Tourism Arrivals and Environmental Intensity: Evidence from Symmetric and Asymmetric Panel ARDL Models

This paper examined the theory of purchasing power parity (PPP) for a group of developed and developing countries from January 2003 to May 2016 using both linear and nonlinear panel autoregressive distributed lag (ARDL) models. In addition, the paper extended a time series nonlinear ARDL model to a panel nonlinear ARDL model in testing for the PPP. Further, several panel tests of unit root were carried out to inspect the stationary properties of the variables. Outcome of the tests showed that the variables are a combination of I(1) and I(0). Since we have a combination of I(1) and I(0), linear and nonlinear panel ARDL models were estimated. The linear ARDL models were not valid since they failed to provide evidence for cointegration. However, the extended nonlinear panel ARDL models provided evidence of cointegration indicating that the PPP theory is valid for this group of countries. Unlike previous studies on the PPP, this study made a significant contribution by the provision of useful policy implication on the results found.

Based on a panel ARDL (AutoRegressive Distributed Lag) model, this paper investigates the environmental and economic consequences of energy tax levied in the four Nordic countries and the G7 countries from 1994 to 2016. Based on the double dividend theory of environmental tax, this paper investigates five variables: energy tax, energy consumption per unit of GDP, income tax, profit tax and capital gains tax, carbon dioxide intensity and fossil fuel burning ratio. A panel ARDL model is established to empirically test the relationship between energy tax and other variables. Experiments show that both the four Nordic countries and the G7 countries have found the existence of green dividends in the long run: the green dividends of the four Nordic countries are reflected in the reduction of carbon dioxide emissions, while those of the G7 countries are reflected in the reduction of fossil fuel use. In terms of blue dividends, the implementation of energy tax in the four Nordic countries can not only reduce distorted taxes in the short term, but also promote economic growth and adjust tax structure in the long term. For the G7 countries, blue dividends are not reflected in the long term. The model used in this paper is a panel ARDL model, which is more suitable for the study of multiple countries, multiple variables and long-term cycles. This model has been seldomly used in previous studies. The application of the panel ARDL model in this paper is not only more scientific and applicable, but also more innovative, which makes up for the shortcomings of previous studies. The research object of this paper selects the energy tax, which is an important part of the environmental tax system, and strives to provide a reference for the implementation of environmental taxation priorities and effects through empirical research. This paper may also serve as a reference for other countries to establish and improve environmental tax. As the first environmental tax law in China, the Environmental Protection Tax Law of the People’s Republic of China was formally implemented on 1 January 2018. This paper chooses G7 countries and Nordic countries as the research objects. As these are important economies in the world, their environmental tax implementation is more perfected and has strong representativeness. This study can provide some experience for the continuous improvement of China’s environmental tax law.

This study aims to approach the Fisher effect issue from a different methodological perspective. To this aim, the nonlinear autoregressive distributed lag (ARDL) model, recently developed by Shin et al. (2014), is applied for South Korea between 2000Q4Ƀ2017Q4. This model allows us to decompose one variable (changes in inflation) into two new variables (increases and decreases in inflation) under the manners of nonlinearity and asymmetry. Hence, it enables us to monitor the Fisher effect in terms of increases and decreases separately. We also apply the linear version of the same model since the nonlinear ARDL model is the extended version of linear ARDL model. While the empirical findings of the nonlinear model support asymmetrically partial Fisher effects in the long-run for 1, 3, 5 and 10-years Korean bond rates, the linear model does not. Additionally, the nonlinear model detects lower size partial Fisher effects when the maturity of interest rates gets longer. Another finding of this study is that the nonlinear model may mathematically identify and introduce a different version of the partial Fisher effect based on singular (separate) effects of each decomposed variable in a parametric manner.

Purpose The COVID-19 pandemic and the onset of the conflict in Ukraine led to a sustained downturn in tourist arrivals (TA) in Russia. This paper aims to explore the influence of geopolitical risk (GPR) and other indices on TA over 1995–2023. Design/methodology/approach We employ a nonlinear autoregressive distributed lag (NARDL) model to analyze the effects, capturing both the positive and negative shocks of these variables on TA. Findings Our research demonstrates that the NARDL model is more effective in elucidating the complex dynamics between macroeconomic factors and TA. Both an increase and a decrease in GPR lead to an increase in TA. A 1% negative shock in GPR leads to an increase in TA by 1.68%, whereas a 1% positive shock in GPR also leads to an increase in TA by 0.5%. In other words, despite the increase in GPR, the number of tourists coming to Russia increases by 0.5% for every 1% increase in that risk. Several explanations could account for this phenomenon: (1) risk-tolerant tourists: some tourists might be less sensitive to GPR or they might find the associated risks acceptable; (2) economic incentives: increased risk might lead to a depreciation in the local currency and lower costs, making travel to Russia more affordable for international tourists; (3) niche tourism: some tourists might be attracted to destinations experiencing turmoil, either for the thrill or to gain firsthand experience of the situation; (4) lagged effects: there might be a time lag between the increase in risk and the actual impact on tourist behavior, meaning the effects might be observed differently over a longer period. Originality/value Our study, employing the NARDL model and utilizing a dataset spanning from 1995 to 2023, investigates the impact of GPR, gross domestic product (GDP), real effective exchange rate (REER) and economic policy uncertainty (EPU) on TA in Russia. This research is unique because the dataset was compiled by the authors. The results show a complex relationship between GPR and TA, indicating that factors influencing TA can be multifaceted and not always intuitive.

The asymmetric effects of oil price changes on unemployment: Evidence from Canada and the U.S

Fluctuations in the oil price profoundly impact many other prices in the economy, as oil is used to produce numerous goods and services. While literature is ample regarding the linear relationship between crude oil prices and food prices, academic discussion on the presence of non-linear relationships is relatively evolving. This paper strives to explore the existence of both linear and non-linear relationships between crude oil price and food price in Bangladesh by employing the autoregressive distributed lag (ARDL) model and the non-linear autoregressive distributed lag (NARDL) model, respectively. The ARDL model indicates that the crude oil price has a linear and positive impact on food price inflation in Bangladesh. The NARDL model finds no asymmetric relationship between the two variables in the long run. As a result, in the long run, the response of food price inflation in Bangladesh is the same whether oil prices increase or decrease. However, the NARDL model reveals that the change in oil prices asymmetrically impacts food price inflation only in the short run. These findings are essential for further study, and the results can be used for policymaking to ensure food security in Bangladesh.

PurposeThis study aims to examine the symmetric and asymmetric impact of external debt on inflation in Sudan from 1970 to 2020 within a multivariate framework by including money supply and the nominal effective exchange rate as additional inflation determinants.Design/methodology/approachThe authors utilize an Auto Regressive Distributed Lag (ARDL) model to examine the symmetric impact of external debt on inflation, while the asymmetric impact is examined using a Nonlinear ARDL (NARDL) model. The existence of a long-run relationship between inflation and external debt is tested using the bounds-testing approach to cointegration, and a vector error-correction model is estimated to determine the short parameters of equilibrium dynamics.FindingsThe linear ARDL model results show that external debt has no statistically significant impact on inflation in the long run. On the contrary, the results of the NARDL model show that positive and negative external debt shocks statistically affect inflation in the long run. The estimated long-run elasticity coefficients of the linear and nonlinear ARDL models reveal that the domestic money supply has a statistically significant positive impact on inflation. In contrast, the nominal effective exchange rate has a statistically significant negative impact on inflation.Practical implicationsThe reliance on symmetric analysis may not be sufficient to uncover the existence of a linkage between external debt and inflation. Proper external debt management is crucial to control inflation rates in Sudan.Originality/valueTo date, no empirical study has assessed the external debt-inflation nexus and its potential asymmetry in Sudan, and the current study aims to fill this gap in the literature.

This paper employs the linear autoregressive distributed lag (ARDL) model, the asymmetric nonlinear ARDL (NARDL) model developed by Shin, et al (2014) to examine the asymmetric effect of oil price and exchange rates pass-through on inflation in Nigeria over a period of 1970 to 2020. The result of the asymmetric test revealed the existence of asymmetries among the variables of the study, suggesting that there is a nonlinear interaction among the variables used in the study. This validates the choice of a non-linear ARDL model for the study. Results of the long-run estimates show that rising (Positive) oil price shocks have a greater impact on inflation than falling (negative) oil price shocks. Furthermore, it is evident from the result that the depreciation of the exchange rate has a much and significant effect on inflation than the appreciation of the exchange rate in Nigeria. However, a rising interest rate increases inflation by 0.84 per cent while a falling interest rate increases inflation by 0.85 per cent. This implies that the effect of negative interest rate on inflation is higher than its positive effect on inflation, though, by a smaller amount of about 0.01 per cent. Again, the short-run dynamic model revealed a high speed of convergence of more than 90% from the short-run disequilibrium. During the study period, the oil price fluctuations showed a significant and incomplete pass-through to both exchange rates and inflation in Nigeria. Moreover, the results suggest that positive oil price changes have a larger impact than the negative ones, that the effect of an oil price shock on inflation and exchange rates is larger in the long-run than in the short-run, and that there is incomplete pass-through effect of oil price on domestic inflation and exchange rates. Based on the findings, the study recommends policies that set oil prices and exchange rates within reasonable limits to check inflation in Nigeria and should diversify its economy as well as withdraw the current subsidy regime completely.

PurposePrevious studies have mostly estimated there to be a symmetric effect in the Foreign direct investment (FDI) inflow regarding the economic growth of Central, East and Southeast European (CESEE) countries. However, for the CESEE countries, as well as for the majority of countries around the world, there has been no study that has estimated the symmetric and asymmetric effect of outward FDI on economic growth. The main objective of this study is to estimate whether the relation between outward FDI and economic growth in CESEE countries is symmetric or asymmetric.Design/methodology/approachThis study includes a sample based on eight CESEE countries. The authors used the linear and non-linear autoregressive distributed lag (ARDL) model and annual data for the period from 1990 to 2020.FindingsIn the long run, in the linear ARDL model, a significant symmetrical effect due to OFDI on the economic growth of Romania and Slovenia was found, while in the non-linear ARDL model, a significant asymmetric effect of OFDI on the economic growth of Bulgaria, Poland, Romania, Russia, Slovenia and Slovakia was found. In six out of the eight countries, asymmetry was found while symmetry was found in the other two. Poorer symmetry results can be ascribed to the lack of linear model neglecting the asymmetric behaviour of the positive and negative change decomposition as part of the OFDI movement, which leads to the wrong conclusion.Originality/valueThis is the first study to evaluate the asymmetric effect of outward FDI on the economic growth of eight CESEE countries.

The long run debt-growth nexus in sub-Saharan Africa is the main focus of this study. Its analysis is conducted on the basis of theoretical discussions and data considerations. The study employs both panel autoregressive distributed lag (PARDL) model and panel non-linear autoregressive distributed lag (PNARDL) model to examine the relationship between external debt and economic growth using a panel dataset of 22 countries from 1985 to 2015. Its results find that the level of investment significantly influences the long-run economic growth in both linear and non-linear models. However, the long-run coefficient of external debt is only significant in the non-linear model. There is strong evidence of error correction as the lagged GDP per capita is highly significant and negative in the two models. Key Words : debt, debt and growth, public debt and growth, linear regression models

Using a panel autoregressive distributed lag (ARDL) model, this paper investigates the impact of environmental, social and governance (ESG) factors on economic growth across BRICS+ countries from 2002 to 2022. The model allows analysis of both short- and long-run effects. We develop an index comprising 17 ESG variables across all three components to study the relationship between ESG and growth. Empirical results show a positive long-run association between ESG and growth for the BRICS+ bloc, however no detectable short-run dynamics. We also find no evidence of a long-term ESG effect on growth for the new member countries that joined in January 2024. This suggests these new members should prioritise developing ESG-related policies. Furthermore, we use a panel nonlinear ARDL (NARDL) model to identify asymmetric effects. Our findings reveal that there are differences in how economic growth responds to positive and negative changes related to ESG factors, highlighting the need to consider asymmetries in the ESG-growth nexus. Finally, the results of this study offer valuable insights to policymakers seeking to promote sustainable development in BRICS+ countries.

Purpose The purpose of this study is to provide empirical evidence on the impact of deviations from the long-run sustainable real exchange rate (RER) equilibrium on real economic growth rate applying panel autoregressive distributed lag model (ARDL) (Pooled Mean Group, Mean Group and Dynamic Fixed Effects estimators) in a dynamic heterogeneous panel setting and panel NARDL for the largest database covering 104 countries during 1995–2022 period developed by Couharde et al. (2017). Design/methodology/approach The EQCHANGE database makes available not only the equilibrium RER but also misalignments according to the Behavioral Equilibrium Exchange Rate approach for each country. One of the main objectives is to examine whether undervaluation or overvaluation RER can imply different responses on economic performance trying to differentiate between short and long run effects. Additionally, the authors consider the World Bank (WB)’s income classifications to compare the asymmetries attending to high-income, upper-middle-income, lower-middle-income and low-income levels. Findings Applying the panel ARDL technique, the results suggest that the RER misalignments have a negative but not significant effect on the short-run, nevertheless a negative and highly significant impact on real economic growth rate is detected on the long-run. Considering the panel NARDL, the asymmetric relationship between RER misalignment and economic growth rate is supported considering all countries in the long-run (in the short-term is not significant). In the long run is detected that undervaluation can promote economic growth rate, rather than overvaluation which can harm the economic performance. Additionally, the WB and the International Monetary Fund (IMF) income’s classifications have been applied and the long-run symmetry test is strongly rejected regardless of income group. Originality/value To the best of the author knowledge, this is the first time the non-linear panel ARDL methodology has been applied for analyzing the impact of deviations from the long-run sustainable RER equilibrium on real economic growth. This allows us to see the asymmetric effect not seen before. The panel ARDL estimation can efficiently performed regardless of the integration level of the variables, additionally, it is consistent even in the presence of endogeneity. Besides, another advantage of this method is that it is possible to reflect not only the short but also the long-run dynamics. Moreover, this analysis offers a comparison between linear panel ARDL and non-linear to compare the advantages from the former. Additionally, this study covers the largest database, in particular, 104 countries during the 1995–2022 period implemented with the Couharde et al. (2017) EQCHANGE database. Finally, it is compared the asymmetries attending to different income classifications.

This article utilised the panel autoregressive distributed lag (ARDL) model to examine the link between macroeconomic developments and carbon emissions for 15 Asian countries from 1990 to 2013. The results of the panel ARDL model showed that there exist long-run equilibrium relationships between principal macroeconomic variables and carbon emissions in the sample. The long-run elasticity of renewable energy and fossil fuels energy consumption with respect to CO2 emissions was calculated as −0.299 and 0.967, respectively. The long-run elasticity of GDP, financial development, urban population density and industry value added share with respect to CO2 emissions was calculated as 0.473, 0.079, −0.633 and −0.10, respectively. FDI was significantly negatively related to CO2 emissions which were calculated as −0.06 in the short-run. These results suggested that FDI inflow was not yet an environmental threat for Asian economies. Renewable energy and upgrades to industry value added share will help various governments mitigate carbon emissions.

BackgroundHuman behavioral responses to changes in risks are often delayed. Methods for estimating these delayed responses either rely on rigid assumptions about the delay distribution (e.g., Erlang distribution), producing a poor fit, or yield period-specific estimates (e.g., estimates from the Autoregressive Distributed Lag (ARDL) model) that are difficult to integrate into simulation models. We propose a hybrid ARDL–Erlang approach that yields an interpretable summary of behavioral responses suitable for incorporation into simulation models.MethodWe apply the ARDL–Erlang approach to estimate the effect of COVID-19 deaths on mobility across US counties from October 2020 to July 2021. A standard panel autoregressive distributed lag (ARDL) model first estimates the effect of past deaths and past mobility on current mobility. The ARDL model is then transformed into an Infinite Distributed Lag (IDL) model consisting of only past deaths. The coefficients of the past deaths are aggregated into an overall effect and fit to an Erlang distribution, summarized by average delay length and shape parameter.ResultsOur results show that on the national level, a one-standard-deviation permanent increase in weekly deaths per 100,000 population (log-transformed) is associated with a 0.46-standard-deviation decrease in human mobility in the long run, where the delay distribution follows a first-order Erlang distribution, and the average delay length is about 3.2 weeks. However, there is much heterogeneity across states, with first- to third-order Erlang delays and 2 to 18 weeks of average delay providing a theoretically cogent summary of how mobility followed changes in deaths during the first year and a half of the pandemic.ConclusionThis study provides a novel approach to estimating delayed human responses to health risks using a hybrid ARDL-Erlang model. Our findings highlight significant variability in the impact and timing of responses across states, underscoring the need for tailored public health policies. This study can also serve as guidelines and an example for identifying delayed human behavior in other settings.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12874-025-02696-x.

This study evaluates and compares the forecasting performances of the ARDL (AutoRegressive Distributed Lag), LSTM (Long Short-Term Memory), and XGBOOST (Extreme Gradient Boosting) models on the MASI (Moroccan All Shares Index). The analysis incorporates daily new COVID-19 cases into the ARDL approach to investigate short-term and long-term relationships with MASI. Cointegration and causality tests are conducted on daily time series data. In terms of accuracy, the ARDL model, especially when including trend and seasonality variables, outperforms LSTM and XGBOOST models. The ARDL model with lags, trend, and seasonality variables achieves the lowest Mean Absolute Percentage Error (MAPE) of 26.7%, with a processing time of 1 second. In comparison, the LSTM and XGBOOST models have MAPE values of 30.5% and 32%, respectively, while requiring significantly longer processing times. These findings suggest that the ARDL model is more efficient and accurate in predicting future values of MASI under pandemic conditions.