Spatial Fixed Effects Research Articles

NONLINEAR AND TIME-VARYING HETEROGENEITY OF THE SPATIAL EFFECT OF G20 COUNTRIES’ MONETARY POLICIES

This study examines how nonlinear and time-varying heterogeneity of the spatial spillover effects of price-based and quantitative monetary policy instruments affect G20 countries’ economic growth by building spatial panel smooth transition models with a trade geography weight matrix. We find spatial dependence between economic growth and monetary policy in G20 countries. Then, we use G7+Spain and BRICS to represent developed and emerging economies, respectively, and re-estimate the spatial fixed effect. G7 countries’ quantitative monetary policy is not significant, and reduced interest rates do not improve GDP; however, both increased money supply and reduced interest rates improve BRICS countries’ GDP. Furthermore, lower interest rates in G7 countries cannot improve GDP during bad times, but during good times, price-based monetary policy effectiveness is improved. Although economic conditions do not affect BRICS countries’ quantitative regulation, they do influence price-based monetary policy’s effects. Neighboring countries’ price-based regulation brings different degrees of spillovers to domestic GDP using both linear and nonlinear models.

Regional early-stage entrepreneurship in the European Union

Purpose: Entrepreneurship is often viewed as a driver of the global economy. However, previous research on the relationship between entrepreneurship and economic growth shows contradictory results depending on the research settings. The purpose of this paper is to investigate how early-stage entrepreneurship - including only enterprises that are less than three and a half years old - affects regional economic growth in the European Union. Methodology: The methodology includes three methods: bivariate correlation, fixed effects regression with region and time fixed effects and spatial fixed effects regression. The panel sample consists of 273 NUTS 2 regions between 2008 and 2017. Results: The results support the hypothesis of this research and show that early-stage entrepreneurship has a mild positive effect on the economic growth of European regions. However, the potential bidirectional nature of this relationship obliterates the ability to comment on the causality of this link. The percentage of people in the active population employed in human resources in science and technology and gross fixed capital formation have a significant and impactful effect on regional GDP. Conclusion: The conclusion can be drawn that the effect of early-stage entrepreneurship on regional economic growth is conditioned by the population density of the region. Although these results show that enterprises founded in densely populated areas such as cities and metropolitan areas tend to have a larger effect on the regional economy, the results are ambiguous.

The role of R&D in the effectiveness of renewable energy determinants: A spatial econometric analysis

This study investigates the effects of renewable energy determinants on various renewable energy resources. The empirical results are based on data from 31 mainly Asia-Pacific countries observed from 2000 to 2018. Different diagnostic tests indicate the existence of spatial effects in the renewable energy model. The spatial Durbin model with the time-period and spatial fixed effects was selected for analysis. Scrutinizing empirical studies suggest a lack of theoretical consensus on how renewable energy determinants work. By examining how R&D affects different renewable energy resources, this study attempts to fill the research gap. In addition, the R&D environment is an important factor in determining the effects of GDP and the development of financial markets and institutions. R&D enhancement reduces the impacts of market expansion on hydropower; however, such effects are incremental for solar, wind, bioenergy, and geothermal energy resources. In an environment with a low level of R&D, financial development has positive effects on hydropower generation and negative impacts on other renewable energy types. However, the expansion of R&D expands reverses such effects gradually. The different results are obtained for different countries at different levels of R&D. Furthermore, trade openness has been a positive factor in the development of hydropower, solar and geothermal energy resources. The results also revealed the negative (positive) impacts of GDP per capita in neighboring countries with a higher (lower) level of R&D on the development of renewable energy of the origin country.

Is it the flood, or the disclosure? An inquiry to the impact of flood risk on residential housing prices

In 2015, the government of Taipei, Taiwan, published a series of flood risk maps based on simulation results. It sparked a broad public debate on flood risk information and the real estate market. This study therefore investigates the effect of making flood risk information public on residential housing prices and makes the following contributions. The first contribution concerns public policy. By analysing the relationship between the published flood risk information and more than 12,000 central Taipei sales records in 2015 and 2016, this study finds that the effect of flood risk on housing prices is significant and negative both before and after the maps were disclosed. Although some local variation in the effect of flood risk after the map disclosure was identified, there is no evidence that the disclosure altered the effect on housing prices. This finding suggests that the disclosure of the flood risk maps impacted the effect of the flood risk locally without influencing the market as a whole. Another contribution concerns perception. The result indicates that homebuyers’ negative attitude towards flood risk mainly focuses on the risk under low-intensity, high-frequency rainfall conditions. The final contribution of this study concerns its methodological perspective. By combining the spatial fixed-effect (SFE) model with geographically weighted regressions (GWRs), this study successfully mitigates the endogeneity problem while addressing the heterogeneity problem.

Effects of Market Structure on Broadband Quality in Local U.S. Residential Service Markets

This paper investigates the effects of changes in market structure on residential service quality in local U.S. broadband markets. Our analysis focuses on identifying causal effects on quality from entry (or exit) by firms using both legacy and non-legacy broadband network technologies. We analyze a very large and highly disaggregated census block-level data set we have constructed using Federal Communication Commission (FCC) data on residential broadband services deployment in populated U.S. census blocks between December 2014 and December 2018. We employ a difference-in-differences framework combined with both spatial fixed effects and spatial instrumental variables. The latter econometric tools are used to address unobserved market heterogeneity and concerns about significant error in measured ISP counts for fixed wireless ISPs. Our results suggest that entry by legacy technology (cable and DSL) service providers has large impacts on available service quality in local broadband markets, effects which diminish rapidly with continued entry. Contrary to expectations, entry by a single new fiber ISP seems to have no impact on available quality; entry by two fiber ISPs, by contrast, raises maximum available quality by a modest but statistically significant 90 Mbps. Entry by a fixed wireless ISP appears to have had a large, negative impact on maximum available download speeds offered by incumbent legacy ISPs. Economic explanations for these results are discussed and evaluated. We establish that (i) both fixed wireless and fiber ISPs compete in sets of speed tiers that are distinct from those offered by legacy ISPs, and speculate that (ii) price competition (as opposed to quality competition) may be more intense in markets with fixed wireless entry.

Spatiotemporal Variations in Energy Consumption and Their Influencing Factors in China Based on the Integration of the DMSP-OLS and NPP-VIIRS Nighttime Light Datasets

With the speedy growth of economic development, the imbalance of energy supply and demand pose a critical challenge for the energy security of our country. Meanwhile, the increasing and excessive energy consumption lead to the greenhouse effect and atmospheric pollution, greatly threatening the survival and development of human beings. This study integrated two nighttime light remote sensing datasets, namely Defense Meteorological Satellite Program’s Operational Linescan System (DMSP-OLS) data and Suomi National Polar-orbiting Partnership (NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) data, to extend the temporal coverage of the study. Then, the distributions of China’s energy consumption from 1995 to 2016 at a 1-km resolution were estimated using different models and the spatiotemporal variations of energy consumption were explored on the basis of the best estimated results. Next, the factors influencing China’s energy intensity on the provincial level were investigated based on the spatial econometric model. The results show that: (1) The integrated nighttime light datasets can be successfully applied to estimate the dynamic changes of energy consumption. Moreover, the panel data model established in our research performed better than the quadratic polynomial model. (2) During the observation period, the energy consumption in China significantly increased, especially in the Yangtze River Delta, the Pearl River Delta, the Beijing–Tianjin–Hebei region, eastern coastal cities, and provincial capitals. (3) Different from the random spatial distribution pattern of energy consumption on the provincial level, the spatial distribution of energy consumption on the prefectural level has significant clusters, and its spatial agglomeration was strengthened year by year during the research period. (4) The spatial Durbin model (SDM) with a spatial fixed effect has been proved to be more suitable to explore the impact mechanism of China’s energy consumption. Among the four socio-economic factors, industrial structure has the greatest impact on the provincial energy intensity in China. Moreover, the changes in industrial structure and foreign direct investment (FDI) can not only influence the local energy intensity but also affect the energy intensity of the neighboring provinces.

Racial-Ethnic Diversity and the Decline of Predominantly-White Mainline and Evangelical Protestant Denominations: A Spatial Fixed-Effects Approach

Predominantly white Mainline and Evangelical Protestant denominations are in a state of stagnation or decline in the United States. This is partially because these denominations fail to thrive in the ever-growing racial-ethnic diversity of the U.S. In fact, despite increases in the non-white population, diversity within these Protestant groups remains low. The organizational ecology approach asserts that the mismatch between the potential pool of adherents and the demographic niche occupied by the white Protestant groups encourages decline in areas that are diversifying. Past research supporting this theory, however, have relied on limited samples/timeframes and neglected the spatial autocorrelation of religious adherent rates and its predictors. This study re-examines the linkage between racial-ethnic minority growth and white denominations’ adherent decline using a fixed-effects spatial Durbin model to predict the adherent rates of Mainline and Evangelical Protestants in U.S. counties from 1980 to 2010. Results from the non-spatial fixed effects model fail to identify a significant association for Mainline denominations, but adjusting for spatial dependency reveals a direct effect for Mainline denominations and a spatial spillover effect for Evangelicals. Future studies on religious change and diversity must take their spatial structure into account.

Spatial risk adjustment between health insurances: using GWR in risk adjustment models to conserve incentives for service optimisation and reduce MAUP.

This paper presents a new approach to deal with spatial inequalities in risk adjustment between health insurances. The shortcomings of non-spatial and spatial fixed effects in risk adjustment models are analysed and opposed against spatial kernel estimators. Theoretical and empirical evidence suggests that a reasonable choice of the spatial kernel could limit the spatial uncertainty of the modifiable area unit problem under heavy-tailed claims data, leading to more precise predictions and economically positive incentives on the healthcare market. A case study of the German risk adjustment shows a spatial risk spread of 86 Euro p.c., leading to incentives for spatial risk selection. The proposed estimator eliminates this issue and conserves incentives for services optimisation.

Learning from man or machine: Spatial fixed effects in urban econometrics

Econometric models with spatial fixed effects (FE) require some kind of spatial aggregation. This aggregation may be based on postcode, school district, county or some other spatial subdivision. Common sense would suggest that the less aggregated, the better inasmuch as aggregation over larger areas tends to gloss over systematic spatial variation. On the other hand, low spatial aggregation results in thin data sets and potentially noisy spatial fixed effects. We show, however, how this trade-off can be substantially lessened if we allow for more flexible aggregations. The key insight is that if we aggregate over areas with similar location premiums, we obtain robust location premiums without glossing over too much of the spatial variation. We use machine learning in the form of a genetic algorithm to identify areas with similar location premiums. The best aggregations found by the genetic algorithm outperform a conventional FE by postcode, even with an order of magnitude fewer spatial controls. This opens the door for spatially sparse FEs, if economy in the number of variables is important. The major takeaway, however, is that the genetic algorithm can find spatial aggregations that are both refined and robust, and thus significantly, lessen the trade-off between robust and refined location premium estimates.

The effects of cultural heritage on residential property values: Evidence from Lisbon, Portugal

We estimate the cultural heritage amenity impact on the residential real estate market of Lisbon, Portugal, paying particular attention to heterogeneity of types and capturing spatial dependencies through a spatial error model. Our heritage amenities include conservation areas, listed historic buildings, and designated heritage which includes churches, palaces, historic buildings and lithic structures. We construct gridded spatial fixed effects which mitigate biases from the modifiable areal unit problem, and further employ a geographic regression discontinuity ensuring the robustness of results. The analysis is complemented with spatial interactions and mixed geographically weighted regressions (MGWR) to explore spatial heterogeneity of impacts.Conservation areas yield 4.1% premiums with spillover benefits of 3.3%, while proximity to designated heritage has a positive price elasticity of 0.0075. This impact is equivalent to an additional designated heritage within 100 m. Ten additional listed buildings within 500 m on the other hand yield 0.5% premiums.We find spatial variation in heritage amenity impacts with MGWR and spatial interactions highlighting common patterns whereby positive price impacts are strongest for the closest properties, and the biggest for landmark amenities. We compare these two manners to evaluate spatial non-stationarity of impacts and highlight the benefits of high-level GIS techniques which are commonly lacking in hedonic studies with urban spatial data.

The Schumpeter creative destruction hypothesis: a spatial assessment on Portuguese regions

Shumpeter characterised the innovative role of the entrepreneur on economic growth (Schumpeter, 1942). By inventing new products, new processes or techniques, the entrepreneur searches for monopoly rents and forces the social and economic structure to evolve. The routines, characterised by accommodative behaviour, which may endure for long periods are suddenly broken, in a sequence of stagnation and evolutionary periods (Schumpeter, 1947). Therefore, according to the Schumpeter theory, innovation is at the core of economic growth, and goes further as it causes social and economic structural changes. The present article leads on with an empirical assessment about the Schumpeter creative destruction hypothesis at the Portuguese Regional level, using a spatial econometric framework, based on Elhorst (2003) and Bhattacharjee et al. (2012). We estimate a model with spatial fixed effect, exploring the relation between R&D intensity at the firm level, productivity and firm demography dynamics, using a panel data for the 28 Portuguese continental NUTsIII regions between 2006 and 2009. The results obtained confirm the association between productivity and firm demographic dynamics – i.e., regions with higher level of productivity present also higher rates of birth and mortality among firms. We have not found, as it would be predictable according to the Schumpeter hypothesis, statistical evidence of association between productivity and R&D intensity at the firm level. This result indicates that Portuguese firms behaviour is more of imitation rather than innovation.

Gender Gap in Personal Bankruptcy Risks: Empirical Evidence from Singapore

Using a unique dataset that merges bankruptcy and motor events with personal data in Singapore, this study finds significant evidence of gender gap in personal bankruptcy risk. We show that the women’ odds in bankruptcy events is 28% of the men’s odds controlling for demographic, housing types and spatial fixed effects. The gender gap is not race-neutral. We find that the gender effect is stronger in Chinese’s bankruptcy events, but weaker in Malay and Indian bankruptcy events, relative to the bankruptcy events in the control group (“other” race). The gender gap is also observed in multiple bankruptcy events and bankruptcy claim amounts. We use motor accident events as an instrument for risk-taking, and find similar evidence of the gender gap in bankruptcy events. Bankruptcy risks of debtors with past motor accident records are significantly higher than those without past motor accident records. The gender gap in bankruptcy risk has important implications for credit modelling and risk pricing in the finance and economic literature.

Gender and Household Financial Decision: Evidence from Personal Bankruptcy

Using a unique dataset that merges bankruptcy and motor events with personal data in Singapore, this study finds significant evidence of gender gap in personal bankruptcy risk. We show that the women’ odds in bankruptcy events is 28% of the men’s odds controlling for demographic, housing types and spatial fixed effects. The gender gap is not race-neutral. We find that the gender effect is stronger in Chinese’s bankruptcy events, but weaker in Malay and Indian bankruptcy events, relative to the bankruptcy events in the control group (“other” race). The gender gap is also observed in multiple bankruptcy events and bankruptcy claim amounts. We use motor accident events as an instrument for risk-taking, and find similar evidence of the gender gap in bankruptcy events. Bankruptcy risks of debtors with past motor accident records are significantly higher than those without past motor accident records. The gender gap in bankruptcy risk has important implications for credit modelling and risk pricing in the finance and economic literature.

Housing Market Fluctuations and the Implicit Price of Water Quality: Empirical Evidence from a South Florida Housing Market

In this study we utilize a hedonic property price analysis to examine changes in the implicit price of water quality given housing market fluctuations over time. We analyze Martin County, Florida waterfront home sales from 2001 to 2010 accounting for the associated significant real estate fluctuations in this area through flexible econometric controls in space and time. We apply a segmented regression methodology to identify housing market price instability over time, interact water quality with these identified market segmentations, and embed these interactions within a spatial fixed effect model to further account for any spatial heterogeneity in the waterfront market. Results indicate that water quality improvement is associated with higher property values. We find no evidence that the economic downturn crowded out concern for the water quality in this area. We further impute an implicit prices of $2614, evaluated at the sample mean, for 1 % point increase in the water quality grade.

An Alternative to the Standard Spatial Econometric Approaches in Hedonic House Price Models

Omitted, misspecified, or mismeasured spatially varying characteristics are a cause for concern in hedonic house price models. Spatial econometrics or spatial fixed effects have become popular ways of addressing these concerns. We discuss the limitations of standard spatial approaches to hedonic modeling and demonstrate the spatial generalized additive model as an alternative. Parameter estimates for several spatially varying regressors are shown to be sensitive to the scale of the fixed effects and bandwidth dimension used to control for omitted variables. This sensitivity reflects the uncertainty associated with the estimates when the appropriate spatial scale of the controls is unknown.

Gender and agricultural productivity: implications of the Farm Input Subsidy Program in Malawi

Alleviating gender differences in agricultural productivity is vital for poverty reduction. While numerous studies suggest that gender differences in agricultural productivity are a result of female farmers having limited access to resources, few studies investigate the role of agricultural interventions in alleviating the constraints to input use and subsequently the gender gap in productivity. This study investigates whether there are gendered gains in agricultural productivity from participating in an input subsidy program and if these gains help reduce the gender gap. Using nationally representative data that is disaggregated at the plot level, this study analyzes the large-scale voucher-based Farm Input Subsidy Program in Malawi. Focusing on the total value of output per hectare, relationships are identified using weighted estimators, where the weights are constructed from propensity scores, and spatial fixed effects, to address the unobservable factors that may confound the relationship between program participation and productivity. The findings suggest that participation in the program improves agricultural productivity for both male and female farmers but it does not provide disproportionate help to female famers to overcome gender disparities in agricultural productivity. This suggests that female farmers face additional constraints to productivity apart from nonlabor input use.

Seeds of distrust: conflict in Uganda

We study the effect of civil conflict on social capital, focusing on Uganda’s experience during the last decade. Using individual and county-level data, we document large causal effects on trust and ethnic identity of an exogenous outburst of ethnic conflicts in 2002–2005. We exploit two waves of survey data from Afrobarometer (Round 4 Afrobarometer Survey in Uganda, 2000, 2008), including information on socioeconomic characteristics at the individual level, and geo-referenced measures of fighting events from ACLED. Our identification strategy exploits variations in the both the spatial and ethnic intensity of fighting. We find that more intense fighting decreases generalized trust and increases ethnic identity. The effects are quantitatively large and robust to a number of control variables, alternative measures of violence, and different statistical techniques involving ethnic and spatial fixed effects and instrumental variables. Controlling for the intensity of violence during the conflict, we also document that post-conflict economic recovery is slower in ethnically fractionalized counties. Our findings are consistent with the existence of a self-reinforcing process between conflicts and ethnic cleavages.

Application of statistical modeling to optimize a coastal water quality monitoring program

The long-term water quality monitoring program implemented by the Massachusetts Water Resources Authority in 1992 is extensive and has provide substantial understanding of the seasonality of the waters in both Boston Harbor and Massachusetts Bay and the response to improvements in effluent quality and offshore transfer of the effluent in September 2000. The monitoring program was designed with limited knowledge of spatial and temporal variability and long-term trends within the system. This led to an extensive spatial and temporal sampling program. The data through 2003 showed high correlation within physical parameters measured (e.g., salinity, dissolved oxygen) and in biological measures such as chlorophyll fluorescence. To address the potential sampling redundancies in the measurement program, an assessment of the impact of reduced levels of monitoring on the ability to make water quality decisions was completed. The optimization was conducted by applying statistical models that took into account whether there was evidence of a seasonal pattern in the data. The optimization used model survey average readings to identify temporal fixed effects, model survey-average-corrected individual station readings to identify spatial fixed effects, corrected the individual station readings for temporal and spatial fixed effects and derived a correlation model for the corrected data, and applied the correlation model to characterize the correlation of annual average readings from reduced monitoring programs with true parameter levels. Reductions in the number of sampling stations were found less detrimental to the quality of the data for annual decision-making than reductions in the number of surveys per year, although there is less of a difference in this regard for dissolved oxygen than there is for chlorophyll. The analysis led to recommendations for a substantially lower monitoring effort with minimal loss of information. The recommendation supported an annual budget savings of approximately $183,000. Most of the savings was from fewer surveys as approximately $21,000 came from the reduction in the number of stations monitored from 21 to 7 and associated laboratory analytical costs.

Agglomeration effects in Europe

The paper estimates agglomeration effects for France, Germany, Italy, Spain, and the UK. Estimation takes into account endogeneity of the spatial distribution of employment and spatial fixed effects. Empirical results suggest that agglomeration effects in these European countries are only slightly smaller than agglomeration effects in the US: the estimated elasticity of (average) labor productivity with respect to employment density is 4.5 percent compared to 5 percent in the US.

Agglomeration-Effects in Europe

The paper estimates agglomeration-effects for France, Germany, Italy, Spain, and the UK. Estimation takes into account endogeneity of the spatial distribution of employment and spatial fixed-effects. Empirical results suggest that agglomeration-effects in these European countries are only slightly smaller than agglomeration-effects in the US: the estimated elasticity of average-labor-productivity with respect to employment-density is 4.5 percent compared to 5 percent in the US.