - Stigler - 1986 |

227 |
An empirical analysis of cross-national economic growth, 1951-1980
- GRIER, TULLOCK
- 1989
(Show Context)
Citation Context ...des human capital accumulation. That feature is ignored for expositional convenience here as it does not aect our basic point. We return to it below. 4 An earlier literature (e.g., Grier and Tullock =-=[52]-=-) studied similar regression equations with growth on the left-hand side and explanatory variables on the right. We distinguish this from the work described in the text only because that earlier resea... |

227 |
On the Concept of Optimal Economic Growth, in The Econometric Approach to Development Planning
- Koopmans
- 1965
(Show Context)
Citation Context ...dependent of any other economic structure beyond the technology specication. (We will see this below when we study the Solow-Swan model [106, 107, 111], its general equilibrium Cass-Koopmans version =-=[23, 65]-=-, and the modication due to Mankiw, Romer, and Weil [78].) The reasoning just given extends naturally to production technologies beyond Cobb-Douglas when the counterpart to P l l (or, more generally... |

225 | Convergence - Barro, Sala-i-Martin - 1992 |

208 | Idea Gaps and Object Gaps in Economic Development - ROMER - 1993 |

206 |
A Convex Model of Equilibrium Growth: Theory and Policy Implications
- Jones, Manuelli
(Show Context)
Citation Context ...behavior occurs also in endogenous growth models that exploit externalities and increasing returns (Romer [98]) or in models with the production technology \asymptotically linear" (Jones and Manuelli =-=[59]-=-, Rebelo [96]). Our denition of balanced-growth equilibrium compares the growth rates _y=y and _kl=kl. This is not sensible for technology (1b) where we see that _A=A appears with _y=y− _H=H but not ... |

204 |
Regional cohesion: Evidence and theories of regional growth and convergence. European Economic Review 40
- Sala-i-Martin
- 1996
(Show Context)
Citation Context ... Levine and Renelt’s paper, yet other control variables have been introduced. We discuss further below the issues raised by these additional regressors. Barro and Sala-i-Martin [10] and Sala-i-Martin =-=[103]-=- assert that with the right conditioning variables, a rate of convergence of 2% per year is uniformly obtained across a broad range of samples. They draw two implications: First, in a CobbDouglas prod... |

198 |
Capital Accumulation and Economic Growth
- Kaldor
(Show Context)
Citation Context ...n discriminating { 2 { across alternative theories of growth. The questions studied in the new empirical growth literature dier from those in earlier empirical work embodying Kaldor’s stylized facts =-=[60]-=- or those in a production function (Solow [107]-Denison [32]) accounting exercise. The new literature emphasizes understanding cross-country patterns of income, not the stability within a single econo... |

195 | Empirics for economic growth and convergence
- Quah
- 1996
(Show Context)
Citation Context ... constructed around (19), quite elaborately varied behavior for the cross-section distribution is consistent with even well-behaved (unconditional)s- convergence. Fig. 10a{c, similar to those in Quah =-=[93]-=-, show three possibilities. It is easy to generate all three from a singlesxed model satisfying the same transition dynamics as given in (19), varying only y(0) and the variance of the regression resi... |

185 |
Interpreting tests of the convergence hypothesis
- Bernard, Durlauf
- 1996
(Show Context)
Citation Context ...ing the data satisfactorily. However, the results of such a procedure can be dicult to interpret within the Solow-Swan (or Cass-Koopmans) growth model.5 Empirical studies such as Bernard and Durlauf =-=[16, 17]-=-, Durlauf and Johnson [39], and Quah [94] seek to circumvent some of the criticisms we have just described. One strand of this work estimates models that explicitly nest the traditional neoclassical s... |

183 | Political Instability and Economic Growth - ALESINA, ROUBINI, et al. - 1996 |

179 | Growth and Convergence in a Multi-country Empirical Stochastic Solow Model
- Lee, Pesaran, et al.
- 1997
(Show Context)
Citation Context ...unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet [21], Caselli, Esquivel, and Lefort [22], Evans [43], Islam [55], Lee, Pesaran, and Smith =-=[69]-=-, and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e.g., Chamberlain [24]), many such studies seek to elim... |

173 | 2004), Polarization
- Duclos, Esteban, et al.
(Show Context)
Citation Context ...exists across economies|that given in Fig. 7. However, that law of motion displays a polarization eect, namely, economies evolve towards one of two distinct steady states (see, e.g., Esteban and Ray =-=[42]-=-). Regardless of the interpretation, however, the observable implications are the same. Already-rich economies converge to a high steady-state level; already-poor ones, to a low steady-state level. { ... |

164 |
On the Mechanics of Economic Development
- Jr, Robert
- 1988
(Show Context)
Citation Context ... more specically on the topic, economists from at least Kaldor [60] on have marshalled cross-country stylized facts as compelling starting points for discussions about economic growth. Indeed, Lucas =-=[74]-=- and Romer [98, 101] use exactly income comparisons across countries to motivate their endogenous growth analyses. { 3 { Section 5 presents empirical models and critically evaluates the empiricalsndin... |

147 |
Convergence in International Output
- Bernard, Durlauf
- 1995
(Show Context)
Citation Context ...ing the data satisfactorily. However, the results of such a procedure can be dicult to interpret within the Solow-Swan (or Cass-Koopmans) growth model.5 Empirical studies such as Bernard and Durlauf =-=[16, 17]-=-, Durlauf and Johnson [39], and Quah [94] seek to circumvent some of the criticisms we have just described. One strand of this work estimates models that explicitly nest the traditional neoclassical s... |

147 |
Convergence Empirics Across Economies with (Some) Capital Mobility
- Quah
- 1999
(Show Context)
Citation Context ...sndings in greater detail below in Section 5, under Clustering and classication. Dynamics similar to those in the Durlauf-Johnson equation (27) also obtain in the model of Galor and Zeira [51]. Quah =-=[92]-=- applies Galor and Zeira’s ideas to study empirically cross sections of economies (rather than cross sections of families as in the original model). Fig. 7|a two-regime counterpart to equation (27)|is... |

145 |
Convergence? Inferences from theoretical models
- Galor
- 1996
(Show Context)
Citation Context ...ce diers froms-convergence; that they are not the same illustrates some of the conceptual diculties associated with statistical convergence measures in general and cross-country growth 17 See Galor =-=[50]-=- for further discussion. { 42 { regressions in particular. But -convergence too is problematic. To understand those diculties, it is convenient to begin with a further look ats-convergence. For simp... |

143 | Empirics for Growth and Distribution
- Quah
- 1997
(Show Context)
Citation Context ...ults of such a procedure can be dicult to interpret within the Solow-Swan (or Cass-Koopmans) growth model.5 Empirical studies such as Bernard and Durlauf [16, 17], Durlauf and Johnson [39], and Quah =-=[94]-=- seek to circumvent some of the criticisms we have just described. One strand of this work estimates models that explicitly nest the traditional neoclassical setup. Another strand seeks to identify th... |

142 | On the evolution of the world income distribution
- Jones
- 1997
(Show Context)
Citation Context ...one overview of the current state of macroeconomists’ knowledge on cross-country growth. Since a number of excellent summaries on this subject already exist (e.g., Barro and Sala-i-Martin [11], Jones =-=[58]-=-, Pritchett [86], D. Romer [97]), it is useful to clarify how our presentation diers. First, our emphasis is empirical: We develop dierent growth models focusing on their observable implications for... |

139 |
Do old fallacies ever die
- Friedman
- 1992
(Show Context)
Citation Context ... interpretations can be given to the eects we have just described| one early manifestation of these is known in the statistics literature as Galton’s fallacy or Galton’s paradox (see, e.g., Friedman =-=[48]-=-, Maddala [76, 3.12], Stigler [108, Ch. 8], or Quah [89]).18 We prefer to regard the situation constructed above as one where knowledge of what happens to the conditional average (the regression repre... |

129 |
Workings of a City: Location
- Benabou
- 1993
(Show Context)
Citation Context ...on describes a variety of empirical approaches that have been used in growth analysis. 11 Models displaying persistent inequality between families due to neighborhood spillover eects (e.g., Benabou =-=[14]-=- and Durlauf [38]) are also driven by endogenous formation of interaction networks. { 38 { Cross section regression:s-convergence The most common approach to growth and convergence applies cross-secti... |

122 |
Productivity Growth, Convergence and Welfare
- DeLong
- 1986
(Show Context)
Citation Context ...those empirical studies that use the neoclassical growth model for their theoretical underpinnings. Cross-section regression analyses, such { 19 { as Barro and Sala-i-Martin [10], Baumol [12], DeLong =-=[29]-=-, Mankiw, Romer, and Weil [78], and Sachs and Warner [102] estimate variants of (19). Mankiw, Romer, and Weil [78], in particular, consider two versions of (19): First, when the term in et is already... |

119 | Nonergodic economic growth
- Durlauf
- 1993
(Show Context)
Citation Context ...to one steady-state level of per capita output, while economies with high initial incomes or capital stocks converge to a dierent steady-state level. Examples of such models include those by Durlauf =-=[37]-=-, Galor and Zeira [51], and Murphy, Shleifer, and Vishny [80]. The model due to Azariadis and Drazen [4] is particularly convenient for illustrating the empirical dierences between this framework and... |

108 |
Trade and convergence among countries
- BEN-DAVID
- 1996
(Show Context)
Citation Context ...do not suggest here that the evidence is decisive one way or the other, merely that circumspection is called for in these univariate time-series analyses. The second line of research, e.g., Ben-David =-=[13]-=-, documents permanent growth and trend path changes|across time samples comparable to that in Jones’s work|for a wide range of countries other than the US. The subtlety of statistical tests on these g... |

88 | Is fixed investment the key to economic growth?”, Quarterly - BLOMSTRÖM, LIPSEY, et al. - 1996 |

70 |
Testing for Convergence: Evidence from Non-Parametric Multimodality Tests
- Bianchi
- 1997
(Show Context)
Citation Context ...k suggests that a richer notion of convergence, one accounting explicitly for the multivariate nature of aggregate socioeconomic characteristics, warrants further study. Distribution dynamics Bianchi =-=[18]-=-, Desdoigts [33], Jones [58], Lamo [67], Quah [88, 89, 92, 94] have studied the predictions of the theoretical growth models in terms of the behavior of the entire cross-section distribution. While th... |

66 |
Classification and Regression Trees, Chapman and
- Breiman, Friedman, et al.
- 1984
(Show Context)
Citation Context ..., while the high-literacy grouping has p = 0:79 and h = −0:07 (not statistically signicant). Finally, high-income economies display p = 0:31 and h = 0:46. 25 Breiman, Friedman, Olshen, and Stone =-=[20]-=- describe the regression-tree procedure and its properties. { 58 { Table 1 gives these regression results for two of the regimes, and compares them with the original MRW regression. We see that, depen... |

62 |
The poor stay poor: non-convergence across countries and regions’, Paper presented at
- Canova, Marcet
- 1997
(Show Context)
Citation Context ...d to { 49 { algorithms of the kind that have been employed. Panel data analysis To permit unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet =-=[21]-=-, Caselli, Esquivel, and Lefort [22], Evans [43], Islam [55], Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional moti... |

59 | Invariant Distributions and the Limiting Behavior of Markovian Economic Models,” Econometrica 50:2
- Futia
- 1982
(Show Context)
Citation Context ...eters (apart from ) influences the long-run growth rate. Distribution dynamics Rigorous expositions of the mathematics underlying a formulation like (35) can be found in Chung [26], Doob [34], Futia =-=[49]-=-, and Stokey and Lucas (with Prescott) [109].27 Since we are concerned here with real-valued incomes, the underlying state space is the pair (R;R), i.e., the real line R together with the collection R... |

54 |
Using Panel Data to Evaluate Growth Theories
- Evans
- 1998
(Show Context)
Citation Context ...n employed. Panel data analysis To permit unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet [21], Caselli, Esquivel, and Lefort [22], Evans =-=[43]-=-, Islam [55], Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e.g., Chamber... |

49 |
Productivity Growth
- Baumol, Wolff
- 1988
(Show Context)
Citation Context ...now organize those empirical studies that use the neoclassical growth model for their theoretical underpinnings. Cross-section regression analyses, such { 19 { as Barro and Sala-i-Martin [10], Baumol =-=[12]-=-, DeLong [29], Mankiw, Romer, and Weil [78], and Sachs and Warner [102] estimate variants of (19). Mankiw, Romer, and Weil [78], in particular, consider two versions of (19): First, when the term in e... |

47 | Trade and growth in East Asian countries: cause and effect?, NBER, Working Paper 5732 - Frankel, Romer, et al. - 1996 |

36 | Health capital and cross-country variation in income per capita - KNOWLES, OWEN - 1995 |

31 |
International Patterns of Growth: Persistence in Cross-Country Disparities." Working Paper, MIT
- Quah
- 1990
(Show Context)
Citation Context ... Unit roots and cointegration An alternative approach to long-run output dynamics and convergence based on time-series ideas has been developed in Bernard and Durlauf [16, 17], Durlauf [36], and Quah =-=[87]-=-. Convergence here is identied not as a property of the relation between initial income and growth over asxed sample period, but instead of the relationship between long-run forecasts of per capita o... |

25 | Aggregate and Regional Disaggregate Fluctuations
- Quah
- 1996
(Show Context)
Citation Context ..., and elsewhere. Using these ideas for studying distribution dynamics rather than analyzing a time-series stochastic process, say, exploits a duality in the mathematics. This is made explicit in Quah =-=[91]-=-, a study dealing not with cross-country growth but business cycles instead. { 69 { where the supremum in this denition is taken over all fAj : j = 1; 2; : : : ; ngsnite measurable partitions of R. E... |

20 |
Output Persistence, Economic Structure, and the Choice of Stabilization Policy
- Durlauf
(Show Context)
Citation Context ... { Time series: Unit roots and cointegration An alternative approach to long-run output dynamics and convergence based on time-series ideas has been developed in Bernard and Durlauf [16, 17], Durlauf =-=[36]-=-, and Quah [87]. Convergence here is identied not as a property of the relation between initial income and growth over asxed sample period, but instead of the relationship between long-run forecasts ... |

20 |
Exploiting Cross Section variations for unit root inference in dynamic data
- Quah
- 1994
(Show Context)
Citation Context ... the individual eects, and conclude that those eects do, indeed, dier across economies. Evans, using a dierent statistical technique, concludes the same. Evans follows Levin and Lin [71] and Quah =-=[90]-=- in taking an underlying probability model where both time and cross-section dimensions in the panel dataset are large. This contrasts with standard panel-data studies where the time dimension is take... |

18 | An alternative interpretation of conditional convergence results - Cho - 1996 |

18 |
Tests of the Convergence Hypothesis: Some Further Results
- Cohen
- 1996
(Show Context)
Citation Context ...erent policy implications, standard cross-country tests of convergence need not provide great insight. While, under the neoclassical model, the conventional cross-country growth equation is 15 Cohen =-=[28]-=- takes this \deconstruction" exercise a step further, and in a dierent direction. He argues that, typically-constructed stocks of human and physical capital show unconditionalsconvergence, even if pe... |

13 |
Changes in the World Income Distribution: A non Parametric Approach to Challenge the Neoclassical Convergence Argument
- Desdoigts
- 1994
(Show Context)
Citation Context ...a richer notion of convergence, one accounting explicitly for the multivariate nature of aggregate socioeconomic characteristics, warrants further study. Distribution dynamics Bianchi [18], Desdoigts =-=[33]-=-, Jones [58], Lamo [67], Quah [88, 89, 92, 94] have studied the predictions of the theoretical growth models in terms of the behavior of the entire cross-section distribution. While this work is often... |

13 |
Specification Searches: Ad Hoc Inference from NonExperimental Data
- Leamer
- 1973
(Show Context)
Citation Context ...mphasized thatsndings of statistical signicance may be fragile due to dependence on additional controls whose presence or absence is not strongly { 47 { motivated by any theory. By applying Leamer’s =-=[68]-=- extreme bounds analysis (thereby identifying the range of coecient estimates for a given regressor generated by alternative choices of additional regressors) they found that only the physical capita... |

12 |
Convergence in Stochastic Growth Models: the Importance of Understanding Why Income Levels Di¤er
- Haan
- 1995
(Show Context)
Citation Context ...cal or endogenous growth analyses. { 40 { interpretation of that 2% rate of convergence.15 Researchers have identied a number of econometric problems with conditionals-convergence analysis. Den Haan =-=[31]-=- and Kocherlakota and Yi [64] argue that how one augments the growth model with stochastic disturbances profoundly aects the inference to be drawn from the data.16 Their point resembles the classical... |

12 | Convergence, endogenous growth and productivity disturbances
- Leung, Quah
- 1996
(Show Context)
Citation Context ...r this, and suggests a vintage human-capital model to explain it. 16 This result on the importance of the stochastic specication is related to but dierent from that in Kelly [61] and Leung and Quah =-=[70]-=-. These authors show that an appropriate stochastic specication can distort, not just statistical inference, but the underlying relation between physical capital’s coecient in the production functio... |

11 |
Making a miracle
- Jr, E
- 1993
(Show Context)
Citation Context ...9] all associate the evolution of A with a measurable input such as research and development expenditure, the number of scientists and engineers, and so on. By contrast, models such as those in Lucas =-=[74, 75]-=- focus on improvement in H|human capital embodied in the labor force|as the source for endogenous growth. When the production technology is (1a) the resulting dynamics in measured per capita income wi... |

10 |
On Endogenous Growth with Productivity Shocks
- Kelly
- 1992
(Show Context)
Citation Context ...ountries that account for this, and suggests a vintage human-capital model to explain it. 16 This result on the importance of the stochastic specication is related to but dierent from that in Kelly =-=[61]-=- and Leung and Quah [70]. These authors show that an appropriate stochastic specication can distort, not just statistical inference, but the underlying relation between physical capital’s coecient i... |

5 |
Accounting for United States Growth
- DENISON
- 1974
(Show Context)
Citation Context .... The questions studied in the new empirical growth literature dier from those in earlier empirical work embodying Kaldor’s stylized facts [60] or those in a production function (Solow [107]-Denison =-=[32]-=-) accounting exercise. The new literature emphasizes understanding cross-country patterns of income, not the stability within a single economy of factor shares or \great ratios" (the ratio of output t... |

4 | How Strongly Do Developing Economies Bene From Equipment Investment - Long, Summers - 1993 |

4 |
Trade and growth: An empirical investigation (Working Paper No. 5476
- Frankel, Romer
- 1996
(Show Context)
Citation Context ...so extends to various attempts in the literature tosnd instruments for the various baseline and augmented Solow-Swan regressors, which are of course typically endogenous themselves. Frankel and Romer =-=[45]-=- use geographic variables to instrument their measure of trade openness. However, that these variables are exogenous with respect to trade openness does not make them legitimate instruments. For examp... |

3 |
Cross-Country Growth Regressions”, Working Paper 97-20
- Benhabib, Spiegel
- 1997
(Show Context)
Citation Context ..., which cannot be reduced to { 49 { algorithms of the kind that have been employed. Panel data analysis To permit unobservable country-specic heterogeneity in growth regressions, Benhabib and Siegel =-=[15]-=-, Canova and Marcet [21], Caselli, Esquivel, and Lefort [22], Evans [43], Islam [55], Lee, Pesaran, and Smith [69], and Nerlove [83] have used panel data methods to study cross-country income data. Fo... |

3 |
The Speci of the Aggregate Production Function: A Cross-Country Empirical Investigation", Working paper
- Duy, Papageorgiou
- 1997
(Show Context)
Citation Context ...is possibility when using the phrase. A useful alternative is to say that (24) implies f(~k) is O(~k) (or big-oh ~k), following standard terminology in statistics and elsewhere. Duy and Papageorgiou =-=[35]-=-snd that a CES specication for the aggregate production functionsts cross-country data better than a CobbDouglas, and moreover that the elasticity of substitution between capital and labor exceeds on... |

3 | How Much Do Distortions Aect Growth - Easterly - 1993 |

3 |
Cross-section Distribution Dynamics
- Lamo
- 1996
(Show Context)
Citation Context ...ergence, one accounting explicitly for the multivariate nature of aggregate socioeconomic characteristics, warrants further study. Distribution dynamics Bianchi [18], Desdoigts [33], Jones [58], Lamo =-=[67]-=-, Quah [88, 89, 92, 94] have studied the predictions of the theoretical growth models in terms of the behavior of the entire cross-section distribution. While this work is often quite technical, it ca... |

2 |
Growth Rate Convergence, Fact or Artifact?", Working paper
- Nerlove
- 1996
(Show Context)
Citation Context ...ry-specic heterogeneity in growth regressions, Benhabib and Siegel [15], Canova and Marcet [21], Caselli, Esquivel, and Lefort [22], Evans [43], Islam [55], Lee, Pesaran, and Smith [69], and Nerlove =-=[83]-=- have used panel data methods to study cross-country income data. Following traditional motivation in panel-data econometrics (e.g., Chamberlain [24]), many such studies seek to eliminate, in the nota... |

2 |
Some Methodological Issues in the Analysis of Longtitudinal Surveys
- Singer, Spilerman
- 1976
(Show Context)
Citation Context ...kedness, 26 Potential inconsistency across M matrices estimated over single- and multipleperiod transitions is a well-known problem from the labor and sociology literature (e.g., Singer and Spilerman =-=[105]-=-). Quah [88] shows that, in the Heston-Summers cross-country application, the long-run properties of interest are, approximately, invariant to the transition period used in estimation. { 62 { after al... |

1 | Back to the Basics: The Positive Eect of Inequality on Growth", Working paper - Forbes - 1997 |

1 |
Convergence of Living Standards: An International Analysis
- Franses, Hobijn
- 1995
(Show Context)
Citation Context ...rlauf and Johnson’s analysis [39] has a classication interpretation. Interest lies in which economies belong to which subgroups. This line of reasoning has been usefully extended. Franses and Hobijn =-=[47]-=- attempt to identify groups of similarly-behaving economies using measures of social indicators in addition to per capita GNP. Unlike Durlauf and Johnson [39] they use clustering algorithms to partiti... |