Non-stationary Panel Data Research Articles

Exploring the finance-real economy link in U.S.: empirical evidence from panel unit root and cointegration analysis

The aim of this article is to analyze the relationships between common shocks affecting the real economy and those underlying co-fluctuations in U.S. financial markets. In order to do this, we test for links between these common factors and also use the econometric theory of non-stationary panel data to estimate the relationships. The estimates prove the existence of significant relationships between financial and macroeconomic factors. It is also shown that there are forces pulling U.S. financial markets to move with the real economy, as seen through nearly instantaneous adjustment to a new equilibrium.

Spatial error correction and cointegration in nonstationary panel data: regional house prices in Israel

We “spatialize” residual-based panel cointegration tests for nonstationary spatial panel data in terms of a spatial error correction model (SpECM). Local panel cointegration arises when the data are cointegrated within spatial units but not between them. Spatial panel cointegration arises when the data are cointegrated through spatial lags between spatial units but not within them. Global panel cointegration arises when the data are cointegrated both within and between spatial units. Spatial error correction arises when error correction occurs within and between spatial units. We use nonstationary spatial panel data on the housing market in Israel to illustrate the methodology. We show that regional house prices in Israel are globally cointegrated in the long run and there is evidence of spatial error correction in the short run.

Disentangling the Short and Long-Run Effects of Occupied Stock in the Rental Adjustment Process

In the current stand of literature on the rental adjustment process starting with Hendershott et al. (Real Estate Economics, 30, 165-183, 2002a, Journal of Real Estate Finance and Economics, 24, 59-87, 2002b) it has become practice to treat the compound variable “occupied stock” as a supply variable. In this study we show that this variable deserves a more critical investigation and that the general view of a supply variable may be misleading. Using panel data covering 30 urban areas for 17 years, we investigate the rental adjustment process in the German office market. The application of recently developed cointegration techniques for non-stationary panel data in conjunction with the corresponding error correction model (ECM) enables us to overcome the data limitations, particularly existent for most European real estate markets. Hence, our primary motivation is (a) to demonstrate how “occupied stock” should be interpreted correctly and (b) to provide useful insights into the long-term relationships and short-run dynamics of real office prime rents. The empirical evidence suggests that a one percent rise in office employment increases real rents on average by 1.64% through higher demand for office space. On the other hand, a one percent increase in the supply of office space decreases real rents in the long run by 2.25%. The results from the error correction model show that deviations from the long-run equilibrium lead to an adjustment process which restores equilibrium within approximately 3 years.

Panel cointegration tests of the sustainability hypothesis in rich OECD countries

This study reexamines the sustainability hypothesis by testing whether government revenues and expenditures for eight rich OECD countries between 1977Q1 and 2005Q4 are cointegrated. For this purpose, a nonstationary panel data approach is adopted, which is general enough to permit for cross-country dependence as well as structural breaks representing major shifts in fiscal policy. In contrast to many earlier studies, the results reported in this study suggest that the sustainability hypothesis cannot be rejected.

Growth Convergence and Spillovers Among Indian States: What Matters? What Does Not?

Convergence and spillovers across countries and within countries are old, but recurrent policy concerns, and India is no exception to this rule. This paper examines convergence and spillovers across Indian states using non-stationary panel data techniques. Results on convergence among Indian states are generally found to be similar, but more nuanced, than previous studies. Generally speaking, there is evidence of divergence over the entire sample period, convergence during sub-periods corresponding to structural breaks, and club convergence. There is strong evidence of club convergence among the high- and low-income states; the evidence for middle-income states is mixed. Dynamic spillover effects among states are small.

Growth Convergence and Spillovers Among Indian States: What Matters? What Does Not?

Convergence and spillovers across countries and within countries are old, but recurrent policy concerns, and India is no exception to this rule. This paper examines convergence and spillovers across Indian states using non-stationary panel data techniques. Results on convergence among Indian states are generally found to be similar, but more nuanced, than previous studies. Generally speaking, there is evidence of divergence over the entire sample period, convergence during sub-periods corresponding to structural breaks, and club convergence. There is strong evidence of club convergence among the high- and low-income states; the evidence for middle-income states is mixed. Dynamic spillover effects among states are small.

A NOTE ON THE POOLING OF INDIVIDUAL PANIC UNIT ROOT TESTS

One of the most cited studies in recent years within the field of nonstationary panel data analysis is that of Bai and Ng (2004), in which the authors propose PANIC, a new framework for analyzing the nonstationarity of panels with idiosyncratic and common components. The problem is that the asymptotic validity of PANIC as a platform for constructing pooled panel unit root tests based on averaging is not fully proven. This paper provides the required results, whose usefulness is verified through simulations.

The sustainability of trade accounts of the G-7 countries

This article has examined the sustainability of the trade accounts of the G-7 countries by analysing the long-run equilibrium between exports and imports. We applied the techniques of the recently developed nonstationary panel data analysis and found that the trade accounts of the G-7 countries are very likely to be nonstationary variables. This implies that trade accounts are not sustainable for the G-7 countries. Therefore, some appropriate policies must be implemented to prevent trade imbalances from increasing.

International term structure of interest rates in the Euro area

This article empirically analyses the international term structure of interest rates in the Euro area over the period from 1999 to 2006. To address the small sample problem, we apply the nonstationary panel data analysis to two data sets: (1) seven countries (Belgium, Finland, France, Germany, Italy, the Netherlands and Spain) from 1999 to 2006 and (2) eight countries (Belgium, Finland, France, Germany, Greece, Italy, the Netherlands and Spain) from 2001 to 2006. Our results support the expectations theory of the term structure of interest rates.

An Examination of Exchange Rate Pass-Through to U.S. Motor Vehicle Products and Auto-Parts Import Prices

A distinctive feature of present globalization is the development of international production sharing activities (i.e. production fragmentation). The recent developments in transportation and communication technologies led to a surge in intermediate goods trade. However, intermediate goods trade is often neglected in the empirical studies of the exchange rate pass-through (ERPT). Using import unit values of 79 motor vehicle products and 245 auto-part products, which are classified by the 10-digit level of Harmonized Tariff Schedule (HTS), this study examines the pass-through of exchange rate changes into the U.S. auto-industry import prices from 5 major trading partners for the period of 1998.01 to 2006.12. Nonstationary panel data estimation techniques and tests for cointegration are employed in this study. Secondly, this study aims to compare the ERPT for the motor vehicle products (final goods) to the ERPT for the auto-parts (intermediate goods) in the U.S. The results suggest that import prices do not respond proportionately to the exchange rates and the estimated pass-through elasticities for motor vehicle products are lower than that for auto-parts.

Discrete choice modeling with nonstationary panels applied to exchange rate regime choice

Abstract This paper develops a regression limit theory for discrete choice nonstationary panels with large cross section ( N ) and time series ( T ) dimensions. Some results emerging from this theory are directly applicable in the wider context of M-estimation. This includes an extension of work by Wooldridge [Wooldridge, J.M., 1994. Estimation and Inference for Dependent Processes. In: Engle, R.F., McFadden, D.L. (Eds.). Handbook of Econometrics, vol. 4, North-Holland, Amsterdam] on the limit theory of local extremum estimators to multi-indexed processes in nonlinear nonstationary panel data models. It is shown that the maximum likelihood (ML) estimator is consistent without an incidental parameters problem and has a limit theory with a fast rate of convergence N 1 / 2 T 3 / 4 (in the stationary case, the rate is N 1 / 2 T 1 / 2 ) for the regression coefficients and thresholds, and a normal limit distribution. In contrast, the limit distribution is known to be mixed normal in time series modeling, as shown in [Park, J.Y., Phillips, P.C.B., 2000, Nonstationary binary choice. Econometrica, 68, 1249–1280] (hereafter PP), and [Phillips, P.C.B., Jin, S., Hu, L., 2007. Nonstationary discrete choice: A corrigendum and addendum. Journal of Econometrics 141(2), 1115–1130] (hereafter, PJH). The approach is applied to exchange rate regime choice by monetary authorities, and we provide an analysis of the empirical phenomenon known as “fear of floating”.

A note on the use of the LLC panel unit root test

One of the single most cited studies within the field of nonstationary panel data analysis is that of LLC (Levin et al. in J Econom 98:1–24, 2002), in which the authors propose a test for a common unit root in the panel. Using both theoretical arguments and simulation evidence, we show that this test can be misleading unless it is based on the same bandwidth selection rule used by LLC.

Purchasing Power Parity for Developing and Developed Countries: What can we Learn from Non-Stationary Panel Data Models?

The aim of this paper is to apply recently developed panel cointegration techniques proposed by Pedroni (1999, 2004) and generalized by Banerjee and Carrion-i-Silvestre (2006) to examine the robustness of the PPP concept for a sample of 80 developed and developing countries. We find that strong PPP is verified for OECD countries and weak PPP for MENA countries. However in African, Asian, Latin American and Central and Eastern European countries, PPP does not seem relevant to characterize the long-run behavior of the real exchange rate. Further investigations indicate that the nature of the exchange rate regime doesn’t condition the validity of PPP which is more easily accepted in countries with high than low inflation.

Testing the stochastic convergence of Italian regions using panel data

This paper examines stochastic convergence in real per capita GDP for Italian regions using recent non-stationary panel data methodologies over the period 1951 to 2002. Economies stochastically converge when regional differences across economies are not persistent, and long-run movements in a region's real per capita GDP are driven by technological shocks. Four panel unit root tests are used to evaluate stochastic convergence and Monte Carlo simulation is performed to obtain critical values. The results indicate that there is no stochastic convergence among Italian regions. They also provide some evidence in favour of stochastic convergence of regions in the Northern part of Italy.

The effectiveness of international knowledge spillover channels

Using panel data from 16 Organization for Economic Cooperation and Development (OECD) countries for the period 1981–2000, we examine the significance of international knowledge spillovers through inward and outward foreign direct investment (FDI), intermediate goods imports, and a disembodied direct channel. Knowledge spillovers through the disembodied direct channel are approximated by using a measure of technological proximity and patent citations between countries. Using estimation models that reflect recent developments in non-stationary panel data econometrics, we show that international knowledge spillovers through inward FDI and the disembodied direct channel are significant and robust. In contrast, outward FDI and imports of intermediate goods are not conducive to international knowledge spillovers.

Current Account Deficit Sustainability: a Panel Approach

Abstract In this paper we attempt to examine the issue of sustainability of current account imbalances in eight East Asia countries using the latest developments in nonstationary panel data analysis. The methods of nonstationary time series panels provide a much more promising explanation than would an analysis based on pure time series or cross section data. The empirical results clearly indicate that the current account imbalances were not on the long-run steady state in the pre-crisis era (1970-1997). This leads to the conclusion that the current accounts of Asia-8 during this period were unstable and did not move towards external account equilibrium. However, strong comovements between exports and imports are found in the extended sample period that includes the post-crisis period (1970-2000). This result implies that large currency depreciations and the economic recovery have brought the Asia-8 economies back on a sustainable path. Thus, current account imbalances may be used as an indicator (or warning signal) in predicting future crises.

Bootstrapping and hypothesis testing in non-stationary panel data

This paper uses the wild bootstrap to compute empirically relevant critical values for the test statistics proposed by Emerson and Kao (2001). Monte Carlo simulations were then performed to evaluate the size and power properties of the bootstrapped tests.

Estimating cross-section common stochastic trends in nonstationary panel data

This paper studies large-dimension factor models with nonstationary dynamic factors, also referred to as cross-section common stochastic trends. We consider the problem of estimating the dimension of the common stochastic trends and the stochastic trends themselves. We derive the rates of convergence and the limiting distributions for the estimated common trends and for the estimated loading coefficients. Generalized dynamic factor models with nonstationary factors are also considered. Cointegration among the factors is permitted. The method is applied to the study of employment fluctuations across 60 industries for the U.S. We examine the hypothesis that these fluctuations can be explained by a small number of aggregate factors. We also test whether some observable macroeconomic variables are the underlying factors.

Convergence in Sub-Saharan Africa: a nonstationary panel data approach

Given the development of time series econometrics and nonstationary data analysis, St. Aubyn (Empirical Economics, 24, 23–44, 1999) demonstrates a new paradigm for testing income convergence, or better defined, income stability, namely testing the stationarity of pair-wise income differentials. In this paper, a panel data set of Sub-Saharan African countries is constructed and panel cointegration and unit root tests are used to investigate the convergence properties of incomes and standards of living within Africa. Overall, little evidence is found to substantiate claims of convergence across Africa, although in some cases, smaller convergence clubs within Africa may be found. In addition the use of nonstationary panel data techniques is proposed for the testing and establishing of coherent convergence clubs.

The Feldstein–Horioka puzzle revisited

Krol (1996) argues that the Feldstein–Horioka puzzle is the result of the estimation technique that using fixed-effect panel regression, low saving-retention coefficients are found. However, Krol’s study has been criticized as revolving around the inclusion of Luxembourg. This paper reexamines the Luxembourg problem by applying Kao and Chiang’s (2001) DOLS and FMOLS estimators to non-stationary panel data of 20 OECD countries. Our empirical evidences show that the inclusion or exclusion of Luxembourg does not affect the estimation results; instead, the power of the estimation technique and of the hypothesis test matters. This result is interpreted as the higher power of cointegration analysis in panel data.