Are Chinese transport policies effective? A new perspective from direct pollution rebound effect, and empirical evidence from road transport sector
aa r X i v : . [ q -f i n . E C ] O c t Are Chinese transport policies effective? A newperspective from direct pollution rebound effect,and empirical evidence from road transport sector
Lu-Yi QIU and Ling-Yun HE , , , ∗
1. College of Economics and Management, China Agricultural University,Beijing 100083, China2. Institute of Resource, Environment and Sustainable Development Research,JiNan University, Guangzhou 510632, China3. School of Economics, JiNan University, Guangzhou 510632, China4. School of Economics and Management, Nanjing University of Information Science and Technology,Nanjing 210044, China* Corresponding author.Emails: [email protected]; [email protected] on December 9, 2016 ∗ Dr. HE is the corresponding author. Dr. HE is a full professor of energy economics and environmentalpolicies. QIU is a Ph.D. candidate supervised by Dr. HE. The authors contribute equally in the project.HE conceived the whole project. QIU calculated and analysed the results under Dr. HE’s supervision. HEand QIU co-wrote the manuscript. The authors would like to thank Dr. YANG Sheng, Dr. CHEN Su-Mei,XU Feng, LIU Li, OU Jia-Jia, WEI Wei, and all other colleagues from both China Agricultural Universityand JiNan University, for all their warm helps, constructive suggestions and pertinent comments. Thisproject is supported by the National Natural Science Foundation of China (Grant Nos. 71273261 and71573258), and China National Social Science Foundation (No. 15ZDA054). bstract The air pollution has become a serious challenge in China. Emissions from motorvehicles have been found as one main source of air pollution. Although the Chinesegovernment has taken numerous policies to mitigate the harmful emissions fromroad transport sector, it is still uncertain for both policy makers and researchers toknow to what extent the policies are effective in the short and long terms. Inspiredby the concept and empirical results from current literature on energy rebound effect(ERE), we first propose a new concept of “pollution rebound effect” (PRE). Then, weestimate direct air PRE as a measure for the effectiveness of the policies of reducingair pollution from transport sector based on time-series data from the period 1986 –2014. We find that the short-term direct air PRE is − . , and the correspondinglong-run PRE is − . . The negative results indicate that the direct air PRE doesnot exist in road passenger transport sector in China, either in the short term or inthe long term during the period 1986–2014. This implies that the Chinese transportpolicies are effective in terms of harmful emissions reduction in the transport sec-tor. This research, to the best of our knowledge, is the first attempt to quantify theeffectiveness of the transport policies in the transitional China. Keywords : Direct rebound effect, Air pollution, Road passenger transport, Policy effectiveness Introduction
China is facing a serious environment problem, especially air pollution resulting fromthe rapid economic growth. According to one study of World Bank (2007), twelve of thetwenty most polluted cities in the world are located in China. This ranking is basedon ambient concentrations of particulate matter less than 10 µm in diameter. The am-bient concentration of PM2 . in China is the most polluted in the world based on thereport of World Bank (2016) (see Fig. 1). The State of Environment (SOE) Report of2016 indicates : “Among the 338 prefecture-level cities, there are eighty percent whoseair quality exceed the standard, and 45 cities exceed the annual average concentrationof fine particulate matter more than doubled in 2015.” Serious air pollution has severeeffects on human health, increasing the risk of lung cancer, respiratory and cardiovas-cular diseases (Kunzli et al., 2000; Hoek et al., 2002; Samet, 2007; Beelen et al., 2008;Brunekreef et al., 2009; Weichenthal et al., 2011), which also increases the residents’medical cost (Yang et al., 2013; Chen and He, 2014; Yang and He, 2016). Now more andmore public pay attention to air quality, which poses more pressure on Chinese govern-ment to make scientific and feasible policies to balance between economic developmentand environment problems.Figure 1: Concentration of ambient PM2.5 in different countries, 2013(Data source: World Development Indicators 2016)To control and decrease air pollution, it is necessary to figure out the main sourcesof air pollution. According to “China Vehicle Environmental Management Annual Re- Ministry of Environmental Protection of the People’s Republic of China, 2016. . Thetransport sector is a major area that policymakers should pay more attention. Thelast decade has witnessed a dramatic increase of the vehicles stock, causing the rapidlygrowing travel demand of the Chinese residents. China’s passenger turnover has risenfrom 1746.67 billion passenger-kilometers (pkm) in 2005 to 3009.74 billion pkm in 2014(National Bureau of Statistics of China, 2006–2015)(see Fig. 2), resulting in serious pol-luted air emissions. The total vehicle emissions in China reached to 45.32 million tonsin 2015. Specifically, the emissions of CO , HC , NO x and PM from vehicles were 34.61,4.30, 5.85 and 0.56 million tons, respectively . The transport sector has been a majorfield of harmful emissions reduction. (cid:2)(cid:3)(cid:4)(cid:5) (cid:6)(cid:7)(cid:7)(cid:8) (cid:6)(cid:7)(cid:7)(cid:9) (cid:6)(cid:7)(cid:7)(cid:10) (cid:6)(cid:7)(cid:11)(cid:7) (cid:6)(cid:7)(cid:11)(cid:6) (cid:6)(cid:7)(cid:11)(cid:8) (cid:6)(cid:7)(cid:11)(cid:9) (cid:12) (cid:4) (cid:13)(cid:13) (cid:3)(cid:14)(cid:15)(cid:3) (cid:5) (cid:1)(cid:16) (cid:17) (cid:5) (cid:14)(cid:18) (cid:19) (cid:3) (cid:5) (cid:1) (cid:20) (cid:21) (cid:22)(cid:23)(cid:23)(cid:22) (cid:18)(cid:14) (cid:1) (cid:24) (cid:25) (cid:26) (cid:27) (cid:11)(cid:9)(cid:7)(cid:7)(cid:11)(cid:10)(cid:7)(cid:7)(cid:6)(cid:7)(cid:7)(cid:7)(cid:6)(cid:6)(cid:7)(cid:7)(cid:6)(cid:8)(cid:7)(cid:7)(cid:6)(cid:9)(cid:7)(cid:7)(cid:6)(cid:10)(cid:7)(cid:7)(cid:28)(cid:7)(cid:7)(cid:7)(cid:28)(cid:6)(cid:7)(cid:7)(cid:28)(cid:8)(cid:7)(cid:7)(cid:28)(cid:9)(cid:7)(cid:7) Figure 2: Passenger turnover in China, 2005-2014.(Data source: National Bureau of Statistics of the People’s Republic of China)Chinese government has implemented several laws and policies to deal with the se-rious air pollution. The law “Prevention and Control of Air Pollution” was introduced in1987 by the National People’s Congress and its Standing Committee. A wide range ofregulations, decisions, orders and quality standards have been issued. For example, theState Council promulgated “Atmospheric Pollution Prevention Action Plan” in 2013. In2004, the mandatory fuel economy standard for passenger vehicles was launched andthe first, second and third phases were implemented in 2005, 2008 and 2010, respec-tively. In some megacities (e.g. Beijing, Shanghai and Guangzhou), regulatory policies Ministry of Environmental Protection of the People’s Republic of China, 2016. Data source: Ministry of Environmental Protection of the People’s Republic of China, 2016. . To accurately understand the actual effect of these policies, a measure of reboundeffect (RE) is necessary, which can provide useful information about the effectiveness ofthe policies for the policymakers.The rebound effect is initially proposed by Jevons (1866). It is generally acknowl-edged that when technological progress causes an increase in efficiency by 1%, a reduc-tion in energy consumption obtaining the same products by 1% is expected, whereasthe actual reduction may be below 1%. Studies have identified three main types of re-bound effects (RE) (e.g., Berkhout et al., 2000; Greening et al., 2000; Frondel et al.,2008; Sorrell and Dimitropoulos, 2008): direct rebound effect, indirect rebound effectand macro-level rebound effect.Direct rebound effect is limited to a single energy service or a single sector. With theimprovement of energy efficiency, energy consumption is not reduced to the expectedlevel in theory because of the decline in the cost of energy product or energy service andthe increase in consumers’ energy demand. The issue of pollution rebound effect (PRE)in transport sector also relates to the improvement in energy efficiency. The governmentencourages to improve the energy efficiency of vehicles to reduce harmful emissionsand save energy from the travel. However, fuel-efficient vehicles make energy servicescheaper, thereby encouraging the increased consumption of those services. For instance,consumers may choose to drive farther and/or more often following the purchase of afuel-efficient vehicle because the operating cost per kilometer has fallen. It may offsetsome savings because of fuel efficiency improvement. So there will be rebound effect ofthe fuel consumption, which results in the harmful emissions also appearing reboundeffect. Indirect rebound effect measures the reallocation of energy savings to spendingon other goods and services that also require energy. Macro-level rebound effect refers tothe impact of energy efficiency improvement on the entire economy. This paper focuseson the direct air pollution rebound effect from transport sector.Based on the definition of rebound effect in energy consumption, we firstly define the Ministry of Environmental Protection of the People’s Republic of China, 2016.
P RE (%) =
P lanedEmissionReductions − ActualEmissionReductionsP lanedEmissionReductions × (1)According to the magnitude, PRE can be classified into five categories, which repre-sent different policy effects (see Table 1). When the size of PRE is greater than 0, itrepresents that pollution rebound effect exists. 0 < PRE < > < Table 1: Categories of PRE based on the sizeSize Existence Policy implicationPRE > = < PRE < = < Some studies have examined the rebound effect in transport sector, but most arefocus on the fuel consumption. Sorrell et al. (2009) provide a review of studies thatinclude transportation and energy in general. They report that for personal automotivetransport, in OECD countries, the mean value of the long-run direct rebound effectis likely to be less than 30% and may be closer to 10% for transport. Small and VanDender (2007), examining motor vehicle transportation in the US, estimate the shortand long-run rebound effect of 4.5% and 22.2%, respectively. Barla et al. (2009) presentestimates of the rebound effect for the Canadian light-duty vehicle fleet. Their resultsimply a rebound effect of 8% in the short term and a little less than 20% in the long term.Hymel and Small (2015), using panel data on U.S. states, confirm the earlier finding ofa rebound effect that declines in magnitude with income, but they also find an upwardshift in its magnitude of about 0.025 during the years 2003–2009. Wang et al. (2012)and Zhang et al. (2015) also estimate the direct rebound effect for passenger transportin China. Although all the current studies conclude that the rebound effect exists infuel consumption for transport sector, the range of the magnitude is very different. Forexample, Wang et al. (2012) estimate the direct rebound effect for passenger transport inurban China, finding that the average rebound effect for passenger transport by urban6ouseholds is around 96%. Zhang et al. (2015), analyzing road passenger transport inthe whole country, eastern, central and western China, reveal that the short-term andlong-term direct rebound effects of the whole country are 25.53% and 26.56% on average,respectively. Furthermore, we can not find any study on the pollution rebound effect.In summary, it is necessary to explore that whether direct air pollution rebound effectexist in the road passenger transport in China. To the best of our knowledge, this paperis the first attempt in current literature to evaluate the direct air pollution reboundeffect. The results can provide useful information for policy makers to understand theeffectiveness of the policies, which aim to reduce harmful emissions of transport sector.The remaining of this paper is structured as follows. Section 2 introduces the meth-ods used to estimate the rebound effect of air pollution as well as data definitions. Sec-tion 3 present the empirical results and detailed discussions. Finally, Section 4 summa-rizes our results and offers some policy implications.
In this paper we explore the existence of direct air pollution rebound effect for the roadpassenger transport sector in China during 1986–2014. To estimate the direct air pollu-tion rebound effect, we firstly need to calculate the emissions reduction from transportsector based on the definition of PRE (see Eq. (1)). Here, we have an assumption thatthe emission factor remains unchanged. The details are provided in Section 2.1. So wecan directly calculate the fuel consumption reduction. According the analysis, we findthat we can estimate the fuel rebound effect, which is equal to the PRE. Following thedefinition by Khazzoom (1980) and Berkhout et al. (2000), we can calculate the reboundeffect according to the elasticity of fuel consumption with respect to fuel efficiency. Inthis paper, we use the elasticity of vehicle kilometers (VKM) with respect to fuel priceas a proxy.
Like the estimation in the literature by Chen and He (2014), we calculate the air pollu-tant emissions of transport by the following equation: H i = F m × A m,i (2)where H refers to the traffic-related harmful gases emissions, F the consumption ofthe fuel from transport, A the emission factor; m , i refer to the fuel type and pollutanttype, respectively. 7n the last thirty years, the technology of vehicles does not have revolutionary inno-vation, which still mainly burn fuel. The emissions still have strong relationship withfuel consumption. So in this paper we suppose the emission factor remains unchanged.When the fuel consumption has rebound effect, then the air pollution rebound effect willoccur. The both two are equal.According to the definition of rebound effect (e.g. Khazzoom, 1980; Berkhout et al.,2000), the estimation can be calculated by the following equation: RE = η E ( F ) + 1 (3)where RE is the rebound effect; F is the consumption of fuel; E is fuel efficiency; η E ( F ) refers to the elasticity of fuel consumption with respect to fuel efficiency.Following the previous study by Odeck and Johansen (2016), the relationship be-tween the elasticity of fuel consumption with respect to fuel efficiency and elasticity ofvehicles kilometers traveled (VKM) demand with respect to fuel price is like that: η E ( F ) = − η P ( V KM ) − (4)where η P ( V KM ) is the elasticity of VKM demand with respect to fuel price. Com-bined Eq. 3 and Eq. 4, the negative η P ( V KM ) is used as a proxy measure for the re-bound effect. We adopt the logarithmic model to measure the short-term direct reboundeffect for road passenger transport, then we estimate the corresponding long-term directrebound effect. Following most studies in the literature (e.g. Small and Van Dender, 2007; Zhang et al.,2015), we choose the logarithmic equations to investigate the size of elasticity of VKMdemand with respect to fuel price. The most common assumption in the literature isthat fuel price and income are the only explanatory variables for VKM demand (e.g.,Alves and Bueno, 2003; Akinboade et al., 2008; Sene, 2012). However, considering thatthe number of vehicles explains some degree of the demand for travel, in this paperwe choose price level, income level and vehicle stock as the determinants for passengertravel following by Odeck and Johansen (2016). After adding the time-lagged VKM,we take the logarithmic operation to all variables before regression in Eq.(5), where
V KM t refers to per capita demands for travel; Y t is real income per capita; P t is the realprice of fuel; V t is vehicle stock; V KM t − i − is the time-lagged VKM; the vector Λ are the8arameters to be estimated; and ǫ t is residuals for VKM demand, at time t .ln V KM t = λ + λ Y ln Y t + λ P ln P t + λ V ln V t + q X i =0 λ vkmi ln V KM t − i − + ǫ t (5)We find that the result of one-ordered lagged ln V KM is not significant throughEq.(5), using the time-series data from 1986–2014. Then we use the two-lagged ln
V KM ,namely, ln
V KM t − . Finally, the VKM demand equation can be written as follow:ln V KM t = λ + λ Y ln Y t + λ P ln P t + λ V ln V t + λ vkm ln V KM t − + ǫ t (6)where the meaning of the variables and parameters are the same as in Eq.(5). Inthis way, − λ P and − λ P − λ vkm are the size of short-term and long-term direct air pollutionrebound effect for road passenger transport in China. By collecting the time-series data from 1986 to 2014, we explore the existence of di-rect air pollution rebound effect for road passenger transport in China and estimate themagnitudes of short-term and long-term direct rebound effects, respectively. The uti-lized data are all the annual average data. National Bureau of Statistics of the People’sRepublic of China, which has collected official statistics on Chinese society , providesmacroeconomic data, such as disposable income, population, vehicle stock and road pas-senger turnover. Diesel price data is obtained from “Price Yearbook of China” (EditorialDepartment of Price Yearbook of China, 1989-2015) and “Price Statistical Yearbook ofChina” (National Bureau of Statistics of China, 1988-1989). Variables that required con-version to their per capita forms are divided by the total annual population. Descriptivestatistics of all variables are shown in Table 2, including the logarithmic forms of vehi-cles kilometers traveled per capita, the disposable income per capita, diesel price andvehicle stock. Here we use per capita road passenger turnover to presents the demand for travel. The real income refersto the disposable income in the whole country. We use the 0 V t refers to road commercial passenger vehicles. able 2: Descriptive statistics of the variables in China, 1986–2014Variable lnVKM lnY lnP lnVMinimum 2.2656 2.7330 3.0336 4.9571Maximum 3.1348 4.3126 3.9354 6.6426Mean 2.7032 3.5524 3.4951 5.6985Std. Dev. 0.2558 0.4790 0.2901 0.5849Skewness -0.1121 -0.1432 0.0843 0.0350Kurtosis 1.8700 1.9057 1.8197 1.3909Observations 29 29 29 29 A specific issue regarding the data’s stationarity properties must be considered usingtime-series data. If two time-dependent variables follow a common trend that causethem to move in the same direction, it is possible to observe a significant correlationbetween them, even if there is no “true” association. This potential problem with time-series data can lead to a spurious regression. To avoid the mistake resulting from spuri-ous regression problems, all the variables are checked for their stationarity properties.We employ both the Dickey and Fuller (1979) test and the Phillips and Perron (1988) testto determine the presence of a unit root. Table 3 presents the tests for the stationarityof the variables.
Table 3: DF and PP test for the presence of unit root in level and differenced variablesVariable DF test 1% critical value 5% critical value 10% critical value lnV KM -1.576 -3.730 -2.992 -2.626 lnP -0.681 -3.730 -2.992 -2.626 lnY -0.994 -3.730 -2.992 -2.626 lnV -1.293 -3.730 -2.992 -2.626 ∆ lnV KM -5.190*** -3.736 -2.994 -2.628 ∆ lnP -5.545*** -3.736 -2.994 -2.628 ∆ lnY -2.800* -3.736 -2.994 -2.628 ∆ lnV -4.117*** -3.736 -2.994 -2.628Variable PP test 1% critical value 5% critical value 10% critical value lnV KM -1.643 -3.730 -2.992 -2.626 lnP -0.667 -3.730 -2.992 -2.626 lnY -0.791 -3.730 -2.992 -2.626 lnV -1.376 -3.730 -2.992 -2.626 ∆ lnV KM -5.193*** -3.736 -2.994 -2.628 ∆ lnP -5.611*** -3.736 -2.994 -2.628 ∆ lnY -2.919* -3.736 -2.994 -2.628 ∆ lnV -4.117*** -3.736 -2.994 -2.628Notes: All variables are I(1) aa *** indicates the significance at 1% level. * indicates the significance at 10% level. ∆ lnV KM denotes thefirst-order difference of lnV KM , with the similar meaning to other variables. As is shown, all the variables are nonstationary at various levels for the reason thatall results of DF test and PP test cannot reject the null hypothesis at the 10% signifi-cance level. Then we test the stationarity of the first order difference of each variable,and the DF and PP tests indicate that the first differences exceed the critical value forall the variables, which indicates that all the variables are stationary in first differ-10nces, i.e., all the series are I(1). These results allow us to conduct the cointegration testto estimate whether a long-term equilibrium relationship exists among these variables,which is presented in Table 4. According to the Johansen (1988) test, the results of thetrace statistic and the max statistic both imply that none cointegration relationship isrejected at the 5% significance level, which means the existence of one long-run relation-ship. Hence, we can conclude that the long-run relationship does exist among vehicleskilometers traveled per capita, the disposable income per capita, diesel price and vehiclestock.
Table 4: Results of Johansen test for cointegrationRank LL Trace statistic 5% critical value Max statistic 5% critical value0 192.45 64.93 54.64 36.37 30.331 210.64 28.56 34.55 16.23 23.782 218.75 12.33 18.17 7.95 16.873 224.92 4.38 3.74 4.38 3.74
Since the variables are found to be cointegrated, we develop logarithmic regressionmodel to estimate the coefficients according to Eq.(6), which is reported in Table 5. Theadjusted R-squared value is relatively high. The estimated coefficients are statisticallysignificant. We can identify the following findings.
Table 5: OLS estimation of the logarithmic regression modelDependent variable lnV KM t Explanatory variables Coefficient SE t-statistic P value lnP t lnY t lnV t lnV KM t − -0.6690 0.3462 -1.93 0.066* λ First, from the direct meaning of estimated coefficients, the static elasticities of traveldemand (VKM) with respect to fuel price, income and vehicle stock are 0.4105, 0.6023and 0.0644, respectively. The elasticity of fuel price and vehicle stock are significantat the 5% significance level. The result of income is significant at the 1% level. Theseresults imply that an increase in the fuel price of 10% would increase travel demandper capita by 4.105%; an increase of disposable income per capita of 10% would cause anincrease in travel distance per capita of 6.023%; and the travel demand per capita wouldincrease 0.644% if vehicle stock increases by 10%. The magnitude of income elasticityis close to the previous studies. For instance, Zhang et al. (2015), based on data of 30provinces from 2003 to 2012, estimate that the elasticity of passenger kilometers with11espect to gross domestic product per capita (PGDP) is 0.7907 in whole China. The resultof vehicle stock is relatively smaller than the income and fuel price. One reason may bethat in this model we mainly use the road commercial passenger vehicles. Althoughthe stock has an increase during the last thirty years, the growth rate of population inChina is much bigger than that of vehicle stock.Attention is drawn to a strong difference: some studies estimate the elasticity oftravel demand with respect to fuel price of a negative value (e.g., Zhang et al., 2015;Odeck and Johansen, 2016; Barla et al., 2009), which means that when the fuel pricesincrease, the travel demand will decrease. However, our result is different from theseresults, which is reasonable in this paper. We mainly study the road passenger trans-port, and use road passenger turnover per capita as the proxy. When the diesel priceincreases, generally the gas price also increases as well. So the cost of taking commercialvehicles is relatively cheaper than taking private cars, resulting in the residents choos-ing commercial vehicles more. Furthermore, this model is based on the time-series data.China experiences a dramatic increase in the last thirty years. The growth rate of dieselprice is largely smaller than the residents’ travel demand. So from the result of thismodel, we find that when the diesel price increases, the travel demand also increases.Second, this paper most concerns the existence of air pollution rebound effect for roadpassenger transport sector. From the results of Table 5 and Eqs.(3) and (4), the short-term PRE can be estimated as -1.4105. This negative estimation means that the directair pollution rebound effect does not exist in road passenger transport sector of wholeChina in the short-term during 1986–2014 based on the Table 1. The harmful gasesemissions reduction is more than 1% in the short-term when the fuel efficiency of vehi-cles improves by 1% based on the unchanged emission factors. This result implies thatthe policies that control air pollution from transport sector are effective. The policiesnot only achieve the initial emissions reduction goal, but also exceed expectations.In addition, the corresponding long-run PRE are obtained as -1.246. The negativeresult also implies direct PRE does not exist in the long-term during 1986–2014. How-ever, the long-run PRE is smaller than the short-term, which means the effect of harm-ful emissions reduction declines than the short-term. This result puts forward a newquestion that whether the PRE will occur for a long time. Considering the existenceof direct energy rebound effect for transport sector in developed countries (e.g., Odeckand Johansen, 2016; Hymel et al., 2010; Barla et al., 2009), it is necessary to study thisproblem further in future. 12
Conclusion
This study, to the best of our knowledge, is the first attempt to explore whether there isdirect air pollution rebound effect for road passenger transport in China, based on time-series data from the period 1986–2014. Our empirical results indicate that direct PREdoes not exist in road passenger transport sector of whole China during 1986–2014. Thepolicies, which aim to reduce harmful emissions of transport sector, are not only fullyachieve the expected benefits, but also exceed the expectations. The results imply thatimproving fuel efficiency of vehicles is a useful policy option for decreasing the transportenergy use, resulting in the reduction of harmful gases emissions.Our empirical study show that the effect of harmful emissions reduction in the long-term declines than the short-term during 1986–2014. With China’s development, the airpollution of transport sector in China may be occur rebound effect in future. The policymakers should consider the possibility of rebound effect to avoid overestimating harmfulemissions reduction achieved by implementing some policies for transport sector.Although there is no study on the air pollution rebound effect, there are studies aboutthe energy rebound effect for transport in China. Wang et al. (2012) and Zhang et al.(2015) both find that there exists direct energy rebound effect for transport. However,their results are very different from each other. Wang et al. (2012), based on the dataof 28 provinces during 1994–2009, find that the average rebound effect for passengertransport by urban households is around 96% by employing the LA-AIDS model, in-dicating that the majority of the expected reduction in transport energy consumptionfrom efficiency improvement could be offset. Zhang et al. (2015), based on the data of 30provinces during 2003–2012, find that the average sizes of short-term and long-term re-bound effect are 25.53% and 26.56% in the whole country through dynamic panel datamodel. This difference may be on account of the different methods and observations(Zhang et al., 2015). This prompts our further research can be conducted to estimatethe different regions of China and use the panel data to get more observations.Furthermore, as for the future work, further research can be conducted to combinethe gasoline and diesel together, and choose more factors influencing travel demand todownsize related bias as much as possible.13 eferences