Journal Of Econometrics Research Articles

Level-Based Estimation of Dynamic Panel Models

Abstract This paper develops an alternative estimator for linear dynamic panel data models based on parameterizing the covariances between covariates and unobserved time-invariant effects. A GMM framework is used to derive an optimal estimator based on moment conditions in levels, with no efficiency loss compared to the classic alternatives like (Arellano, M., and S. Bond. 1991. “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies 58 (2): 277–297), (Ahn, S. C., and P. Schmidt. 1995. “Efficient Estimation of Models for Dynamic Panel Data.” Journal of Econometrics 68 (1): 5–27) and (Ahn, S. C., and P. Schmidt. 1997. “Efficient Estimation of Dynamic Panel Data Models: Alternative Assumptions and Simplified Estimation.” Journal of Econometrics 76: 309–321). Still, we show analytically and by Monte Carlo simulations that the new procedure leads to efficiency improvements for certain data generating processes. The framework also leads to a very simple test for unobserved effects.

Nexus of FDI, population, energy production, and water resources in South Asia: a fresh insight from dynamic common correlated effects (DCCE).

The purpose of this study is to explore the empirical relationship between foreign direct investment (FDI), population, energy production, and water resources in South Asia. The newly developed approach dynamic common correlated effects (DCCE) by Chudik and Pesaran (Journal of Econometrics 188:393-420, 2015a) for measuring co-integration has been applied in the present study. This procedure provides significant robust outcomes in the presence of cross-sectional dependence among the cross-sectional units. The findings confirmed that earlier models, such as mean group (MG), pooled mean group (PMG), and augmented mean group (AMG), which have been used in the literature for long data, provide misleading results in the presence of cross-sectional dependence among the cross-sectional units. A statistically significant and negative result has been observed between FDI, population, energy production, and water resources in South Asia. The governments of South Asian economies must encourage green FDI initiatives for water management, ensuring water security, securing natural resources for enhancing the sustainable development of regional economies.

Impact of Exchange Rate Management on the Nigerian Economic Growth: Empirical Validation

Adewuyi, (2002). Balance of Payments Constraints and Growth Rate Differences under Alternative Police Regimes. Nigerian Institute of Social and Economic Research (NISER) Monograph Series No. 10, Ibadan, Nigeria. Adebiyi, M. A. (2007). An Evaluation of Foreign Exchange Intervention and Monetary Aggregates in Nigeria (1986-2003). The University of Munich Finance Journal, 4 (2),1-19. Aghin, P.; Bacchetta, P.; Ranciere, R. & Rogoff, K. (2006). Exchange rate volatility and productivity growth: The role of financial development. Journal of Monetary Economics, 56 (4), 494–51. Bonser-Neal, C. & Tanner, G. (1996). Central Bank Intervention and Volatility of Exchange Rates: Evidence from the Options Market. Journal of International Money and Finance, 18 (2), 23-45. Dominguez, K. (1990). Market Responses to Coordinated Central Bank Intervention. Carnegie Rochester Conference Series on Public Policy, 32 Dominguez, K. (1998). Central Bank Intervention and Exchange Rate Volatility. Journal of International and Finance, 15 (4). Dominguez, K & Frankel, J. A. (1993). Does Foreign Exchange Intervention matter? The Portfolio effect. American Economic Review, 83 (5), 231-259 Dubas, J.M., Lee, B.J., & Mark, N.C. (2005). Effective Exchange Rate Classifications and Growth. NBER Working Paper No. 11272 Frankel, J.A. ((1992). In search of the Exchange rate Premium: A Six-Currency Test assuming mean-variance optimization. Journal of International Money and Finance,1(2),19-32 Gosh, A. (1992). Is it Signinaling? Exchange Rate intervention and the Dollar Deutssche-Mark Rate. Journal of International Economics, 32(4), 45-67 Granger, C.W.J. & Newbold, P. 1974). Spurious regression in Econometrics, Journal of Econometrics 2 (4) 111-120. Harris R.G. (2002). New Economy and the Exchange Rate Regime. Centre for International Economics Studies, Discussion paper, No 111. Humpage, O. (1989). On the Effectiveness of Foreign Exchange Market Intervention. Federal Reserve Bank of Cleveland. Kaminsky, G & Lewis, K (1996). Does Foreign Exchange Intervention signal future monetary policy? Journal of Monetary Economics, 37(2), 66-89 Nwankwo (G.O) (1980). Money and capital markets in Nigeria Today. University of Lagos Press, Nigeria. Odusola A.F. and Akinlo, A.E. (2001). Output, Inflation, and Exchange Rate in Developing Countries: An Application to Nigeria. Developing Economies, 39(2). Oloyede, J. A. (2002). Principles of International Finance. Forthright Educational Publishers, Lagos. Rano-Aliyu S.U. (2009). Impact of Oil Price Shock and Exchange Rate Volatility on Economic Growth in Nigeria: An Empirical Investigation. Research Journal of International Studies 10(4) 23-45. Rognoff, K. (1984). The effects of sterilized intervention: An analysis of weekly data. Journal of Monetary Economics, 2 (2), 14-34. Sarno, L. & Taylor, M. P. (2001). Official intervention in the Foreign Exchange Market: Is it effective, and if so, how does it work? Journal of Economic Literature, 3(1), 39-56. Simatele, M.C.H.(2003). Financial Sector Reforms and Monetary Policy in Zambia. Ph.D Dissertation, Economics Studies, Department of Economics, School of Economics and Commercial Law, Goteborg University. Unugbro, A.O (2007). The Impact of Exchange Rate Fluctuation on Capital Inflow: The Nigerian Experience. The Nigeria Academic Journal of Social Sciences, 6(4),1-21

The Nexus between Government Expenditure and Economic Growth: Evidence of the Wagner’s Law in Kuwait

Abstract This study is the first attempt to examine the validity of the Wagner’s law hypothesis by employing time-series data over the period from 1970 to 2015 in Kuwait. In this paper, the causal relationship between government expenditure and economic growth is tested by conducting the Granger non-causality test developed by (Toda, H. Y., and T. Yamamoto. 1995. “Statistical Inference in Vector Autoregressions with Possibly Integrated Processes.” Journal of Econometrics 66 (1): 225–250.) and (Dolado, J. J., and H. Lütkepohl. 1996. “Making Wald Tests Work for Cointegrated VAR Systems.” Econometric Reviews 15 (4): 369–386.). The empirical results support the unidirectional causality running from government spending to economic growth. This occurs only when real government expenditure per capita is a proxy for state activity and real gross domestic product (GDP) per capita is a measure of economic growth. This implies that Wagner’s law does not apply for Kuwait’s economy, and the Keynesian proposition of government spending as a policy instrument that encourages and leads economic growth is supported by the data used.

Learning environment and primary school efficiency

Purpose Existing studies of school efficiency primarily specify teacher inputs as the number of teachers and perhaps the student-teacher ratio. As a result, there is no direct qualitative recognition of the learning environment. The purpose of this paper is to incorporate the learning environment directly into the assessment of school efficiency. Design/methodology/approach The authors employ data envelopment analysis to derive efficiency scores and the double-bootstrap truncated regression approach in Simar and Wilson’s (2007) Journal of Econometrics to quantify the sources of efficiency in 430 Queensland state primary schools. In the first stage, the outputs of student National Assessment Program-Literacy and Numeracy scores and the inputs of full-time equivalent teaching staff and cumulative capital expenditure per student are used to measure efficiency. In the second stage, the authors specify an index of community socio-educational advantage, class size, the share of teachers with postgraduate qualifications, funds spent on professional development, and surveyed opinions from parents/caregivers, students, staff and principals on the learning environment to explain these measures of efficiency. Findings Socio-economic background and the teaching environment affect school efficiency. Although not all variables related to teacher contribution are significant, there is evidence to suggest that teachers have a positive influence on student performance hence school efficiency. Teachers ability to clearly explain the requirements of schoolwork tasks and listening to student opinions sets an ideal student engagement environment which can have a profound impact on student learning. Practical implications From a policy perspective, policy makers should target resources at inefficient schools aimed at enhancing student learning through teacher development and, at the same time, providing financial and non-financial educational assistance to students and their families from a low socio-educational background. Originality/value This is the first large-scale primary school efficiency analysis to incorporate the Simar and Wilson (2007) approach to explaining the determinants of efficiency, including teaching environment from the perspective of students, teachers and other stakeholders.

Testing for cointegration with threshold adjustment in the presence of structural breaks

Abstract In this paper, we develop new threshold cointegration tests with SETAR and MTAR adjustment allowing for the presence of structural breaks in the equilibrium equation. We propose a simple procedure to simultaneously estimate the previously unknown breakpoint and test the null hypothesis of no cointegration. Thereby, we extend the well-known residual-based cointegration test with regime shift introduced by (Gregory, A. W., and B. E. Hansen. 1996a. “Residual-based Tests for Cointegration in Models with Regime Shifts.” Journal of Econometrics 70: 99–126) to include forms of nonlinear adjustment. We derive the asymptotic distribution of the test statistics and demonstrate the finite-sample performance of the tests in a series of Monte Carlo experiments. We find a substantial decrease of power of the conventional threshold cointegration tests caused by a shift in the slope coefficient of the equilibrium equation. The proposed tests perform superior in these situations. An application to the “rockets and feathers” hypothesis of price adjustment in the US gasoline market provides empirical support for this methodology.

Threshold Effect of Bank-specific Determinants of Non-performing Assets: An Application in Indian Banking

The article investigates role of bank-specific factors on non-performing assets (NPAs) in Indian banking system in a panel threshold framework (Hansen, 1999, Journal of Econometrics, 93(2), 345–368), using an unbalanced panel of 82 scheduled commercial banks over the period of 1995–1996 to 2010–2011. We consider capital to risk-weighted assets ratio (CRAR) and credit growth as alternative threshold variables (and regime dependent) along with relevant bank-specific variables treated as regime independent. Findings reveal that CRAR exerts negative and significant impact on NPAs once it reaches a critical threshold. Possible implication is that banks extend less risky loans in a high CRAR regime than in low CRAR regime that helps reduce NPAs. With credit growth as threshold as well as regime dependent, we observe statistically significant non-linear effect of credit growth on NPAs. Beyond threshold, credit growth exerts significant negative effect on NPAs that may imply that banks extend good quality loans. However, we cannot rule out the possibility of evidence of ‘ever-greening hypothesis’ of bad debts in Indian banking, that is, banks just roll over previous bad debts into fresh performing loans. JEL Classification: G21, G28, C13, C33

Asymmetries in Volatility: An Empirical Study for the Peruvian Stock and Forex Markets

Asymmetric autoregressive conditional heteroskedasticity (EGARCH) models and asymmetric stochastic volatility (ASV) models are applied to daily data of Peruvian stock and Forex markets for the period of 5 January 1998–30 December 2011. Following the approach developed in [Omori, Y, S Chib, N Shephard and J Nakajima (2007). Stochastic volatility with leverage: Fast likelihood inference. Journal of Econometrics, 140, 425–449], Bayesian estimation tools are used with Normal and [Formula: see text]-Student errors in both models. The results suggest the significant presence of asymmetric effects in both markets. In the stock market, negative shocks generate higher volatility than positive shocks. In the Forex market, shocks related to episodes of depreciation create higher uncertainty in comparison with episodes of appreciation. Thus, the Central Reserve Bank faces relatively major difficulties in its intention of smoothing Forex volatility in times of depreciation. The model with the best fit in both markets is the ASV model with Normal errors. The stock market returns have greater periods of volatility; however, both markets react to shocks in the economy, as they display similar patterns and have a significant correlation for the sample period studied.

Time and frequency domain causality Testing: The causal linkage between FDI and economic risk for the case of Turkey

ABSTRACTThis study aims to explore the causal relationship between economic risk and foreign direct investment (FDI) inflows for the case of Turkey. With the aim of establishing robust findings for the research in mind, both traditional and modern causality techniques are utilized; time domain Granger (1969, “Investigating Causal Relations by Econometric Models and Cross-Spectral Methods.” Econometrica 37: 424–438.), Toda and Yamamoto (1995, “Statistical Inference in Vector Autoregressions with Possibly Integrated Processes.” Journal of Econometrics 66 (1–2): 225–250.), Fourier Toda-Yamamoto and frequency domain Breitung and Candelon (2006, “Testing for short- and long-run causality: A frequency-domain approach.” Journal of Econometrics 132 (2): 363–378.) spectral causality test. Our empirical findings reveal that; economic risk changes in Turkey significantly lead to changes in FDI inflows. However, there is no evidence of causality running from FDI to economic risk. The findings imply that economic risk is an essential determinant of FDI inflows in Turkey. Our findings are compatible with historical macroeconomic developments in Turkey and imply important policy implications. The results of this study can be generalized for other emerging economies that have similar macroeconomic environments, in order to create useful policy implications regarding FDI inflow.

РЕГИОНАЛЬНАЯ ДИНАМИЧЕСКАЯ СТОХАСТИЧЕСКАЯ МОДЕЛЬ ОБЩЕГО РАВНОВЕСИЯ КАК ИНСТРУМЕНТ АНАЛИЗА ФИСКАЛЬНОЙ ПОЛИТИКИ

Using the tools of regional dynamic models for analyzing the economy of the constituent entities of the Russian Federation, in particular, for studying regional business cycles is currently an urgent task. It is determined by the need to develop system conceptions about the factors, conditions and prerequisites for the development of regions, about the features and trends of the dynamics of their sector and territorial structure. The purpose of the article is to develop a dynamic stochastic multi-sector model for analyzing the effects of regional economic policy. The scientific novelty of the research concerns the development and implementation of dynamic models with microeconomic justification to formalize the processes of regional development, the sustainability of regional policy and spatial development. Similar class of models, that forms the theoretical foundation of contemporary macro-economics, is currently used for the analysis of national economy mostly. Models of such class that describe the processes in the regional economy are practically absent. The original tools for the construction of a regional dynamic stochastic general equilibrium model, suggested by the authors, describe the structure of a real sector of the economy of Sverdlovsk region. Parameterization of the model was made on the empirical data basis about the economy of Sverdlovsk region for 2003–2016. The behaviour of the following economic operators has been considered in the model: households; firms operating in the real sector of economy, the regional and federal government, and the Central Bank. Fiscal multipliers for three sectors of the economy – tradable goods sector, non-tradable goods sector and resource sector have been calculated with impulse response functions. The analysis of fiscal multipliers has revealed that the shock of the effective tax rate on individual income and the sock of regional costs have the most significant effect on the output in the above considered sectors of economy among all the rest fiscal shocks. The use of the tools in the form of a historical decomposition of regional variables demonstrates the results of the impact of supply and demand shocks in a time perspective on the output in the three sectors of the regional economy. The results of temporal decomposition of the variations of the endogenous variables mentioned above suggest that the cyclic processes in the regional economy of Sverdlovsk region during the study period are largely due to factors of supply rather than demand. The research results may be used both for the analysis of the regional economic policy priorities and for the development of measures aimed at the decrease of possible crisis phenomena in the regional economy. The trend to the construction of multi-sector models of regional economy in the framework of general equilibrium approach with the macroeconomics justification and rational expectations of economic operators described in the article stresses the importance and prospects of further studies. In particular, to reflect the specifics of the regions, it is necessary to take into account the institutional factors of each region in the model. This issue is an interesting theme for further research in the field of modeling of regional social and economic systems. Keywords region, regional economic policy, dynamic stochastic model, tradable and non-tradable goods sector, resource sector, fiscal multipliers, demand shocks, supply shocks, impulse response functions, historical decomposition of variations of endogenous variables. Acknowledgements The article has been written according to the Plan of Research and Development of the Institute of Economics, the Ural Branch of the Russian Academy of Sciences for 2019–2021.

How Does Governance Modify the Relationship between Public Finance and Economic Growth: A Global Analysis

How Does Governance Modify the Relationship between Public Finance and Economic Growth: A Global Analysis

The Faster Growth of Larger, Less Crowded Locations

Recent population growth across U.S. locations (metros, micros, and remaining counties) has been strongly correlated with initial size and density. From 2000 to 2017, growth was positively correlated with initial population except among medium and large metros (those with population of at least 500,000 in 2000). Among medium and large metros, the relationship was essentially flat, except for a downturn across the very largest. Growth across the medium and large metros also turns out to be strongly correlated with measures of population density: negative with summary measures of pop-weighted density and strongly positive with spikes in density from the 95th-percentile tract to the 99th-percentile tract. I interpret the latter as capturing nearby urban amenities. Size and density also account for a surprisingly high share of the variation in growth from 2000 to 2017. A kernel regression of growth on population has R-squared of 0.24, up from 0.05 over the previous 40 years. For the medium and large metros, regressing growth on the internal distribution of density (25th percentile, 75th, 95th, and 99th) gives an R-squared of 0.37. Overall, I interpret the results as suggesting that both agglomerative benefits and agglomerative costs have increased during recent decades. An online supplement has all variables (including weather, coastal proximity, and the full distribution of pop-weighted density within metros from the 1st to 99th percentiles), replication code, and a stata program to estimate standard errors robust to spatially-correlated disturbances (Conley, Journal of Econometrics, 1999).

Implementing Valid Two-Step Identification-Robust Confidence Sets for Linear Instrumental-Variables Models

In this article, we consider inference in the linear instrumental-variables models with one or more endogenous variables and potentially weak instruments. I developed a command, twostepweakiv, to implement the two-step identification-robust confidence sets proposed by Andrews (2018, Review of Economics and Statistics 100: 337–348) based on Wald tests and linear combination tests (Andrews, 2016, Econometrica 84: 2155–2182). Unlike popular procedures based on first-stage F statistics (Stock and Yogo, 2005, Testing for weak instruments in linear IV regression, in Identification and Inference for Econometric Models: Essays in Honor of Thomas Rothenberg), the two-step identification-robust confidence sets control coverage distortion without assuming the data are homoskedastic. I demonstrate the use of twostepweakiv with an example of analyzing the effect of wages on married female labor supply. For inference on subsets of parameters, twostepweakiv also implements the refined projection method (Chaudhuri and Zivot, 2011, Journal of Econometrics 164: 239–251). I illustrate that this method is more powerful than the conventional projection method using Monte Carlo simulations.

An empirical investigation of direct and iterated multistep conditional forecasts

SummaryWhen constructing unconditional point forecasts, both direct and iterated multistep (DMS and IMS) approaches are common. However, in the context of producing conditional forecasts, IMS approaches based on vector autoregressions are far more common than simpler DMS models. This is despite the fact that there are theoretical reasons to believe that DMS models are more robust to misspecification than are IMS models. In the context of unconditional forecasts, Marcellino et al. (Journal of Econometrics, 2006, 135, 499–526) investigate the empirical relevance of these theories. In this paper, we extend that work to conditional forecasts. We do so based on linear bivariate and trivariate models estimated using a large dataset of macroeconomic time series. Over comparable samples, our results reinforce those in Marcellino et al.: the IMS approach is typically a bit better than DMS with significant improvements only at longer horizons. In contrast, when we focus on the Great Moderation sample we find a marked improvement in the DMS approach relative to IMS. The distinction is particularly clear when we forecast nominal rather than real variables where the relative gains can be substantial.

Regression discontinuity designs with unknown state-dependent discontinuity points: estimation and testing

Abstract This paper extends Regression discontinuity designs with unknown discontinuity points developed by (Porter, J., and P. Yu. 2015. “Regression Discontinuity Designs with Unknown Discontinuity Points: Testing and Estimation.” Journal of Econometrics 189: 132–147.) to allow for state-dependent discontinuity points. We discuss the estimation of the model, and propose test statistics for treatment effect and state dependency in the discontinuity points. We conduct Monte Carlo simulations to compare the proposed estimator with these based on the constant discontinuity RDD and the classic fuzzy RDD, and find that overlooking the state dependency can lead to biased estimates of treatment effects, while the proposed estimator works well and is robust when applied to constant discontinuity RDDs. Monte Carlo experiments also point out that the sizes and powers of the proposed test statistics are generally satisfactory. The model is illustrated with an empirical application.

How important are fixed effects and time trends in estimating returns to schooling? Evidence from a replication of Jacobson, Lalonde, and Sullivan, 2005

SummaryA substantial and rapidly growing literature has developed around estimating earnings gains from 2‐year college degrees using administrative data. These papers almost universally employ a person‐level fixed‐effects strategy to estimate earnings premia net of fixed attributes. We note that the seminal piece on which these papers build—Jacobson, Lalonde, & Sullivan, Journal of Econometrics, 2005, 125(1–2), 271–304—provides theoretical and empirical evidence for the importance of additionally differencing out individual time trends. The subsequent literature has not followed suit. Through replication we ask whether this matters. We show that it does, and further that these person‐level time trends need not be computationally burdensome in large administrative data. We recommend them as a unifying econometric standard for future work.

The ET Interview: Professor Hashem Pesaran

Since graduating with a PhD in Economics from Cambridge, Hashem Pesaran has been a prominent contributor to the theoretical econometrics literature, undertaking work on testing of non-nested models, cointegration, forecasting in the presence of model instability, forecasting of multinomial data, and estimation and testing of time-series and panel data models. His work also covers several areas of applied econometrics, including studies of oil prices, prices in financial markets, rational expectations and economic surveys, analysis of panel data models with cross sectional dependence, and spillover effects and interdependencies across international economies. Hashem is the founding editor of the Journal of Applied Econometrics and has served on the editorial boards of numerous journals in economics and econometrics. He is a Fellow of the British Academy, the Econometric Society, and the Journal of Econometrics. Previously, he was head of the Economic Research Department of the Central Bank of Iran and Under-Secretary of the Ministry of Education, Iran. He has held positions as a Professor of Economics at Cambridge University and at UCLA. He is currently an emeritus Professor of Economics at Cambridge University, a life time Fellow of Trinity College, Cambridge, and holds the John Elliott Distinguished Chair in Economics at the University of Southern California.

Estimating Dynamic Common-Correlated Effects in Stata

In this article, I introduce a new command, xtdcce2, that fits a dynamic common-correlated effects model with heterogeneous coefficients in a panel with a large number of observations over cross-sectional units and time periods. The estimation procedure mainly follows Chudik and Pesaran (2015b, Journal of Econometrics 188: 393–420) but additionally supports the common correlated effects estimator (Pesaran, 2006, Econometrica 74: 967–1012), the mean group estimator (Pesaran and Smith, 1995, Journal of Econometrics 68: 79–113), and the pooled mean group estimator (Pesaran, Shin, and Smith, 1999, Journal of the American Statistical Association, 94: 621–634). xtdcce2 allows heterogeneous or homogeneous coefficients and supports instrumental-variable regressions and unbalanced panels. The cross-sectional dependence test is automatically calculated and presented in the estimation output. Small-sample time-series bias can be corrected by “half-panel” jackknife correction or recursive mean adjustment. I carry out a simulation to prove the estimator's consistency.

Theory and Practice of Total-Factor Productivity Estimation: The Control Function Approach using Stata

Alongside instrumental-variables and fixed-effects approaches, the control function approach is the most widely used in production function estimation. Olley and Pakes (1996, Econometrica 64: 1263–1297), Levinsohn and Petrin (2003, Review of Economic Studies 70: 317–341), and Ackerberg, Caves, and Frazer (2015, Econometrica 83: 2411–2451) have all contributed to the field by proposing two-step estimation procedures, whereas Wooldridge (2009, Economics Letters 104: 112–114) showed how to perform a consistent estimation within a single-step generalized method of moments framework. In this article, we propose a new estimator based on Wooldridge's estimation procedure, using dynamic panel instruments à la Blundell and Bond (1998, Journal of Econometrics 87: 115–143), and we evaluate its performance by using Monte Carlo simulations. We also present the new command prodest for production function estimation, and we show its main features and strengths in a comparative analysis with other community-contributed commands. Finally, we provide evidence of the numerical challenges faced when using the Olley–Pakes and Levinsohn–Petrin estimators with the Ackerberg–Caves–Frazer correction in empirical applications, and we document how the generalized method of moments estimates vary depending on the optimizer or starting points used.

Does More Consumption Promote Real GDP Growth?

AbstractThis paper employs Hansen's (1999) panel threshold regression model [Journal of Econometrics 39 (1999) 345–68] based on a time series dataset of 109 countries from 1960 to 2007 to investigate the threshold relationship between the change in real GDP per capita and the consumption size (consumption‐income ratio, APC). The results show that the consumption level should not exceed the 49.68% threshold of real GDP per capita for each country regardless of the income level. Also, the relationship between the change in real GDP per capita and the consumption size seems to have ‘Armey curve’ or ‘inverted‐U shape’ characteristic. In order to promote real GDP growth, our results suggest that the high‐income, low‐APC countries should encourage more consumption while the low‐income, high‐APC countries should encourage more saving.