The relationship between maximum walking distance and the final destination parking tariff: a panel analysis

The relationship between maximum walking distance and the

final destination parking tariff

Citation for published version (APA):

van der Waerden, P. J. H. J., Borgers, A. W. J., & de Bruin - Verhoeven, M. (2016). The relationship between maximum walking distance and the final destination parking tariff: a panel analysis. Paper presented at 5th Symposium of the European Association for Research in Transportation (hEART 2016), Delft, Netherlands.

Document status and date: Published: 01/01/2016 Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

The Relationship between Maximum Walking Distance and the Final Destination Parking Tariff: a Panel Analysis

Peter VAN DER WAERDEN

Urban Planning Group, Eindhoven University of Technology,

PO Box 513, 5600 MB Eindhoven, The Netherlands, p.j.h.j.v.d.waerden@tue.nl Aloys BORGERS

Urban Planning Group, Eindhoven University of Technology, PO Box 513, 5600 MB Eindhoven, The Netherlands, a.w.j.borgers@tue.nl

Marloes DE BRUIN

School of Education, Eindhoven University of Technology,

PO Box 513, 5600 MB Eindhoven, The Netherlands, m.d.bruin.verhoeven@tue.nl

Abstract

This paper describes a study regarding car drivers’ preferences of walking distance between parking facility and final destination in relation to parking tariff and personal characteristics. The included parking tariffs ranged from 1.50 euro to 3.50, in steps of 50 eurocents. The following distance categories were specified: less than 100 meter, between 100 and 250 meter, between 250 and 500 meter, between 500 and 1000 meter, and more than 1000 meter. Based on data gathered in the City of ‘s-Hertogenbosch in 2013, a random effects ordered probability model with panel effects is estimated. The data of 422 respondents were included in the analyses. The model goodness-of-fit shows that the optimal model performs quite well. The estimation results show that parking tariff and personal characteristics gender, age, education, home location, and car type influence the probability of distance categories significantly.

Keywords: walking distance, parking tariff, ordered logit

1. INTRODUCTION

Previous studies showed that the attractiveness of parking facilities is mainly determined by the parking tariff and the walking distance between parking and the car driver’s final destination (e.g., Guan et al., 2005; Marsden, 2006; Shoup, 2006; Van Ommeren et al., 2012; Kobus et al., 2013). In general, it appears that the lower the parking tariff and the shorter the walking distance, the higher the attractiveness of a parking facility is (e.g., Hakkesteegt & Radema, 1979; Van der Waerden, 2012). However, it is also known that car drivers make a trade-off between the parking costs and the maximum distance they are willing to walk (e.g., Bonsall & Palmer, 2007). Some car drivers are prepared to pay more for a parking facility that is closer to their final destination. The same holds for the opposite: pay less for a parking facility at a larger distance from the final destination. Most municipalities in the Netherlands offer more expensive parking facilities close to the main destinations and cheaper parking facilities at some distance from the destination (Figure 1).

Car drivers’ opinion regarding the relationship between the maximum walking distance from parking facility to final destination and the accompanying parking tariff of the parking facility is still unclear. Also the influences of context variables like personal and trip characteristics on this relationship have not been investigated in detail yet. The aim of this study is to provide more insight into the relationship between walking distance and parking tariff, and the influence of personal characteristics on this relationship. The study is an extension of a previous study of Van der Waerden et al. (2015). In that study, only the maximum distance was considered.

The remainder of the paper is organized as follows. First, the adopted research approach is described. This section is followed by a description of the data collection method and the composition of the sample. In the next section, the setup and the results of model analyses are presented. The paper ends with the conclusions and some suggestions for future research.

Figure 1 Tariff structure and walking distance in the center of Amsterdam, the Netherlands

2. RESEARCH APPROACH

The study was set up in the city of ‘s-Hertogenbosch, in the South of the Netherlands. The city is the capital of the province ‘Noord-Brabant’ and hosts approximately 145,000 inhabitants.

The historical

city center includes approximately 500 stores and 250 bars and restaurants. The city center is

surrounded by 7 public parking facilities (Figure 2; in blue) and 3 Park & Ride facilities

(Figure 2; in red). Users of these facilities were invited to participate.

To get more insight into the relationship between acceptable walking distance and parking tariff, the following research approach was adopted. The parking tariffs ranged from 1.50 euro to 3.50, in steps of 50 eurocents. The following distance categories were specified: less than 100 meter, between 100 and 250 meter, between 250 and 500 meter, between 500 and 1000 meter, and more than 1000 meter. The selected categories of both the distance and the tariff are in line with the existing levels in the city of ‘s-Hertogenbosch.

Figure 3 Screenshot of the online questionnaire

Respondents were invited to identify the maximum walking distance they are prepared to walk given various parking tariffs (Figure 3). The question was included in an extensive online questionnaire covering several visit (frequency, motif, etc.) and parking (frequency, duration, preferences, etc.) related topics. The questionnaire ended with some questions regarding personal characteristics of the respondent: gender, age, education, origin, and car type.

3. DATA COLLECTION

In the spring of 2013, approximately 8000 invitation cards for the online questionnaire were distributed across the various parking facilities (see Figure 2). In collaboration with the municipality, the invitation cards were handed to car drivers who used one of the parking facilities in the city center or a Park and Ride facility at some distance from the city center.

In total, 502 car drivers filled out the questionnaire. The data of 422 car drivers could be used for the analyses. Some details of the respondents are presented in Table 1. Because of a lack of information, it is difficult to compare the sample with the composition of the visiting population of the city center of ‘s-Hertogenbosch. The sample includes more males than females, more older than younger respondents, and more higher than lower educated respondents. These findings are not very different from findings in other comparable cities in the Netherlands. However, given the number of observations per category of these characteristics, the effects of these characteristics can be estimated.

Table 1: Personal characteristics of the respondents

Characteristic Levels Frequency Percentage Coding*

Gender Female

Male 172 250 40.8 59.2 -1 1

Age Younger than 35 years From 35 to 55 years Older than 55 years

83 214 125 19.7 50.7 29.6 1 0 0 1 -1 -1 Education Professional Higher professional Scientific 147 174 101 34.8 41.2 23.9 1 0 0 1 -1 -1 Home location ‘s-Hertogenbosch

Adjacent municipalities Other 114 104 204 27.0 24.6 48.3 1 0 0 1 -1 -1 Car type Small

Medium Large 55 221 145 13.3 52.4 34.4 1 0 0 1 -1 -1 Total 422 100.0

* Coding used in the model analysis

Table 2 presents for each parking tariff category, the distance category that respondents have chosen. The table clearly shows that the maximum walking distance respondents accept decreases with an increase of the parking tariff. In the case of 1.50 euro, almost 80 percent of the respondents accepts a maximum walking distance of 500 meters or more. This percentage decreases with the increase of the parking tariff. Almost 80 percent of the respondents accepts a maximum walking distance of less than 100 meters when the parking tariff is 3.00 euro or higher.

Table 2: Observed frequencies of parking tariff and maximum walking distances (percentages)

Tariff <100 m 100-250 m 251-500 m 501-1000 m >1000 m Total 1.50 euro 2.00 euro 2.50 euro 3.00 euro >3.00 euro 11.6 18.5 33.4 53.3 77.3 11.8 24.2 30.3 30.3 10.4 32.0 32.5 26.5 9.7 7.1 22.5 17.5 4.0 2.4 1.2 22.0 7.3 5.7 4.3 4.0 100.0 100.0 100.0 100.0 100.0 Total N 38.8 819 21.4 452 21.6 455 201 9.5 183 8.7 100.0 2110 4. ANALYSES

For each respondent the five observations (the selected max. distance for each of the five tariff categories) are included in the data set. In the analyses, these five observations are considered as panel data. The panel data are analyzed using a random effects ordered probability model. The selected maximum walking distance is used as dependent ordinal scaled variable. It ranges from 0 (Less than 100 meters) to 4 (More than 1000 meters). The model includes the parking tariff levels as independent variables and various personal characteristics (gender, age, education, home location and car type) as context variables. The model is estimated using Nlogit version 5.0 (Econometric Software Inc).

The random effects ordered probability model is formulated as follows (Green & Hensher, 2010):

it i it it

x

u

y











j

y



u



y



u

Where,

y*

it is the continuous latent utility or “measure” for answer category i and time period t;

xit is a set of K covariates that are assumed to be strictly independent of εit;

β is a vector of K parameters that is the object of estimation and inference;

ui is threshold parameter of answer category i;

yit is the observed discrete outcome of answer category i and time period t;

)

(

~

f



it



)

(

~

g



u

it for all t. The latent utility , y*it, is observed in discrete from through a censoring mechanism:

yit = 0 if y*it ≤ u0,

yit = 1 if u0 < y*it ≤ u1,

yit = 2 if u1 < y*it ≤ u2

yit = ….

yit = J if uJ-1 < y*it ≤ uJ.

The results of the model estimation are presented in Table 2. According to the Chi-square test (-2[LLrestricted – LLoptimal]), the estimated random effects model significantly outperforms the ordered probability model with fixed parameters only. The model represents the observed preferences satisfactory (McFadden’s Rpseudo-squared equal to 0.260).

Table 3 Parameter estimates of the random effects ordered logit model

Attributes Levels Parameter Significance

Parking tariff Gender Age Education Home location Car type Thresholds

Std deviation of random effect

Male

Female

Younger than 35 years From 35 to 55 years

Older than 55 years

Professional Higher professional Scientific ‘s-Hertogenbosch Adjacent municipalities Other Small Medium Large Mu(01) Mu(02) Mu(03) Sigma -3.6024 2.7201 -2.7201* -0.1135 3.2504 -3.1369 2.9564 2.1298 -5.0862 -4.8078 0.9762 3.8316 -9.9540 6.8213 3.1327 2.6592 5.8987 8.3503 9.6381 0.000 0.000 0.333 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Goodness-of-fit

Log-likelihood optimal model Log-likelihood restricted model Chi-squared (1 degree-of-freedom) McFadden’s Rpseudo-squared

Adjusted Rpseudo-squared

-2298.2653 -3107.2936 1618.06 (0.000)

0.260 0.256 * _{Value based on coding (see Table 1)}

The estimated threshold values result into the following cutting points of the included distance categories: yit = 0 (<100 m) if y*it ≤ 0, yit = 1 (100-250 m) if 0 < y*it ≤ 2.6592 yit = 2 (251-500 m) if 2.6592 < y*it ≤ 5.8987 yit = 3 (501-1000 m) if 5.8987 < y*it ≤ 8.3503 yit = 4 (> 1000m) if y*it > 8.3503

Most included parameters are significant at 99 percent significance level. Only the effect of the age group younger than 35 years is not significant. The estimated parameters show the following trends. The negative parameter of the attribute Parking Tariff shows that the probability of choosing a higher distance category deceases with an increasing parking tariff. For the other parameters, a negative sign indicates that the probability of a lower distance category increases and the probability of a higher distance category decreases. For a positive sign the opposite holds. For example, it seems that males accept the longer distance categories, while females prefer the shorter distance categories. Women appear to be more distance sensitive than men. As expected, older respondents prefer shorter distance categories. The same is valid for respondents living in the city of ‘s-Hertogenbosch. Respondents with a higher educational level prefer shorter walking distances. Respondents who use a small car when traveling to the city center also prefer the smaller walking distance categories.

5. CONCLUSIONS

The study described in this paper aims to understand the relationship between maximum walking distance between parking facility and final destination, and parking tariff. Based on car drivers’ responses on an online questionnaire, a random effects ordered probability model with panel effects is estimated. The model’s goodness-of-fit shows that the model performs quite well. Almost all parameters of the included attribute levels are significant and effects are as expected. This finding shows that not only the parking tariff but also personal characteristics play a role when car drivers evaluate the maximum walking distance.

The results of this study could be used by both municipalities and parking management organizations. Both stakeholders can optimize the parking tariff of individual parking facilities based on the location of the facility and the main type of car drivers who use the facility. Of course, when setting the tariff structure also attention has to be paid to additional services offered by the parking facility.

At this moment the study has some limitations. First, no attention is paid yet to respondents who filled in only one distance category for all parking tariffs (so-called: ‘non-trading behavior’). The influence of these respondents on the estimation results could be explored in more detail. Second, the attributes levels of the personal characteristics are all set to three (except gender). Some detailing or redefining of these levels is possible. Also the interrelationship between the various characteristics is worthwhile to investigate in more detail. Finally, the model could be extended with trip related characteristics.

Acknowledgement

The authors want to thank Koen van Waes and Marcel Berends of the Municipality of ‘s-Hertogenbosch for their contribution to the study.

6. REFERENCES

Bonsall, P. & Palmer, I. (2004) Modeling Drivers’ Car Parking Behavior Using Data from a Travel Choice Simulator, Transportation Research C, 12, pp. 321-347.

CROW (2003) Walking Distances in Shopping Areas (in Dutch), In: From Parking Management to

Mobility Management, Part 7. Center for Research and Contract Standardization in Civil and Traffic

Engineering, Ede, the Netherlands

Econometric Software Inc. (2012) NLOGIT version 5, Economic Software Inc., Plainview, USA. Green, W.H. & Hensher, D.A. (2010) Modeling Ordered Choices: A Primer, Cambridge University Press, Cambridge, UK.

Guan, H., Sun, X, Liu, X., and Liu, L. (2005) Modeling Parking Behavior of Better Control and Pricing: A Case Study from One of the Busiest Retail Shopping Areas in Beijing, China, Compendium

of Papers of the 84th Annual Meeting of the Transportation Research Board, Washington DC, USA.

Hakkesteegt, P. & Radema, B. (1979) Parking Facilities in Urban Centers (in Dutch), Verkeerskunde, pp. 118-123.

Kobus, M.B.W., Gutiérrez-i-Puigarnau, E., Rietveld, P. & Van Ommeren, J.N. (2013) The On-Street Parking Premium and Car Drivers’ Choice between Street and Garage Parking, Regional Science and

Urban Economics, 43, pp. 395-403.

Marsden, G. (2006) The Evidence Base for Parking Policy – A Review, Transport Policy, 13, pp. 447-457.

Shoup, D.C. (2006) Cruising for Parking, Transport Policy, 13, pp. 479-486.

Van der Waerden, P. (2012) Pamela, Parking Analysis Model for Predicting Effects in Local Areas, Thesis, Eindhoven University of Technology, Eindhoven, The Netherlands.

Van der Waerden, P., Timmermans, H. & De Bruin, M. (2015) Car Drivers’ Characteristics and the Maximum Walking Distance between Parking Facility and Final Destination. Paper presented at the 2014 World Symposium on Transport and Land Use Research, Delft, the Netherlands.

Van Ommeren, J.N., Wentink, D. & Rietveld, P. (2012) Empirical Evidence on Cruising for Parking,